falta el logo - efccnaefccna.org/images/stories/publication/niv_book.pdf · sección de neumología...

Finanantial support by

Endorsed by Spanish Pediatric Intensive Care Society

FALTA El logo

All rights reserved. No part of this publication may be reproduced or distributed in any form or bymeans, or stored in a data base or retrieval system, without the prior written permission of thepublished.

© 2009 by ErgonC/ Arboleda, 1. 28221 Majadahonda (Madrid)Pza. Josep Pallach 12. 08035 Barcelona

ISBN: ???Depósito Legal: ???

Writing the second edition of “Non-Invasive Ven-tilation in Pediatrics” was truly a team effort, dra-wing on the input and assistance of numerousindividuals.

The first edition was well received and provedhighly useful for the courses organized by the Res-piratory Group of the SECIP. Therefore, we wouldfirst like to thank all of the students and participantsin these courses. These individuals, in continuing toamass experience at their centers, have shown usthat our text has addressed a real need in the medi-cal field.

We also express our gratitude to the doctors andhospital staff from our wards: thanks to these pro-fessionals, some wards have been able to quintuplethe number of patients treated with NIV in less thanthree years, and others have been able to incorpo-rate NIV as a routine procedure.

We are also indebted to all of the pediatric inten-

sivists whose wards provided data for the EPIVE-NIP study, as well as those who contributed to eva-luating the quality control indices in the new chapteron healthcare assistance, quality control and clini-cal training for NIV.

We would also like to thank A. Alarcon Allen(Medical Attending of Neonatal Department. Hos-pital Sant Joan de Deu, Barcelona) and RubénD. Restrepo (MD RRT FAARC, Associate Professor,Department of Respiratory Care, The Universityof Texas Health Science Center at San Antonio,San Antonio, Texas) who supervised the englishtraslation of this book and make many useful advi-ces.

Lastly, those of us at Hospital Sant Joan de Déuwould like to thank the Invest4Children Foundation,which has enabled us to provide NIV to far morechildren than we could have ever dreamed of fouryears ago.

Acknowledgements

Alberto Medina VillanuevaMartí Pons Òdena

Federico Martinón-Torres

We dedicate this book

to our wives and children

for keeping us up all night

after we get off of duty

Carmen Antelo LandeiraSección de Neumología Pediátrica.Hospital Infantil La Paz (Madrid)

Marta Los Arcos SolasUnidad de Cuidados Intensivos Pediátricos.Hospital Universitario Central de Asturias (Oviedo)

Isabel Barrio Gómez de AgüeroSección de Neumología Pediátrica. Hospital Infantil «La Paz» (Madrid)

Mónica Balaguer GargalloUnidad de Cuidados Intensivos Pediátricos.Hospital Universitario Sant Joan de Déu (Barcelona)

José Manuel Blanco GonzálezUnidad de Cuidados Intensivos Pediátricos. Hospital Universitario Sant Joan de Déu (Barcelona)

Francisco José Cambra LasaosaUnidad de Cuidados Intensivos Pediátricos.Hospital Universitario Sant Joan de Déu (Barcelona)

Félix Castillo SalinasServicio de Neonatología. Hospital Vall d'Hebrón (Barcelona)

Andrés Concha TorreUnidad de Cuidados Intensivos Pediátricos.Hospital Universitario Central de Asturias (Oviedo)

Lola Elorza FernándezServicio de Neonatología. Hospital Universitario La Paz (Madrid)

Antonio Esquinas RodríguezServicio de Cuidados Intensivos.Hospital Morales Meseguer (Murcia)

Miguel José dos Santos FélixLaboratorio de sueño y ventilación.Hospital Pediátrico de Coimbra (Portugal)

José Ramón Fernández LorenzoServicio Neonatologia-UCI neonatal. Hospital Clinico Universitario. Santiago de Compostela (La Coruña)

Paula Fernández DeschampsUnidad de Cuidados Intensivos Pediátricos.Hospital Niño Jesús (Madrid)

Maria Luisa Franco FernéndezHospital General Universitario Gregorio Marañón (Madrid)

Mirella GaboliUnidad de Cuidados Intensivos Pediátricos.Hospital Universitario de Salamanca

M ª Ángeles García TeresaUnidad de Cuidados Intensivos Pediátricos.Hospital Niño Jesús (Madrid)

Julio García-Maribona Rodríguez-MaribonaUnidad de Cuidados Intensivos Pediátricos.Hospital Universitario Central de Asturias (Oviedo)

Teresa Gavela PérezUnidad de Cuidados Intensivos Pediátricos.Hospital Niño Jesús (Madrid)

Teresa Gili BigatàUnidad de Cuidados Intensivos Pediátricos.Corporació Sanitaria Parc Taulí (Sabadell. Barcelona)

Pedro Gómez de Quero MasiaUnidad de Cuidados Intensivos Pediátricos.Hospital Universitario de Salamanca

Authors

Miguel GonçalvesUnidad de Fisiopatología y Rehabilitación.Departamento de Medicina Pulmonar.Hospital Sao Joao. Porto (Portugal)

Margarita González PérezUnidad de Cuidados Intensivos PediátricosHospital Universitario Central de Asturias (Oviedo)

Manuel Gresa MuñozComplejo Hospitalario UniversitarioMaterno-Insular de Las Palmas

Antonio Gutiérrez LasoHospital Universitario La Fe (Valencia)

Emili Ibiza PalaciosUnidad de Cuidados Intensivos Pediátricos.Hospital Universitario La Fe (Valencia)

Maria Helena Lopes EstêvãoLaboratorio de sueño y ventilación.Hospital Pediátrico de Coimbra (Portugal)

Jesús López-Herce CidSección de Cuidados Intensivos Pediátricos.Hospital General Universitario Gregorio Marañón (Madrid)

Jon López de Heredia GoyaUCI NeonatalHospital de Cruces (Bilbao)

Antonio Losada MartínezHH.UU. Virgen del Rocío. Hospital de la Mujer. Hospital Infantil (Sevilla)

Carmen Martínez CarrascoSección de Neumología Pediátrica.Hospital Infantil La Paz (Madrid)

Federico Martinón-TorresServicio de Críticos y Urgencias.Hospital Clínico Universitario (Santiago de Compostela)

Juan Mayordomo ColungaUnidad de Urgencias Pediátricas.Hospital Universitario Central de Asturias (Oviedo)

Alberto Medina VillanuevaUnidad de Cuidados Intensivos Pediátricos.Hospital Universitario Central de Asturias (Oviedo)

Sergio Menéndez CuervoUnidad de Cuidados Intensivos Pediátricos.Hospital Universitario Central de Asturias (Oviedo)

Xavier Miracle EchegoyenServicio de Neonatología. Hospital Clínic-Casa Maternitat (Barcelona)

Vicent Modesto i AlapontUnidad de Cuidados Intensivos Pediátricos.Hospital Universitario La Fe (Valencia)

Juan Carlos Monroy Mogollón†Unidad de Cuidados Intensivos Pediátricos.Hospital Universitario Sant Joan de Déu (Barcelona)

Julio Moreno HernandoServicio de Neonatología. Unidad de Cuidados IntensivosNeonatales. Hospital Universitario Sant Joan de Déu (Barcelona)

Mª José Pérez GarcíaUnidad de Cuidados Intensivos Pediátricos.Hospital Niño Jesús (Madrid)

Martí Pons ÒdenaUnidad de Cuidados Intensivos Pediátricos.Hospital Universitario Sant Joan de Déu (Barcelona)

Javier Pilar OriveUnidad de Cuidados Intensivos Pediátricos.Hospital de Cruces. Baracaldo

Raquel Porto AbalUnidad de Cuidados Intensivos Pediátricos.Hospital Niño Jesús (Madrid)

Corsino Rey GalánUnidad de Cuidados Intensivos Pediátricos.Hospital Universitario Central de Asturias (Oviedo)

Antonio Rodríguez NuñezServicio de Críticos y Urgencias. Hospital Clínico Universitario (Santiago de Compostela). Departamento de Pediatría.Universidad de Santiago de Compostela

Adolfo Valls SolerServicio de Neonatología. Catedrático de Pediatría de la Universidad del País VascoHospital de Cruces (Bilbao)

Silvia Vidal i MicóUnidad de Cuidados Intensivos Pediátricos.Hospital Universitario La Fe (Valencia)

Reading medical texts that are more than tenyears old usually provokes one of two reactions: weeither laugh at the inaccuracy of the author’s hypo-theses, or we are taken aback by the author’s pres-cience.

The aim of this second edition of “Non-invasiveVentilation in Pediatrics” was to incorporate all ofthe knowledge accumulated in the past four years.Nonetheless, due to the limited body of literatureon NIV, there are many aspects of this treatment

that are dictated primarily by experience and byeach author’s opinion.

We hope that the colleague that reads this textnow, does so with a critical eye, remembering thathistory repeats itself and that, consequently, manyof the opinions published in this text will be scien-tifically refuted in the future. Nonetheless, with allhumility, we hope that some of the hypotheses pre-sented here will ultimately be confirmed.

Prologue

AF: Assisted flowARDS: Acute respiratory difficulty syndromeARF: Acute respiratory failureAV: assisted volumeAVAPS: Average volume assured pressure

supportBiPAP: Bi-level positive airway pressureCCHS: Congenital central hypoventilation

syndromeCm: centimetersCNS: Central nervous systemCOPD: Chronic obstructive pulmonary

diseaseCPAP: Continuous positive airway pressureCRF: Chronic respiratory failureE: ElasticityEPAP: Expiratory positive airway pressureF: FlowFig: FigureFiO2: Fraction of inspired oxygenHMV: Home mechanical ventilationHz: HertzIFD: Infant Flow DrivIFDA: Infant Flow Drive AdvanceIPAP: Inspiratory positive airway pressureIPPV: intermittent positive pressure ventila-tionIV: IntravenousFRC: Functional residual capacityFVC: Functional vital capacityL: LitersLPM: Liters per minuteMI-E: Mechanical insufflation-exsufflationMin: minutes (s)n-CPAP: Nasal continuous positive airway

pressuren-IPPV: Nasal intermittent positive pressure

ventilationNIV: Non-invasive ventilationn-SIPPV: Nasal synchronized intermittent

positive pressure ventilationNMD: Neuromuscular diseaseNNPV: Non-invasive negative pressure

ventilationNRDS: Neonatal respiratory difficulty

syndromeOSAS: Obstructive sleep apnea syndromePaO2: Arterial oxygen pressurePaCO2: Arterial carbond dioxide pressurePACV: Pressure-assisted/controlled

ventilationPAV: Proportional assisted ventilationPCV: Pressure-controlled ventilationPEEP: Positive-end expiratory pressurePEP: positive expiratory pressurePICU: Pediatric intensive care unitPIP: Peak inspiratory pressurePmusc: Respiratory muscle pressurePP: Positive pressurePS: Pressure supportR: ResistanceRF: Respiratory frequencyRPM: Respirations per minuteS: Spontaneous modeSaO2: Trancutaneous oxygen saturationSec: Second(s)SMA: Spinal muscular atrophyS/T: Spontaneous/timed modeT: Timed modeTi: Inspiratory timeV: Volume of air that enters the lungsVCV: Volume-controlled ventilationV/Q: Ratio of ventilation to perfusionVt: Tidal volume

Abbreviations

1. History and clinical use of non-invasive ventilation 1C. Rey, M. Pons, J. Mayordomo and M. Los Arcos

2. Physiology of respiratory dynamics. Application to non-invasive ventilation 7V. Modesto, S. Vidal and E. Ibiza

3. Indications and contra-indications of non-invasive ventilation in patients with acute respiratory failure 11M. Gáboli and M. Pons

4. Conditions and settings for delivering non-invasive ventilation 17M. Gáboli, J. Mayordomo, P. Gómez de Quero and M. Pons

5. Non-invasive ventilation interfaces 25A. Concha, A. Medina, M. Pons and F. Martinón-Torres

6. Non-invasive ventilation devices 37S. Menéndez, A. Medina, F. Martinón-Torres and C. Rey

7. Non-invasive ventilation modes and methods for pediatric patients 47J. López-Herce and J. Pilar

8. Non-invasive ventilation methodology for acute pediatric pathologies 59M. Pons, T. Gili and A. Medina

9. Caring for non-invasive ventilation patients 65J. García-Maribona, M. González, J.M. Blanco and J.C. Monroy†

10. Monitoring of non-invasive ventilation for pediatric patients 75M.A. García, R. Porto and P. Fernández

11. Complications and technical difficulties in non-invasive ventilation 81M. Pons and M. Balaguer

12. Sedation in non-invasive ventilation 87M. Los Arcos, J. Mayordomo and M. Gáboli

13. Failure analysis and predictive factors in non-invasive ventilation 91M. Pons, A. Medina, F. Martinón-Torres and J. Mayordomo

Contents

14. Non-invasive ventilation with heliox 99F. Martinón-Torres

15. Humidification and bronchodilator therapy in non-invasive ventilation 107A. Esquinas and M. Pons

16. Fiberoptic bronchoscopy in non-invasive ventilation 115M.I. Barrio, M.C. Antelo and M.C. Martínez

17. Neonatal non-invasive ventilation 119F. Castillo, D. Elorza, M.L. Franco, J. Fernández, M. Gresa, A. Gutiérrez, I. López de Heredia, J. Miracle, J. Moreno, A. Losada and A. Valls Soler Grupo Respiratorio Sociedad Española de Neonatología

18. Indications of non-invasive ventilation in chronic pediatric pathologies 125M.H. Estêvão and M.J. dos Santos

19. Methodology for pediatric non-invasive ventilation in the home 137M.Á. García Teresa, T. Gavela and M.J. Pérez

20. Respiratory physiotherapy in non-invasive ventilation 147M.R. Gonçalves

21. Healthcare assistance, quality control and clinical training for non-invasive ventilation of acute pediatric patients 157T. Gili, A. Medina and M. Pons

22. Ethical aspects of non-invasive ventilation 163F.J. Cambra, M.H. Estêvão and A. Rodríguez Núñez

23. Summary and algorithms 169A. Medina, M. Pons and F. Martinón-Torres

Index 179

Translated from the Spanish by Gregory Qushair for SciLingua, except for Chapter 20, ”Respiratoryphysiotherapy in Non-Invasive Ventilation”, which was originally written in English by M.R. Gonçalves.

INTRODUCTIONNon-invasive ventilation (NIV) can be defined

as ventilation which does not require artificial entry—via tracheotomy or endotracheal intubation—beyondthe vocal chords of the patient’s respiratory track.Various mechanisms have been used to deliver NIV,and different eras have witnessed the predominanceof certain devices and techniques over others. Galenowas the first to describe experimental ventilation,observing how inflation of a dead animal with airvia the larynx caused its bronchi to fill and its lungsto expand. Vesalio is believed to have later performedthe first ever artificial ventilation during the autopsyof a Spanish nobleman whose lungs he inflated withair.

This chapter provides a historical summary ofnegative pressure ventilation, ventilation usingrocking beds and pneumatic belts, and ventilationvia nasal masks and face masks. Current clinical useof NIV in Spain is then reviewed.

NEGATIVE PRESSURE VENTILATIONThe first techniques used for mechanical

ventilation were based on non-invasive systems thatgenerated negative pressure over the chest wall. Thefirst “tank” negative pressure respirator wasdeveloped by Dalziel in Scotland in 1832. Jonespatented an apparatus similar to Dalziel’s, inKentucky in 1864, recommending its use for thetreatment of asthma, bronchitis, paralysis, neuralgias,

rheumatism, seminal weakness and dyspepsia. Figure1 shows a negative pressure respirator presented byWoillez at the French Academy of Medicine in 1876.The patient was placed into a cylindrical containerwith their head sticking out. An adjustable elasticrubber collar was used around the patient's neck toobtain a perfect seal. Manual expansion of the pumpcreated a sub-atmospheric pressure in the tank,whereas pressure on the pump caused the pressureof the tank to return to atmospheric level. Negativeextrathoracic pressure was transmitted via the chestwall, and analogously to the action of negativepleural pressure during spontaneous respiration, thesystem generated a flow of air from the mouth tothe lung.

Substitution of an electric source for manualpressure, and a leather diaphragm for the pump,provided so-called “tank” respirators, or “iron lungs”,whose use peaked during the polio epidemic of the1930's to the 1950's. The first negative pressurerespirator to be widely used in the clinic was theDrinker-Shaw iron lung, developed in 1928. A majorpolio epidemic in 1931 led Emerson to manufacturea smaller and simpler respirator. This apparatus wascheaper and easier to operate than its predecessorsand boasted several technological improvements;hence, it was widely used and enabled several livesto be saved during epidemics of respiratory paralysiscaused by polio.

From the 1960’s to the 1970’s, a negativepressure ventilator similar to iron lungs was used to

History and clinical use of non-invasive ventilation

C. Rey, M. Pons, J. Mayordomo and M. Los Arcos

Chapter 1

treat hyaline membrane disease in neonates. In thisrespirator, the neonate’s head was placed inside ofa chamber into which humidified oxygen could beadded, while the rest of its body was maintained ina hermetically sealed chamber to which negativepressure was applied.

The aforementioned systems had numerouslimitations: they easily led to obstructions of theupper airway through closure of the glottis, interferedwith physical examination of the patient, causedskeletomuscular disorders, and were burdened bytheir large size, among other factors. Due to theseproblems, use of these respirators was eventuallyabandoned, as new forms of mechanical ventilationwere sought.

ROCKING BEDS AND PNEUMATIC BELTSRocking beds and pneumatic belts, or

pneumobelts, are NIV tools that were developedand that reached their peak use during the finalperiod of the polio epidemic. Both are based onexploiting gravity to help the movements of thediaphragm and were especially utile for patients witha paralytic or very weak diaphragm.

In the early 1930’s Eve described the use ofmanual rocking to favor ventilation in two patientswith acute respiratory paralysis. The techniqueconsisted of positioning the patient in a bed thatrocked 45° up and down. The change in weight ofthe abdominal viscera moved the diaphragm up anddown alternately, creating a new method of artificialrespiration. The technique was later accepted by the

British Navy as a recommended way of revivingvictims of near-drowning. Subsequent studiesdemonstrated the utility of this type of artificialrespiration as compared to other methods. Indeed,it remained in the revival protocols until the 1960’s,when mouth to mouth resuscitation took over asthe initial form of artificial respiration. Automaticrocking beds were introduced in the 1940's as a typeof ventilatory aid. In 1950, the American Council onPhysical Medicine and Rehabilitation accepted arocking bed designed by McKesson as a tool forweaning polio patients off of iron lungs. This bedfacilitated nursing care and provided the patientwith more freedom; however, it was very noisy andheavy. The Emerson rocking bed, used extensivelyduring the 1950’s and 1960’s, was quieter andlighter. Nonetheless, control of the polio epidemicthrough vaccination led to a drastic fall in demandfor these beds. It is estimated that in the UnitedStates only 80 rocking beds are still in use; they areemployed for polio survivors. The majority of patientshave sought other forms of ventilation assistance.

Pneumobelts deliver intermittent abdominalpressure by assisting diaphragmatic movements.They also arose in the final period of the polioepidemic, and were designed to overcome thelimitations of the ventilators available at that time.Pneumobelts enabled total freedom of the upperextremities and the mouth, making it easier for thepatient to remain seated. A diurnal ventilator wasfor patients in wheelchairs. As in the case of rockingbeds, once the polio epidemic had come undercontrol, pneumobelts lost their primary indication.Albeit several modifications were eventually madeto the original mechanism, including a combinationof pneumatic belt and intermittent positive pressureventilation, pneumobelts are now very rarely used.

NASAL MASKS AND FACE MASKSThe first non-invasive positive pressure ventilation

(NPPV) systems were used at the beginning of the20th century by surgeons to perform operationsthat implied opening of the thorax. Brauer isconsidered the first surgeon to have used a positivepressure system, which consisted of a small cabininto which the patient’s head was introduced. Asalternatives to these small cabins, other surgeonsdesigned face masks and hermetically sealed helmets

2 C. Rey et al.

Figure 1. Manually operated negative pressure ventilator.

to generate positive pressure in the airway. However,with the advent of translaryngeal intubationtechniques, these NPPV systems were abandoned.

Outside of surgery, the limitations of negativepressure NIV systems, coupled with the technologicaladvances made during WWII, led to standard use ofpositive pressure mechanical ventilation viaendotracheal or tracheotomy tubes. In 1907 thecompany Dräger developed one of the first NIVventilators, the Pulmotor (Fig 2.), which was usedfor resuscitation. In 1935 Barach reported the useof a respirator that provided continuous positiveairway pressure (CPAP) via a mask for patientssuffering from various forms of acute respiratoryfailure (ARF). Nonetheless, it was not until the 1970’sthat NVPP began to emerge as a slightly lessaggressive alternative, simultaneously circumventingthe complications associated with endotracheal tubeswhile improving the quality of life of the patient—namely, by preserving the defense mechanisms ofthe airway and enabling patients to speak andswallow. This type of ventilation assistance employedintermittent positive pressure via a mouthpiece. Initialexperiences with this method appeared favorable:it had been reported that NIV could controlhypercapnia in patients with acute respiratory failure.However, in subsequent studies, including a multi-center study by the US National Institutes of Health(NIH), it was found that NIV combined withnebulizer therapy for patients with chronicobstructive pulmonary disease did not provide anyadded advantage compared to the nebulizer therapyalone.

In the 1980’s continuous positive airway pressurevia the nasal passage (nasal CPAP, or n-CPAP) wasfirst used for patients with sleep apnea and affordedgood results. Intermittent positive pressure ventilationvia the nasal passage (nasal IPPV, or n-IPPV) was soonemployed, improving ventilation in patients withchronic respiratory failure (CRF), especially duringsleep. Multiple studies on NVPP were subsequentlyperformed in patients with ARF or CRF, sleep apnea,or respiratory difficulty after extubation, withincreasingly positive results obtained.

CLINICAL USE OF NIV (Spain)Use of NVPP has undergone surged in the past

5 years. In 2002 a survey in Spain that was

performed using an electronic hospital mailing list(UCIP-net) revealed that only four pediatric intensivecare units (PICUs) were equipped with NIV-specificventilators and had written NIV protocols, andtherefore, performed NIV for their patients. However,in the past 5 years, there has been an exponentialincrease in the use of NIV in pediatrics; hence, alarge percentage of Spanish PICUs now perform thistype of ventilation assistance.

From October 2004 to February 2004, anepidemiological study on pediatric NIV wasperformed by eight PICUs in Spain. The studyencompassed 109 cases of NIV use in 104 patients(mean age: 12 months) in which the overall efficacywas 78%. In the PICU of Hospital UniversitarioCentral de Asturias (HUCA), which has seven bedsand averages approximately 300 annual admissions,during the past 3 years there have been 119 casesof NIV use in 93 patients (mean age: 3 years; agerange: 1 month to 16 years). NIV was indicatedfor type I respiratory failure (RF) in 25% of the cases,for type II RF in 50%, and for post-extubation RF inthe remaining 25%. NIV-specific ventilators wereused in 95% of the cases. The primary interface usedwas oral-nasal (74%); in one case, a full face maskwas employed. The average duration of the NIV was35 hours. Favorable progress was obtained:mechanical ventilation was not required in 80% ofthe cases. Failure of NIV was greater in post-extubation RF (36%) and type I RF (27%) than intype II RF (8%). There were no complications in 80%of the cases; the complications in the remaining 20%were principally minor: self-limited skin lesions (16%),

3History and clinical use of non-invasive ventilation

Figure 2. The Pulmotor.

pneumothorax (3%) and upper airway bleeding(1%).

Figure 3 shows an important trend: the changein the use of NIV compared to conventionalmechanical ventilation (CMV) in the PICU at HUCA.As observed in the Figure, use of NIV has risen whileuse of CMV has steadily decreased, leading theformer to surpass the latter for the first time ever in2007.

Figure 4 reveals a similar trend at the PICU ofHospital Universitario Sant Joan de Déu de Barcelona,based on data collected from 287 cases between2002 and 2006. Note that there is a major differencein the mean age of the two treatment groups (NIVgroup: 7.6 years; CMV group: 2.5 years). Moreover,the NIV group exhibited a lower rate of nosocomialinfection as well as a shorter mean hospital stay.

In the past decade NIV has been the subject ofover 1,500 scientific articles, including 14 meta-analyses. There is growing interest in pediatric NIV,as demonstrated at the 5th World Congress onPediatric Critical Care, held in June 2007, whichfeatured a roundtable discussion and fifteencommunications on NIV.

In summary, NIV has represented a majoradvance in the treatment of ARF, and its utility hasbeen well established in several clinical researchpublications. When properly applied, NIV is a non-aggressive respiratory support method with few sideeffects. Nonetheless, it has yet to be systematically

employed for all children with ARF. Several moreyears will be needed until all patients for whom NIVwould be useful will be able to receive it.

REFERENCES

1. Drinker P, Shaw LA. An apparatus for the prolonged adminis-tration of artificial respiration: I. Design for adults and children.J Clin Invest 1929;7:229-47.

2. Crone NL. The treatment of acute poliomyelitis with the respi-rator. N Engl J Med 1934;210:621-3.

3. Colice GL. Historical perspective on the development of mecha-nical ventilation. En: Tobin MJ Ed. Principles and practice ofmechanical ventilation. McGraw-Hill, INC. New York 1994;1-36.

4. Levine S, Levy S, Henson D. Negative pressure ventilation. CritCare Clin 1990;3:505-31.

5. Eve FC. Actuation of the inert diaphragm. Lancet 1932;2:995-7.

6. Plum F, Whendon GD. The rapid-rocking bed: its effect on theventilation of poliomyelitis patients with respiratory paralysis. NEngl J Med 1951;245:235-40.

7. Hill NS. Use of the rocking bed, pneumobelt, and other non-invasive aids to ventilation. En: Tobin MJ Ed. Principles and prac-tice of mechanical ventilation. McGraw-Hill, INC. New York1994;413-25.

8. Carter HA. McKesson respiraid rocking bed accepted. JAMA1950;144:1181.

9. The Intermittent Positive Pressure Breathing Trial Group. Inter-mittent positive pressure breathing therapy of chronic obstruc-tive pulmonary disease. Ann Intern Med 1983;99:610-20.

10. DiMarco AF, Connors AF, Altose MD. Management of chronicalveolar hipoventilation with nasal positive pressure breathing.Chest 1987;92:952-4.

4 C. Rey et al.

Figure 3. Change in use of conventional mechanical ventila-tion and non-invasive ventilation in the Pediatric Intensive CareUnit of Hospital Universitario Central de Asturias, 2000 to 2007.Data are presented as percentages relative to the total num-ber of admitted patients. CMV: conventional mechanical ven-tilation; NIV: non-invasive ventilation

2000 2002 2004 2006

35

30

25

20

15

10

5

0

CMVNIV

Year

2001 2003 2005 2007Perc

enta

ge o

f tre

ated

pat

ient

s re

lativ

eto

tot

al n

umbe

r of

pat

ient

s

Figure 4. Change in use of non-invasive ventilation in thePediatric Intensive Care Unit of Hospital Universitario SantJoan de Déu de Barcelona, 2000 to 2008.NIV: non-invasiveventilation; post-extub: post-extubation.

NIVPost-extub. NIVTotal NIV

Year

Num

ber

of p

atie

nts

2002

1009080706050403020100

2003 2004 2005 2006 2007 2008

11. Belman MJ, Soo Hoo GW, Kuei JH, Shadmehr R. Efficacy of posi-tive vs negative pressure ventilation in unloading the respiratorymuscles. Chest 1990;98:850-6.

12. Medina A, Prieto S, Los Arcos M, Rey C, Concha A, MenéndezS, et al. Aplicación de ventilación no invasiva en una unidad decuidados intensivos pediátricos. An Esp Pediatr 2005;62:13-9.

13. Corrales E, Pons M, López-Herce J, Martinón-Torres F, García MA,Medina A, González Y. Estudio epidemiológico de la ventilaciónno invasiva en las Ucip de España. XXII Congreso Nacional de laSECIP, 2005. An Pediatr (Barc) 2007;67:91-100.

14. Pons M, Mayordomo J, Barón Ruiz I, Palomeque Rico A. Com-

parative analysis between conventional mechanical ventilationand non invasive ventilation in 4 years in our picu. World pedia-tric intensive care medicine congress. Geneve 2007. Pediatr CritCare Med 2007;8(3, Suppl):A268.

15. Hess DR, Fessler HE. Respiratory controversies in the critical care set-ting. Should noninvasive positive-pressure ventilation be used in allforms of acute respiratory failure? Respir Care 2007;52:568-78.

16. Burns KE, Adhikari NK, Meade MO. Noninvasive positive pres-sure ventilation as a weaning strategy for intubated adults withrespiratory failure. Cochrane Database Syst Rev 2003;(4):CD004127.

5History and clinical use of non-invasive ventilation

INTRODUCTION Unfortunately, physicians do not have the

opportunity to administer non-invasive ventilation(NIV) until nearly a decade after having studiedphysiology. Hence, the authors of this chapterthought it would be apropos to review the basics ofhow humans breathe, how the body adjusts tosituations of respiratory failure (RF), and how NIVcan reverse RF.

RESPIRATORY PHYSIOLOGYIn order to deliver air to the lungs during

inspiration, the respiratory musculature must createsufficient pressure to overcome two forces: theincrease in the recoil pressure of the lung once itis filled, which is a static force, and the frictionassociated with airway flow, which is a dynamicforce. During expiration, air exits the lungspassively.

To enable better understanding of how NIV withpressure support (PS) functions, this chapter firstreviews the static properties of respiratory mechanics,and then reviews the dynamic properties: how theflow of air is produced from the atmosphere towardsthe alveoli during inspiration and in the oppositedirection during expiration.

Static properties of respiratory mechanicsThe static properties of respiratory mechanics

are studied in conditions of airflow equal to 0 L/s

(liters per second). The pressure inside of the mouthcorresponds to the atmospheric pressure (Patm),which at sea level is equal to 760 mm Hg; however,for pulmonary physiology calculations, its value istaken as 0 cm H2O (the reference pressure). Theinterior pressure in the alveoli (Palv) and the interiorpressure in the pleural space (Ppl) are static pressuresand must be measured in the absence of airflow.The static difference between these two pressuresis defined here as the transpulmonary pressure (PTP),or pressure of pulmonary retraction, such that PTP =Palv - Ppl. The static difference between the interiorpressure in the pleural space and the atmosphericpressure is defined here as the elastic recoil pressureof the chest wall (PW), such that PW = Ppl - Patm. Thesum of PTP and PW is called the total recoil pressureof the respiratory system.

Within the scope of this chapter, there is onlyone main concept of interest from static respiratorymechanics: the volume of the lung only changeswhen the magnitude of PTP changes. Though itseems counterintuitive, it is not the value of Palv thatcauses the volume of lung to change, but rather thevalue of PTP. As the lung fills with air each value ofpulmonary volume corresponds to a specific valueof PTP. Regardless of the values of Palv and Ppl, at a PTP

value of +5 cm H20, the lung will fill with air to avolume equal to its functional residual capacity (FRC);at a PTP value of + 30 cm H20, the lung will be attotal lung capacity (TLC); and at a PTP value of +3cm H20, the lung will be at its residual volume (RV).

Physiology of respiratory dynamics.Application to non-invasive ventilation

V. Modesto, S. Vidal and E. Ibiza

Chapter 2

For greater detail on this concept, the reader isreferred to other sources(1).

Respiratory dynamicsTo enable a more detailed study of the dynamic

behavior of the respiratory system, two types ofdiagrams are used in this chapter (shown in Figures1 to 7 below). The first type shows a graduatedvertical bar which reflects the pressure values of thesystem (in cm H2O), whereby positive valuescorrespond to super-atmospheric pressure, andnegative values, to sub-atmospheric pressure. Thesecond type shows volume plotted against time; inthis plot, a point on the curve shows the position ofthe respiratory cycle with respect to time for eachof the phases.

Analogously to fluid dynamics, respiratorydynamics is governed by the general principle thatair always moves along the pressure gradient. If thereis no gradient between the mouth and the alveoli,then there is no flow of air. If the Palv is lower thanthe Patm, then air enters the alveoli, whereas if Palv

is higher than Patm, then air exits the alveoli. OncePalv and Patm balance out, the flow of air stops.

Before an inspiration, the respiratory musculature(i.e. the diaphragm and supporting muscles) is atrest (Fig. 1). The natural tendency of the chest wallto retract leads to a sub-atmospheric pressure in thepleural space (Ppl) of ca. -5 cm H2O; however, at thispoint Palv = Patm = 0 cm H2O, so there is no flow ofair. Hence, the transpulmonary pressure (PTP) = 0 cmH2O - (- 5 cm H2O) = + 5 cm H2O, which drives thelung volume to FRC, the volume of the respiratorysystem at rest.

Inspiration begins with contraction of theinspiratory muscles immediately before air entersthe body (Fig. 2). This leads to a decrease in pressurein the pleural space (Ppl). At this initial instant, no airhas yet entered the body: the volume of the lunghas not changed, meaning that, as explained above,the PTP initially remains constant. This can onlyhappen if, immediately before air begins to enter,the drop in Ppl causes a drop in Palv of equalmagnitude. It is this decrease in Palv at levels lowerthan Patm that establishes the gradient which causesair to enter.

The decrease in Palv generates a flow of air intothe system, causing the lungs to fill with air and

8 V. Modesto, S. Vidal , E. Ibiza

Figure 1. Phase of apnea before inspiration. PTP: transpul-monary pressure; PW: elastic recoil pressure of the chest wall.

Figure 2. Initial phase of inspiration: contraction of the diaph-ragm. PTP: transpulmonary pressure; PW: elastic recoil pressu-re of the chest wall.

Figure 3 A and B. Final phase of inspiration. A) The drop inalveolar pressure begins the flow of air inwards. The lunggains volume, which causes the transpulmonary pressure (PTP)to increase until B), the moment of maximum muscular con-traction. The maximum PTP coincides with the maximum lungvolume, but Palv balances out with the atmospheric pressu-re, so no more air enters.

Super-atmospheric pressure

Atmospheric pressure(cm H2O)

Sub-atmospheric pressure

Inspiration

Alveolar

Intrapleural

Time

PTP

PW

Expiration

Volu

me




Inspiration

Alveolar

Intrapleural

Time

PTP

PW

Expiration

Volu

me

Inspiration

Time

Expiration




Volu

me

Inspiration

Alveolar

Intrapleural

Alveolar

Intrapleural

Time

PTPPTP

Expiration

Volu

me

consequently, increase in volume (Fig. 3A). Thechange in lung volume leads the PTP to graduallyincrease: contraction of the muscles causes a furtherdrop in Ppl, but, as air enters the alveoli, the decreasein Palv is not substantial. This situation remains steadyduring the entire inspiration until, in the final phase,muscular contraction peaks, pulmonary volume andPTP reach their maximum values, and Palv and Patm

balance out, causing the flow of air to stop andproducing the inspiratory pause.

After the inspiratory pause there is a brief totalstop in respiratory muscle contraction (Fig. 4). Thelung remains at maximum volume, meaning thatthe value of PTP is equal to that during the inspiratorypause (i.e. no volume has yet been lost); however,the stop in contraction causes the value of Ppl toreach zero. At this point, PTP is entirely the result ofPalv, which reaches its maximum value due to thetendency of lung tissue to passively retract. Theincrease in Palv to levels above Patm establishes thegradient which causes air to exit (Fig. 5).

Once all of the air has left (Fig. 6), the flow stops:Palv and Patm balance out to 0 cm H2O, Ppl returns toits rest level of -5 cm H2O, PTP reaches 5 cm H2O,and lung volume is equal to FRC. The process thenbegins again.

The physiological basis of pressure support If upon inspiration the patient is unable to

generate sufficient force to contract the muscles inorder for Ppl to reach a given value (e.g. < -7 cm

H2O), then the situation shown in Figure 7A arises.As Palv is equal to Patm, the maximum PTP will be +7cm H2O, and the lung will barely be able to fill withair. This is the clinical scenario of a patient enteringinto respiratory failure. If at this point, and only forthe duration of the inspiration, the airway can bequickly pressurized to enable Palv to rise up to a pre-established value (known as pressure support, orSP), then with the same level of muscle effort (Ppl

= –7 cm H2O) a much higher value of PTP can bereached, and the lung can reach a much greaterinspiratory volume (Fig. 7B). When the inspirationstops, the airway depressurizes just as in spontaneousventilation: the stop in muscular contraction drives

9Physiology of respiratory dynamics. Application to non-invasive ventilation

Figure 4. Abrupt stop in muscular contraction. The volumeand the transpulmonary pressure remain constant, but thealveolar pressure reaches its maximum value. PTP: transpul-monary pressure.

Figure 5. Exiting of air during passive expiration. PTP: trans-pulmonary pressure.

Figure 6. Pre-inspiratory apnea.

Atmospheric pressure (cm H2O)







Inspiration

Time

PTP

PTPAlveolar

Intrapleural

Alveolar

Intrapleural

PTP

Expiration

Volu

me


Inspiration

Time

Expiration

Volu

me


Inspiration

Time

ExpirationVo

lum

e

Alveolar

Intrapleural

Alveolar

Intrapleural

the value of Ppl to zero. Again, at this point PTP isgoverned by Palv, which reaches its maximum valuedue to the tendency of lung tissue to passivelyretract. As explained above, the increase in Palv tovalues above Patm generates the gradient whichenables air to exit the body.

Ventilation with pressure support requires a wayof knowing when the inspiration begins (i.e. theinspiratory trigger), establishment of a pressure valueat which to pressurize the airway during inspirationonly, and a way of knowing when the inspiration ends(i.e. the expiratory trigger). Hence, use of pressuresupport, in what is known as spontaneous/timed (S/T)mode in NIV-specific ventilators, is dictated by thefollowing key factors: • Inspiratory synchrony• Expiratory synchrony• Rapid compensation for leaks in the system, in

order to reach pre-established pressure valuesduring inspiration. Obtaining good resultsdepends on the specific characteristics of eachventilator, especially for patients with highphysiological requirements (e.g. high respiratoryfrequency, short inspiratory time, and weak peakinspiratory flow), including infants, prematureneonates, and patients suffering fromneuromuscular diseases.

REFERENCES

1. Ball WC (1996) Interactive Respiratory Physiology. Johns Hop-kins School of Medicine. http://oac.med.jhmi. edu/res_phys/

10 V. Modesto, S. Vidal , E. Ibiza

Figure 7 A and B. A) Respiratory failure: the patient is notcapable of generating transpulmonary pressure (PTP) greaterthan + 7 cm H2O. B) Use of pressure support (PS): with thesame effort, the patient reaches a higher PTP, and the lungfills to a greater volume.

Inspiration

Time

Expiration




Volu

me

Alveolar

Intrapleural

PTP

Alveolar

Intrapleural

PTP

PS

INDICATIONS

Non-invasive ventilation in acute respiratoryfailure

The indications of non-invasive ventilation (NIV)in acute respiratory failure (ARF) are beingincreasingly better defined; consequently, it isbecoming safer and more successful. As with othermedical procedures, NIV was first used in adultpatients, and then based on these experiences, itwas gradually extended to specific areas inpediatrics. The rise in NIV use stems from a desireto avoid the complications inherent to invasiveventilation as well as an aim to better use resources.NIV can reduce hospital stays and the costs ofhospital admissions and improve patient comfort.However, careful selection of candidate patients,availability of adequate material, and physician andnursing care continue to be the determinant factorsfor successful treatment of ARF with NIV—muchmore so than in the case of chronic respiratoryfailure (CRF).

The objectives for NIV in ARF comprise:ameliorating patient symptoms, reducing theworkload of respiratory muscles, and improving gasexchange in patients who do not require intubationand conventional mechanical ventilation (CMV);however, NIV is never intended for substituting thesetechniques if they are clearly required. NIV ismaintained while the patient is treated for the cause(e.g. infection or surgery) of respiratory failure (RF).

When selecting an ARF patient for NIV treatment,the following factors should be considered:a. Clinical criteria: symptoms and signs of respiratory

difficulty (e.g. moderate or severe dyspnea, highrespiratory frequency, use of accessory respiratorymuscles, and paradoxical breathing).

b. Blood gas criteria: PaCO2 > 45 mm Hg and pH <7.35; or (PaO2/FiO2) < 250; or, if arterial bloodgas test is not available, the ratio of peripheralsaturation (SatO2) to FiO2 < 320, when SatO2

< 98%, which is the criterion used in the PALIVE(Pediatric Acute Lung Injury MechanicalVentilation) multi-center study on acuterespiratory distress syndrome (ARDS), which usedcriteria from an article on adult patients.

c. The cause of ARF.d. Lack of contra-indications (see separate section

below). In the case of any doubts, a one-hourtest treatment of NIV could be administered, inwhich the patient is closely monitored in ordernot to delay intubation (if required).The following factors are predictors of success

for use of NIV in ARF:a. The process or condition triggering the ARF is

relatively mild.b. A high level of cooperation from the patient and

a high level of patient-ventilator coordination.c. Optimal adaptation of the interface to the patient

(i.e. in size and shape).d. Improved gas exchange, heart rate and respiratory

frequency in the first two hours of NIV.

Indications and contra-indications ofnon-invasive ventilation in patients

with acute respiratory failure

M. Gáboli and M. Pons

Chapter 3

e. A moderate level of initial hypercapnia (PaCO2

between 45 and 90 mm Hg).f. A moderate level of respiratory acidosis (pH

between 7.20 and 7.35).g. Acute respiratory distress syndrome (ARDS) in

which (PaO2/FiO2) > 150. The clearest sign ofsuccess of NIV is a drop in respiratory frequency.Contrariwise, agitation, worsening of respiratorydifficulties, and worsening of gas exchange areall indicative of failure of NIV. Hence, monitoringof vital signs and blood gases (when indicated)should be prioritized in the first few hours oftreatment (see Chapter 10).

Indications of NIV in the acute patient• During the course of neuromuscular diseases

(NMDs), affectation of respiratory muscles leadsto progressive chronic respiratory failure,although this still allows the patient to maintaina certain minimum quality of life. However, inthe presence of a precipitating factor—typicallya respiratory infection, convulsions, or medicalintervention, or an intercurrent illness whichlimits daily physical activity or demands greatermuscle effort—ARF arises, primarily via alveolarhypoventilation. In a recent study, NIV wasshown to be a safe and effective first line oftreatment for infants and children suffering fromNMD. The authors of the study emphasize theimportance of recognizing patients who canmaintain sufficient respiration without use of aventilator, whom they consider as idealcandidates for NIV, given that conventionalmechanical ventilation (CMV) via endotrachealintubation with sedation and analgesia wouldfurther weaken respiratory muscles, complicateweaning and often imply definitive tracheotomy.They even propose that use of conscious sedationto facilitate coupling of the NIV interface to thepatient should be considered before ruling outNIV.

• Analogously to the case of NMDs, diseases ofthe chest wall and of the spinal cord imply arestrictive form of respiratory failure and a highpropensity to atelectasis. In these cases, NIV isnot only utile during flare-ups, but also duringthe perioperative period of surgery to correctthoracic defects, in which it facilitates weaningfrom mechanical ventilation.

• Indication of NIV to facilitate extubation andprevent reintubation in adult patients is currentlyaccepted by the majority of authors. Patientswho could most benefit from use of NIV afterextubation or even after decannulation are thosewith chronic lung disease. Hence, the typicalpediatric patient who could benefit from NIVwould be a preterm infant who has undergonetracheal intubation for administration ofsurfactant, for whom extubation may be advisedto diminish any side effects of mechanicalventilation on their immature lungs.

• In patients with acute airway obstruction, suchas that which occurs in laryngitis and in certainpharyngeal-tonsillar infections, NIV can markedlyreduce the muscle effort required for breathingand improve gas exchange. It must be notedthat use of NIV here is not free of risks and shouldbe performed in the presence of experts inmanaging complicated airways, with adequatematerial available for establishing a definitiveairway and only if the patient maintains theirphysiological reflexes for airway protection.

• For ARF related to bronchiolitis, NIV can beindicated at different moments and for differentreasons. For the youngest patients, who presentwith apneas, CPAP tends to be indicated as inthe case of apneas in preterm infants. Patientstend to respond favorably, except in the case ofsevere apneas and/or if the infection developsinto pneumonia. NIV then takes on the sameindications and limitations described for severecases of hypoxemia. For patients with obstructiverespiratory failure, with an increase in respiratorymuscle effort as well as secondary retention ofCO2, NIV with two levels of pressure (bi-levelpositive pressure airway ventilation, or BiPAP)tends to be indicated, namely for improving gasexchange and reducing respiratory effort.Nonetheless, a recent random study whichdemonstrated that use of CPAP alone wassufficient to improve hypercapnia warrantsmention. The benefits of NIV employing mixturesof helium and oxygen (which are less dense thanair) are also noteworthy; these are detailed inChapter 14. The greatest challenge in this typeof pathology resides in the fact that adequatematerial is often difficult to find for these patients(typically, infants younger than 3 months).

12 M. Gáboli, M. Pons

• Indication of NIV in pediatric patients with severehypoxemia due to asthma remains highlycontroversial. The complications implied in CMVfor this pathology have led to the search foralternate ventilatory strategies. NIV has beenshown to improve oxygenation in patients duringsevere asthmatic attacks. NIV treatment ofchildren with this pathology has scarcely beenstudied, and the few existing reports deal withsmall patient populations: 33 (Mayordomo-Colunga), five (Carroll) and three (Akingbola).Nonetheless, in the study by Mayordomo-Colunga, NIV only failed in one case. This isaccomplished through a minor adjustment ofthe V/Q mismatch, whereby using a PEEP whichdoes not exceed 80 to 90% of the auto-PEEP incombination with an inspiratory positive airwaypressure (IPAP) or a pressure support (PS) lowersthe level of respiratory effort required to maintainan adequate flow volume for gas exchange. Thegreatest challenge in clinical practice is to ensurethat the patient couples well to the ventilator,without any agitation; this often requires lightsedation (see Chapter 12). However, in respiratoryfailure with hypercapnia due to an asthmaticattack, NIV has not proven superior to CMV, nordoes it lead to a lower number of requiredintubations. Early administration of NIV (i.e.before total respiratory failure), and use ofventilators that are very sensitive to the respiratoryefforts of the patient, and that consequentlyenable good synchronization, may be importantfactors in the success of the treatment.

• Indication of NIV in severe hypoxemic respiratoryfailure (i.e. [PaO2/FiO2] < 300), such as that whichoccurs in ARDS or in acute pulmonary lesions(ALI), is also the subject of intense debate. Themajority of studies on adult patients do notdifferentiate among the different causes of ARDS(i.e. whether it is primarily pulmonary orsystemic), nor do they distinguish between ALIof infectious or traumatic origins. Currently, themost widely accepted conclusion is that NIV cannot be recommended universally for ARDSpatients ([PaO2/FiO2] < 150); rather, it shouldbe reserved for hemodynamically stable patientswithout metabolic acidosis, and should only beadministered in an ICU by staff with NIVexpertise. In a recent study of adults, the authors

observed that up to 54% of patients with ARDSthat had initially been treated with NIV did notrequire intubation. They also reported that a(PaO2/FiO2) value > 175 after one hour of NIVis indicative that the treatment will be successfuland stated that for patients with (PaO2/FiO2) <175, NIV must be considered even if all theintubation criteria are not strictly met. Patientswho benefitted from NIV had significantly lesscomplications, especially those of an infectiousnature, and required fewer days in the ICUcompared to patients treated from the beginningwith endotracheal intubation and mechanicalventilation. Few promising studies on severelyhypoxemic pediatric patients exist, and nouniversal guidelines have yet been establishedfor this pathology. It seems reasonable toextrapolate the findings from adult patients andthen proceed cautiously in the most severepediatric patients (according to the level ofseverity upon admission), without delayingintubation in those patients who, after one hourof treatment, do not exhibit any clinical or bloodgas improvement (see Chapters 13 and 23).

• Among situations of ARF in adult patients, use ofNIV is probably most well-documented for acutepulmonary edema (APE). NIV, in either CPAP orBiPAP modes, has been shown to rapidly improvegas exchange and lead to lower intubation andmortality rates and is relatively cost-effective. NIVin APE currently tends to begin in the ER and iseven used during extra-hospital transport. APEis much less frequent in pediatric patients thanin adult patients, owing to the rarity of cardiacischemia in the pediatric age. Cardiogenic APE(CAPE) may be the first sign of severe cardiacinsufficiency (e.g. myocarditis, myocardiopathyor a congenital deformation). In this case, if thereis any clear cause of hemodynamic instability,then NIV must be used cautiously, in the PICUand with strict monitoring. Contrariwise, if theAPE is due to hypervolemia and a periodicdecompensation of a known and controlledcardiac condition that arises in the post-operativeperiod of corrective cardiac surgery, then NIV,owing to its mildness, should be attempted asthe first treatment option in tandem with thespecific medical treatment. Among non-cardiogenic edema, a noteworthy type is that of

13Indications and contra-indications of non-invasive ventilation in patients with acute respiratory failure

APE due to negative intrathoracic pressure, whichis caused by a spontaneous forced respirationagainst an obstruction in the upper airway. Thisscenario has been reported in patients after ear,nose and throat (ENT) surgery. In these cases, inwhich the mechanism appears to be an increasein permeability due to a change in pressure inthe pulmonary arterioles and capillaries, if theairway is permeable and safe, then the responseto NIV tends to be very good and the patienttends to improve within a few hours. Lastly, if theedema is found within the context of an ALI (e.g.due to inhalation of toxic material, infection ortrauma), then the level to which the pulmonaryparenchyma has been affected and the grade ofthe hypoxemia should be evaluated as describedabove for ARDS.

• NIV has also proven effective for ARF in patientswho have undergone autologous bone marrowtransplant. The efficacy of NIV for this patientgroup is limited to certain etiologies such asAPE following hyperhydration before achemotherapy session. More severe etiologies(e.g. pulmonary hemorrhage) call for intubationand CMV. NIV seems to involve fewercomplications than CMV. NIV has likewiseproven utile in pediatric ARF patients affectedby oncological pathology.

• Albeit anecdotal, NIV has been used in one casein which intubation was impossible. Althoughthis application can not be considered a trueindication of NIV, it should nonetheless beconsidered in extreme situations.

• Finally, the palliative indication of NIV inoncology and neurology patients should not beforgotten. In these patients NIV can improve anintercurrent process or ease symptoms ofrespiratory difficulty.

CONTRA-INDICATIONSAll techniques, regardless of their novelty or

sophistication, have their respective limitations andcontra-indications. The main contra-indications ofNIV are described below (Table IV, chapter 23).

When protection of the airway is requiredFor situations in which ventilation is indicated

for protection of the airway (e.g. coma or active

digestive hemorrhage) NIV is absolutely contra-indicated, since, as with use of a laryngeal mask, itcan not guarantee this protection. The onlyexception to this rule is for patients with hypercapnicencephalopathy, who can derive neurologicalbenefits from a short (2 to 3 hours) test treatmentof NIV.

Severe respiratory failureContra-indication of NIV in severe RF is

supported by data on adult patients: a highermortality rate has been observed in patients thatreceived intubation after preliminary NIV treatment.However, patients for whom intubation is not a validoption due to the causative disease are an exception.As previously mentioned, in ARDS with (PaO2/FiO2)< 150 indicates a high risk of technical failure of NIV;hence, NIV should generally be considered contra-indicated for severe RF.

Fixed obstruction of the airwayNIV is contra-indicated in these situations, which

take too long to resolve for it to be effective.

Abundant and thick secretionsRestricted access to cleaning the airway

–especially in the case of an oral-nasal interface inpatients with limited coughing ability– is a high riskfactor for NIV failure.

VomitingAs with abundant secretions, the presence of

vomit makes maintaining a well-positioned interfacefor continuous administration of NIV nearlyimpossible.

Hemodynamic instability: shock For patients in this severe state, the concept of

energy conservation should be applied: respiratoryeffort is eliminated, and therefore, NIV should notbe used. In post-operative cardiac patients, thepresence of arrhythmias can often be considered acontra-indication for NIV.

Craniofacial malformations, trauma and burnsNIV is impossible to administer in patients with

lesions in the area where the NIV interface wouldbe positioned. Moreover, positive pressure in thepresence of ethmoidal fractures has been associated


with orbital herniation. Some authors havepostulated that application of NIV in patients withcerebrospinal fluid (CSF) fistulae implies an increasedrisk of post-traumatic meningitis.

PneumothoraxRegardless of how it is delivered, positive pressure

always has negative consequences for lungs withpneumothorax. However, experience with adultpatients does not contra-indicate use of NIV indrained pneumothorax.

Recent gastrointestinal surgeryDehiscence of esophageal sutures has been

described in patients who received NIV during thepost-operative period. The entry of a large volumeof air into the digestive tract with esophageal and/orgastric distension implies a risk during the immediatepost-operative period. However, currently there arepublications that demonstrate the efficacy of NIV atlow pressure without complications.

Despite the existence of certain contra-indications, in the Guidelines of the British ThoracicSociety, use of NIV is accepted as long as intubationis planned or the indication is palliative.

REFERENCES1. Antonelli M, Conti G, Esquinas A, Montini L, Maggiore SM,

Bello G, et al. A multiple-center survey on the use in clinicalpractice of noninvasive ventilation as a first-line interventionfor acute respiratory distress syndrome. Crit Care Med 2007;35:18-25.

2. Hamel DS, Klonin H. The role of noninvasive ventilation for acuterespiratory failure. Respir Care Clin N Am 2006;12:421-35.

3. Venkataraman ST. Mechanical ventilation and respiratory care.En: Fuhrman BP, Zimmerman JJ. Pediatric Critical Care. 3ª ed.Mosby Elsevier.

4. Nava S, Navalesi P, Conti G. Time of non-invasive ventilation.Intensive Care Med 2006;32:361-70.

5. Teague WG. Non-invasive positive pressure ventilation: currentstatus in paediatric patients. Paediatric Respiratory Review2005;6:52-60.

6. Bach JR, Niranjian V. Noninvasive ventilation in children. In BachJR editor, Noninvasive mechanical ventilation. Philadelphia: Han-ley & Belfus, 2002;203222.

7. Nørregaard O. Noninvasive ventilation in children. Eur Respir J2002;20:1332-42.

8. Todd W Rice, Artur P Wheeler, Gordon R Bernard, Douglas L Hay-den, David Schoenfeld, Lorraine B Ware. Comparison of theSpO2/FiO2 ratio and the PaO2/FiO2 in patients with acute lunginjury or ARDS. Chest 2007;132:410-7.

15Indications and contra-indications of non-invasive ventilation in patients with acute respiratory failure

Table I. Causative processes of acute respiratory failure(ARF) for which use of non-invasive ventilation (NIV) shouldbe considered

Decompensated diseases of the central nervoussystem• Congenital (e.g. respiratory sequelae in infantile cere-

bral paralysis)• Acquired (nearly always palliative indication: e.g. cere-

bral tumors)• Central hypoventilation with hypercapnia• Apneas in preterm infants• Apneas in infants

Decompensated abnormalities of the chest wall andof the spinal column• Malformations of the chest wall• Ankylosing spondylitis• Kyphoscoliosis• Achondroplasia• Obesity hypoventilation syndrome

Decompensated neuromuscular diseases in whichrespiratory muscles are affected• Diseases of the second motor neuron (e.g. spinal mus-

cle atrophy)• Guillain-Barré syndrome without signs of bulbar affec-

tation• Diseases of, or damage to, the phrenic nerve• Myasthenia gravis and other congenital myasthenic

syndromes• Myopathies (e.g. congenital, mitochondrial, metabo-

lic or inflammatory, or deposition diseases)• Muscular dystrophies• Myotonic dystrophy• Polio and its after-effects • Botulism (Use NIV with caution)

Diseases of the upper respiratory tract• Obstruction of the upper airway (e.g. laryngitis or

laryngotracheitis)

Decompensated pulmonary diseases• Acute pulmonary edema• Pneumonia• Atelectasis • Bronchiolitis • Cystic fibrosis • Asthma• Neonatal acute respiratory distress syndrome

Other scenarios• Apnea after adenotonsillectomy• Post-operative period after surgical correction of sco-

liosis• Pulmonary complications caused by sickle cell anemia• Early extubation• As support in procedures involving sedation• Severe respiratory failure in terminal illness (palliati-

ve indication)

9. Elliott MW, Confalonieri M, Nava S. Where to perform nonin-vasive ventilation? Eur Respir J 2002;29:1159-66.

10. Antonelli M, Conti G. Noninvasive positive pressure ventilationas treatment for acute respiratory failure in critically ill patients.Critical Care 2000;4:15-22.

11. Piastra M, Antonelli M, Caresta E, Chiaretti A, Polidori G, ContiG. Noninvasive ventilation in childhood acute neuromuscularrespiratory failure: a pilot study. Respiration 2006;73:791-8.

12. Nava S, Gregoretti C, Fanfulla F, Squadrone E, Grassi M, CarlucciA, et al. Noninvasive ventilation to prevent respiratory failureafter extubation in high-risk patients. Crit Care Med 2005;33:2465-70.

13. Ezingeard E, Diconne E, Guymarc’h S, Venet C, Page D, GeryP, et al. Weaning from mechanical ventilation with pressure sup-port in pacients failing a T-tube trial of spontaneous breathing.Intensive Care Med 2006;32:165-9.

14. Ferrer M, Esquinas A, Arancibia F, Abuer TT, Gonzalez G, Carri-llo A, et al. Noninvasive ventilation during persisting weaningfailure: a randomized controlled trial. Am J Respir Crit Care Med2003;168:70-6.

15. Girault C, Daudenthun I, CHevron V, Tamion F, Leroy J, Bon-marchand G. Noninvasive ventilation as a systematic extubationand weaning technique in acute-on-chronic respiratory failure;a prospective, randomized study. Am J Respir Crit Care Med1999;160:86-92.

16. Rana S, Jenad H, Gay PC, Buck CF, Hubmayr RD, Gajic O. Fai-lure of no-invasive ventilation in patients with acute lung injury:observational cohort study. Critical Care 2006;10:R79.

17. Codazzi D, Nacoti M, Passoni M, Bonanomi E, Rota Sperti L,Fumagalli R. Continuous positive airway pressure with modifiedhelmet for treatment of hypoxemic acute respiratory failure ininfants and a preschool population: a feasibility study. PediatrCrit Care Med 2006;7:455-60.

18. Teague WG. Noninvasive ventilation in the pediatric intensivecare unit for children with acute respiratory failure. Pediatr Pul-monol 2003;35:418-26.

19. Masip J. Noninvasive ventilation. Heart Fail Rev 2007;12:119-24.

20. Crummy F, Naughton MT. Non-invasive positive pressure ven-tilation for acute respiratory failure: justified or just hot air? InternMed J 2007;37:112-8.

21. Ikeda H, Asato R, Chin K, Kojima T, Tanaka S, Omori K, et al.Negative-pressure pulmonary edema after resection of media-tinum thyroid goiter. Acta Otolaryngol 2006;126:886-8.

22. Meert AP, Berghmans T, Hardy M, Markiewicz e, Sculier JP. Non-invasive ventilation for carcer patients with life-support techni-ques limitation. Support Care Cancer 2006;14:167-71.

23. Rabitsch W, Staudionger T, Locker GJ, Köstler WJ, Laczika K, FrassM, et al. Respiratory failure after stem cell transplantation: impro-ved outcome with non-invasive ventilation 2005;46:1151-7.

24. Nava S, Cuomo AM. Acute respiaratory failure in the cancerpatient: the role of non-invasive mechanical ventilation. CriticalReviews in Oncology/Hematology 2004;51:91-103.

25. Piastra M, Antonelli M, Chiaretti A, Polidori G, Polidori L, ContiG. Treatment of acute respiratory failure by helmet-deliverednon-invasive pressure support ventilation in children withacute leukemia: a pilot study. Intensive Care Med2004;30:472-6.

26. Campion A, Huvenna H, Letemtres S, Noizet O, Biroche A, Die-pendaele JF et al. Non invasive ventilation in infant with severeinfection presumable due to respiratory syncitial virus: feasibi-lity and failure criteria. Arch Pediatr 2006;13:1404-9.

27. Soma T, Hino M , Kida K, Kudoh S. A Prospective and Rando-mized Study for Improvement of Acute Asthma by Non-invasi-ve Positive Pressure Ventilation (NPPV). Intern Med.2008;47:493-501.

28. Soroksky A, Stav D, Shpirer I. A Pilot Prospective, Randomized,Placebo-Controlled Trial of Bilevel Positive Airway Pressure inAcute Asthmatic Attack. Chest 2003;123:1018-25.

29. Beers SL, Abramo TJ, Bracken A, Wiebe RA. Bilevel positive air-way pressure in the treatment of status asthmaticus in pedia-trics. American Journal of Emergency Medicine 2007; 25:6–9.

30. Akingbola OA, Simakajornboon N, Hadley Jr EF, Hopkins RL.Noninvasive positive-pressure ventilation in pediatric statusasthmaticus. Pediatr Crit Care Med 2002;3:181-4.

31. Carroll CL, Schramm CM. Noninvasive positive pressure venti-lation for the treatment of status asthmaticus in children. AnnAllergy Asthma Immunol 2006; 96:454-9.

32. Mayordomo-Colunga J, Medina A, Rey C, Díaz JJ, Concha A, LosArcos M, Menéndez S. Predictive factors of non invasive venti-lation failure in critically ill children: a prospective epidemiolo-gical study. Intensive Care Med 2009;35:527-36.


INTRODUCTION When considering the conditions and settings

required for administering non-invasive ventilation(NIV), the first aspect to evaluate is the limitationsof the equipment to be used.

Firstly, as with other apparatuses, fixed NIVventilators require AC power and gas tanks (oxygenand medicinal air) and can only be used in hospitals(and only in designated areas). Portable NIVventilators enable greater flexibility, although themajority of them do not have an oxygen blender,and therefore, have limited use for treatment ofseverely hypoxemic patients. Moreover, appropriateventilators and interfaces must be available for eachtype of patient, which in pediatrics impliesmaintaining a broad array of materials andaccessories.

Secondly, personnel caring for the pediatricpatient must have a certain level of expertise inNIV. Although NIV-specific ventilators tend to berelatively easy to operate, no personnel shouldinitiate NIV or change any NIV parameter if theylack fundamental knowledge of the method, donot have experience in treating acute patients, orare not qualified to detect and treat any possiblecomplications.

Lastly, hospital staff must consider theavailability of materials for cardio-respiratorymonitoring (in principle, non-invasive) of patientsand for performing CPR.

If all of the aforementioned conditions have beenmet, then NIV can basically be performed anywhereby using a portable ventilator. However, it ispreferable to maintain acute patients, and to performany changes in the ventilatory parameters for chronicpatients, in a hospital. Application of NIV outside ofthe hospital should be reserved for pre-establishedtreatment of chronic patients.

The following sections provide an overview ofNIV in the pediatric intensive care unit (PICU),general ward, intermediate care areas, ER, deliveryroom and other hospital areas (e.g. sleeplaboratories, diagnostic radiology and the operatingroom), during transport, and lastly, at home.

It should be underscored that factors such asventilator type, and the availability of trained,experienced personnel on hand 24 hours a day, canhave greater influence on the outcome of NIV thanthe location where it is performed.

NIV IN THE PEDIATRIC INTENSIVE CARE UNITThe location which best ensures successful NIV

of acute patients is the PICU, where patients canmost quickly be rescued via intubation andconventional mechanical ventilation (CMV). Someauthors recommend initiating NIV in the PICU forpatients with acute respiratory failure (ARF), whilerecognizing its potential efficacy for patients withstable chronic respiratory failure (CRF).

Conditions and settings for deliveringnon-invasive ventilation

M. Gáboli, J. Mayordomo, P. Gómez de Quero and M. Pons

Chapter 4

Whenever NIV is performed outside of thePICU, it should be treated as an early form ofintervention, but never with the same criteria asthose with which CMV is indicated, since the riskof failure—and consequently, any risks for thepatient—can be extremely high. In contrast to thecase of adults who are not in the ICU, NIV is notused as respiratory support for children who arenot in the PICU.

Only one prospective randomized controlledtrial study on the efficacy of NIV in pediatric acuterespiratory failure have been performed.

Criteria for initiating NIV in the pediatric intensive care unit

Initiation of NIV in the PICU is recommended forpatients presenting with respiratory acidosis,respiratory failure pathology (e.g. pneumonia, acutelung injury [ALI], acute respiratory distress syndrome[ARDS] and asthma). Patients who do not show signsof improvement within the first 2 hours of NIVtreatment in the general ward should be transferredto the PICU to continue NIV under closer observation.The criteria for starting NIV in the PICU comprise: • Respiratory insufficiency which requires FiO2 >

0.4.• Apneas.• pH < 7.30, either initially or after 2 hours of

ineffective NIV treatment in the general ward.• Respiratory therapist, medical or nursing staff in

the general ward has limited experience withNIV.

• Patient or family are not cooperating.

Material • NIV-specific ventilator equipped with an oxygen

blender.• Conventional ventilator with an NIV option.• Interface (preferably oral-nasal).• Assisted cough for neuromuscular disease

patients.

PersonnelAdministration of NIV requires trained medical

and nursing staff, plus respiratory physicaltherapists for neuromuscular disease (NMD)patients. If the latter are not available, they canbe replaced with nurses to perform physicaltherapy.

NIV IN THE GENERAL WARD AND IN INTERMEDIATE CARE AREAS

Recent studies on adult patients show that NIVperformed in the general ward reduces the need forintubation and shortens hospital stay. One multi-center study reported a lower hospital mortality rate.Nonetheless, this technique requires closemonitoring of the patient and a clearly establishedprotocol for transferring the patient to the ICU. Thecriteria may vary among hospitals in function of theavailability of nursing staff and the level of trainingof the personnel. Another factor to consider forpediatric patients is the amount of stress thatadmission into the PICU could imply.

Intermediate respiratory care units (IRCUs), whichexist in countries such as the United States and Italy,have been shown to be cost-effective and can evenimprove patient life expectancy.

NIV in acute pediatric patients can rarely bestarted in the general ward, although some centershave reported periodic experiences in cancer andNMD patients with heightened ARF or CRF. This typeof administration requires qualified personnel whoare available 24 hours a day.

The ideal option for patients in these areas is toprovide an intermediate care unit staffed with trainedpersonnel. The criteria for initiating NIV in the generalward or in intermediate care areas comprise:• Respiratory insufficiency which requires FiO2

< 0.4.• Initial pH > 7.3• Expert staff available 24 hours a day• Patient and family are cooperating

Material• NIV-specific ventilator (no blender required).• Assisted cough for neuromuscular disease

patients.

Personnel Medical and nursing staff, respiratory therapists

must be trained in NIV use.

NIV IN PEDIATRIC EMERGENCIESTo date, in most hospitals NIV in pediatric

patients has generally been administered in thePICU. The ever growing body of evidence on theefficacy of NIV for ARF (primarily Type II), and the

18 M. Gáboli, J. Mayordomo, P. Gómez de Quero, M. Pons

increasing amount of experience with it at manyhospitals, has led to administration of this techniquein less ideal scenarios, such as pediatric emergencies.Moreover, better results have been reported forearly initiation of NIV; hence, the child who is toreceive NIV should be transported to the PICU asquickly as possible.

Several articles on adult patients have beenpublished that demonstrate that NIV can be safelyadministered in the ER to select ARF or CRF patients,such as those suffering from heightened chronicobstructive pulmonary disease (COPD) or acutepulmonary edema (APE), terminal patients, or thosewho have rejected more advanced support methods.Indication of NIV in the ER is less established forasthma and for pneumonia and other types ofhypoxemic or Type I ARF.

All of the aforementioned articles underscorethat the expertise of hospital staff is key to the successof NIV. Given that ER staff may not be ideally suitedto the task, or that the ER may provide less thanoptimal conditions for patient monitoring, NIV inthe ER should only be administered to patients withthe least severe diagnoses.

The NIV mode (CPAP or BiPAP) to be employedin the ER should be the same one used in the ICU.Albeit the recent literature does not clearly favor onemode over the other, certain publications havedescribed CPAP as being simpler.

Indications and ModesAs there is no published list of indications of NIV

for pediatric emergencies, the authors of this chapterhave compiled one here. It must be emphasize thatdespite the fact that NIV in the ER is not performedfor severe ARF patients, hospital staff must alwayshave the material required for emergency intubationprepared and at hand.1. Asthmatic patients with marked respiratory effort

who do not respond sufficiently to conventionalinhalation therapy and oral or parenteralcorticosteroids: These patients can benefit fromNIV in BiPAP mode. The ideal patient for NIVin the ER would be an older child capable ofcooperating who is suffering a moderate tosevere asthmatic attack.

2. NMD patients with ARF, above all, those withupper airway respiratory infections: Thesepatients can respond very well to NIV, given that

their lack of muscle force enables good patient-ventilator synchrony. Some of these patients mayreceive NIV at home, which can facilitatecooperation on their part or the part of theirfamilies.

3. Infants suffering from bronchiolitis (Type I or II):These patients should initially be treated withCPAP via nasal prongs or nasal tube, which canafford better adaption and less air leakage thando nasal or oral-nasal masks.

4. Immunodepressed ARF patients for whomeventual intubation could imply seriouscomplications and for whom rapid administrationof NIV can prevent worsening of respiratoryfailure.

5. Older children with pneumonia and associatedmarked respiratory effort: These patients requireeven closer monitoring than those cited above,as Type I ARF implies much greater risk of NIVfailure than does Type II ARF. CPAP providesbetter gas exchange, although BiPAP decreasesthe load on inspiratory muscles. These patientsshould only be treated by ER personnel withbroad experience in NIV and should beimmediately transferred to the PICU. Patientswith ALI—and of course, those with ARDS—aswell as infants with pneumonia or some othercause of Type I ARF should be immediatelytransferred to the PICU for initiation of NIV orCMV.

Potential problemsThe main problems that can arise with use of

NIV in pediatric emergencies comprise:• Lack of experience or training among hospital

personnel• Rejection of NIV by hospital personnel, above

all, if it fails• Insufficient monitoring • Excessive workload for medical and nursing staff,

namely due to the close monitoring required forthese patients, especially at the onset of NIVadministration

NIV DURING TRANSPORT OF PEDIATRICPATIENTS

Mechanical ventilation during transport ofcritically ill children has improved substantially,

19Conditions and settings for delivering non-invasive ventilation

especially due to advances in portable ventilatorsand monitoring systems, which now provide similarperformance to that typically obtained in the ICU.Although the outlook for NIV during transport maybe promising, there is no specific literature withstrong evidence from which any recommendationson its use could be made.

One of the first things to ensure before transportbegins is that the patient has a permeable airway;if there is any doubt, then the patient should beintubated. This would be the principal limitation onthe application of NIV during transport. One of themajor risks implied in using NIV is a delay inintubation, which must be avoided at all costs beforeand during transport.

Types of transportTransport can be classified into two main groups:

intra-hospital and extra-hospital. NIV is much easierto administer inside the hospital, since it enablesmuch greater safety in terms of controlling thepatient’s airway.

Portable ventilators and ventilation systemsA broad range of commercially available portable

ventilators is now available. However, few of theseare specifically adapted for NIV. Although the Osiris2® and 3®, and the Oxylog 3000®, can perform leakcompensation for adaptation to NIV, use of thesedevices during transport of pediatric patients isproblematic. The Osiris 2® and 3® are only equippedfor positive pressure ventilation with mixtures of air:they can not be used with 100% O2. The Oxylog3000® features a flow sensor with a minimumthreshold of 3 liters per minute (lpm), which isinadequate for young pediatric patients with reducedinspiratory strength. Other ventilators which can bemade portable are not specific for NIV and are notamenable to the difficult working conditions impliedin transporting patients. Conventional ventilators(e.g. Servo-i® and Savina®) can be used, althoughthey are excessively large and are very susceptibleto damage from continuous use in inadequateconditions.

Home ventilators (e.g. VS Ultra® and LegendAir®) meet several of the requisites for use of NIVduring transport of pediatric patients: they are small,flexible, and easy to program. However, they areseverely limited by their lack of an oxygen blender;

hence, their use implies incorporation an oxygen T-piece in line with the tubing and the flow meter.The maximum flow of O2 supplied is 15 lpm andthe FiO2 does not exceed 50%. As such, theseventilators can not be used in hypoxemic patientswhose O2 requirements exceed said value or whosecondition is highly likely to worsen during transport.Some of these ventilators are equipped with oxygenenriching systems that enable FiO2 levels near 80%.

There are now NIV ventilators that can be usedas both conventional and portable ventilators (e.g.Elisée® and Carina®). They feature oxygen blenders,graphic monitoring, and highly sensitive triggeringsystems and are small and light. Although theseventilators are very expensive—which limits theirpracticality for use exclusively during transport—they remain an attractive option due to their multi-functionality (i.e. CMV, NIV and portable ventilation).

Ventilators that feature continuous flow (e.g.Babylog 2000®, CF120® and BabyPac 100®) can beused to administer CPAP in neonates and infants.These are practical for initial treatment of patientsin early phases of Type I ARF (i.e. hyaline membranedisease [HMD], Type I-pattern bronchiolitis, etc.) orType II ARF (i.e. apneas in neonatals or secondaryto bronchiolitis) who can benefit from early CPAPtreatment instead of administration of pure O2. Useof BiPAP with these ventilators is difficult, given thatnone of them are equipped with a trigger, andtherefore, can only perform intermittent mandatoryventilation (IMV). Consequently, these ventilators maysuffer from a high level of desynchronization; hence,they are not the best devices for use during transport.

Another option is the Boussignac CPAP system,which is now used in the pre-hospital stabilizationand transport of adult patients with acute Type IARF. Albeit no studies on its efficacy in childrenhave been published, the authors of this chapterbelieve that it could prove highly utile, as it ischeap, simple (it does not require a respirator) andcomfortable and enables pulmonary recruitmentwhile simultaneously providing effective oxygentherapy.

Indications and ventilation modesPatients should be diligently selected, as the

priority during transport is to maintain permeabilityof the airway. In terms of choosing the ventilationmode, the easiest option seems to be CPAP used


either with ventilators featuring continuous flow orwith systems that are directly adaptable to flowmeters (e.g. Boussignac CPAP or Vygon® CPAP), asthis enables faster adaptation to the patient than useof BiPAP, which demands more careful monitoringand longer adaptation time. Furthermore, systemsfor administering BiPAP generally do not feature anoxygen blender; therefore, they have limited use insituations in which patient control is difficult, suchas in transport.

Based on the premises outlined above, NIV intransport may be easier for patients with Type I ARF(e.g. HMD, near drowning, and meconiumaspiration) than those with Type II. Nonetheless, inType I patients, NIV should only be administered tothose who do not have major respiratory difficultiesor heightened oxygen requirements and who arenot hemodynamically compromised; it is contra-indicated for patients with ALI or ARDS. In Type IIpatients, NIV should be limited to patients whoseoxygen needs are less than 50%. In any case, whenevaluating NIV for use during transport of a pediatricpatient, the contra-indications of NIV, and the specificrisks implied in transporting the patient, shouldalways be very carefully considered.

NIV AT HOME

Indications NIV at home is primarily indicated for CRF that

demands intermittent mechanical ventilation andfor respiratory sleep disorders.

Objectives• Improve the patient’s quality of life: avoiding

flare-ups, preserving cardiopulmonary function,maintaining adequate growth and developmentat each age, and improving the chances ofacademic and social integration.

• Increasing survival rate.

ModesNIV can be performed at home with either

positive pressure (NPPV) or negative pressure(NNPV). The former is typically preferred as it is moreeffective, easier to use during transport and doesnot imply obstruction of the upper airway (anoccasional consequence of NNPV).

ConditionsNIV can be performed at home for patients who

have maintained respiratory autonomy and theability to expectorate and who do not have anydifficulty swallowing. The upper airway must bepermeable, without any fixed obstruction ordeformation that substantially alters it morphology.

In order for parents to care for their children athome, it is crucial that they receive a detailedexplanation of the objectives sought with NIV andthat they give their full approval and cooperation.Patients must also be taught about the specific NIVtreatment that they are to receive and its benefitsin a manner appropriate to their age, cognitive leveland emotional state, and must understand that thesuccess of the treatment will be greatly facilitatedby their cooperation.

Family members and others who are to care forthe child must also be trained in home use of theapparatuses, rapid detection of major problems thatcould arise during NIV administration (e.g.disconnection of the apparatus, mask leaks, sores,and hypoventilation with hypoxia), and maneuversto solve said problems, and must know how toquickly alert emergency services in the event thatthese maneuvers prove insufficient.

It is absolutely critical that there be a team ofhospital personnel to periodically monitor the patient'sprogress, evaluate any required changes in the NIVparameters, quickly solve any complications that mayarise, and maintain close contact with equipmentproviders. This team would preferably have at leastone doctor with expertise in respiratory pathologyand critical pediatric illnesses, as well as nursing staff,a rehabilitation physician and a physical therapist thatis trained in respiratory pathology. The team and thepatient (or family or care provider) must be able tocontact each other by telephone at any time.

If the patient can not be easily transferred to theteam, or if the level of home care can not beguaranteed to meet that provided in the hospital,then home administration of NIV should be seriouslyreconsidered.

Control guidelines: Initially, it would be advisablefor the team to visit the patient at 2 to 4 weeks afterthe start of home NIV, and then at 2 to 6 months(according to the stability of the patient), as well asanytime that the patient suffers from some type ofintercurrent condition that affects the NIV.


NIV IN OTHER LOCATIONS

NIV in sleep laboratoriesRespiratory sleep disorders in pediatrics

encompass a broad array of clinical situations suchas obstructive sleep apnea (OSA) and sleep apneahypopnea syndrome (SAHS), increased resistance inthe upper airway, nocturnal asthma symptoms, andother conditions involving chronic alteration ofrespiratory function.

Sleep laboratories outfitted for children providethe ideal environment to diagnose the respiratorychanges that arise during this period, and can alsoserve as the ideal location to begin NIV (ifindicated).

NIV in diagnostic radiologyOne of the latest areas to which NIV has been

adapted is the radiology ward. Radiologicalexamination can be affected as the patient breathes.As infants and small children can not cooperateduring radiological exams by pausing at the end ofeach deep breath, they must often be sedated toensure that they do not move during chest CTs(whether standard or high-resolution). Some authorshave suggested using NIV (administered via facialmask) to induce a controlled respiratory pause of8 to 12 seconds, during which the images can becaptured. The pause is achieved by increasingpressure in the airway in synch with the patient'sinspiration. NIV can afford better quality radiologicalexams and avoids the interference caused byendotracheal tubes during exploration of the larynxor trachea. This indication of NIV demands closemonitoring of the sedated patient.

NIV in the operating roomNIV can be used in the operating room for

procedures which are brief, relatively painless anddo not directly compromise the airway in childrenwho can not cooperate. There have been reports ofcases in which intubation was impossible where NIVwas used for ventilation until spontaneous breathingwas achieved.

REFERENCES1. Teague WG. Non-invasive positive pressure ventilation: current

status in pediatric patients. Paediatric Respiratory Reviews2005;6:52-60.

2. Norregaard O. Non invasive ventilation in children. Eur RespirJ 2002;20:1332-42.

3. Nava S, Carbone G, DiBattista N, Bellone A, Baiardi P, Cosenti-ni R, et al. Noninvasive ventilation in cardiogenic pulmonaryedema: a multicenter randomised trial. Am J Respir Crit CareMed 2003;168:1432-7.

4. Masip J, Betbesé AJ, Páez J, Vecilla F, Cañizares R, Padró J, et al.Non-invasive pressure support ventilation versus conventionaloxygen therapy in acute cardiogenic pulmonary oedema: a ran-domized study. Lancet 2000;356:2126-32.

5. Biarent D, Bingham R, Richmond S, Maconochie U, Wyllie J,Simpson S, et al. European Resuscitation Council Guidelinesfor Resuscitation 2005. Section 6. Paediatric life support. Resus-citation 2005;67(S1):S97-S133.

6. Massa F, Gonzalez S, Alberti A, Wallis C, Lane R. The use of nasalcontinous positive airway pressure to treat obstructive sleepapnea. Arch Dis Child 2002;87:438-43.

7. Padman R, Hyde C, Foster O. The pediatric use of Bilevel posi-tive airway pressure therapy for obstructive sleep apnea sin-drome: a retrospective review with analysis of respiiratoryparameters. Clin Pediatr 2002;41:163-9.

8. Long FR, Castile RG, Brody AS, Hogan MJ, Flucke RL, Filbrun DA,McCoy KS. Lungs in infants and young children: improved thin-section CT with noninvasive controlled-ventilation technique-initial expererience. Radiology 1999;212:588-93.

9. Fauroux B, Boffa C, Desguerre I, Estournet B, Trang H. Long-termnoninvasive mechanical ventilation for children at home: a natio-nal survey.Pediatric Pulmonology 2003;3:119-25.

10. Miske LJ, Hickey EM, Kolb SM, Weiner DJ, Panitch HB. Use of themechanical in-exsufflator in pediatric patients with neuromus-cular disease and impaired cough. Chest 2004;125:1406-12.

11. Medina A, Pons M, Esquinas A. Ventilación no invasiva en Pedia-tría. Madrid: Ed Ergon 2004.

12. British thoracic Society guideline. Non invasive ventilation inacute respiratory setting. Thorax 2002;57:192-211.

13. International Consensus Conferences in Intensive Care Medici-ne. Noninvasive Positive pressure ventilation in acute respiratoryfailure. Am J Resp Crit Care Med 2001;163:283-91.

14. Elliot MW, Confalonieri M, Nava S. Where to perform noninva-sive ventilation? Eur Resp J 2002;19:1159-66.

15. Plant PK, Owen JL, Elliot MW. Early use of non invasive ventila-tion for acute exarcerbation of chronic obstructive pulmonarydisease on general respiratory wards: a multicentre randomisedcontrolled trial. Lancet 2000;355:1931-35.

16. Fort PA, Boussarie C, Hilbert G, Habachi M. Ventilation non inva-sive en prehospitalier. Etude d'interet et de faisabilite (7 cas).Presse Med 2002;31:1886-9.

17. Pelosi P, Severgnini P, Aspesi M, Gamberoni C, Chiumello D, Fachi-netti C, Introzzi L, Antonelli M, Chiaranda M. Non-invasive venti-lation delivered by conventional interfaces and helmet in theemergency department. Eur J Emerg Med 2003;10:79-86.

18. Antro C, Merico F, Urbino R, Gai V. Non-invasive ventilation asa first-line treatment for acute respiratory failure: "real life" expe-rience in the emergency department. Emerg Med J 2005,22:772-7.

19. Yosefy C, Hay E, Ben-Barak A, Derazon H, Magen E, Reisin L,Scharf S. BiPAP ventilation as assistance for patients presenting


with respiratory distress in the department of emergency medi-cine. Am J Respir Med 2003;2:343-7.

20. Templier F, Dolveck F, Baer M, Chauvin M, Fletcher D. Boussignaccontinuous positive airway pressure system: practical use in a pre-hospital medical care unit. Eur J Emerg Med 2003;10:87-93.

21. Brochard, L. Ventilation non invasive dans l'insuffisance respi-ratoire aigue. Rev Prat 2003;53:980-4.

22. Browning J, Atwood B, Gray A. Use of non-invasive ventilationin UK emergency departments. Emerg Med J 2006;23:920-1.

23. Crane SD, Elliott MW, Gilligan P, Richards K, Gray AJ. Randomi-sed controlled comparison of continuous positive airways pres-sure, bilevel non-invasive ventilation, and standard treatment inemergency department patients with acute cardiogenic pul-monary oedema. Emerg Med J 2004;21:155-61.

24. Cross AM, Cameron P, Kierce M, Ragg M, Nelly AM. Non-inva-sive ventilation in acute respiratory failure: a randomised com-parison of continuous positive airway pressure and bi-level positiveairway pressure. Emerg Med J 2003;20:531-4.

25. L'Her E, Moriconi M, Texier F, Bouquin V, Kaba L, Renault A,Garo B, Boles JM. Non-invasive continuous positive airwaypressure in acute hypoxaemic respiratory failure--experienceof an emergency department. Eur J Emerg Med 1998:5;313-8.

26. Merlani PG, Pasquina P, Granier JM, Treggiari M, RutschmannO, Ricou B. Factors associated with failure of noninvasive posi-tive pressure ventilation in the emergency department. AcadEmerg Med 2005;12:1206-15.

27. Thys F, Roeseler J, Delaere S, Palavecino L, El Gariani A, MarionE, Meert P, Danse E, D’Hoore W, Reynaert M. Two-level non-inva-sive positive pressure ventilation in the initial treatment of acuterespiratory failure in an emergency department in Belgium. EurJ Emerg Med 1999;6:207-14.

28. Yañez LJ, Yunge M, Emilfork M, Lapadula M, Alcántara A, Fer-nández C, et al. A prospective, randomized, controlled trial ofnoninvasive ventilation in pediatric acute respiratory failure.Pediatr Crit Care Med 2008;9:484-9.


INTRODUCTION In non-invasive ventilation (NIV), interfaces are

devices used as a bridge between the patient andthe tubing of the mechanical ventilator. They arepositioned in or around the nose or mouth andadjusted to obtain a semi-hermetic seal.

Successful NIV requires an NIV-compatibleinterface (i.e. features and material) and fasteningsystem. Indeed, failure of NIV within the first fewhours of treatment is usually due to patientdiscomfort with the interface or to a lack of patient-ventilator synchronization.

The type of interface chosen for each situationdepends on two main factors: the patient’s age andtype of respiratory failure (RF). The availability ofmaterials in each ward (see Table I) is also influential.

There are various types of interfaces, the mostwell-known of which are nasal and oral-nasal masks.Other systems include nasal pillows, nasal prongs,pharyngeal cannulae, full face masks, and helmets.

Although pediatric use of NIV has increased, nogeneral guidelines on its use have yet been established,nor have any studies been published in which strongevidence could be used to support specificrecommendations. Hence, physicians must developtheir pediatric clinical practices according to thepediatric consensuses and data on adult patients.

CLASSIFICATION OF NIV INTERFACESNIV interfaces can be classified according to

various characteristics:

• Material: silicone, silicone gel or a mixture ofboth.

• Location: nasal, nasal prong, intranasal, oral-nasal, or oral.

• Safety: with or without an anti-asphyxia valve• Exhalation: vented or non-vented.• Use: single use or reusable.

CONVENTIONAL INTERFACES: MASKSThe most popular masks in pediatric NIV are

nasal and oral-nasal masks. These must be fabricatedwith non-allergenic materials and should be adaptedto the patients face to minimize the possibility ofleaks. In neonatal NIV, masks are no longer widelyused due to their tendency to obstruct the nasalairway as well as the difficulty they present inobtaining an adequate seal. There are new maskdesigns which can be effective alternatives to othernasal interfaces that are more complicated to attachor that cause local lesions.

Nasal masks are generally indicated for Type IIchronic respiratory failure (CRF) or Type I acuterespiratory failure (ARF), but only for patients thatare not very dyspneic, and can cooperate and keeptheir mouth closed—otherwise, leak compensationwould make the mask intolerable. Oral-nasal masksare best suited for Type I and advanced Type II ARF,in which the patient can not breathe through thenose alone, especially for dyspneic patients that tendto breathe through the mouth. They are alsoindicated for cases in which it is impossible to keep

Non-invasive ventilation interfaces

A. Concha, A. Medina, M. Pons and F. Martinón-Torres

Chapter 5

the patient's mouth closed during use of a nasalmask.

No studies on the efficiency of each system forpediatric patients have been published; hence, useof these interfaces depends on personal experienceand on the tolerance and efficacy in each patient.

Given that pediatric ARF patients that meetintubation criteria currently tend to be admitted intothe PICU, the most widely used—and possibly, themost effective—interface is the oral-nasal mask, asin the case of adults, indicated above.

Use of nasal and oral-nasal masks in neonatesand infants is difficult due to the limited selectionof materials available for these ages (Fig. 1) as wellas to the challenges encountered in maintainingthese systems during feeding, crying, movement,etc.

Each type of interface has its respectiveadvantages and disadvantages (see Table II).

Masks tend to be fabricated primarily fromsilicone mixed with plastic. The former is used toprovide better comfort and tolerance, although in

some cases adaptation to the patient may still beinsufficient. Some interfaces are moldable, whichfacilitates adaptation to the patient’s face (Figs. 2and 4). For example, once run under hot water,silicone gel (Fig. 2) interfaces can be molded toachieve better coupling to the patient’s nose.

26 A. Concha y cols.

Table I. General recommendations for choosing an NIV interface. ARF: acute respiratory failure

ARF Age Choice Alternatives

TYPE I Neonates Short double nasal prongs Nasal-pharyngeal tubeNasal mask

Infants Large nasal mask used an oral-nasal mask Nasal maskShort nasal prongs

4 to 6 years Oral-nasal mask Nasal maskNasal-pharyngeal tube

6 to 12 years Oral-nasal mask Nasal maskHelmet

> 12 years Oral-nasal mask Full face maskHelmetNasal mask

TYPE II Neonates Short double nasal prongs Nasal-pharyngeal tube

Infants If FiO2 < 0.5: nasal mask, nasal prongs Short nasal prongsIf FiO2 > 0.5: large nasal mask used as Nasal-pharyngeal tubeoral-nasal mask Nasal-pharyngeal tube

1 to 6 years Oral-nasal mask Nasal maskNasal-pharyngeal tube

6 to 12 years If FiO2 < 0.5: nasal mask Oral-nasal maskIf FiO2 > 0.5: oral-nasal mask

> 12 years If FiO2 < 0.5: nasal mask Oral-nasal maskIf FiO2 > 0.5: oral-nasal mask Full face mask

Figure 1. Respironics® oral-nasal mask used on an infant.

Regardless of the interface material, one of themost common complications encountered inadministration of NIV is skin irritation caused bypressure, which can lead to sores and skin necrosis.This effect can be minimized through pre-applicationof special dressings (e.g. Comfeel®, Askina® andDuoderma®), a practice that should be madestandard in all acute patients before beginning NIV.

If required, nasal masks can be custom made,although this can be difficult as it requires thepatient’s cooperation. This is generally done forchronic patients for whom adequate masks are notcommercially available.

Masks may have ports for pressure lines, CO2

measurement or other uses; these must remain

closed when not in use. Said connectors can also beused to administer O2, although this is not an idealmethod since it generates a turbulent flow inside ofthe mask, which increases discomfort.

Masks may be vented (Fig. 3), whereby theexhaled air is eliminated through a vent in the maskitself, or non-vented (Fig. 4), whereby the exhaled

27Non-invasive ventilation interfaces

Table II. Advantages and disadvantages of oral-nasal and nasal masks

Type Advantages Disadvantages

Oral-nasal • More effective than nasal masks (dyspneic • Higher incidence of claustrophobiapatients tend to breathe through the mouth) • More complicated maintenance in the event of

• Avoids nasal resistance coughing or vomiting• Less comfortable in the long-term• More dead space

Nasal • More comfortable and tolerable in the • Loss of efficacy when the mouth is open, long-term requiring greater work by the diaphragm

• Lower incidence of leakage implica mayor trabajo diafragmático• Less dead space • Leakage via the mouth. Increased flow to • Lower incidence of claustrophobia compensate for leaks causes de-adaptation.• Lower risk in the event of vomiting• Enables expectoration and feeding without

removal of the mask

Figure 2. SleepNet nasal interface used as an oral-nasal inter-face on an infant.

Figure 3. Resmed® vented interface.

air is eliminated through the tubing and the maskdoes not have any vents. Both types of masks canbe used with NIV-specific ventilators. However, if aconventional ventilator with double tubing is used,then non-vented masks should be employed and allof the ports should be closed to minimize any leaksthat the ventilator is not able to offset.

Some masks are equipped with an anti-asphyxiavalve (Fig. 5). These valves are intended to captureair outside of the tubing to prevent re-inhalationby the patient in the event of ventilator failure orpower outtage. These valves act as gates which areopened by the flow of air emitted by the ventilator.Should this flow of air prove insufficient, the gatecloses, enabling the patient to breathe ambient air.If the expiratory positive airway pressure (EPAP) istoo low or a conventional ventilator withdiscontinuous flow is being used, the anti-asphyxiavalve can open and close intermittently; in this case,the preferred option is to use a different mask,otherwise, the EPAP must be increased. Theseinterfaces should not be used in conventionalventilators equipped with an NIV option, since theseventilators do not compensate for leaks well, and

therefore, should only be used with masks withoutan anti-asphyxia valve (Fig. 6).

However, masks with an anti-asphyxia valve aresuitable for home NIV patients who cannot adaptto nasal interfaces: in the event of a power outage


Figure 5. Hans-Rudolph oral-nasal interface with anti-asphyxiavalve.

Figure 4. Draëger Novastar non-vented oral nasal interface.

Figure 6. Respironics oral-nasal mask without anti-asphyxiavalve.

they prevent asphyxiation, as many home NIVventilators are not equipped with an internal battery.

Pediatric care providers should be have a broadarray of interfaces (Fig. 7) on hand in order to choosethe best mask for each child and allow for thepossibility of changing the interface over the courseof the day to eliminate pressure points in the samearea of the child’s face. It should be mentioned thatnasal and oral-nasal masks designed for adults areoften used for children.

The number of brands of masks continues to rise(see Table III). Respironics and Resmed® are the twobrands in Spain that currently offer the greatestvariety of sizes for pediatric patients.

Respironics provides a wide selection of reusablemasks: nasal, nasal gel and oral-nasal, including fordifferent nose widths. It also has small nasal masksfor infants and toddlers. Its Comfort Flap® adaptors,used on the cushion of its nasal masks (Fig. 8), createan air cushion that enables greater patient comfortby facilitating adaptability and reducing tension inthe straps. Respironics also offers a mask (ContourDeluxe™) whose cushion has dual flaps.

Resmed® offers masks (Infant Mask System)which are designed for patients aged 0 to 6 years.These masks feature a rigid pediatric frame that canadapt to two different sized cushions, a frontal T-support that provides greater stability and a largercontact surface area while lessening the risk ofpressure sores, and a five-point head strap. Resmed®

used to offer the smallest commercially availablenasal mask for infants; however, they stoppedmanufacturing it because the cushion was so soft

that it blocked the nostrils. Their Mirage Kidsta® nasalmask is utile for patients over 6 years old. Its specialharness is primarily supported by the cheeks andthe lower nose, thereby reducing pressure on thenasal bridge and forehead. Moreover, it affordsextremely low levels of leakage, thanks to its double-wall cushion, which inflates upon entry of air intothe mask. Its design provides patients with a clearfield of vision, allowing them to read and helpingthem to acclimate (see Chapter 11, Figure 3).

Lastly, there are oral interfaces, available in twomouth sizes (small and large), indicated for patientswith mouth-breathing or chronic nasal obstruction(see Chapter 19, Figure 5).

ALTERNATIVE INTERFACESFor cases in which NIV can not be administered

via mask, alternative—albeit less effective—interfacesare available.


Figure 7. Nasal and oral-nasal masks. Choosing the best maskfor each child requires having a wide variety of models andsizes available.

Figure 8. Respironics nasal mask with adaptor. Top: Mask andadaptor. Bottom: Adaptor coupled to the mask (ComfortFlap®).


Table III. The most widely used nasal and oral-nasal masks. Models available in pediatric sizes are listed in bold.

Brand Type Brand Sizes* Sizing by age (years)

Respironics® Oral-nasal • Spectrum • P, S, M, L Lite profile <1• Comfort Full 2 • S, M, L Petite 1-3• Image3, Image3 SE vented/novented • S, M, L Small 3- 6

Medium 7-14Large > 14

Nasal • Profile Lite (modeable gel, vented) • P, S, MS, M, MW, L, LN• ComfortGel (modeable + comfort flap) • P, S, M, L• Comfort Classic® • S, M• Contour Deluxe (no vented) • P, S, M-L• ComfortLite 1 and 2 (3 interfaces: • P, S, M, L + 6 sizes

pillow cushions, simple cushions or direct seal cushionsdirect seal cushions) • S, M, L

• ComfortCurve (3 seals) • S, M, SW• Comfort Select • P, S, MS, M, MW, • Nasal Gold Seal (7 sizes) L, LN• Simplicity • S, M

Resmed® Oro-nasal • Mirage Quattro • XS, S, M, L S: 1 to 4• Ultra Mirage (vented or non-vented) • S, M, L shallow or standard M: 4 to 10• Mirage serie 2 (vented or non-vented) • S, M, L shallow or standard L: 10 to adult• Mirage Liberty (oral and nasal pillows) • Oral: S, L + Nasal: S,

M, LNasal • Infant mask system • Size < 2 y, cushion: S, L Infant mask

• Mirage Kidsta • S (> 6 years) system < 2• Sullivan Mirage (vented or non-vented) • S, M, L shallow o S S: 1-4t• Ultra Mirage II (vented or non-vented) • St, L, shallow, wide M: 4-10• Mirage Activa • St, L, shallow L: > 10• Mirage Vista 2 • St or deep

Hans Oro-nasal • Hans Rudolph VIP 7500, 7600 • P, XS, S, M, L Rudolph® Nasal • Alizes (vented or non-vented) • S M, LDräger® Oro-nasal • ClassicStar (vented or non-vented) • S, M, L >30 kg

• NovaStar (vented or non-vented) • S, M, LFisher & Oro-nasal • FlexiFit 431, 432 (vented or non-vented) • S, M, L Paykel® Nasal • Infinity 481 • L

• Aclaim2 • 1 size 2 silicone seals• FlexiFit 405 • 1 size, 2 seals: S/M-M/L• FlexiFit 406 “Petite” (vented) • 1 size P • FlexiFit 407 (vented or non-vented) • 1 size, silicone seal

Koo Oro-nasal • Bluestar, Bluestar Plus, inflatable • Child, S, M, L, Ultra LMedical® Nasal • Moonlight Deluxe • S, M, LWeinmann®Oro-nasal • YARA (vented or non-vented) • S, M, L

Nasal • Joyce (vented or non-vented) • S, M, LSleepNet Oro-nasal • MOJO gel • S, M, LCorp.® Nasal • IQ (vented or non-vented) • S, M, L

• MiniME (vented or 3 caps P, S, XS) • Single sizeTyco® Nasal • Breeze SleepGear with nasal pillows or • 7 pillows-3 nasal masksHealthcare Dream Seal masck • St, L, Plana

• DreamFit® (vented) • 6 siezes• SoftFit Ultra Nasal

Vygon® Oro-nasal • CPAP Boussignac • 6 type: neonate, infant, child, adult S, M, L

*Sizes: P = petite; XS = extra small; S = small; M = medium; MN = medium-narrow; MW = medium-wide; L = large, LN =large-narrow; Std. = standar

Nasal pillow systems (Adams-type)Nasal pillow systems arose as an alternative to

nasal masks for treating patients with sleepingdisorders. They provide patients with better visionand greater comfort (Fig. 9). They can also be usedas an alternative interface for patients with pressuresores on the nasal bridge. There are several models

of these systems available, some of which are listedin Table IV.

A noteworthy example of Adams circuits is theairway delivery and management (ADAM) interfacefrom Puritan Bennett®. This system is coupled to thenostrils through nasal pillows (available in sevensizes) and supported by the head, over which thetube is fastened via Velcro® straps to avoid pressureon the nasal bridge.

HelmetHelmets were initially designed for maintenance

of respiratory failure in hypoxemic adults. Althoughthey are employed as an alternative to conventionalmasks in patients with ARF, to date few experienceswith them in pediatric patients have been reported.

Helmets are transparent PVC cylinders that coverthe patient’s head. They have a strap at neck leveland are fastened at armpit level via an abdominalbelt or a harness. These systems are advantageousin that they do not imply any points of facial contactand feature ports for nasal-gastric exploration andfor catheters. Given their large dead space, use ofhelmets in NIV requires a high flow (> 35 liters perminute [lpm]) to reduce the risk of CO2 re-inhalation.

There are currently two models of helmets onthe market, both of which are made by StarMed®.These are available in various sizes according to thecircumference of the patient's neck:• A model (Castar R®) for bi-level positive airway

pressure (BiPAP) ventilation (available in small,medium, large and extra-large). It features two


Figure 9. Adams circuit used for an adolescent patient.

Table IV. An overview of various nasal pillow masks

Brand Name Sizes

Respironics® • ComfortLite 1 and 2 (3 interfaces: pillow • Pillow cushions P, S, M, L + simple cushions: cushions, simple cushions or direct seal cushions) P, S, M, L + 6 direct seal cushions

• OptiLife (with chin-support band) • P, S, M, L

Resmed® • Mirage Liberty (nasal pillows or oral pillow) • Nasal pillows: S, M, L + oral pillow: , L• Mirage Swift I y II • 3 pillows: S, M, L

Fisher & Paykel® • Infinity 481 • 4 seal sizes: XS, S, M, L • Opus 482 • 3 silicone seal sizes: S,M,L

Tyco® Healthcare • Breeze® SleepGear® CPAP (with nasal pillow • 30 kg. 7 pillow sizes and 3 mask sizesPuritan Bennett® or Dream Seal mask) • 7 pillows

• Adam Circuit

*Sizes: P = petite; XS = extra small; S = small; M = medium; L = large.

inflatable nasal pillows at its base and top thatare designed to reduce internal volume.

• A model (Castar®) for continuous positive airwaypressure (CPAP) ventilation and oxygen therapy.CPAP is generated through a PEEP valveconnected to the expiratory circuit. In contrastto its BiPAP counterpart, this model comes inspecific pediatric sizes. The Infant Low helmet(Fig. 10) is indicated for children weighing 3 to10 kg, and the Infant High, for 10 to 15 kg. Thesystem has an anatomically moldable pillow tohelp support the child’s head. The adult helmetweights 400 g, and the pediatric one, 260 g.

Full face masksFull face masks are only available in adult sizes;

hence, their use in pediatric medicine is limited toadolescent patients (Fig. 11). They are a goodalternative for patients in whom a good seal can notbe achieved with oral-nasal masks; patients withpressure sores; and patients who suffer fromclaustrophobia, as they do not obstruct the patient’sfield of vision. These masks minimize the risk of leaks,but have similar disadvantages as oral-nasal masks.

Alternative systems for neonatesCurrently, neonatal NIV is almost exclusively

administered through nasal devices or interfaces.Masks that are strongly fastened to the face, anddevices that require sealing at neck level, have beenrejected in pediatric medicine due to serious

complications associated with their use, includingan increase in intracerebral hemorrhage (ICH) andpost-hemorrhagic hydrocephalus. Noteworthyexamples of alternative interfaces for neonates andinfants are listed below:

Nasal prongsNasal prongs (tubes) are indicated for neonates

and small infants with Type I and Type II RF. Singlenasal prongs are simple and widely-used, althoughthe majority of authors consider them to be inferiorto double nasal prongs.

Double nasal prongsDouble nasal (bi-nasal) prongs are the preferred

interface for neonatal NIV. Several models are nowcommercially available (e.g. Argyle®, Hudson®,Vygon®, Inca®, BabyFlow®, Alladin/Infant Flow® andGiulia®). They are simple, safe and effective. However,they carry a risk of causing nasal trauma, and somemodels may imply a significant increase in thepatient’s breathing work. Hence, new nasal interfaceshave been designed to minimize these effects andto reduce any additional resistance in the airway dueto the device as well as any fluctuations in thepressure delivered to the airway. Noteworthyexamples of new nasal prong systems include:• BabyFlow® (Dräger®): Employs a set of adaptors

that substitute the flow sensor of the Babylog8000 ventilator and that contain a silicone doublenasal prong available in different sizes. Eliminating


Figure 10. Helmet interface for administering CPAP to aninfant. Model: Castar®. Figure 11. Full face mask used for NIV during weaning from

conventional mechanical ventilation of a patient sufferingfrom ARDS secondary to a hemophagocytic syndrome.

the flow sensor means that when BabyFlow® isused with the Babylog 8000, only CPAP can beadministered; however, if it is used with the Evitaventilator (which contains an internal flow sensor),then BiPAP can be administered (Fig. 12).

• Infant Flow®: Designed to maintain sufficientflow during patient inspiration (i.e. the Bernouilliand Venturi effects) while minimizing resistanceto expiration (i.e. the Coanda effect, and fluidicflip). For more information on Infant Flow®, thereader is referred to Chapter 17, “Neonatal non-invasive ventilation”. This system can be usedwith short double nasal prongs, or soft siliconenasal masks of various sizes, to enable theexhaust flow to exit as gas enters. This increasesits effective internal diameter, thereby decreasingleaks around the prongs (Fig. 13). The largereffective diameter and fine walls of the prongs,and the fact that no gas enters during expiration(due to fluidic flip), together reduce addedbreathing work normally implied by this type ofinterface. The Infant Flow® is D-shaped, wherebythe area supported by the septum is flat—adesign intended to minimize pressure sores andprevent septal necrosis, which are the maincomplications encountered with nasal CPAP. Itincludes a guide for selecting the appropriatetube in function of the patient’s nostrils (thelargest tolerated diameter should be applied).

Nasal cannulaeFor treatment of apnea in premature infants,

CPAP administered via nasal cannulae with a flow

of up to 2.5 L/min can provide comparable resultsto those obtained with conventional CPAP via nasalprongs. It has been shown that cannulae with anexternal diameter of 3 mm and a flow of up to 2L/min can increase intra-esophageal pressure andreduce asynchrony in thoracicoabdominalmovements. Nevertheless, optimal parameters offlow, cannula size and humidification level have yetto be established.

Nasal-pharyngeal tubesNasal-pharyngeal tubes have been used to

deliver CPAP for over 30 years and can be utile forneonates and infants. They enable feeding throughnasal-gastric tubes without interruption of ventilation.However, they have their disadvantages: comparedto other interfaces they are less tolerant, less effective(they are prone to leakage from the mouth and the


Figure 12. BabyFlow®

system used with theEvita XL® respirator foran infant.

Figure 13. Top: Infant Flow® system used for an infant. Bot-tom: interfaces that are compatible with the system (nasal maskand nasal prongs).

contra-lateral nostril), and harder to fasten. Leakagefrom the mouth can be minimized through the useof a pacifier. A shortened endotracheal tube insertedthrough a nostril can be used as a nasal-pharyngealtube (Fig. 14). The depth of insertion should becalculated externally from the patient, such that thetube is positioned in the oropharynx without causingdiscomfort. The reason for cutting the tube is toreduce dead space. It is attached to the nose usingsurgical tape. This system has primarily been usedfor delivering CPAP, although it can also be employedfor intermittent positive pressure ventilation. Thesetubes can be used in tandem with either Benveniste-type devices (Fig. 15) with pre-warmed and pre-humidified airflow or ventilators with highly sensitivetrigger systems that are capable of detecting thesepatients’ small variations in flow, which can be madeeven smaller by major leaks.

FASTENING SYSTEMSThe majority of interfaces are adapted to the

child’s face through elastic straps or caps (Fig. 16),which are adjusted tightly enough to prevent airleaks but loosely enough to avoid generatingexcessive pressure. Caps are easier and faster toposition than straps, but allow less perspiration andtherefore, are uncomfortable in warm environments.Straps require more effort to achieve a good fit, butthey enable better perspiration, thereby allowingthe patient to stay cooler. Just as in the case of masks,choosing the best fastening system for each childand each interface requires having a wide variety ofsystems on hand. The BabyFlow® and Infant Flow®

systems each feature specific fastening systems indifferent sizes. Respironics and Resmed® offer variouscaps and strap systems (Fig. 17) that are adaptableto different masks. However, most masks nowincorporate their own fastening system (Fig. 18),which eliminates the difficulties of coupling interfacesto fastening equipment. Helmets are fastened eitherunder the armpits (for adolescents) or through anintegrated piece of cloth that passes under the body(for infants) (see Fig. 19).

DISADVANTAGES AND COMPLICATIONSPediatric NIV has its drawbacks. Above all:

• It puts high demands on medical and nursingstaff in terms of attending to the patient, namely,


Figure 14. NIV administered with the Respironics BiPAP®

Vision® system via a nasal-pharyngeal tube (shortened endo-tracheal tube) in an infant with pneumonia.

Figure 15. NIV administered using the Benveniste device.

Figure 16. Different fastening systems on young patients:cap; elastic straps; and a ventilation cap designed for theInfant Flow® system.

for adjusting the interface to minimize leaks andobtain good adaptation to the ventilator.

• There is a limited selection of commerciallyavailable masks, which leads to a low rate ofsuccess with, and a lack of confidence in, thetechnique.

• Due to factors such as age and pathology NIVis only suitable for a relatively small populationof RF patients. This makes obtaining experiencefor more complex cases difficult. The primarycomplications associated with NIV interfaces arecovered in Chapter 11, “Complications andtechnical problems in non-invasive ventilation”.

RECOMMENDATIONS FOR STARTING NIVThe following recommendations are intended

as a guide to starting NIV and should be adapted toeach patient’s situation (i.e. comfort level, technicaloptions, etc.):A. The choice of interface depends on various

factors, which are listed below in order ofimportance:1. Size and age of the patient.2. Phase of the disease and/or blood gas

levels.3. Type of respiratory failure (RF).4. Availability of material.5. The patient’s level of cooperation.Table II lists major interfaces NIV. These are to

be chosen according to the patient’s age, type ofRF, and specific needs as well as in function of thecapacities of each ward.

B. One or more alternatives should be made readyfor patients whose facial profile, pathology (e.g.insufficient strength to activate the inspiratorytrigger, and pressure sores) or comfort level arenot compatible with the first interface chosen.

Figure 18. Initial protocol for starting NIV in function of theinterface.

Choose the interface

Apply protective dressings

Adapt patient

Position the harness of fastening system

Attach the interface (requires 2 people)

Connect the ventilator

Confirm proper functioning

Program disconnection (rest) periods

AgeType of ARFAvailability of material

Explain treatment to patient before beginning(age permiting)Sedate patient (if required)

Cooperating children should beable to maintain the interface

Do not adjust the interface too tightly(two fingers should fit underneath)The mask should be perpendicularto the face and symmetric from topto bottom

Check level of adaptationCheck for leaks

Figure 17. Bonnet that incorporates a strap to facilitate clo-sing of the mouth.

Figure 19. A helmet fastened to an infant.


REFERENCES

1. de Carvalho WB, Johnston C. The fundamental role of inter-faces in noninvasive positive pressure ventilation. Pediatr CritCare Med 2006;7:495-6.

2. Medina A, Pons M, Martinón-Torres F, Rey C, Prieto S. Inter-fases en pediatría. En: Medina A, Pons M, Esquinas A (eds).Ventilación no invasiva en Pediatría. Madrid: Editorial Ergon2004;35-44.

3. Kwok H, McCormack J, Cece R, Houtchens J, Hill NS. Con-trolled trial of oronasal versus nasal mask ventilation in thetreatment of acute respiratory failure. Crit Care Med2003;31:468-73.

4. Hess DR. Noninvasive ventilation in neuromuscular disea-se: equipment and application. Respir Care 2006;51:896-911.

5. De PA, Davis PG, Faber B, Morley CJ. Devices and pressuresources for administration of nasal continuous positive air-way pressure (NCPAP) in preterm neonates. Cochrane Data-base Syst Rev 2002;4:CD002977.

6. Roy B, Cordova FC, Travaline JM, D'Alonzo GE Jr, Criner GJ.Full face mask for noninvasive positive-pressure ventilationin patients with acute respiratory failure. J Am OsteopathAssoc 2007;107:148-56.

7. Medina A, Prieto S, Los Arcos M, Rey C, Concha C, Menén-dez S, Crespo M. Aplicación de ventilación no invasiva enuna unidad de cuidados intensivos pediátricos. An Pediatr(Barc) 2005;62:13-19.

8. García Teresa MA, Martín Barba C. Ventilación no invasiva.En J López-Herce, C Calvo, M Lorente, D. Jaimovich (eds).Manual de Cuidados Intensivos Pediátricos. Editorial Publi-med. Madrid 2001;655-60.

9. Raposo G, Correia M. Ventilación mecánica no invasiva. EnF Ruza. Tratado de Cuidados Intensivos Pediátricos. Edicio-nes Norma-Capitel. Madrid 2003;690-5.

10. Fauroux B, Lavis JF, Nicot F, Picard A, Boelle PY, Clément A,et al. Facial side effects during noninvasive positive pressu-re ventilation in children. Intensive Care Med 2005;31:965-9.

11. Teague WG. Non-invasive positive pressure ventilation: currentstatus in paediatric patients. Paediatr Respir Rev 2005;6:52-60.

12. Sinuff T, Kahnamoui K, Cook DJ, Giacomini M. Practice gui-delines as multipurpose tools: a qualitative study of nonin-vasive ventilation. Crit Care Med 2007;35:776-82.

13. Chiumello D. Is the helmet different than the face mask in deli-vering noninvasive ventilation? Chest 2006;129:1402-3.


A broad array of devices for non-invasiveventilation (NIV) is now available. For example,invasive ventilators used in the pediatric intensivecare unit (PICU), designed for intubated patientswith little or no leakage, now increasingly featurean NIV option. There are also ventilators which arespecifically developed for NIV. Some of these areintended for the hospital use, whereas others areintended for home use; the former tend to be morecomplex than the latter. Furthermore, there areventilators designed for neonatal NIV, and ventilatorsdesigned for use during transport, which alsoincreasingly feature the option to administer NIV.

This chapter provides a broad overview of variousNIV systems and ventilators. Due to the constanttechnological advances in the field, only certainmodels have been described here.

NEGATIVE PRESSURE NIV SYSTEMSNon-invasive negative pressure ventilation

(NNPV) systems are based on the classic iron lung.They work by applying a sub-atmospheric pressurearound the chest wall of the patient, which generatesa pressure gradient from the mouth to the alveoli,and consequently, a flow of air into the lungs duringinspiration. During expiration the externallyproduced negative pressure stops, allowing air toexit the alveoli. This flow usually occurs passively,although some devices apply a positive pressureto drive expiration.

There are three main modes of NNPV: negativepressure applied during inspiration; negative pressureduring inspiration and positive pressure duringexpiration; and continuous negative pressure, whichis analogous to continuous positive airway pressure(CPAP). All of these methods tend to involve a lotof equipment and are therefore difficult to use duringtransport. The main complication with NNPV is thatit can cause obstructive sleep apnea through collapseof the upper airway. One of the advantages of NNPVis that it increases the venous flow to the heart; henceit may be indicated in acute respiratory failure (ARF)patients with low cardiac output after right heartsurgery. These methods can be used individually orin combination with positive pressure ventilationdelivered through an endotracheal tube. However,NNPV systems have generally been replaced bymodern positive pressure systems.

There are various types of NNPV systems:

Iron lungs or tank ventilatorsThese are the most mechanically effective

systems for guaranteeing a total seal of the chestwall area. However, they are bulky and thereforedifficult to transport.

Cuirasses and jacketsCuirasses (shells) and jackets (ponchos) tend

to be made of plastic which adapts to the frontand lateral areas of the chest wall and theabdomen. Jackets also feature nylon cords that

Non-invasive ventilation devices

S. Menéndez, A. Medina, F. Martinón-Torres and C. Rey

Chapter 6

reach the neck, wrists and elbows to aid in sealingthe system. Both cuirasses and jackets connect toa central unit equipped with a pump that generatesa negative inspiratory pressure and in some cases,a positive expiratory pressure. Continuous negativepressure is sometimes used to prevent alveolarcollapse. Compared to iron lungs, cuirasses andjackets are more comfortable and manageable butless effective, as they do not completely surroundthe patient’s chest wall. Moreover, they may bedifficult to adapt to patients with chest wallmalformations, and therefore, may need to becustom made.

POSITIVE PRESSURE NIV SYSTEMSNon-invasive positive pressure ventilation (NPPV)

consists of facilitating the entry of gas into the lungsvia cyclic application of a positive pressure in theairway using extra-tracheal systems (e.g. face ornasal masks, and nasal or pharyngeal cannulae).

Conventional invasive ventilatorsAny conventional mechanical ventilation (CMV)

device can be used to deliver NIV, although withserious limitations. Indeed, before NIV-specificventilators entered the market, CMV ventilators wereused for NIV (Fig. 1). Using volume-controlledventilation (VCV) modes, leaks were compensatedfor by increasing the tidal volume (Vt) more or lessrandomly. On one hand this strategy led to excessivepeak inspiratory pressure (PIP) in cases of lowleakage, while on the other hand, it could notsufficiently compensate for high leakage. In certaincases, this compromised the patient’s tolerance.Whereas the most common modes (S, S/T) used inNIV-specific ventilators are equivalent to pressuresupport ventilation (PSV), the most widely usedmodes in CMV ventilators are those which are time-controlled or time-assisted with a predefinedinspiratory time (Ti). This is because in CMVventilators used with PSV the transition frominspiration to expiration occurs when the inspiratoryflow drops below a pre-defined maximum value,which in NIV may be delayed or may not even occurwhen large increases in inspiratory flow are requiredto compensate for a high level of leakage (seeChapter 13, Fig. 1). An excessively high or never-ending Ti will not be well tolerated by patients;

hence, modes in which the Ti is specificallyprogrammed are preferred.

CMV ventilators have other disadvantages whichlimit their utility for NIV. The inevitable leaks inherentto NIV can interfere with key aspects of ventilatorfunction, diminishing patient-ventilator synchronyand increasing the possibility that endotrachealintubation will be required. Choosing an appropriateinterface and attaching it tightly still can not totallyeliminate leakage and may cause patient discomfortor even pressure sores. Leaks can complicateactivation of the inspiratory trigger at the beginningof the cycle, thereby increasing the level ofrespiratory effort required for activation. Likewise,they lengthen the response time of the device,which often leads to auto-cycling. After activationand during gas delivery, leaks reduce the efficacywith which the device achieves pressurization, tothe point that the programmed PIP may not bereached during the Ti. As explained above, usingvariable flow to compensate for leakage at the endof the cycle can delay activation of the expiratorytrigger. This phenomenon tends to be especiallymarked in obstructive RF cases, where it causes aslower drop in inspiratory flow. Contrariwise, inrestrictive RF cases, leaks tend to shorten the Tiexcessively. In both cases patient-ventilatorsynchrony can become compromised. In contrastto CMV ventilators, NIV-specific ventilators (seebelow) are designed with applications that offereffective compensation for a wide array of leaks,adjusting for trigger sensitivity, response time,pressurization performance and cycling criteria tothe extent of leakage.

38 S. Menéndez y cols.

Figure 1. NIV with the Servo 900® ventilator.

Table I summarizes the main features of the CMVventilators with NIV option that are most widelyused in the PICU. The Table includes the variable ofinspiratory pressure-time product (PTPt) andpressure-time product at 500 milliseconds ofrespiratory effort (PTP500). PTPt represents the areaunder the pressure-time curve from the beginningof the inspiratory effort (negative pressure) to thereturn to the starting value by pressurization of thesystem, whereas PTP500 is used to gauge thepressurization performance of the ventilator. PTPt ismeasured in cm H2O per second, and its value isdirectly proportional to the respiratory effort requiredto activate the inspiratory trigger. Low PTPt valuesrepresent less work to activate the trigger in thepresence of leaks. PTP500 is measured relative to anideal pressure (i.e. the target pressure) whose value

is taken as 100%. High PTPt values in the presenceof leakage indicate a high pressurization capacity.

NIV-specific ventilatorsAlthough commercially available NIV respirators

include volumetric models (see Table II), the mostpopular devices employ continuous flowpressometric turbines that cycle with a deceleratingflow and by alternating between two preset pressurelevels. These are generally labeled as bi-level positiveairway pressure (BiPAP) ventilators. Despite havinghigh levels of leakage, they provide effective andwell-synchronized ventilation.

The simplest respiratory support mode offeredby BiPAP ventilators is continuous positive airwaypressure (CPAP). When the level of CPAP must begradually brought above 12 cm H2O, or when

39Non-invasive ventilation devices

Table I. Principal conventional mechanical ventilators featuring an NIV option.

CMV modes NIV modes PTPt PTP500(cm H2O/sec) (% ideal value)

Elisée (ResMed-Saime) (Fig. 2) A/C, SIMV (PCV and VCV) 0.15-0.2 < 10%PS ± VG

Esprit (Respironics) (Fig. 3) A/C, SIMV (PCV and VCV) S, S/T < 0.05 20 to 30%

Evita XL (Drägerwerk) (Fig. 4) SIPPV, SIMV ± ASB, MMV± ASB, BIPAP±ASB, APRV NIV < 0.05 10 to 20%

Extend (Täema) A/C, SIMV (PCV and VCV) VCV, PCV, SIMV, PSV 0.05 to 0.1 20%PSV, APRV, PRVC, MRV

Newport e500 (Newport MI) A/C, SIMV (PCV and VCV) 0.15 to 0.2 20 to 30%PSV, VTPS, VTPC

Servo-i (Maquet) (Fig. 5) VCRP, PC, VC, PS, VS, PS , PC , PS + 0.15 to 0.2 10 to 20%BiVent, Automode and NAVA VNI Backup,

(Neurally Adjusted CPAP Nasal Ventilatory Assist) (500 g- 10 kg)

Servo-s (Maquet) VCRP, PC, VC, PS, PS , PC , PS + 0.15 to 0.2 10 to 20%BiVent PC Backup

Vela (Viasys) A/C, SIMV (PCV and VCV) A/C, SIMV, PSV 0,1 10 to 20%PSV, PRVC

Raphael (Hamilton) A/C, SIMV (PCV and VCV) NIV < 0.05 40 to 50%SPONT, ASV, APRV DuoPAP

840 Ventilator (Tyco-Nellcor A/C, SIMV (PCV and VCV) NIV 0.012 N/PPuritan Bennett) (Fig. 6) PS, Bi level, VV +, PAV+

CMV: conventional mechanical ventilation; NIV: non-invasive ventilation; PTPt: inspiratory pressure-time product; PTP500:pressure-time product at 500 ms from start of inspiratory effort; PCV: pressure-controlled ventilation; VCV: volume-contro-lled ventilation; PS, PSV and ASB: pressure-support ventilation; N/P: information not provided by manufacturer.

40 S. Menéndez y cols.

there is a high level of respiratory effort startingfrom the very beginning, a high level of respiratorysupport is required. The two levels of pressure inBiPAP comprise the inspiratory positive airwaypressure (IPAP) and the expiratory positive airwaypressure (EPAP); the latter is lower than the former.BiPAP ventilators offer two distinct types ofinspiratory cycles: spontaneous (S) and timed (T).S cycles are started by the patient by activatingthe inspiratory trigger and stopped by the patientonce the virtual curve of the ventilator intersectsthe real curve of the patient (see Fig. 7). Incontrast, T cycles are started by the ventilator ata preset frequency (Fr) and stopped after a presetTi.

The most widely used mode in BiPAP ventilatorsis S/T mode, whereby spontaneous modes in whichthe difference between IPAP and EPAP is applied asPRESSURE SUPPORT are alternated with timed modesthat, in the absence of any respiratory effort by thepatient, are applied using a preset Fr. In both theS and the T cycles, the Vt obtained by the patient isa function of any difference between IPAP and EPAPas well as of any respiratory effort.

State of the art devices feature ventilation modesthat support variable levels of pressure support. These

modes include proportional assisted ventilation(PAV®; Respironics) (Fig. 8), average volume assured

Figure 5. NIV with the Servo-i® ventilator, delivered throughnasal-pharyngeal tube.

Figure 3. Esprit® .

Figure 2. Elisée® .

Figure 4. Evita XL®.

Figure 6. Puritan-Bennet 840.

pressure support (AVAPS®; Respironics) (Fig. 9),adaptive servo-ventilation (ASV®; ResMed®) (Fig. 10),and pressure support with tidal volume (PS·TV®;Saime). No comparative studies on the efficacy ofthese modes in pediatric patients have beenpublished to date.

As with CMV ventilators, an important aspect inNIV-specific ventilators is the ability to offset theeffect of leaks on the respiratory effort needed toactivate the inspiratory trigger. The most effectivedevices are those that operate by flow, especially forpatients with obstructive respiratory conditions ortrapped airways, or for whom reaching the presetauto–PEEP threshold may imply additional respiratoryeffort. In terms of the expiratory trigger, both leakagecompensation via additional flow, and obstructiveairway pathologies, can delay the drop in inspiratoryflow below the cycling threshold, therebylengthening the Ti and worsening patient-ventilatorsynchrony. Technological advances have led to thedevelopment and patenting of various sophisticated

triggering systems. One example is the Autotrack®

system by Respironics, which in less than 100 ms

Table II. Principal NIV-specific ventilators

Model NIV modes O2blender Graphic display Setting Accessories

BiPAP Vision® (Respironics) CPAP, S/T, T, PAV Yes Yes Hospital

Supportair® (Airox) VCV (A/C, SIMV), Yes Yes HospitalPCV (C, A/C), PSV (S, ST)

Carina (Dräger) (Fig. 11) CPAP, PS, PC Yes Yes Hospital Internal battery BIPAP, VG

LTV 1000® (Breas) VCV (C, A/C, SIMV), PCV (PSV) Yes Yes Hospital/Home

BiPAP Synchrony® CPAP, S, S/T, T No (optional O2 valve) No Home/Hospital(Respironics)

VPAP III ST-A® (Res Med) CPAP, S, S/T, T No No Home/Hospital Attachable humidifier

VS Ultra® (Saime) PCV (A/C), PSV (S, S/T), No (optional O2 No Home/Hospital BatteryPS·TV, VCV (A/C) valve)

VIVO 40® (Breas) (Fig. 12) CPAP, PSV, PCV No No Home/Hospital Internal battery

BiPAP AVAPS® (Respironics) AVAPS No No Home

AutoSet CS 2® (Res Med) SVA No No Home

BiPAP Serena (Saime) (Fig. 13) S, S/T No No Home

GoodKnight 425 ST BiLevel ® S, APNEA No No Home(Puritan Bennett)

ASV: adaptive servoventilation; AVAPS: average volume-assured pressure support; PS-TV: pressure support with tidal volume;CPAP: continuous positive airway pressure; GV: guaranteed volume; PAV: proportional assisted ventilation; PCV: pressure-con-trolled ventilation; PSV: pressure support ventilation; S: spontaneous; S/T: spontaneous/timed; T: timed; VCV: volume-con-trolled ventilation.

Figure 7. Auto-track® automatic triggering mechanism.

EPAP

PRESSURE

FLOW

Approximate flowin the patiente

Secondary signalPoint of intersection,and activation of the

EPAP cycle

Point of intersection, andactivation of the

IPAP cycle

IPAP

41Non-invasive ventilation devices

(the threshold of conscience inspiratory effort)combines the inspiratory response with a cyclingcriterion that oscillates according to the respiratoryresistance of each patient.

BiPAP ventilators are generally simpler andcheaper than CMV ventilators. For instance, theyfeature tubing with a single inspiratory circuit, whichobviates quantification of the small expiratory tidalvolume that escapes through the inevitable leaksbetween patient and interface. Expiratory gasesescape the system through a constant outflow whichis typically located at the patient-end of the tubing.However, this design does not totally prevent re-inhalation of expired gas. Given that the re-inhaledvolume in pediatric patients can imply a largepercentage of the physiological Vt, re-inhalation canhave major consequences for the patient’s acid-baseequilibrium. These effects can be diminished through

adequate positioning of the controlled outflow inthe interface, by increasing the outflow itself (byincreasing the EPAP), and by using expiratory valvessuch as Plateau (Respironics) valves.

NIV-specific ventilators designed for hospital use(see Table II) typically feature a built-in oxygenblender to facilitate treatment of hypoxemic patientsthat have high FiO2. State of the art BiPAP systems

S. Menéndez y cols.

Figure 12. Vivo 40® .

Figure 11. Carina® .

42

Figure 9. BiPAP Synchrony®.

Figure 10. VPAP III.

Figure 8. BiPAP Vision® with optional PAV mode.

for hospital use are generally equipped with screensfor graphic monitoring of major respiratoryparameters.

NIV-specific devices intended for extra-hospitaluse generally lack an O2 blender but may featurespecific valves for connecting to O2 or can beequipped with an oxygen T-piece, preferably at thenear end of the inspiratory handle in order to avoidturbulence that can bother the patient. However,given that these devices generally employ flowsgreater than 30 L/min, but that the maximum O2

flow offered by conventional flow-meters is 15 L/min,it is difficult to achieve an FiO2 of > 50% whetherusing O2 valves or a T-piece. As such, extra-hospitalNIV ventilators have limited utility for hypoxemicpatients.

Neonatal NIV ventilators used with nasal cannulae

Neonatal NIV requires equipment that isspecifically adapted to the functional and anatomicalneeds of neonatal patients. The Infant Flow® (EME)and Infant Flow Advanced® (EME) systems deliverNIV to neonates through a patented piece thatsimultaneously acts as nasal interface and pressuregenerator (Figs. 14 and 15).

In the Infant Flow® system the aforementionedpiece enables high-velocity bursts of warm andhumidified gas to be generated during inspiration(i.e. the Bernoulli effect), providing active assistanceand nearly constant maintenance of the CPAP level.During expiration the increase in intranasalpressure, together with the pressure generator—which is based on fluid dynamics, fluidic flip andthe Coanda effect—directs said bursts of gas andthe expired gas towards the expiratory branchof the piece with a minimum response time andwithout producing any major oscillations in CPAP.The small size and long-lasting battery of the Infant

Flow® system make it amenable to early perinataluse and enable uninterrupted ventilation duringtransport between the delivery room and theneonatology ward.

The Infant Flow Advanced® includes the optionto add additional gas to the base volume atprogrammable values of 0 to 5 L/min (flow), 1 to 120respirations/min (respiratory frequency, f), 0 to 11cm H2O (pressure) and 0.1 to 1 sec (Ti). This feature,together with the specific impedance trigger activatedby the patient’s abdominal muscles, extends theventilation modes possible with the Infant Flow systemto include non-synchronized and synchronizedintermittent positive pressure ventilation (IPPV andSIPPV). The reliability of said triggering system islimited, given that it can be inadvertently activatedby non-respiratory abdominal muscle movements.Lastly, use of the trigger as an apnea sensor enablesprogramming of a safety mode (backup) at respiratoryfrequencies adjustable from 0 to 30 rpm. The latestInfant Flow model is known as SiPAP® (Fig. 16).

The Giulia® (Ginevri) ventilator (Fig. 17) provides

Non-invasive ventilation devices 43

Figure 13. Serena® .

Figure 14.Top: Structure of the pressure generator used inthe Infant Flow® and Infant Flow Advance® systems. Bottom:Characteristic change in flow from inspiration to expiration(i.e. fluidic flip) of these systems (see text for more details).

EXPIRATORYLIMB

NASALINTERFACE

EXPIRATORYLIMB

PRESSURETRANSDUCER

EXPIRATORYLIMB EXPIRATORY

LIMB

FRESH GASSTREAM

CONNECTION TONASAL INTERFACE

(PATIENT)

INSPIRATION EXPIRATION

CONNECTION TONASAL INTERFACE

(PATIENT)

FRESH GASINLET

CPAP, SIPPV and synchronized intermittentmandatory ventilation (SIMV). It employs a flowtrigger that is connected to nasal cannulae andwhose activation is not interrupted by non-respiratory abdominal muscle movements. Itsgraphic display shows pressure and flow values andrates.

Portable ventilators with an NIV modePortable ventilators used during transport or

emergency situations are also becoming evermoresophisticated: in some cases they now offer the samelevel of performance as the devices used in the PICU.Noteworthy examples include the Oxylog 3000®

(Dräger), LTV1000® (Pulmonetic Systems) and Osiris2 and 3 (Taema), whose leak alarms can be modifiedor turned off, and which compensate for leaks,allowing mask delivery of NIV to patients weighingapproximately 5 kg. The Oxylog 3000®, which onlyoffers NIV in the pressure-controlled modes of BIPAP,BIPAP/ASB, CPAP and CPAP/ASB, employs aninspiratory sensor with programmable flow (> 3L/min). NIV during transport can also beadministered with simple CPAP devices such as theBoussignac (see Chapter 12).

Other NIV methods

Phrenic electrostimulationPhrenic electrostimulation, also known as

diaphragmatic tracking or diaphragmatic stimulation,can be considered as a type of NIV which does notrequire a permanent airway. It consists of positioningelectrodes that electrically stimulate one or both ofthe phrenic nerves, causing the diaphragm to contract,and consequently, generating an inspiratory flow. Abdominal compressors (pneumobelts)

Abdominal compressors, also known aspneumobelts, are inflatable belts that are placedin between the patient’s pubis and navel and thenconnected to a positive pressure ventilator. Uponinflation, the belt generates a pressure that movesthe diaphragm upwards, thereby favoring expiration;upon deflation, it favors inspiration.

S. Menéndez y cols.44

Figure 15. Infant Flow Advance®.

Figure 17. Giulia. Ginevri®.

Figure 16. SiPAP® .

Rocking bedsRocking beds allow patients to alternate between

the Trendelemburg and anti-Trendelemburgpositions, facilitating passive movement of thediaphragm by gravity (alternating between thecaudal and cephalic directions), thereby aidinginspiration and expiration. Rocking beds enablerespiratory frequencies of approximately 30respirations/min and Vt of approximately 500 mL.

REFERENCES1. Hill NS. Noninvasive ventilation for chronic obstructive pulmo-

nary disease. Respir Care 2004;49:72-87.

2. Hess DR. The evidence for noninvasive positive-pressure ven-tilation in the care of patients in acute respiratory failure: asystematic review of the literature. Respir Care 2004;49:810-29.

3. Peter JV, Moran JL, Phillips-Hughes J, Warn D. Noninvasive ven-tilation in acute respiratory failure--a meta-analysis update. CritCare Med 2002;30:555-62.

4. Bernstein G, Knodel E, Heldt GP. Airway leak size in neonatesand autocycling of three flow-triggered ventilators. Crit CareMed 1995;23:1739-44.

5. Calderini E, Confalonieri M, Puccio PG, Francavilla N, Stella L,Gregoretti C. Patient-ventilator asynchrony during noninvasiveventilation: the role of expiratory trigger. Intensive Care Med1999;25:662-7.

6. Prinianakis G, Delmastro M, Carlucci A, Ceriana P, Nava S. Effectof varying the pressurisation rate during noninvasive pressuresupport ventilation. Eur Respir J 2004;23:314-20.

7. Schettino GP, Tucci MR, Sousa R, Valente Barbas CS, Passos AmatoMB, Carvalho CR. Mask mechanics and leak dynamics duringnoninvasive pressure support ventilation: a bench study. Inten-sive Care Med 2001;27:1887-91.

8. Peerless JR, Davies A, Klein D, Yu D. Skin complications in theintensive care unit. Clin Chest Med 1999;20:453-67.

9. Richard JC, Carlucci A, Breton L, Langlais N, Jaber S, MaggioreS et al. Bench testing of pressure support ventilation with threedifferent generations of ventilators. Intensive Care Med2002;28:1049-57.

10. Tassaux D, Gainnier M, Battisti A, Jolliet P. Impact of expiratorytrigger setting on delayed cycling and inspiratory muscle wor-kload. Am J Respir Crit Care Med 2005;172:1283-9.

11. Nava S, Ambrosino N, Bruschi C, Confalonieri M, Rampulla C.Physiological effects of flow and pressure triggering during non-invasive mechanical ventilation in patients with chronic obs-tructive pulmonary disease. Thorax 1997;52:249-54.

12. Ferguson GT, Gilmartin M. CO2 rebreathing during BiPAP ventila-tory assistance. Am J Respir Crit Care Med 1995;151:1126-35.

Non-invasive ventilation devices 45

INTRODUCTIONAlthough non-invasive ventilation (NIV) can be

delivered using either positive or negative pressure,the techniques used to administer each form arevery different. This chapter provides an overview ofthe modes and methods for positive pressure NIV(NPPV), which is the most common type of NIV inboth adult and pediatric patients.

Until only a few years ago, NIV was performedusing continuous positive airway pressure (CPAP)systems or NIV-specific ventilators that offered fewventilation modes and were primarily indicated forobstructive sleep apnea (OSA). When ICU personneldid not have an NIV ventilator available, theyadapted conventional invasive ventilators toadminister NIV using volume-controlled or pressure-controlled methods. NIV delivered this way oftenfailed, especially in very small children, because theventilators and materials did not meet the specificneeds of pediatric patients. However, in the past fewyears, new ventilation modes suitable for NIV havebeen developed and existing ones have beenimproved. Moreover, new ventilators for homeventilation of pediatric patients, capable of bothinvasive and non-invasive ventilation, have beenintroduced. Lastly, many of the latest systems forinvasive ventilation now feature NIV modes.Consequently, the range of indications of pediatricNIV has expanded. The recent development of highflow oxygen therapy, together with CPAP and NPPV,has provided an array of new treatment strategies

for pediatric patients with respiratory failure (RF).Physicians can now choose among options of varyingdegrees of aggressiveness before resorting tointubation.

This chapter provides an overview of the currentNIV methods of high flow oxygen therapy, CPAPand NPPV.

HIGH FLOW OXYGEN THERAPY

PrincipleIn high flow oxygen therapy a strong flow of

air—ideally stronger than the peak inspiratory flowof the patient—is mixed with oxygen, and thendelivered through nasal cannulae. The gas ishumidified (95 to 100% relative humidity) andwarmed to near body temperature (33 to 43 °C) tomake it more tolerable.

Mechanism of actionThe high flow is believed to create a small

positive pressure (i.e. CPAP) gradient that improvesoxygenation and decreases respiratory effort;however, this has yet to be fully demonstrated.Regardless of oxygen concentration, the humidityin the air has benefits the body by improving ciliarymovement and removal of secretions. Lastly, usingairflow similar to the insufflation of tracheal gascleanses the anatomic dead space, therebyimproving ventilation.

Non-invasive ventilation modes and methods for pediatric patients

J. López-Herce and J. Pilar

Chapter 7

Interfaces • Nasal cannulae of different sizes (according to

age). Conventional nasal cannulae can beemployed as long as the same caliber and lengthrecommended by the manufacturer are used.

• Can be used in tracheotomized patients.

Indications• Patients with high oxygen requirements .• Moderate respiratory failure (RF).• Apnea in premature infants.

Commercially available systems in SpainOf the numerous high flow oxygen systems

produced, there are currently two available in Spain:Fisher-Paykel (Fig. 1) and Vapotherm (Fig. 2). Thelatter can deliver flows of approximately 1 to 40L/min. These models feature two modes: low flow(1 to 8 L/min: for premature infants, neonates andsmall infants) and high flow (5 to 40 L/min: forchildren and adults). Each mode uses differentcartridges for humidifying the gas. The nasal cannulaeused are sized according to the patient’s age.

Programming the flowThe flow is programmed according to tolerance:Premature infants, neonates and small infants:

low flow (1 to 8 L/min).

Children: approximately 5 to 20 L/min. Adults: approximately 8 to 40 L/min.

Advantages• Simple and easy to use, and well-tolerated by

the patient.• Suitable for any age.• The air can be enriched with nitric oxide or

heliox.

RisksInfection: In order to minimize the risk of

infection, hospital staff must closely follow theprotocols for changing the interfaces and cleaningthe system. This risk has been reduced in newermodels which use disposable interfaces.

CONTINUOUS POSITIVE AIRWAY PRESSURE

Principle and mechanism of actionIn CPAP, which is the simplest NIV mode, a

continuous positive pressure is maintainedthroughout the respiratory cycle. This pressure isdelivered via continuous airflow and/or a pressurevalve, enabling the patient to breathe spontaneously(Fig. 3). Spontaneous breathing• The child determines the respiratory frequency

and Vt according to their breathing work.• The continuous pressure in CPAP keeps the

airway open, increases the respiratory functionalcapacity and reduces alveolar collapse.

48 J. López-Herce Cid, J. Pilar Orive

Figure 1. Fisher-Paykel.

Figure 2. Vapotherm.

InterfaceCPAP can be administered through any interface

(e.g. nasopharyngeal tube, nasal or oral-nasal mask,nasal cannulae, and short nasal prongs).

Delivery methodsCPAP can be delivered using various systems and

ventilators. However, it should be remembered thatrespiration and programming differ among devices.

CPAP systems: These basically comprise an airand oxygen flow source, tubing and an expiratoryresistance (positive end expiratory pressure [PEEP])valve. An air heater and humidifier are connectedto the inspiratory tubing. The patient can freely drawin air. In these systems neither the volume nor therespiratory frequency are monitored; hence, a flowmeter must be connected to the tubing to measurethe pressure. CPAP systems can either be assembledfrom separate components available in the ward(Fig. 4) or purchased fully pre-assembled. There arealso commercially available simplified CPAP systemswhich can be used during transport and which donot include an air heater and humidifier.

NIV ventilators: These generally only maintainCPAP using continuous flow in the airway. Thepatient can freely draw in air.

Continuous flow neonatal ventilators: Thesemaintain CPAP with continuous flow in the airwayand with a PEEP valve. The patient can freely drawin air.

Conventional invasive ventilators: These maintaincontinuous pressure in the airway via a PEEP valve,but generally do not maintain a continuous airflow.In order to draw in air, the patient must open theinspiratory sensitivity valve at each breath. Thisrequires greater respiratory effort than do the othersystems. This problem can be solved by introducingcontinuous airflow by connecting a T-piece to the

tubing at the inspiratory loop of the ventilator.However, this modification can alter pressure andvolume readings.

An ideal CPAP system would administercontinuous flow, such that the patient could breatheand draw in air at will, and would monitor at leastvolume, respiratory frequency, and pressure. Theairflow would have to be strong enough for the

49Non-invasive ventilation modes and methods for pediatric patients

Figure 3. Plot of time vs. pressure for CPAP (P: pressure).

Figure 5. Bubble CPAP system.

Figure 4. Benveniste system assembled from separate com-ponents.

Spontaneous breathsP

4

0

patient to draw in air without limits (i.e. at least twoto three times the normal minute volume), but notso strong as to cause the patient discomfort.

Indications• Obstructive sleep apnea syndrome (OSAS).• Central and/or obstructive apneas in premature

infants with bronchiolitis.• Bronchiolitis with respiratory failure.• Respiratory failure (without alveolar

hypoventilation): pneumonia, chest wall trauma,cardiogenic pulmonary edema.

• Post-extubation respiratory failure.• Respiratory failure with weak respiratory effort

(if BiPAP is not available).

Programming• Parameters to program:

– CPAP systems: flow of air and/or oxygen andregulation of the expiratory resistance valve.

– NIV-specific ventilators or conventional(invasive) ventilators with an NIV option: CPAPor EPAP, and FiO2 (if available), and alarms forvolume, pressure, frequency and apnea (onlyavailable in certain models)

– Neonatal ventilators: flow, PEEP, FiO2, andalarms for volume, pressure, frequency andapnea. The minute volume and apnea alarmsmust often be set at the minimum to preventthem from continuously going off due toleakage.

– Conventional invasive ventilators: inspiratorysensitivity, PEEP, FiO2, and alarms for minutevolume, pressure and apnea.

• Starting parameters: Low CPAP levels(approximately 4 to 6 cm H2O) are generallyrecommended for initiating ventilation.

• Adjustments: The CPAP level should beadjusted in increments of 2 cm H2O accordingto the patient’s tolerance and respiratory needs.Levels above 12 cm H2O are usually not welltolerated. If the patient does not show any signsof improvement, hospital staff should considerswitching to another NIV mode.

Advantages• Simple, easy to use and adjust, and can be

delivered with a simple apparatus .

Disadvantages and secondary effects • Does not assist the patient with each breath.• Does not provide safety breaths for patients who

are hypoventilating or have constant apnea;hence, monitoring of cardio-respiratory signalsand of transcutaneous oxygen levels is crucial.

• Over-inflation: Use of excessive CPAP can leadto pulmonary over-inflation (infants withbronchiolitis are at especially high risk) anddiminished veinous return.

NIV MODES IN CONVENTIONAL INVASIVEVENTILATORS

Types Non-invasive mechanical ventilation can be

administered using conventional invasive ventilatorsunder volume-control or pressure-control. Nearlyany ventilation method or mode can be used:controlled, assisted/controlled, intermittentmandatory ventilation (IMV), pressure supportventilation (PSV) and CPAP. The respectiveadvantages and disadvantages of each method ormode to deliver NIV are similar to those for deliveringinvasive ventilation.

ProgrammingProgramming of invasive ventilators for NIV is

similar to that used for patients with endotrachealintubation, the parameters of which are summarizedin Table I. However, there are certain differences thatmust be considered:

Volume: In volume modes, a greater Vt (up totwice as large) can be programmed for NIV than forinvasive ventilation in order to compensate for thedead space of the mask and for any leaks.


Figure 6. Hudson system.

Respiratory frequency: For the patient tomaintain spontaneous breathing, good synchronyis crucial: setting the respiratory frequency and theinspiratory to expiratory ratio very close to thepatient’s actual values indirectly provides aninspiratory time similar to that of the patient.Patients typically have very short inspiratory times(0.2 to 0.5 seconds); hence, there is no rationalefor artificially extending the inspiratory time, whichin turn increases the risk of asynchrony. Someventilators feature the option to independentlylimit the inspiratory time.

Inspiratory sensitivity: Flow-sensitivity methodsare generally better than pressure-sensitivitymethods. For NIV in conventional invasive ventilators,adjusting the sensitivity is crucial: given the deadspace of the mask and the existence of leaks, flow

sensors, such as those for pressure, may not detectthe patient’s inspiration or may detect it late. Theseproblems could lead to ventilator-patient asynchronywith respiratory exhaustion and risk of barotrauma.Furthermore, in CPAP and PSV modes, if the patientis unable to open the ventilator valve, then they willnot draw in any air, and therefore, will functionallyremain in apnea with a severe risk of respiratoryarrest. In assisted/controlled methods and PSV andSIMV-pressure support modes, if the trigger is toosensitive, then autocycling can occur and possiblycompromise ventilation by causing patient-ventilatorasynchrony.

Expiratory sensitivity: regulates the end of theventilation cycle. Depending on the ventilatorand the mode used, the end of inspiration maybe:


Table I. Starting values of NIV parameters for invasive ventilators.

Peak pressure or pressure support*: 8 to 10 cmH2O Once the patient tolerates the mask well, this value should be increased in increments of 2 cm H2O until reachingthe peak value (typically 14 to 20 cm H2O) that provides the best improvement of respiratory failure without pro-voking intolerance or generating leaks.

Tidal volume**: 15-30 mL/kg. Low values (10 to 12 mL/kg) should be used initially, and then increased according to the patient’s chest wall expan-sion and tolerance.

PEEP: 4 cmH2O. For patients with hypoxemia and/or atelectasis, this value should be increased to a maximum of 6 to 10 cm H2O.

Inspiratory time or percent: 0.2 to 0.5 sec or 33%.

Respiratory frequency: 2 to 5 respirations/min less than the patient's respiratory frequency***.

FiO2: 0.21 to 1 according to the pathologyFor ARF patients, high values can be used initially and then decreased in function of the saturation.

Ramp slope, flow speed, and inspiratory delay: These are adjusted based on the patient’s tolerance. As a general rule,the younger the patient, the lower the flow or the slower the ramp during the adaptation phase****.

Inspiratory sensitivity: maximum sensitivity; Flow of 1 to 2 L/min or Pressure of 0.5 to 2 cm H2O (flow trigger is pre-ferred).

Expiratory sensitivity****: 40 to 70% of the maximum inspiratory flow, depending on leakage.

Alarms: for apnea, respiratory frequency and pressure; The expiratory volume alarm should be turned off for very lowvalues caused by leakage.

*For PCV, and pressure-assisted/controlled ventilation (PACV), pressure-controlled SIMV (PC-SIMV), and PSV modes.**For volume-controlled ventilation, including volume-assisted/controlled ventilation (VACV), and volume-controlled SIMV (VC-SIMV) modes.***For controlled, assisted/controlled, and SIMV modes.**** Not programmable on all ventilators.

a. A fixed value set by the operator (e.g. byprogramming the inspiratory time in controlled,assisted/controlled or IMV modes).

b. A fixed value set by the manufacturer (e.g. 25%of the inspiratory peak value in the pressuresupport).

c. Hence, just as in invasive ventilation, theexpiratory trigger tends to be adjusted as closelyas possible to the extubated patient'sphysiological condition: it is programmed toapproximately 5 to 25% of the achieved peakflow (provided there is no leakage). In NIV,ventilators try to compensate for constantleakage, and the decrease in peak flow is much slower. This translates into an undesiredprolongation of the inspiratory time, causing thepatient discomfort. Consequently, the expiratroytrigger should be programmed at a higher value(40 to 70% of the achieved peak flow),according to the inspiratory time that the patientwould use without NIV (see Chapter 13, Fig. 1).

IndicationsInvasive ventilators in NIV mode are indicated

for hospital use when volume-controlled or pressure-controlled methods are required, and when an NIVventilator with built-in oxygen blender is eitherunavailable or unable to provide good ventilationor adaption to the patient.

Advantages and disadvantages of each methodand mode• In volume-controlled methods, the pressure is

variable and there is a higher risk of leakage,with lower tolerance by the patient as well asgastric distension, and skin irritation and/ornecrosis.

• Pressure-controlled methods generallycompensate better for leaks and are bettertolerated by the patient. Moreover, thedecelerating flow distributes air better. However,in these methods the Vt is variable, and therefore,if patient resistance or compliance changes, thensufficient ventilation can not be guaranteed.

• PSV can provide better adaptation to the patient.However, leakage in PSV can extend theinspiratory flow even after the patient has startedexpiration, thereby forcing the patient to increasetheir respiratory effort in order to counteract the

ventilator. This scenario can be avoided by usingpressure-control, whereby the inspiratory timeis limited by programming, which is thepreferred method of many physicians.

Disadvantages of using invasive ventilators for NIV • Invasive ventilators, whether used with pressure-

controlled or volume-controlled methods,generally offer worse adaptation in BiPAP modethan do NIV ventilators.

• Invasive ventilators, whether used with pressure-controlled or volume-controlled methods, donot provide sufficient leak compensation.

• In invasive ventilators the alarms for low minutevolume and for apnea often sound continuouslydue to dead space in the mask and to leaks. Insome models, these alarms can not be turned off.

• Invasive ventilators are generally harder tomaintain than NIV ventilators.

VENTILATION WITH CONTINUOUS FLOW ANDBI-LEVEL POSITIVE AIRWAY PRESSURE

This mode is available in NIV ventilators and insome conventional ventilators that have an NIVoption.

PrincipleIn this mode, a turbine generates pressure at

two levels (inspiratory positive airway pressure [IPAP]and expiratory positive airway pressure [EPAP]) witha continuous flow throughout the respiratory cycle.Although it is traditionally known as BiPAP (bi-levelpositive airway pressure), there are legal issuesconcerning use of this acronym; hence, this modeis given a different name by each manufacturer. Itis the most widely used NIV mode for all types ofpatients and clinical scenarios. The ventilatorconstantly monitors the patient’s respiratory effort,using a highly sensitive flow sensor attached to thecircuit, which enables synchronization withspontaneous breathing as well as compensation forany leaks in or near the mask.

Parameters to program (see Table II)• IPAP: controls the ventilation; During the

inspiratory phase a higher IPAP generates agreater tidal volume (Vt).


• EPAP: improves the residual functional capacityand oxygenation.

• Inspiratory ramp slope or flow speed: regulatesthe speed at which air enters; These parametersare crucial for ventilator-patient adaptation.Longer ramp times and/or slower flows lead toslower entry of air, consequently enabling thepatient to better adapt at the onset of ventilation.Not all ventilators feature programmable rampslope or flow speed.

• FiO2: For ventilators that are not equipped withan oxygen blender, but that do have an oxygenvalve (e.g. BiPAP synchrony®, VS Ultra®), the FiO2

delivered can be determined from themanufacturer’s data table. For ventilators inwhich oxygen is added through the tubing ormask, the FiO2 can be calculated based on theairflow delivered by the ventilator and the flowof added oxygen. However, it is difficult toachieve FiO2 > 50% in these ventilators.

• Respiratory frequency: This parameter is onlyactivated as rescue parameter when the detectedrespiratory frequency falls below theprogrammed value. If the programmed value istoo close to the patient’s actual value, theventilator could interfere with the patient’srespiratory rhythm. The majority of NIV-specific

ventilators can not detect the inspiratory effortsof infants, especially those younger than 3months. This problem can be solved by settingthe respiratory frequency and the inspiratorytime close to the patient’s actual values, in hopesthat the child adapts to the programmedrhythm.

• Inspiratory time (as time or percent): controlsthe duration of the IPAP in rescue breaths.

• Alarms for apnea, volume and pressure.

Bi-level pressure ventilation modes• The oldest BiPAP ventilators featured different

ventilation modes:– CPAP: the ventilator maintains a continuous

flow that allows the patient to breathe freely.– Spontaneous (S) mode: equivalent to pressure

support with constant flow; The ventilatormaintains a CPAP (EPAP) until the patientdraws in, at which point it switches to apressure support (IPAP). The frequency andduration of the inspiration are controlled bythe patient. The patient triggers all inspirations,and the ventilator helps the patient with eachbreath.

– Spontaneous/timed (S/T) mode: the ventilatoracts as pressure support for the patient’s


Table II. Starting values of NIV parameters for bi-level pressure ventilators

IPAP: 8 to 10 cm H2O.Once the patient tolerates the mask well, this value should be increased in increments of 2 cm H2O until reachingthe peak value (typically 10 to 20 cm H2O) that provides the best improvement of respiratory failure without pro-voking intolerance or generating leaks.

EPAP: 4 cm H20*

Inspiratory time or percent: 0.4 to 0.5 sec or 33%**.

Respiratory frequency: 10 respirations/min less than the patient's respiratory frequency.

FiO2**: according to pathology.

Ramp slope, flow speed, and inspiratory delay****: ramp slope: 0.05 to 0.4 sec; or flow speed: 10 to 30 L/min accor-ding to age (in the adaptation phase: the younger the patient, the lower the flow).

Alarms: for apnea, respiratory frequency, tidal volume, minute volume, and pressure*****

*In some bi-level pressure ventilators, the minimum programmable EPAP is 4 cm H20.**Only in S or S/T modes.***Some bi-level pressure ventilators (e.g. BiPAP Vision®, and Carina®) feature a built-in oxygen inlet, whereas in others, theoxygen flow must be introduced through the tubing or the mask.**** Not programmable on all ventilators.***** Depends on ventilator type.

spontaneous breaths. If the patient does nottrigger a minimum number of respirations,then the ventilator runs a cycle comprising anIPAP, an inspiratory time and an EPAP at aprogrammed frequency. S/T is the most widelyused bi-level mode for both acute respiratoryfailure (ARF) and chronic respiratory failure(CRF) patients.

– Timed (T) mode: the ventilator delivers aprogrammed number of breaths regardless ofthe patient’s respiratory efforts. The authorsof this chapter believe that this mode shouldonly be used if the ventilator can not detectthe patient’s breaths, as commented above.

• The latest NIV models and conventionalventilators with NIV option only feature twomodes: CPAP and an S/T mode, which guaranteea minimum number of respirations if therespiratory frequency falls below a preset value.For spontaneous breathing, pressure supportis delivered until the IPAP is reached.

IndicationsS/T mode tends to be the preferred bi-level

mode for initiating NIV in nearly all cases; however,certain pathologies can initially be treated with CPAP(e.g. acute pulmonary edema [APE]).

Advantages • For invasive ventilators S/T bi-level modes

generally provide better adaptation than volume-controlled or pressure-controlled methods.

Disadvantages• Some pediatric patients, especially very young

(small) children, can not tolerate the flowrequired to reach the required pressure.

• Certain modes and ventilators can not regulatethe pressure ramp.

MIXED VENTILATION MODESBoth conventional and NIV ventilators now often

feature mixed mode options for ventilation whichis controlled by volume but regulated or cycled bypressure. These mixed modes include averagevolume assured pressure support (AVAPS), found inthe Respironics BiPAP® devices, and Draëger’sguaranteed volume (GV).

PrincipleIn this mode a target Vt is programmed. The

ventilator measures the patient’s expiratory Vt andthen modifies the IPAP breath by breath to reachthe target value. This mode can be used with S, S/Tor assisted/controlled methods.

Parameters to program• Target Vt: 200 to 1,500 mL.• Minimum IPAP: at least 4 cm H2O above the

EPAP.• Maximum IPAP: up to 30 cm H20; The ventilator

automatically gradually adjusts the IPAP deliveredto the patient between the minimum andmaximum values to reach the target Vt. Thechange in IPAP between breaths is less than 1cm H2O.

• EPAP: improve the residual functional capacity(RFC) and oxygenation.

• Inspiratory time (as time or percent).• Slope of the inspiratory ramp or flow.• Expiratory tidal volume alarm (can be turned

off).

IndicationsOlder children and adolescents with

hypoventilation.

Advantages • Ensures ventilation with low risk of hypoventilation

or hyperventilation.• Provides better adaptation to the patient.

Disadvantages• For patients who have major leaks, the expiratory

Vt does not accurately reflect the real Vt.• Given that the minimum programmable Vt is

200 mL, this mode can only be administered toolder children.

• There is very little clinical experience with thismode in NIV for adult patients and no experiencewith pediatric patients.

PROPORTIONAL ASSISTED VENTILATION

PrincipleProportional assisted ventilation (PAV) is a

partially synchronized mode in which the ventilator


delivers a variable pressure support and flow whichare instantly adjusted according to the patient’srespiratory effort.• The ventilator continuously measures the volume

and flow generated by the patient through apneumotacometer connected to the tubing andthen calculates the elasticity and resistance.

• The ventilator administers a pressure and a flowto compensate for the elasticity and resistance.This support is directly proportional to thepatient’s inspiratory effort.

• The ventilator can amplify the patient’srespiratory effort at a preset proportion, withoutthe need for a preset volume or pressure (flow,pressure, and volume are not programmed inPAV). This mode acts as an additional musclewhich is controlled by the respiratory effort ofthe patient, who determines the respiratoryfrequency and the depth of each breath.

• Just as in PSV, in PAV the patient determines therespiratory pattern (respiratory frequency andinspiratory time), and the patient's effortcontributes to the entry of air.

Fundamentals • In spontaneous breathing the pressure generated

by the respiratory muscles serves to overcomethe forces of elasticity (E) and resistance (R) inthe respiratory system. Elasticity is proportionalto the volume of air introduced into the lung (V),whereas resistance is proportional to the flowspeed (F). The pressure of the respiratory muscles(Pmusc) is equal to the product of the elasticityand volume plus the product of resistance andthe flow: Pmusc = (E x V) + (R + F).

• PAV delivers respiratory aid through flowassistance (FA), measured in cm H2O/L/sec, andvolume assistance (VA), measured in cm H2O/L.In PAV the pressure created by the respiratorymuscles is equal to the sum of the elasticity andresistance minus the pressure by the ventilator,such that: Pmusc = V x (E–VA) + F x (R–FA).

Parameters to programPAV ventilators offer an array of ventilation

options to choose from according to the patient’spathology: 1. Programmed ventilation: all parameters are

programmed by the operator.

• Flow assistance (FA) compensates for resistanceby increasing the Vt while decreasing therespiratory frequency. The most widely usedlevel for adults is 2 cm H2O/L/sec (range: 0 to12 cm H2O/L/sec).

• Volume assistance (VA) compensates forelasticity by increasing the Vt while reducingthe patient's respiratory effort. 5 cm H2O/L(range: 0 to 25 cm H2O/L).

• Percent assistance (for elasticity and resistance)delivered by the ventilator: 100%, althoughsome devices start at 20% and then increaseuntil reaching the maximum value toleratedby the patient.

• PEEP: 4 cm H20.• Maximum pressure.• Maximum Vt.• Alarms for pressure and volume.• Sensitivity: automatically set in the BiPAP

Vision®.2. The operator chooses the patient’s respiratory

pattern (e.g. normal, obstructive, restrictive ormixed), the percent assistance, the PEEP, themaximum pressure, and the maximum Vt, andthe ventilator automatically adapts the assistedflow and assisted volume according to thepatient’s respiratory effort and the measuredelasticity and resistance.

Steps for programming proportional assistedventilation

The most difficult aspect of PAV is the initialprogramming, since, theoretically, the elasticity andresistance of the patient’s respiratory system shouldbe measured before beginning. If these values arenot known, there is a risk of overestimating orunderestimating the patient’s respiratory needs.However, studying respiratory mechanics in asedated, non-intubated sedated patient is verycomplicated. Several methods have been reportedfor adults, although the majority of these are rathercomplex—above all, in the critical patient. Thesemeasurements are even more difficult to obtain inpediatric patients.

In clinical practice1. If a respiratory pattern is selected (e.g. normal,

obstructive, restrictive or mixed), then thepercent assistance is programmed. Some authors


have reported starting at 20% and then graduallyincreasing every 35 min (to 40, 60, 80, 90 and95%), until reaching maximum assistance, oruntil respiratory assistance has improved (i.e.dyspnea and respiratory frequency have bothdecreased), or until the patient shows signs ofintolerance. Others start directly with 80 to 100%assistance.

2. If programmed ventilation is chosen, then theparameters are set as follows:• Flow assistance (FA): a starting value of 1 cm

H2O/L/sec is used, and then the value isincreased in increments of 1 cm H2O/L/sec.

• Volume assistance (VA): a starting value of 2cm H2O/L is used, and then the value isincreased in increments of 2 cm H2O/L untilexcessive pressure is reached (i.e. patientdiscomfort, leakage and highly extendedinspiratory time), at which point the value isdecreased until well tolerated by the patient.

• Some authors have reported use of thefollowing fixed values (which are later adjustedaccording to the patient’s needs): FA of 2 cmH2O/L/ sec and VA of 5 cm H2O/L.

• Percent assistance: a starting value of 20 to30% is used, and then the value is graduallyincreased as described above.

Adjustments to ventilation assistanceThe parameters for ventilation assistance should

be adjusted for the clinical scenarios listed below.• Hypoventilation: the ventilator values for elasticity

and resistance should be increased.• Hyperventilation: the ventilator values for

elasticity and resistance should be decreased.• Hypoxemia: the FiO2 and/or PEEP should be

increased

Indications and efficacy• PAV has proven utile in both invasive ventilation

and NIV for adult ARF and CRF patients. • The indications of PAV in NIV are the same as

PSV: RF due to a disease of the airway or to lungparenchyma. Patients must maintain a normalrespiratory frequency and minimum breathingwork.

• Owing to the scarce amount of clinicalexperience with PAV in pediatric patients, noclear pediatric indications have yet been

established. Currently, PAV can be used incooperative older children for whom elasticityand resistance can be measured and/or in whomthe response to changes in FA and VA can bewell evaluated.

Comparison with other ventilation modesOnly a few studies comparing the efficacy and

side effects of PAV to those of other NIV modes havebeen published. The majority of these did not findany differences in patient mortality, length ofmechanical ventilation, frequency of intubation, orsecondary effects. Some authors have reported thatfor invasive ventilators, PAV requires lower pressureand provides a quicker drop in respiratory frequencythan does PSV. Moreover, the flow adjusts betterto the patient, providing better tolerance andrequiring fewer modifications. Nonetheless, morestudies comparing PAV to other NIV modes areneeded.

Advantages • In PAV, the ventilator instantly adapts to the

patient’s respiratory effort, providing ventilationclose to spontaneous breathing, which in turnenables better patient adaptation and comfort.

• PAV can reduce the patient’s breathing workusing less pressure than do other ventilationmodes.

Disadvantages and side effects• PAV is hard to program at the beginning of

ventilation due to the difficulty in measuringelasticity and resistance, especially in criticalpatients, and above all, in children (the youngerthe patient, the more difficult the measurement).

• Asynchrony: the phenomenon of escape—whereby the respirator continues deliveringairflow even after the patient has completed theirinspiration—has been reported in adult patientswhen an FA of 80% and a VA of 45% were used,despite the fact that these values were lowerthan the patients’ measured resistance andelasticity, respectively. Studies using models haverevealed that this problem is due to a lack ofexpiratory sensitivity (i.e. the ventilator can notdetect when the patient has begun expiration).This problem can lead to asynchrony, discomfortand hyperinflation.


• Risk of hypoventilation: PAV generates greatervariability of the controlled volume than doesPSV. If the patient reduces their respiratory effortby taking shallow breaths, then the ventilatoronly provides minimal assistance.

• Currently there is almost no clinical experiencewith pediatric PAV.

• PAV is only available in some invasive ventilatorsand some NIV ventilators.

INDICATION OF NON-INVASIVE VENTILATIONMODES ACCORDING TO PATHOLOGY

Among NIV modes, double pressure modes withcontinuous flow seem to be the best tolerated bypatients. However, there have been very fewcomparative studies on the efficacy and toleranceof NIV modes, none of which have dealt withpediatric patients.

Chronic respiratory failure• For adult CRF patients, including those with a

heightened condition, assisted/controlled, PSV,S/T and PAV modes have all been shown toimprove minute ventilation, respiratory frequencyand arterial gas levels. In some cases volume-controlled and pressure-controlled methods havebeen reported to be better at decreasingrespiratory effort than PSV.

• For pediatric CRF patients the NIV mode ormethod used varies according to the pathology:for cystic fibrosis is most often treated with PSV;restrictive pulmonary disease and centralhypoventilation, with volume-controlledmethods; and OSA, with CPAP or bi-levelpressure ventilation.

Cardiogenic acute pulmonary edemaThe most widely used modes for cardiogenic

acute pulmonary edema are CPAP and S/T, as theydecrease the respiratory frequency, correct respiratoryacidosis and improve hemodynamics.

Acute respiratory failureFor adult and pediatric ARF patients, bi-level

pressure (S and S/T), assisted/controlled, PSV, andPAV (only studied in adults) have all been shownto improve minute ventilation, reduce respiratoryfrequency and breathing work and improve blood

gas levels, thereby lowering intubation rates. Nostudy has shown any of these modes to besuperior.

REFERENCES

1. Pons Odena M, Cambra Lasaosa FJ. Ventilación no invasiva. EnGrupo de Respiratorio de la SECIP ed, Manual de Ventilaciónmecánica en pediatría. Publimed, Madrid, 2003:95-107.

2. García Teresa MA, Martín Barba C. Ventilación no invasiva. EnLópez-Herce J, Calvo C, Lorente M eds. Manual de CuidadosIntensivos Pediátricos. Publimed 2001:655-60.

3. Martínez Carrasco C, Barrio Gomez de Aguero I, Antelo LandeiraC, Díaz Lobato S. Ventilación domiciliaria vía nasal en pacientespediátricos. An Esp Pediatr 1997;47:269-72.

4. Essouri S, Chevret L, Durand P, Haas V, Fauroux B, Devictor D.Noninvasive positive pressure ventilation: five years of experiencein a pediatric intensive care unit.Pediatr Crit Care Med.2006;7:329-34.

5. Loh LE, Chan YH, Chan I. Noninvasive ventilation in children: areview. J Pediatr (rio J) 2007;83 (2 suppl):S91-99

6. Teague WG. Noninvasive ventilation in the pediatric intensivecare unit for children with acute respiratory failure. Pediatr Pul-monol. 2003;35:418-26.

7. Norregaard O. Noninvasive ventilation in children. Eur Respir J2002;20:1332-42.

8. Akingbola OA, Hopkins RL. Pediatric noninvasive positive pres-sure ventilation. Pediatr Crit Care Med. 2001;2:164-9.

9. Medina Villanueva A, Prieto Espunes S, Los Arcos Solas M, ReyGalan C, Concha Torre A, Menendez Cuervo S, et al. Aplicaciónde la ventilación no invasiva en una unidad de cuidados inten-sivos pediátricos. An Pediatr (Barc) 2005;62:13-9.

10. Essouri S, Nicot F, Clement A, Garabedian EN, Roger G, LofasoF, et al. Noninvasive positive pressure ventilation in infants withupper obstruction: comparison of continuous and bilevel posi-tive pressure. Intensive Care Med 2005;31:574-80.

11. Young HK, Lowe A, Fitzgerald DA, Seton C, Waters KA, Kenny E,et al. Outcome of noninvasive ventilation in children with neu-romuscular disease. Neurology 2007;68:198-201.

12. Larrar S, Essouri S, Durand P, Chevret L, Haas V, ChabernaudJL, et al. Effects of nasal continuous positive airway pressure ven-tilation in infants with severe acute bronchiolitis. Arch Pediatr2006;13:1397-403.

13. López-Herce Cid J, Moreno de Guerra Girón M, Sánchez Sán-chez C, Carrillo Álvarez A. Ventilación mecánica no invasiva enla hipoventilación central alveolar congénita. An Esp Pediatr2000;52:198-9.

14. Carroll CL, Schramm CM. Noninvasive positive pressure venti-lation for the treatment of status asthmaticus in children. AnnAllergy Asthma Immunol 2006;96:454-9.

15. Courtney SE, Barrington KJ. Continuous positive airway pres-sure and noninvasive ventilation. Clin Perinatol 2007;34:73-92.

16. Fauroux B, Boffa C, Desguerre I, Estournet B, Trang H. Long-termnoninvasive mechanical ventilation for children at home: a natio-nal survey. Pediatr Pulmonol 2003;35:119-25.


17. Bunn HJ, Roberts P, Thomson AH. Noninvasive ventilation for themanagement of pulmonary hypertension associated with conge-nital heart disease in children. Pediatr Cardiol 2004;25:357-9.

18. Barreiro TJ, Gemmel DJ. Noninvasive ventilation.Crit Care Clin.2007;23:201-22.

19. British thoracic society standards of care committee. Non-invasiveventilation in acute respiratory failure. Thorax 2002;57:192-211.

20. International consensus conferences in intensive care medicine:noninvasive positive pressure ventilation in acute respiratory fai-lure. Am J Respir Crit Care Med 2001;163:283-91.

21. Ambrosino N, Rossi A. proportional assist ventilation (PAV): a sig-nificant advance or a futile struggle between logic and practi-ce? Thorax 2002;57:272-6.

22. López-Herce Cid J, Carrillo Álvarez A. Nuevas modalidades deventilación mecánica. An Pediatr (Barc) 2003;59:82-102.

23. Gay PC, Hess DR, Hill NS. Non invasive proportional assist ven-tilation for acute respiratory insufficiency: comparison with pres-sure support ventilation. Am J Respir Crit Care Med 2001;164:1606-11.

24. Polese G, Vitacca M, Bianchi L, Rossi A, Ambrosino N. Nasal pro-portional assist ventilation unloads the inspiratory muscles of sta-ble patients with hypercapnia due to COPD. Eur Respir J 2000;16:491-8.

25. Younes M, Webster K, Kun J, Roberts D, Masiowski B. A methodfor measuring passive elastance during during proportional assistventilation. Am J Respir Crit Care Med 2001;164:50-60.

26. Younes M, Kun J, Masiowski B, Webster K, Roberts D. A methodfor noninvasive determining of inspiratory resistance during pro-portional assist ventilation. Am J Respir Crit Care Med 2001;163:829-39.

27. Du H-L, Ohtsuji M, Shigeta M, Chao DC, Sasaki K, Usuda Y, etal. Expiratory asynchrony in proportional assist ventilation. AmJ Respir Crit Care Med 2002;165:972-7.

28. Storre JH, Seuthe B, Fiechter R, Milioglou S, Dreher M, Sorich-ter S, et al. Average Volume Assured Pressure Support in Obe-sity Hypoventilation: a Randomized Cross-Over Trial. Chest 2006;130;815-21.

29. Thia LP, McKenzie SA, Blyth TP, Minasian CC, Kozlowska WJ, CarrSB. Randomised controlled trial of nasal continuous positive air-ways pressure (CPAP) in bronchiolitis. Arch Dis Child 2008;93:45-7.

30. Cambonie G, Milési C, Jaber S, Amsallem F, Barbotte E, PicaudJC, Matecki S. Nasal continuous positive airway pressure decre-ases respiratory muscles overload in young infants with seve-re acute viral bronchiolitis. Intensive Care Med2008;34:1865-72.


This chapter is intended as a guide to non-invasiveventilation (NIV) for acute patients in intensive care.Since the available data on pediatric patients is notstrong enough to use as the basis for any clinicalguidelines, the authors of this chapter provide certainrecommendations drawn from the relevant literatureand from their own experience. It should beunderscored that NIV for acute patients must alwaysbe performed in the ICU by well trained and judiciouspersonnel using the best possible materials.

Careful selection of patients by hospital staff canprevent certain inevitable failures that could lead tohigher patient mortality, and consequently, a loss ofconfidence in NIV. As the indications and contra-indications of NIV are described in detail in Chapter3, herein are presented only those pathologies orclinical scenarios for which NIV has been widely usedand has afforded good results.

In the first section of this chapter, therapeuticstrategies are outlined according to the latestclassification of acute respiratory failure (ARF). Ideally,special cases would have been treated here in detail;however, due to the limited pool of clinical evidenceon pediatric patients, this was not possible.

TYPE I ACUTE RESPIRATORY FAILUREType I acute respiratory failure (RF) is

characterized by V/Q mismatch without alveolarhypoventilation. In pediatric patients it tends to arise

with pneumonia, acute pulmonary edema (APE),chest wall trauma, neonatal respiratory distresssyndrome (NRDS) and acute respiratory distresssyndrome (ARDS) (Fig. 1).

IndicationSince NIV provides slow physiological

improvement, it has limited use for hypoxemic ARFpatients; nonetheless, it can prove somewhat effective,especially in cooperative children. NIV is further limitedfor this pathology because it requires rapid adaptation.However, with sufficiently experienced personnel andthe right equipment, NIV can be employed to treathypoxemic ARF patients, thereby avoiding intubation.In some publications, NIV is not recommended forpatients with PaO2/FiO2 < 150.

The authors of this chapter believe that NIVshould not be started based on the same arterialblood gas criteria as that used to start conventionalmechanical ventilation (CMV) (i.e. PaO2 < 60 mmHg and FiO2 0.5). Indeed, the decision to initiateNIV must be made with more prudence, as thistreatment will be preventative, rather thansubstitutive. The hemoglobin saturation index ([SatO2]/FiO2) could prove interesting as an alternativebasis on which to select patients for whom NIVis indicated; however, it has yet to be validated asa marker in pediatrics.

Early initiation of treatment strategies thatoptimize oxygenation is especially important for

Non-invasive ventilation methodologyfor acute pediatric pathologies

M. Pons, T. Gili and A. Medina

Chapter 8

those pathologies in which oxygen therapy alonedoes not compensate for the ARF (see Chapter 10;and Chapter 23, Figure 5).

Choosing an interfaceFor younger pediatric patients, the preferred

interface is an oral-nasal mask that adapts well tothe patient's face. For adolescent patients, theinterface of choice is a full face mask, although goodresults have been reported for cooperative adolescentpatients treated with nasal masks. For infants,medium-sized nasal interfaces (e.g. oral-nasal) canbe used (see Chapter 5, Fig. 2). Despite a lack ofcontrolled studies, and based solely on periodicexperiences, a helmet interface used with aconventional or specially designed ventilator can bea valid option for pediatric patients. Nasal interfacescan be tried for infants or for cooperative patientsover 10 years, as well as for those patients for whoman oral-nasal interface generates too much leakageand/or causes discomfort. Lastly, it should beremembered that for adolescent patients with TypeI RF, full face masks can be employed if oral-nasalmasks fail.

Choosing a ventilatorThe ideal ventilator would be an NIV-specific

ventilator equipped with an oxygen blender. If thisis not available, then the next choice would be aconventional ventilator with NIV software. Somehome ventilators feature an oxygen valve thatenables enrichment of the airflow with up to 60 to80% oxygen, depending on the recommendedvalues for pressure and respiratory frequency (seeChapter 6). In NIV-specific ventilators lacking ablender, oxygen (for patients requiring less than50%) can be delivered via a T-piece at the patient-end of the tubing. In certain unpublished cases ofinfants younger than 6 months suffering frombronchiolitis, good results have been obtained withnasal masks and special neonatal ventilators.Conventional ventilators without an NIV optionshould never be used for Type I ARF patients, as thechances of success are minimal. The only exceptionto this rule is that in treatment of ARDS, conventionalventilators used with a helmet can deliver a positiveend-expiratory pressure (PEEP) of 10 to 12 cm H2O.This technique has been employed by certainpediatrics groups in Italy.

Programming

NIV-specific ventilatorsStart with a continuous positive airway pressure

(CPAP) of 4 to 8 cm H2O, with FiO2 of 0.5 to 1. If norapid improvement is observed, then increase theCPAP up to 10 cm H2O. If this proves insufficient,then spontaneous/timed (S/T) mode should beinitiated with the following starting values: inspiratorypositive airway pressure (IPAP) of 8 to 10 cm H2Oand expiratory positive airway pressure (EPAP) of 5to 6 cm H2O. The IPAP should then be increasedin increments of 2 cm H2O; effective IPAP rangesfrom 10 to 22 cm H2O, with the best tolerated valuesbetween 10 and 14 cm H2O. The recommendedEPAP is 6 to 8 cm H2O.

Due to the lack of published pediatric clinicalexperience with pressure assist ventilation (PAV), thismode is only explained from a theoretical perspective(see Chapter 7, “Non-invasive ventilation modesand methods for children”).

Conventional ventilators featuring an NIV option• CPAP:

– Recommendations: Use a PEEP of 4 cm H20,and then gradually increase up to a maximumof 10 to 12 cm H20. Use respiratory frequency

60 M. Pons, T. Gili, A. Medina

Table I. Acute respiratory failure

Type I• Pneumonia • Acute pulmonary edema • Chest wall trauma • Acute respiratory distress syndrome• Neonatal respiratory distress syndrome • Bronchiolitis

Type II• Bronchiolitis • Asthmatic status • Obstructive apneas• Central apneas • Neuromuscular diseases

– Duchenne’s disease – Infantile spinal atrophy– Guillain-Barré syndrome– Myasthenia gravis

as rescue parameter with an inspiratory time(Ti) < 0.5 sec.

• PSV: – Recommendations: Use a PEEP of 4 cm H20

and a pressure support of 4 cm H20 (abovethe PEEP). Set the end of the inspiratory cycle,or the expiratory trigger, to 40 to 70% of the

peak flow reached. Use respiratory frequencyas rescue parameter for an inspiratory time(Ti) < 0.5 sec.

• Pressure-assisted/controlled ventilation (PACV):If expiratory synchronization can not beachieved, then PACV mode should be usedinstead, with the respiratory frequency and Ti

61Non-invasive ventilation methodology for acute pediatric pathologies

Figure 1. Algorithm for Type I acute respiratory failure (ARF).

*If pressure support ventilation (PSV) is used, then the respiratory frequency is not programmed because it is not required bythe ventilator; in PSV respiratory frequency is only used as a rescue parameter, analogously to the case of S/T mode in NIV-specific ventilators.ARF: acute respiratory failure; FiO2: fraction of inspired oxygen; PSV: pressure support ventilation; ST: spontaneous/timed; A/C:assisted/controlled; IPAP: inspiratory positive airway pressure; EPAP: expiratory positive airway pressure; f: respiratory frequency;Ti: inspiratory time; ins: inspiratory; exp: expiratory; PIP: peak inspiratory pressure; PEEP: positive-end expiratory pressure.Vt: tidal volume; NIV: non-invasive ventilation.

Type I ARF

Interface Oral-nasal

Ventilator with blender NIV-specific Conventional with NIV option

Initial setting CPAP: 5 to 10 cm H2OFiO2: 50 to 100%

No improvement S/STIPAP: 8 to 10 cm H2OEPAP: 5 to 6 cm H2O

Ramp: mediumFiO2: 50 to 100%

f (rescue): 10 to 15rpm less than patient

A/C or PSVPIP: 3 to 5 cm H2O

PEEP: 5 to 6 cm H2ORamp: slow

f (rescue): 2 to 5 rpm less than patient*Maximum Ti: similar to patient's value

Insp. trigger: minimumExp. trigger: 40 to 70%

FiO2: 50 to 100%

Test for efficacy Δ IPAP: 2 cm H2O every 5 minutes in function of the Vt reachedΔ EPAP: according to recruitment, tolerance and oxygenation

Δ Ramp: according to tolerance and Vt reached

Target values IPAP: 10 to 22 cm H2OEPAP: 5 to 8 cm H2O

Vt: 8 to 10 mL/kgReduction in f: ≥ 10 rpm within 1 h

FiO2 < 40% within 24 h

Non-vented interface

adjusted to the values reached spontaneouslyby the patient. Starting parameters: use a PEEPof 4 cm H2O, a peak inspiratory pressure (PIP)of 4 cm H2O, and respiratory frequency and Ti

similar to those of the patient.

TYPE II ACUTE RESPIRATORY FAILUREType II ARF is characterized by alveolar

hypoventilation. It tends to be associated withconditions that affect the respiratory impulse, as wellas airway obstruction, neuromuscular weakness,

chest wall abnormalities, and morbid obesity.Oxygen therapy alone is not sufficient to treat TypeII ARF; the conditions causing alveolarhypoventilation must be remedied (Fig. 2).

Choosing an interfaceNasal masks are well tolerated but must be

adapted to, as the patient has to keep their mouthclosed in order to avoid leaks in the circuit. Saidadaptation tends to be slow, though it can rangefrom just a few minutes to several days. For patientswho are hypoxemic, younger than 6 years, or unable


Figure 2. Algorithm for Type II acute respiratory failure (ARF).


Type II ARF


NIV NIV-specific Conventional with NIV option

Non vented interface

Initial setting S/TIPAP: 8 to 10 cm H2OEPAP: 4 to 5 cm H2O

Ramp: slowFiO2: as low as possible

f (rescue): 10 to 15rpm less than patient




Insp. trigger: minimumExp. trigger: 40 to 70%FiO2: as low as possible

Test for efficacyΔ IPAP: 2 cm H2O every 5 minutes in function of the Vt reached


Target values IPAP: 10 to18 cm H2OEPAP: 5 to 7 cm H2O

Vt: 8 to 10 mL/kgReduction in f within 3 to 6 h

NasalFiO2 ≥ 50% FiO2 < 50%

FiO2 < 50% without blender

FiO2 ≥ 50% with blender

CPAPApnea

BronchiolitisEPAP: 4 to 10 cm H2OFiO2: as low as possibleIf respiratory workload doesn’t

improve consider S/T or PSV

to fully cooperate, ventilation should be initiatedwith an oral-nasal interface, preferably one with anexpiratory port to minimize dead space.

Choosing a ventilatorThe preferred ventilator would be an NIV-specific

device that delivers two levels of pressure (IPAP andEPAP) in synch with the patient’s breathing.

Programming

Non-invasive-specific ventilatorsS/T mode. Due to hypoventilation, ventilation

should be started with two levels of pressure, exceptfor in patients with apnea or signs of bronchiolitis.

Starting parameters: Use an IPAP of 8 cm H2Oand an EPAP of 4 cm H2O; Once the patient toleratesthe interface well, raise the IPAP gradually inincrements of 2 cm H2O until reaching 14 to 20 cmH2O. In patients with atelectasis, the EPAP shouldbe raised to 5 to 7 cm H2O. FiO2 is delivered at levelssufficient for maintaining hemoglobin saturationbetween 92 and 95%. The clearest sign that properalveolar ventilation has been achieved is a reductionin oxygen demand, since in these patients,hypoxemia is secondary to hypoventilation.

CPAP. Obstructive sleep apnea (OSA) is a Type IIARF etiology which tends to be treated with CPAP,since, once the airway obstruction has been cleared,hypoventilation disappears. Recently, it has beenalso demonstrated for patients with bronchiolitis.

Starting parameters: Begin with a CPAP of 4 cmH2O, and then gradually increase the value up to 10cm H2O; if this is not sufficient for opening theairway, then switching to an S/T mode or PSV, A/Cwould probably afford better results.

Conventional ventilators featuring a non-invasive ventilation option

PSV. Recommendations: Use a PEEP of 4 cm H20and a pressure support of 4 cm H20 (above thePEEP). Set the end of the inspiratory cycle, or theexpiratory trigger, to 40 to 70% of the peak flowreached. Use respiratory frequency as rescueparameter for an inspiratory time (Ti) < 0.5 sec.

Pressure-assisted/controlled ventilation (PACV). Ifexpiratory synchronization can not be achieved, then(PACV) mode should be used instead, with therespiratory frequency and Ti adjusted to the values

reached spontaneously by the patient. Startingparameters: use a PEEP of 4 cm H2O, a peakinspiratory pressure (PIP) of 4 cm H2O, and respiratoryfrequency and Ti similar to those of the patient.

Conventional ventilators Conventional ventilators can be programmed

in any mode, although PACV with flow trigger isrecommended, despite the fact that it does notprovide optimal synchronization or compensate forleakage. In this mode, the respiratory frequency isset to 2 to 5 respirations/min (rpm) less than thepatient breathes spontaneously. In PSV mode, inwhich the Ti does not finish until the programmedpressure is reached, leakage extends the Ti, leadingto asynchrony and, consequently, failure (see Chapter7). Conventional ventilators should always be usedwith interfaces that are non-vented (i.e. that lack anexpiratory port) and that do not have an anti-asphyxia valve, since leaks in CMV are not controlledthrough the tubing or interface (see Chapter 5, Fig.6).

Regardless, the authors of this chapter believethat CMV should only be used as a last resort, andonly in the least severe Type II ARF patients (i.e.oxygen requirement < 50%).

Starting values for parameters:• Set sensitivity to the minimum; if possible, use

flow trigger if it does not lead to autocycling.• Set the respiratory frequency close to the

patient's value and then synchronize.• Use the same values as those used in

conventional ventilators with an NIV option.• In the event of high leakage, an oxygen flow can

be added to the expiratory loop to “trick” thealarms for low minute volume and for apnea (seeChapter 11, Fig. 5).

• The patient may need to be sedated.• Puede requerirse sedación.

STOPPING NON-INVASIVE VENTILATIONStopping NIV has not been studied

systematically. The majority of publications on NIVrecommend maintaining treatment as long aspossible in the first 24 hours, and then adding restperiods (e.g. for eating, receiving physical therapyor medication, or bathing) according to clinical

63Non-invasive ventilation methodology for acute pediatric pathologies

criteria. Some physicians have recommended thefollowing criteria for NIV patients to be given a restperiod: respiratory frequency < 24 respirations/min,heart rate < 110 beats/min, hemoglobin saturation> 90% with FiO2 < 4 L/min, and pH > 7.35.

CRITERIA FOR FAILURE OF NON-INVASIVE VENTILATION

The criteria for failure of NIV must be establishedbefore treatment is started. Hospital staff mustunderstand the limits of NIV in their ward accordingto their own experience and the available materials.The following factors should be considered as signsof NIV failure, especially within the first 4 to 6 hoursof treatment:• No improvement in symptoms (e.g. respiratory

frequency).• Patient-ventilator desynchronization.• Appearance of contra-indications (e.g. a decrease

in consciousness, hemodynamic instability,arrhythmias).

• General deterioration of the patient's condition.• No improvement in blood gas levels.• Appearance of complications that are

incompatible with NIV (e.g. abundant secretionsor severe hypoxemia).

• Decision by the patient’s parents to endtreatment; The option of intubation should bespecifically evaluated for each patient. Forpatients in whom NIV proves ineffective and forwhom intubation was ruled out early on,alternative therapies should be sought.

REFERENCES1. British thoracic society standards of care committee. Non-invasive

ventilation in acute respiratory failure. Thorax 2002;57:192-211.

2. International consensus conferences in intensive care medicine:noninvasive positive pressure ventilation in acute respiratory fai-lure. Am J Respir Crit Care Med 2001;163:283-91.

3. Pons Odena M, Cambra Lasaosa FJ. Ventilación no invasiva. EnGrupo de Respiratorio de la SECIP editores. Manual de Ventila-ción mecánica en pediatría. Madrid: Publimed, 2003;95-107.

4. García Teresa MA, Martín Barba C. Ventilación no invasiva. En:López-Herce, C Calvo, M Lorente, D. Jaimovich (eds). Manualde Cuidados Intensivos Pediátricos. Madrid: Publimed, 2001;655-60.

5. Teague GT. Non invasive ventilation in the pediatric intensivecare unit for children with acute respiratory failure. Pediatric pul-monology 2003;35:418-26

6. Padman R, Lawless ST, Kettrick RG. Noninvasive ventilation viabilevel positive airway pressure support in pediatric practice. CritCare Med 1998;26:169-73.

7. Nirajan V, Bach JR. Noninvasive management of pediatric neu-romuscular ventilatory failure. Crit Care Med 1998;26:2061-5.

8. Fortenberry JD, Del Toro J, Jefferson LS, Evey L, Haase D. Mana-gement of pediatric acute hypoxemic respiratory insufficiencywith bilevel positive pressure (BiPAP) nasal mask ventilation. Chest1995;108:1059-64.

9. Fortenberry JD. Noninvasive ventilation in children with respi-ratory failure. Crit Care Med 1998;26:2095-6.

10. Todd W Rice, Artur P Wheeler, Gordon R Bernard, Douglas L Hay-den, David Schoenfeld, Lorraine B Ware. Comparison of theSpO2/FiO2 ratio and the PaO2/FiO2 in patients with acute lunginjury or ARDS. Chest 2007;132:410-7.

11. Medina A, Prieto S, Los Arcos M, Rey C, Concha C, MenéndezS, Crespo M. Aplicación de ventilación no invasiva en una uni-dad de cuidados intensivos pediátricos. An Pediatr (Barc)2005;62:13-9.

12. Corrales E, Pons M , López-Herce J, Martinón-Torres F, GarcíaMA, Medina A, González Y. Estudio epidemiológico de la ven-tilación no invasiva en las Ucip en España. An Pediatr (Barc)2007;67:98.

13. Mayordomo J, Medina A, Rey C, Fernández B, Díaz J, Concha A.Utilidad de la ventilación no invasiva dependiendo del tipo deinsuficiencia respiratoria y marcadores de fracaso. Comunica-ción Oral XXIII Congreso de la SECIP. Sevilla 2007.

14. Thia LP, McKenzie SA, Blyth TP, Minasian CC, Kozlowska WJ, CarrSB. Randomised controlled trial of nasal continuous positive air-ways pressure (CPAP) in bronchiolitis. Arch Dis Child 2008;93:45-7.

15. Cambonie G, Milési C, Jaber S, Amsallem F, Barbotte E, PicaudJC, Matecki S. Nasal continuous positive airway pressure decre-ases respiratory muscles overload in young infants with seve-re acute viral bronchiolitis. Intensive Care Med2008;34:1865-72.



INTRODUCTIONThanks to its efficacy, rapid application, ease

of use, flexibility, and comfort for patients, non-invasive ventilation (NIV) is being increasingly usedin the pediatric intensive care unit (PICU).Consequently, hospital staff must have expertisein this technique and be able to resolve anycomplications that may arise during its use. Thischapter provides an overview of patient care in NIV,encompassing ventilation equipment set-up, patientpreparation, and troubleshooting.

PREPARATIONBefore administering NIV to a patient, hospital

personnel must first prepare all of the componentsof the ventilation system and ready themselves forany complications that could arise during treatment.It is crucial that patients are very closely monitoredin the first few hours of NIV to determine its efficacy;hence, hospital staff must anticipate any distractionsthat would force them to leave the patientunattended.

Hospital staff must anticipate and avoid anycircumstance which could interfere with NIVtreatment once it has begun. The following pointsshould be considered:

Setting up the equipmentThe NIV system must be adapted to the specific

ventilator chosen, including its power needs, andall of its components, to which it must be coupledcorrectly. A basic NIV system comprises the ventilator

body, an anti-bacterial filter, a humidifier, a watertrap, tubing, an expiratory valve, and interface withits harness (or other fastening system). The authorsof this chapter propose the following set-upsequence:1. Place the equipment close to the patient,

ensuring that the tubing does not limit thepatient’s mobility or autonomy. Choose a safe,stable location, protected from sunlight or othersources of heat (e.g. heaters, lamps andunfiltered windows) that could heat theventilator or its components to temperaturesabove 55 °C.

2. Connect the ventilator to the electrical outlet (orother power source): its Power/On indicatorsshould light up.

3. Attach the (non-hydrophobic) anti-bacterial filterbetween the ventilator’s air outlet and thetubing, ensuring that it can withstand theprogrammed airflow (the manufacturers of thesefilters generally specify the minimum andmaximum intended flows). Be sl as rere units PICU)4. Connect the filter tothe humidifier; this normally requires a smallpiece of tubing. Fill the humidifier with distilledwater to the level indicated by the manufacturer.

5. Connect the far (ventilator) end of the tubingto the humidifier, and connect the near(patient) end of the tubing to the expiratoryvalve or interface, according to the modelsused. If the interface contains an expiratoryvalve, ensure that the outlet capacity is notblocked by the patient’s clothes or the bed. The

Caring for non-invasive ventilationpatients

J. García-Maribona, M. González, J.M. Blanco and J.C. Monroy†

Chapter 9

outlet must not be directed at the patient, asthe airflow would cause local irritation ordiscomfort at the site where it was positioned.If a conventional ventilator with NIV option isused, then do not attach an expiratory valveor use ventilated interfaces: these ventilatorsalready contain expiratory tubing or branchesthat divert expired air, and any leaks from aleak control outlet on an expiratory valve orinterface would interfere with ventilatorfunction. If specific tubing is to be used forhumidification (i.e. heated tubing whichcontains an internal filament for insulatingheated air to avoid water condensation), thencheck the connections between the humidifier’stemperature sensors and the tubing, as faultyconnections could give false readings, deliveringair at excessively high or low temperatures thatcould cause nasal passage burns, andconsequently, patient discomfort.

6. If the ventilator has an independent near-endpressure segment, then connect this segmentto the specific outlet on the ventilator at oneend, and to the pressure intake on the interfaceor expiratory valve (depending on the modelsused) at the other end.

7. If the ventilator does not have a specific oxygenintake, then the system can be outfitted with anoxygen T-piece, which enables variable deliveryof oxygen into the airflow delivered to thepatient. The T-piece should be positionedaccording to the following factors:• If it is placed at the ventilator end, then a more

homogeneous airflow will be obtained at theinterface, but the FiO2 will be highly variable,as the oxygen will be diluted by the air insidethe tubing.

• If it is placed at the patient (interface) end,then the FiO2 will be more stable, but therewill major turbulence in the airflow that thepatient draws in, which will cause the patientsome discomfort.

8. Choose the interface according to the patient’sage (size) and pathology and choose the harnessaccording to the shape of the patient’s face andthe ventilation mode selected (see Chapters 5and 6). Use of more than one interface for eachpatient is recommended: by alternating betweenvarious interfaces that are supported on different

areas of the patient’s body, the risk of pressuresores can be reduced.

9. Wash the interface with a liquid soap that is freeof conditioners—if unavailable, use a detergentwhich is free of ammonia and chlorine. Rinse theinterface thoroughly in warm water, and thendry it carefully. Interfaces can normally be useddirectly without a prior washing step. They mustbe used exclusively for one patient only, andwashed after each use and if the patient vomitsor has abundant secretions, in which case theanti-asphyxia and security valves must bechecked for correct functioning (e.g. ensure thatexhaust ports are open and have the correctdiameter).Once all of the materials have been prepared,

the following conditions must be met before startingNIV:• Ensure that all ventilation equipment works

correctly, (e.g. aspirators, oxygen source,humidifier and pulse oximeter). Running theequipment briefly before initiating treatmentis recommended.

• Have the following ready for immediate use: anaspiration system (vacuum flow meter, tubing,and container for secretions); a sufficient quantityof sterile, disposable aspiration probes (of theappropriate caliber) for removing vomit orsecretions; and a container of water for cleaningthe aspiration system after each use.

• Have the following ready and close to thepatient: a self-inflating resuscitation bag with itscorresponding interface (adjusted to the patient'ssize and tidal volume) connected to an oxygensource capable of delivering a minimum flow of10 L/min.

• Ensure that appropriate resuscitation materialsand medication are available for each patient.

Preparing the patient

Conditioning the patient (Fig. 1)The protocol for conditioning the patient before

beginning NIV treatment should include thefollowing steps:1. Check that the prescribed treatment corresponds

to the patient.2. Confirm that the patient does not have any

contra-indications to NIV.

66 J. García-Maribona y cols.

3. Check the permeability of the patient’s airways,aspirating their secretions and removing anyobjects (e.g. nasal prongs) that could potentiallycompromise permeability. Pacifiers should notbe used in small children who are treated viaface mask, since in the event of vomiting, thesecould become an additional obstacle toremoving the vomit. Furthermore, as patientswith the most severe forms of acute respiratoryfailure (ARF) breathe primarily through theirmouth, a pacifier could markedly limit theircapacity for ventilation. Contrariwise, pacifierscan be extremely useful for controlling pressurein patients treated with nasal interfaces, whichleave the mouth free. Specifically, pacifiers tendto improve the basal tone of the muscles thatboth keep the mouth closed at rest and ensurethat it does not easily open upon application ofpositive pressure to the airway.

4. Check the patient for gastric distention, whichis a common side effect in NIV amongneuromuscular disease (NMD) patients forwhom an IPAP above 15 cm H2O can overcomethe closing pressure of the cardia(approximately 25 cm H2O in healthyindividuals). If there is any doubt as to thepatient’s ability to control gastric distensionsecondary to NIV, then treat the patient withan open nasogastric tube.

5. Prevent pressure sores by protecting the zonesof the patient’s body that will be subject tocontinuous pressure from the interface or harnessstraps. The most sensitive areas are listed belowaccording to the source of irritation.• Irritation caused by the interface: nasal root,

forehead, cheeks (in patients with prominentcheeks), and edges of the cheekbones (inpatients with little adipose tissue)

• Irritation caused by the harness: cheeks, earpavilions, occipital margin, upper back neck,the chin (in harnesses with a chin strap), plusareas that come into contact with roughsurfaces (e.g. seams, Velcro and rigid edges)

• Irritation caused by tubing rubbing on thepatient’s body: chest and abdomen.

• Irritation caused by other materials: specialcare should be taken with areas that comeinto contact with nasal-gastric probes or withsome other device (e.g. catheter or drain)

inserted between the patient’s skin and theinterface or harness (see Fig. 1).

For use in preventing pressure sores, somemanufacturers provide padding (e.g.Microfoam®) with their interfaces. If this is notavailable, then soft-adhering hydrocolloid stripscan be used (Fig. 2). Hospital staff should applythese bandages to cover vulnerable areas of thepatient’s body at the very onset of ventilation;they should not wait until having observed theinitial signs of pressure sores (reddened skin) toact.The patients should also be given brief pausesfrom treatment (1 minute per hour) to alleviatepressure, especially on the nasal bridge area.During these breaks, the most sensitive areasshould be moisturized and massaged. Aspreviously mentioned, switching between twoor more well-adjusted interfaces with differentsupport areas can help minimize localizedpressure. Children’s faces should be periodicallytreated with an appropriate moisturizing lotionas well as with hyper-oxygenated oils (e.g.Corpitol and Mepentol, in Spain), which favorvascularization in areas submitted to pressureand can prevent or heal pressure sores up tograde 1. Use of barrier creams is recommendedfor scratched or damp areas. Lastly, hospitalpersonnel should prevent the tubing fromscratching or cutting the child’s skin.

6. Ensure that the proper analgesic is administeredin each case; this is especially important forpatients experiencing chest pain, since they tendto hypoventilate this area. The patient’s pain—

67Caring for non-invasive ventilation patients

Figure 1. Equipment positioned between the patient and theharness can increase the risk of pressure sores.

and subsequently, the efficacy of the analgesic—should be characterized and registered usinganalog and/or visual scales.

7. Plan occasional rest periods from ventilationduring which the patient can be cared for (e.g.fed, bathed, medicated, or attended to forsecretions), their sensitive skin areas can betreated, interfaces can be alternated, andventilator equipment can be repositioned,cleaned or otherwise maintained. These pausesare designed to not interfere with ventilationtherapy. Various criteria have been proposed forthese periods: some authors have recommended20 to 30 minutes every 4 to 6 hours, some haverecommended 5 to 15 minutes every 3 to 6hours, and some have recommended adjustingthe pauses to each patient’s tolerance.

8. For NIV administered in the PICU or similar ward,patients often must be mildly sedated viacontinuous infusion during the initial phases oftreatment.

Accommodating the patientOnce the patient has been conditioned, they

must be properly accommodated: the recommendedposition, when possible, is for the patient to be seatedat a 45° angle, with their legs bent and their poplitealspace supported. This enables relaxation of theabdominal muscles and allows broader diaphragmaticmovements using less force (see Fig. 3). This initialposture can be supplemented with auxiliary positionssuch as: • Both axillae supported to relieve the chest wall

and spinal column of their support load andenable broader costal movement

• Both sides of the face cushioned to secure thehead in a neutral position

• Cervical region of the spinal column supportedby cushions or a pillow to prevent the neck frombending over the chest wall (since excessivespinal extension can compromise the openingand permeability of the airway, especially in smallchildren).

• Lumbar region of the spinal column supportedto favor lordosis and prevent the chest wall fromsagging over the abdomen, thereby enablingbroader diaphragmatic movementsThe aforementioned positions provide the

patient with the greatest comfort and respiratoryefficacy possible in the context of their pathology.A comfortable patient will always be more willingto cooperate with hospital staff. Patients withrespiratory sleep disorders suffer from more apneasin the supine position; hence, it is recommendedthat they adopt the lateral semi-supine position. Inthis case, for small children or those who tend toadopt other positions, an object (free of sharp edges)can be placed behind the patient’s back, such thatthe supine position becomes uncomfortable, andconsequently, they gravitate towards the semi-supineposition. When positioning the patient, especiallyvery young children, hospital staff should anticipatethe possibility of the patient falling by using safetyrails or similar equipment.

Educating the patient and their caretakersThe patient and their caretakers should be

familiarized with the maneuvers required to preventcomplications inherent to NIV and acquire goodhabits to this end. These include:• Optimizing breathing work: Due to the elasticity

of their chest wall, infants and small children


Figure 2. Application of hydrocolloid dressings before initia-tion of non-invasive ventilation.

Figure 3. The patient’s posture must be corrected to favorrespiratory movements.

tend to take abdominal or diaphragmaticbreaths, which are the most effective. As patientsage, they gradually shift to costal respiration,which is less effective and consumes moreenergy. Since NIV patients remain consciousduring treatment, they should be taught howto make proper respiratory movements throughdiaphragmatic breathing and forced expirationexercises in which they must optimize respiratoryefficacy. However, it is difficult for critical patientsto be receptive to breathing instructions; hence,hospital staff must wait until these patients arein the proper physical and physiological state tomodify their respiratory behavior.

• Effective cough: One of the advantages of NIVis that it enables children to maintain their reflexand ability to cough. However, patients oftenuse short, weak coughs (which are not veryeffective). Hence, hospital staff should teachthese patients how to cough correctly, usingexercises for forced inspiration, laryngealocclusion, and contraction of the diaphragm insynch with glottal relaxation. Neuromusculardisease (NMD) patients require assisted cough(either manual or automatic).

• Preventing and relieving gastric distension andits consequences: Application of positive pressureto the airway can cause gastric distension, whichin turn can lead to vomiting and regurgitation.Discontinuous NIV treatment should not startuntil at least 1 hour after the patient’s last meal.There is less risk of compromising the airwaywith vomit when nasal interfaces or tubes areused, as the mouth is not blocked. Children whorequire uninterrupted NIV can be taught todetect their own gastric distension and to controlit through burping. The patient and theircaregivers must be advised of the possibility ofvomiting and how to respond if it occurs (theyshould have towels or swabs on hand, plus acontainer for the vomit). Patients treated withoral-nasal interfaces must be taught how toquickly and correctly remove these; this willprovide them with a greater sense of controlover the situation and reduce their fear.

• Maintaining the seal on the interface: The patientand their caregivers must understand theimportance of maintaining a proper seal betweenthe mask and the skin to avoid leakage. For the

patient to collaborate, they must be taught howto properly adjust the straps on the interface andhow to avoid dragging the mask or tubing. Thepatient may need to use a safety strap to affixthe tubing to their clothes in order to minimizeany problems with the interface.

• Valsalva’s maneuvers: these are designed to keepthe Eustachian tubes and sinus drainage areaspermeable in order to prevent otitis and sinusitis.With these movements, the Eustachian tubesretain their ability to balance pressure betweenthe middle ear and the oropharynx. There aretwo main exercises:1. Keeping their mouth closed and nose

pinched, the patient expels air towards thenostrils, increasing the pressure from theoropharynx until feeling that the Eustachiantubes have opened (the patient will feel theirears click), thereby allowing air to enter intothe middle ear.

2. Keeping their mouth closed and nosepinched, the patient swallows saliva; upontraveling towards the esophagus, the boluswill produce a negative pressure in theoropharynx, which in turn will cause the tubesto reopen (and compensate for the differencein pressure).

Getting the patient involved in their own treatment

This is achieved by honing the structural,technical and human elements that surround thepatient. By ensuring that the child does not perceivethe therapy as dangerous or strange, hospital staffwill make the child feel comfortable and gain theircooperation. This can be accomplished by addressingthe following points:• Minimizing sources of stress around the child:

Both hospital and home NIV requiresurroundings in which the patient feels calm andcomfortable. This enables objective evaluationof the efficacy of the treatment. The mostfrequent complication in NIV—and one of themain causes of treatment failure—is a lack ofpatient adaptation due to agitation ordisordered, ineffective breathing. These problemscan be drastically reduced by creating the rightenvironment—namely, by minimizing thevolume and number of alarms or of any


unnecessary sounds; moderating the volumeand content of conversations held around thepatient; lighting the area in function of the child’sactivity level (i.e. intense light for activities, softlight for resting, and near or total darkness forsleeping); carefully decorating the room;respecting and facilitating the patient’s intimacyand modesty; and providing the child with age-appropriate activities such as games (see Fig. 4).

• Fostering confidence in the efficacy of NIV:Patients, especially those who suffer from chronicconditions with episodes of ARF, and theircaregivers are skeptical of any treatment thatthey are not familiar with. Educating patientsand their caregivers on NIV can reduce theiranxiety and foment success of the treatment.

• Gaining the trust of the patient: Upon meetingthe patient for the first time, hospital staff shouldbe amiable, not make any sudden movements,look directly at the patient when speaking,focusing on an imaginary point just above theireyes, try to bring themselves to the patient’sheight, and use a moderate tone of voice,avoiding major fluctuations. Doctors and nursesshould introduce themselves by name, state theirtreatment objectives, ask the patient their name,and refer to the patient by their name. Once aninitial feeling of trust has been established,hospital staff should have greater physical contactwith the patient.

• Encourage patient cooperation by using rewardsand supportive behavior and phrases.

Clinical evaluation before treatmentIt is imperative that the patient is clinically

evaluated before being treated with NIV, in order toasses the specific consequences of the treatment forthem. Said evaluation encompasses two main areas:1. Monitoring and recording of vital signs: heart rate,

respiratory frequency, oxygen saturation, bloodpressure, and venous or arterial blood gas levels.

2. Evaluation and recording, for subsequent follow-up, of:• Pain (location, and intensity according to

scales).• Permeability of airways (based on the

presence, quantity, consistency andappearance of nasal, oral and trachealsecretions).

• Respiratory state (cyanosis, dyspnea, or signsof increased work of breathing, such asparadoxical breathing, use of accessoryabdominal muscles, sagging of the xyphoidprocess and flaring of the nostrils), includingexistence and efficacy of cough.

• Mental state (agitation, anxiety, depression,confusion, and ability to cooperate).

• Diameter of the abdomen and occurrenceof vomiting.

• Dermatological state (existence, location andgrade of skin sores, and level of topical andsystemic hydration); Scales such as the Bramenscale are useful for gauging the susceptibilityof the patient's skin to scores. These are basedon a broad range of factors, including thepatient’s level of mobility, activity, perception,local and general skin hydration, and nutrition.

• Signs of conjunctivitis and of otitis.

INITIATING NIVWhen starting NIV, the following steps should

be followed:1. Attach the cap or harness evenly, such that once

interface is connected, it will remain well-positioned on the patient’s face or head and willnot shift to one side. Care should be taken toavoid that the labels and other sharp parts ofthe straps do not come into direct contact withthe patient. In the smallest children (neonatesand infants) a crown-type harness should beused to avoid excessive pressure on the occipitalzone and the upper neck in order to preventvascular complications.


Figure 4. Making the patient as comfortable as possible, andminimizing their causes of stress, will make them more coo-perative.

2. Start the ventilator, and if required, connect itto an oxygen source and start the humidifier.

3. Connect the chosen interface to the tubing andbegin ventilation. Air should begin to flowthrough the interface, causing a noise that couldfrighten the patient; assure them that everythingis fine.

4. Attach the interface to the patient’s face.Whenever possible, the patient should attachthe interface themselves, using the pressurerequired to obtain a good seal, allowing a certainlevel of air leakage if needed. This way, thepatient will gradually become accustomed tothe interface and its effects, overcome any fearor anxiety, and better adapt to the system.Alternatively, NIV treatment can be begun byfirst attaching the interface to the patient’s face,and then starting the ventilator. However, thissecond method is more difficult for patients withmajor dyspnea, anxiety or agitation, since,depending on the expertise of personalattending them, they may desaturate. This canbe prevented by introducing a flow of oxygenat the point where the interface connects to thetubing. This method is advantageous, as itprovides for better attachment of the interface,obviating subsequent repositioning andresealing. Moreover, during adaptation patientsdo not have to experience the bothersome leakcompensation that occurs in the first method,which can annoy them to the point that theyinitially reject NIV. When warmed, interfaces thathave silicone gel seal strips afford betteradaptation to the patient's face and bettersealing. Warming can be accomplished byrunning the interface under hot water; thisliquefies the gel, making it more malleable. Oncethe interface reaches a suitable temperature, itcan then be delicately positioned on the patient'sface.

5. Once the interface has been optimally adaptedto the patient’s face, attach it to the patient usingthe straps on the harness or cap. Care should betaken to ensure that the support areas on theinterface remain positioned on the patient’sprotected areas (i.e. against skin sores). Interfacesshould not be attached too tightly: there shouldbe enough room for two fingers to fit betweenthe straps and the patient. The optimal fit will

provide the best seal with the least tension. Fororal-nasal masks, the best fit is obtained by firstadjusting the lower part of the mask to thepatient's chin, and then adjusting the upper(nostril area) part using the frontal straps; in thereverse order, the masks tends to place too muchpressure on the nostril area and leak through thechin area. The mask should be adjusted to bewell centered on the patient’s face, andsupported on the aforementioned protectedareas, and to reduce leakage in the system. Itis crucial that the nostrils, nasal bridge andforehead remain well protected. It should beremembered that when nasal or oral-nasalinterfaces are used for infants, there is a tendencyto place too much support on the frontalcushion; this gives the interface an exaggeratedincline which causes leakage in the lower portion.

6. Gradually adjust the ventilation parameters untilobtaining a suitable reduction in breathing workand improved vital signs. Pennock et al. havereported that effective adaptation can beachieved within 15 to 30 min of patient-interfaceinteraction under strict supervision by hospitalstaff. The authors of this chapter have learnedfrom experience that this initial period ofventilation is fundamental for deciding whetherthe patient should continue receiving thetreatment or should be switched to conventionalmechanical ventilation (CMV). The level ofpatient care can be somewhat reduced after saidperiod; however, the patient should remainunder strict vigilance for the first 12 hours, whichis a crucial window of time for adaptation andfor monitoring any side effects. This level ofvigilance should be extended, and may need tobe held permanently, for patients withpathologies that imply an especially high risk ofcomplications such as cardiorespiratory instabilityor difficulty in sedation (e.g. infants with severeARF secondary to bronchiolitis).Patients are considered to respond well to NIV

if they exhibit the following: good adaptation,improved respiratory and mental states, reductionor elimination in breathing work and in dyspnea,and no signs of gastric distension. Blood gas levelstake longer to correct, and therefore, should not beused as criteria for determining the success of NIVbefore at least 1 hour of treatment (see Chapters 13


and 23). Acidosis and hypercapnia may requireseveral hours to correct.

Hospital staff should act according to thefollowing guidelines for providing successfultreatment and avoiding any related complications:1. Ensure that airways are permeable by aspirating

and humidifying secretions as often as needed. 2. Protect and monitor areas of the patient's skin

that are susceptible to irritation from pressureor scraping (from the interface or tubing) toavoid skin sores or cuts.

3. Protect the patient’s eyes from the effects ofleakage from the interface; prevent conjunctivitisby using eye drops and ointments if needed.

4. Monitor functioning of the ventilator and itsaccessories.

5. Help the patient maintain the best posture fortheir needs.

6. Adjust the interface frequently to avoid or correctexcessive leakage (one of the main causes offailure in NIV).

7. Monitor and record vital signs: heart rate,respiratory frequency, oxygen saturation, bloodpressure, temperature, etc.

8. Avoid contamination of the system by changingthe antibacterial filters and washing the interfacesevery 24 hours, eliminating condensed moisturein the tubing, and, when required, refilling thewater in the humidifier using all requiredprotocols for sterility.

9. Watch for signs of gastric distension by listeningfor borborygmus with a stethoscope in thepatient’s epigastric region and monitoring anyincrease in abdominal diameter. If needed, inserta nasogastric tube to reduce the patient’s gastrictension.

10. Prevent otitis by periodically hydrating thenostrils with an isotonic saline solution andaspirating the secretions, frequently providingthe patient with liquids in small quantities, andencouraging the patient to perform Valsalva’smaneuvers.

11. Ensure that the patient maintains adequatebodily hygiene by bathing them at least once aday with warm water and mild soap. For youngchildren, use age-specific products.

12. Feed each patient according to their specificneeds, including their recommended pausesin NIV. During flare-ups of respiratory failure,

continuous enteral feeding (if possible,transpyloric) is recommended. More-stablepatients can be fed frequently with smallquantities of easy to swallow, energy-rich foods.

13. Ensure that the patient is involved in their careand participates in their treatment.

14. Make the patient as comfortable as possible.15. Ensure that the proper medication and treatment

are provided to each patient.Patient care in non-invasive ventilation is

summarized in the algorithm shown in Figure 5.

HOME CARETransferring an NIV patient from the hospital

to their home requires the contributions of allmembers of the multi-disciplinary NIV team.Nurses, technicians and therapists have anespecially important role here, both in the hospitalas well as in primary care teams at communityhealth clinics. In the hospital, their primary tasksare to teach patients and their caregivers, as well


Figure 5. Algorithm for patient care in non-invasive ventilation(NGT: nasogastric tube; CPR: cardiopulmonary resuscitation).

PacifierNGTCPR materials

Prepare the materials Interface and harnessVentilatorTubingHumidifierAccessories (Table VIII)

Prepare the patientHydrocolloid dressingsEnsure correct posture: layingback at 45° angle

Set-up the ventilationcircuit

1. Position the cap or harness2. Attach the interface3. Connect the ventilator4. Adjust the cap straps

Patient care: painand discomfort Check-up on the patient often

Sedation/analgesia

Scheduledventilation pauses

Local massagesLocal hydrationAspiration of secretionsNutritionHygieneGeneral care

Patient education Effective work of breathingEffective coughControl gastric distensionControl and treat vomitingControl leakageValsalva maneuver

Prevent andtreat side effects

ConjunctivitisSoresAbdominal distensionOtitisAir escape

Confirm that airwaysare accessible

as other NIV team members, and to develop atreatment plan specific for each child and eachventilation system. Outside of the hospital, nurseshave several duties:1. Evaluate the area where the patient is to be

treated and the individualized treatment plandeveloped for each patient according to thefollowing factors:• Accessibility of the patient’s home to

emergency and support personnel andequipment.

• Availability of private and public transport(from home to hospital).

• Proximity of the patient’s home to emergencyservices (i.e. response time).

• Hygienic conditions in the home: cleanliness,ventilation, light, temperature, and facilitiesfor personal hygiene. Since home NIVtreatment aims to help patients reintegrateinto society, it is important to avoid creatinga hospital-like environment in the patient’shouse.

• Quality and power capacity of the electricalinstallations in the patient’s home (e.g. fuseboxes).

• Availability of public health resources for thepatient: medical care at school, social securityand social services

• Level of competence of the patient’scaregivers.

2. Oversee an individualized plan for each patientthat comprises:• Detailed pharmaceutical prescriptions and an

outline of the course of treatment, writtenin clear language for the patient and theirfamily and caregivers

• Training for the patient’s family or caregiversin the treatment protocol, includingmaintenance of the ventilation equipmentused.

CONCLUSIONSThere is no doubt as to the benefits of NIV for

the patient, their family, health services and thepublic. Hospital personnel must accept their role instarting NIV treatment with decisiveness and clarity,basing their clinical practice on technical expertise,experience and constant learning.

REFERENCES1. New therapeutic technology for routine use. Chest 1994;

105:441-44.

2. Peñalver Hernández F. Inicio de la Ventilación Mecánica.Aspec-tos de control de Enfermería. En: Esquinas A, Blasco J, Halles-tad D editores. Ventilación Mecánica no Invasiva en Emergencias,Urgencias y Transporte Sanitario. Granada: Ediciones AlhuliaSL 2003. p. 185-208.

3. Knopp B. Fisioterapia Torácica. En: Logston Boggs R, Wooldrid-ge-King M (eds). Terapia Intensiva Procedimientos de la AACN,3ª ed Buenos Aires: Editorial Panamericana 1995. p. 87-95.

4. Prevos S. Manejo de las úlceras por presión. En: Logston BoggsR, Wooldridge-King M (eds). Terapia Intensiva Procedimientosde la AACN, 3ª ed. Buenos Aires: Editorial Panamericana, 1995.p. 740-5.

5. Carrión Camacho MR, Terrero Varilla M. El Paciente Crítico conVentilación Mecánica No Invasiva. Modos ventajas desventajasy principales cuidados de enfermería. En: Esquinas A, Blasco J,Hallestad D (eds). Ventilación Mecánica no Invasiva en Emer-gencias Urgencias y Transporte Sanitario. Granada: EdicionesAlhulia SL 2003. p. 209-30.

6. Goldberg A. Ventilación mecánica en domicilio aspectos orga-nizativos. En: Ruza F (ed). Tratado de Cuidados Intensivos Pedia-tricos. 3ª ed. Madrid: Ediciones Norma-Capitel 2002. p. 695-700.

7. García Teresa MA. Ventilación mecánica en domicilio aspectosprácticos. En: Ruza F (ed). Tratado de Cuidados Intensivos Pedia-tricos. 3ª ed. Madrid: Ediciones Norma-Capitel 2002. p. 701-13.

8. Raposo Rodrigues G, Correira M. Ventilación mecánica no inva-siva. En: Ruza F (ed). Tratado de Cuidados Intensivos Pedia-tricos. 3ª ed. Madrid: Ediciones Norma-Capitel 2002. p.695-700.

9. García Teresa MA. Procedimientos no invasivos para mejorar laoxigenación y ventilación. En: Casado Flores J, Serrano A (eds).Urgencias y tratamiento del niño grave. Madrid: Ediciones Ergon2000. p. 142-149.

10. Barcons M, Mancebo J. Ventilación no invasiva con presión desoporte. En: Net Á, Benito S (eds). Ventilación mecánica. 3ºed. Barcelona: Springer-Verlag Iberica 1998.p. 369-76.

11. Estopá R. Ventilación a domicilio. En: Net Á, Benito S (eds). Ven-tilación mecanica. 3ª ed. Barcelona: Springer-Verlag Iberica 1998.p. 253-368.

12. West JB. Fisiopatología pulmonar. 5ª ed. Madrid: Editorial Médi-ca Panamericana SA 2000.

13. Spessert CK, Weilitz PB, Goodenberger DM. A protocol for initia-tion of nasal positive pressure ventilation. Am J Crit Care 1993;2:54-60.

14. St John RE, Thomson PD. Noninvasive respiratory monitoring.Crit Care Nurs Clin North Am 1999;11:423-35.

15. González Pérez M, García-Maribona Rodríguez-Maribona J, Medi-na Villanueva JA. Ventilación Mecánica no Invasiva en Pedia-tría. En Puesta al día en Urgencias Emergencias y Catástrofes. EdArán. 2007;7:48-55.

16. Chevrolet JC, Jolliet P, Abajo B, Toussi A, Louis M. Nasal positivepressure ventilation in patients with acute respiratory failure.Difficult and time-consuming procedure for nurses. Chest1991;100:775-82.

17. American Thoracic Society. Home mechanical ventilationof pedia-


tric patients. Am Rev Respir Dis 1990;141:258-9.

18. Abad Corpa E, Hernández González M. Complicaciones de la ven-tilación mecánica no invasiva. Enfermeria Global 2002;1:1-12.

19. Peñas Maldonado J, Valverde Mariscal A, Martos López J. Ven-tilación mecánica domiciliaria. En: Barranco Ruiz F, Blasco Mori-lla J, Mérida Morales A, Muñoz Sanchez MA. Principios deUrgencias emergencias y Cuidados Críticos: Edición Electrónica,http://www.uninet.edu/tratado/c0204i.html

20. Artacho R, García de la Cruz JI, Panadero JA, Jurado Solís A, Dega-

yón H, Guerrero A, et al. Ventilación mecánica no invasiva uti-lidad clínica en urgencias y emergencias. Emergencias 2000;12:328-336.

21. González Pérez M, Medina Villanueva JA, García-Maribona Rodrí-guez-Maribona J. Ventilación No Invasiva. Tratado Enfermería deCuidados Críticos Pediátricos y Neonatales. Edición electrónica,http://www.eccpn.aibarra.org/

The aim of non-invasive ventilation (NIV) is to aidthe patient’s pulmonary function and to improve anysymptoms, signs or blood gas alterations resultingfrom respiratory failure (RF). Monitoring of NIVcomprises tracking of various parameters in the patient(e.g. clinical, analytical and functional) and in theventilator to evaluate the treatment as early as possibleto determine if it should be continued, modified orstopped. The techniques and frequency of monitoringdiffer between acute respiratory failure (ARF) andchronic respiratory failure (CRF) patients and varywith the severity of the pathology, the location oftreatment (i.e. hospital or home), and the time whichthe patient has been receiving NIV.

MONITORING OF NON-INVASIVE VENTILATIONIN ACUTE RESPIRATORY FAILURE PATIENTS

In ARF patients, NIV is used to amelioraterespiratory function without resorting to intubation,by improving symptoms, reducing work ofbreathing, and improving gas exchange.

Clinical monitoring This type of observation is crucial for determining

the efficacy of NIV. Patients should be monitoredcontinuously, especially in the first 4 to 6 hours oftreatment (or more, based on their needs). Thisdemands that hospital personnel remain at thepatient’s bedside to make any adjustments to theparameters, change the interface or switch thepatient to intubation.

First and foremost, hospital staff must determinethe level of comfort and adaptation of pediatricpatients, as these factors can dictate treatmentsuccess. They depend on the type of ventilator,ventilation mode, IPAP and EPAP values, patient-ventilator synchrony, the sensitivity of the ventilatortrigger, the type of interface, the pressure that theinterface places on the patient’s face, and theamount of leakage in the ventilation circuit. Thepatient’s age, pathology, level of cooperation, andpersonality can also be influential. Any sources ofdiscomfort must be recognized and corrected.

Hospital staff should endeavor to minimize cryingand speaking in children treated with nasal masks,since these cause a loss in efficacy of ventilation andan increase in discomfort from leakage, which canworsen the RF. Breathing work and polypnea shouldbe closely watched: NIV should reduce both.Dyspnea should improve, and any use of anyaccessory muscles in breathing should diminish,within 1 to 2 hours of treatment. Lack of patient-ventilator synchrony can lead to failure of NIV. It isimportant to remember that neuromuscular disease(NMD) patients, and patients with an alteredrespiratory center, may not present with dyspnea,despite having severe RF. It should also beremembered that normal respiratory frequencydepends on age and that children can tolerategreater tachypnea than adults. Hospital staff shouldalso watch for cyanosis, observe the patient’s chestwall mobility, and monitor the patient’s level ofconsciousness as a marker of hypercapnia.

Monitoring of non-invasive ventilationfor pediatric patients

M.Á. García Teresa, R. Porto Abal, P. Fernández Deschamps

Chapter 10

Listening to the patient’s chest with astethoscope is important for checking for adequateentry of air, adjusting the IPAP level and trackingtheir respiratory pathology.

Constant monitoring of the patient’s heart rateand respiratory frequency is mandatory: successfulNIV should reduce both, although in many publishedcases, these could not be used to gauge the efficacyof the treatment. Periodic measurement of bloodpressure is required.

Daily chest wall and abdominal radiographycould be considered for assessing pulmonarypathology and discard complications.

Gas exchangeMeasurement of arterial blood gases (arterial

oxygen pressure [PaO2], arterial carbon dioxidepressure [PaCO2], and arterial pH) is considered tobe the gold standard for evaluating pulmonary gasexchange. However, it is rarely used in pediatricpatients that do not have already have an arterialcatheter placed, which is often the case in pediatricNIV patients. Repeated puncturing of arteries ispainful, and therefore, can cause the patient to cry,which leads to false blood gas readings.Furthermore, arterial puncture can be technicallydifficult in very young children. Nonetheless, itshould be strongly considered for Type I ARFpatients when ARDS is suspected (see Chapters 13and 23).

Oxygenation in these patients is monitoredcontinuously and non-invasively using pulseoximetry. It is absolutely crucial during the first 24hours of NIV and preferably would be maintainedthroughout the entire course of treatment. Thishighly simple technique simultaneously measuresoxy-hemoglobin, arterial blood pressure wave andheart rate. Saturation values < 80% are less reliable,but are still clinically useful. Values > 98% mayindicate that PaO2 = 100 mm Hg (in which casehyperoxia would go undetected).

For patients with acute lung injury (ALI) or acuterespiratory distress syndrome (ARDS), who areincluded in the most severe cases of Type I ARD, anda transcutaneous SatO2 < 98%, the hemoglobinsaturation index (SatO2)/FiO2 has proven utile forvaluating the PaO2/FiO2 index, using the followingformula: PaO2/FiO2 = [(SatO2/FiO2) - 64]/0.84. Thisindex can be employed to determine the degree of

intrapulmonary shunt before and after NIVtreatment, and for ARDS patients, can be used forcloser blood gas monitoring to avoid prolongingfailed NIV or preventing the patient from becomingcontra-indicated (PaO2/FiO2 < 150).

Transcutaneous PO2 can also be measuredcontinuously and non-invasively through anelectrode attached to the patient’s skin. Inhemodynamically stable children this value correlateswell with PaO2. However, this method is typicallynot employed due to technical difficulties: calibrationand attachment of the electrode, warming of theskin, long response time, and the need to frequentlyreposition the electrode to avoid burning the patient.It is primarily indicated in neonatal pathologies, asit can prevent hyperoxia. It is also useful in infants,children and adolescents.

The success of NIV can be assessed by usingcapillary PCO2 obtained from a punctured arteryafter heating the site of extraction (e.g. heel, earlobeor fingers). This value correlates well with PaCO2,although it can become bothersome for children ifrepeated several times in a short period. Althoughit is not common practice, venous PCO2 (PvCO2)measurements can be used; these are easy to obtain,since nearly all children tend to already have avenous catheter placed. PvCO2 is 5 to 6 mm Hghigher than PaCO2. Generally, absolute readings arenot evaluated, but rather changes in readings, whichare caused by NIV or by alterations in the patient’sstate.

There are two non-invasive techniques formonitoring PCO2: capnography and transcutaneousPCO2 measurement.

Capnography continuously measures andrecords the concentration of CO2 in exhaled air. ForNIV and spontaneous ventilation, sidestreamcapnographs, which measure outside of therespiratory circuit, should be used. For NIV, a thinpiece of tubing is connected to the exhalation porton the mask, from which the exhaled air travels tothe capnograph. For spontaneous ventilation, thetubing is attached to the nostrils (analogously tooxygen cannulae) or to the mouth (if the child isbreathing through their mouth). The ideal locationwould be the hypopharynx. Exhaled CO2 (end-tidalCO2, or ETCO2) readings and curves in NIV are oftenerroneous to due to the constant flow in bi-levelpressure (BiPAP) systems or to major leakage. If the

76 M.Á. García Teresa, R. Porto Abal, P. Fernández Deschamps

ETCO2 value increases and the baseline does notreach zero, this is a sign of re-inhalation.

Transcutaneous monitoring of CO2 has the sameadvantages and disadvantages as transcutaneousmonitoring of PO2. Measured CO2 values are higherthan those for PaCO2, although in hemodynamicallystable patients these two parameters correlatestrongly (see Fig. 1).

Monitoring of acid-base equilibrium: In somereports on adult patients, the patient’s blood pHat 2 hours of NIV is used as a predictive factor fortreatment success.

Pulmonary mechanics and monitoring of theventilator

For patients treated with invasive ventilation,the curves for flow, volume and pressure, and thevalues for lung compliance and resistance aremonitored using pneumotacographs and pressuresensors whose measurements are analyzed by anddisplayed on the ventilator. Currently, there arecommercially available non-invasive ventilators thatprovide continuous monitoring of pressure, flowand volume curves (Fig. 2). Both conventional andNIV ventilators used in the hospital monitor theestimated tidal volume and minute volume, thepeak pressure, percent of breaths initiated by thepatient, and leakage level, and feature alarms forapnea, low minute volume and high or lowrespiratory frequency. Said alarms should be usedin function of the patient’s age and the treatmentobjectives.

Monitoring of side effects Area in which the interface is supported. These

areas—above all, the nasal bridge—must be watchedwith extreme care. Interfaces tend to be attached tootightly, and children often can not vocalize that theinterface is hurting them. Children have more sensitive

skin than adults: a light rash can develop into a pressuresore within a few days, forcing NIV to be stopped.

Gastric distension. This should be monitored andtreated, since it can worsen the patient’s respiratorystate via restrictive pulmonary compromise.Preventive care using an open nasogastric tube isrecommended.

Leakage. Leaks can be observed clinically (i.e. theyare felt or heard by the patient or cause the patient'shair to move) or, in certain ventilators, measured (seeFig. 3). A leak rate < 30 L/min is preferable for avoidingpressure sores. The airflow from leaks can causeconjunctivitis by drying out the patient’s eyes.

Pneumothorax. This should be monitored inhigh risk patients (e.g. those with chest walltrauma).

Monitoring of non-invasive ventilation for pediatric patients 77

Figure 2. The display of an NIV ventilator showing pressu-re, flow and volume curves for use in monitoring. This ven-tilator also measures the estimated expiratory tidal and minutevolumes, peak pressure, percentage of breaths initiated bythe patient, and leakage.

Figure 1. Transcutaneous PCO2 monitor.

Figure 3. Leakage plotted on an NIV display.

MONITORING OF NON-INVASIVE VENTILATIONIN CHRONIC RESPIRATORY FAILURE PATIENTS

In CRF patients, NIV is used to amelioratesymptoms, improve the patient’s diurnal andnocturnal blood gas levels, better their quality of lifeby improving the quality of their sleep, enhancetheir functional state, reduce the duration of theirhospital stay, preserve their pulmonary function andprolong their life.

In some patients, ventilation is begun for flare-ups (usually caused by an infection), whereas forothers, it is indicated for gradually worsening CRF.

Monitoring at the beginning of NIVClinical observation. As previously explained,

the emphasis here is on evaluating the comfortand adaptation of the child. For home NIVpatients, this process starts in the hospital andcontinues at their home. Patient-ventilatorsynchrony must be observed, as should chest wallmovements. The patient’s chest wall should belistened to with a stethoscope during entry of airin NIV. As mentioned above, leakage must beclosely watched.

Gas exchange. This is measured usingcontinuous pulse oximetry, since for the majorityof pediatric CRF patients requiring NIV, hypoxiawill be indicative of hypoventilation. ETCO2 valuescan also be used, with the same limitations asthose explained for PCO2: the response time istoo slow for detecting temporary periods ofhypoventilation; and the transducer signal can bealtered by the patient’s movements, whichrequires that the detector be frequentlyrepositioned and recalibrated. Neither arterial norcapillary blood gases are typically measured dueto the discomfort that they cause the patient.Alternatively, measurements can be taken from avenous source (treated with anesthetic creambefore being punctured): the variations in PCO2

compared to the baseline throughout the nightcan be evaluated to determine if the child cantolerate a higher IPAP (if needed).

Blood pH. Respiratory acidosis compensated forby a high level of bicarbonate is indicative of chronichypoventilation.

Monitoring of NIV side effects. As previouslymentioned, the highest priority here is protectingthe nasal bridge from pressure sores.

Monitoring of patients at home

OrganizationPatients receiving home NIV must also be

monitored for adaptation, ventilation efficacy andside effects. Indeed, the home ventilation set-uptypically includes pulse oximetry and oxygen therapy.Hence, before the patient is sent home from thehospital, their parents should be taught basicmonitoring skills.

Close monitoring of patients is crucial to the successof NIV. Their parents must have 24 hour telephoneaccess to hospital staff to call for assistance. Thetechnician from the home ventilation company shouldvisit the patients at their home several times duringthe first few weeks of treatment and should advise thepatient's physician of any problems that they detect.

Monitoring in the hospital can be done on anoutpatient or inpatient (one night) basis, accordingto the patient’s needs. The patient can make theirfirst visit to the hospital 4 to 6 after having beenreleased from the hospital, plus additional visits asneeded. Once the patient has been stabilized, theycan be examined with less frequency; dependingon the pathology, they should visit the hospital twoto four times annually. The patient’s nocturnal pulseoximetry should be recorded at home, and basedon the results they should be evaluated for hospitaladmission to make adjustments to their ventilation.

Monitoring should be especially vigilant forinfants and small children (15 days after having beenreleased from the hospital, and then monthly duringtheir first year of life), as these patients requireperiodic adjustments to their ventilator and mask.

Pediatric patients often present withcomplications associated with NIV and consequently,require multidisciplinary care (e.g. cardiology,pulmonology, neurology, rehabilitation, nutritionand traumatology). Hence, their appointments withdifferent specialists should be coordinated.

Methodology

Improving symptoms and signsThe patient should be monitored for improvements

in or resolution of symptoms which are signs ofsuccessful NIV. Adolescents present with the samesymptoms as adults (morning headache, difficultyconcentrating, enuresis, severe exhaustion, and

M.Á. García Teresa, R. Porto Abal, P. Fernández Deschamps78

hypersomnolence) and may suffer from dyspnea. Veryyoung children tend to show irritability, psychomotordelay, low academic performance, enuresis,malnutrition, and poor sleep, in which they often wakeup from nightmares. They may present withsomnolence or show signs of pulmonary hypertensionor cor pulmonale (e.g. hepatomegaly or edemas).

Improved or normalized lung pressure, asdetermined by echocardiography, and gaining ofweight are both indicative of NIV success.

Gas exchangeGas exchange should be measured diurnally and

nocturnally to determine the effects of NIV. For adultNMD or kyphoscoliosis patients, effective NIVprovides major improvements in diurnal PaO2 andPaCO2.

Oxygenation is evaluated using pulse oximetrywhile the patient is awake as well as when they areasleep. Appearance of desaturation during NIV maybe due to persistent hypoventilation, intermittentairway obstruction, excessive leakage (due to maskadjustments), or patient-ventilator asynchrony.

The efficacy of NIV can be gauged bycapnography while the patient is not connected tothe ventilator. Normalization of diurnal PCO2 isneither required nor desirable in NIV, as it may implyan increase in the IPAP that the child can not tolerate,or extended use of the ventilator, which the childmay not accept. No ideal value of PCO2 has beenestablished, as it varies with each patient. In somechildren, the symptoms of cor pulmonale mayimprove, and the signs may disappear, despite thefact they have a PCO2 of 50 to 60 mm Hg.

In order to achieve normoventilation in childrenwith central sleep hypoventilation—whetheridiopathic (e.g. Ondine’s Curse) or secondary (e.g.Arnold Chiari malformation, medullar lesion, etc.)—gas exchange must be measured while the patientis sleeping.

Quality of life of the patient and their familyHospital staff should assess the quality of life of

the patient and their family by asking them specificquestions on health (nutrition, infections, and hospitaladmissions), functionality, autonomy, academicprogress, socialization and activities (e.g. playing,or doing sports). The quality of the patient's sleep canbe indirectly determined by gauging improvements

in symptoms or via nocturnal pulse oximetry orpolygraphy at home. Full polysomnography is nota routine monitoring technique: it is very expensive,only available at a few centers and in high demand.However, some authors have reported using it toadjust ventilation in complicated cases.

Pulmonary functionPatients should be monitored by spirometry as

often as needed according to their age, level ofcooperation, pathology, and rate of respiratorydeterioration. NMD patients with vital capacity < 40%of the theoretical value may begin to retain CO2; iftheir condition worsens despite treatment, then theyshould be ventilated for longer periods.

Side effects and complicationsIn addition to the side effects and complications

described in this and other chapters, long-termventilation patients should be observed for possiblefacial bone deformities, and those treated with oralinterfaces should be monitored for dental damage.

REFERENCES1. Bernet V, Hug MI, Frey B. Predictive factors for the success of

noninvasive mask ventilation in infants and children with acuterespiratory failure. Pediatr Crit Care Med 2005;6:660-4.

2. Essouri S, Chevret L, Durand P, et al. Noninvasive positive pres-sure ventilation: Five years of experience in a pediatric intensivecare unit. Pediatr Crit Care Med 2006;7:329-34.

3. Fauroux B, Boffa C, Desguerre I, et al. Long-term noninvasivemechanical ventilation for children at home: A national Sur-vey. Pediatric Pulmonology 2003;35:119-25.

4. Hill NS. Management of long-term noninvasive mechanical ven-tilation. En: Hill NS (ed). Long term mechanical ventilation. Mar-cel-Dekker, New York, 2000.

5. International Consensus Conferences in Intensive Care Medici-ne: noninvasive positive pressure ventilation in acute Respiratoryfailure. Am J Respir Crit Care Med 2001;163:283.

6. Liesching T, Kwok H, Hill NS. Acute applications of noninvasivepositive pressure ventilation. Chest 2003;124:699-713.

7. Mueller GA, Eigen H. Pulmonary function testing in pediatricpractice. Pediatr Rev 1994;15:403-11.

8. Ruza Tarrio F, De la Oliva Senillosa P. Monitorización: mediciónde gases, mecánica ventilatoria. En: Ruza F (ed). Tratado de Cui-dados Intensivos Pediátricos, 3ª ed. Norma-Capitel, Madrid, 2003.

9. Teague WG. Noninvasive ventilation in the pediatric intensivecare unit for children with acute respiratory failure. Pediatr Pul-monol 2003;35:418-26.

10. Rice TW, Wheeler AP, Bernard GR, Hayden DL, Schoenfeld D, WareLB. Comparison of the SpO2/FiO2 ratio and the PaO2/FiO2 inpatients with acute lung injury or ARDS. Chest 2007;132: 410-7.

Monitoring of non-invasive ventilation for pediatric patients 79

COMPLICATIONSComplications in non-invasive ventilation (NIV)

are defined here as any adverse effect in the patientwhich arises after treatment has begun and whichcan be attributed exclusively to the treatment.

EPIDEMIOLOGYReported rates of complications in adult patients

range from 10 to 20%. The most frequentcomplication is skin necrosis where the interface issupported, which accounts for 70% of allcomplications. Gastric distension, pneumothorax andaspiration each do not occur in more than 3% of NIVpatients. The findings of the EPIVENIP epidemiologicalstudy, performed in Spanish pediatric intensive careunits (PICUs) from 2004 to 2005, correlate well withthe aforementioned figures (Fig. 1).

COMPLICATIONSComplications in NIV can be classified into five

groups:

Complications due to the interfaceSkin irritation. This occurs in the area where the

interface is supported, typically in the nasogenianarea. Periodic use of moisturizing creams can helpalleviate this irritation.

Skin sores. Until very recently, this was one of themost common problems in patients treated with

continuous NIV. Although the most frequent locationis the nasal bridge, skin necrosis can appear at anypoint in which the interface is supported. The rateof skin necrosis has been markedly reduced throughthe use of special anti-sore dressings (e.g. Comfeel®).Home NIV patients can be further protected bealternating between different interfaces or by usingan Adams-type interface (Fig. 2).

Conjunctivitis. Air leaking out from the sides ofmasks can cause conjunctivitis and even corneal sores.This can be prevented or treated by adjusting themask or changing to an interface which better suitsthe patient's face. Special care should be taken forpatients with lagophthalmos (Fig. 3). Some nasalinterfaces contain a bubble-like similar to the ComfortFlap™ by Respironics, which, upon being filled bythe inspiratory airflow, molds itself to the patient’sface to minimize leakage (Fig. 8, chapter 5).

Airway obstruction. Tracheal tubes used as nasal-pharyngeal interfaces in infants younger than 3months can obstruct the nostrils if the permeabilityof the tubes and of the nostrils is not adequatelyprotected; hence, hospital staff should take specialcare with these patients. Likewise, the inner plasticlayer of nasal and oral-nasal interfaces can block thenasal fossa; this can be resolved by trimming theplastic (see Fig. 4).

Maxillar hypoplasia. Home ventilation over severalyears has been associated with maxillarunderdevelopment and secondary malocclusion ofthe cheekbone.

Complications and technical difficulties in non-invasive ventilation

M. Pons, M. Balaguer

Chapter 11

Hypercapnia. Oral interfaces with large deadspace carry a higher risk of hypercapnia, especiallyfor small children with high respiratory frequency.Currently, for small infants, medium sized-nasalinterfaces are used as oral-nasal interfaces andprovide good results. For the patients describedabove, or for those with Type II ARF, vented interfacesare preferable to non-vented ones; if these are notavailable, then a Plateau valve can be used in placeof the expiratory valve as a preventative measure.

Complications related to the interface straps,belts or harness

Accidental disconnection. In patients who arehighly dependent on ventilatory support, accidentaldisconnection is as dangerous as accidentalextubation: they may suffer from hypoxemiasecondary to alveolar derecruitment, plus relatedproblems. Despite not being described in theliterature, this is a very real danger.

Axillary vein thrombosis. Helmet interfaces areattached through belts under the armpits, andtherefore, imply a risk of thrombosis of the axillaryvein. As a preventative measure, some centers attachweights to the belts to minimize the pressure on theaxillary zone. Helmets designed for infants featurea special harness that is supported around the diaperarea (see Chapter 5, Figure 19).

Basal ischemic lesions. In neonates, interfaceswhich have been attached too tightly have led to

ischemic lesions: due to bone plasticity, the excessivepressure from the straps or belts compresses thecerebral basal circulation.

Complications related to pressure in the airwayInspiratory blockage. This phenomenon is related

to a glottal reflex which closes the vocal chords in theevent of hypocapnia due to excessive pressure. Theincrease in resistance in the airway tends to facilitateleakage and patient-ventilator desynchronization. It

82 M. Pons , M. Balaguer

Figure 1. Results of the EPIVENIP study on complications inpediatric non-invasive ventilation.

100

80

60

40

20

0

92

112 2 1 1 1 1

Non

e

Der

mat

itis

Nas

al h

emor

rhag

e

Con

junc

tiviti

s

Gas

tric

dis

tens

ion

Pneu

mon

ia

Pneu

mom

edia

stin

um

Obs

t. o

f pro

ngs

Nº

of c

ases

of c

omp

licat

ions

Figure 2. Adams interface.

Figure 3. Protecting the eyes of a child with lagophthalmosfrom the effects of air leaks.

is critical to anticipate this complication, which isresolved by lowering the inspiratory peak airwaypressure (IPAP).

Gastric distension. This complication tends toarise only when inspiratory pressures > 25 cm H2Oare used; however, in neuromuscular disease(NMD) patients, it can occur with inspiratorypressures < 20 cm H2O due to the weakness of theirdiaphragm and lower esophageal sphincter (LES).Gastric distension can be dangerous because itcarries a risk of vomiting. In one exceptional case,it was associated to compartmental abdominalsyndrome with hemodynamic deterioration;however, the patient was a 65 year old who hadrefused to accept insertion of a nasogastric tubebefore beginning NIV.

Food aspiration. There is a risk of vomiting andfood aspiration for patients fed orally or vianasogastric tube, especially those treated with a facemask. This risk can be exacerbated by use of sedationto improve patient adaptation; hence, hospital staffshould take precautions such as temporarilysuspending enteral feeding or intubating the patientwith a transpyloric tube.

Pneumothorax. Pneumothorax has been mostwidely reported in chest wall trauma patients withfractured ribs. However, according to the EPIVENIPstudy of 2005, it only occurs in less than 1% of allPICU NIV patients (see Fig. 1). Recent studysdescribed a percentage of 3%.

Air embolism. Neurological lesions brought onby air embolism have been described in patientssuffering from pneumothorax.

Pulmonary hyperinflation. This can occur in theevent of dynamic hyperinflation caused byadministering an expiratory positive airway pressure(EPAP) higher than the patient’s intrinsic positive-end expiratory pressure (PEEP).

Hemodynamic deterioration. This is rare, but a dropin preload and hypotension, can be detrimental forhemodynamics, especially in hypovolemic patients.

Complications related to humidificationDue to lack of humidification. Humidification of

the airflow leads to alteration of the nasal mucosawith blockage of ventilation, leading to a loss oftreatment efficacy.

In home-NIV patients with oral leakage, a lackof humidification has been shown to increase nasal

resistance, diminish the tidal volume and cause thepatient discomfort.

Lack of humidification can also lead to formationof mucous layers that can facilitate airway obstructionand NIV failure and complicate intubation.

Due to excess humidification. Patients exhibit poortolerance to conventional mechanical ventilation(CMV) humidifiers when the target temperature isset to 39 °C.

Complications related to NIV indicationsThe most severe complications in NIV are the

result of trying to treat contra-indicated patients;fortunately, this rarely occurs.

In adult ARF patients, a higher mortality rate hasbeen observed for those that first undergo a trialperiod of NIV before being intubated than for thosethat start with CMV. In adult patients with acutepulmonary edema (APE) of cardiac origin, a higherrate of heart attacks has been observed for thosewho are treated with NIV than for those who areonly treated with continuous positive airway pressure(CPAP).

In NIV patients who have had recent esophagealor gastric surgery, dehiscence is a potentialcomplication. Nonetheless, some authorsrecommend use of NIV for morbidly obese patientsthat have had gastroplasty, since these patients havean elevated risk of respiratory complications in thepostoperative period.

Esophageal-pleural fistula has been described inan NMD patient that had suffered a spontaneousesophageal rupture secondary to Boerhaave’ssyndrome three years before beginning home NIVtreatment.

Orbital herniation has been described in a fewpatients with ethmoidal fractures. It should bereiterated that facial fractures are a contra-indicationto NIV.

TECHNICAL PROBLEMS

Non-invasive ventilation using a conventionalventilator

This is rarely used today. Technical problems withthis method are due to the fact these ventilators arenot designed to handle strong leakage (which isintrinsic to NIV).

83Complications and technical difficulties in non-invasive ventilation

Expiratory volume alarmBackflow of expiratory gas in the tubing is

minimal; hence, this alarm should either be set tothe minimum or turned off. In ventilators in whichthe alarm can not be shut off, an external airflow oroxygen flow can be added to the expiratory end ofthe tubing to prevent the alarm from sounding (seeFig. 5).

AutocyclingNon-compensated leaks in the ventilator cause

a drop in flow that the ventilator erroneouslyinterprets as inspirations, leading to autocycling. Ifthe leaks can not be improved by adjusting theinterface, then the inspiratory trigger should be setto the minimum to prevent it from inadvertentlygoing off.

Opening of the anti-asphyxia valveOral-nasal interfaces from certain manufacturers

(e.g. Respironics and Hans Rudolph) feature an anti-asphyxia valve which serves as a security mechanismwhen the ventilation system loses pressure. One ofthe consequences of non-compensated leaks is thatthe PEEP can not be maintained, which then causesthe anti-asphyxia valve to open. Albeit this problemcan be minimized by increasing the PEEP, the bestsolution is to avoid using these interfaces withconventional ventilators.

DesynchronizationMany conventional ventilators lack an adjustable

expiratory trigger, which implies that active work is

needed to start expiration: the patient must fightagainst the ventilator, causing them greaterexhaustion instead of relaxation. For these ventilators,synchrony can be improved by shortening theinspiratory time.

Desynchronization can also occur if the ventilatorfails to detect the patient’s inspiratory efforts, as isoften the case with small infants or with ventilatorsthat lack a flow trigger. In these scenarios, the bestmode to use is pressure assisted/controlledventilation (PACV), using a respiratory frequencysimilar to the patient’s spontaneous respiratoryfrequency.

Lastly, desynchronization can also arise if thepatient’s airway is blocked with secretions that theycan not expel; hence, they must be checked for thispossibility.

In conventional ventilators that feature an NIVoption, adjustment of the expiratory trigger oftenenables optimum synchronization; in those that lackthis option, the best strategy is to use modesequivalent to PACV.

Non-invasive ventilation using a non-invasive-specific ventilatorI

Despite providing better support for leakagethan conventional ventilators, NIV-specific ventilatorsstill have their drawbacks.

Oxygen therapyFor models that lack an oxygen blender or an

oxygen input valve (e.g. VPAP II, BiPAP S/T and BiPAPHarmony), the airflow delivered to the patient must


Figure 4. Profile Lite interface modified for use as an oral-nasal interface by trimming its internal membrane.

Figure 5. Connecting an external oxygen flow to the expi-ratory end of the ventilator.

be enriched with oxygen either through the tubingor through one of the inlets available on theinterface. The maximum FiO2 possible with thismethod is 0.5. Some newer models (BiPAPSynchrony, VPAP III, VsUltra, and Legendair) featurean oxygen input valve. For these models, themaximum FiO2 possible is 0.8; the FiO2 can becalculated based on data tables provided by themanufacturers.

Positioning an oxygen T-piece close to theventilator-end of the tubing minimizes turbulencebut provides a lower concentration of oxygen.Contrariwise, positioning the T-piece at the interface-end will provide a greater FiO2 but greaterturbulence. If the patient’s hypoxemia can not becompensated for by this type of oxygen input, theneither a ventilator equipped with an oxygen blenderor a conventional ventilator should be used.

HypercapniaUsing a single arm circuit facilitates re-inhalation

of CO2, above all in patients with tachypnea. Usingan EPAP of at least 4 cm H2O allows exhaled air tobe forced out of the leak-control area. If this provesinsufficient, then the EPAP can be increased to upto 8 cm H2O; or the fragment of tubing in whichthe expiratory valve is located can be cut, and thenreplaced with a Plateau valve (see Chapter 13, Figs.2 and 3). Some authors have recommended addingan oxygen flow of 4 to 6 L/min into the interface ornear the expiratory valve to diminish re-inhalationof CO2. Alternatively, hospital staff can switch to aninterface which has less dead space or which featuresexhalation ports, provided that the patient adaptsto it. All of the aforementioned methods shouldbe considered, keeping in mind that the tidal volume

administered to the patient has been optimized ata particular IPAP.

DesynchronizationThe most frequent cause of desynchronization

with NIV-specific ventilators is excessive flow usedto compensate for leakage from a poorly adjustedor sized interface. As explained above forconventional ventilators, desynchronization canoccur if the ventilator fails to recognize the patient’sinspiratory efforts. This is especially true for infantsyounger than 6 months and for ventilators withimpedance-based triggers. The best option forthese scenarios is to use controlled modes (S/T orT), adjusting the respiratory frequency accordingto the patient’s spontaneous respiratory frequency.Another cause of desynchronization is inadequateinspiratory ramp time: in infants, if the time is tooshort (0.05 seconds), then it will lead to patientdiscomfort from elevated initial flow; inadolescents, if the time is too long (0.4 seconds),then it will make patient feel that they can notdraw in air, or it will slow the response of the

85Complications and technical difficulties in non-invasive ventilation

Figure 6. Oxygen T-piece for enriching the ventilator airflowwith oxygen.

Figure 7. Plateau valve.

OXYGENINLETS

PLATEAUVALVE

inspiratory trigger. Sedation should always be alast resort for achieving good patient-ventilatorsynchronization.

REFERENCES1. Hill NS. Complications of noninvasive ventilation. Respir Care

2000;45:480-1.

2. García Teresa MA, Martín Barba C. Ventilación no invasiva. EnManual de Cuidados Intensivos Pediatricos. Ediciones Publimed.Madrid 2001;655-660.

3. British thoracic Society standards of care committee. Non-invasi-ve ventilation in acute respiratory failure. Thorax 2002;57:192-211.

4. Akingbola OA, Hopkins RL. Pediatric non invasive positive pres-sure ventilation. Pediatr Crit Care Me 2001;2:164-169.

5. Padkin AJ, Kinnear WJ. Supplemental oxygen and nasal intermit-tent positive pressure ventilation. Eur Resp J 1996;9:834-836.

6. Ferguson Gt, Gilmartin M. CO2 rebreathing during BiPAP venti-latory assistance. Am J Resp Crit Care Med 1995;151:1126-1135.

7. Hurst JR, Polkey MI, Goldstraw P, Yung B. Oesophagopleural fis-tula as a novel cause of failed non-invasive ventilation. Thorax2003;58:642-643.

8. Bart L. DE Keulenaer, Adelard de Backer, Dirk R Schepens et al.Abdominal compartment syndrome related to non invasive ven-tilation. Intensive Care Medicine 2003;29:1171-1181.

9. Hung S-C, Hsu H-C, Chang S-C. Cerebral air embolism com-plicating bilevel positive airway pressure therapy. Eur Respir J1998;12:235-237.

10. Corrales E, Pons M, López-Herce J, Martinón-Torres F, García MA,Medina A, González Y. Estudio epidemiológico de la ventilaciónno invasiva en las Ucip de España. XXII Congreso Nacional de laSECIP, 2005. An Pediatr (Barc) 2007;67:91-100.

11. Tuggey JM, Delmastro M, Elliott. The effect of mouth leak andhumidification during nasal non-invasive ventilation. Respir Med2007;101:1874-9.

12. Jounieaux V, Aubert C, Dury M, Delguste P, Rodestein PO. Effectsof nasal positive pressure hyperventilation on the glottis in nor-mal awake subjects. J Appl Physiol 1995;79:176-85.

13. Jounieaux V, Aubert C, Dury M, Delguste P, Rodestein PO. Effectsof nasal positive pressure hyperventilation on the glottis in nor-mal sleeping subjects. J Appl Physiol 1995;79:186-93.



INTRODUCTIONCompared to conventional mechanical ventilation

(CMV), non-invasive ventilation (NIV) offers lesscomplications (e.g. volutrauma, barotrauma,pneumonia, and laryngo-tracheal sores) and lowerrequirements for sedation in critical pediatric patients.Using less sedation helps conserve the patient’s airwaydefense mechanisms, their capacity to speak andswallow, and their cough reflex, which enables themto spontaneously eliminate secretions. In certain cases,it also helps the patient maintain their ability to befed orally. Nonetheless, NIV often demands that thepatient be sedated. However, literature on sedationin NIV, such as pharmacological guidelines on the useof sedatives, is scarce.

A recent study performed at HospitalUniversitario Central de Asturias (HUCA) revealedthat 63% of pediatric NIV patients required sometype of sedation. Among sedated patients, 84% hadto be sedated by continuous infusion, and 16%, byperiodic bolus.

Sedatives must be used with extreme caution inNIV, as they can diminish the patient’s respiratoryefforts, thereby reducing their state of consciousness(a contra-indication to NIV), and consequently,causing the treatment to fail.

PRINCIPLES Firstly, a clear distinction between sedatives and

analgesics must be made, as the majority of sedativesdot not have analgesic effects. Relieving the NIVpatient of their pain will help improve theirrelationship with hospital staff and help them to

better tolerate the interface and treatment. Hospitalstaff should likewise care for patients by treating anyfever, alleviating hunger or thirst, and ensuring thatthey do not feel too hot or cold, all of which aresensations which can cause discomfort and stress,especially in very young children.

Since NIV is intended to alleviate any anxiety oragitation felt by the patient from respiratorypathologies, it is crucial that other factors influencingcomfort be separately evaluated and treated (e.g.neurological or metabolic problems, pain). Table Ilists measures for minimizing the need for sedationin pediatric NIV patients suffering from anxiety oragitation. The expertise and attention of hospitalstaff are fundamental for successful treatment (seeChapter 9).

Pharmacological sedation comprises two levels:1. Conscious sedation, or anxiolysis, in which the

patient’s level of consciousness is minimallyaffected, their airway remains open, their reflexesare conserved, and they are able to maintainadequate physical and verbal response; This isthe preferred level for NIV.

2. Deep sedation or hypnosis, in which the patient’slevel of consciousness is markedly reduced andcontrolled by drugs, and they can not be easilyawoken and may lose their protective reflexesand their ability to respond to physical or verbalstimulation. This level should only be usedperiodically for NIV and with close observationof the patient.For any pharmacological sedation strategy,

hospital staff must ensure that the patient will beadequately monitored. Hence, acute patients are

Sedation in non-invasive ventilation

M. Los Arcos, J. Mayordomo and M. Gáboli

Chapter 12

normally started on NIV treatment in the pediatricintensive care unit (PICU) (see Chapter 4).Monitoring of NIV patients is covered in Chapters3 and 10. If the patient is to be sedated deeply, thenhospital staff may consider monitoring the sedationby employing the Bispectral® (BIS®) index, which iswidely used in anesthesia.

BIS® is most valuable in pediatrics for monitoringsedation in short, painful procedures, and is beingincreasingly used for monitoring chronic patients.Not all sedatives induce an electroencephalographicresponse which can be translated into a drop in BIS®

index value. Those which do generate this type ofresponse include benzodiazepines (e.g. midazolam)and propofol. It has been clearly demonstrated thata BIS® value between 60 and 80 enables sedationwith amnesia, with the possibility of maintainingairway reflexes and, in the absence of analgesia, theability to respond to physical or pain stimuli. To avoidany undesired side effects of sedation, the BIS® valueof sedated NIV patients should be monitored so thatit does not drop below said range.

The expertise of hospital staff is fundamentalwhen choosing and administering a sedative. Eachhospital ward will most likely have in-housepediatric sedation protocols that can be adaptedto NIV, especially to the start of treatment. Thepatient’s age is another decisive factor in the choiceof sedative.

Lastly, pharmacological sedation is not indicatedfor chronic NIV, especially in the home, unless itforms part of a palliative course of treatment.

INDICATIONS OF SEDATION IN NON-INVASIVEVENTILATION FOR ACUTE RESPIRATORY FAILURE PATIENTS

No clear indications on the use of sedatives inpediatric NIV patients have been established fromany clinical studies. Pharmacological sedation istypically required at the onset of NIV, as support fornon-pharmacological methods and to ensure thattreatment begins correctly.

The authors of this chapter, based on the limitedpublished evidence and on their own experience, havecompiled a list of scenarios in which pharmacologicalsedation can help in the administration of NIV: • Adapting the patient to the interface: Positioning

the interface on the child can cause them

discomfort. In some cases, an initial bolus issufficient: upon tranquilization, the patient willfeel that their breathing has improved, andconsequently, better adapt to treatment.

• Improving patient-ventilator synchronization:Sedation can help alleviate synchronizationproblems related to ventilator triggers, especiallyfor very young children. Hence, infants are thepatient group which most requires sedation forinitiating NIV. Data on the patients of the authorsof this chapter reveal that 68% of infants under24 months require sedation, as compared only53% of those over 24 months.

• Controlling hypoxic agitation: Hypoxemia canagitate respiratory failure (RF) patients. Hence,administration of a sedation bolus at the onsetof NIV can help these patients adapt totreatment. Contrariwise, hypercapnia may inducelight somnolence in RF patients, which canactually be valuable for initiating NIV.

• Protecting the lungs by reducing air trapping: Theagitation inherent to the onset of NIV treatmentor due to alterations of the interface (e.g.repositioning, or aspiration of secretions) canfacilitate or even lead to a higher level oftachypnea, greater air trapping or strongeroscillations in pressure. Depending on thepatient, their pathology, and the ventilationmode used, these effects can have severeconsequences. Periodic or continuous sedationof these patients may reduce this risk.

88 M. Los Arcos, J. Mayordomo, M. Gáboli

Table I. Non-pharmacological sedation methods for pedia-tric non-invasive ventilation.

• Whenever possible, allow a parent to accompany thechild. Encourage physical and verbal contact with thepatient while projecting a sense of calm

• Provide the patient with access to familiar objects (e.g.dolls, pets, pacifiers and books) and music

• Respect the patient’s sleep-wake cycles• Maximize the patient’s comfort level (e.g. temperatu-

re, clothing, light, noise, etc.)• Ensure that hospital personnel present themselves in

the least aggressive manner possible, instilling the patientwith a sense of security rather than of fear. Avoid nega-tive speech

• Minimize the aggressiveness of treatment methods suchas extractions and examinations

• Emphasize non-invasive monitoring techniques

DRUGSNIV patients are typically sedated with drugs

commonly used in the ICU. The dose used shouldbe the minimum quantity required to enableadaptation of the patient to NIV without causingany interference in their respiratory efforts. Sedativesare administered intravenously via periodic bolus orcontinuous infusion. Sometimes, a single injectionat the onset of treatment is sufficient for couplingthe patient to the ventilator. However, for cases inwhich adaptation to NIV is more laborious (e.g.treatment of small infants), continuous infusion maybe required.

The most common sedatives for pediatric NIVare described below and summarized in Table II.1. Midazolam. Midazolam is the most widely used

benzodiazepine for sedation of children andadults, owing to its short half-life (45 to 60min) and its minor side effects. Typical dosesrange from 0.05 to 0.2 mg/kg for direct bolusand from 0.05 to 0.3 mg/kg for continuousinfusion. For patients that can not be treatedintravenously, midazolam can be administeredeither intranasally or orally, as it functionsrelatively quickly by these routes. However, ifit is administered sublingually or intranasaly,the dose should be doubled, and if it isadministered orally or rectally, the dose shouldbe tripled.

2. Fentanyl. Fentanyl is a synthetic opioid whichis 100 times more potent than morphine andhas fewer side effects. It has a relatively shorthalf-life (30 to 60 minutes). Typical doses rangefrom 1 to 3 µg/kg for direct bolus and from 0.5to 4 µg/kg for continuous infusion.

3. Ketamine. Ketamine is a potent hypnotic agentand analgesic with a dissociative effect. It hasminimal effects on respiratory function butdoes produce a bronchodilator effect byrelaxing smooth bronchial muscles. However,it increases production of secretions, whichcan compromise NIV in patients that havedifficulty eliminating secretions (atropine [0.01mg/kg] can be co-administered with Ketamineto prevent sialorrhea and bronchorrhea). Themost important side effects of Ketamine arehallucinations and delirium; these can generallybe prevented by co-administering the drugwith low doses of benzodiazepines (e.g.midazolam [0.05 to 01 mg/kg]). Ketamine hasa short half-life: when administered by IV, ittakes effect in less than one minute andremains active for 15 to 20 minutes. The typicalIV dose of Ketamine is 0.5 to 1 mg/kg deliveredin 2 to 3 minutes (repeatable), and eventually,0.25 to 2 mg/kg/h. Ketamine is contra-indicated for NIV in patients with poorlycontrolled hypertension, hepatic failure,

89Sedation in non-invasive ventilation

Table II. Most widely used sedatives in pediatric non-invasive ventilation

Drug IV Dose (mg/kg) Indications & Comments

Midazolam 0.1 to 0.2 Sedative, anxiolytic, amnesiac and non-analgesicMinimal effects on hemodynamicsRapid administration leads to respiratory arrestEffects are reversible with flumazenile

Fentanyl 0.002 Very strong analgesic; sedativeMinimal effects on hemodynamicsHigh dose or rapid administration leads to chest wall rigidityEffects are reversible with naloxone

Ketamine 0.5 to 2 Strong analgesic; amnesiac and hallucinogenBronchodilator (indicated for asthmatics), induces sialorrhea and bronchorrhea (preventable with atropine), hallucinations (preventable with benzodiazepines) and laryngospasm (highly problematic in NIV)

Propofol 1-2 Potent hypnoticMinimal effects on hemodynamics or respiratory efforts at this doseNot recommended for children under 3 years

Remifentanilo Remifentanil Bolus delivery not recommended.Takes effect within 1 minute; effect wanes 3 to 5 minutes after stopping IV

aneurysms, and pulmonary hypertension. Itshould be used with extra caution in patientsthat have airway infections orlaryngotracheomalacia, due to the risk oflaryngospasm. Ketamine can be alsoadministered intramuscularly, in which casethe dose should be doubled; or orally, rectally,nasally or sublingually, in which case the doseshould be tripled or quadrupled and wherebythe absorption is slower and less predictable.

4. Propofol. Propofol is a fast-acting sedative usedfor short periods of deep sedation. Typical dosesare 1 to 2 mg/kg for an initial bolus and 0.5 to4 mg/kg/h for continuous infusion. Themanufacturer of propofol does not recommendusing the drug for more than 48 hours in smallchildren or in doses greater than 4 mg/kg/h,due to a risk of lactic acidosis with myocardialfailure, which is potentially fatal. Despite thelarge body of clinical experience with the drugin young children, including infants, themanufacturer does not recommend its use inchildren under 3 years old.

5. Remifentanil. Remifentanil is an extremely fast-acting (1 minute) synthetic opioid whoseeffects do not last as long as those of Fentanyl.Administration of remifentanil by continuousinfusion is recommended over bolus injectiondue to a high risk of apneas in the latter. Afterinfusion is stopped, the drug’s effects wanequickly (3 to 5 minutes), and it does notaccumulate in tissue. The typical dose for

continuous infusion is 0.05 to 0.5 µg/kg/minwith or without an initial bolus of 1 µg/kg.

6. Chloral hydrate. Chloral hydrate is widely usedfor painless pediatric procedures. It can beadministered orally or rectally. Its mainadvantage is its long latency (30 to 60 minutes);its half-life is 10 hours. Typical doses range from25 to 50 mg/kg, without exceeding 1 g/doseand 2 g/day.

7. Anti-psychotics (neuroleptics). Mild anti-psychotics(e.g. levomepromazine) can also be used assedatives. The typical dose is 1 mg/kg, whichcarries a risk of causing extrapyramidalsymptoms.

REFERENCES1. Krauss B, Green SM. Sedation and analgesia for procedures in

children. N Engl J Med 2000;342:938-945.

2. Playfor S, Jenkins I, Boyles C, Choonara I, Davies G, HaywoodT et al. Consensus guidelines on sedation and analgesia in cri-tically ill children. Intensive Care Med 2006;32:1125-1136.

3. Devlin JW, Nava S, Fong JJ, Bahhady I, Hill NS. Survey of seda-tion practices during noninvasive positive-pressure ventilation totreat acute respiratory failure.Crit Care Med 2007;35:2298-302

4. Jakob SM, Lubszky S, Friolet R, Rothen HU, Kolarova A, TakalaJ. Sedation and weaning from mechanical ventilation: effects ofprocess optimization outside a clinical trial. J Crit Care2007;22:219-28.

5. Constantin JM, Schneider E, Cayot-Constantin S, Guerin R, Ban-nier F, Futier E, et al. Remifentanil-based sedation to treat nonin-vasive ventilation failure: a preliminary study. Intensive Care Med2007;33:82-7.

90 M. Los Arcos, J. Mayordomo, M. Gáboli

INTRODUCTION Introducing non-invasive ventilation NIV into

the ICU without proper training of personnel canlead to a high rate of treatment failure withdangerous consequences, especially in the first hourof ventilation. Unprepared hospital staff may thendevelop an aversion to the technique, therebylimiting its future use. Furthermore, even after anICU has surpassed the learning curve of adoptingNIV, it still must face myriad other factors that canlead to treatment failure (Table I). This chapter hasbeen written as a practical checklist for preventingor correcting failure of NIV. It includes descriptionsof clinical situations in which the patient's stateeither worsens or remains stagnant duringtreatment.

PROBLEMS ANALYSIS

DesynchronizationDesynchronization can occur because of

problems with either the interface or the ventilator:

Inadequate interfaceFor acute patients, especially those who are

hypoxemic, nasal interfaces lead to a higher rate oftreatment failure. These patients frequently breathethrough their mouths, generating leakage which isthen compensated for by the ventilator, ultimatelycausing patient discomfort and leading to patient-ventilator desynchronization. This can be remedied by:• Using a pacifier to minimize leakage through the

mouth; a sweetener may be added to it to makeit more desirable to the child.

• Excessive controlled leakage. Sometimesvented interface can produce autocycling anddesynchronization.

• Changing the type of nasal interface or switchingto an oral-nasal interface.

Non-invasive-specific ventilatorsThe most common causes of desynchronization

are use of excessive flow to compensate for leakageand use of a poorly adjusted or incorrectly sizedinterface.

Another cause is failure of the ventilator to detectthe patient’s respiratory efforts, which often occurswith patients younger than 6 months or whenimpedance triggers are used.• In these situations, the best option is to use

controlled modes (S/T, T), adjusting the respiratoryfrequency to the patient’s spontaneous respiratoryfrequency.

• Sedation may be useful for achieving bettersynchronization and improving the patient'stolerance of the ventilation.When using a ventilator that features differentmodes, the preferred mode is the one which willafford the best overall synchronization.

• As a general rule, for patients older than 4 to 6months, NIV-specific ventilators are superior toconventional ventilators with an NIV option.However, the inspiratory (flow) trigger inconventional ventilators can detect and

Failure analysis and predictive factorsin non-invasive ventilation

M. Pons, A. Medina, F. Martinón-Torres and J. Mayordomo

Chapter 13

synchronize the respiratory efforts of youngerpatients.Another cause of desynchronization is inadequateramp time: if it is too short (0.05 sec), it will leadto excessive initial flow, causing the patientdiscomfort; in contrast, if it is too long (0.4 sec),the patient will feel short of breath or will sufferfrom uncompensated tachypnea due to failureto reach the target inspiratory positive airwaypressure (IPAP) or due to discomfort from theslow response of the inspiratory trigger.

• Use of a humidifier in the respiratory circuit candiminish the sensitivity of the inspiratory trigger,especially in the case of infants.

Conventional ventilators with non-invasive optionThe problems most frequently encountered with

these ventilators are listed below:• Insufficient leak compensation

– This can be resolved by increasing the positive-end expiratory pressure (PEEP) to providegreater flow.

• Insufficient expiratory trigger sensitivity. Thisextends the inspiratory time after the patient hasalready begun expiration (Fig. 1; and see Chapter

7), and is especially problematic in pressuresupport mode.

• This can be resolved by adjusting the expiratorytrigger and, if possible, limiting the inspiratorytime.

• In assisted/controlled (A/C) mode, this can beresolved by adjusting the inspiratory time closeto the patient’s value.

Inadequate etiological treatment of the cause ofrespiratory failure

ICU personnel understand that most of thetechniques they employ are simply methods for gainingtime and ensuring that the efficacy of the etiologicaltreatment, together with the natural course of theillness, enable patients to recuperate. Hence, if thepatient does not improve as predicted, the treatmentstrategy (e.g. antibiotics, diuretics)—and even the initialdiagnosis—may need to be reconsidered.

SecretionsFor patients that have difficulty eliminating

secretions (e.g. neuromuscular disease patients),NIV can easily fail if it is not supplemented with

92 M. Pons, A. Medina, F. Martinón-Torres, J. Mayordomo

Table I. Systematic checklist for possible failure of non-invasive ventilation

1. Desynchronizationa. Inadequate interfaceb. NIV-specific ventilator:

• Leakage in the interface• Inspiratory trigger not activated• Inadequate respiratory circuit• Inadequate ramp

c. Conventional ventilator:• Insufficient leakage compensation• Correct expiratory trigger

2. Confirm adequate etiological treatment for the causeof RF

3. Facilitate drainage of secretions via physiotherapy4. Check for any new complications:

a. Pneumothoraxb. Aspiration pneumonia

5. Persistent hypoxemia:a. Switch to a ventilator with oxygen blenderb. Determine if the EPAP should be increasedc. Increase the FiO2

6. Persistent or newly arising hypercapnia:a. Check for leakage in the interfaceb. Confirm that the circuit has leakage controlc. Correct any re-inhalation:

• Increase the EPAP• Change to Plateau valve• Use an interface with less dead space (if possible)

d. Prevent desynchronization:• Adjust the respiratory frequency and the I/E ratio• Adjust the inspiratory and expiratory triggers (if

possible)• Determine if the EPAP should be increased

e. Ensure adequate ventilation:• Check chest wall expansion• Increase the IPAP or delivered volume• Determine if the mode or ventilator should be

changed

FiO2: fraction of inspired oxygen; EPAP: expiratory positive airway pressure; IPAP: inspiratory positive airway pressure; I/E: inspiratory toexpiratory; RF: respiratory failure

additional respiratory physiotherapy treatment (e.g.manual or mechanical assisted cough) (see Chapter20).

In patients treated with a nasal interface, NIVmay fail if their nasal airway is blocked by secretions.Current NIV systems can not detect this problem,since blockage of nasal flow does not necessarilyinterfere with functioning or monitoring of theventilator flow. Hence, this risk must be anticipatedand treated clinically.

Another source of blockage—and consequently,of NIV failure—is edema caused by overly aggressivemanipulation of the nasal airway.

Assessing new complications that arise duringnon-invasive ventilation

If a patient stabilized with NIV quicklydeteriorates, and leakage is not the cause, thenhospital staff must evaluate the patient forcomplications such as pneumothorax, aspirationpneumonia, and hemodynamic instability. If anynewly arisen condition can not be resolved byadjusting the ventilation mode or parameters, thenintubation should be seriously considered.

Persistent hypoxemiaManaging hypoxemia tends to be slower in NIV

patients, whose oxygen needs often can not bereduced in the first 6 to 8 hours of treatment,especially in the case of Type II patients. Uncontrolled

hypoxemia will force the patient to maintain a highlevel of breathing work, ultimately leading torespiratory failure within hours. Hence, the possiblecauses of hypoxemia must be analyzed:

Insufficient oxygen in the airflowFor ventilators lacking an oxygen blender or

input valve (see Chapter 6, "Non-invasive ventilationdevices"), the airflow must be enriched with oxygeneither via the tubing or through a port on theinterface. The maximum FiO2 possible with thismethod is 0.5, although the level will vary in functionof leakage, parameter settings and the ventilatorused. There are recent models that include anoxygen valve whereby the FiO2 delivered can becalculated from manufacturer’s data tables(maximum FiO2 = 0.8).

Insufficient alveolar recruitmentIf the patient’s FiO2 persistently exceeds toxic

levels and can not be reduced, they may besuffering from insufficient alveolar recruitment—whether due to their pathology or because theprogrammed pressure is too low. The besttreatment strategy for this scenario is to increasethe expiratory positive airway pressure (EPAP)

93Failure analysis and predictive factors in non-invasive ventilation

Figure 1. Flow-time plot for pressure support mode: a) withoutleakage, and b) with leakage; The inspiratory time is lengthe-ned; Increasing the value of the expiratory trigger to highervalues of the peak flow reduces the inspiratory time and impro-ves synchronization.

Figure 2. Increasing the EPAP leads to a decrease in re-bre-athed CO2. The maximum decrease is observed at an EPAPof 8 cm H2O (with permission of the author, GT Ferguson,reference 3).

while carefully maintaining a sufficient pressuregradient for ventilating the patient. Hence, inventilators in which the IPAP is not above the EPAPor PEEP, it should also be increased.

Appearance or persistance of hypercapniaIf hypercapnia is observed at any time during

NIV, whether during periodic blood gas,transcutaneous or capnography measurements, thenthe patient should be evaluated for the followingcomplications: a. Obstruction of the airway by secretions.b. Re-inhalation. This can easily be identified by

capnography from a characteristic pattern onthe respiratory graph: the curve never reachesbaseline, indicating the presence of CO2 in theinhaled air. Re-inhalation can be resolved by thefollowing measures:• Increasing the EPAP: Using a single arm

respiratory circuit facilitates re-inhalation ofCO2, especially in patients with severetachypnea. Using an EPAP of at least 4 cm H2Owill force the exhaled air out of the exhalationvalve (see Fig. 2); the EPAP can be increasedup to 8 cm H2O. Alternatively, the section oftubing located at the exhalation valve can beremoved, and then replaced with a Plateauvalve (see Fig. 3).

• Some authors have recommended addingan external oxygen flow (4 to 6 L/min) tothe interface to minimize re-inhalation ofCO2.

• Another cause of re-inhalation is use of aninterface with large dead space; hence, as long

as the patient adapts well, hospital staff canswitch to an interface which has less deadspace or which features expiratory ports.

c. Inadequate ventilation. When taking any of themeasures listed below, hospital staff must alwaysensure that they properly adjust the IPAP toprovide the patient with the optimal tidalvolume.• Check chest wall movement.• Check leakage in the interface: Persistent leaks

not only cause desynchronization, they alsodiminish the tidal volume delivered to thepatient. Leaks can sometimes be caused byaccidental opening of an interface port; hence,all ports should be checked carefully.

• Check the respiratory circuit: The ICUs towhich the authors of this chapter pertainexperienced the following problems when firstincorporating NIV: on one occasion, hospitalstaff did not add an exhalation valve in thecircuit used with an NIV-specific ventilator; onanother, an inexperienced nurse deliberatelyclosed an exhalation valve in the ventilationset-up, thinking that it would be a source ofuncontrolled leakage.

• Consider any possible changes in ventilationmode or ventilator: This is especially importantin patients with restrictive thoracic diseases, forwhom pressure-metric ventilators can notguarantee an adequate tidal volume. Hospitalstaff should opt for pressure modes thatguarantee volume or for volume modes directly.

The term “non-invasive” should never beinterpreted as requiring less care; indeed, NIV alwaysdemands continuous reevaluation of the patient’s


Figure 3. The segment of tubing at the exhalation valve should be cut out, and then replaced with a Plateau valve.

PLATEAU VALVE

condition based on strict monitoring, early anticipationof any possible complications, appropriate adjustmentof the ventilation mode and parameters, and, whenrequired, careful transition of the patient to CMV.

As reiterated throughout this book, the successof NIV depends on the experience of hospital staffand on the quality of patient care. Hence, a checklist(see below and Table I) to avoid or correct NIVfailure, or for use in unequivocally switching to CMVif failed NIV can not be resolved, is an invaluablecomponent of any ward's treatment arsenal.

PREDICTIVE FACTORS OF FAILURE IN NON-INVASIVE VENTILATION

Predictive failure analysis is crucial to improvingNIV and defining its indications more clearly. Variousstudies on adult patients have identified predictivefactors of NIV failure within the general population as

well as according to pathology (e.g. pulmonary edema[APE], chronic obstructive pulmonary disease [COPD],and immunosupression). These factors comprise pH;pCO2; (PaO2/FiO2), both before and 2 hours afterinitiation of NIV; and a decrease in respiratoryfrequency. However, there have been very few studieson pediatric patients, and these have either been basedon retrospective data or have only analyzed a specificpathology (bronchiolitis) (Tabla II).

In the most detailed pediatric study published—performed in 2005 on 42 patients (including 6neonates and 11 post-operative cardiac surgerypatients)—the success rate for NIV was 57% andthe only predictive factor for NIV failure was an FiO2

requirement > 0.8 (with a negative predictive valueof 71%). However, this study was limited by theheterogeneity of the patient population and of theNIV indications (elective, post-extubation), as wellas by the fact that it was based on use of a


Table II. Predictive factors for failure of non-invasive ventilation published to date

Authors Year Study type # Patients Predictive factors

Essouri 2006 Retrospective 114 Multivariate: • SDRA • High PELOD scoreUnivariate: • PRISM • Decrease in pCO2 within 2 hours• Decrease in pCO2 within 2 hours

Bernet 2005 Prospective 42 • FiO2 > 80% after 1 hour

Joshi 2007 Retrospective 45 • Diseases of the pulmonary parenchyma • Age: < 6 years• FiO2 > 60% in first 24 h • pCO2 ≥ 55 mmHg in first 24 h

Larrar 2006 Prospective 53 • PRISMBronchiolitis (nasal CPAP) • Decrease in pCO2 within 2 hours

Campion 2006 Prospective (bronchiolitis) 69 • Apneas• High pre-NIV pCO2

• PRISM

Mayordomo 2008 Prospective 116 • Type I failure • PRISM• RR decrease at hours 1 and 6

ARDS: acute respiratory distress syndrome; PELOD: pediatric logistic organ dysfunction; PRISM: pediatric risk of mortality;PCO2: partial pressure of CO2 in the blood; FiO2: inspired fraction of oxygen; NIV: non-invasive ventilation; CPAP: conti-nuous positive airway pressure. RR: respiratory rate; ARF: acute respiratory failure.

conventional ventilator with NIV option (i.e. theseventilators provide worse synchronization andleakage compensation than NIV-specific ventilators).

A recently published prospective study of 87adult NIV cases reported an overall success rate of86%. This study established an algorithm basedon the physiopathological classification of eachpatient’s ARF, taking into account that one of theindependent factors associated with failure washaving Type I ARF.

Patients that respond to NIV tend to improveclinically within the first hour of treatment,experiencing a drop in tachypnea and retractions.Contrariwise, patients that respond poorly to NIV,whether due to poor adaptation or to worsening oftheir pathology, suffer from increased work ofbreathing and ultimately require intubation.

The factors which have shown the mostsensitivity for determining NIV success comprise adecrease in FiO2, a decrease in respiratory frequency,an increase in tidal volume, improved blood pH, and

improved PaO2/FiO2. Hence, for Type I ARF patients,especially the most severe patients without anycontra-indications (i.e. patients with mild acuterespiratory distress syndrome [ARDS]), it is criticalthat arterial blood gases be measured before NIV isstarted; if the patient’s PaO2/FiO2 does not improveabove > 175 within 1 hour of NIV, then thetreatment should be stopped.

To identify patients at the highest risk for NIVfailure (and therefore, those that will requireintubation), the authors of this chapter havedeveloped an algorithm based on clinical data fromthe study by Mayordomo et al. and on blood gasdata obtained from severe Type I ARF patients(specifically, ARDS patients).

The predictive factors for NIV failure identifiedin a study of 101 pediatric/adult bronchiolitis patients(success rate: 68%) comprised presence of apneas,initial hypercapnia, and high PRISM. As explainedabove, these types of predictive factors are invaluablefor ascertaining the potential for success (or failure)


Figure 4. Algorithm for analyzing failure of non-invasive ventilation in Types I and II acute respiratory failure. RF: respiratoryfrequency; PaO2: arterial oxygen pressure; FiO2: fraction of inspired oxygen.

Indications (Tables I and II)Contra-indications (Table IV)

Type I Type II

Selected patient

Classify patient (Table III)

≥ 12 monthsLow PRISM

> 6 months> 5 kg

Low PRISM

High risk Continue

No Yes

Measure SatO2/FiO2

ARDS suspected

Arterial blood gases and chest X-ray

PaO2/FiO2

< 150 > 150

Improvement andPaO2/FiO2

> 175

Intubation*

Intubation Continue

No Yes

ARDS no suspected

Lowering RF ≥ 10

High risk Continue

No Yes

1 hour

1 hour

FiO2 < 40%

1 to 24 hours

High risk Low risk

High risk Continue

No Yes

↓ FR

1 to 6 years

Intubation Continue

No Yes

*If intubation is delayed, close monitoring should be done

of NIV when choosing a course of treatment forcritical pediatric patients. The predictive factors forsuccess (or failure) of NIV in pediatric patients thathave been published to date are summarized in TableII and Fig. 4.

Patients with post-extubation acute respiratory failure

No study on ARF in patients treated with NIVpost-extubation has been published to date. Indeed,these patients are typically grouped together withNIV patients that have not previously received CMV.An exception to this trend is the work of Joshi, whoexcluded this patient group when analyzing theefficacy of NIV because of the lack of pre-BiPAP valuesfor them. The grouping together of children thathave previously received CMV with those that havenot can lead to errors in NIV analysis, as an invasiverespiratory treatment can have major effects onvariables such as heart rate and respiratory frequencyas well as on the gas pressures to be administeredin NIV.

In adult patients, the most recent works evaluatepost-extubation ARF alone. The meta-analysis ofAgarwal includes the four fundamentals. Inagreement with these studies, use of NIV for post-extubation ARF must be differentiated according tothe following two patient groups:1. Patients that develop ARF post-extubation: This

group has been analyzed in two randomizedstudies (Keenan and Esteban). Adult ICU patientsfrom this group that were treated with NIV didnot exhibit lower rates of reintubation or ofmortality compared to those given theconventional medical treatment. In fact, theauthors even observed a statistically insignificanttrend of higher mortality in the NIV patients,and in both studies, there was a higher rate ofreintubation in this group.

2. Patients at a high risk of developing post-extubation ARF that are given NIV as preventativetreatment (by being phased out of CMV andinto NIV): This group has also been analyzedin two randomized studies (Nava and Ferrer).The authors found that ICU patients given NIVin combination with the conventional medicaltreatment had lower rates of intubation and ofmortality than those given the conventionalmedical treatment alone.

A prospective study on 27 cases of post-extubation ARF, which did not include thedistinction made in the studies described above,was performed at the pediatric intensive care unit(PICU) of Hospital Universitario Central de Asturias.In the multivariate analysis, the only variableindependently associated with the success or failureof NIV was the FiO2, at 1 and 6 hours. Using thesedata, the authors sought cut-off points: at 1 hour,an FiO2 ≥ 50% had a sensitivity of 80%, a specificityof 82.53%, a positive predictive value (PPV) of72.23%, and a negative predictive value (NPV)of 87.50%; and at 6 hours, an FiO2 ≥ 50% had asensitivity of 80%, a specificity of 94.12%, a positivepredictive value (PPV) of 80%, and a negativepredictive value (NPV) of 94.12%.

The fact that no significant differences wereobserved for any of the more specific indicators ofa patient’s oxygenation state (e.g. PaO2/FiO2)between the successful NIV group and the failed NIVgroup is most likely due to the small sample size:arterial blood gas measurements were only availablefor 15 patients. The same explanation probably holdstrue for pCO2.

The aforementioned data reveal that we shouldclosely monitor the oxygenation level of patientstreated for post-extubation ARF. Indeed, an increasein oxygen needs is a red flag for possible NIV failure.

REFERENCES1. British thoracic Society standards of care committee. Non-inva-

sive ventilation in acute respiratory failure. Thorax 2002;57:192-211

2. M Pons. Contraindicaciones, complicaciones y problemas téc-nicos en ventilación no invasiva. En: Medina A, Pons M, Esqui-nas A (eds). Ventilación no invasiva en Pediatría. Madrid. Ed.Ergon 2004;79-86.

3. Ferguson GT, Gilmartin M. CO2 rebreathing during BiPAP venti-latory assistance. Am J Resp Crit Care Med 1995;151: 1126-35.

4. Antro C, Merico F, Urbino R, Gai V. Non-invasive ventilation asa first-line treatment for acute respiratory failure: "real life" expe-rience in the emergency department. Emerg Med J 2005;22:772-7.

5. Rodríguez Mulero L, Carrillo Alcaraz A, Melgarejo Moreno A, Rene-do Villarroya A, Párraga Ramírez M, Jara Pérez P, Millán MJ, Gon-zález Díaz G. Predictive factors related to success of non invasiveventilation and mortality in the treatment of acute cardiogenic pul-monary edema Med Clin (Barc) 2005;124:126-31.

6. Hilbert G, Gruson D, Vargas F, Valentino R, Gbikpi-Benissan G,Cardinaud JP, Guenard H. Non-invasive ventilation in immuno-suppressed patients Rev Mal Respir 2003;20:68-76.


7. Cadiergue V, Philit F, Langevin B, Durieu I, Bertocchi M, GuerinC, Robert D. Outcome of adult patients with cystic fibrosis admit-ted to intensive care with respiratory failure: the role of non-inva-sive ventilation Rev Mal Respir 2002;19:425-30.

8. Putinati S, Ballerin L, Piattella M, Panella GL, Potena A. Is it pos-sible to predict the success of non-invasive positive pressure ven-tilation in acute respiratory failure due to COPD? Respir Med2000;94:997-1001.

9. Moretti M, Cilione C, Tampieri A, Fracchia C, Marchioni A, NavaS. Incidence and causes of non-invasive mechanical ventilationfailure after initial success. Thorax 2000;55:819-25.

10. Bernet V, Hug MI, Frey B. Predictive factors for the success ofnoninvasive mask ventilation in infants and children with acuterespiratory failure. Pediatr Crit Care Med 2005;6:660-4.

11. Antonelli M, Conti G, Moro ML, Esquinas A, Gonzalez-Diaz G, Con-falonieri M et al. Predictors of failure of noninvasive positive pres-sure ventilation in patients with acute hypoxemic respiratory failure:a multi-center study. Intensive Care Med 2001;27: 1718-28.


13. Keenan SP, Powers C, McCormack DG, Block G. Noninvasivepositive-pressure ventilation for postextubation respiratorydistress: a randomized controlled trial. JAMA 2002;287:3238-44.

14. Esteban A, Frutos-Vivar F, Ferguson ND, Arabi Y, Apezteguia C,Gonzalez M, et al. Noninvasive positive-pressure ventilationfor respiratory failure after extubation. N Engl J Med 2004;350:2452-60.

15. Ferrer M, Esquinas A, Arancibia F, Bauer TT, Gonzalez G, Carri-llo A, et al. Noninvasive ventilation during persistent weaningfailure: a randomized controlled trial. Am J Respir Crit Care Med2003; 168:70-6.

16. Nava S, Gregoretti C, Fanfulla F, Squadrone E, Grassi M, CarlucciA, et al. Noninvasive ventilation to prevent respiratory failureafter extubation in high-risk patients. Crit Care Med 2005;33:2465-70.

17. Agarwal R, Aggarwal AN, Gupta D, Jindal SK. Role of noninva-sive positive-pressure ventilation in postextubation respiratoryfailure: a meta-analysis. Respir Care 2007; 52:1472-9.

18. Mayordomo Colunga J. Aplicación de la ventilación no invasi-va en una unidad de cuidados intensivos pediátricos. Doctoralthesis. Universidad de Oviedo. 2008.


Non-invasive ventilation (NIV) is highly effectivefor respiratory care of pediatric patients. Owing toconstant advances in NIV ventilators and interfaces,as well as the increasing familiarity of hospital staffwith this technique, clinical use of NIV is increasingboth in frequency, scope and efficacy. Patientsreceiving NIV may simultaneously requiresupplementary techniques either at the onset oftreatment or throughout the progression of theirpathology. These techniques may be complementaryto NIV and may even enable synergistic effects.Alternatively, NIV may be indicated simply to improvethe respiratory state of a given patient before theyare treated with a different, complimentary procedure.

Combinations of helium and oxygen, or heliox,have been used in ventilation therapy. The majorityof published cases have dealt with treatment of adultpatients for flare-ups of chronic obstructive pulmonarydisease (COPD). This chapter provides an overviewof the theoretical basis, existing clinical evidence, andpractical guidelines for use of heliox in pediatric NIV.

PROPERTIES OF HELIOXHelium was first introduced into the medical

arena in the 1930's by Barach, who showed that

when it was combined with oxygen, the resultingmixture (which he dubbed "heliox") improved airflowin patients with obstructive laryngeal, tracheal orlower airway lesions. Since then, studies on pediatricuse of heliox have demonstrated its efficacy fortreatment of various conditions, including upperairway obstruction caused by diverse pathologies,as well as asthma and acute bronchiolitis.

Helium, a noble gas, is inert, colorless andodorless and has very low density. If the nitrogen ininspired air (composed of 78% N2 and 22% O2) isreplaced with helium, which is seven times lessdense, a mixture (78% He and 22% O2) is obtainedwhich is three times less dense than normal air. Thetherapeutic utility of heliox lies in this very differencein density: when a patient breathes heliox insteadof instead of air, airway resistance to gas flow isreduced, leading to a reduction in respiratory work.Furthermore, heliox also improves gas exchange,above all in alveolar ventilation: in small airways,where elimination of CO2 is facilitated by diffusion,it diffuses four to five times faster in heliox than innormal air.

CLINICAL APPLICATIONS OF HELIOXAs an inert mixture, heliox lacks any intrinsic

therapeutic effects. However, it can act as atherapeutic buffer, maintaining the patient inimproved conditions, delaying onset of musclefatigue and respiratory failure, and obviating the use

Non-invasive ventilation with heliox*

F. Martinón-Torres

Chapter 14

*This work has been partially funded by Consellería de Sanidade-Xunta de Galicia (RHI07/2) and Instituto de Salud Carlos III-Ministerio de Sanidad y Consumo (Intensificación de la actividadinvestigadora F.Martinón-Torres, 2009).

of more aggressive treatments, until either othertherapies can be administered or the patient'scondition spontaneously resolves itself.

Heliox is most effective at the highest possibleconcentrations of helium (typically, 60 to 80%). Itsprincipal clinical applications are based on its physicalproperties: it tends to be used for predominantlyobstructive respiratory diseases in which alveolarventilation is compromised and airway resistance iselevated, forcing the patient to do greater work ofbreathing. Table I summarizes the main clinicalindications of heliox in pediatric medicine.

According to the literature heliox is typicallyadministered to patients with spontaneous breathingor to intubated patients using modified conventionalmechanical ventilation (CMV) devices.

For patients with spontaneous breathing, thepreferred method is via mask with reservoir andunidirectional valves at flows of 10 to 15 L/min. Ifrequired, supplementary oxygen can be deliveredthrough nasal prongs or cannulae, but at < 2 L/min,since a higher flow could excessively reduce thehelium concentration.

Warming and humidification of heliox, which isespecially important for very small children, can readilybe achieved by modifying standard equipment usedfor air-oxygen mixtures. Oxygen hoods or tents arenot ideally suited to heliox therapy since they lead togreater mixture with air, with the helium (less dense)going to the top, and the air (more dense) going tothe bottom, close to the patient’s airway. Simple facemasks or nasal cannulae may be amenable to helioxtherapy; however, they tend to cause dilution of theheliox flow with external air, thereby markedlyreducing the chances of successful treatment.

The delivery of invasively administered helioxdepends on the ventilator used. The heliox isintroduced via the pressurized-air inlet on theventilator. Hospital staff must first confirm that theventilator is heliox compatible, consider the variousconsequences of the physical properties of helioxon the ventilation parameters and performance (e.g.recorded volumes, FiO2 and flow measurements,and trigger sensitivity), and then take any requiredcorrective measures. The safest way to ventilate apatient with heliox is to use a pressure-controlledventilation (PCV) mode, following pressure goals.Heliox has also been delivered using high-frequencyjet ventilation (HFJV), high-frequency percussive

ventilation (HFPV) and high-frequency oscillatoryventilation (HFOV).

There is major clinical evidence for the efficacyof heliox as driving gas for drug nebulization in bothadult and pediatric patients. A 20 to 25% greaterflow is used for heliox than for air or oxygen, as thenebulization time for a given volume of solution willbe longer if driven with heliox than if driven with airor oxygen. Hospital personnel must ensure thatduring nebulization, the patient's entire minutevolume be filled using heliox. The best resultsreported for drug nebulization have been obtainedby using Y-shaped masks with reservoirs. Alternatively,an ultrasonic nebulizer can be used.

100 F. Martinón-Torres

Table I. Primary clinical applications of heliox in pedia-tric medicine. Heliox has shown the greatest efficacy intreatment of obstructive diseases of the upper airway;among all its indications, this one is most widely suppor-ted by clinical evidence. In contrast, use of heliox to treatconditions in which the lower airway is compromisedremains controversial

A. Obstruction of the upper airway:A1.Infectious etiology (e.g. laryngitis, epiglottitis and

tracheitis)A2. Inflammatory etiology:

1. Post-extubation subglottal edema 2. Post-radiotherapy edema3. Angioedema4. Edema caused by inhalatory lesions5. Spasmodic or recurring croup

A3. Mechanical etiology1. Foreign bodies2. Paralysis of the vocal chords3. Subglottal stenosis

A4. LaryngotracheomalaciaA5. Tumor etiology:

1. Expansion of the larynx and trachea2. Extrinsic compression of the larynx, trachea and

bronchiB. Obstruction of the lower airway:

B1. Acute asthma attacksB2. BronchiolitisB3. Bronchial hyper-reactivityB4. Neonatal respiratory distress syndrome B5. Bronchopulmonary dysplasia (BPD)

C. Driving gas for drug nebulizationD. Fiberoptic bronchoscopy (FOB) and/or instrumental

manipulation of the airwayE. Other conditions:

E1. HyperammoniemiaE2. Pneumothorax

RATIONALE FOR USING HELIOX IN NON-INVASIVE VENTILATION

Work on NIV of adult patients with helioxpublished by Jolliet et al. in the late 1990's sparkedinterest in use of this treatment for acute flare-upsin pulmonary lung disease patients. Since then, itsefficacy for these patients has been demonstrated.The first successful case of pediatric NIV using helioxinstead of air-oxygen mixtures was in the mid-1970's. It was used to wean infants that hadundergone post-operative cardiac surgery frommechanical ventilation. However, this practice didnot spread (nor did NIV at that time), and no otherreports on the use of heliox in pediatric NIVappeared until the authors of this chapter publishedwork on treatment of infants with acute refractorybronchiolitis.

Clinical data on adult mechanical ventilation,and experimental data from simulations, bothdemonstrate that at a given target pressure, helioxprovides greater minute volumes –and therefore,better alveolar ventilation– than do air-oxygenmixtures.

Increasing the expiratory flow reduces air trappingand generation of auto-PEEP, thereby improving lungcompliance, restoring the mechanical advantage toinspiration, and decreasing work of breathing. Thisin turn expands the tidal volume generated per cmH2O of pressure used. Moreover, NIV with helioximproves the patient’s elimination of CO2 and reducestheir sensations of dyspnea.

Heliox has complimentary effects when usedwith NIV; indeed, the combination of heliox andNIV can even be considered synergistic. As positivepressure, NIV can help reduce the load on respiratorymuscles, prevent or correct atelectasis, preventairway collapse, and promote and facilitatedistribution of heliox throughout blocked airways.Heliox can further reduce work of breathing andimprove CO2 elimination. It can also provide higherexpiratory flow than air-oxygen mixtures at the samepressure, which in turn can help improve passiveexpiratory lung mechanics, reducing the risk ofbarotrauma by gas trapping and thereby limitingany deleterious side effects from continuous positiveairway pressure (CPAP) in blocked airways.

Application of continuous positive pressure canlower the patient's FiO2 requirements, enabling themto be treated with a higher concentration of helium,

and consequently, increasing the likelihood oftreatment success.

Hypoxia secondary to development of atelectasishas been described in patients treated with heliox.Neonates and small infants are at especially high riskfor this complication. However, this highly seriousside effect can easily be prevented throughconcomitant use of CPAP.

Given that heliox can reduce the pressuregradient required to maintain a given flow, lesspressure can be used to obtain the target tidal andminute volumes, which in turn diminishes peakpressures and minimizes the risk of barotrauma orvolutrauma.

Lastly, heliox delivered by NIV can delay or evenobviate endotracheal intubation of the patient andsubsequent invasive mechanical ventilation, each ofwhich carries its respective complications. Althoughendotracheal intubation is the preferred method forairway isolation, it can produce local inflammation,ischemia of the mucosa, subglottal edema, and/orstenosis. Likewise, invasive positive pressureventilation can cause lung damage by altering lungmechanics, specifically, barotrauma, volutrauma,atelectrauma and biotrauma.

INDICATIONS OF HELIOX IN PEDIATRIC NON-INVASIVE VENTILATION

The authors of this chapter have compiled a listof indications of heliox used with NIV for pediatricpatients, based on the physical properties of helioxas well as on published data from simulations andfrom clinical studies on adult patients. These comprise:1. Children with spontaneous breathing that are

already receiving heliox therapy and that are notmeeting their oxygenation requirements and/orwhose FiO2 requirement is either > 0.40 orincreasing.

2. Children already receiving NIV:• For whom the ventilation level achieved is

inadequate and/or that require higherpressure, or whose pressure requirements areincreasing, and/or…

• That show insufficient improvement in dyspneaor in clinical respiratory score, and/or…

• Whose pathology requires protectiveventilatory strategy (i.e. the lowest possiblepressure for ensuring a target CO2 level).

101Non-invasive ventilation with heliox

PROTOCOL FOR USING HELIOX WITH NON-INVASIVE VENTILATIONa Equipment designed for heliox delivery: currently,

there are four commercially available ventilatorsdesigned for heliox delivery: Aptaer® HelioxDelivery System (GE Healthcare), Inspiration® (E-Vent Medical Ltd.), Avea® (Vyasis Healthcare)and Helontix Vent® (Linde Gas Therapeutics).The main characteristics of these machines aresummarized in Table I. They remain limited indistribution and use; indeed, none of thesedevices have appeared in any of the few workson pediatric patients published to date. TheAptaer® features inspiratory and expiratorytriggers, enabling use of pressure support withheliox. The Inspiration® is a conventionalventilator that includes an option for NIV deliveryof heliox. The Avea® offers all the standard modesused for invasive ventilation, plus the option forinvasive or non-invasive ventilation with heliox.The Helontix Vent® was specifically developedfor heliox therapy. It allows use of pressuresupport and only comes in a single model foradults and older children.

b. Equipment adapted for heliox delivery: helioxcan be safely delivered by modifying standardNIV ventilators according to literatureguidelines. However, as with use of anyequipment beyond its manufacturer'sindications, extreme caution must be appliedwhen using this approach. Hospital staff musthave technical expertise in both the ventilatorand in heliox therapy to minimize any potentialrisks in this method.1.When heliox is used with NIV, the average flow

and volume readings are not reliable, unlessan external pneumotacograph (which is notinfluenced by gas density) is used. Hence,ventilation should be programmed andcontrolled in function of the programmedpressures (which are not affected by heliox)and the resulting blood gas values (arterial O2

saturation and CO2 levels).2.NIV delivery of heliox instead of air-oxygen

mixtures generates greater inspiratory andexpiratory flows at the same pressure andprovides far superior CO2 diffusion. In termsof protecting the patient's lungs, this impliesthat the pressure gradient required to reach

a given target CO2 level will be lower withheliox than with an air-oxygen mixture.

3. If pure helium is used–which is generally notrecommended and is prohibited by legislationon therapeutic gases in most Europeancountries–, then the patient must becontinuously monitored by oximetry toprevent hypoxic conditions. Nonetheless, thesafest and most practical clinical option is touse helium-oxygen mixtures at predeterminedconcentrations (typically, 80:20 or 70:30), andthen add any required supplemental oxygenaccording to the patient's needs.

There are two ways of adapting standard NIVequipment for delivery of heliox: pre-dilutionconnection and post-dilution connection.

Pre-dilution connectionIn this approach heliox is directly connected to

the ventilator through a pressurized-gas inlet or ablender. It can be employed with the BiPAP Vision®

(Respironics), and the Infant Flow, Infant FlowAdvance and SiPAP® (EME/Viasys) systems. Thehelium or heliox source is connected to theventilator's air inlet (probably the best option) oroxygen inlet. In the former case, the FiO2 is regulatedusing the ventilator's FiO2 control; whereas in thelatter case, the FiO2 control is set to 1.0 to ensurethat only heliox is being delivered. If supplementaryoxygen is required, it can be added through a T-piece attached to the respiratory circuit. Alternatively,an oxygen blender can be attached to the ventilator'sair inlet or oxygen inlet, such that helium and oxygencan be added simultaneously before entering theventilator.

For cases in which the gas flow is directlyregulated (e.g. in the Infant Flow® system), thefollowing factors are used to convert flow meterreadings from air-oxygen values to heliox values: for80% He and 20% O2, multiply by 2.1; for 70% Heand 30% O2, multiply by 1.7; and for 60% He and40% O2, multiply by 1.4. These conversions enabledetermination of the real amount of heliox requiredto maintain the target pressure. If heliox is introducedthrough the air inlet, then the automatic limits ofthe alarm should be adjusted by first temporarilysetting the FiO2 control to the same percentage asthe oxygen in the heliox used, and then resetting itto 0.21. This aim of this technique is to avoid false


alarms caused by discrepancy between theconcentration set on the oxygen blender and theconcentration detected by the oximeter (whoseminimum value is the percentage of oxygen in theheliox introduced at the air inlet).

NIV delivery of heliox using the BiPAP Vision®

ventilator has been validated in an experimentalmodel. However, if this device is used, the initialleakage test of the expiratory valve should not beperformed; paradoxically, this will provide betterfunctioning of the ventilator, including greateraccuracy of the trigger. Using this method withthe FiO2 set at 1.0 enables safe and accurateadministration of helium at concentrations > than60%.

The NIV ventilators best suited to modificationfor heliox delivery (i.e. those for which heliox causesthe least interference in functioning, including inmeasurements) comprise, in decreasing order ofpreference: Servo 300®, Galileo®, Galileo Gold® andVeolar FT®. For volume controlled ventilation (VCV),there are specific correction factors for tidal volumeand FiO2 to be used with the most popular NIVventilators (see References).

Post-dilution connectionIn this method, the heliox is introduced directly

into the respiratory circuit tubing, after theventilator, ideally, as close to the patient (interface)end as possible. In this set-up, the concentration ofhelium that the patient receives is primarily dictatedby the flow of helium used. It requires external

oximetry to guarantee adequate FiO2 delivery.Furthermore, since the parameter values registeredby the ventilator are not reliable, externalpneumotacography is also needed. This set-up isan option in some ventilators (e.g. BiPAP Vision®),whereas it is the only option in others, whichgenerate pressure by drawing in ambient air (e.g.BiPAP S/T-D30, Knightstar 335 [Mallinckrodt],Quantum PSV [Respironics] and Sullivan VPAP II ST[ResMed]). Analogously to the case in standard NIVof trying increase the FiO2 by introducing oxygenin the circuit or interface, this approach is limitedby the fact that the operator can not determineexactly how much helium is being introduced intothe circuit, and therefore, delivered to the patient),which carries a risk of affecting ventilator function.Nonetheless, it has been validated experimentally.Using any of the aforementioned ventilators withheliox (80% He and 20% O2) flows of 18 L/minensures delivery to the patient of helium atconcentrations > 60%. Furthermore, at this flowrate, heliox only causes minimal interference (BiPAPS/T?D30 and the Quantum) with ventilator functionor does not interfere at all (Knightstar, Sullivan, andBiPAP Vision), nor does it alter the programmedinspiratory or expiratory pressures in any of thesemodels. Logically, with this heliox delivery approach,the final concentration of helium delivered to thepatient is influenced by the tidal volume obtained:higher inspiratory pressure and greater tidal volumeat a constant heliox flow correlate with a lowerconcentration of helium administered to the patient.


Table II. Main features of the commercially available ventilators designed for non-invasive delivery of heliox

Model Aptaer® Avea® Event® Helontix Vent®

Distributor Datex Ohmeda Vyasis Healthcare Event Inspiration LTD Linde Gas Therapeutics

Adjustable FiO2 No Yes Yes Yes

Adjustable PEEP No Yes Yes Yes

Pressure support Yes Yes Yes Yes

Pediatric model No Yes Yes No*

Standard ventilation modes No Yes Yes No

Heliox consumption Low Medium Medium Low

Difficulty for portable use Medium High High Low

Price Low High High Low**

*Pediatric model currently in development. **Device provided free with gas purchase.

SIDE EFFECTS AND DRAWBACKS

Side effectsAlthough heliox is inert (non-toxic), when used

with NIV it may lead to certain side effects, namely:1. Hypoxemia. The primary side effect is insufficient

oxygenation, since this treatment implies useof the lowest FiO2 possible to maximize thepositive effects of helium. Hence, stronglyhypoxemic patients, who are characterized byhigh oxygen requirements, can not be treatedwith heliox (at least not at any therapeuticallysignificant concentration of helium). Hypoxemiacan also be caused by inadvertent use of ahypoxic mixture (FiO2 < 21%) in cases in whichthe helium and oxygen are mixed manually.This can be avoided by using tanks of pre-mixedheliox at known concentrations, or byemploying continuous oximetric monitoring(with an alarm) of the mixture delivered to thepatient.

2. Hypothermia. One of the most therapeuticallyimportant physical properties of heliox is itshigh thermal conductivity (six to seven timesthat of air); hence, prolonged heliox treatmentat temperatures under 36 °C carries a risk ofhypothermia. Neonates and small infants areat the highest risk for this side effect, whichcan be prevented by ensuring proper warmingand humidification of the heliox and bycarefully monitoring the patient's bodytemperature.

Drawbacks1. Cost. There have not been any studies on the

cost effectiveness of NIV heliox treatment.Compared to other gases, heliox is much moreexpensive: two to three times as much as air,and eight to ten times as much as oxygen. Usingnon-NIV-specific devices for delivery of helioxcan lead to increased costs due to internal gasconsumption, leakage compensation and orexcessive internal leaks. However, NIV-specificventilators also have their respective drawbacks,which vary according to model.

2. Technical difficulties. The physical properties ofheliox– above all, its low density– can interferewith fundamental ventilator operation, especiallyrecording of FiO2 volumes, as well as triggerfunctioning, flow measurements, and theautomatic leakage compensation featured onsome models. These technical difficulties vary withthe ventilator used, and are primarily dictated bythe type of valves and pneumotacograph used.In any case, pressure transducers are generallynot affected by heliox; hence, treatment shouldbe adjusted according to the programmedpressures.

CONCLUSIONSUsing heliox with NIV is a promising therapeutic

option for children with various respiratorypathologies that do not respond to conventionaltreatment. Heliox is complementary to NIV, and


Figure 1. BiPAP Vision® (Respironics®) non-invasive ventilatoradapted for heliox delivery: the heliox is introduced throughthe device's pressurized-oxygen inlet.

Figure 2. InfantFlow Advance® (EME®) non-invasive ventila-tor adapted for heliox delivery: the heliox is introducedthrough the device's pressurized-air inlet.

together, they may even have synergistic effects.Albeit there is only a small body of literature on helioxin pediatric NIV –focusing on treatment of severeacute bronchiolitis patients that can not be treatedwith standard therapies–, the results have beenpositive. The ideal ventilators for heliox delivery aredevices which are specifically designed for this task;however, their use remains limited. Alternatively,other ventilators can be modified to administer helioxin NIV, although these must be used with extraprecautions. Theoretically, NIV with heliox embodiesa new strategy for lung treatment and protection.However, this technique must be further evaluatedin clinical studies, and its indications and treatmentguidelines must be more clearly established.

REFERENCES1. Austan F, Polise M. Management of respiratory failure with nonin-

vasive positive pressure ventilation and heliox adjunct. HeartLung 2002;31:214-8.

2. Castello Munoz A, Carreira Sande N, Bouzon Alejandro M, PerezValle S, Rodriguez Nunez A, Martinon Sanchez JM, et al. Utili-dad del heliox en el manejo de una obstruction severa de lavía aérea por hemangiomatosis cervical múltiple. An Pediatr(Barc) 2007;67:61-4.

3. Chatmongkolchart S, Kacmarek RM, Hess DR. Heliox deliverywith noninvasive positive pressure ventilation: a laboratory study.Respir Care 2001;46:248-54.

4. Colebourn CL, Barber V, Young JD. Use of helium-oxygen mix-ture in adult patients presenting with exacerbations of asthmaand chronic obstructive pulmonary disease: a systematic review.Anaesthesia 2007;62:34-42.

5. Hess D, Chatmongkolchart S. Techniques to avoid intubation:noninvasive positive pressure ventilation and heliox therapy. IntAnesthesiol Clin 2000;38:161-87.

6. Hess DR. Heliox and noninvasive positive-pressure ventilation: arole for heliox in exacerbations of chronic obstructive pulmo-nary disease? Respir Care 2006;51:640-50.

7. Hilbert G. Noninvasive ventilation with helium-oxygen ratherthan air-oxygen in acute exacerbations of chronic obstructivedisease? Crit Care Med 2003;31:990-1.

8. Jolliet P, Tassaux D, Roeseler J, Burdet L, Broccard A, D'Hoore W,et al. Helium-oxygen versus air-oxygen noninvasive pressure sup-

port in decompensated chronic obstructive disease: A pros-pective, multicenter study. Crit Care Med 2003;31:878-84.

9. Jolliet P, Tassaux D, Thouret JM, Chevrolet JC. Beneficial effectsof helium: oxygen versus air:oxygen noninvasive pressure sup-port in patients with decompensated chronic obstructive pul-monary disease. Crit Care Med 1999;27:2422-9.

10. Martinon-Torres F, Crespo Suarez PA, Silvia Barbara C, CastelloMunoz A, Rodriguez Nunez A, Martinon Sanchez JM. Ventila-ción no invasiva con heliox en lactante con síndrome de dis-trés respiratorio agudo. An Pediatr (Barc) 2005;62:64-7.

11. Martinón-Torres F, Martinón Sánchez JM. Helio: Utilidad e indi-caciones. En: Casado Flores. Urgencias y tratamiento del niñograve: síntomas guía, técnicas y procedimientos (2ª ed). Madrid,Ergon, 2007;248-254.

12. Martinón-Torres F, Medina Villanueva A, Pons Odena M. Heliox.En: Esquinas Rodríguez A. Consensos clínicos en ventilacónmecánica no invasiva 2008, Editorial Aula Médica, Madrid; 596-603.

13. Martinon-Torres F, Rodriguez Nunez A, Martinon Sanchez JM.More about heliox and bronchiolitis. Pediatrics 2002;110:198-9.

14. Martinon-Torres F, Rodriguez Nunez A, Martinon Sanchez JM.More about heliox and bronchiolitis. Pediatrics 2002;110:198-9.

15. Martinon-Torres F, Rodriguez Nunez A, Martinon Sanchez JM.Heliox therapy. Pediatrics 2002;110:847-8.

16. Martinón-Torres F, Rodríguez Núñez A, Martinón Sánchez JM.Heliox: Perspectivas de aplicación en pediatría. An Esp Pediatr1999;128:42-6.

17. Martinon-Torres F, Rodriguez-Nunez A, Martinon-Sanchez JM.Heliox questions. Pediatrics 2003;111:441-3.

18. Martinon-Torres F, Rodriguez-Nunez A, Martinon-Sanchez JM.Nasal continuous positive airway pressure with heliox in infantswith acute bronchiolitis. Respir Med 2006;100:1458-62.

19. Martinon-Torres F, Rodriguez-Nunez A, Martinon-Sanchez JM.Nasal continuous positive airway pressure with heliox versus air-oxygen: a crossover study. Pediatrics 2008;121:1190-1195.

20. Martinón-Torres F. Ventilación con helio. En: Ruza Tarrío F (ed).Tratado de Cuidados Intensivos Pediatricos. Madrid, Norma Capi-tel, 2003;677-681.

21. Piva JP, Barreto SSM, Zelmanovitz F, Amantea S, Cox P. Helioxversus oxygen for nebulized aerosol therapy in children withlower airway obstruction. Pediatr Crit Care Med 2002;3:6-10.

22. Rodriguez Nunez A, Martinon Sanchez JM, Martinon Torres F.Gases medicinales: oxígeno y heliox. An Pediatr (Barc) 2003;59:74-81.


HUMIDIFICATION

IntroductionThere is a general consensus on the need to

condition inspired gases in order to maintain adequateciliary function, conserve the rheological characteristicsof respiratory mucosa, facilitate gas exchange andpromote clinical progress. Humidification of inspiredgases is essential for patients treated with medicinalgases, oxygen therapy, invasive mechanical ventilation(via oral-tracheal tube or tracheotomy), and non-invasive ventilation (NIV).

Mouth breathing in acute patients leads to thefollowing complications:• Loss of heat.• Increased resistance in the nasal cavity.• Drying out of mucosa.• Thickening of secretions.

Indications of gas humidification in pediatricrespiratory medicine

Any gas administered to a pediatric patientshould be humidified. Insufficient humidification, orunderhumidification, is especially detrimental topatients presenting with abundant bronchialsecretions, chronic respiratory pathology, highinspiratory fraction, deficient pre-treatment hydrationstate, or hypoxemic acute respiratory failure (ARF);in patients receiving prolonged NIV treatment; andin patients whose ventilation is marked by highventilator flows and/or major leakage.

The primary indications of gas humidification inpediatric respiratory medicine are summarized inTable I.

Methodology

HumidifiersChoosing a humidifier, rating its efficacy and

evaluating its clinical and morphological effects onrespiratory mucosa depend on various factors. Ideally,a humidifier should:1. Have low resistance in the inspiratory and

expiratory phases of the respiratory cycle. 2. Not interfere with monitoring for hypercapnia

(in order to avoid generating re-inhalation).3. Not interfere with the ventilator's triggers.4. Not cause heat loss in the patient's respiratory

circuitThe humidifiers used to condition respiratory

gases can be passive or active. Some activehumidifiers contain a heated wire circuit.

Passive humidifiers (artificial noses), also called heatand moisture exchangers (HMEs)

Passive humidifiers are currently the most widelyused humidification systems for conventionalmechanical ventilation (CMV). They contain bothhydrophilic and hydrophobic components, deliveringheated and humidified air at low inspiratory andexpiratory resistance and with low internal volume.However, these systems are ineffective in NIV for thefollowing reasons:1. Their ability to maintain adequate heat and

moisture exchange is limited at high oxygenflows for prolonged periods of time.

2. They are not recommended for use with nasalmasks due to the risk of losing gas exhaledthrough the mouth.

Humidification and bronchodilatortherapy in non-invasive ventilation

A. Esquinas and M. Pons

Chapter 15

3. When used with a face mask, the effort requiredof the patient to set off the ventilator's triggermay cause patient-ventilator desynchronization.

4. They can increase resistance in inspiratory andexpiratory flows.

Active humidifiers (with or without a heated wire cir-cuit)

Active humidifiers with servo-control providerelative humidification from 33% (those without aheated wire circuit) to 100% (those with a heatedwire circuit). There is a general consensus that thesesystems prevent drying out of mucosal membranesin NIV, although they are less effective in patientstreated with nasal masks. These systems have severaldrawbacks:1. They interfere with the ventilator's triggers and

delivered volumes. Infants younger than 6months and very weak patients (whether due tothe stage of their respiratory failure or to aneuromuscular disease [NMD]) are especiallyprone to this complication.

2. In intubated patients, it has been demonstratedthat active humidifiers which generate aerosolscarry an increased risk of nosocomial pneumonia.

3. They use more water than passive humidifiersand are more difficult to maintain.

4. They are more expensive than passive humidifiers.There is a wide range of commercially available

humidifiers and of tubing, the majority of which hasan internal diameter of 22 mm (see Fig. 1). Tubing

diameter is important in NIV-specific ventilators:some ventilators have problems adapting to thestrong resistance generated by use of high flows.

The authors of this chapter are only aware ofone type of tubing available with an internaldiameter of 25 mm; this model contains an internalheated wire circuit to reduce condensation in therespiratory circuit (Fig. 2).

Target temperature and humidityInspiratory gases are typically conditioned to a

temperature of 37 °C and a relative humidity of100%. However, since the delivered gas in NIV doesnot directly reach the trachea but instead must firstpass through the nasal cavity or cross the face, 37°C may prove excessive, causing the patientdiscomfort and leading them to reject treatment.Hence, some humidifiers now feature the option toheat gas to 34 °C (Fig. 3); this temperature can beused during the adaptation phase of NIV if thepatient does not tolerate 37 °C well.

The clinical complications related to gashumidification are summarized in Table II.

108 A. Esquinas, M. Pons

Table I. Indications of humidification in pediatric medici-ne: clinical situations in which humidification is critical

• Patients with deficient drainage of secretions– NMDs

• Patients with thick secretions– Cystic fibrosis– Dehydration

• Patients with hypoxemic ARF– High FiO2 requirements– High ventilators flows – High risk of leakage

• Patients with limited cooperation– Infants– Prolonged NIV (> 72 hours)

ARF: acute respiratory failure; FiO2 fraction of inspired oxygen;NIV: non-invasive ventilation; NMD: neuromuscular disease.

Figure 1. Active humidification without a heated wire circuit,using 22 mm tubing (Intersurgical®).

Figure 2. Active humidification with a heated wire circuit,using 25 mm tubing (Fisher-Paykel®).

Underhumidification can have minor side effects(e.g. drying out of mucous membranes, andbleeding) or cause major complications (e.g.uncontrollable dyspnea, uncontrollable secretions,and patient intolerance to the interface), and mayeven result in the patient needing to be urgentlyintubated. These effects are listed in Table III.

BRONCHODILATOR THERAPYA common delivery method for bronchodilator

is aerosol (inhalatory) therapy, in which the patientinhales a suspension of the drug, which is thendelivered directly to the target organ. Compared toother routes, aerosol therapy is advantageous in thatit uses much smaller doses (which translates to lowercosts), is much faster and leads to lowerconcentrations of drug in the blood (i.e. it has abetter therapeutic index).

Fundamentals of aerosol therapyAn aerosol is a suspension of particles in an air

current. The efficacy of drugs administered by aerosolis primarily influenced by the following factors:

1. Particle size. The size of drug particles (measuredin microns) in an aerosol used for respiratorydiseases dictates where in the respiratory pathwaythey will be deposited (see Table IV). The idealparticle size for respiratory drugs is 1 to 8 microns.At this size, particles can reach the walls of thedistal airways via sedimentation and diffusion. Incontrast, very small drug particles (< 1 micron)have poor transportability and a high probabilityof being exhaled, whereas very large particles (>8 microns) tend to agglomerate and quicklydeposit in the upper airway (tongue andoropharynx), from where they can be swallowedand absorbed, ultimately causing systemic sideeffects. It should be noted that nebulizers—thedevices which generate drug aerosols—do notafford a uniform particle size; rather, they producea wide range of particle sizes.

2. Aerosol flow rate. The patient's inspiratory flowinfluences the quantity and type of particlesdeposited as well as the mechanism of deposition.The preferred aerosol flow rate is 30 to 60 L/min.High inspiratory flows (> 100 L/min) favordeposition by impact and provide a high a rateof penetration, whereas low inspiratory flows (<30 L/min) favor sedimentation but carry a riskthat the patient will only inhale a small quantityof the drug. Deposition of particles in children'slower airways is hindered by the combination ofhigh flow speeds and the decreasing (from topto bottom) diameter of their airways.

3. Hydrophilicity: The affinity of particles for waterdetermines the extent to which they can changesize.

109Humidification and bronchodilator therapy in non-invasive ventilation

Table II. Factors that influence complications with humi-dification in non-invasive ventilation

1. Type of acute respiratory failure (hypoxemic or hyper-capnic)

2. Type of interface (e.g. nasal, oral-nasal, face or hel-met)

3. Duration of NIV

4. Severity of the pulmonary pathology that is affec-ting the airways

5. Related pathologies (e.g. cardiopathy or neuromus-cular disease [NMD])

6. NIV parameter values: FiO2, pressures, flows and lea-kage

7. Patient's hydration level

8. Intercurrent factors (e.g. fever, and loss of liquids)

9. Patient's nutritional state: namely, protein metabolism

10. Metabolism of the respiratory epithelium during mucusproduction

11. Vasoactive drugs

12. Diminished cough reflex and decreased consciousness

FiO2 fraction of inspired oxygen; NIV: non-invasive ventilation.

Figure 3. Button for selecting the temperature.

4. Airway geometry. The numerous bifurcations andnarrowness of the respiratory airways amplifyturbulence in the gas flow and increase theprobability that the drug particles will impact inthe most proximal areas.

5. Type of nebulizer or inhaler used,

Nebulizers and inhalersThe generation and deposition of aerosols during

mechanical ventilation—with or without isolationof the airway—strongly influences the efficacy of thedrug, especially in the case of bronchodilators, andcan ultimately dictate the success of NIV. There aretwo ways to generate aerosols: via nebulizers or viainhalers. The choice of device depends on thepatient's clinical profile.

The most widely used nebulizers and inhalersamenable to NIV are described below, and theirrespective advantages and disadvantages aresummarized in Table V.

NebulizersThere are two types of nebulizers: jet and

ultrasonic.1. Jet nebulizers. These devices operate by the

Bernoulli and Venturi effects. In the former, whenpressurized gas (air, oxygen or heliox) is pushedthrough the open end of a capillary tube whoseopposite end is submerged in a liquid at ambient

pressure, the resulting sub-atmospheric pressurecauses the liquid to be aspirated upwards to thegas-end, where, upon contact with the gas, itfragments into an array of different sized particlesby diffusion (< 1 micron), gravity (1 to 6 microns)and inertia (> 10 microns).

2. Ultrasonic nebulizers. These devices exploit thepiezoelectric ability of quartz to vibrate uponexposure to an electric current: the vibrationsare transmitted through a plastic membrane thatcontains the liquid to be nebulized. The vibratingliquid is converted into an aerosol which istemporarily stored in a chamber, from whereit is delivered to the patient by a stream ofpressurized air generated by a ventilator.Experience with ultrasonic nebulization in NIVis very limited; however, it appears to be thenebulization method which least interferes withNIV (see Fig. 4).

InhalersInhalers produce fixed doses of very small sized

particles by subjecting a solution to a pressurizedflow; the dose size varies with the device. They aremore widely used than nebulizers for stable patients(i.e. those not receiving mechanical ventilation) ableto achieve a minimum level of synchronization andcooperation. This indication is based on the work ofFuller; however, other published studies havereported that no differences in clinical outcome wereobserved between treatment with inhalers andtreatment with nebulizers.

Metered dose inhalers (MDIs) (Fig. 5) use apressurized canister to generate different sizedparticles (from 2 to 4 microns). MDIs have threemain components:1. A variable capacity canister which contains the

active drug and propellants, either as a liquid


Table III. Signs of inadequate heating or humidification ofinspired gases

Signs of deficient heating or humidification1. Nasal or pharyngeal irritation or discomfort2. Nasal dryness3. Nosebleed4. Nasal blockage5. Rhinorrhea6. Coughing7. Sneezing8. Bleeding in the airway9. Viscous respiratory secretions10. Atelectasis11. Decreased lung compliance12. Patient fails to follow treatment

Signs of excessive heating or humidification1. Irritation or pain of the airway due to burns2. Diffuse micro-atelectasis3. Over-hydration or water poisoning4. Excessive condensation in the respiratory circuit tubing.

Table IV. Distribution by size and by location of bron-chodilator particles delivered by aerosol therapy

Size Location Percentage(microns)

>10 Mouth, nose and trachea 1005 Between larynx and bronchioles 90-953 Between bronchioles and acini 851 Acini 600,5-0,6 Alveoli 35

solution or gas suspension, at a pressure of 3 to4 atmospheres.

2. A metering valve which releases a controlled andreproducible dose of the micronized drug eachtime the patient presses down on the canister.

3. A plastic case which covers the canister; Pressingdown on the canister opens the valve, whichreleases the aerosol through a small opening.

Nebulization and inhalation in non-invasiveventilation

The preferred aerosol therapy option formechanical ventilation patients depends on thepatient's clinical profile (i.e. age, respiratory frequency,level of respiratory difficulty, level of consciousness,and blood gas levels) and on ventilation treatmentfactors (e.g. adaptation, synchronization, stage ofNIV). Just as in patients with spontaneous breathing,in NIV patients there may be a slight preference forthe use of inhalers. However, nebulizers are probablythe best choice for pediatric patients—especially theyoungest ones—with acute pathologies and those inthe initial phases of NIV.

Indications of aerosol therapy in pediatricmedicine

According to the few existing publications onaerosol therapy in pediatric patients, there are twomain pathologies for which this treatment isindicated:• Status asthmaticus. It seems obvious that once

these patients have been started on NIV, theyshould be maintained on continuousbronchodilator therapy to avoid respiratoryfailure. In less severe patients, inhalers providebetter peak expiratory flow than do nebulizers.

• Cystic fibrosis. Nebulization used with pressuresupport NIV of 10 cm H20 provides 25% betterdrug deposition than does nebulization alone.

MethodologyThe following guidelines should be followed

when administering aerosol therapy to NIV patients:1. Positioning the inhaler or nebulizer. Inhalers with

spacers should be placed at the far (ventilator)end of the respiratory circuit for two reasons:firstly, this reduces impact of the drug on themask walls, and secondly, it enables coordinationof the start of ventilation with that of drug

delivery. In contrast, nebulizers should beintroduced into the respiratory circuit close tothe mask (typically a face or oral-nasal mask, toprevent loss of the drug).

2. NIV parameters. The inspiratory positive airwaypressure, pressure support (where required), andexpiratory positive airway pressure should beadjusted to achieve optimal inspiratory flowsand tidal volume without compromising patientcomfort and tolerance. The lowest level possibleof FiO2 is recommended.

3. Cardiorespiratory monitoring. The patient's vitalsigns should be checked before and duringaerosol therapy to evaluate the efficacy of thetherapy and indentify any side effects.

4. Drug dose, nebulization time, and nebulizer gasparticle size and flow rate. The drug dose shouldbe calculated according to the patient's weight(see Table VI). The nebulization time must bemonitored (it should be at least 20 to 30 min)and the particle size in the nebulizer gas flowshould be controlled (the target particle sizedictates the position of the nebulizer in the circuit).For jet nebulizers, maximizing particle depositionin the upper airway generally requires an oxygenflow of 4 L/min, whereas focusing delivery on thelower airway generally requires an oxygen flowof 6 to 8 L/min (12 L/min for heliox).

5. Supplementary devices. If the respiratory circuitis equipped with a Plateau (Respironics®) valve,then the nebulizer should be positioned inbetween the valve and the interface to avoid lossof elasticity in the valve's diaphragm. For thedouble-circuits of conventional ventilators, thenebulizer or inhaler should be positioned in theinspiratory arm. Other devices introduced intothe respiratory circuit (e.g. aspirators) cancompromise deposition of aerosols in the circuit.

6. Humidification. As previously mentioned, nasalmasks should not be used with pasive humidifiers.

7. Evaluation of bronchial secretions. The patient'sbronchial secretions should be evaluated tooptimize the aerosol therapy. All bronchialsecretions should be eliminated before beginningtherapy.

Precautions in aerosol therapyThe main complications in aerosol therapy derive

from incorrect administration of the drug, including


overdoses. There are two precautionary factors thatmust be remembered when using nebulizers:1. Children treated with nebulizers can suffer from

over-hydration and excessive moisture in theairways. This leads to increased density of thedelivered air, causing greater resistance to thepassage of the airflow, and consequently,increased dyspnea in the patient. In extremecases, patients can face a risk of drowning fromliquid which has condensed in the airways.

2. If distilled water is used as the solvent for thedrug to be nebulized, it can dilute the surfactantof the alveoli, thereby increasing the patient'sdyspnea and worsening their respiratorycondition. Moreover, since it is hypotonic, watercan irritate the mucous membranes of the

airways and may ultimately cause edemas orbronochiospasms. Therefore, the recommendedsolvent for drug nebulization is an isotonic 0.9%saline solution or, for bronchiolitis patients, a3.0% hypertonic solution.

Bronchodilator drugsThere is a broad array of bronchodilator drugs

that can be used in aerosol therapy. However, thegeneral opinion at the 2000 Consensus Conferenceof the American Association for Respiratory Care(AARC) was that literature on aerosol therapy in NIVis rather limited and that most clinical experience in


Table V. Respective advantages and disadvantages of nebulizers and inhalers

Nebulizers Inhalers

Advantages Require less coordination ComfortableAllow for high doses (including continuous dosing) CheapDo not release chlorofluorocarbons (CFCs)

Disadvantages Expensive Deposit drugs in the larynxWaste drugs Difficult to administer in high dosesNot available for all drugs Not available for all drugsRequire a source of pressurized gasSlower actingGreater risk of infections

Figure 5. Metered dose inhaler (MDI) canister for use with amechanical ventilator.

Table VI. Dosing of the most common aerosol therapydrugs

Drug Dose Interval

Salbutamol • MDI: 2 (4 to 8) puffs • Nebulization: 0.2 mg/kg/dose 4 to 6 hours

Minimum: 1 mg; Maximum: 5 mg • Continuous nebulization:

0.3 to 0.5 mg/kg/hour ContinuousMaximum: 30 mg/hour

Terbutaline • MDI: 2 to 3 puff 4 to 6 hours• Nebulization: 1mL/kg/day 8 hours

Ipratropium • MDI: 4-8 puffbromide • Nebulization: < 4 years:

250 μg/dose; 4 to 6 hours> 4 years: 500 μg/dose

Beclometasone • MDI: 100-400 mg/day 12 hours

Fluticasone • MDI: 100-2.000 μg/day 12 hours

Budesonide • MDI: 100-400 μg/day 12 hours

Figure 4. Ultrasonic nebulizer.

this area has been with delivery of bronchodilators(see Table VI).

REFERENCES

HUMIDIFICATION

1. Schulze A. Respiratory gas conditioning in infants with an artifi-cial airway. Semin Neonatal 2002;7:369-77.

2. Kacmarek R. Essential gas delivery features of mechanical venti-lators. Respiratory Care 1992;37:1045-55.

3. Branson R. Humidification for patients with artificial airways. Res-piratory Care 1999;44:630-42.

4. Holland AE, Denehy L, Buchan CA, Wilson JW. Efficacy of a hea-ted passover humidifier during noninvasive ventilation: a benchstudy. Respir Care 2007;52:38-44.

5. Hill NS. Complications of noninvasive positive pressure ventila-tion. Respiratory Care 1997;42:432-42.

6. Tuggey JM, Delmastro M, Elliott MW. The effect of mouth leakand humidification during nasal non-invasive ventilation. RespirMed 2007;101:1874-9.

7. Richards GN, Cistulli PA, Ungar G, Berthon-Jones M, Sullivan CE.Mouth leak with nasal continuous positive airway pressure incre-ases nasal airway resistance. Am J Respi Crit Care Med1996;154:182-6.

8. Shelly MP, Lloyd GM, Park GR. A review of the mechanism andmethods of humidification of inspired gases. Intensive Care Medi-cine 1998;14:1-9.

9. Branson RD, Davis K, Jr, Brown, R, Rashkin M. Comparisons ofthree humidification technique during mechanical ventilation:patient selection, cost, and infection considerations. RespiratoryCare 1996;41:809-16.

10. Wood KE, Flaten AL, Backes WJ. Inspissated secretions: a life thre-atening complication of prolonged noninvasive ventilation. Res-piratory Care 2000;45:491-3.

11. Fonkalsrud EW, Sánchez M. A comparative study of the effectsof dry vs humidificated ventilation on canine lungs. Surgery1975;78:373-9.

12. Constantinidis J, Knobber D, Steinhart H. Morphological andfunctional alterations in the nasal mucosa following nCPAP the-rapy. Fine–structural investigations of the effect of nCPAP –maskapplication on the nasal mucosa. Otolaryngol 2000;120:432-7.

13. Rashad K, Wilson K. Effect of humidification of anaesthetic gasesand static compliance. Anesth Analgesia Curr Res 1969;40:127-9.

14. Fontanari P, Burnet H, Zattara– Hartmann M, Jammes Y. Chan-ges in airway resistance induced by nasal inhalation of cold dry,dry or moist air in normal individuals. Journal Applied Physio-logy 1996;81:1739-43.

15. Strohl K, Arnold J, Decker MJ, Hoekje PL, McFadden ER. Nasalflow resistive responses to challenge with cold dry air. JournalApplied Physiology 1992;72:1243-6.

16. Clini E. Patient ventilator interfaces: practical aspects in the chro-nic situation. Monaldi Arch Chest Disease 1997;52:76-9.

17. Teresa Martins de Araújo M, Seergio Barron Vieria, Elisardo CorralVazquez, Bernard Fleury. Heated humidification or face mask to

prevent upper airway dryness during continuous positive airaypressure therapy. Chest 2000;117:142-7.

18. Togias AG, Nacleiro RM, Pround D, Fish JE, Adkinson NF Jr, Kagey-Sobotka A et al. Nasal challenge with cold, dry air results in rele-ase of inflammatory mediators. Journal Clinical Invest1985;76:1375-81.

19. Dreyfuss D. Mechanical ventilation with heated humidifiers orheat and moisture exchanges: effects on patient colonizationand incidence of nosocomial pneumonia. American Journal Cri-tical Care Medicine 1995;1515:986-92.

20. Thomachot L. Do the components of heat and moisture exchan-ges filters affect their humidifying efficacy and the incidence ofnosocomial pneumonia. Critical Care Medicine 1999;27:923-8.

21. Meduri GU. Noninvasive mechanical positive-pressure ventila-tion via face mask. First line intervention in patients with acutehypercapnic and hypoxemic respiratory failure. Chest 1996;109:179-93.

22. Bott J, Carroll MP, Coway JH, Keilty SE, Ward EM, Brown AM etal. Randomised controlled trial of nasal ventilation in acute ven-tilatory failure due to chronic obstructive airways diseases. Lan-cet 1993;341:1523-30.

23. Antonelli M, Conti G, Moro ML, Esquinas A, Gonzalez- Diaz G,Confalonieri M, et al. Predictors of failure of noninvasive positi-ve pressure ventilation in patients with acute hypoxemic respi-ratory failure: a multi-center study. Intensive Care Med 2001;27:1718-28.

BRONCHODILATOR THERAPY

1. Coleman DM, Kelly HW, McWilliams BC. Determinants of aero-solised albuterol delivery to mechanically ventilated infants. Chest1996;109:607-13.

2. Faroux B, Itti E, Pigeot J, Isabey D, Meignan M, Ferry G, et al.Optimization of aerosol deposition by pressure support in chil-dren with cystic fibrosis. Am J Respir Crit Care Med 2000;162:2265-71.

3. Dhand R, Tobin M. Inhaled bronchodilatator therapy in mecha-nically ventilated patients. Am J Respir Crit Care Med 1997;156:3-10.

4. Mouloudi E, Katsanoulas K, Anastasaki M, Askitopoulou E, Geor-gopulos D. Bronchodilator delivery by metered dose inhaler inmechanically ventilated COPD patients: influence of end-inspi-ratory pause. Eur Resp J 1998;12:165-9.

5. Hess D. Aerosol delivery during mechanical ventilation. Miner-va Anaesthesiology 2002;68:321-5.

6. Special problems in aresol delivery. Artificial airways. ConsensusConference on aerosols and delivery devices. Respiratory Care2000;45:636-45.

7. Martín LD, Bratton SL, Walker LK. Principios y práctica de lasmedidas e apoyo respiratorio y ventilación mecánica. En: RogerMC, Helfaer MA (eds). Cuidados Intensivos en pediatría. Méxi-co 2ªed. McGraw-Hill Interamericana. 2000; 152-201.

8. Mouloudi E, Georgopoulos D. Treatment with aerosols in mecha-nically ventilated patients: is it worthwile? Curr Opin Anaes-thesiol 2002;15:103-9.

9. King D, Earnshaw SM, Delane JC. Pressurised aerosol inhalers.The cost of misues. Br J Pract 1991;41:48-9.


10. Turner MO, Gafni A, Swan D, Fitzgerald JM. A review and eco-nomic evaluation of bronchodilator delivery methods in hospi-talized patients. Arch Intern Med 1996;156:2113- 8.

11. Giner J, Basualdo LV, Casan P, Hernández C, Macián y cols. Nor-mativas sobre la utilización de fármacos inhalados. En normati-vas SEPAR. Arch Bronconeumol 2000;36:34-43.

12. Parkes SN, Bersten AD. Aerosol kinetics and bronchodilatadorefficacy during continuos positive airway pressure delivered byface mask. Thorax 1997;52:171-5.

13. Pollack CV Jr, Fleisch KB, Dowsey K. Treatment of acute bron-chospasm with beta-adrenergic agonist aerosols delivered by anasal bilevel positive airway pressure circuit. Ann Emerg Med1995;26:552-7.

14. Mouloudi E, Katsanoulas K, Anastasaki M, Hoing S, GeorgepoulosD. Bronchdilators delivery by metered-dose inhaler in mechani-cally ventilated COPD patients: influence of tidal volume. Inten-sive Care Med 1999;25:1215-21.

15. Hess D, Fisher D, Williams P, Pooler S, Kacmarek RM. Medicationnebulizer performance. Effects of diluent volume, nebulizer flow,and nebulizer brand. Chest 1996;110:498-505.

16. Newman SP, Pavia D, Moren F, Sheahan NF, Clarke SW. Depo-sition of pressurized aerosols in the human respiratory tracy. Tho-rax 1981;36:52-5.

17. Hess DR, Acosta FL, Ritz RH, Kacmarek RM, Camargo CA. Theeffect of heliox on nebulizer function using a betaagonist bron-chodilator. Chest 1999;115:184-9.

18. Mouloudi E, Prinianakis G, Kondili E, Georgopulos D. Effect ofinspiratory flow rate on beta2 agonist induced Bronchodila-tion in mechanically ventilated COPD patients. Intensive CareMed 2001;27:42-6

19. Segal MS, Salomon A, Dulfano MJ, Herschfus JA. Intermittentpositive pressure breathing; its use in inspiratory phase of res-piration. N Engl J Med 1954;2025-32.

20. Chatmongkolchart S, Schettino GP, Dillman C, Kacmarek RM,

H. In vitro evaluation of aerosol bronchodilator delivery duringnoninvasive positive pressure ventilation: effect of ventilator set-tings and nebulizer position. Crit Care Med 2002;30:2515-9.

21. Dolovich MB, Killian D, Wolff RK, Obminski G, Newhouse MT.Pulmonary aerosol deposition in chronic bronchitis: intermittentpositive pressure breathing versus quiet breathing. Am Rev RespDis 1977;115:397-402.

22. Mouloudi E, Prinianakis G, Kondii E, Georgopoulos D. Effect ofinspiratory flow rate on b2-agonist induced Bronchodilation inmechanically ventilated COPD patients. Intensive Care Med2001;27:42-6.

23. Thomas SH, Langford JA, George RJ, Geddes DM. Aerosol depo-sition in the human lung: effect of high-frequency oscillation onthe deposition characteristics of an inhaled aerosol. Clin Sci (Lond)1998;75:535-42

24. Good M, Fink JB, Dhand R, Tobin MJ. Improvement in aerosoldelivery with helium-oxygen mixtures during mechanical ven-tilation. Am J Respir Crit Care Med 2001;163:109-14.

25. Esquinas AM, Carrillo A, González Diaz G. Nebulization with gasheliox vs Oxygen. Ar. Evaluation of therapy and cardiorespira-tory effects in hypercapnic COPD. A pilot study. American Jour-nal Respiratory and Critical Care Medicine 2001;163:A680.

26. Shenfield GM, Evans ME, Walker SR, Paterson JW. The fate ofnebulized salbutamol administered by intermittent positive pres-sure respiration to asthmatic patients. Am Rev Resp Dis1973;108:501-5.

27. Salcedo Posadas A. Casado Flores J, Serrano A. Asma. En: Urgen-cias y tratamiento del niño grave. Madrid. Editorial Ergon, 2000;202-3.

27. Marik PE, Varon J, Fromm R. The management of acute severeasthma. J Emerg Med 2002;23:257-68.

28. Mandelberg A. Nebulized 3% hypertonic saline solution treat-ment in hospitalized infants with viral bronchiolitis. Chest2003;123:481-7.


Fiberoptic bronchoscopy (FOB), or flexiblebronchoscopy, is an essential diagnostic tool forvarious respiratory diseases. Moreover, as FOB canbe employed from the patient's bedside, it can beused for critical patients. Nonetheless, this techniquehas certain limitations which can adversely affectpatients.

Fiberoptic bronchoscopy in pediatric patientsThe main risks involved in FOB for pediatric

patients depend on the patient's age and respiratorypathology. Hence, for each patient, hospital staffshould determine the suitability of this technique,the most adequate entry route and any precautionswhich should be taken during examination.

Pediatric airways and the size of fiberoptic bronchoscopes

There are currently three sizes (based on externaldiameter) of pediatric flexible bronchoscopes: 2.2,3.5 and 4.9 mm. Given that children's airways arerelatively small and increase with age, then theyounger the child, the greater the resistance thata bronchoscope will imply.

Functional residual capacity of the lungs andmetabolic oxygen consumption

Children have lower functional residual capacity(FRC) and higher metabolic oxygen consumptionthan do adults. Moreover, in pediatric patients with

an underlying pathology, the ventilatory reserve maybe even more reduced, thereby further complicatingFOB and increasing its inherent risk of altering bloodgas levels.

Analgesia and sedationExamining patients by FOB causes temporary

alterations in lung mechanics and in gas exchange,which can lead to varying degrees of hypoxemia andhypercapnia. These changes are primarily dictated bythe age and underlying pathology of each patient.Moreover, treating the patient before examinationwith analgesics and/or sedatives that can affectrespiratory function may exacerbate these effects.

Entry routesBronchoscopy is typically performed transnasally,

although it can also be done via face mask, laryngealmask or endotracheal tube. The choice of routedepends on the patient's clinical profile.

TransnasalThe transnasal route is used for the most stable

patients. It enables complete examination of theairways. Supplementary oxygen is usually deliveredvia nasal cannulae.

Laryngeal maskEntry via laryngeal mask allows for safer

oxygenation and ventilation of the patient; hence,

Fiberoptic bronchoscopy in non-invasive ventilation

M.I. Barrio, M.C. Antelo and M.C. Martínez

Chapter16

it is used in less stable patients. However, since thesemasks are supported directly on the glottis, this routedoes not enable examination of the supraglotticarea.

Endotracheal tubeExamination via endotracheal tube is performed

for the most critical patients. Hospital staff mustchoose an appropriately sized bronchoscope whichwill permit simultaneous exploration and ventilation.Ventilator parameters and oxygen delivery may haveto be temporarily increased during the procedure.This route only permits visualization of the bronchialbranches and sample collection; it does not allowfor examination of the trachea or the laryngeal area.If these areas need to be explored, then theendotracheal tube must be gradually slid upwardsto provide access to the bronchoscope. Thismaneuver requires that all the equipment for possiblereintubation be prepared in advance.

Face maskBronchoscopy via face mask allows full

visualization of the airways as well as delivery ofinhaled anesthetics, supplementary oxygen,continuous positive pressure ventilation (i.e. CPAP)or intermittent mechanical ventilation (IMV). Thesemasks can also be adapted to the patient's existingNIV system.

Clinical basics of fiberoptic bronchoscopyduring non-invasive ventilation

Several articles have been published in the pastfew years in which the risk of gasometricinterference during bronchoscopy was reducedby introducing the instrument through a mask.Erb et al. used a specially designed mask with aterminal opening equipped with a siliconemembrane. To examine adult NIV patients,Antonelli et al. used a face mask featuring a T-piece(Vitalsigns, Inc.) for insertion of the bronchoscope.They recommended this method for hypercapnicand/or hypoxic patients that requirebronchoalveolar lavage (BAL) and whose PaO2is not greater than 75 mm Hg or whose SatO2,with supplementary oxygen and spontaneousbreathing, is not greater than 90% (based on theguidelines published by the American ThoracicSociety in 1990). Da Conceiçao et al. used a

FibroxyTM (Peters) mask modified with twoopenings: an upper one, for connection to NIV;and a lower one, covered with a siliconemembrane, for introduction of the bronchoscope.Maitre et al. recently reported good resultsobtained for hypoxic patients using the Vygon®

Boussignac CPAP system adapted to a face mask.Their set-up features two connections: a standardoxygen inlet and a funnel-shaped inlet comprisingfour microchannels which, when connected to astrong flow of oxygen, produces four high-pressurestreams that provide CPAP. The CPAP can bemeasured using a pressure transducer. Since thissystem is open to the atmosphere, the jet flowor pressure do not have to be changed before theflexible bronchoscope is introduced.

Adaptors for performing fiberoptic bronchoscopy with mechanical ventilation

FOB can be performed in pediatric NIV patientsor in unstable patients that require simultaneouspositive pressure ventilation by using the followingdevices:

Face mask and T-pieceThe bronchoscope can be introduced through

a T-piece connected to the terminal opening of aface mask (see Fig. 1). The instrument should firstbe inserted into the T-piece and mask, and thenentire set-up should be positioned on the patient,and then attached using elastic straps. If the patientis pre-medicated and sedated according to standardFOB procedure, then the NIV system can be

116 M.I. Barrio, M.C. Antelo, M.C. Martínez

Figure 1. Face mask with terminal opening and T-piece.

connected using the same parameter settings earlierused for their ventilation, and oxygen delivery canbe slowly increased according to their needs.

During CPAP ventilationFOB in unstable non-ventilated patients that,

despite receiving large volumes oxygen, have limitedoxygen saturation, can be performed using theVygon® Boussignac CPAP kit (Fig. 2), which isadapted to a face mask.

CONCLUSIONS FOB and NIV are not only compatible, they can

actually be complimentary. Choosing the best entryroute and materials for each patient depends

primarily on their age and pathology. There arecommercially available adaptors for facilitatingsimultaneous bronchoscopy and NIV in pediatricpatients.

REFERENCES1. Antonelli M, Conti G, Riccioni L, Meduri GU. Noninvasive posi-

tive-pressure ventilation via face mask during bronchoscopy withBAL in high-risk hypoxemic patients. Chest 1996;110:724-8.

2. Antonelli M, Conti G, Rocco M, et al. Noninvasive positive-pres-sure ventilation vs. conventional oxygen supplementation inhypoxemic patients undergoing diagnostic bronchoscopy. Chest2002;121:1149-54.

3. Da Conceicao M, Favier JC, Bidallier I, Armanet L, Steiner T, GencoG, et al. Intubation fibroscopique sous ventilation non invasiveavec un masque facial endoscopique. Ann Fr Anesth Reanim2002;21:256-62.

4. Erb T, Hammer J, Rutishauser M, Frei FJ. Fibreoptic bronchos-copy in sedated infants facilitated by an airway endoscopy mask.Paediatr Anaesth 1999;9:47-52.

5. Goldstein RA, Rohatgi PK, Bergofsky EH, Block ER, Daniele RP,Dantzker DR, et al. Clinical role of bronchoalveolar lavage inadults with pulmonary disease. Am Rev Respir Dis 1990;142:481-6.

6. Maitre B, Jaber S, Maggiore SM, Bergot E, Richard JC, BakthiariH, et al. Continuous positive airway pressure during fiberopticbronchoscopy in hypoxemic patients. A randomized double-blind study using a new device. Am J Respir Crit Care Med2000;162:1063-7.

7. Nussbaum E, Zagnoev M. Pediatric fiberoptic bronchoscopy witha laryngeal mask airway. Chest 2001;120:614-6.

8. Slonim AD, Ognibene FP. Enhancing patient safety for pediatricbronchoscopy: alternatives to conscious sedation. Chest 2001;120:341-2.

Fiberoptic bronchoscopy in non-invasive ventilation 117

Figure 2. Vygon® Boussignac CPAP system.

INTRODUCTIONThe high incidence of respiratory disease among

neonates, combined with advances in respiratorymedicine, has led to a wide arsenal of strategies forneonatal respiratory assistance. Since mechanicalventilation of neonates was begun in the 1960's, themortality rate of newborns with respiratorypathologies has decreased dramatically. In 1971,Gregory first reported the use of continuous airwaypressure as an early treatment for respiratory distresssyndrome (RDS). In 1973, Agostino reported thetreatment of a series of low birth weight neonatesusing nasal continuous positive airway pressure (n-CPAP). Non-invasive ventilation gradually becamewidespread, having been applied through myriaddevices such as negative pressure chambers, facechambers and face masks; however, none of thesemethods entered common practice due to theirrespective secondary effects. Currently, the mostwidely used NIV devices are nasal interfaces.

DELIVERY METHODS

Nasal continuous positive airway pressureNasal CPAP (n-CPAP) is the continuous delivery

of a gas to the airway at a fixed pressure via thenostrils.

Nasal intermittent positive pressure ventilation Nasal intermittent positive pressure ventilation

(n-IPPV) is the combination of n-CPAP with periodiccycles of positive pressure whose frequency andparameters are set by the operator. Albeit thismethod can theoretically be used in synchrony withventilators that detect the inspiratory efforts ofpatients (synchronized intermittent positive pressureventilation [SIPPV]), currently there is only onecommercially available system with a sufficientlysensitive flow trigger for neonatal patients, andclinical experience with this device remains verylimited. Good results have recently been reportedfor patients that were removed from mechanicalventilation and in patients with apneas. N-IPPV hasbeen shown to reduce the number of days of oxygentherapy and ventilation in preterm infants.

PHYSIOLOGICAL EFFECTSThe respiratory system of neonates, especially of

preterm infants, is characterized by its immaturity: thechest wall is extremely unstable and the airways areextremely compliant, making them highly prone tocollapse. Furthermore, neonates have low levels of lungsurfactant and a weak central respiratory impulse, furthercomplicating their ability to combat respiratory diseases.

INTERFACES AND PRESSURE-GENERATORS

Standard interfacesThere is a wide array of commercially available

interfaces for neonatal NIV. The challenge with using

Neonatal non-invasive ventilation

Grupo Respiratorio Sociedad Española de Neonatología*

Chapter 17

*F. Castillo, D. Elorza, M.L. Franco, J. Fernández, M. Gresa,A. Gutiérrez, I. López de Heredia, J. Miracle, J. Moreno, A.Losada, A. Valls i Soler.

this equipment is ensuring a proper seal betweenthe interface and the airway. There are two mainclasses:1. Pharyngeal interfaces. These are primarily used

in the nasopharyngeal area. They can be single-sided or double-sided, although the former isused in the majority of cases. These are easy toadjust and have only minor leakage, but theyincrease resistance in the airway.

2. Nasal interfaces. These encompass nasalcannulae, nasal prongs and nasal masks. Theyare more widely used than pharyngeal interfacesand are more effective for resolving failedextubation in premature infants. However, theyare difficult to adjust, can cause facial deformitiesand trauma, and suffer from frequent leakage.The most common nasal interface is short nasalprongs (see Chapter 5).

Pressure generators Both continuous- and variable-pressure

generators are used for neonatal NIV. In the formercase, the pressure is generated during expiration;the typical set-up comprises a ventilator and one ofthe aforementioned interfaces. In the latter case, thepressure is generated through the flow of thedelivered gas, which is regulated by the interfaceand the ventilator. Together, they maintain stablepressure during the length of the respiratory cycle,generating less resistance during expiration, which

improves the patient's work of breathing. N-IPPVcan be used in this type of system with or withoutsynchronization. Currently available devices forsynchronizing the patient's breathing efforts operateby detecting changes in movement (abdominalimpedance) or in flow (see Chapter 6).

INDICATIONSNeonatal NIV was initially used to treat

pathologies marked by a low FRC, such as hyalinemembrane disease and transitory tachypnea. It hasalso been used to treat patients that have just beenremoved from conventional mechanical ventilation(CMV) to prevent reintubation and apneas. As aresult of the ever increasing clinical experience withneonatal NIV, the range of indications of thistechnique has expanded to encompass nearly allneonatal respiratory pathologies.

CLINICAL APPLICATIONSClinical use of NIV in neonates, especially in

extremely premature infants, can be complicatedand requires coordinated hospital care. N-CPAP isinitially delivered at a minimum pressure of 4 to 5cm H2O with the minimum FiO2 required to maintainadequate transcutaneous oxygen saturation (SatO2)for the patient's gestational age and lung pathology(see Table II). The inspired gas should be properlyheated and humidified. The pressure should begradually raised in increments of 1 to 2 cm H2O untilan adequate lung volume is reached; in clinicalpractice, this pressure ranges from 5 to 8 cm H2O,although it varies according to each patient andtheir pathology.

Ideally, n-IPPV should be used as SIPPV;however, currently available devices are ratherlimited in the quality of synchronization that theyprovide. They enable regulation of inspiratory andexpiratory pressures, inspiratory time andrespiratory frequency. The expiratory pressure istypically set at 4 to 6 cm H2O; the inspiratory time,at least 2 cm H2O above the expiratory pressure;the inspiratory time, from 0.4 to 0.6 seconds; therespiratory frequency varies in function of thepatient's size and pathology; and the FiO2 used isthe minimum value that affords the desired SatO2

(see above, and Table II).

120 Grupo Respiratorio Sociedad Española de Neonatología

Table I. Indications of neonatal non-invasive ventilation

• Diseases in which the FRC is altered (e.g. RDS andTNT) and/or that cause increased work of breathing

• Weaning from CMV

• In premature infants: apneas

• Pulmonary edema

• Paralysis and/or paresis of the diaphragm

• Laryngotracheomalacia

• Restrictive diseases of the upper airway (e.g. BPD)

• To decrease left-right shunt through persistent duc-tus

BDP: bronchopulmonary dysplasia; CMV: conventional mecha-nical ventilation; FRC: functional residual capacity; RDS: res-piratory distress syndrome; TNT: transient neonatal tachypnea

ENSURING SUCCESSFUL NON-INVASIVE VENTILATION

The success of NIV is determined primarily bythe technical quality of the treatment and by thequality of patient care, which is especially importantfor minimizing any side effects caused by thetreatment.

Technical factorsIt is essential that the proper interface (i.e. model

and size) for each patient is chosen and positionedcorrectly. It must be fastened with just the rightamount of pressure to guarantee a good seal andconstant gas flow; too little pressure will generateleakage, whereas too much will cause local pressuresores. Proper heating and humidification of theinspired gas is also fundamental, since the high gasflows in NIV can have deleterious effects on thepatient's bronchial secretions, which in turn can leadto technical failure of the treatment and/oraggravation of the patient's respiratory pathology.Inspired gases are normally conditioned to 37 °Cand 100% relative humidity (44 mg/mL at 37 °C).

Patient careProper care of the patient's upper airway—

namely, correct positioning of their head andremoval of bronchial secretions which can block thisairway—will prevent increased airway resistance,and consequently, avoid increased work of breathing.

NIV patients are normally fed via oral-gastrictubes; however, some patients may be able to befed orally. The technique must be used carefully toensure the patient's comfort and prevent gastricdistension, which can occur if patients with increasedwork of breathing swallow the delivered gas. Indeed,the patient's comfort level is a good indicator of howwell the technique has been applied. Hospital staffmust have sufficient expertise to be able to detect

any decline in the patient's wellbeing and to resolveany problems that arise during treatment. Hospitalstaff should cooperate closely with the child's parentsand make them active participants in the child's care.

The neonate should be closely monitored,primarily by standard non-invasive methods suchas pulse oximetry, transcutaneous blood gasmonitoring, and chest wall impedancemeasurements. Chest wall impedance results canbe used to quickly assess respiratory patterns inpremature infants, who are normally difficult toexamine directly because of the high level ofhumidity and the barriers against luminalcontamination inherent to neonatal incubators.

Lung X-rays provide information on lung volume(e.g. collapse and over-compliance) which isimportant for fine-tuning the level of pressure usedin NIV.

Lastly, each ward should establish and put intopractice its own protocols for neonatal NIV.

FAILURE OF NON-INVASIVE VENTILATIONWhen faced with negative results in NIV, hospital

staff should first check the following factors beforeconcluding that the technique has failed:• Ensure that the patient's airway is correctly

positioned and that their neck is not subject toexcessive flexing or rotation (especially inextremely premature infants).

• Confirm that neither the interface nor the airwayis blocked by secretions (a frequent cause offailure in NIV).

• Check that the ventilator is operating correctlyand that the interface is positioned adequately.

• Ensure that the patient's mouth is closed (leakagethrough the mouth can lead to pressure lossesof 2 to 3 cm H2O).Failure in NIV is characterized by the following

signs and symptoms:• The patient can not reach the target PaO2 or

SatO2, and/or has high FiO2 requirements (basedon their gestational age and pathology).

• The patient requires administration of exogenouslung surfactant.

• In ARF patients: the PaCO2 rises above 60 mmHg at a blood pH of < 7.25

• The patient experiences apneas that either requirestrong resuscitation or occur more than three

121Neonatal non-invasive ventilation

Table II. Providing neonates with adequate oxygenation

• Pre-term neonates: PaO2: 50 to 60 mmHgSatO2: 88 to 92%

• Full-term neonates: PaO2: 50 to 70 mm HgSatO2: 92 to 95%

Source: Sociedad Española de Neonatología, Grupo RespiratorioNeonatal, 2001.

times per hour, with concomitant desaturationand/or bradycardia.

REMOVING PATIENTS FROM MECHANICALVENTILATION

Patients should be taken off NIV gradually oncethey reach the target PaO2 or SatO2 with low FiO2

(0.3 for ARF) or if they have not experienced apneasin the previous 24 hours. The pressure should bedecreased in increments of 1 cm H2O in function ofthe patient's response, until ultimately reaching 4cm H20. The procedure for removing patients fromn-IPPV is similar to that used in invasive mechanicalventilation.

COMPLICATIONSIf hospital personnel are well-trained and

experienced in NIV and in the ventilators used in

their ward, they can avoid the most commoncomplications that arise during treatment(summarized in Table III).

INTERPRETATIVE SUMMARY OF COCHRANECOLLABORATION ABSTRACTS

N-CPAP is widely used in the delivery room;however, whether this treatment reduces the needfor subsequent administration of lung surfactant ormechanical ventilation is currently unknown. ForRDS patients, n-CPAP is often used as an alternativeto mechanical ventilation despite a lack of clinicalevidence clearly demonstrating its efficacy. Likewise,many moderate RDS patients given n-CPAP are notadministered any lung surfactant, despite a lack ofevidence that n-CPAP obviates treatment withsurfactant. Early treatment with n-CPAP has beenshown to reduce the need for mechanical ventilation(RR 0.55; CI 95%: 0.32 to 0.96; NNT 6). Also, n-CPAP reduces the need for reintubation if it is appliedafter extubation at a pressure of 5 cm H20. Thereare no data supporting the superiority of single ordouble nasal prongs in n-CPAP. However, doublenasal prongs have been shown to be better atreducing the need for reintubation (RR 0.59; CI 95%:0.41 to 0.85; NNT 5).

RECOMMENDATIONSRecommendations for neonatal NIV are listed

below. The level of each recommendation (indicatedby a letter from A to D) and the strength of clinicalevidence for each one are summarized in Table IV. 1. N-CPAP should be used early for all patients at

risk for developing RDS, above all, premature

122 Grupo Respiratorio Sociedad Española de Neonatología

Table III. Complications in neonatal non-invasive venti-lation

• Gastric distension

• Increased upper airway secretions, leading to obs-tructive apneas

• An incorrectly positioned interface can cause defor-mations or necrosis of the nasal septum

• Excessive pressure can diminish minute volume andincrease PaCO2

• Air escape

• Increased vascular resistance in the lungs, complica-ting venous return and weakening the heart

• Favor left-right shunt through persistent ductus

Table IV. Grades of recommendation and levels of evidence

Grade of recommendation Level of evidence

A A directly applicable, high quality meta-analysis, systematic RCT review, or RCT

B Another type of RCT meta-analysis, or a systematic review of case control studies, or a lower quality RCT, with a high probability that the relationship is casual

C Case control or cohort study with a low risk of bias or of confounding factors

D Evidence derived from series of cases, individual cases, or expert opinion

RCT: randomized clinical trial. Adapted from the "SIGN Guidelines Developer's Handbook" (www.sign.ac.uk/guidelines/fulltext/50).

infants at 30 weeks of age that are not receivingCMV, until their clinical condition can beaccurately evaluated (D).

2. N-CPAP and early administration of lungsurfactant should be considered for all RDSpatients as a means to reduce the need for CMV(A).

3. Double nasal interfaces are preferred over nasal-pharyngeal tubes since, when used at pressuresof at least 6 cm H20 (A), they reduce the needfor intubation (C).

4. After extubation, premature infants weighingless than 1.5 kg should be treated with n-CPAPto avoid reintubation (A).

REFERENCES1. Gregory GA, Kitterman JA, Phibbs RH, et al. Treatment of the

idiopathic respiratory distress syndrome with continuous positi-ve airway pressure. N Engl J Med 1971;284: 1333-40.

2. AARC Clinical Practice Guideline. Application of continuous posi-tive airway pressure to neonates via nasal prongs or naso-pharyngeal tube. Respir Care 1994;39:817-823.

3. Klausner JF, Lee AY, Hutchison AA. Decreased imposed work witha new nasal continuous positive airway pressure device. PediatrPulmonol 1996;22:188-94.

4. Lin CH, Wang ST, Lin YJ, Yeh TF. Efficacy of nasal intermittentpositive pressure ventilation in treating apnea of prematurity.Pediatr Pulmonol 1998;26:349-53.

5. Foster SJ. Nasal deformities arising from flow driver continuouspositive airway pressure. Arch Dis Child Fetal Neonatal Ed1998;78:F157-8.

6. Friedlich P, Lecart C, Posen R, Ramicone E, Chan L, RamanathanR. A randomized trial of nasopharyngeal-synchronised intermit-tent mandatory ventilation versus nasopharyngeal continuouspositive airway pressure in very low birth weight infants follo-wing extubation. J Perinatol 1999;19:413-418.

7. Grupo Respiratorio de la Sociedad Española de Neonatología.Recomendaciones sobre ventiloterapia convencional neonatal.An Esp Pediatr 2001;55:244-250.

8. Barrington KJ, Bull D, Finer NN. Randomized trial of nasal synchro-nized intermittent mandatory ventilation compared with conti-nuous positive airway pressure after extubation of very low birthweight infants. Pediatrics 2001; 107:638-41.

9. Khalaf MN, Brodsky N, Hurley J, Bhandari V. A prospective ran-domized, controlled trial comparing synchronized nasal inter-mittent positive pressure ventilation versus nasal continuouspositive airway pressure as modes of extubation. Pediatrics2001;108:13-7.

10. Courtney SE, Pyon KH, Saslow JG, Arnold GK, Pandit PB, HabibRH. Lung recruitment and breathing pattern during variable ver-

sus continuous flow nasal continuous positive airway pressurein premature infants: an evaluation of three devices. Pediatrics2001;107:304-8.

11. Mazzella M, Bellini C, Calevo MG, Campone F, Massocco D, Mez-zano P, et al. A randomised control study comparing the InfantFlow Driver with nasal continuous positive airway pressure inpreterm infants. Arch Dis Child Fetal Neonatal Ed 2001;85:F86-90.

12. Davis PG, Lemyre B, De Paoli AG. Nasal intermittent positivepressure ventilation (NIPPV) versus nasal continuous positive air-way pressure (NCPAP) for preterm neonates after extubation En:The Cochrane Library, issue 1, 2003. Oxford:Update Software.

13. De Paoli AG, Davis PG, Faber B, Morley CJ. Devices and pres-sure sources for administration of nasal continuous positive air-way pressure (NCPAP) in preterm neonates. Cochrane DatabaseSyst Rev 2008(1):CD002977.

14. Donn SM, Sinha SK. Invasive and noninvasive neonatal mecha-nical ventilation. Respir Care 2003;48:426-39.

15. Henderson-Smart DJ, Subramaniam P, Davis PG. Continuouspositive airway pressure versus theophylline for apnea in pre-term infants. En: The Cochrane Library, Issue 1 2003. Oxford:Update Software.

16. Ho JJ, Henderson-Smart DJ, Davis PG. Early versus delayed initia-tion of continuous distending pressure for respiratory distresssyndrome in preterm infants . En:The cochrane Library, Issue1, 2003. Oxford: Update Software.

17. Lemyre B, Davis PG, De Paoli AG. Nasal intermittent positivepressure ventilation (NIPPV) versus nasal continuous positive air-way pressure (NCPAP) por apnea of prematurity. En: The Coch-rane Library, Issue 1, 2003. Oxford: Update Software.

18. Subramaniam P, Henderson-Smart DJ, Davis PG. Prophylacticnasal continuous positive airways pressure for preventing mor-bidity and mortality in very preterm infants En: The cochraneLibrary, Issue 1, 2003. Oxford: Update Software.

19. Bancalari E, del Moral T. Continuous positive airway pressure:early, late, or stay with synchronized intermittent mandatoryventilation? J Perinatol 2006;26:S33-S37.

20. Owen LS, Morley CJ, Davis PG. Neonatal nasal intermittent posi-tive pressure ventilation: What do we know in 2007? Arch DisFetal Neonatal Ed. 2007; 92:414-418.

21. Verder H. Nasal CPAP has become an indispensable part of theprimary treatment of newborns with respiratory distress syndro-me. Acta Paediatr 2007;96(4):482-4.

22. Morley CJ, Davis PG, Doyle LW, Brion LP, Hascoet JM, Carlin JB.Nasal CPAP or intubation at birth for very preterm infants. N EnglJ Med 2008;358:700-8.

23. Thia LP, McKenzie SA, Blyth TP, Minasian CC, Kozlowska WJ, CarrSB. Randomised controlled trial of nasal continuous positive air-ways pressure (CPAP) in bronchiolitis.Arch Dis Child 2008;93:45-7. Epub 2007 Mar 7.

24. Te Pas AB, Davis PG, Kamlin CO, Dawson J, O'Donnell CP, Mor-ley CJ. Spontaneous breathing patterns of very preterm infantstreated with continuous positive airway pressure at birth. PediatrRes 2008 (in press).

123Neonatal non-invasive ventilation

INTRODUCTIONMedical and technological advances, an

increasing survival rate of critical patients, and abetter general prognosis for sufferers of chronicrespiratory diseases have all led to a rise in thenumber of pediatric patients and clinical situationstreated with long-term mechanical ventilation.Although conventional mechanical ventilation(CMV) via tracheal tube is the most effective formof mechanical ventilation, tracheostomies areproblematic and involve various complications,and consequently, are avoided whenever possible.The increasing use of polysomnography (PSG) inthe past fifteen years has enabled bettercharacterization of respiratory sleep disorders.When non-invasive ventilation (NIV) was first used,it was primarily indicated for children andadolescents with neuromuscular diseases (NMDs).Since then, its range of indications has expandedto encompass obstructive sleep apnea syndrome(OSAS) and parenchymal lung disease. This broadarray of indications comprises two main groups:pathologies that can be clinically improved (e.g.bronchopulmonary dysplasia [BPD] andlaryngotracheomalacia) and those which are stableor progressive (e.g. NMDs and congenital centralhypo-ventilation syndrome [CCHS]). Homerespiratory care has undergone majordevelopments in the past two decades, owing ingreat part to technological refinements ofventilation equipment.

CRITERIA FOR INITIATING NON-INVASIVE VENTILATION

When healthy individuals are asleep, they havegreater upper airway resistance, reduced tidal volume(above all, during REM sleep), lower respiratoryfrequency, reduced intercostal muscle tone andweaker central ventilatory impulse. Consequently,their PaCO2 rises by 3 to 7 mm Hg while their PaO2

drops by 3 to 9 mm Hg. Children are especiallysusceptible to these changes due to the followinganatomical and functional factors:a. A higher percentage of REM sleep (40 to 70%),

and more daytime sleep, than adults.b. A respiratory system which is still undergoing

growth and development.c. Greater upper airway resistance, especially at

the level of the subglottal area and the choanae,than adults.

d. Highly compliant chest wall.e. Diaphragm has a lower number of fatigue-

resistant muscle fibers than in adults.Hence, the number of clinical situations in

which children can benefit from home NIV isgreater than for adults. The respiratory physiologyof children, combined with the muscular hypotoniainherent to REM sleep, together facilitate moresevere respiratory sleep disorders which can havediurnal consequences.

Given that the first signs of these disorders oftenappear during sleep, PSG is one of the most widelyapplicable tools for diagnosing respiratory disorders.

Indications of non-invasive ventilationin chronic pediatric pathologies

M.H. Estêvão, M.J. dos Santos

Chapter18

Symptoms of respiratory disease may beaccompanied by hypoxemia, with or withoutconcomitant hypercapnia. However, children aresometimes treated with mechanical ventilationdespite not having experienced major changes inblood gas levels, as the disorder may manifest itselfin polypnea and intensive work of breathing, whichin turn take a massive toll on a child's energy reservesand can ultimately affect their ponderal growth.

Altered respiratory function in children has beenassociated with disturbed sleep, which, in very youngchildren, may manifest itself as daytime irritability,and in older children, may cause problems withlearning, behavior or growth.

Respiratory infections are frequent among pediatricpatients, and in some cases (depending on therespiratory pathology) are nearly permanent and caneven be fatal. The respiratory compromise resultingfrom these infections can lead to cardiovascularoverload, which can translate into pulmonary or arterialhypertension and even cor pulmonale. Hence, pediatricpatients with these types of chronic respiratoryconditions require NIV. Although NIV probably willnot prolong a patient's life—for example, in the caseof progressive NMDs—, it can be useful for relievingsymptoms (e.g. lack of air, exhaustion, and poor sleepfor the child or their family). It also limits the severityof flare-ups. NIV can be extremely useful inprophylactic treatments, enabling expansion of thechest wall and growth of the lungs. Ventilation isprimarily provided while the patient is asleep, whenrespiratory conditions are the most aggravated andthe consequences of hypoventilation are markedlyworse. Nocturnal NIV can provide a diurnal reserve ofrespiratory strength.

Correcting blood gas levels and decreasing workof breathing provide the patient with better qualitysleep (i.e. more regular sleep with disappearance ofapneas, greater relaxation and decreased sweating),and consequently, improvements in their diurnalstate (i.e. better mood, greater attention span andfewer headaches). The decrease in energyexpenditure resulting from less respiratory effort canfacilitate weight gain, and the improvements inventilatory mechanics reduce the frequency ofrespiratory infections. Lower morbidity rates translateto fewer hospital admissions, higher quality of lifefor the patient and their family, including fewer daysof lost work, and savings in healthcare costs.

The objectives of long-term home NIV aresummarized below:a. Maintain and prolong the patient's life.b. Improve the quality of life of the patient and

their family.c. Maintain adequate (i.e. age-appropriate) growth

and development of the patient.d. Promote lung growth and expansion and

prevent chest wall deformities.e. Reduce morbidity.f. Facilitate return of the patient into their family

circle.g. Optimize the cost to benefits ratio of the patient's

healthcare

AGENIV is being used for younger and younger

patients. This is most likely the result of the growingbody of clinical experience with the technique aswell as of technological advances—namely, thedevelopment of ever-smaller masks. Not long ago,children that required ventilation assistance in thefirst few years of life were initially tracheotomized,and then, gradually transitioned to NIV several yearslater. However, NIV is now used for very youngpatients (see Fig. 1).

One of the major complications to considerwhen starting early NIV is the effects of the pressurefrom the mask on the child's facial bone structure.Children's facial bones are constantly growing andhighly moldable; by age four, only 60% of the facialstructure has been formed. The risk of facialdeformation in NIV is very high in patients that sleepseveral hours per day and that can not sleep withoutventilation (e.g. CCHS patients).

126 M.H. Estêvão, M. Félix

Figure 1. Child with congenital central hypoventilation syndro-me that started non-invasive ventilation (NIV) at the age of11 days.

INITIATING NON-INVASIVE VENTILATION:DELIVERY METHOD AND LOCATION

Acute pediatric patients are typically started onNIV, or transitioned off of CMV, in the ICU. In chronicsituations and other select cases treatment is begunin the general ward. There are also reports of NIVbeing started at home; however, this method canbe problematic. Since the initial phases of NIV canultimately detect the success of treatment, thenhospital staff may consider admitting the patientinto the hospital for 3 to 5 days, so that they andtheir families can familiarize themselves with theequipment, and the patient can get accustomed tosleeping with the mask on while the ventilator isrunning. The patient and their family should also befamiliarized with the typical complications and sideeffects in NIV and how to resolve them.

In infants, NIV treatment is started by attachingthe mask to the child after they have fallen asleep.For older patients, hospital personnel first explaineach step of the procedure to the patient,endeavoring to slowly gain the patient's confidence,and then eventually start NIV over the course of theday in increasingly longer periods. These patientsare then transitioned to night ventilation, after whichthey usually improve quite rapidly and accepttreatment, especially in the case of obstructive sleepapnea syndrome (OSAS). Use of gradually risingpressure (i.e. a pressure ramp) is often required tofacilitate adaptation.

The success of the treatment is highly dependenton the extent to which the parents can motivatetheir child, especially in the case of home ventilation.In order for the patient's parents to fully appreciatethe benefits of NIV, they must be made familiar witheach step of the treatment. Albeit NIV is safe andtolerated by the majority of pediatric patients, therates of NIV failure reported in the literature rangefrom 10% to over 40%. The expertise of hospitalstaff is also critical to successful treatment.

Use of negative pressure ventilators hascontinued to decline because they are difficult totransport and imply complications such as frequentobstructive sleep apneas. Volumetric ventilatorshave specific indications, are bulky and difficult toacquire (see Chapter 6, "Non-invasive ventilationdevices"). Children generally adapt well to bi-levelpressure ventilators; hence, these have become themost widely used type of ventilator. However, even

these devices may occasionally present adaptionproblems.

Ventilation should begin at low pressure (3 to 4cm H2O) and then the pressure should be increasedaccording to the patient's level of adaptation. If thereis an obstructive component, the preferredventilation mode is continuous positive airwaypressure (CPAP). If desaturation persists even aftercorrection of the obstruction, then the mode mayhave to be switched to bi-level positive airwaypressure (BiPAP). Children tolerate BiPAP slightlybetter than CPAP. Some authors (e.g. Teague) havereported starting with an inspiratory positive airwaypressure (IPAP) of 6 to 10 cm H2O and an expiratorypositive airway pressure (EPAP) of 5 cm H2O. Highexpiratory pressures may be required for patientswith atelectasis, hypoventilation associated with V/Qmismatch, or obstructive apneas. The tidal volumeof infants or NMD patients may be too small toactivate the ventilator's trigger (especially duringREM sleep); hence, respiratory frequency shouldalways be used as a rescue parameter for thesepatients (via spontaneous/timed [S/T] mode).

With few exceptions, nasal interfaces are themost widely used type of interface for starting NIV.However, their use in small children may becomplicated by the limited number of commerciallyavailable age-appropriate models. These patientsmay need a custom molded interface; however, thisis not always an option.

Oral interfaces have been used in some pediatricpatient groups, but these require a high degree ofpatient cooperation and are not always amenableto nocturnal ventilation. Indeed, prolonged nocturnalventilation using oral interfaces can cause dentalproblems in small children.

Oral-nasal interfaces may be an alternative forpatients with mouth breathing, nasal obstruction orsignificant mouth leakage. Oral leaks often occur inNMD patients due to their muscular weakness ordue to repositioning of their lower jawbone. Oral-nasal interfaces can also be used in acute cases.However, their range of indications remainssomewhat limited due to their inherent risk offacilitating aspiration of vomit.

There is no such thing as a perfect interface. Forpatients that require several hours of ventilation perday, rotating among different interfaces can preventor at least reduce the risk of pressure sores by shifting

127Indications of non-invasive ventilation in chronic pediatric pathologies

the pressure from the interface to different areas ofthe patient's face, and consequently, varying thelocal point of maximum pressure and friction.

Once the interface has been positioned andventilation has begun, the patient must bemonitored for respiratory parameters (includingSatO2 and SatCO2 at the end of each expiration, ifpossible), recruitment of accessory muscles inbreathing, and sweating, which is especiallyimportant in the patient's head (sweating improvesas the patient becomes better ventilated).

Once the patient's hemoglobin level hasincreased, reflecting lowered respiratory distress andreduced sweating, hospital staff should explain thebenefits of maintaining the patient on regular home

ventilation to the patient's parents. In some patientsdesaturation is small, but as long as they are wellventilated, then their heart rate will decrease.Ventilation parameters can be adjusted accordingto the results of either nocturnal pulse oximetry (Fig.2) or PSG (Figs. 3 and 4).

Costal movement ensures lengthwise growth ofthe ribs—and, therefore, an increase in chest wallparameter—as well as growth of the lungs. Anyparalysis of the respiratory musculature, includingthe inter-costal muscles, leads to stunted growth ofthe ribs by 10 cm and ultimately causes paradoxicalmovement of the ribs. Children that experience thisshortening early in life will be unable to perform anyadequate costal movements. Hence, propermovement achieved through early application ofNIV can favor growth of the rib cage.

INDICATIONS OF NON-INVASIVE VENTILATIONIN CHRONIC PEDIATRIC PATHOLOGIES

NIV has an ever expanding range of indicationsin chronic pediatric pathologies (see Table I). Theseindications are classified and summarized below.Several peculiar cases are also described.

Pathologies of the respiratory system

Compromised airways1. Craniofacial malformations:

• Retrognathia.• Micrognathia (e.g. Treacher-Collins and Pierre-

Robin syndromes).• Hypoplasia of the mid-face (e.g. Pfeiffer,

Crouzon and Apert syndromes).• Macroglossia (gigantism syndrome).


Figure 2. Two year old girl with severe microretrognathia.She presented with severe obstructive sleep apnea syndro-me. Nocturnal pulse oximetry before (A, B) and after (C, D)continuous positive airway pressure (CPAP).

Figure 3. Infant being monitored by polysomnography toadjust ventilation parameters.

A)

B)

C)

D)

2. Major hypertrophy of the adenoids and of thetonsils and/or soft palette:• In patients who are temporarily or

permanently contra-indicated to surgery.• In diseases characterized by abnormal growth

of lymph tissue (e.g. sickle cell disease).3. Malacia of a segment of the respiratory tract:

• Isolated.• Associated with polymalformative syndrome.• Resulting from repeated infections.Upper-airway obstruction facilitated by atony of

the airways may occur in sleeping children withanatomical alterations or malformations (e.g.hypertrophy of the tongue or of the tonsils andpalate) or with conditions marked by increased upperairway resistance (e.g. under-development orrepositioning of the jaws, narrow palate, large softpalate, or abnormal airway curvature at the base ofthe skull). Although some of these patients maybe treated with surgery (e.g. glossopexy, lowerjawbone osteotomy, facial repositioning surgery, ortonsillectomy), they can also benefit from temporaryNIV. For some patients, ageing and growth of theface and airways may lead to spontaneousimprovements in symptoms. Contrariwise, inchildren with craniofacial anomalies, hypoxemia or

hypercarbia secondary to airway obstruction mayhave deleterious effects on their neurologicaldevelopment. Indeed, many of these patients sufferfrom hyperactivity or major delays in growth anddevelopment.

Laryngotracheomalacia is a frequent cause ofstridor in infants and is the most common causeof partial obstruction of the upper airway. Thiscondition can cause severe complications such asairway obstruction, leading to sudden death,pulmonary hypertension, cor pulmonale, stuntedgrowth, and delayed intellectual development;however, most children overcome this condition byage two. Surgery (e.g. epiglottoplasty) may notbe sufficient for some patients. However, beforehospital personnel consider treating these patientsby tracheotomy, which is associated with a highdegree of morbidity, they should consider NIV, whichreduces work of breathing. EPAP frees blockedairways and keeps the upper airway open, andincreases the FRC, which in turn dilates the pharynx.NIV is similarly advantageous for lower airwaymalacia disorders.

Lung diseases1. Cystic fibrosis.2. Bronchiectasis.3. Bronchopulmonary dysplasia (BPD).4. Diffuse interstitial pulmonary diseases.

Children with healthy lungs do not suffer fromrespiratory alterations that occur during sleep orin response to circadian rhythms. However, inpatients with a lung disease involving limitedpulmonary reserve, the normal effects of sleep onrespiration can lead to clinically significant ventilatoryand gas exchange anomalies.


Table I. Chronic pathologies for which non-invasive ven-tilation (NIV) is indicated.

1. Pathologies of the respiratory systemA. AirwaysB. Lungs

2. Neurological diseasesA. CentralB. Peripheral

3. Neuromuscular diseases and chest wall diseases4. Mixed pathologies5. Other pathologies

Figure 4. Child with achondroplasia and severe obstructivesleep apnea syndrome being monitored by polysomnographyto adjust ventilation parameters. As the continuous positiveairway pressure increases, the total number of apneas andhypopneas per hour (A + H index), and the number of micro-arousals (A/W), decrease, and the level of SatO2 increases asREM sleep begins.

Cystic fibrosis, bronchiectasis, BPD, and diffuseinterstitial pulmonary diseases are all exacerbatedduring sleep, especially due to hypoxemia duringREM sleep. Therefore, patients suffering from thesediseases can benefit from NIV (see above).

The first use of NIV for cystic fibrosis patients wasonly as temporary support during lung transplants.However, its indications for this disease have sinceexpanded to encompass treatment of symptomsincluding acute hypercapnia and respiratory musclefatigue and to assist in kinesitherapy for improvingdiminished oxygen saturation. Late-stage cysticfibrosis patients develop hypoxemia and hypercapnia,especially during sleep. Gozal demonstrated that NIVmarkedly improves alveolar ventilation during sleep,as opposed to oxygen therapy alone, which improvesoxygenation without improving sleep patterns orarousals. The same author also demonstrated thatNIV improves subjective feelings of dyspnea, facilitatesperformance of everyday tasks and increases toleranceof respiratory kinesitherapy.

BPD, which is a chronic pulmonary disease thatstrikes premature infants with respiratory difficulty,causes respiratory failure of varying degrees whichcan become aggravated during sleep, but whichgradually improves. BPD patients are typically treatedwith oxygen therapy, and in some cases (e.g. tocorrect hypoxemia) by ventilation via nasal interface.Long-term hypoxemia during growth stages cancause pulmonary hypertension and ultimately, corpulmonale, and can lead to poor weight gain.

Neurological diseases• Central:

1. Congenital:– Primary: central hypo-ventilation syndrome

(CCHS).– Secondary:

- Type II Arnold-Chiari malformation.- Leigh syndrome.- Moebius syndrome.- Pyruvate kinase deficiency.- Carnitine deficiency.

2. Acquired:– Asphyxia.– Infections.– Trauma.– Tumors.– Heart attacks.

• Peripheral.Alveolar hypoventilation is typically only observed

in patients with neuromuscular or lung diseases andis classified as congenital or acquired. In the formercase, it can be secondary to brain malformations (e.g.Type II Arnold-Chiari malformation) or functionalanomalies (e.g. pyruvate kinase or carnitinedeficiencies). In the latter case, it may be result fromvarious factors (e.g. vascular, inflammatory, orinfectious).

CCHS—which in 2003 was associated withheterozygous mutations of the PHOX2B gene byDebra Weese-Mayer and Jeanne Amiel—ischaracterized by reduced involuntary control ofrespiratory function; however, it affects voluntarycontrol to a much lesser extent. Given that NREMis the sleep phase in which automatic respiratoryfunction dominates, it is also the state during whichventilation is most compromised in CCHS.However, the consequences range in severity fromminor perturbations of NREM sleep to alterationsof both REM sleep and diurnal behavior. Thesymptoms of CCHS generally appear by age one,and normally affect patients on a daily basis;however, in the past few years, cases have beenindentified in which symptoms did not arise untillate infancy or even adulthood. CCHS does notdisappear with age.

Children with CCHS normally require ventilatoryassistance during sleep, and the most severe casesmay also require it during the day. In the past, allpediatric patients were initially tracheotomized, andthose patients treated with primarily nocturnal


Figure 5. Polysomnograph of a child with a neuromusculardisease. These diseases are characterized by the relationshipbetween desaturations (→) and REM sleep (↑), and ofteninclude obstructive apneas (OA), obstructive hypopneas (OH)and arousals.

hypoventilation were then transitioned to NIV. NIVis now used increasingly more often and in everyounger patients (see Fig. 1).

Children with severe CCHS may require adiaphragmatic pacemaker during the day, to improvetheir mobility, and NIV during sleep. Whateverventilation mode is used, the primary objective is toprevent episodes of hypoxia and their consequenceson cardiopulmonary (e.g. hypertension) and cerebral(e.g. convulsions and mental retardation) function.Indeed, the prognosis for these patients can beimproved if hypoxia can be avoided.

One of the complications that affect youngchildren treated with ventilation is hypoplasia of themid-face caused by prolonged used of masks.However, it is quite rare: only a few hundred caseshave been reported worldwide.

Type II Arnold-Chiari malformation is a complexdeformation that affects the CNS, bones and softtissue. It consists of a downward shift of the medullaand the cerebellum via the foramen magnum. It isoften associated with myelomeningocele. Type II isthe Arnold-Chiari malformation which is most relatedto respiratory sleep disorders. Compression of thespinal cord can cause central apneas due to reducedresponse to hypercapnia or by affecting theperipheral response to hypoxia via compression ofthe ninth cranial nerve. Obstructive sleep apnea mayresult from vocal cord abductor paralysis (VCAP) orfrom compromised function of the pharyngeal dilatormuscle caused by lesions in the ninth or tenth cranialnerve.

Decompression of the posterior fossa is the firsttreatment strategy employed for Arnold Chiarisyndrome. However, this may not be sufficient,especially in patients with obstructive sleep apneas;NIV would then be required.

Patients with high spinal cord lesions, which areoften caused by trauma, need immediate ventilationassistance if the lesion is above C4; if the lesion isbelow C4, they will have some respiratory autonomy,but if they suffer any complication, then they willrapidly deteriorate.

NEUROMUSCULAR AND CHEST WALL DISEASES

Respiratory failure is often a major cause ofmorbidity and mortality in acute and chronic NMD

patients, although this varies highly according tothe pathology (see Table II). Hence, respiratory failurein NMD patients with normal lungs may causehypoxemia and hypercapnia comparable to thatfound in severe lung diseases. NMD patients havedifficulty breathing primarily because they can notexpand their chest wall (i.e. lungs, rib cage anddiaphragm).

Respiratory complications caused by NMD havebeen known for quite some time; however, they areoften underestimated and occur more frequentlythan assumed. The role of sleep in thesecomplications was only recently discovered. TheCNS controls respiratory musculature differentlyduring sleep than during waking hours, which hasmajor consequences on the body's ability to captureO2 or release CO2.

The limitations that NMDs place on respiratorymusculature start during sleep and continue duringwaking hours. The first respiratory anomaly to occuris disturbance of REM sleep by micro-arousals, whichcan be considered as both a type of adaptiveresponse that shortens REM sleep time as well as aconsequence of desaturation. This leads to REM sleepdeprivation (see Fig. 5), which has serious cerebraland respiratory consequences: loss of function ofthe frontal lobe, and consequently, diminishedconcentration and alertness; mood swings, whichin children translates to irritability; decreasedresponse to hypoxia and to hypercapnia; andcompromised function of the respiratorymusculature. Furthermore, stabilization of the airwaysby dilatory muscles is diminished, obstructive apneasarise during sleep, and muscle activity in the chestwall decreases, which leads to hypoventilation.Respiration depends on proper functioning of thediaphragm.

Specific problems occur in REM sleep when theNMD implies affectation of the diaphragm. Forexample, the tidal volume decreases markedly asaccessory muscles become paralyzed and respirationbecomes dependent exclusively on the diaphragm.Infants with this type of NMD are especiallyvulnerable because their chest wall is highlycompliant and is characterized by very low tidalvolumes, and because they sleep a lot and have ahigh proportion of REM sleep. In healthy infants,diaphragmatic work may correspond to 10% of thebaseline metabolic rate, whereas infants with a


compromised diaphragm have a very low metabolicreserve.

All NMDs may imply secondary conditions suchas scoliosis or pulmonary hypoplasia, which in turncontribute to the development of atelectasis,respiratory infections and respiratory failure (andconsequently, all of the aforementioned problems).Hence, these NMD patients ultimately require NIVduring sleep.

Medical and technological advances in NIV havefomented use of HMV, which in turn has led to newapproaches to many NMDs that were previouslyconsidered fatal without exception. Home NIV hasincreased the life expectancy of pediatric NMDpatients and improved their quality of life.

Spinal muscular atrophy (SMA) is a congenitalautosomal recessive disorder of the cells of theanterior horn of the spinal cord and frequently affectsthe motor nuclei of the fifth and sixth cranial nerves.It involves mutations in the 5q 11-13 region ofchromosome 5. SMA is characterized by muscleweakness in the extremities of the limbs and

progressive affectation of the chest wall muscles.SMA does not affect the motor neurons of thephrenic nerve and only affects the diaphragm in itslatter stages. One in 25,000 newborns suffers fromSMA.

SMA is typically classified into three typesaccording to the age of onset of the symptoms andto the maximum level of motor function: Type I(Acute SMA, or Werdnig-Hoffman disease), whichoccurs in infants; Type II (Intermediate SMA orOppenheim disease), in older infants; and Type III(Mild SMA, Kugelberg-Welander disease or JuvenileSMA), in older children.

SMA is one of the most deleterious NMDs interms of respiratory consequences. In childrenwith Type I, reduced movement is detected shortlyafter birth. Their prognosis is rather grim:respiratory failure occurs early on, generally aspart of a respiratory infection, and rapidly worsensin the absence of ventilatory assistance. There isa general consensus that NIV is beneficial for allSMA patients except those with Type I. Early NIV


Table II. Effects of the most common neuromuscular diseases on respiratory function (adapted from Givan, DC)

Lesion Disease Muscular/respiratory affectation

Motor neuron Spinal muscular atrophy G Distal spinal muscular atrophy Distal muscles of the limbs

Peripheral nerve Charcot-Marie-Tooth syndrome UA; D Riley-Day syndrome UAPhrenic nerve lesions DLeukodystrophies G

Neuromuscular junctions Congenital myasthenia GMyasthenia gravis G

Muscular (types) Duchenne and Becker syndromes G; MDystrophynopathies Emery-Dreifuss syndrome G; M; D Non-dystrophynopathies Muscular dystrophy (MD) of the waist D; I; A

Facioscapulohumeral (FSH) MD, UA; neck ± D oculopharyngeal MD (OPMD)Merosine-deficient GMerosine-positive UA

Myotonic dystrophy Congenital and non-congenital UA; M; D; decreased central drive; apnea

Congenital myopathie Nemaline; central core; minicore; multicore; G; D; UA centronuclear; myotubular; fiber disproportion

Metabolic myopathies Acid maltase deficiency (Pompe's disease) UAS DMitochondrial myopathie Dehydrogenase deficiency Apnea; G

Type I respiratory complex deficiency G; MType IV respiratory complex deficiency G

D: diaphragm; G: general; I: intercostal; M: myocardial; UA: upper airway.

treatment has been proposed for stimulatingdevelopment of the chest wall and growth of thelungs. Prophylactic use of NIV (i.e. beforesymptoms of hypoventilation appear) favors chestwall development and may even reducesymptoms, especially in children with Type II (seeFig. 6). However, the prognosis for each childshould be determined on an individual basis, andnot according to their type of SMA.

Children with Type II may be able to sit downwithout help. Muscle function in these patients isonly partially compromised, and may go unnoticedbefore onset of a respiratory infection that ultimatelyleads to respiratory failure.

Distinguishing between Type I and Type II maybe difficult when diagnosing patients that exhibitmarked muscular hypotonia before the age of 6months. Dubowitz developed a grading scale foreach type whereby the condition is ranked from 1to 9: for example, an infant with Type I.9 is closerto Type II than to more severe levels of Type I (I.1or I.2).

Children with Type III may have total freedomof movement. Respiratory failure in Type III typicallyoccurs in late adolescence or early adulthood.

Duchenne myopathy is the most common formof muscular dystrophy (MD) among infants, affectingone in 3,500 newborns. It is an X-linked recessivegenetic disease. The severity of Duchenne MD isreflected in the extent to which the patient isdeficient in the protein dystrophin. The least severeform is known as Becker MD.

Pulmonary function in MD patients ischaracterized by three phases: in the first, up to age

ten, the patient has normal lung growth anddevelopment; in the second, pulmonary functionbegins to plateau and the patient begins to losemuscle strength; and in the third, the patient suffersa gradual loss of forced vital capacity.

The first signs of diminished respiratory functionin MD patients appear during sleep: obstructive sleepapneas, in younger children, and then hypoventilation,later on. These complications typically arise duringREM sleep, when vital capacity is below 30% of thetheoretical value, and increase as the patient ages.Patients with mild to medium nocturnal hypoxia areoften asymptomatic; they do not begin to displaysymptoms until respiratory perturbations affect theirREM sleep. This is one of the greatest sources ofcontroversy regarding the benefits of NIV. In onestudy, NIV treatment led to significant improvementsin nocturnal blood gas exchange but did not affectsleep patterns, strength of respiratory musculature,number of hospitalizations, or rate of invasiveventilation. Contrariwise, another study reported alower number of hospitalizations among patientstreated with NIV compared to those treated bytracheostomy. Long-term NIV does not appear to bebeneficial.

In myopathies and congenital MDs the level ofmuscle weakness, and its limitations on respiratoryfunction, are related to the myopathic or dystrophicprocess. In some cases, NIV may prove insufficient,and the patient must be switched to CMV bytracheostomy.

Isolated scoliosis, unless it is very severe, rarelycauses respiratory failure that requires CMV.Patients that do require invasive ventilationtypically show signs of muscle weakness, rigidspine syndrome or parenchymal lung disease (seeFig. 7).

MIXED PATHOLOGIESRespiratory difficulties may be caused by

complex mixed etiologies. The clearest examples of these pathologies are

the genetic syndromes listed below, as well asobesity-hypoventilation syndrome:• Down's syndrome.• Mucopolysaccharidosis (MPS).• Prader-Willi syndrome.• Achondroplasia.


Figure 6. Four yearold boy with spinalmuscular atrophy thatstarted non-invasiveventilation (NIV) atthe age of 18 months.He does not have anychest wall deforma-tion or symptoms.

Down's syndrome is one of the most widespreadgenetic disorders, affecting 1.5 in every 1,000newborns. The particular facial characteristics ofchildren with Down's syndrome are partially due totheir abnormal cranial structure. This malformation,combined with the fact that Down's syndromepatients have narrow upper airways, suffer fromglossoptosis or macroglossia and are predisposedto other conditions (e.g. obesity, hypothyroidismand generalized hypotonia), leads to a 30 to 50%higher rate of obstructive sleep apnea syndrome(OSAS) in children with Down's syndrome ascompared to healthy children. Pulmonaryhypertension induced by airway obstruction occursmore quickly in children with Down's syndrome, asit is facilitated by other factors such as pulmonaryhypoplasia, congenital cardiopathy (in 30 to 40%of patients), and recurrent respiratory infections.OSAS often goes unnoticed by hospital staff andpatient's families: many of its sequelae (e.g. growth,alterations in behavior or sleep, and pulmonaryhypertension) are often confused with effects of thesyndrome itself.

Airway obstruction in Down's syndrome patientsis usually caused by hypertrophy of the tonsils andadenoids; therefore, it is often treated by surgicalresection of the affected tissue. However, patientsthat do not suffer this hypertrophy or for whomsurgery does not provide the desired results must betreated with NPPV. Good results have been reportedfor home NPPV via nasal mask. Other treatments forrespiratory sleep disorders in Down's syndromepatients include uvulopalatopharyngoplasty (UPPP),tongue reduction, midface advancement, andtracheostomy.

Patients with mucopolysaccharidosis (MPS) oftensuffer from compromised respiratory function: their

tongue, nasal mucous membranes, oropharynx,epiglottis, tracheal wall and bronchi becomeenlarged, ultimately leading to total obstruction ofthe airways. This progressive process is associatedwith the characteristic cranial shape of MPS patients,which favors recurrent respiratory infections as wellas airway obstruction, which can be severe and causeOSAS.

MPS patients also suffer from bone malformations(e.g. scoliosis, hyperkyphosis, thoracolumbargibbosity and/or lumbar hyperlordosis) that can affectrespiratory function. The extent of respiratorycompromise depends on the type of MPS: the mostsevere consequences occur in Type I (Hurlersyndrome) and Type II (Hunter syndrome). Thesepatients are particularly difficult to intubate;furthermore, due to their spinal column abnormalities,they must be intubated with extreme caution. Themultifaceted etiology of these patients complicatestheir treatment. In the past, sleep apneas in thesepatients were treated by tonsillectomy and/oradenoidectomy. However, due to the diffuseaffectation of the airway and to the continuous localdeposits of polysaccharides that occur in MPS, theseapproaches are only temporary solutions. Hence,NPPV has become the preferred method for long-term efficacy.

Prader-Willi syndrome is characterized by infantilehypotonia, obesity, hypogonadism, altered behavior,and mild mental retardation. It affects one in 15,000newborns. Nearly 70% of Prader-Willi patients carrya deletion in the q11q13 region of the long armof chromosome 15.

Respiratory problems in Prader-Willi syndromehave a mixed etiology, encompassing peripheral andcentral mechanisms: muscular hypotonia, reducedlung function, facial dimorphism with arched palateand narrow airways, hyperplasia of the tonsils andadenoids, obesity, and malfunctioning of thehypothalamus and of chemical receptors. NIV is acommon treatment for Prader-Willi patients withrespiratory sleep disorders; however, it is often difficultto administer due to the behavioral perturbationsthat frequently accompany the syndrome.

Achondroplasia is an autosomal dominantdisorder that is manifested in various phenotypes(see Fig. 8). Achondroplasia patients are predisposedto respiratory sleep disorders (including suddendeath) through various mechanisms:


Figure 7. Non-invasiveventilation of a child witha severe chest wall defor-mity.

1. Obstruction of airways:A. Mechanical:

• Short cranial base.• Hypoplasia of the middle third of the face.• Relative hypertrophy of the tonsils and

adenoids.B. Neurological:

• Respiratory control centers are compromisedby compression of the brain stem at the(stenosed) foramen magnum

2. Restriction:A. Small chest wall.B. Reduced anterior-posterior diameter.The relative severity of each respiratory condition

may vary with the phenotype of achondroplasia.Respiratory disorders in these patients are bestcharacterized via PSG (Fig. 8).

Patients with severe obstructive sleep apneasthat can not be resolved through resection of thetonsils and the adenoids or through another surgicalprocedure (e.g. decompression of the spinal cord)must be treated with NIV.

NIV has also been used for conditions such ascardiopathies, although to a much lesser extent.

At Hospital Pediátrico de Coimbra, the largestpatient group treated with NIV comprises NMDpatients (approx. 40%), followed by those with sometype of airway obstruction (22%).

The decision to initiate NIV is influenced byseveral factors:• The technical limitations found in adult patients

are exacerbated in pediatric patients, who oftenhave structural alterations in the upper airway,high respiratory frequencies, and mouth leakageand suffer from patient-ventilator asynchrony.

• Each family's definition of quality of life for theirchild.

• Obtaining the cooperation of the child and theirfamily to optimize treatment.

REFERENCES1. Kamm M, Burfer R, Rimensberger P, Knoblauch, Hammer Jürg.

Survey of children supported by long-term mechanical ventila-tion in Switzerland. Swiss Med WKLY 2001;131:261-6.

2. Jardine E, Wallis C. Core guidelines for the discharge home ofthe child on long term assisted ventilation in the United King-dom. Thorax 1998;53:762-7.

3. Hardart MKM, Truog RD. Spinal muscular atrophy - type I. ArchDis Child 2003;88:848-50.

4. Gozal D. Nocturnal ventilatory support in patients with cysticfibrosis: a comparison with supplemental oxygen. Eur Respir J1997;10:1999-2003.

5. Fauroux B, Boule M, Lofaso F. Chest physiotherapy in cystic fibro-sis: improved tolerance with nasal pressure support ventila-tion. Pediatrics 1999;193:E32.

6. Fauroux B, Sardet A, Foret D. Home treatment for chronic res-piratory failure in children: a prospective study. Eur Respir J1995;8:2062-6.

7. Beckerman RC, Brouillette RT, Hunt CE (ed). Respiratory ControlDisorders in Infants and Children. Baltimore: Williams & Wilkins1992.

8. Voter KZ, Chalanick K. Home oxygen and ventilation therapiesin pediatric patients. Curr Op Pedia 1996;8:221-5.

9. Gates AJ. Home Ventilation. In: Hilman BC (ed). Pediatric Respi-ratory Disease: Diagnosis and Treatment. Philadelphia: WB Saun-ders Company; 193:913-9.

10. Toder DS. Home Care of Children Dependent on RespiratoryTechnology. Pediatrics in Review 1997;18:274-81.

11. Nicholas S. Hill (ed). Noninvasive Positive Pressure Ventilation:Principles and Applications. New York: Futura Publishing Com-pany, Inc 2001.

12. Teague WG. Pediatric Application of Noninvasive Ventilation.Respiratory Care 1997;42:414-23.

13. Simonds AK, Ward S, Heather S, Bush A, Muntoni F. Outcomeof paediatric domiciliary mask ventilation in neuromuscular andskeletal disease. Eur Respir J 2000;16:476-81.

14. Barois A, Leclair-Richard D, Abinum M-F, Golovtchan C. Princi-pes de la prise en charge d'un enfant atteint de maladie neu-romusculaire. In: Barois A (ed). Maladies neuromusculaires(Progrès en pediatrie). Paris; Doins éditeurs 1998;20-42.

15. Fauroux B, Pigeot J, Polkey MI, et al. Chronic Stridor Causedby Laryngomalacia in Children. Am J Respir Crit Care Med2001;164:1874-8.

16. Gaultier C (ed). Pathologie respiratoire du sommeil du nourris-son et de l'enfant. Paris: Vigot 1989:109-11.

17. Ferber R, Kryger M (eds). Principles and Practice of Sleep Medicinein the Child. Philadelphia: WB Saunders Company 1995:217-30.

18. Gaultier C. Troubles respiratoires au cours du sommeil de l'en-fant. Rev Prat (Paris) 1989;39:21-5.

19. American Thoracic Society. Idiopathic Congenital Central Hypo-ventilation Syndrome - Diagnosis and Management. Am J Res-pir Crit Care Med 1999;160:368- 73.

20. Gozal D. Congenital Central Hypoventilation Syndrome: AnUpdate. Pediatr Pulmonol 1998;26:273-82.

21. Marcus CL, Jansen MT, Poulsen MK, Keens SE, Nield TA, Lips-ker LE, Keens TG. Medical and psychological outcome of chil-


Figure 8. Seven monthold girl with achondro-plasia. Despite havingundergone spinal corddecompression, she stillrequires non-invasiveventilation (NIV).

dren with congenital central hypoventilation syndrome. J Pediatr1991;119:888-95.

22. Loughlin GM, Caroll JL, Marcus CL. Sleep and Breathing in Chil-dren. A Developmental Approach. New York: Marcel Dekker, Inc.2000.

23. Voter KZ, Chalanick K. Home oxygen and ventilation therapiesin pediatric patients. Curr Opin Pediatr 1996;8:221-5.

24. Sivak ED, Shefner JM, Sexton J. Neuromuscular disease and hypo-ventilation. Curr Opin Pulm Med 1999;5:355-62.

25. Barois A, Estournet-Mathiaud B. ventilatory support at in chil-dren with spinal muscular atrophies (SMA). Eur Respir Rev 1992;10:319-22.

26. Simonds AK. Paediatric non-invasive ventilation. In: Simonds AK(ed). Non-Invasive Respiratory Support. London: Arnold, 2001.

27. Hernandez MEC, Elliot J, Arens R. Clinical outcomes of noctur-nal non-invasive ventilation in patients with Duchenne muscu-lar dystrophy. Am J Respir Crit Care Med 2000;161:A555.

28. Bach JR, Ishikawa Y, Kim H. Prevention of pulmonary morbidityfor patients with Duchenne muscular dystrophy. Chest 1998;112:1024-8.

29. Andrews TM. Airway Obstruction in Craniofacial Anomalies.In: Meyer CM III(ed). The Pediatric Airway. JB Lippincott Com-pany 1995;247-61.

30. Levanon A, Tarasiuk A, Tal A. Sleep Characteristics in childrenwith Down. JPediatr 1999;134:755-60.

31. Stebbens VA, Dennis J, Samuels, MP, Croft CB, Southfall DP. Sleeprelated upper airway obstruction in a cohort with Down's syndro-me. Arch Dis Child 1991;66:1333-8.

32. Estêvão MH. Ventilação Não Invasiva no Domicílio em Pediatria.Acta Pediatr Port 2000;2:135-41.

33. Guilleminault C, Pelayo R, Clerk A, Leger D, Bocian RC. Homenasal continuous positive airway pressure in infants with sleep-disordered breathing. J Pediatr 1995;127:905-12.

34. Macieira L, Estêvão MH. Síndrome de Down e Obstrução res-piratória. Acta Pediatr Port 2000;31:269-72.

35. Nargozian CD. The difficult airway in the pediatric patient withcraniofacial anomaly. Anesthesiology Clinics of North Amer1998;16:839-52.

36. Morehead JM, Parsons DS. Tracheobronchomalacia in Hunter'ssyndrome. Int J Pediatr Otorhinolaryngol 1993;26(3):255-61.

37. Schlüter B, Buschatz D, Trowitsch E, Aksu F, Andler W. Respira-tory control in children with Prader-Willi syndrome. Eur J Pediatr1997;156:65-8.

38. Tasker RC, Dundas I, Laverty A, Fletcher M, Lane R, Stocks J. Dis-tinct patterns of respiratory difficulty in young children withachondroplasia: a clinical, sleep, and lung function study. ArchDis Child 1998;79:99-108.

39. Liner LH, Marcus CL. Ventilatory management of sleep-disor-dered breathing in children. Curr Opin Pediatr 2006;18:272-6.

40. Donna-Schwake C, Podlewski P, Voit T, Mellies U. Non-InvasiveVentilation Reduces Respiratory Tract Infections in Children WithNeuromuscular Disorders. Pediatr Pulmonol 2008;43:67-71.

41. Serra A, Polese G, Braggion C, Rossi A. Non-invasive proportio-nal assist and pressure support ventilation in patients with cysticfibrosis and chronic respiratory failure. Thorax 2002;57:50-4.

42. Simonds AK (ed). Non-Invasive Respiratory Support – A practi-cal Handbook. London: Hodder Arnold 2007.

43. Amiel J, Laudier B, Attié-Bitach T, et al. Polyalanine expansionand frameshift mutations of the paired-liked homeobox genePHOX2B in congenital central hypoventilation syndrome. NatGenet 2003;33:459-61.

44. Weese-Mayer D, Berry-Kravis E, Zhou L, et al. Idiopathic con-gential central hypoventilation syndrome: analysis of genes per-tinent to early autonomic nervous system embryologicdevelopment and identification of mutations in PHOX2B. Am JMed Genet 2003;123A:267-78.

45. Weese-Mayer DB, Berry-Kravis EM, Zhou L. Adult Identified withCongenital Central Hypoventilation Syndrome–Mutation inPHOX2b Gene and Late-Onset CHS (letter). Am J Respir Crit CareMed 2005;171:88.


INTRODUCTIONHome mechanical ventilation (HMV) was first

used in the mid-20th century as a treatment forchronic respiratory failure (CRF) caused by thewidespread polio epidemic. Now it is used inresponse to the increased survival rate of CRFpatients that depend on respiratory support and/oroxygen therapy. The improved prognosis for thesepatients is due to several factors, including medical,diagnostic and therapeutic advances, technologicaloutfitting of pediatric and neonatal intensive careunits, and broader ethical and psychological criteriafor prolonging patients’ lives with mechanicalsupport.

HMV has been made possible due to continuingdevelopments and improvements in ventilationequipment amenable to home care, which in turnhave stemmed from financial investment in thefield.

Discharging patients from the hospital, andcaring for them at home, can be laborious tasksthat demand the contributions of variousprofessionals and institutions. To ensure the besttreatment for the patient, these efforts should befocused in a specific home care program.

INVASIVE VENTILATION VERSUS NON-INVASIVE VENTILATION

Patients can be ventilated at home by twomethods:

• Invasive ventilation (via tracheostomy or by usinga diaphragmatic pacemaker)

• Non-invasive ventilation (NIV) using positivepressure (NPPV) or negative pressure (NNPV).The most widely used techniques are invasive

ventilation via tracheostomy and NPPV. Althoughthis chapter is focused on NPPV, details for the otherventilation methods are provided.

The choice of method depends on factors suchas the age at which treatment begins, pathology,degree of respiratory autonomy, and the extent towhich the child and the parents are willing or ableto cooperate. Hence, ventilation via tracheostomyis most frequently used in small children or as vitalsupport therapy in patients lacking respiratoryautonomy, whereas NIV is more widely used in olderchildren, adolescents and adults as an electivetherapy aimed at preventing flare-ups, preservingpulmonary function and increasing survival rates.

Up to the past decade, ventilation viatracheostomy has been the predominant form patternof pediatric HMV. Indeed, an overview of nineteenarticles published in between 1983 and 1998 revealsthat of the 455 total children treated, 71% weretreated via tracheostomy, as compared to only 13%for NPPV. However, data from an epidemiologicalstudy performed in Spain from June 1998 to June2003 reflect a major surge in use of non-invasiveventilation: 51% treated via tracheostomy and 44%with NPPV. This change is primarily due to majortechnical advances in NIV ventilators and interfaces.

Methodology for pediatric non-invasive ventilation in the home

M.Á. García Teresa, T. Gavela Pérez, M.J. Pérez García

Chapter 19

Some authors have reported that tracheostomycan be avoided or removed in pediatricneuromuscular disease (NMD) patients regardlessof the number of hours of ventilation that theyrequire per day. However, in order to receive NIVthese patients must be mentally stable; able tocooperate; have sufficient bulbar muscle functionto avoid aspiration of secretions which are producedby desaturation below 95%; have cough peak flow(either spontaneous or manually-assisted cough)above 2.7 L/sec; have sufficient lung compliance toenable non-invasive ventilation at a pressure < 40cm H2O; and should not require supplementaloxygen (i.e. SatO2 > 95% maintained with or withoutventilation) or have any physical conditions thatcould impede use of an NIV interface (e.g. nasalor oral mask).

Non-invasive positive pressure ventilation (NPPV)

Advantages• Technically simple, and easy to administer and

remove• Patient care in NIV is much simpler than in

invasive ventilation, making it much easier todischarge the patient from the hospital to beginhome treatment

• Can be used intermittently• Does not involve any of the complications of

invasive ventilation with tracheostomy• Compared to NNPV, NPPV does not cause

obstructive apneas, involves more portableequipment, and enables greater patient mobility

Disadvantages• Not amenable to all patients, especially young

ones• Very limited selection of interfaces for small

children• Interfaces are very uncomfortable and often

cause pressure sores• High risk of ineffective ventilation due to

excessive leakage, patient-ventilator asynchrony,persistent hypoventilation, or intermittent airwayobstruction

• When the patient moves, the interface can easilybecome de-adjusted, and consequently, provokeleaks

• More limited access to patient’s secretions

• Chronic use is difficult for patients that requiremore than 20 hours per day of mechanicalventilation or for patients that have difficultycoughing or swallowing

Other home ventilation methods

Non-invasive negative pressure ventilation In NNPV, a negative pressure ventilator creates

a negative pressure around the chest wall and/orabdomen, causing the lungs to inflate duringinspiration; return of the lungs to ambient pressuregenerates expiration. This method is less effectivethan NPPV. Its efficacy depends on the surface areaof the body submitted to negative pressure and onthe leakage in the system. NNPV has been used inchildren and adults with NMDs, kyphoscoliosis,tuberculosis sequelae, and, more recently, has beenapplied to chronic obstructive pulmonary disease(COPD) patients; however, it is generally ineffectivefor children that require continuous mechanicalventilation. It is now rarely used, having beenreplaced by NPPV.

Equipment that has been used for NNPV is listedbelow:A. Iron lung (tank): a metal cylinder into which the

patient’s body is introduced (their head remainsoutside). Once inside, the patient is inaccessibleand he can not move. Iron lungs were frequentlyused during the polio epidemic.

B. Cuirasses. A ventilator attached to the front ofthe chest wall and abdomen.

C. Ponchos. A plastic device that covers the upperbody, leaving the neck and limbs free.

D. Pneumobelt. A device that covers theabdomen.

Advantages• Does not involve any of the complications of

invasive ventilation with tracheostomy.• Easy to learn and use.• If it is only used at night, then the patient is

completely free of apparatuses during the day.• Cheaper than invasive ventilation with

tracheostomy.• The patient’s face remains uncovered.

Disadvantages• Less effective ventilation.

138 M.Á. García Teresa, T. Gavela Pérez, M.J. Pérez García

• Requires the patient to be immobile, hence it isimpractical for patients that require several hoursof mechanical ventilation per day.

• The patient remains inaccessible and poorlyvisible to hospital staff, which complicates theirmedical evaluation and care.

• Some devices cause back and shoulder pain dueto use of the supine position.

• Negative pressure can cause upper airwayobstruction, especially in children andadolescents, ultimately leading to obstructiveapnea. These patients then require ventilationvia tracheostomy.

• NNPV patients that have bulbar dysfunction andproblems with swallowing can developbronchoaspiration.

• In patients that do not have a tracheostomy,access to the airway is not guaranteed in thecase of emergency.

Ventilation with diaphragmatic pacemakerDiaphragmatic pacemakers are small devices

implanted in the chest wall—either surgically, or,more recently, via laparoscopy—that stimulate thephrenic nerve to induce contraction of thediaphragm. They are used for patients withcongenital or secondary central hypoventilationsyndrome and spinal cord injury. Currently, there isone Spanish hospital experienced with thistechnique: Hospital de Parapléjicos de Toledo.

Advantages• Enables patient mobility, affording them a greater

sense of independence with associatedpsychological benefits.

• Provides more natural (physiological)respiration.

Disadvantages• The pacemaker must be implanted and

controlled by experienced hospital staff. • Children and adolescents generally require

tracheostomy in order to avoid obstruction ofthe upper airway.

• The pacemakers are not equipped with alarms,and therefore, can go undetected if they fail tooperate.

OBJECTIVES OF HOME MECHANICAL VENTILATION

According to the American College of ChestPhysicians, the objectives of HMV comprise: 1. Prolong the patient’s life and improve their

quality of life2. Provide an environment that stimulates the

potential of each individual 3. Reduce morbidity4. Improve the patient's physical and psychological

wellbeing5. Reduce the patient’s healthcare costs 6. Facilitate the patient’s physical growth and

139Methodology for pediatric non-invasive ventilation in the home

Table I. Indications of pediatric home non-invasive ven-tilation

1. Alterations of the CNS• Congenital and acquired disorders of the respiratory

control center (e.g. central hypoventilation, whetheridiopathic or secondary to a tumor, trauma or infec-tion)

• Myelomeningocele and Type II Arnold-Chiari syndro-me

• Spinal cord injury

2. Neuromuscular diseases• Infantile spinal muscular atrophy• Congenital hypotonias• Myasthenia gravis• Paralysis of the phrenic nerve or diaphragm• Myopathies• Duchenne muscular dystrophy• Guillain-Barré syndrome• Botulism

3. Skeletal disorders• Kyphoscoliosis• Chest wall deformities

4. Respiratory diseases• Obstruction of the upper airway:

– Craniofacial deformation syndromes (e.g. PierreRobin, and Treacher-Collins)

– Laryngotracheomalacia– Obstructive sleep apnea syndrome (OSAS)

• Bronchopulmonary alterations:– Cystic fibrosis– Sequelae of acute respiratory distress syndrome

(ARDS)– Pulmonary fibrosis

5. Metabolic diseases

psychosocial development and improve theability of children to perform activitiesAnd the specific objectives of NIV comprise:

Primary objectives1. Alleviate the symptoms and signs of the

respiratory pathology.2. Improve the patient’s diurnal and nocturnal gas

exchange.

Secondary objectives1. Improve the patient’s quality of sleep, functional

state and pulmonary function.2. Extend the patient’s life.3. Reduce healthcare costs.

INDICATIONSCRF patients that require mechanical ventilation

are candidates for HMV. The pathologies that causeCRF are listed in Table I.

Clinical criteria. Before being discharged fromthe hospital, patients that have received NIV for flare-ups must be medically stable, and their infectionmust be controlled.

Some pediatric CRF patients are admitted to thehospital for NIV as elective therapy. However, sinceno specific clinical criteria have been established fortreating pediatric patients, the criteria for adultpatients have been included here (Table II).

Family criteria. Before assuming the responsibilityof HMV, the child's parents—and the child, if theyare mature enough— should be fully informed ofthe diagnosis and prognosis, as well as of theadvantages, disadvantages and risks of the plannedtreatment. Sufficiently old patients should becooperative and motivated and should befamiliarized with the treatment objectives.

ADVANTAGES• Favors the physical, psychological, intellectual

and social development of the child.• Facilitates family life.• Avoids the secondary effects associated with

prolonged hospitalization (e.g. nosocomialinfections; iatrogenic complications; lack ofaffection, family contact or positive stimuli;altered sleep patterns; feeling of being in a hostile

environment; no or inadequate academicstudies).

• Increases the number of available beds in thePICU (the number of beds occupied by chronicpatients is often high) and reduces healthcarecosts.

DISADVANTAGESFor the family:

• Stress and fear related to the responsibility ofcaring for the child; this is especially heightenedin the first few weeks following discharge of thechild from the hospital, but gradually decreasesas the caregivers become more comfortable withthe treatment.

• Change of habits and adaptation to a newsituation at home.

• May imply additional costs.• Prolonged home care may cause the patient's

family to feel exhausted, or to reject or feelresentment towards the situation.


Table II. Clinical criteria for using home ventilation as elec-tive therapy

Stable or slowly progressing chronic respiratory failure (CRF)1. Symptoms of hypoventilation or altered sleep (e.g. mor-

ning headaches, restless sleep, nightmares, diurnal hyper-somnolence, hyperactivity, cognitive or behavioralproblems, malnutrition, recurrent respiratory infections,enuresis, or late-developing cor pulmonale)

2. Changes in blood gas levels:• PaCO2 >45 to 50 mm Hg during waking hours• Nocturnal hypoventilation (SatO2 < 88% for longer

than 5 minutes uninterrupted)

3. Severe alterations in pulmonary function: forced vitalcapacity (FVC) < 20% of theoretical value (or < 50% inrapidly progressing pathologies)

4. Repeated hospitalization for respiratory flare-ups

5. The patient does not have any contra-indications fornon-invasive ventilation (NIV) (e.g. major difficulty expul-sing secretions or swallowing, severe difficulty coughingaccompanied by chronic aspiration, patient is non-coo-perative)

6. The patient suffers from hypoxia and hypercapnia des-pite receiving adequate physical therapy, bronchodi-lator treatment (in the case of a lung disease), antibioticsand diet

RISKSThe greatest danger in HMV is ventilator failure

(e.g. due to a power outtage) while the child issleeping. For patients treated with NIV as vitalsupport, inadvertent disconnection of the ventilatorcould cause fatal severe hypoxemia, or ischemic-hypoxic encephalopathy. Complications can bedetected early on by monitoring the patient via pulseoximetry.

EQUIPMENT

VentilatorsVentilators for NIV are described in detail in

Chapter 6. For home care, they should be simple,light and equipped with alarms; if possible, theyshould include an internal battery to preventdisconnection during a power outage as well as toenable ventilation of the patient during transportto the hospital for treatment of a flare-up. Forpatients that require several hours of mechanical

ventilation per day, internal batteries areindispensable, as is the option to connect theventilator to an external battery and adapt it to awheelchair (see Fig. 1).

Pulse oximetersPulse oximeters for HMV should be simple,

dependable, equipped with sonic alarms, battery-powered and highly portable. Pulse oximetry canbe used as substitute for or complement tocardiorespiratory monitoring (see Fig. 2).


Figure 1. Ventilator adapted to a wheelchair.

Figure 2. Pulse oximeter.

Figure 3. Portable aspirator for secretions.

Aspirator for removing secretionsAspirators can be powered either electrically (AC

or DC) or mechanically (by pedal). They should beeasy to use and have adjustable aspiration pressure.Aspiration can also be performed using manualdevices such as syringes, suction bulbs or manualaspirators. This may be required for NMD patientstreated with NIV (see Fig. 3).

Assisted-cough devicesDevices that help patients expel secretions are

extremely valuable for NMD patients (see Fig. 4).These are indicated in NMD patients with poorcontrol of secretions for whom manual assistedcough provides ineffective (< 160 L/min) cough peakflow.

InterfacesNIV interfaces are detailed in Chapter 5. The

most common interfaces for HMV are nasal masks.Pressure sores are a serious enough problem thatthey can lead to temporary suspension of NIV.Hence, interfaces which do not place pressure onthe nasal bridge are recommended for patientssuffering from sores: these include Adams circuits,the Simplicity (Respironics®) nasal mask and theBreeze (Puritan Bennett®) nasal pillows (seecorresponding figures in Chapter 5). NMD patients

that require diurnal ventilation can be treated witha combination of nasal and oral interfaces (Fig. 5).

Humidifiers1. Compressed oxygen: Oxygen gas is stored in

a traditional gas tank or cylinder. However, eventhe smaller of the two available sizes is heavy,and therefore, difficult to move.

2. Liquid oxygen: Liquid oxygen is stored in a tankcooled to -183 °C, from where it is readilytransferred to a small and light (i.e. highlyportable) refillable vessel. This is the most widelyused method in pediatric home oxygen therapy,as it enables a high level of patient autonomyand mobility (see Fig. 6). It should be noted thatthese vessels vent, even when not in use,emptying within 8 hours.

3. Oxygen concentrators: These devices extractoxygen from ambient air; hence, unlike the


Figure 4. Assisted-cough device (CoughAssist™ from JH Emer-son Co.)

Figure 5. Oral interface for daytime support of non-invasiveventilation delivered by nasal interface.

Figure 6. Liquid oxygen vessel.

containers above, there is no need for refilling.They deliver a maximum flow of 4 L/min andare either AC or DC powered, the latter of whichimplies an additional cost. Oxygen concentratorsare used for patients with low oxygen needs thatdo not require mobility during sleep or areimmobile, as well as for home care patients.

PREPARING THE PATIENT FOR DISHCARGEFROM THE HOSPITAL

The difficulty involved in discharging the patientfrom the hospital to begin home care depends oneach case: it may range from extremely difficult toextremely easy (in the case of children who had beenadmitted to begin NIV).

The patient's discharge plan should becoordinated by a healthcare team comprised ofspecialized doctors and nurses. This team can formpart of a larger home care group that handles otherpathologies and therapies (e.g. enteral and parenteralfeeding, and IV treatments). This multidisciplinarygroup should include medical specialists (e.g. forintensive care, pulmonary medicine, rehabilitation,nutrition, trauma, cardiology, neurology andneurosurgery), nurses, social workers, physicaltherapists, and counselors.

The location where treatment is started will varyaccording to each center, and may include PICUs,intermediate care units, hospitals for chronic patients,and pulmonary medicine or rehabilitation wards.

Guidelines for discharging patients from thehospital to start home ventilation1. Ensure that HMV treatment is indicated and that

the parents accept the responsibility. Hospitalstaff should foster parent-child relationships tohelp the parents accept the situation. Moreover,the patient and their family should be placed intouch with other patients who have already beendischarged and their families.

2. Train at least two parents or other caregivers.They should spend as much time as possible withthe child—if possible, in a private room. Thechild and their caregivers should gradually betransitioned out of the ward where the child wasadmitted, until the caregivers becomeindependent and acquire the skills to resolvedifficult situations. Hospital staff should employ

training material such as pamphlets, videos anddummies.The caregivers should be taught how to:• Position and attach the interface.• Check for leaks and pressure sores related to

the interface.• Operate all equipment (e.g. ventilator, pumps,

humidifier, aspirator for secretions, monitoringdevices, oxygen sources, and assisted-coughdevices).

• Interpret alarms and adjust all ventilationparameters (if the patient's doctor deems itsuitable).

• Administer aerosol therapy and oxygentherapy.

• Evaluate signs and symptoms of any changesin the patient's respiratory condition (e.g.difficulty breathing, changes in coloration orsecretions).

• Perform physical, speech, and occupationaltherapies as well as respiratory rehabilitationtherapy (including use of assisted-coughdevices).


Table III. Equipment required for beginning home venti-lation

1. For non-invasive ventilation (NIV):• Ventilator (pressure-controlled or volumetric)• Interface (e.g. nasal mask)• Complete respiratory circuits for the ventilator• Phonendoscope• Pulse oximeter

2. For oxygen therapy:• Oxygen source

– Concentrated oxygen gas (in small or large tanksor cylinders)

– Liquid oxygen (tank plus refillable vessel)– Oxygen concentrator

• Oxygen tubing, nasal cannulae and masks

3. Patient specific materials:• Feeding pump, silicone nasogastric tube, and gas-

trostomy connection• 10 and 50 mL syringes• Apnea monitoring device• Wheelchair with external battery• Nebulizer• Portable aspirator for secretions, and probes• Assisted-cough device

• Respond to emergency situations; this includeslearning CPR.

• Feed the child; this includes becoming familiarwith special diets and feeding techniques and,if required, gastrostomy care or how to changea nasogastric tube.

• Care for (e.g. bathing and dressing) and playwith the child.

Family is critical to the success of HMV. Theyshould be provided with the maximum supportpossible to perform home treatment. The timerequired for training depends on the pathology,whether the child is ventilated via tracheostomyor NIV, and the caregivers' motivation, amongother factors. Children who are admitted to thehospital for elective NIV therapy typically stayfor one or two days, as patient-ventilatoradaptation is normally achieved at home.

3. Provide the caregivers with all necessaryequipment and materials for HMV (summarizedin Table III). Presently, the Spanish public healthsystem provides rapid and easy access to homecare material (e.g. ventilators, humidifiers,aspirators, pulse oximeters, and assisted-coughdevices) through leasing plans with oxygentherapy companies. The amount and type ofpaper work involved depends on regional healthservices; the oxygen therapy companies canprovide information on these requirements.Monthly use of disposable materials must bemanaged in conjunction with the patient'scommunity health clinic (CAP) or thecorresponding hospital.

4. Other aspects for the healthcare team to considerbefore discharging the patient:• The patient's home and family environment

should be evaluated. The home should havesufficient space, adequate electricity, waterand heating, and a telephone line. The family'swork and economic situation should bedetermined, and other family members shouldbe encouraged to help with the child's care.

• The team should contact the pediatrician fromthe patient's community health clinic (CAP)to provide them with clinical information andto discuss any possible home visits and withthe corresponding hospital.

• The team should contact extra-hospitalemergency services (in Spain: 112) to provide

them with information on the patient for usein the event of any emergencies.

The parents should:• Advise their electric company of their child's

treatment, underscoring that the ventilationequipment requires dependable electricity.

• Make a list of emergency phone numbers (112,plus the phone numbers for the hospital andthe oxygen therapy [ventilation equipment]company).

FOLLOWING UP ON THE PATIENT AFTER HOSPITAL DISCHARGE

Hospital visits. The child should be given periodiccheck-ups. These should be planned according tothe amount of time since the child was discharged,their pathology and any secondary conditions, andthe distance between their home and the hospital.Check-ups by different specialists should becoordinated on the same day to save the family trips.In some cases the patient may have to be admittedto the hospital for one or two days.

Home visits. These should be periodic and serveto monitor the patient's progress. The number ofvisits depends on the distance between their homeand the hospital, as well as on the hospital personneland time available. The companies that lease theventilation equipment are responsible for itsmaintenance and for replacing disposable materials.

Trouble shooting. The parents should have 24hour telephone access to the hospital staff to resolveany problems that may arise; in some cases, theymay have to go to the hospital. Internet monitoringis now common practice: high speed (DSL/cable)connections enable bidirectional communication inreal time between the patient's home and thehospital with high quality audio and video.

Monitoring of HMV patients: This is detailed inChapter 10, "Monitoring of non-invasive ventilationfor pediatric patients".

Unplanned hospital visits: Respiratory infectionsand other common diseases that strike infantsfrequently lead to unplanned medical check-ups.Any blood tests given to the patient should includemonitoring of blood gas exchange, which will revealimportant information on their respiratory status.Hospital staff should avoid admitting the child tothe hospital if possible. Children treated with


ventilators are sometimes admitted to the ER out offear or uncertainty regarding their condition, ratherthan because of a medical requirement—this isespecially common among pediatric tracheostomypatients. The parents of children treated with HMVare typically highly informed, know their child well,and are capable of evaluating the situation andresolving any problems that arise. Home care patientswho are admitted to the hospital will probably besubjected to an exhaustive array of analytical testsand highly aggressive treatments, and may have toremain away from home for a prolonged period oftime. This is especially true if the child is new to thehospital.

HOME ASSISTANCEHMV is not simply a treatment: it is a way of life

for the child and their family, who tend to sufferfrom other medical, psychological and socialproblems that must be addressed in an integratedfashion. As children treated with HMV are chronicallyill and require continuous, specialized care, theirfamilies are key to treatment success. However,parents may become overwhelmed withresponsibility; hence, they require constant supportand resources from medical professionals andinstitutions.

In the United States, nurses or expert caregiverstypically attend to homecare children for 16 to 24hours daily in order to facilitate treatment and enablethe child's parents to go to work. Paying for thesehealthcare professionals accounts for 60 to 75%of the costs of HMV.

The Spanish public health system still does nothave personnel or funds earmarked for pediatricHMV care, although this may become a reality withthe passing of a new patient assistance law (Leyde Dependencia). Currently, certain professionalassociations raise funds for children with specialmedical needs. Moreover, patients who have sufferedspinal cord injuries in traffic accidents receivepayments from insurance companies to cover theexpenses of professional caregivers.

In most cases the family is in charge of care. Theyshould seek assistance from friends and relatives,and should take full advantage of the availablehuman and material resources, including those ofthe healthcare services of the autonomous

community where they reside (e.g. their communityhealth clinic, city and regional health services, theMinistry of Education, and charity organizations).

SCHOOLINGThe parents should be encouraged and

supported in their efforts to ensure proper schoolingfor their child, which often entails surpassing severalobstacles stemming from the family itself, the child'spathology, academic professionals or the publicschool system. Spanish public education institutionsemploy health professionals that can attend to thechild during school hours.

ASSOCIATED PROBLEMS

NutritionPediatric CRF patients often present with

malnutrition or, in the case of NMD patients, obesity.Their diet must be optimized for physical growth,pulmonary development and nutrition, sincemalnutrition diminishes the strength of respiratorymuscles. Some patients may have to be fed viagastrostomy or nasogastric tube.

Psychomotor developmentMany pediatric CRF patients have compromised

motor skills (fine and gross) and psychologicaldevelopment and present with cognitive disabilities.These are due factors including the underlyingpathology and any intercurrent conditions. Hence,the contributions of physical, occupational andspeech therapists, to the treatment of these childrenare invaluable. Respiratory physical therapy—especially coughing exercises—is crucial for avoidingrespiratory infections and preserving respiratoryfunction. For patients that require a wheelchair forphysical therapy, the ventilator must be connectedto a battery and then attached to the wheelchair.

REFERENCES

1. Bach JR, Niranjan V. Noninvasive mechanical ventilation in chil-dren. En: Bach JR (ed). Noninvasive mechanical ventilation. Phi-ladelphia PA. Hanley and Belfus, 2002;203-22.

2. Bach JR. Home mechanical ventilation for neuromuscular venti-latory failure: conventional approaches and their outcomes. En:


Bach JR (ed). Noninvasive Mechanical Ventilation. Philadelphia,Hanley & Belfus, 2002;103.

3. Carnevale FA, Alexander E, Davis M, Rennick J, Troini R. Dailyliving with distress and enrichment: the moral experience offamilies with ventilator-assisted children at home Pediatrics2006;117:e48-60.

4. Fauroux B, Boffa C, Desguerre I et al. Long-term noninvasivemechanical ventilation for children at home: A national Sur-vey. Pediatric Pulmonology 2003;35:119-25.

5. Fauroux B, Lofaso F. Domiciliary non-invasive ventilation in chil-dren. Rev Mal Respir 2005;22:289-303.

6. García Teresa MA. Ventilación mecánica a domicilio: aspectosprácticos. En: Fuza F (ed). Tratado de Cuidados Intensivos Pediá-tricos, 3ª ed. Norma-Capitel, Madrid, 2003.

7. Guilleminault C, Pelayo R, Clerk A, et al. Home nasal continuouspositive airway pressure in infants with sleep-disordered brea-thing. J Pediatr 1995;127:905-12.

8. Hill N. Management of long-term noninvasive mechanical ven-tilation. En: Hill NS (ed). Long term mechanical ventilation. Mar-cel-Dekker, New York, 2000.

9. Holecek MS, Nixon M, Keens TG. Discharge of the technology-dependent child. En: En Levin DL, Morriss FC (eds). Essentials ofPediatric Intensive Care (2ª ed). N York Churchill Livingstone1997:1537-48.

10. Liptak GS. Home care for children who have chronic conditions.Pediatr in Rev 1997;18:271-81.

11. Make BJ, Hill NS, Goldberg AI et al. Mechanical Ventilation beyondthe intensive care unit. Report of a Consensus Conference of Ame-rican College of Chest Physicians. Chest 1998;113:289S-344S.

12. Martínez Carrasco C, Barrio I, Antelo C, et al. Ventilación domi-ciliaria vía nasal en pacientes pediátricos. An Esp Pediatr 1997;47:269-72.

13. Miske LJ, Hickey EM, Kolb SM, Weiner DJ, Panitch HB. Use of themechanical in-exsufflator in pediatric patients with neuromus-cular disease and impaired cough. Chest 2004;125:1406-12.

14. Norregaard O. Noninvasine ventilation in children Eur Respir J2002;20:1332-42.

15. Panitch HB, Downes JJ, Kennedy JS et al. Guidelines for homecare of children with chronic respiratory insufficiency. PediatrPulmonol 1996;21:52-56.

16. Pilmer SL. Prolonged mechanical ventilation in children. PediatClin N Am 1994;41:473-512.

17. Simond AK, Ward S, Heather S, Bush A, Muntoni F.Outcome ofpaediatric domiciliary mask ventilation. Eur Respir J 2000;16:476-8.

18. Simonds AK. Respiratory support for the severely handicappedchild with neuromuscular disease: ethics and practicality. SeminRespir Crit Care Med 2007;28:342-54.

19. Anita K, Simonds, MD. Recent Advances in Respiratory Care forNeuromuscular Disease. Chest 2006;130:1879-86.

20. Sritippayawan S, Kun SS, Keens TG, Davison SL. Initiation ofhome mechanical ventilation in children with neuromusculardiseases. J Ped 2003;142:481-5.

21. Teague WG. Long-term mechanical ventilation in infants andchildren. En: Hill NS (ed). Long-term mechanical ventilation.Marcel Dekker. New York 2001;177-213

22. Teague WG. Non-invasive positive pressure ventilation: currentstatus in paediatric patients. Paediatr Respir Rev 2005;6:52-60.


INTRODUCTIONNeonatal respiratory failure requiring respiratory

support is most common in neuromuscular diseases(NMDs) such as spinal muscular atrophy (SMA) Type1, but also occurs in certain congenital malformationssuch as Pierre Robin syndrome, congenital centralhypoventilation, and bronchopulmonary dysplasia.It may result from subglottic stenosis due to damagefrom intubation or tracheostomy.

Surveys in the United States, Western Europe,and Japan indicate that pediatric use of homemechanical ventilation (HMV) is rapidly increasing.A Swedish survey noted that 80% of 541 HMV usershad neuromuscular respiratory impairment and thatthe great majority used non-invasive ventilation(NIV). A survey of pediatric HMV patients in theUnited Kingdom reported that 141 children requiredlong term respiratory assistance, of which 33 usedintermittent positive pressure ventilation (IPPV) withtracheostomy, 103 used nocturnal nasal IPPV (n-IPPV), and nine used negative pressure bodyventilators. Furthermore, 96 of these patients hadprimary respiratory impairment, and the remainderhad other conditions. Some children were treatedwith more than one non-invasive ventilation method.However, the majority of children that appeared torequire more than nocturnal ventilator use, and themajority of those who had difficulty with weaningafter intubation during chest infections, requiredtracheostomy. When the tubes had not beenremoved after the acute episode, the children usually

remained dependent on full time ventilation.Interestingly, most of the children in studied in thissurvey attended mainstream schools, but only one-half of the tracheostomy IPPV users resided in thecommunity.

Despite major advances in managing respiratoryfailure in the pediatric intensive care unit (PICU),most pediatric patients still require conventionalmechanical ventilation (CMV), which entailsintubation, mechanical ventilation, and thenweaning. This chapter reviews respiratoryphysiotherapy and intervention strategies in thisarea, describing the most important techniques forensuring successful NIV in children. Respiratoryphysiotherapy should be evaluated according todifferent factors, including measurement of bronchialmucus transport and of expectorated mucus,pulmonary function, use of medication, frequencyof acute exacerbations and quality of life.

GOALS OF INTERVENTIONAs with adult patients, the goals of intervention

in children are to maintain constant lung complianceand normal alveolar ventilation and to maximizecough flows for adequate clearance ofbronchopulmonary secretions. Techniques foraugmenting normal mucociliary clearance and coughefficacy have been used for many years in patientswith respiratory disorders of various etiologies. Recent

Respiratory physiotherapy in non-invasive ventilation

M.R. Gonçalves

Chapter20

advances have enabled more effective and morecomfortable treatment for the majority of patients.In most countries, postural drainage with manualchest percussion and shaking has been replaced bymore independent and efficacious techniques suchas the active cycle of breathing; autogenic drainage;high frequency chest wall oscillation; intrapulmonarypercussive ventilation; mechanical insufflation-exsufflation (MI-E); and use of equipment such aspositive expiratory pressure (PEP) masks, the R-CCornet®, and the Flutter® mucus clearance device.However, the evidence supporting use of thesemethods is varied, and the literature on the clinicalindications for each one is confusing and sometimesconflicting. This may be related to differences in theintensity, duration and frequency of the treatmentsused by physiotherapists in different areas of theworld.

Elimination of airway mucus and other debris iscritical to effective mechanical ventilation (invasiveor non-invasive) of chronic and acute patients withrespiratory or oxygenation impairment. In patientswith primarily respiratory impairment, 90% ofepisodes of respiratory failure result from an inabilityto clear airway mucus during intercurrent chest colds.Moreover, because of difficulties with cooperationin very young patients, acute episodes of respiratoryfailure must be properly managed, so that the childcan reach the age at which cooperation is no longera problem and morbidity and mortality can beavoided without hospitalization. Older children arecapable of following directions and imitating atherapist’s demonstration of deep breathing,coughing and active exercise.

Chest physiotherapy in children is focused onimproving ventilation and the efficacy of breathing,increasing general strength and endurance (especiallyof respiratory muscles), improving posture, andaddressing relaxation, breathing control, and pacing.Hospital staff should involve the patient’s family intheir treatment whenever possible. In some cases theparent may perform most of the techniques andrepeat instructions to their child under the directguidance of the therapist. This reinforces parent andfamily education for later use in home therapy.

Lung MobilizationInfants with NMDs are unable to take deep

breaths. Consequently, their lungs and chest walls

do not grow normally, which leads to severelyrestricted pulmonary volumes, low pulmonarycompliance, and undergrowth of the thoracic cage.

Range-of-motion therapy in pediatric patientsmaintains compliance, promotes more normal lunggrowth and prevents chest deformity. Because infantscannot air stack (see below) or cooperate to receivemaximal insufflations, those with spinal muscularatrophy (SMA) or other NMDs who have paradoxicalchest wall motion require nocturnal high-span bi-level positive pressure ventilation (BiPAP) to rest theirinspiratory muscles, prevent pectus escavatum, andpromote growth of their lungs. Use of high-spanBiPAP during sleep—even without maximalinsufflation therapy—prevents rib cage deformitiesin children with SMA Type 1, and possibly, in otherinfants with paradoxical breathing.

For patients older than 9 months inspiration-timeddeep insufflations can be delivered via an oral-nasalinterface in tandem with abdominal compression, toprevent abdominal expansion while air is directedinto the upper thorax. Children can cooperate withdeep insufflation therapy by 11 to 30 mouths of age.This therapy is usually performed two or three timesdaily with delivery of ten to fifteen maximalinsufflations via manual resuscitator or by mechanicalinsufflation-exsufflation (MI-E), using a CoughAssist™device (J.H. Emerson Co., Cambridge, MA, USA) ata pressure of 40 cm H2O and timed to inspiration.Periodic maximal insufflations may complementnocturnal BiPAP in preventing pectus escavatum andpromoting lung growth; however, there are currentlyno published data to support this approach.

148 M.R. Gonçalves

Figure 1. A 2 year old girl with SMA Type 1 undergoing high-span nasal BiPAP (n-BiPAP) after having been successfully extu-bated in the ICU.

Once accustomed to positive pressureinsufflations, some infants as young as 11 monthscan use MI-E via oronasal interface. Anothertechnique that promotes lung expansion andincreases cough efficacy is air stacking. Thistechnique, which is only efficient in cooperativepatients (children older than 2 years), consists of thefollowing: the subject takes a deep breath, thenholds their breath, and is then an extra air volumeof air (delivered from a resuscitation bag or volumeventilator through an oral or nasal interface). Saidvolume produces lung distension, enabling thepatient to cough with greater volume, and therefore,more effectively.

Another factor to consider is inspiratory apnea,which is assured by functioning of the glottis. Deepinspiration stretches the airways and increases thecontraction force of the expiratory muscles and theretraction force of the lung parenchyma. Inspiratoryapnea (with glottic closure) facilitates airwaydistribution to the most peripheral areas of the lungand increases intrathoracic pressure.

Extrinsic pulmonary development is clearlyinterrelated with musculoskeletal and motordevelopment of the trunk. Therefore, it is crucialthat doctors assess patients’ functional range ofmotion through the trunk. This requires elongationof the pectoral, sternocleidomastoid, upper trapezius,and rectus abdominis muscles. Manual lowering ofthe rib cage is also needed for maximizing chestmobility. Helping the patient achieve proprioceptiveinput of the scapulae on the posterior thorax helpsreinforce active thoracic extension and anterior chestexpansion.

Maintenance of alveolar ventilationDuring the past decade use of non-invasive

ventilation (NIV) to provide mechanical ventilationor respiratory support to various groups of pediatricpatients has grown. As part of a multidisciplinaryteam that provides NIV to children, respiratoryphysiotherapists play an important role by activelyparticipating in each stage of treatment, especiallyin ensuring that patients maintain adequate alveolarventilation. For many NMD patients, respiratorysupport is indicated for unloading retained carbondioxide from the muscles and for substituting thefailing respiratory pump, particularly during sleep.NIV is based on the cyclical application of a positive

pressure (or volume) to the airways. Previous studieshave used volume-targeted ventilators, but mostrecent pediatric studies have employed positive-pressure-targeted units.

Choosing an appropriate ventilator for childrendepends on age; vital capacity; lung and chest wallcompliance; the extent of gas exchange impairment;and the patient’s size, mobility, and level ofcooperation. Patient triggering is not always possiblefor very small and/or weak patients, for whom rateadjustment is highly important. This is typically thecase for infants extubated to nasal BiPAP (n-BiPAP).To facilitate patient ventilator synchrony for patientsunable to trigger the machine, the respiratoryphysiotherapist must set the spontaneous timedmode of pressure-cycled machines to be used witha back-up rate slightly higher than that of thespontaneous breathing rate of the patient. In general,back-up rates of 25 to 30 respirations per minute(rpm) are required for infants, 15 to 20 rpm for smallchildren, and 10 to 12 rpm for adolescents.

Low-span nasal BiPAP is generally used in childrenof all ages with obstructive apneas or primaryrespiratory insufficiency and adequate pulmonarycompliance. EPAP can replace the normal positiveend expiratory pressure (PEEP) generated by thelarynx during coughing, talking or crying. EPAP levelsmay be increased in the rare patient with NMD andsevere hypoxemic insufficiency or air trapping. EPAPis rarely needed for children with primary respiratoryimpairment and normal lung compliance, howeverit is often beneficial for older children with SMA TypeI.

Optimized clearance of airway secretionsFor pediatric patients who cough sufficiently

enough to eliminate secretions, chest physicaltherapy and assisted coughing are used as for adultpatients. Chest percussion with postural drainagecan help mobilize peripheral airway secretions butcauses oxy-hemoglobin desaturation and imposesrespiratory work that can also fatigue respiratorymuscles. These problems can be alleviated, at leastfor patients with cystic fibrosis, by the use of nasalpeak inspiratory pressure (PIP) combined with PEEP.In a study of 16 cystic fibrosis patients aged 13 to14 years, who performed forced expiratorymaneuvers, respiratory rates were significantly lowerand tidal volumes, maximal inspiratory and

149Respiratory physiotherapy in non-invasive ventilation

expiratory pressures, and SatO2 significantly higherwhen nasal PIP plus PEEP was used. Furthermore,among the patients treated with PIP plus PEEP, 15patients reported less fatigue, and 10 of 16 patientsreported that expectoration was easier with nasalPIP.

The critical level for effective coughing in adultairways, as measured at the mouth, appears to be160 liters per minute (lpm). However, no one hasdetermined the analogous level for children. Lowertotal flows may be effective in the airways of smallchildren because cough effectiveness is more afunction of air velocity than air flow. In infants withineffective cough, expectoration can be facilitated,and fatigue eased, by providing deep insufflationsand n-IPPV or high-span n-BiPAP. Infants youngerthan 1 year learn how to first close the glottis to holdthe IPAP and then quickly open it to expel secretions.

MECHANICAL INSUFFLATION- EXSUFFLATION(MI-E)

Mechanical assisted cough devices such as theCoughAssist™ (J.H. Emerson Co., Cambridge, MA,USA) deliver deep insufflations followed immediatelyby deep exsufflation. The insufflation and exsufflationpressures and the delivery times are independentlyadjustable. Abdominal thrusts are applied inconjunction with the exsufflation, except after a meal.MI-E can be provided via an oral-nasal interface, asimple mouthpiece, or an invasive airway tube (e.g.tracheostomy tube). When delivered via the latter,the cuff, when present, should be inflated. Whetherused via the nose or mouth or via invasive, indwellingairway tubes, routine airway suctioning misses theleft lung in approximately 90% of cases; hence, 80%of pneumonias occur in the left lung. MIE via anairway tube provides the same exsufflation flows inthe left and the right airways without the discomfortor airway trauma of tracheal suctioning and can beeffective when suctioning is not. Furthermore, patientsinvariably prefer MI-E to suctioning for comfort andeffectiveness and find it less tiring. Therefore, deepsuctioning, whether via airway tube or via the upperairway, can essentially be discontinued for mostchildren.

The CoughAssist™ (Fig. 4) can be manually orautomatically cycled. Manual cycling facilitatescoordination of inspiration and expiration with

insufflation and exsufflation between the care giverand the user, but requires hands to deliver anabdominal thrust, to hold the mask on the patient,and to cycle the machine. Automatic mode enablesthe patient to work alone with the machine andincrease their cough efficacy without the help of acaregiver (Fig. 5). Although the machine can bemanaged automatically by programming theinsufflations, exsufflations, and pause times, themanual mode allows for better synchronization, andmakes it easier for patients to coordinate theirinsufflation and cough with the machine. This isespecially true for infants, whereby the cycles shouldfollow their rapid respiratory frequency and chestmovement.

A single treatment session comprises roughly fivecycles of MI-E followed by a short period of normalbreathing or ventilator use to avoid hyperventilation.Multiple treatments are given in one sitting until nofurther secretions are expulsed and any secretion- ormucus-induced oxygen desaturations are reversed.For patients with chest colds, treatment may berequired as frequently as every few minutes over a24 hour period. Although no medications are usuallyrequired for effective MI-E in children with weakmuscles, liquefaction of sputum using heated aerosoltreatments may facilitate exsufflation when secretionsare inspissated.

150 M.R. Gonçalves

Figure 2. The CoughAssist™ device (J.H. Emerson Co., Cam-bridge, MA, USA).

MI-E is labor intensive and often difficult fornonprofessional caregivers in both acute and chronicsettings. Moreover, MI-E of children with advancedNMDs is nearly impossible without the use oftracheostomy tubes, unless their families andcaregivers provide virtually all of the care during upperrespiratory tract infections. After taking full care ofthe hospitalized patient, hospital staff can not simplyinstruct the family just before discharge and thenexpect them to be able to prevent another episode.Although the home or ICU patient’s family often hasthe time and motivation to use MI-E along withabdominal thrusts as often as every 15 minutes, andto use oximetry as feedback to maintain normalsaturation (without supplemental oxygen), therespiratory physiotherapy and nursing staff can notbe expected to do this.

Caregivers provide the positive pressure of MI-Etimed to diaphragm movement (abdominalprotrusion) via oral-nasal interface to 9- to 30-monthold infants on a daily basis to allow maximal lungexpansion and to accustom them to the techniqueso that they will use MI-E effectively during chestinfections. MI-E via the upper airway can be usedeffectively for children as young as 11 months: evenat this age patients can avoid crying or closing theirglottis. Between 2.5 and 5 years of age most childrenbecome able to cooperate and cough on cue withMI-E. Children that are too young to cooperate withmanually and mechanically assisted coughing, orthose who are not introduced to these methodsbefore a cold develops into pneumonia andrespiratory failure, are hospitalized and usually

intubated (a catheter is passed via the nose or mouthinto the lungs for respiratory support and airwaysecretion evacuation). It is often the case for smallchildren that until they are old enough to cooperate,each time they develop a chest infection, they needto be hospitalized and intubated.

When airway secretions need to be cleared insmall children unable to cooperate with MI-E, MI-Ewith an exsufflation–timed abdominal thrust(manually assisted coughing [MAC]) is used via anoral-nasal interface on manual mode to correlate theinsufflation and exsufflation with the child’s inspirationand expiration. MAC is avoided for the first 1.5 hoursafter a meal; hence, if assisted cough is needed, onlyMI-E, or MAC without aggressive manual thrusting,is performed. Although timing MI-E to an infant’sbreathing and crying may be helpful, when thepatient does not cooperate, MI-E can only be effectivewhen used via translaryngeal or tracheostomy tubewith the cuff inflated.

In addition to the standard hand placements forapplying thrusts to the chest and abdomen to increasecough flows, other placements can be used to applythrusts to specific lung regions. Likewise, thrusts canbe applied to stretch specific areas of the anterior andposterior chest walls. There are guidebooks thatdemonstrate these hand placements and thrustingtechniques in children (Fig. 3).

When MI-E is delivered with oral-nasal interfaces,pressures of +20 to -20 cm H2O are used initially, andthen augmented quickly, as tolerated, to the moreeffective settings of +40 to -40 cm H2O. The maximalefficacy of MIE at pressures of +40 to -40 cm H2O hasbeen demonstrated in experimental models and inboth adult and pediatric populations. Although thesevalues are generally adequate for most patients, highersettings are often required when compliance decreases(due to obesity or scoliosis) or resistance increases.When used via pediatric translaryngeal ortracheostomy tubes, a severe drop-off in flows andin pressures often demands that full lung insufflationand subsequent emptying last for at least 4 secondseach. Pressures much greater than 40 cm H2O areusually required to achieve the clinical goal of fullchest expansion and subsequent complete emptyingof the lungs in 2 to 4 seconds. However, in infants,even the maximum capabilities of the CoughAssistTMcan be suboptimal for effective cough flows acrossthe tubes.


Figure 3. An example of hand placement in thoracic expi-ratory mobilization on a child, used for airway secretion cle-arance and lung expansion.

Although there have not been any extensivestudies on the use of MIE in small children, there arepublished reports that this treatment enablesconsistent extubation of NMD patients followinggeneral anesthesia—despite their inability to breatheon their own—and ventilation of them with non-invasive IPPV. It has also permitted hospital staff toavoid intubation or to quickly extubate patients withacute respiratory failure (ARF) and with profuse airwaysecretions due to intercurrent chest infections. MI-Ein a protocol with manually assisted coughing,oximetry feedback, and home use of non-invasiveIPPV was shown to effectively hospitalizations,respiratory complications, and mortality in pediatricNMD patients. However, it may be ineffective forpatients that cannot cooperate sufficiently to keeptheir airway open, or those with a fixed upper airwayobstruction, or those whose upper airway dilatormuscles cannot maintain sufficient patency to allowfor a peak cough flow (PCF) > 160 lpm.

MI-E is very effective for resolving ARF in NMDpatients, but is rarely ever needed for stable patientswith intact bulbar function that can air stack tomaximum lung volumes and that can close their glottisagainst high pressures with an abdominal thrust.However, even in stable patients, routine use of MI-Emay be advisable, just to enable them to practice thetechnique so that they can apply it during upperrespiratory tract infections, when it is needed the most.

CHEST PHYSIOTHERAPY TECHNIQUES Approaches to preventing retention of airway

secretions include the use of medications to reduce

mucus hypersecretion or to liquefy secretions as wellas facilitation of mucus mobilization. Complimentarymethods include chest physiotherapy, which hasbeen shown to be very effective at preventingpulmonary complications in infants that haveaccumulated bronchopulmonary secretions. Theprinciples of chest physiotherapy techniques in adultscan be applied in children.

Mucus mobilization can be facilitated by the useof special breathing techniques, manual ormechanical chest vibration, direct oscillation of theair column, and postural drainage, all with or withoutdirected assisted coughing. The goal is to transportmucus from the peripheral airways to the centralairways, from which it can be more easily eliminated.Gravitational forces can enhance mucus transportwhen bronchi are vertically positioned. Posturaldrainage is the facilitation of airway drainage byhaving the patient assume gravity-assisted positions.Knowledge of the anatomy of the tracheobronchialtree is vital to effective treatment. Each lobe to bedrained must be aligned so that gravity can facilitateprogression of the secretion to the upper airways.This is probably most effective for relatively largequantities of mucus with low adhesiveness. Ninepostural positions have been described for drainingthe large bronchi. Localization of airway mucus isessential. The goal is vertical positioning of thesecretion-encumbered bronchi for sufficient time toallow drainage (generally, approximately 20minutes). The time required probably depends onthe quantity, viscoelasticity and adhesiveness of themucus. This intervention is more efficient whencomplemented with manual chest mobilizations,

152 M.R. Gonçalves

Figure 4. A 3 year old boy with SMA Type 2 being treatedwith oscillating pressure under a chest shell to help clear air-way secretions.

Figure 5. CoughAssist™ used with a face mask.

vibrations and breathing exercises. Mucusmobilization implies certain disadvantages: posturaldrainage is relatively time-consuming and mayrequire use of a specially modified bed or table.Vibration is a sustained co-contraction of the upperextremities of a caregiver to produce a vibratoryforce that is transmitted to the thorax over aninvolved lung segment. It is applied throughoutexhalation concurrently with mild compression ofthe infant’s chest wall. Vibration is proposed toenhance mucociliary transport from the peripheralof the lungs fields to the larger airways. Sincevibration is used in tandem with postural drainageand percussion, many studies do not distinguish theeffects of vibration from those of other elements.

The chest wall or abdomen can be vibratedexternally by rapidly applying oscillating pressure(high frequency chest wall oscillation [HFCWO])through a vest (e.g. the thAIRapy™ vest, AmericanBiosystems, Inc., St Paul, MN, USA), by cyclingoscillating pressures under a chest shell (e.g. theHayek oscillator, Breasy Medical Equipment, Inc.,Stamford, CT, USA) (Fig. 2), or by using chestvibrators. Vibration is either applied during the entirebreathing cycle or only during expiration. HFCWOmay act as a physical mucolytic, reducing both thespinnability and viscoelasticity of mucus andenhancing clearance by coughing. Moreover, it hasdemonstrated efficacy in assisting mucus clearancein patients with disorders associated with mucushyper-secretion and preserved muscle function, suchas cystic fibrosis (CF). HFCWO is an external, non-invasive respiratory modality proven effective formobilizing airway secretions from the smallperipheral airways and improving mucus rheologyin patients with CF. Indeed, it has become a majorcomponent of airway clearance techniques for thesepatients.

Mechanical vibration can be performed atfrequencies up to 40 Hz. Oscillation at 16 Hz can beapplied orally by high-frequency positive-pressureventilation or jet ventilation. One such air columnoscillator, based on high frequency jet ventilation,has been developed by the Bird Corporation (Exeter,UK). This hand-held device delivers 30 mL sine waveoscillations through a mouthpiece at 20 Hz. Lowerfrequency air column oscillators include theIntrapulmonary Percussive Ventilator (IPV), and thePercussionator, Impulsator, and Spanker Respirators

(Percussionaire Corp., Sandpoint, ID, USA). Theycan deliver nebulized medications while providingintrapulmonary percussive ventilation: mini-burstsof high-flow air delivered to the lungs at 2 to 7 Hz.This technique has been shown to be as effective asstandard chest physiotherapy and to assist mucusclearance in patients with secretion encumbrancefrom various etiologies, such as CF, acuteexacerbations of COPD and Duchenne muscledystrophy. IPV can be used with a mouthpiece orface mask, or endotracheal or tracheostomy tube.The primary aims of this technique are to reducesecretion viscosity, promote deep lung recruitment,improve gas exchange, deliver a vascular “massage”,and protect the airway against barotrauma. Themain contraindication is the presence of diffusealveolar hemorrhage with hemodynamic instability.Relative contraindications include active or recentgross hemoptysis; pulmonary embolism;subcutaneous emphysema; bronchopleural fistula;esophageal surgery; recent spinal infusion; spinalanesthesia or acute spinal injury; presence of atransvenous or subcutaneous pacemaker; increasedintracranial pressures; uncontrolled hypertension;suspected or confirmed pulmonary tuberculosis;bronchospasm; empyema; large pleural effusion;and acute cardiogenic pulmonary edema.

Another method of promoting airway clearanceis use of positive expiratory pressure (PEP) breathing.PEP creates a backpressure to stent the airways openduring exhalation and promotes collateral ventilation,allowing pressure distal to the obstruction toaccumulate. This method of clearing the airwaysprevents their collapse, easing mobilization ofsecretions from the peripheral airways toward thecentral airways. Application of PEP breathing is basedon the hypothesis that mucus in peripheral smallairways is mobilized more effectively by coughingor forced expirations if the alveolar pressure andvolume behind mucus plugs are increased. A maskor mouthpiece apparatus provides a controlledresistance (10 to 20 cm H2O) to exhalation andrequires a slightly active expiration; tidal volumeinspiration is unimpeded. PEP increases the pressuregradient between open and closed alveoli, thustending to maintain patency in alveoli. It alsoincreases functional residual capacity (FRC), thusreducing the resistance in collateral and smallairways. As a voluntary maneuver, PEP requires


children to cooperate; therefore, it is difficult to applyin very small infants. Nonetheless, exhalation canbe coordinated through PEP when infants are crying.

The Flutter® is a device which provides a formof intermittent PEP. The patient exhales through asmall pipe, which provides PEP, oscillation of theairways (at 6 to 20 Hz) and accelerated expiratoryflow rates to loosen secretions and move themtowards the central airways. Mobilization of mucusis thought to occur via widening of the airways byincreased expiratory pressure and air flow oscillationsgenerated by the oscillation ball located inside thedevice. However, carefully controlled studies needto be conducted before this technique can be widelyrecommended. The RC Cornet® is a device basedon the same physiological principles as the Flutter®.It features a curved plastic tube containing a flexible,latex-free valve hose. During expiration through theRC Cornet®, a PEP and oscillatory vibration of theair within the airways are generated. Since itfunctions independently of gravity, it can be usedin any position. The flow, pressure and frequency ofthe oscillations can be adjusted to the patient’sneeds. As with the Flutter®, secretions mobilizedto the central airways are cleared by coughing orhuffing.

As demonstrated in clinical practice, these threetechniques have major effects on sputumviscoelasticity. Viscoelasticity has been reported tobe significantly lower in patients treated with thesedevices compared to standard treatment. Thisreduction has been hypothesize to improvemucociliary and cough clearability; however,interestingly, there have not been any publishedstudies that compare the differences in the quantityof expectorated sputum during treatment with eachof these devices.

CONCLUSIONMost physiotherapy interventions for ventilated

children take place in the hospital, above all, in thepediatric intensive care unit (PICU). NIV is criticalto the management of respiratory insufficiency inPICU patients, and its success depends on theestablishment of a good respiratory physiotherapyprogram.

Many PICU patients need to be intubatedbecause of acute illness. A good extubation protocol

depends on the total achievement of the goalsdescribed in this chapter. After extubation, airwaysecretions continue to be removed by somecombination of physiotherapy techniques for a fewdays until the airways are determined clear byauscultation. The family must continue to providemost airway secretion management, and therefore,must be trained in managing MI-E, respiratoryresuscitation—including manual resuscitation—and respiratory support. The family needs thisknowledge and these skills to avoid futurehospitalizations and to resuscitate childrensusceptible to bradycardias or sudden respiratoryarrests associated with mucus plugging or fatigue.

Transmitted sounds from the throat andpharyngeal secretions are not a reason to initiatechest physiotherapy; nocturnal n-BiPAP and diurnalinsufflations should be continued, and the patientcan be discharged whether or not they can breatheautonomously. Most patients with good secretionclearance, and lung expansion associated with goodalveolar ventilation, can wean from full-timeventilator use to nocturnal-use at 1 or 2 weeks post-discharge.

In conclusion NIV is a very efficient techniquefor the respiratory management of children;however in most cases, secretions are excessiveand NIV alone is likely to fail. Respiratoryphysiotherapy is crucial for these patients. It shouldbe based on the intervention goals described inthis chapter to enable efficient treatment while thechild is hospitalized, and prevent hospitalizationswhile the child is at home, where family membersmust be trained to meet said goals, and therefore,maximize the child’s potential for pulmonaryrehabilitation.

REFERENCES1. Jardin E, O´Toole M, Paton JY, Wallis C. Current status of long-

term ventilation of children in the United Kingdom: Question-naire survey. Br Med J 1999;318:295-8.

2. Midgren B, Olofson J, Harlid R, Dellborg C, Jacobsen E, Norre-gaard O. Home mechanical ventilation in Sweden, with refe-rence to Danish experiences. Respir Med 2000; 94:135-8.

3. Niranjan V, Bach JR. Spinal muscular atrophy type I: a non-inva-sive respiratory management approach. Chest 2000;117:1100-5.

4. Frownfelter D, Dean E. Principles and practice of cardiopulmo-nary physical therapy. 3rd ed. Mosby-Year Book, Inc, 1996.

154 M.R. Gonçalves

5. Bach JR, Baird JS, Plosky D, Nevado J, Weaver B. Spinal muscu-lar atrophy type I: management and outcomes. Pediatr Pulmo-nol 2002;34:16-22.

6. Bach, JR. Management of patients with neuromuscular disea-se: Hanley & Belfus, Inc, 2003.

7. Niranjan V, Bach JR. Noninvasive management of pediatric neu-romuscular ventilatory failure. Crit Care Med 1998;26:2061-5.

8. Norregaard O. Noninvasive ventilation in children. Eur Respir J2002;20:1332-42.

9. Bach, JR. Noninvasive Mechanical Ventilation. Hanley & Belfus,Inc, 2002.

10. Fauroux B, Boule M, Lofaso F, Zerah F, Clement A, Harf A, et al.Chest physiotherapy in cystic fibrosis: Improved tolerance withnasal pressure support ventilation. Pediatrics 1999;103:E32.

11. Bach JR, Saporito LR. Criteria for extubation and tracheostomytube removal for patients with ventilatory failure: a differentapproach to weaning. Chest 1996;110:1566-71.

12. Evans JN, Jaeger MJ. Mechanical aspects of coughing. Pneumo-logie 1975;152:253-7.

13. Bach JR, Ishikawa Y, Kim H. Prevention of pulmonary morbidityfor patients with Duchenne muscular dystrophy. Chest1997;112:1024-8.

14. Bach JR. Pulmonary rehabilitation in neuromuscular disorders.Seminars in respiratory medicine, 1993;14:15-29.

15. Ross J, Dean E, Abbound RT. The effects of postural drainagepositioning on ventilatory homogeneity in healthy subjects. PhysTher 1992;72:794-9.

16. Webber BA, Pryor JA. Pysiotherapy skills: Techniques andadjuncts. In Webber BA, Pryor JA (eds). Physiotherapy for Res-piratory and Cardiac Problems. London, Churchill Livingstone1993;113.

17. Hubert J. Mobilisations du Thorax Belgium. Les edicions Medi-cales et Paramedicales de Charleroi, Montignies-sur-Sambre1989.


INTRODUCTIONPediatric intensive care units (PICUs) should be

equipped for non-invasive ventilation (NIV). Thischapter provides an overview of the human andmaterial resources required for administration of NIVand for training and research in this technique. Theseresources are the same as those employed forconventional mechanical ventilation (CMV). Sharingof clinical and research experiences among healthprofessionals from the field enables continuouslearning.

MATERIAL RESOURCES

Hospital infrastructureNIV in acute pediatric patients can be initiated

in any hospital ward (e.g. ER, pediatric ward orPICU).The location depends on the specificcharacteristics of each hospital. In the majority ofSpanish hospitals, NIV is started in the PICU. It canalso be administered in extra-hospital environmentssuch as the home or during transport.

The space required for administering NIV issimilar to that of any PICU bed. The mainrequirements are agreeable surroundings for thepatient, sufficient space for the required equipment(e.g. ventilator, monitor, and IV pumps), andstandard gas outlets (three for oxygen, one for air,and two for vacuum) and utilities (e.g. electrical

outlets with surge protectors, lighting, airconditioning, and heating). Furthermore, since NIVhas proven utile for immunologically compromisedpatients, a room equipped for isolating a patient inthis condition may also be valuable.

Equipment1. Monitoring of physiological parameters: clinical

evaluation of ventilated patients requires that theirrespiratory rate and oxygen saturation becontinuously monitored. If possible, transcutaneousCO2 should also be monitored. Electrocardiogramsand blood pressure measurements are onlyrequired for patients at risk for hemodynamiccomplications.

2. NIV-specific material:a. Ventilators: Ideally, an NIV-specific ventilator

would be used. A home ventilator withoutoxygen blender can be used to initiatetreatment. However, any ventilator for CMVthat is equipped with an NIV module can beemployed. In certain cases CMV ventilatorscan be used as long as the problems inherentto their use in NIV are resolved (see Chapter13).

b.Tubing: The tubing used depends on theventilator. If a non-vented interface is used,then the tubing should include an exhalationor leak control valve or, for ventilators lackingan expiratory loop, a Plateau valve. If the

Healthcare assistance, quality controland clinical training for non-invasive

ventilation of acute pediatric patients

T. Gili, A. Medina and M. Pons

Chapter 21

ventilator used does not include pressurizedoxygen or lacks an oxygen blender, then thetubing should also be equipped with anoxygen connection.

c. Interfaces: A wide variety of interfaces (e.g.nasal, oral-nasal or face masks, and nasaltubes) should be available to enable optimalfitting for each patient. It is important toorganize these according to type and size.Many interfaces are designed for only one use;however, most hospitals tend to clean andreuse them. If this is the case, then theinterfaces must be carefully stored with theirrespective harnesses (since mismatching canlead to poor attachment) and inventoried tofacilitate any required replacement. One usefulclassification system is to separate theinterfaces by age group (neonates, infants,toddlers, children, adolescents and adults) andthen by model (type, manufacturer and size),which enables rapid location of any interfacein stock.

d.Active humidifiers: Any simple humidifier canbe connected to the respiratory circuit throughthe tubing.

e. Other accessories: Oxygen connections,Plateau valves, filters, jet or ultrasonicnebulizers, aspirator for secretions, assistedcough device (CoughAssist®), etc.

3. Manual resuscitation bag connected to anoxygen source and kept at the patient’s bedside.

4. Material for endotracheal intubation and CMVin the event that NIV fails.

HUMAN RESOURCESNIV treatment of children in the PICU should be

performed by all medical personnel: doctors,respiratory therapists, nurses and auxiliary staff.Furthermore, multidisciplinary healthcare teamsshould be created to evaluate and care for theventilated child and to ensure proper monitoring ofthe child after they have been discharged from thehospital, especially those who are to receive homemechanical ventilation (HMV). Said team shouldinclude intensive care physicians, pulmonaryspecialists, physical therapists and nutritionists.

The healthcare team will decide when the patientis admitted to and discharged from the PICU,

determine the type of ventilation to be used and thematerial required, program the respirator, help setup the material, and be in charge of clinicalmonitoring of the patient.

The team should also develop NIV protocols andevaluate the treatment results. It may considerdesignating one doctor and one nurse from the wardto share the responsibilities of maintaining ventilationequipment in optimal order, updating the treatmentprotocols, supervising team member training, andkeeping close records, upon which QC will based.These two professionals will be in charge ofcoordinating training, teaching, treatment evaluationand research in NIV. Another valuable option,especially for small PICUs with low patient volume,is periodic rotations of doctors or nurses in otherhospitals. This ensures good NIV training and enablessharing of experiences among professionals, whichis extremely useful in the medical profession.

The team nurses or respiratory therapists will beresponsible for continuous care of the patient. Theywill also be required to learn the treatment protocolsand how to operate the equipment, help set up thematerials, become familiar with aspects of thetreatment that may cause failure during initiation ofNIV, and be able to quickly and independently solveany problems that arise. It is crucial that nursingsupervisor(s) collaborate with the nurses in maintainingNIV materials, conducting training activities, andfomenting teamwork among the medical staff.

Auxiliary personnel should collaborate closelywith the nurses in patient care.

Just as in CMV, safe delivery of NIV requires thatwell-trained hospital staff be present 24 hours a day,365 days a year. NIV training should encompassindications, contra-indications and disadvantagesof the technique, as well as early recognition of signsthat the patient is clinically deteriorating or that thetreatment is failing, in order to prevent any delaysin endotracheal intubation and subsequent initiationof CMV.

It is well known that the first 6 to 8 hours of NIVimply an additional workload for hospital personnel,especially nurses or respiratory therapists. In themajority of Spanish hospitals, NIV is initiated byintensive care staff (whether in the ICU or PICU); thiswork requires a high level of personal motivation andforce. Therefore, hospital wards that use NIV shouldanticipate problems in this time frame and take

158 T. Gili, A. Medina, M. Pons

adequate measures to resolve any problems. Ifpreventative measures are not taken, then there isa strong chance that the treatment will fail andhospital staff will lose their motivation, andconsequently, a highly useful ventilatory supportoption for many patients could be lost.

As with any technique in the PICU, NIV demandsthat all personnel be capable of delivering thistechnique at any time; the importance of adequatetraining can not be reiterated enough.

QUALITY CONTROLThe majority of published studies report an

efficacy rate of nearly 75% for NIV in intensivepediatric patients. However, this rate can dropsubstantially in function of the patient's age andpathology; this is especially apparent in infantsyounger than 12 months and in patients with TypeI acute respiratory failure (ARF).

Evaluating the efficacy of NIV involves more thansimply counting the number of patients for whomintubation has been avoided and confirming thatthe morbidity and mortality are not higher inintubated patients that had first received NIVcompared to those that did not: secondary variablesmust also be measured. These include the averagelength of hospital stay for patients that have receivedNIV alone compared to that of clinically similarpatients (i.e. by factors including age, type and stageof pathology) that had first been treated with CMV,as well as the frequency of nosocomial infections.

Although the aforementioned variables can beused to compare NIV against the gold standard ofCMV, the best way to evaluate NIV is by assessingthe quality of the treatment itself and how it hasprogressed within a given ward.

A record of the efficacy of NIV is undoubtedly thebest form of quality control for evaluating thistechnique within a PICU, although a record ofcomplications can also be useful. Despite the limitedbody of literature on quality control in NIV, the authorsof this chapter have proposed the following fourquality indices for NIV of critical pediatric patients:

Initial failure index (IFI)The IFI is the percentage of patients for whom

NIV fails within the first hour. It is calculated bydividing the number of these patients by the total

number of patients treated with NIV, and thenmultiplying the result by 100%. The authors of thischapter consider that the IFI should be < 10%(ideally < 5%) after the first year doing NIV.

The authors believe that failure of NIV withinthe first hour is typically due to inappropriateindication, an inadequate interface or ventilator,or a poorly adjusted interface—all of which areassociated with inexperienced or untrainedpersonnel—and therefore, think that once a wardadopts NIV, it should closely monitor its IFI. As theward gradually acquires experience and confidencewith NIV, it can begin to treat evermore complexpatients or patients that are borderline contra-indicated (and consequently, whose condition cantemporarily worsen). Hence, the IFI can serve asa red flag in reviewing individual cases, evaluatingpatient selection protocols, and preventingexcessively optimistic use of NIV, which can bepernicious for patients.

Early failure index (EFI)The majority of NIV failures occur between the

first and twelfth hours. The causes of failure includerapid progression of the disease, inadequateetiological treatment of the cause of ARF, patient-ventilator desynchronization (due to multiplefactors), and non-compensated hypoxemia (e.g.due to slow recruitment or to use of a ventilatorlacking an oxygen blender).

The EFI is the percentage of patients for whomNIV fails between the first and twelfth hours oftreatment. It is calculated analogously to the IFI.Its maximum value should be 10 to 15%, which isthe overall intubation rate reported in the majorityof publications. Although these articles typicallydo not mention whether intubation occurred beforeor after the first hour of NIV, the authors of thischapter have determined that experienced wardshave an IFI < 5% (see Table I).

The EFI serves to track the progress of NIV inwards that have already learned the technique andhave an acceptable IFI. It can also be useful forevaluating the efficacy of NIV in hypoxemic patientswith a PaO2/FiO2 of nearly 200 that are treated withmore than two hours of NIV despite having clinicalfactors consistent with a poor prognosis (e.g. age< 12 months or no decrease in respiratory rate afterthe first hour of NIV).

159Healthcare assistance, quality control and clinical training for non-invasive ventilation of acute pediatric patients

Late failure index (LFI)Many NIV failures also occur after the twelfth

hour and are often due to the same reasons thatcause failure between the first and twelfth hours,such as inadequate humidification and progressionof the disease (e.g. appearance of new atelectasesin bronchiolitis).

The LFI is the percentage of patients for whomNIV fails after the twelfth hour. It is also calculatedanalogously to the IFI. The LFI should beapproximately 5-15%; hence, the EFI and the LFIshould sum to approximately 20%, which is the

overall intubation rate (see above and Table I). Thecut-off between EFI and LFI could be set at 18 or24 hours instead of 12 hours; this is open to debateand will ultimately depend on analysis of data fromdifferent wards.

Delayed failure index (DFI)Some patients are at risk for delayed failure of

NIV, namely, those given NIV either as rescuetreatment for failed extubations during weaning oras elective therapy for weaning from CMV andavoiding tracheostomy. Albeit no data on this patient


Table I. Quality control indices for non-invasive ventilation (NIV)

Center Year IFI EFI LFI # NIV Patients

Bilbao (Cruces) 2004 0% 50% 0% 42005 0% 6% 6% 162006 5% 0% 5% 192007 5% 0% 15% 25

Oviedo 2004 0% 13.63% 4.54% 222005 3.12 3.12 15.62 322006 6.66% 6.66% 2.22% 452007 1.85% 5.55% 11.11% 54

Lisboa (F Fonseca) 2005 0% 14% 7% 142006 0% 12% 5% 422007 4% 12% 16% 25

Salamanca 2004 0% 23% 0% 132005 0% 9% 18% 222006 7.7% 15.4% 7.7% 262007 0% 10.5% 0% 19

Castelló 2005 0% 0% 50% 32006 10% 15% 10% 192007 0% 0% 25% 8

Albacete 2005 0% 0% 50% 42006 0% 7.7% 7.7% 132007 0% 12% 0% 16

Sabadell 2004 14.2% 14.2% 0% 82005 0% 0% 0% 32006 11% 0% 0% 92007 0% 0% 14% 14

Barcelona (HSJD) 2002 6.6% 6.6% 6.6% 152003 0% 11% 3.7% 272004 0% 14.7% 11.7% 342005 0% 9.8% 15.7% 51

IFI: initial failure index; EFI: early failure index; LFI: late failure index; NIV: non-invasive ventilation; HSJD: Hospital Sant Joande Deu

group have been published, the delayed failure index(DFI) may be > 50%. Based on the authors’experience, DFI can range from 75% in the generalpopulation to 100% in post-operative cardiacpatients.

The duration of NIV in these patients is highlyvariable, and treatment may remain relativelyunstable for several days; hence, the actual numberof hours is not very important in determining delayedfailure. For many of these patients NIV implies anextended stay in the PICU until they can be treateddefinitively with CMV.

The authors believe that a DFI of < 75% ispossible in the subgroup of patients for whom NIVas rescue treatment is used as last resort.

The authors present here data from eightintensive care units (seven Spanish and onePortuguese) with varying levels of expertise in NIVthat support the use of these indices in qualitycontrol of NIV. Nonetheless, these indices should bevalidated in the near future.

Lastly, a record of NIV complications is aninvaluable tool for establishing preventative strategiesand improving treatment, as exemplified in the dropin cases of skin necrosis due to pressure from NIVinterfaces.

ACADEMIC ACTIVITIES

Teaching and trainingVentilatory assistance for acute pediatric patients

is a complex discipline which requires a firm base inpediatric medicine, a strong domain of treatmenttechnology, and the skills to perform the diagnosticand therapeutic procedures indicated for respiratoryfailure patients. This is especially true in the caseof relatively recent treatment strategies such as NIV.

Hence, healthcare professionals seeking to learn NIVor update their knowledge on this technique requirewell organized training programs. Those requiringtraining include intensive care physicians, pulmonaryspecialists, physical therapists, pediatrics residents,and nurses. Once trained, these professionals canteach other personnel in their ward through specifictraining courses and help develop treatmentstrategies by sharing their NIV experiences with theirpeers.

Hospital wards should develop in-houseacademic programs for NIV similar to those used forother techniques. These would include clinicalmeetings with presentation of clinical cases, literaturereviews, and updating of protocols and equipment.Multidisciplinary meetings are especially valuable.These should encompass intensive care physicians,pulmonary specialists, physical therapists, radiologistsand any other specialists that have contributed tothe patients’ care.

Proper training will expand the range ofindications of NIV, increase the rate of treatmentsuccess and improve early application of thistechnique, while reducing the frequency of majorcomplications and decreasing the workload onhospital staff that must start and maintain NIVtreatment.

ResearchContinued research on NIV should provide

valuable scientific results that will enable new orimproved practical applications (see Table II), andtherefore, should be stimulated. However, there isa need for common protocols for multicenter studiesand clinical trials, which would be of great interestto professionals in the field. Creation of theseprotocols could be facilitated by assembling national-level work groups with international connections.

161Healthcare assistance, quality control and clinical training for non-invasive ventilation of acute pediatric patients

Table II. Future areas of NIV research.

• The immediate effects of NIV in various pathologies.

• The mid-term and long-term effects of NIV in various pathologies.

• NIV safety.

• Technical variables (interfaces and triggers).

• Structured training for healthcare professionals.

• Impact of NIV on development (namely, pulmonary function).

• Guidelines for diagnosis, treatment, and patient care in NIV.

In these groups professionals could share theirexperiences, and then use them as the basis forprotocols. Therefore, each ward should maintainrecords on its NIV treatment for use in research aswell as for quality control. These records shouldbe simple, practical and well-supported by IT.

The results from NIV research should becommunicated through the appropriate channels(e.g. journals, conferences and training courses),not only those related to intensive care, but alsothose dealing with pediatric specializations and withadult medicine.

It is crucial that pediatric ARF patients receiveearly non-invasive ventilatory support according tocriteria distinct from those used for intubation. Thisrequires that healthcare professionals be familiarizedwith all of the available resources and latest criteriafor NIV treatment of ARF.

REFERENCES1. Tratado de Cuidados Intensivos Pediátricos. 3ª ed. Francisco Ruza

Tarrio y colaboradores. Ediciones Norma-Capitel 2003.

2. Norregaard O. Noninvasive ventilation in children. Series “Nonin-vasive ventilation in acute and chronic respiratory failure”. Euro-pean Respiratory Journal 2002;20:1332-42.

3. Brochard L. Mechanical ventilation: invasive versus non-invasi-ve. European Respiratory Journal 2003;547:31-7.

4. Díaz Lobato S, Mayoralas Alises S. Ventilación no invasiva. Archi-vos de Bronconeumología 2003;39:566-79.

5. Torres A, Ferrer M. Grupo de trabajo de Cuidados RespiratoriosIntermedios de la Sociedad Española de Neumología y CirugíaTorácica. Unidad de Cuidados Respiratorios Intermedios. Defi-nición y características. Archivos de Bronconeumología2005;41:505-12.

6. Essouri S MD, Chevret L MD, Durand P MD, Haas V MD, Fau-roux B MD, PhD, Devictor D MD. Noninvasive positive pressu-re ventilation: Five years of experience in a pediatric intensivecare unit. Pediatr Crit Care Med 2006; 7:329-33.

7. Medina A, Prieto S, Los Arcos M, Rey C, Concha A, MenéndezS, et al. Aplicación de ventilación no invasiva en una unidadde cuidados intensivos pediátricos. An Esp Pediatr 2005;62:13-9.

8. Fortenberry JD. Noninvasive ventilation in children with respi-ratory failure. Crit Care Med 1998;26:2095-6.



INTRODUCTIONThe aim of treating a sick child is to cure them,

or, if this is not possible, than at least to improvetheir condition. Medical technology for treatingdiseases and prolonging patients’ lives is continuallyprogressing. Indeed, the prognoses for manypatients, especially those with chronic conditions,are constantly improving as therapeutic techniquesbecome more advanced. However, it is now moreimportant than ever to carefully assess a newtechnique before adopting it, establishing itstherapeutic boundaries and evaluating its suitabilityfor each patient so that unnecessary or ineffectivetreatments can be avoided.

In most treatment scenarios, healthcareprofessionals have developed a clear strategy whichhas been accepted by the child’s parents. In thesesituations, the benefits clearly outweigh the risks.

Starting non-invasive ventilation (NIV) is avaluable option for children with diverse pathologies,primarily acute ones; it is an effective treatment thatobviates the need for more aggressive forms ofventilation assistance. However, in certainpathologies, especially chronic ones, NIV can rangefrom highly effective to futile: for some patients, itmay improve the quality of their life and that of theirfamily, whereas for others, it may simply extend thetime during which they remain in a compromisedstate, diminishing their quality of life and that oftheir family, without offering any hope of curingtheir pathology.

Defining the point at which a child’s quality oflife becomes unacceptable due to intellectual or

physical limitations can be extremely difficult.Healthcare professionals must carefully inform thepatient and their family of all treatment options withtheir respective benefits and risks. Only the patient—or their family, in the case of very young or otherwiseincapable children—can then decide what they arewilling to sacrifice or endure in the name of medicaltreatment.

Another important and potentially controversialethical issue at the social justice level is the economicand social costs that starting and maintaining certaintherapies or techniques places on the publichealthcare system. This is especially relevant to theSpanish system, which is being stretched thinnerand thinner as it endeavors to continue offering highquality health services. Social justice dictates that allpatients should have equal access to these servicesand that if resources are limited, then the systemmust give priority to those patients for whom a giventreatment will prove most effective.

INDICATIONS OF NON-INVASIVE VENTILATIONThere is strong evidence that the drop in

mortality rates of pediatric patients is partly due toimprovements in neonatal and pediatric intensivecare. However, technological advances have broughtincreased morbidity with severe sequelae thatsometimes imply major intellectual, cognitive andphysical (especially respiratory) limitations amongsurvivors. Furthermore, modern techniques enabletherapeutic options that can prolong the lives ofchildren with various pathologies but that place

Ethical aspects of non-invasive ventilation

F.J. Cambra, M.H. Estêvão and A. Rodríguez Núñez

Chapter 22

these children and their families in overwhelminglystressful situations.

There are two ethical principles from goodclinical practice which are particularly relevant toNIV: beneficence (doing everything possible for thepatient’s well-being) and non-maleficence (avoidingharm to the patient). There may be cases in whichventilatory support keeps the patient alive at thecost of a severe loss in quality of life which affectsnot only them, but their family and society as well.NIV may be an excellent treatment option for somepatients, especially when used only for part of theday, whereas for others, it may put heavy demandson them and their families.

NIV is often proposed for patients who can notbe weaned off of conventional mechanical ventilation(CMV). In these scenarios, the treatment optionscomprise extubating the child and allowing theirpathology to run its natural course, or transitioningthe child to NIV. In the latter case, a stipulation canbe placed whereby the patient will not be returnedto CMV should the treatment prove ineffective. NIVmay be employed as elective therapy for patientswhose pathology and its natural progression areknown.

NIV is indicated for various pathologies, andits therapeutic scope has broadened as healthcareprofessionals gain experience with this techniqueand its related equipment (see Chapters 3 and 21).The benefits of NIV are most apparent for diseasesin which respiratory failure is temporary or relativelystable.

For certain conditions (e.g. laryngeomalacia,tracheobronchomalacia, bronchopulmonary dysplasia[BPD], and certain facial malformations) ventilatoryassistance enables decreasing or even eliminatingwork of breathing, which in turn allows for bettergrowth and development of the airways, whosestructures can then gradually mature. NIV is especiallybeneficial for patients with facial deformations thatare treated by surgery: it can be used until theirairways are sufficiently developed for undergoingsurgery. In many of these cases NIV can be stoppedafter a given period of time.

NIV provides clear clinical benefits for patientswhose respiratory failure is stable, such as those withscoliosis.

The most challenging questions regardingindication of NIV typically arise over patients whose

condition is very severe or progressively deteriorating.This is most clearly illustrated in neurological orneuromuscular diseases, especially Type I spinalmuscular atrophy (SMA).

Type I SMA is an extremely severe progressiveneuromuscular disease (NMD) which involves a neartotal loss of voluntary muscle activity, but whichdoes not affect upper cerebral activities (e.g. sensoryorgan or intellectual activities). Patients with thisdisease remain incapacitated and require ventilatoryassistance to survive. Explaining the prognosis of anSMA patient to their family can prove challenging:these patients suffer from muscular debilitation sograve that they can not even maintain effectivebreathing, yet their cognitive function remainsnormal. As reflected in the literature, there is nogeneral consensus on the best method of ventilatoryassistance for infants with SMA, which in turninfluences parents' decisions as well as the availabilityof human and material healthcare resources. Authorsthat are against the use of NIV for these patientsclaim that it is futile and only serves to lengthen thechild’s suffering, whereas those in favor affirm thatit improves the child’s quality of life and amelioratesthe family’s sense of powerlessness to stop thedisease.

NIV has recently been employed to treat cysticfibrosis patients. Initially its use was limited totemporary ventilation support during lungtransplants, but it is now applied early on in cysticfibrosis patients and has demonstrated good results.

NIV can also prove valuable for cancer patientssuffering from intercurrent respiratory compromisewhich, if treated by CMV, would lead to a very poorprognosis. Indeed, NIV is an ideal option for thesepatients since it improves their clinical state whilelimiting lung damage. Alternatively, NIV can maketerminal patients more comfortable during end-of-life care.

When assessing indication of NIV, healthcareprofessionals may consider experimental use of thistechnique. In these cases, they must take intoaccount all of the ethical stipulations that dictateexperiments involving humans—specifically, pediatricpatients—and, in the absence of other options,consider using NIV as compassionate treatment.

For very severely ill or progressively deterioratingpatients the only alternative to NIV is CMV viatracheotomy. This is an invasive method whereby

164 F.J. Cambra, M.H. Estêvão, A. Rodríguez Núñez

access to the airways is created surgically via thetrachea—which carries inherent risks—and whichimplies a greater probability of tracheopulmonaryinfections, requires more frequent aspiration ofsecretions, limits the patient’s oral communicationand feeding, and may demand greater dependenceon mechanical ventilation.

Homecare is easier for children treated with NIVthan for those treated with CMV. This obviatesprolonged hospital stays (often in intensive care)and their negative consequences on the child’spsychological and emotional development.

When assessing the suitability of an NIVtreatment program for a given child, healthcareprofessionals should consider the following generalobjectives:1. Prolong their patient’s life.2. Optimize cardiopulmonary function.3. Reduce morbidity.4. Improve the quality of life of the patient and

their family.5. Maintain age-appropriate growth and

development.6. Optimize the cost to benefits ratio of the medical

care.

DECISION MAKINGIn medical decision-making, the principle of

autonomy dictates that capable adult patients mustconsent to or decide on their care. A doctor mustalways follow the choice of their patient, regardlessof what their own decision would be.

When discussing treatment options with thirdparties for patients who are not capable of expressingtheir own opinion—as occurs in pediatric medicine—the main challenge is to determine the best interestsof these individuals. Furthermore, decision makingin pediatrics can be very complex for patients that,despite legally being minors, have some degree ofautonomy because of their age or level of maturity.The parents or caregivers of pediatric patients areresponsible for deciding on medical intervention.Their authority, which is legally and ethicallyindisputable, is based on trust and expressed onbehalf of the (incapable) patient, and is presumedto reflect the will and best interests of the patient.Parents are granted this authority precisely becausethey want the best for their children.

Pediatric patients do not have any recognizedauthority over their health until they reach legal adultage. However, if the acquisition of autonomy isviewed as a gradual process that occurs over severalyears, then in terms of decision making, an infantcan not be equated with a 7 year old, nor can a childof 7 be equated with an adolescent of 14. From alegal perspective minors should not all be groupedtogether as being incapable of making choicesconcerning their health: each patient should betreated on an individual basis and be granteddecision-making power in function of their level ofmaturity. However, as a general rule, children under12 can be considered as incapable of makingdecisions, and those over 14, as capable. It is hardto establish a general rule for patients between 12and 14; each patient should be carefully evaluatedto determine their level of maturity. Apart from ageand maturity, the consequences of the procedureto be realized must also be taken into account, suchthat a higher level of maturity is demanded forsituations with greater risks (e.g. initiating CMV).

The patient (when capable), their family, andthe professionals responsible for their healthcaremust assess the advantages and disadvantages ofthe procedure together.

It is important to highlight the potential forconflicting opinions between parent and child. Forcapable children these situations can be especiallydifficult if some type of compromise is not reached;indeed, they may require the intervention of thehospital’s ethics committee or the Department ofJustice. Regardless of their age or capacity fordecision making, the child’s dignity must always berespected, and they must always be informed oftheir condition in an age-appropriate fashion.

From the very beginning of treatment, doctorsact based on the principle of beneficence, applyingtheir knowledge to improve the child's wellbeing.The only possible restrictions on their involvementcould be the wishes of the child themselves (whencapable) or of the parents or caregivers (for non-capable children). Furthermore, during conflicts ofopinion between the doctor and patient, there arelimits to autonomy: the priority is always the bestinterest of the child.

The concept of quality of life is difficult to definefor children, and doctors, patients (when capableof expressing themselves) and parents do not always

165Ethical aspects of non-invasive ventilation

share the same definition. Surveys on quality of lifefor children treated with ventilatory support typicallyrefer to CMV, as information on NIV remains scarce.

Doctors may have a more distanced perspective,implying less personal feelings, whereas familiesoften can not remove the emotional element fromtheir decision making. Social, cultural and religiousfactors also influence families’ treatment choices. Animportant aspect to consider when valuating thefamily’s opinion is their need to feel that they havedone everything possible to maintain their child inthe best conditions possible. In the event of an earlydeath, healthcare professionals must ensure that thefamily does not feel responsible, even if they hadunderstood the gravity of the child’s prognosis.

Life is one of the most precious things on Earth,but it is not the only one; it is not absolute. Thequality of a life prolonged by ventilatory supportmust be deemed sufficiently good to justify thetreatment. This demands that the doctor have thedeepest knowledge possible of the pathologyaffecting the child and of the natural course of the

disease. Doctors must remain up to date on currenttherapies, since medical decisions can vary infunction of experience and available resources.

Usually there is no ideal treatment option;different decisions may be equally correct from amoral perspective. Each case is different and eachfamily is unique. Health professionals must recognizethe rights of the parents to define what is best forthe child, and, as long the principle of maleficenceis not violated, must accept the parents' decisioneven if they do not agree with it. Furthermore, theymust accompany the parents during the decisionmaking process and comfort them over theirdecision, regardless of whether the family has optedfor a particular treatment or decided against it.

Despite the difficulty inherent in defining qualityof life, NIV can offer certain clear benefits thatfacilitate decision making: improved nocturnal sleepwith a consequent increase in diurnal activity;decreased rate of respiratory infections; fewer hospitaladmissions; and greater physical autonomy (e.g.eating and speaking).


Table I. Bioethical evaluation of non-invasive ventilation according to the principles of Beauchamp and Childress

Principle Role of non-invasive ventilation (NIV)

Beneficence Improves the clinical condition of respiratory failure patients; Improves their comfort level and quality of life; Enables verbal expression and good interaction with surroundings; Shortens hospital and PICU stays.

Non-maleficence Has few side effects; Obviates intubation and tracheostomy; Has lower risk of ventilator-induced pulmonary lesions and of respiratory infections.

Autonomy A treatment option that children and/or their parents can choose once informed of its indications, advantages, disadvantages and alternatives.

Distributive justice Enables distribution of resources (i.e. invasive methods are dedicated to most severe patients); Easy to use in home care, which leads to savings in the hospital resources.

Table II. Questions to ask when determining the suitability of NIV for a given patient

Question Comments

Can the patient benefit from NIV? Valuation of the technique (principle of beneficence)How can harm to the patient be avoided during NIV? Risks inherent to the technique (principle of non-maleficence)Has the patient and/or family participated in Respect the patient's autonomy. The child and their family decision-making on the type of ventilation to be used? should be well-informed.What quality of life is expected for the patient before, It is only ethically acceptable to ventilate the patient if theduring and after treatment? treatment is expected to be beneficial according to objective

and subjective terms that fall under the concept of quality of life.

Accurate information is critical to decisionmaking. It should be clearly and sincerelycommunicated to all parties involved (including thechild, if capable), with all the treatment risks andbenefits carefully explained. Healthcare professionalsmust ensure that in every conversation with thepatient, family and/or caregivers, the information iswell understood; if needed, they must be willing toexpand on or repeat what they have said ifrequested.

THE ROLE OF SOCIETYNIV is dictated by a series of contextual factors

that, depending on the patient, can either facilitateor complicate its clinical application. Materialresources are limited; hence, the public healthadministration and healthcare institutions mustdevelop strategies to organize and distribute theseresources in function of their priorities. The bioethicsprinciple of justice stipulates that all individualsshould have equal access to health resources (sinceeach person deserves the same consideration andrespect), and the premise of distributive justicedeems that limited resources must be maximized.Hence, NIV should be provided to those patientsthat stand to benefit the most.

NIV can be considered a highly efficient resource,since its use for both chronic and acute patientsimplies major savings as compared to CMV: NIVequipment is cheaper, simpler and longer lasting,and NIV is relatively easy to administer in the home,which reduces workloads on hospital staff. Homemechanical ventilation (HMV) programs should beencouraged as much as possible for both NIV andCMV patients.

In terms of resource distribution, doctors mustevaluate each patient to determine which form ofmechanical ventilation would be most appropriate—and therefore, how resources will be used—according to fair criteria. They must also try andobtain all the materials that they deem necessaryfor their patients, endeavoring to convince thehospital administration of the various benefits ofNIV—namely, in terms of beneficence, non-maleficence and quality of life for the patient, as wellas cost-effectiveness as a health service. Doctors mustalways efficiently manage their available resources.For example, NIV systems should be used for as

many patients as possible. This in turn may requirethe development of specific treatment programs,training courses, technical protocols and clinicalguidelines.

REFERENCES1. Abel F, Cambra FJ. Neonatología y discapacidad severa en Diag-

nóstico prenatal, neonatología y discapacidad severa: proble-mas éticos. Instituto Borja de Bioética. Fundación MapfreMedicina. Ed. Mapfre 2001.

2. American Academy of Pediatrics. Ethics and the Care of CriticallyIll Infants and Children. Pediatrics 1996;98:149-52.

3. Bach JR, Niranjan V, Weaver B. Spinal muscular atrophy type I:noninvasive respiratory management approach. Chest 2000;117:1100-5.

4. Bertrand P, Fehlmann E, Lizama M, Holmgren N, Silva M, Sán-chez I. Asistencia ventilatoria a domicilio en niños chilenos: 12años de experiencia. Arch Bronconeumol 2006;42:165-70.

5. Branthwaite MA. Ethical and medico-legal aspects of assistedventilation. En: Simonds AK (ed). Non-Invasive ventilatory sup-port. London: Arnold, 1991;282-291.

6. Burns JP. Research in children. Crit Care Med 2003;31:S131-6.

7. Carnevale FA, Alexander E, Davis M, Rennick J, Troini R. Dailyliving with distress and enrichment: the moral experience offamilies with ventilator-assisted children at home. Pediatrics2006;117:e48-60.

8. Edwards EA, Hsiao K, Nixon GM. Pediatric home ventilatory sup-port: the Auckland experience. J Paediatr Child Health2005;41:652-8.

9. Goldberg AI. Health care networks for long-term mechanicalventilation. En: Hill NS. Long-term mechanical ventilation. Ed.Marcel Dekker, Inc. New York, 2001;411-29.

10. Gomez Rubí JA. Conflictos de justicia en el paciente crítico. En:Gómez Rubí JA. Etica en Medicina Crítica. Ed. Triacastela, Madrid,2002;205-24.

11. Eng D. Management guidelines for motor neurone diseasepatients on non-invasive ventilation at home. Palliat Med2006;20:69-79.

12. Hardart MKM, Burns JP, Truog RD. Respiratory Support in SpinalMuscular Atrophy Type I: A Survey of Physician Practices andAttitudes. Pediatrics 2002;110.

13. Jonsen AR, Siegler M, Winslade WJ. Clinical ethics. A practicalapproach to ethical decisions in clinical medicine. McGraw-Hill,New York, 2002.

14. Lumeng JC, Warschausky SA, Nelson VS, Augenstein K. The qua-lity of life of ventilator-assisted children. Pediatric Rehabilitation2001;4:21-7.

15. Marcus CL. Ventilator Management of Abnormal BreathingDuring Sleep. En: Loughlin GM, Carroll JL, Marcus CL (eds). Sleepand Breathing in Children. A developmental Approach. NewYork: Marcel Dekker AG, 2000;797-811.

16. Rodríguez Núñez A. Aspectos éticos de la ventilación mecánica.En: Grupo de Respiratorio de la Sociedad Española de CuidadosIntensivos Pediátricos. Manual de Ventilación Mecánica en Pedia-tría. Publimed. Madrid, 2003.

167Ethical aspects of non-invasive ventilation

17. Simón P, Barrio IM. La capacidad de los menores para tomardecisiones sanitarias: Un problema ético y jurídico. Rev Esp Pediatr1997;53:107-18.

18. Simonds AK. From intensive care unit to home discharge inthe 24h ventilator-dependent patient. En: Roussos C (ed) Mecha-nical Ventilation from Intensive Care to Home Care. EuropeanRespiratory Monograph 1998;3:364-79.

19. Street K, Henderson J. Are we treading a fine line? BMJ 2001;323:388-9.

20. Tejedor J.C, Crespo D, Niño E. Consentimiento y confidenciali-dad en medicina del niño y adolescente. Med Clin (Barc) 1998;111:105-11.

21. Zawistowski CA, Frader JE. Ethical problems in pediatric criticalcare: consent. Crit Care Med 2003;31:S407-10.


INTRODUCTIONThe demands that non-invasive ventilation (NIV)

places on hospital staff require that hospital wardsdevelop clear treatment protocols. This chapter aimsto establish general clinical guidelines, based on theevidence presented throughout this book, thatshould be adapted to each patient and each ward.Furthermore, the protocols and algorithms presentedhere should be frequently updated, as continuedinterest in NIV leads to greater technical expertiseand clinical experience among professionals in thefield.

INDICATIONS (CHAPTER 3)The indications of NIV in respiratory failure are

becoming increasingly clear and are based on thecriteria for pediatric acute respiratory failure (ARF)(see Table I). These can be extrapolated to establisha general classification for the different acute andchronic pathologies which can be treated with NIV(Table II).

From a practical perspective, when using NIVthe type of ARF should be classified according tocertain physiopathological criteria:

Type I ARF. Type I ARF is characterized byimbalance between ventilation and perfusion (V/Qmismatch) without alveolar hypoventilation. It firstappears as hypoxemia with no increase in arterialcarbon dioxide pressure (PaCO2). Patients mayexhibit condensations when examined by chest X-rays.

Type II ARF. The main feature of Type II ARF isalveolar hypoventilation. It is defined by the presence

of hypercapnia (PaCO2 > 45 mm Hg) with a normalalveolar-arterial oxygen gradient (A-a O2). Patientsmay present with hypoxemia due to hypoventilation.Chest X-rays do not reveal any condensations(excluding atelectases).

Following the classification above, the authorsof this chapter propose that pathologies should begrouped as shown in Table III (modified from Teague& Dobyns et al.). Acute bronchiolitis can be classifiedunder each type of ARF in function of its clinicalcharacteristics. It is important to remember thatatelectasis observed in chest X-rays is not necessarilyindicative of Type I ARF. Despite causing V/Qmismatch in the affected area, it may occur in TypeII ARF, specifically, as the consequence of pre-existinghypoventilation. Furthermore, atelectasis in Type IARF may be generated in adjacent segments bypneumonia.

CONTRA-INDICATIONS (CHAPTER 3) Early detection and pre-assessment of contra-

indications to NIV is crucial for avoiding erroneoustreatment of patients that should instead be treatedwith conventional mechanical ventilation (CMV). Thesecontra-indications are constantly updated; indeed,certain conditions which only a few years ago wereconsidered absolute are now considered relative (e.g.pneumothorax and recent gastric surgery). In fact,the British Thoracic Society guidelines accept the useof NIV for patients with contradictions as long asintubation is planned or the treatment is palliative.Nonetheless, there are certain absolute conditionswhich must be carefully evaluated (Table IV).

Summary and algorithms

A. Medina, M. Pons and F. Martinón-Torres

Chapter 23

CONDITIONS FOR NON-INVASIVE VENTILATION (CHAPTER 4)

NIV can be delivered nearly anywhere if aportable ventilator is used. However, a hospitalenvironment is preferred for treating acute patientsand for controlling changes in ventilationparameters in chronic patients, whereas extra-hospital treatment should be reserved for chronicpatients who have already begun NIV. Choosing

the ward in which NIV should be started dependson the patient, the available materials and thepersonnel (Table V).

MATERIALSPositive pressure NIV of pediatric patients

requires various materials that often differ from, orare used differently than, those for adult patients:

170 A. Medina, M. Pons, F. Martinón-Torres

Table I. Criteria for respiratory failure in children

Clinical criteria Physiological criteria

Weak cough, retention of respiratory secretions Vital capacity < 15 mL/kgIncreased use of accessory muscles Maximum inspiratory force < 20 cmH2OHigh respiratory frequency (based on age) PaCO2 > 45 mm Hg and pH < 7.35Paradoxical breathing PaO2/FiO2 < 300 mmHgIncompetent swallowing SatO2 < 97% for ambient airReduced activity level or diminished function

mL: milliliters; kg: kilograms; cm H2O: centimeters water; mm Hg: millimeters mercury; PaCO2: arterial carbon dioxide pres-sure; PaO2: arterial oxygen pressure; Sat O2: oxygen saturation; FiO2: inspired fraction of oxygen

Table II. Processes causing acute or chronic respiratory failure in pediatric patients for which non-invasive ventilation(NIV) is indicated

Acute respiratory failure (ARF)A. Decompensated central nervous system (CNS) diseases:

1. Apneas in pre-term/full-term infantsB. Decompensated respiration due to chest wall or spinal

abnormalities1. Obesity-hypoventilation syndrome

C. Upper airway diseases:1. Upper airway (e.g. laryngitis, and laryngotracheitis)

D. Lung diseases1. Asthma2. Bronchiolitis3. Pneumonia4. Atelectasis5. Acute pulmonary edema (APE)

E. Other situations1. Apneas after tonsillectomy2. Postoperative period after corrective surgery for

scoliosis3. Pulmonary complications from sickle cell disease4. Early extubation5. As ventilatory support during sedation procedures6. Severe respiratory failure in terminal illness (i.e.

palliative indication)

Chronic respiratory failure (CRF)A. Respiratory sleep disorders:

1. Obstructive sleep apnea syndrome (OSAS)2. Primary or acquired central alveolar hypoventilation

B. Neuromuscular diseases (NMDs) that affect respiratorymusculature1. Diseases of the 2nd motor neuron (e.g. spinal

muscular atrophy [SMA])2. B303. Diseases of or damage to the phrenic nerve4. Myasthenia gravis and other congenital myasthenic

syndromes5. Myopathies (e.g. congenital, mitochondrial, metabolic,

or inflammatory; and deposition diseases)6. Muscular dystrophies (MDs)

C. Upper airway diseases:1. Tracheomalacia

D. Respiratory diseases of the lower tract or parenchyma1. Bronchopulmonary dysplasia (BPD)2. Cystic fibrosis3. Bronchiectasis

171Summary and algorithms

Table III. Physiopathological classification of acute respiratory failure (ARF)

Type I ARF Type II ARF

ALI/acute respiratory distress syndrome (ARDS) Respiratory center:Neonatal distress syndrome • Drugs (opiates, barbiturates and anesthetics)Bronchoaspiration • Central congenital hypoventilation syndrome (CCHS)Acute bronchiolitis Upper motor neuron:Cardiogenic pulmonary edema • Spinal column traumaCystic fibrosis • Syringomyelia Pulmonary embolism (air, fat or blood) • Demyelinating diseasesInterstitial pulmonary disease • TumorsPost-obstructive pulmonary edema • Anterior horn neuronsRadiation • PoliomyelitisSepsis • Werdnig-Hoffmann syndromeSevere pneumonia (of bacterial, viral, fungal or Lower motor neuron:

parasitic origin) • Post-thoracotomy lesion of the phrenic nerveInhalation of poisons or toxic gases • Guillan-Barré syndromeMultiple transfusions Neuromuscular junction:Trauma (pulmonary contusion) • Botulism, multiple sclerosis (MS) and myasthenia gravis

• Antibiotics that block neuromuscular function• Organophosphorous compound poisoning• Tetanus

Pleura and chest wall:• Lesions from chest wall burns with retraction• Massive pleural effusion, and morbid obesity• Muscular dystrophy (MD) and pneumothorax

Increased airway resistance:• Obstruction of the larynx (e.g. laryngitis, diphtheria,

epiglottis, aspiration of foreign bodies, post-extubation edema, and paralysis of the vocal chords)

• Lower airway obstruction (e.g. emphysema, asthma, and acute bronchiolitis)

ALI: acute lesion injury; ARF: acute respiratory failure; ARDS: acute respiratory distress syndrome.

Table IV. Contra-indications to the use of non-invasive ventilation (absolute contra-indications are shown in bold)

1. Neurological:• Inability to protect airway: compromised bulbar

function, paralysis of the vocal chords, and alteredlevel of consciousness

• Severe psychomotor retardation

2. Craniofacial alterations:• Facial trauma or burns• Facial surgery

3. Gastrointestinal (GI):• Upper digestive tract surgery (esophageal or upper

GI)• Profuse vomiting• Active digestive hemorrhaging• Obstruction of the intestinal tract

4. Respiratory:• Severe respiratory failure (relative)• Undrained pneumothorax• Fixed obstruction of the upper airway• Upper airway surgery• Abundant heavy secretions• ARDS with PaO2/FiO2 < 150

5. General:• Severely compromised clinical state• Hemodynamic instability or shock• Post-operative arrhythmias following cardiac sur-

gery• Congenital cardiopathies affecting pulmonary flow

ARDS: acute respiratory distress syndrome; PaO2: arterial oxygen pressure; FiO2: fraction of inspired oxygen.

a. Ventilators.b. Interfaces.

c. Attachment systems.d. Tubing and associated equipment.


Table V. Criteria, materials and hospital staff requirements for use of non-invasive ventilation in the pediatric intensi-ve care unit or in the intermediate care or general wards

PICU Intermediate-care or general ward

Criteria ARF with FiO2 < 0.4 ARF with FiO2 < 0,4pH < 7.30 at any time Initial pH > 7.30General ward staff is inexperienced General ward staff is experienced Patient and/or family are non-cooperative hospitalizaciónApneas 24 hour continuous care by experienced staff

Patient and/or family are cooperativeMaterials Non-invasive specific ventilator with oxygen Non-invasive specific ventilator with oxygen

blender blenderConventional ventilator with non-invasive module Interface (oral-nasal or nasal)Interface (preferable oral-nasal) For NMD patients: assisted cough deviceFor NMD patients: assisted cough device

Hospital staff Trained physician Trained physicianTrained nurse Trained nurseFor NMD patients: respiratory physiotherapist(if unavailable, nurse to patient ratio of 1:1)

PICU: pediatric intensive care unit; ICW: intermediate care ward; ARF: acute respiratory failure; FiO2: fraction of inspired oxy-gen.

Table VI. Choice of interface in function of the patient's age and type of acute respiratory failure

RF Age Preferred interface Alternatives

Type I Neonate Short double nasal prong Nasopharyngeal tubeNasal mask

Infant Large nasal mask used as an oral-nasal mask Nasal maskShort nasal prongHelmet

1 to 6 years Oral-nasal mask Nasal maskNasopharyngeal tube

6 to 12 years Oral-nasal mask Nasal maskHelmet

> 12 years Oral-nasal mask Full face maskHelmetNasal mask

Type II Neonate Short double nasal prong Nasopharyngeal tubeNasal mask

Infant FiO2 < 0.5: nasal mask, or short nasal prong Short nasal prongFiO2 > 0.5: Large nasal mask used as an oral-nasal mask Nasopharyngeal tube

Helmet1 to 6 years Oral-nasal mask Nasal mask

Nasopharyngeal tube6 to 12 years FiO2 < 0.5: nasal mask Oral-nasal mask

FiO2 > 0.5: oral-nasal mask> 12 years FiO2 < 0.5: nasal mask

FiO2 > 0.5: oral-nasal mask Oral-nasal maskFull face mask

ARF: acute respiratory failure; FiO2: fraction of inspired oxygen.

INTERFACES (CHAPTER 5)Choosing the right interface for each patient

primarily depends on their age and their type ofrespiratory failure, as well as on the availability ofmaterials in the ward (Table VI). The interface shouldbe applied according to the algorithm shown in Fig.18, chapter 5.

VENTILATORS (CHAPTER 6)The principal criterion for choosing a ventilator

is the patient's oxygen needs: hospital staff shoulddetermine the patient's FiO2 requirements and checkwhether the ventilator in question is equipped with

an oxygen blender (see Table VII). Other influentialfactors are the need for synchronization or for aspecific NIV mode.

AccessoriesAuxiliary materials depend on the patient,

ventilator and interface. Table VIII summarizes theaccessories required for NIV of acute patients.

TREATMENT PROTOCOL FOR ACUTE PATIENTS(CHAPTER 8)

No sufficiently strong evidence exists forestablishing class A or B recommendations. Hence,the authors of this book have referred to theprotocols published to date as well to expert opinion.Based on the preceding chapters, the authors of thischapter propose the following algorithms for acutepatients (Figs. 1 and 3).

PATIENT CARE (CHAPTER 9)The efficacy of NIV is highly dependent on the

technical and theoretical expertise of the healthcare


Table VII. Overview of mechanical ventilators

Ventilator Supports NIV Oxygen Inspiratory Expiratory Compensación Monitoringtype Modes blender sensitivity sensitivity de fugas

CPAP (assembled Yes Yes Yes No No No Pressureby hand) CPAPCPAP (specific) Yes Yes Yes No No No Pressure

CPAPNIV ventilator Some Yes Someones Automatic Automatic Yes Pressurefor home use models CPAP by flow

Pressure Volume

NIV ventilator Some Yes Yes By flow Generally Yes Pressurefor hospital use models CPAP Generally automatic Volume

Pressure automatic ApneasVolume

Conventional Yes Yes Yes Yes* Yes* No Pressure(invasive) CPAP Pressure and flow Volume*ventilator with Pressure Apneasstandard modes VolumeConventional Yes Yes Yes Yes Yes Yes Pressure(invasive) ventilator CPAP Pressure and/or flow Automatic or Volumewith NIV mode Pressure Automatic or manual Apneas

manual

IMV: invasive mechanical ventilation; CPAP: continuous positive airway pressure. *Does not work well in NIV

Table VIII. Accessories required for non-invasive ventila-tion.

Hydrocolloid dressings TubingHumidifier NebulizerOxygen T-piece Plateau valveOxygen lines Pressure linesFlow meter

professionals involved, as well as the quality of patientcare, including troubleshooting of complicationsthat may arise during ventilation. NIV patient careis summarized in Figure 4.

MONITORING (CHAPTER 10)Monitoring in NIV encompasses close observation

of different parameters in the patient (e.g. clinical,analytical and functional) and in the ventilator. Theobjective is to ensure effective treatment, detect anyside effects or other complications early on, and then,if required, apply the necessary clinical criteria todecide whether ventilation should be continued,modified or stopped.

Figure 5 shows the protocol for monitoring NIVpatients before and during treatment. This algorithmis based on clinical monitoring and on arterial bloodgas levels.

1. Clinical monitoring. This is used before andduring treatment to detect any contra-indicationsto NIV or any changes in NIV mode parameters.

2. Arterial blood gases. Since NIV is non-invasive, there is a risk that hospital staff will notmonitor the patient sufficiently. To avoid this, thepatient’s ARF must be classified physiopathologically(see Table III), as recommended by current guidelinesfor adult patients and according to the data reportedby Mayordomo-Colunga et al. For patients withARDS (according to the data of Antonelli et al.), aPaO2/FiO2 > 175 after one hour of NIV is a predictivefactor for treatment success, whereas a PaO2/FiO2 <175 indicates that CMV should be considered evenif the patient does not strictly meet intubationcriteria. Hence, blood gases should be closelymonitored before, and 1 hour after, NIV has begunto avoid unnecessarily prolonging failed NIV orallowing the patient to enter into a contra-indicatedstate (PaO2/FiO2 < 150).

3. Pulse oximetry. NIV patients should be shouldbe systematically and regularly monitored by pulseoximetry, the results of which are the best indicatorof their level of oxygenation. Moreover, for adultpatients with either acute pulmonary lesions (ALI)


Figure 1. General algorithm (ARF:acute respiratory failure; NIV: non-invasive ventilation).

Hydrocolloid dressingsInterface and harness (Table VI)Ventilator (Table VII)TubingHumidifierAccessories (Table VIII)

Connect the patient tothe ventilator

Patient care algorithm (Fig. 7)Ventilation modeSetting

Type I ARF algorithm(Fig. 2)

Type II ARF algorithm(Fig. 3)

Indications (Table I)ARF diagnosis (Table II)Contra-indications (Table IV)Conditions for use (Table V)

Choose the patient andthe location of treatment

Classify the patient Types I and 2 ARF (Table III)Post-extubation

Choose the material ARF TypeAge-sizeMaterial available

Adequacy and adaptation Patient care algorithm (Fig. 7)

Evaluate efficacy LeakageAdaptationMonitoring (Fig. 4)

Prevent failure Failure algorithm (Figs. 5 and 6)(Table IX)

Patient care Patient care algorithm (Fig. 7)

Weaning from NIV


Figure 2. Algorithm for Type I acuterespiratory failure (ARF).

*If pressure support ventilation (PSV) is used, then the respiratory frequency is not programmed becauseit is not required by the ventilator; in PSV respiratory frequency is only used as a rescue parameter,analogously to the case of S/T mode in NIV-specific ventilators.ARF: acute respiratory failure; FiO2: fraction of inspired oxygen; PSV: pressure support ventilation; ST:spontaneous/timed; A/C: assisted/controlled; IPAP: inspiratory positive airway pressure; EPAP: expiratorypositive airway pressure; f: respiratory frequency; Ti: inspiratory time; ins: inspiratory; exp: expiratory;PIP: peak inspiratory pressure; PEEP: positive-end expiratory pressure. Vt: tidal volume; NIV: non-invasiveventilation.

Type I ARF


Ventilator with blender NIV-specific Conventional with NIV option

Initial setting CPAP: 5 to 10 cm H2OFiO2: 50 to 100%

No improvement S/STIPAP: 8 to 10 cm H2OEPAP: 5 to 6 cm H2O

Ramp: mediumFiO2: 50 to 100%

f (rescue): 10 to 15 rpmless than patient




Insp. trigger: minimumExp. trigger: 40 to 70%

FiO2: 50 to 100%

Test for efficacy Δ IPAP: 2 cm H2O every 5 minutes in function of the Vt reachedΔ EPAP: according to recruitment, tolerance and oxygenation


Target values IPAP: 10 to 22 cm H2OEPAP: 5 to 8 cm H2O

Vt: 8 to 10 mL/kgDecrease in f: ≥ 10 rpm within 1 h

FiO2 < 40% within 24 h

Non-vented interface

Figure 3. Algorithm for Type II acuterespiratory failure (ARF).


Type II ARF


NIV NIV-specific Conventional with NIV option

Non vented interface

Initial setting

Test for efficacy

Target values IPAP: 10 to18 cm H2OEPAP: 5 to 7 cm H2O

Vt: 8 to 10 mL/kgReduction in f within 3 to 6 h

NasalFiO2 ≥ 50% FiO2 < 50%

FiO2 < 50% without blender

FiO2 ≥ 50% with blender

CPAPApnea

BronchiolitisEPAP: 4 to 10 cm H2OFiO2: as low as possible

If respiratory workload doesn’timprove consider S/T or PSV

S/TIPAP: 8 to 10 cm H2OEPAP: 4 to 5 cm H2O

Ramp: slowFiO2: as low as possiblef (rescue): 10 to 15 rpm

less than patient




Insp. trigger: minimumExp. trigger: 40 to 70%FiO2: as low as possible

Δ IPAP: 2 cm H2O every 5 minutes in function of the Vt reachedΔ Ramp: according to tolerance and Vt reached


Figure 4. Algorithm for analyzing failure of non-invasive ven-tilation in post-extubation patients. PaO2: arterial oxygen pres-sure; FiO2: fraction of inspired oxygen; AFR: acute respiratoryfailure.

Elective NIV forrisk of ARF,atelectasis, oracute pulmonaryedema

POST-EXTUBATION

Classify patient

High risk Low risk

1 hour

FiO2 < 50%

Post-extubationARF

No Yes

1 to 6 hours

FiO2 < 50%

Evaluate intubation Continue

No Yes

High risk Low risk

Figure 6. Algorithm for monitoring non-invasive ventilation. HR: heart rate; RF: respiratory frequency; FiO2: fraction of inspi-red oxygen.

Optimize treatment

1 hour

HRRFRespiratory difficulty indexPulse oximetryTranscutaneous capnometryFiO2

HRRFRespiratory difficulty indexPulse oximetryTranscutaneous capnometryFiO2

HRRFRespiratory difficulty indexPulse oximetryTranscutaneous capnometryFiO2Chest X-rays if needed

Type I Type II

SatO2/FiO2Measure blood gases if ARDS is suspected

LeakagePatient synchronization

LeakagePatient adaptation

LeakageAbdominal distensionPatient adaptationSigns of pneumothoraxPressure sores

Classify the patient

Connect patient to the ventilator

Optimize treatment

6, 12, 24 hours

SatO2/FiO2

Type I Type II

Monitoring

SatO2/FiO2Measure blood gases if ARDS is suspected

Figure 5. Algorithm for patient care in non-invasive ventilation(NGT: nasogastric tube; CPR: cardiopulmonary resuscitation).

PacifierNGTCPR materials

Prepare the materials Interface and harnessVentilatorTubingHumidifierAccessories (Table VIII)

Prepare the patientHydrocolloid dressingsEnsure correct posture: layingback at 45° angle

Set-up the ventilationcircuit

1. Position the cap or harness2. Attach the interface3. Connect the ventilator4. Adjust the cap straps

Patient care: painand discomfort Check-up on the patient often

Sedation/analgesia

Scheduledventilation pauses

Local massagesLocal hydrationAspiration of secretionsNutritionHygieneGeneral care

Patient education Effective work of breathingEffective coughControl gastric distensionControl and treat vomitingControl leakageValsalva maneuver

Prevent andtreat side effects

ConjunctivitisSoresAbdominal distensionOtitisAir escape

Confirm that airwaysare accessible


Table IX. Systematic checklist for possible failure of non-invasive ventilation

1. Desynchronizationa. Inadequate interfaceb. NIV-specific ventilator:

• Leakage in the interface• Inspiratory trigger not activated• Inadequate respiratory circuit• Inadequate ramp

c. Conventional ventilator:• Insufficient leakage compensation• Correct expiratory trigger

2. Confirm adequate etiological treatment for the causeof RF

3. Facilitate drainage of secretions via physiotherapy4. Check for any new complications:

a. Pneumothoraxb. Aspiration pneumonia

5. Persistent hypoxemia:a. Switch to a ventilator with oxygen blenderb. Determine if the EPAP should be increasedc. Increase the FiO2

6. Persistent or newly arising hypercapnia:a. Check for leakage in the interfaceb. Confirm that the circuit has leakage controlc. Correct any re-inhalation:

• Increase the EPAP• Change to Plateau valve• Use an interface with less dead space (if possible)

d. Prevent desynchronization:• Adjust the respiratory frequency and the I/E ratio• Adjust the inspiratory and expiratory triggers (if

possible)• Determine if the EPAP should be increased

e. Ensure adequate ventilation:• Check chest wall expansion• Increase the IPAP or delivered volume• Determine if the mode or ventilator should be

changed

FiO2: fraction of inspired oxygen; EPAP: expiratory positive airway pressure; IPAP: inspiratory positive airway pressure; I/E: inspiratory toexpiratory; RF: respiratory failure

Figure 7. Algorithm for analyzing failure of non-invasive ventilation in Types I and II acute respiratory failure. RF: respiratoryfrequency; PaO2: arterial oxygen pressure; FiO2: fraction of inspired oxygen; Rx: radiografía tórax.

Indications (Tables I and II)Contra-indications (Table IV)

Type I Type II

Selected patient

Classify patient (Table III)

≥ 12 monthsLow PRISM

> 6 months> 5 kg

Low PRISM

High risk Continue

No Yes

Measure SatO2/FiO2

ARDS suspected

Arterial blood gases and chest X-ray

PaO2/FiO2

< 150 > 150

Improvement andPaO2/FiO2

> 175

Intubation*

Intubation Continue

No Yes

ARDS no suspected

Lowering RF ≥ 10

High risk Continue

No Yes

1 hour

1 hour

FiO2 < 40%

1 to 24 hours

High risk Low risk

High risk Continue

No Yes

↓ FR

1 to 6 years

Intubation Continue

No Yes

*If intubation is delayed, close monitoring should be done

or ARDS—the patients included under the mostsevere cases of Type I ARF—and a transcutaneousSatO2 < 98%, the hemoglobin saturation quotient([SatO2]/FiO2) has proven utile for indirectmeasurement of PaO2/FiO2, according to theformula: PaO2/FiO2 = [(SatO2/FiO2)-64]/0.84. Hence,this index can be useful for measuring the level ofintrapulmonary shunt in patients before and duringNIV treatment.

4. Capnometry. For acute indications of NIV,determination of CO2 levels via capnography (side-stream or micro-stream) or conventionaltranscutaneous capnometry allow evaluation of thepatient’s minute volume, and therefore, of theirresponse to NIV, especially in Type II ARF patients.

FAILURE ANALYSIS (CHAPTER 13)Any ward that adopts NIV must have

protocols for evaluating its success or failure.Various studies have been performed in whichthe authors attempted to identify predictivefactors for failure of NIV in adult patients.However, very few studies of this type have beendone for infants, and those that do exist are eitherbased on retrospective data or focus on singlepathology (bronchiolitis).

Patients generally respond clinically to NIVwithin the first hour of treatment: a positiveresponse consists of reductions in tachypnea andin chest retractions, whereas a negative response,whether due to poor adaptation or the underlyingpathology, leads to increased work of breathing,ultimately obligating the patient to be intubated.The parameters which have proven most utile fordetermining the success of NIV comprise a decreasein FiO2, a decrease in respiratory frequency, anincrease in delivered tidal volume, improved pH,and improved PaO2/FiO2.

To identify NIV patients at a high risk fortreatment failure, and for whom CMV shouldtherefore be considered, the authors of this chapterhave developed the algorithms shown in Figures 6and 7, which incorporate clinical and arterial blood

gas data. Furthermore, it is essential to use achecklist (see Table IX) to recognize all factors(whether or not they can be corrected) that causeNIV to fail.

REFERENCES1. Teague WG. Noninvasive ventilation in the pediatric intensive

care unit for children with acute respiratory failure. Pediatr Pul-monol 2003;35:418-26.

2. Dobyns EL, Carpenter TC, Durmowicz AG, Stenmark KR: Acuterespiratory failure. In: Chernick V, Boat TF, Wilmott RW, Bush A(eds). Disorders of the respiratory tract in children. Seventh edi-tion. Philadelphia, Saunders, 2006;224-42.

3. British thoracic Society guideline. Non invasive ventilation inacute respiratory setting. Thorax 2002;57:192-211.

4. International consensus conferences in intensive care medicine:Noninvasive Positive pressure ventilation in acute respiratory fai-lure. Am J Resp Crit Care Med 2001;163:283-91.

5. American Thoracic Society, European Respiratory Society, Euro-pean Society of Intensive Care Medicine and Societé de Réani-mation de langue francaise. International Consensus conferencesin intensive care medicine: non-invasive positive pressure venti-lation in acute respiratory failure. Am J Respir Crit Care Med2001;163:283-91.

6. Antonelli M, Conti G, Esquinas A, Montini L, Maggiore SM, BelloG, et al. A multiple-center survey on the use in clinical practiceof noninvasive ventilation as a first-line intervention for acuterespiratory distress syndrome. Crit Care Med 2007;35:18-25.

7. Simonds AK. Home Ventilation. Eur Respir J 2003;22 (Suppl47):538-46.

8. Norrengaard O. Non invasive ventilation en children. Eur RespirJ 2002;20:1332-42.

9. Elliot MW, Confalonieri M, Nava S. Where to perform noninva-sive ventilation? Eur Resp J 2002;19:1159-66.

10. Schönhofer B, Sortor-Leger S. Equipment needs for noninvasivemechanical ventilation. Eur Respir J 2002;20:1029-36.

11. Delclaux C, L'Her E, Alberti C, Mancebo J, Abroug F, Conti G, etal. Treatment of acute hypoxemic nonhypercapnic respiratoryinsufficiency with continuous positive airway pressure deliveredby a face mask: A randomized controlled trial. JAMA 2000;284:2352-60.

12. Ferrer M, Esquinas A, Leon M, Gonzalez G, Alarcon A, Torres A.Noninvasive ventilation in severe hypoxemic respiratory failure:a randomized clinical trial. Am J Respir Crit Care Med 2003;168:1438-44.



AAchondroplasia 15, 127, 131-134 Acute lung injury 11, 15-18, 64, 76, 79 Acute respiratory distress syndrome (ARDS)

11-21, 59, 60, 64, 76, 79, 95, 96, 137,169, 172, 174, 176

Acute respiratory failure (ARF) 3, 5, 11, 13-26, 31, 36, 37, 45, 51, 54, 56-67, 70,71, 75, 79, 82, 83, 86, 90, 95-98, 105-107, 119, 120, 150, 157, 160, 167-176

Advance 4, 58, 100, 114 Aerosol therapy 103, 107-110, 141Air trapping 88, 99, 147 ALI 13, 14, 18, 19, 21, 76, 169, 172 Alveolar derecruitment 82 Alveolar-arterial oxygen gradient 167 Analgesia 12, 88, 90, 111, 113 Ankylosing spondylitis 15 Application of NIV 15, 17, 20, 126 Aspiration of secretions 88, 136, 163 Asthma 1, 13, 15-19, 22, 57, 97, 98, 103,

112, 168, 169 Asthmatic status 60 Asynchrony 33, 45, 51, 56, 58, 63, 79, 133,

136Atelectasis 12, 15, 51, 63, 99, 108, 125, 130,

167, 168 Atelectrauma 99 Atrophy 15, 60, 130-133, 137, 145, 146, 152,

153, 162, 165, 168 Auscultation 152 Auto-PEEP 13, 99 Autologous 14AVAPS 41, 54Axillary vein thrombosis 82 BBabylog 20, 32, 33 Barotrauma 51, 87, 99, 151 Bernoulli effect 43 Biotrauma 99BiPAP® 34, 54Botulism 15, 137, 169 Boussignac 20-23, 30, 44, 114, 115 Bronchiectasis 127, 128, 168 Bronchioles 108Bronchiolitis 12, 15, 19, 20, 50, 57, 58, 60,

63, 64, 71, 95-103, 110, 112, 121, 158,167-169, 176

Bronchodilators 108, 111 Bronchopulmonary dysplasia (BPD) 98, 118,

123, 127, 128, 145, 162, 168 Bronchorrhea 89 Bulbar 15, 136, 137, 150, 169

CCapnography 76, 79, 94, 174Caps 30, 34Central alveolar hypoventilation 168

Cerebrospinal fluid 15Chest retractions 176Chin strap 67Chronic respiratory failure (CRF) 3, 11, 12,

16, 17-21, 25, 54-57, 75, 78, 133-135,138, 143, 160, 168

CI 120Coanda effect 33, 43Comfort flap 30Congenital central hypo-ventilation syndrome

(CCHS) 123, 124, 128, 129, 169Congenital hypotonias 137 Conjunctivitis 70, 72, 77, 81 Cor pulmonale 79, 124, 127, 128, 138 Cough flow 150 CPAP 3, 12, 13, 19-21, 30-39, 41, 43, 44, 47-

64, 83, 95, 99, 114-126, 171Croup 98 Cystic fibrosis 15, 57, 98, 106-111, 127, 128,

133-137, 147, 151, 153, 162, 168, 169

DDead Space 27, 31, 34, 47, 50-52, 63, 82,

85, 92, 94, 175 Decannulation 12 Dehiscence 15, 83 Demyelinating 169 Desynchronization 20, 64, 83-85, 91-94, 106,

157, 175 Diphtheria 169Duchenne's 60Dyspnea 11, 56, 70, 71, 75, 79, 99, 107, 110,

128 Dystrophy 15, 130-137, 151, 153, 169

EEducation 143, 146 Effective cough 69, 149 Effects of NIV 79, 159 Effusion 151, 169 Elastance 58 Enuresis 78, 79, 138 EPAP 28, 40-42, 50-54, 60, 63, 75, 83, 85,

92-4, 125, 127, 147, 175Epiglottitis 98Esprit 39 Ethics 144, 163, 165 Ethmoidal 14, 83 Evita 33, 39, 40 Expiratory loop 63, 155

FFacial malformations 162 Facial trauma 169 Failure of NIV 3, 12, 14, 25, 64, 75, 91, 97,

157, 158, 174 Fastening systems 34 Feeding 26, 27, 33, 72, 83, 141, 142, 163

Fixed obstruction of the airway 14 Flow meter 20, 49, 66, 100, 171 Fluidic flip 33, 43 Forced vital capacity (FVC) 131, 138

GGalileo® 101 Gas exchange 11-13, 19, 75-79, 97, 105, 113,

127, 131, 138, 142, 147, 151 Gastric distension 15, 52, 67, 69, 71, 72, 77,

81, 83, 119, 120 Gastrointestinal surgery 15 Glossoptosis 132

HHarmony 85 Headache 78Heliox 48, 97-103, 108, 109, 112 Helmets 2, 25, 31, 34, 82 Hemodynamic instability 13, 14, 64, 93, 151,

169 Home care 21, 72, 133, 135, 138-144, 164,

166Humidifiers 83, 105-109, 111, 140, 142, 156 Hyaline membrane disease (HMD) 2, 20, 118 Hyperammoniemia 98 Hypercapnia 3, 12, 13, 15, 58, 72, 75, 82,

85, 88, 92, 94, 96, 105, 113, 124, 128,129, 138, 167, 175

Hyperhydration 14 Hyperinflation 56, 83 Hyperkyphosis 132 Hypertrophy 127, 132, 133 Hypervolemia 13 Hypopharynx 76 Hypoxemia 12-14, 51, 56, 63, 64, 82, 85, 88,

92, 93, 102, 113, 124, 127-129, 139,157, 167, 175

IIndications to NIV 66, 167, 172 Infant Flow 43, 44, 100, 121 Inhalation therapy 19 Inhaler 108, 109, 110, 111, 112Interfaces 17, 22, 25-36, 48, 60, 63, 66-72,

77, 79, 81-84, 91, 97, 111, 117, 118,121, 125, 135, 136, 140, 149, 156, 159-171

Interior pressure in the alveoli 7 Interior pressure in the pleural space 7IPAP 13, 40, 41, 52-54, 60, 63, 67, 75-79, 83,

85, 92, 94, 125, 148, 175Iron lung 1, 37, 136

KKinesitherapy 128Knightstar 101Kyphoscoliosis 15, 79, 136, 137

Index

LLaryngospasm 89, 90Laryngotracheomalacia 90, 98, 118, 123, 127,

137Leakage 19, 27, 29, 33-41, 50-52, 56, 60, 63,

69-85, 91-96, 101-107, 118, 119, 125,133, 136, 175

Lordosis 68Lumbar hyperlordosis 132Lung compliance 77, 99, 108, 136, 145, 147

MMacroglossia 126, 132 Malacia 127 Masks 1, 2, 19, 25-38, 60, 62, 71, 75, 81, 98,

105, 106, 109, 114, 117, 118, 124, 129,140, 141, 146, 156

Maxillar hypoplasia 81 Micrognathia 126Mucopolysaccharidosis (MPS) 131, 132Myasthenia gravis 15, 60, 130, 137, 168, 169Myelomeningocele 129, 137Myocarditis 13Myopathy 131

NNasal fossa 81Nasal mucosa 83, 111Nasal prongs 19, 25, 26, 32, 33, 49, 67, 98,

118, 120, 121Nasal septum 120 Nasogenian area 81 Nebulizers 107-110, 156 Negative pressure ventilation (NPV) 1, 4, 5,

37, 97, 136Neuromuscular diseases 10, 12, 15, 60, 123,

127, 130, 137, 144, 145, 162, 168 NNT 120 Nocturnal 22, 78, 79, 124-128, 131-134, 138,

145, 146, 152, 164Non-invasive positive pressure ventilation

(NIPPV) 2, 15, 16, 22, 23, 36, 38, 98,121, 136, 144, 176

Nosocomial 4, 106, 111, 138, 157 Nursing care 2, 11

OObstruction of airways 133 Obstructive pulmonary disease 3, 4, 19, 22,

45, 95, 97, 103, 136 Obstructive sleep apnea (OSA) 22, 37, 47, 50,

57, 63, 123-129, 132, 137, 168 Oral-nasal 3, 14, 18, 19, 25-32, 49, 60, 63,

69, 71, 81-84, 91, 107, 109, 125, 146-149, 156, 170

Orbital herniation 15, 83 Oropharynx 34, 69, 107, 132Otitis 69, 70, 72Oxygen blender 17-21, 42, 52, 53, 60, 84,

85, 92, 93, 100, 101, 155-157, 171, 175 Oxygen T-piece 20, 43, 66, 85, 171 Oxygen therapy 20, 22, 32, 47, 60, 62, 78,

84, 105, 117, 128, 135, 140-142

PPacifier 34, 67, 91 Paradoxical breathing 11, 70, 146, 168 Parenchyma 14, 56, 95, 147, 168 Parenchymal lung disease 123, 131 PEEP 13, 32, 41, 49-51, 55, 56, 60-63, 83,

84, 92, 94, 99, 101, 147, 148Phrenic electrostimulation 44Pleura 169Pleural effusion 151, 169Pleural pressure 1Pneumatic belt 2Pneumobelt 4, 136Pneumonia 12, 15, 18, 19, 34, 50, 59, 60,

87, 92, 93, 106, 111, 149, 167-169, 175Pneumothorax 4, 15, 77, 81, 83, 92, 93, 98,

167, 169, 175Poliomyelitis 4, 169Polypnea 75, 124Polysomnography (PSG) 79, 123, 126, 127,

133Popliteal space 68Post-extubation 3, 4, 50, 95-98, 169, 174Ppl 7, 8, 9, 10Preparing the patient 66, 141Pressure sores 29, 31-35, 38, 66, 67, 77, 78,

119, 125, 136, 140, 141Profile 30, 35, 84, 108, 109, 113Programming 43, 48-52, 55, 60, 63, 148Propofol 88-90Proportional assist ventilation 58Proportional assisted ventilation (PAV) 39-42,

54, 54-60Protection of the airway 14Psychomotor development 143PTP 7-10Pulmonary function 75, 78, 79, 131, 135,

138, 145, 159Pulmonary hyperinflation 83Pulmonary physiology 7Pulse oximetry 76-79, 119, 126, 139, 172Puritan Bennett 39, 41Pw 7, 8

QQuality control 155-160Quality of life 3, 12, 21, 78, 79, 124, 130,

133, 137, 145, 161-165

QRebreathing 45, 86, 97REM sleep 123, 125-131Residual functional capacity 53, 54Respiratory center 75, 169Respiratory insufficiency 18, 58, 64, 144, 147,

152, 176 Respiratory work 97, 147 Retrognathia 126 RF 3, 7, 11, 14, 25, 32, 35, 38, 47, 48, 56,

59, 60, 75, 88, 92, 96, 170, 174, 175 Rhinorrhea 108 Rocking bed 2, 4 RR 95, 120

SScoliosis 15, 130-132, 149, 162, 168 Sedation 12-15, 71, 83, 86-91, 113, 115, 168Shock 14, 169 Sialorrhea 89 Skin necrosis 27, 81, 159 Sleep apnea hypopnea syndrome (SAHS) 22 Spinal muscular 130-134, 137, 145, 146, 152,

153, 162, 165, 168 Spinal muscular atrophy (SMA) 130-134, 137,

145-147, 150, 152, 153, 162, 165, 168 Status asthmaticus 16, 57, 109 Straps 29, 31, 34, 67, 69-71, 82, 114 Stridor 127, 133 Surgical tape 34 Swallowing 21, 136-138, 168 Synchrony® 41, 42, 53 Syringomyelia 169

TTechnical problems 35, 83 Tidal volume 38, 41, 42, 51-54, 66, 77, 83,

85, 94, 96, 99, 101, 109, 112, 123, 125,129, 151, 176

Tonsillectomy 127, 132, 168 Total recoil pressure of the respiratory system

7Tracheitis 98 Tracheomalacia 168 Tracheostomy 131, 132, 135-137, 142-153,

158, 164 Training 18, 19, 73, 91, 141, 142, 155-160,

165Transpulmonary pressure 7, 8, 9, 10 Trigger 10, 20, 34, 35, 38-45, 51, 52, 54, 61,

63, 75, 84, 86, 91-93, 98, 101, 102,106, 117, 125, 147, 175

Tubing 20, 25, 28, 42, 49, 53, 55, 60, 63, 65-72, 76, 84, 85, 93, 94, 101, 106, 108,141, 155, 156, 170, 171

UUvulopalatopharyngoplasty (UPPP) 132

VV/Q mismatch 13, 59, 125, 167 Vented interfaces 82 Ventilation in the home 135, 137, 139, 141,

143 Vision 29, 31, 32, 101 VPAP 41, 42, 84, 85, 101

WWeaning 2, 5, 12, 16, 32, 90, 98, 118, 145,

153, 158 Werdnig-Hoffman 130

180 Index

falta el logo - efccnaefccna.org/images/stories/publication/niv_book.pdf · sección de neumología...

Documents