novel uses of noninvasive ventilation - respiratory care · novel uses of noninvasive ventilation...

Novel Uses of Noninvasive Ventilation

Joshua O Benditt MD

IntroductionContinuous Positive Airway Pressure and Noninvasive Ventilation in the

Postoperative PeriodContinuous Positive Airway Pressure and Noninvasive Ventilation During

BronchoscopyNoninvasive Ventilation During Percutaneous Gastrostomy Tube

PlacementNoninvasive Ventilation in Pandemic Respiratory InfectionsContinuous Positive Airway Pressure for Obesity Hypoventilation

SyndromeSummary

Noninvasive ventilation (NIV) and continuous positive airway pressure (CPAP) have been used invarious unusual settings to assist breathing. NIV is now frequently used to treat exacerbations ofchronic obstructive pulmonary disease and chronic respiratory failure in neuromuscular disease.This paper discusses CPAP and NIV for postoperative hypoxemia, preventing intubation in high-risk bronchoscopy, respiratory failure in pandemics, obesity hypoventilation syndrome, and respi-ratory support during percutaneous endoscopic gastrostomy tube placement. Key words: noninva-sive ventilation, NIV, continuous positive airway pressure, CPAP, bronchoscopy, pandemic, obesityhypoventilation syndrome. [Respir Care 2009;54(2):212–219. © 2009 Daedalus Enterprises]

Introduction

In the past 30 years, dramatic strides have been made inthe development of noninvasive ventilation (NIV). Nega-tive-pressure ventilation devices were used during the po-lio epidemics up through the 1950s, then positive-pressure

ventilation via nasal or face mask gained acceptance, andnasal and face masks underwent rapid development in the1980s.1 Mask were initially used with continuous positiveairway pressure (CPAP) to treat sleep apnea, but soon themasks were used with volume-control and pressure-con-trol ventilators to treat patients with neuromuscular andchest-wall restrictive diseases2-4 as well as acute respira-tory failure. NIV has also been extensively used for exac-erbations of chronic obstructive pulmonary disease (COPD)and is now considered almost the standard of care foracute hypercarbic exacerbations of COPD in the emer-gency department and intensive care unit (ICU).5 NIV hasyet to be shown to have efficacy for chronic respiratoryfailure in COPD.6 This paper discusses novel applicationsof CPAP and NIV, such as for postoperative hypoxemia,prevention of intubation in high-risk bronchoscopy pa-tients, respiratory failure in pandemics, obesity hypoven-tilation syndrome, and during percutaneous endoscopic gas-trostomy tube placement.

Joshua O Benditt MD is affiliated with the Division of Pulmonary andCritical Care Medicine, Department of Medicine, University of Wash-ington Medical Center, Seattle, Washington.

Joshua O Benditt MD presented a version of this paper at the 42ndRESPIRATORY CARE Journal Conference, “Noninvasive Ventilation inAcute Care: Controversies and Emerging Concepts,” held March 7-9,2008, in Cancun, Mexico.

The author reports no conflicts of interest.

Correspondence: E-mail: [email protected].

212 RESPIRATORY CARE • FEBRUARY 2009 VOL 54 NO 2

Continuous Positive Airway Pressure andNoninvasive Ventilation in the Postoperative Period

Respiratory insufficiency in patients undergoing majorsurgery, especially of the chest or upper abdomen, is rel-atively common.7 After abdominal surgery, respiratorycomplications occur in approximately 10% of patients, andreintubation represents 30% of those complications.8 Pain,splinting, and respiratory muscle dysfunction are likelycontributors to hypoxemia and respiratory insufficiency.9

Common methods to prevent this in the postoperative pe-riod have included incentive spirometry, chest physiother-apy, mucolytics, and intermittent positive-pressure breath-ing.10-13 Recently there have been trials of CPAP and NIVto reduce postoperative respiratory complications. Caseseries have used preventive CPAP or NIV in the recoveryroom or ICU,14,15 and the results are promising, with re-ports of prevention of reintubation ranging up to 100%.However, the lack of control groups limits interpretationof those studies. Three randomized controlled trials havebeen performed: 2 with CPAP and one with NIV.

Squadrone et al performed a large multicenter trial thatinvolved 15 centers in Italy.16 They included patients whohad undergone upper-abdominal surgery of at least 90 minand had a ratio of PaO2

to fraction of inspired oxygen (FIO2)

of � 300 mm Hg. Patients were excluded if they had ahistory of cardiac disease, COPD, pH � 7.30, PaCO2

� 50 mm Hg, hematocrit � 30%, or coma. They random-ized 104 subjects to the control arm, who received FIO2

of0.5 via air-entrainment mask, and 105 subjects to the treat-ment arm, who received FIO2

of 0.5 and CPAP at 7.5 cm H2O.At the end of a 6-hour treatment period both groups wereplaced on FIO2

of 0.3. If PaO2/FIO2

was � 300 mm Hg, thepatient was placed back on his or her assigned treatment.With standardized criteria for the decision to intubate, therewas a markedly lower intubation rate and lower rates ofpneumonia and sepsis in the treatment arm (Fig. 1).

In a similar study that used CPAP to prevent post-ex-tubation respiratory complications in patients who under-went repair of thoracoabdominal aneurysm, Kindgen-Milles et al compared controls to patients who were treatedwith CPAP for 12–24 hours after surgery. The patientswere extubated in the ICU when standard extubation cri-teria were met, then randomized. The control group re-ceived oxygen via non-occlusive face mask at 25 L/min;other routine medical therapies (eg, morphine for pain, andstress ulcer prophylaxis); and CPAP for 10 min every4 hours. The experimental group received the same stan-dard therapies plus CPAP at 10 cm H2O for 12–24 hoursafter the procedure,17 and they had better oxygenation,fewer pulmonary complications, and shorter hospital staythan the controls.

Auriant and colleagues performed a randomized trial ofNIV in postoperative respiratory failure.18 Patients were

included if they met 3 of 4 criteria for respiratory insuf-ficiency after extubation following lung-resection surgery:respiratory rate � 25 breaths/min, accessory muscle use,PaO2

/FIO2� 200 mm Hg, and chest radiograph abnormal-

ities such as atelectasis or consolidation. Subjects wererandomly assigned to standard postoperative care and sup-plemental oxygen or standard postoperative care plus NIVvia oronasal mask and a bi-level pressure-support ventila-tor adjusted to generate a tidal volume of 8–10 mL/kg, tomaintain oxyhemoglobin saturation � 90%, and to keeprespiratory rate � 25 breaths/min. The study was stoppedafter the first interim analysis because of dramatic differ-ences in reintubation rate and mortality. Approximately21% of the subjects in the NIV arm were reintubated,according to standardized criteria, compared to 50% in thecontrol arm. In-hospital mortality was 37.5% in the controlarm, and 12.5% in the treatment arm.

Although the above-described 3 studies were small, itappears that positive airway pressure improves outcomesin patients who develop postoperative respiratory insuffi-ciency. It is not known whether CPAP or NIV is the bettermode.

Continuous Positive Airway Pressure andNoninvasive Ventilation During Bronchoscopy

Bronchoscopy in patients with respiratory disease canpresent a substantial challenge. The bronchoscope occu-pies approximately 10% of the normal airway and canincrease work of breathing and decrease PaO2

by 10 –20 mm Hg, which can cause respiratory complications andcardiac arrhythmia.19,20 Several investigators have evalu-ated CPAP and NIV for preventing these complications inat-risk patients, and case series have had success withCPAP and NIV.21-23

Fig. 1. Kaplan-Meier estimates of intubation risk in the group thatreceived supplemental oxygen alone (control) versus the groupthat received continuous positive airway pressure (CPAP). Thecontrol patients had a higher risk of intubation (P � .005 via logrank test). (From Reference 16, with permission.)

NOVEL USES OF NONINVASIVE VENTILATION

RESPIRATORY CARE • FEBRUARY 2009 VOL 54 NO 2 213

Maitre et al24 conducted a randomized controlled trial ofbronchoscopy with CPAP versus with high-flow oxygenin patients at high risk for respiratory complications (PaO2

� 125 mm Hg on high-flow mask). The mask/CPAP de-vice (Fig. 2) delivers either high-flow oxygen or high-flowoxygen plus CPAP (at 7.5 cm H2O) during bronchoscopy.The study outcomes were oxygenation during and afterbronchoscopy, and respiratory complications, including in-tubation. Fifteen patients were entered into each arm of thestudy. Figure 3 shows the oxygenation results. Seven pa-tients in the oxygen group needed some form of ventila-tory support (4 intubations) in the 6 hours following theprocedure, as opposed to only one patient in the CPAPgroup, although that difference is not statistically signifi-cant.

In another randomized controlled trial, Antonelli et alused NIV in high-risk patients undergoing bronchosco-py.25 Patients who required bronchoscopy and had a PaO2

/FIO2

� 200 mm Hg were randomized to receive an FIO2of

0.9 via face mask, or NIV via oronasal mask, with inspira-tory pressure of 15–17 cm H2O, expiratory pressure of5 cm H2O, and FIO2

of 0.9. During bronchoscopy, FIO2was

reduced to maintain oxygen saturation (measured via pulseoximetry) � 92%. PaO2

/FIO2was substantially higher both

during bronchoscopy (261 mm Hg vs 139 mm Hg) andone hour after bronchoscopy (176 mm Hg vs 140 mm Hg),although both groups started at essentially the same PaO2

/FIO2

at baseline (143 mm Hg vs 155 mm Hg). WhetherNIV reduces the requirements for bronchoscopy-associ-ated intubation in high-risk patients remains unknown. Itdoes appear that both CPAP and NIV are safe and effec-

tive for maintaining oxygenation in relatively hypoxemicpatients undergoing bronchoscopy.

Noninvasive Ventilation During PercutaneousGastrostomy Tube Placement

For patients with neuromuscular disease, swallowingproblems and malnutrition can become major and life-threatening issues. Percutaneous endoscopic gastrostomytube placement may prevent large-volume aspiration bybypassing the dysfunctional swallowing muscles, thus pro-tecting the airway from aspiration and providing adequatecalories. Sedation is required for this procedure, and forpatients with respiratory muscle insufficiency and low vi-tal capacity, sedation may add significant risk of respira-tory complications. Intubation of patients with neuromus-cular disease is associated with prolonged and sometimespermanent invasive mechanical ventilation.26 In fact, theAmerican Academy of Neurology issued a practice param-eter that percutaneous gastrostomy tubes should be placedbefore the forced vital capacity drops below 50% of pre-dicted, to reduce the risk of respiratory failure.27

Unfortunately, some patients fall below that level priorto receiving a percutaneous gastrostomy tube. NIV hasbeen used in several groups of neuromuscular patientsduring percutaneous endoscopic gastrostomy tube place-ment, even when the forced vital capacity is quite low(Fig. 4). Boitano et al used bi-level NIV to support 5patients with amyotrophic lateral sclerosis and forced vitalcapacity that ranged from 21% to 44% of predicted duringpercutaneous endoscopic gastrostomy tube placement, andhad no complications.28 Similarly, in a group of patients

Fig. 2. Oronasal mask (Peters Laboratories, Bobigny, France) witha device (Vygon Laboratories, Ecouen, France) that delivers con-tinuous positive airway pressure (CPAP) with or without supple-mental oxygen during bronchoscopy. The bronchoscope is in-serted through an opening in the front of the mask. (Adapted fromReference 24, with permission.)

Fig. 3. Oxyhemoglobin saturation (measured via pulse oximetry[SpO2

]) before, during, and after fiberoptic bronchoscopy (FOB) in2 groups: one that received high-flow oxygen; another that re-ceived high-flow oxygen plus continuous positive airway pressure(CPAP). FIO2

� fraction of inspired oxygen. Full face mask � oro-nasal mask. (From Reference 24, with permission.)



with Duchenne muscular dystrophy, Birnkrant et al hadsuccessful and uncomplicated percutaneous endoscopicgastrostomy tube placement in patients in whom vital ca-pacity was nearly unmeasureable.29

Noninvasive Ventilation in PandemicRespiratory Infections

Recently, because of the threats of severe acute respi-ratory syndrome (SARS), avian flu, and bioterrorism, therehas been widespread discussion of respiratory-care issuesin pandemic and mass-casualty situations.30-35 NIV hasbeen suggested as a possible treatment for patients withmilder respiratory failure, or to decrease the intubationrate. The major debate is on the safety of NIV, becauseinfectious exhaled aerosol can enter the ambient air throughthe exhalation ports in the NIV mask or tubing. The evi-dence on this topic is of 2 kinds: reports of NIV use duringthe SARS epidemic of 2003, and bench modeling of drop-let release and spread with an NIV model.

Cheung et al reported their experience with NIV andSARS at the Pamela Youde Nethersole Eastern Hospital inHong Kong.36 During March and April of 2003 they usedNIV to treat 20 patients with SARS. To reduce the riskfrom droplets they employed 3 strategies: oronasal maskto prevent large leaks through the mouth; exhalation valve(Whisper-Swivel II, Respironics, Murrysville, Pennsylva-nia), which they thought should have a lower likelihood ofproducing jets of droplets; and viral/bacterial filter (Air-life, Allegiance Healthcare, McGaw Park, Illinois) in-linebefore the exhalation valve. The mean duration of NIVwas 84.3 hours, and they indicated that 14 intubationswere avoided. Those treated with NIV had shorter ICUstay than the intubated patients (3.1 d vs 21.3 d). Duringthat time they also monitored the health and SARS sero-logic status of the 105 clinicians who came into contact

with the patients on NIV. There were no seroconversionsor symptoms consistent with SARS in any of the clini-cians, although 3 of them declined the blood draw. Theclinicians were protected initially with N95 masks, andlater with powered air-purifying respirators, although theproportion of each was not reported.

In a related study, Yam et al retrospectively comparedthe above results from the NIV strategy at Pamela YoudeNethersole Eastern Hospital to those from 13 other hospi-tals in Hong Kong that only used intubation and invasiveventilation during the SARS epidemic.37 Data were ob-tained from the Hong Kong Hospital Authority SARS da-tabase, which contains data on all patients with confirmedSARS during that period in 2003. Compared to the hos-pitals that used invasive ventilation only (451 patients), thehospital that used NIV (42 patients) had a lower adjustedodds ratio for intubation (0.36, 95% confidence interval[CI] 0.64 – 0.79, P � .01) and death (0.24, 95% CI0.08�0.72, P � .01). Patients also improved more rapidlyafter steroid pulse rescue. The severity of disease at pre-sentation was slightly worse at the NIV hospital.

In a retrospective epidemiologic study of clinicians ex-posed to the SARS virus during the outbreak of 2003,Fowler and colleagues found that direct involvement withintubation of a SARS patient carried the highest risk ofcontracting the infection, and working with patients onNIV did not have a statistically significant risk of devel-oping infection in the clinician.38 However, the number ofinstances of NIV use was quite small (6 uses of NIVdevices), which makes it difficult to interpret the findings.

In another retrospective analysis of clinician exposureto a SARS-infected index case, Scales39 reported that of 22clinicians present in the room while the patient was onNIV, 4 developed SARS and were in the room for � 31 min.Only one of those clinicians was wearing a mask. Of 18clinicians who did not develop SARS, only one was in theroom for � 31 min and was wearing a mask, gown, andgloves. Thus, the epidemiologic data are mixed at best,and the actual risk of SARS spread with NIV is not known.Obviously, only a prospective study could determine theeffectiveness of NIV versus invasive ventilation in themanagement of SARS.

Hui et al40 conducted a bench study of aerosol disper-sion during NIV. A mannequin wearing an oronasal NIVmask was exposed to various ventilating pressures whileknown concentrations of aerosol particles were instilledinto the artificial lung. The aerosol that exited the maskexhaust ports was visualized with laser light and recordedwith a computer (Fig. 5). The dispersion maps (Fig. 6)showed that, at an inspiratory pressure of 18 cm H2O andan expiratory pressure of 4 cm, dispersion ranged up to0.5 m from the mask, which is certainly enough to con-taminate clinicians.

Fig. 4. Patient with amyotrophic lateral sclerosis undergoing per-cutaneous endoscopic gastrostomy tube placement with nonin-vasive ventilatory support.



At present, despite conflicting data, we must assumethat there is at least some contamination risk from infec-tious droplets when treating an infected patient with NIV,and that clinicians must take careful precautions.

The American Association for Respiratory Care recentlyissued a statement concerning ventilation in mass-casualtysituations, including pandemics.41 They recommendedagainst NIV in pandemic flu, because flu often progressesto acute respiratory distress syndrome, for which NIV isnot the standard of care. Also, the 1–2-hour respiratory-therapist time commitment to initiate NIV would probablybe impossible in a pandemic situation. The debate on thiscontinues.

Continuous Positive Airway Pressurefor Obesity Hypoventilation Syndrome

Obesity hypoventilation syndrome describes an individ-ual with obesity (body mass index � 30 kg/m2), awakehypercapnia (PaCO2

� 45 mm Hg), and sleep-disordered

breathing.42 Ninety percent of the sleep-disordered breath-ing in obesity hypoventilation syndrome consists of ob-structive sleep apnea.43 The remaining 10% of these pa-tients have an apnea-hypopnea index � 5 events/h, andtheir problem has been referred to as sleep-related hy-poventilation, which is defined by a nocturnal PaCO2

ele-vation of 10 mm Hg. Compared to eucapnic obese pa-tients, patients with even mild obesity hypoventilationsyndrome (PaCO2

45–50 mm Hg) have a lower quality of

Fig. 5. A: Neodymium-yttrium-aluminum-garnet (Nd:YAG) laser thatgenerates the2-dimensional laser “light sheet” (wavelength527nm)B: Laser light sheet aligned on the mannequin’s median sagittalplane, showing the pattern of particles in the exhaled gas. (FromReference 40, with permission.)

Fig. 6. Dispersion map of particles released from a ResMed me-diummaskduringbi-level-positive-airway-pressureventilation.Theparticles disperse up to 0.5 m with 18 cm H2O inspiratory pressureand 4 cm H2O expiratory pressure. (From Reference 40, withpermission.)

Fig. 7. Pathophysiology of obesity hypoventilation syndrome.(Based on data in Reference 42.)



life, more daytime hypersomnolence, and higher health-care costs.44 The generation of obesity hypoventilation syn-drome is complex. Figure 7 presents a potential physio-logic explanation. The reason for the elevated daytimePaCO2

has not been entirely elucidated, but most expertsagree that it involves renal buffering of PaCO2

with HCO3–

that persists during the daytime.45-48

Positive airway pressure can ameliorate obesity hypoven-tilation syndrome. The sleep-disordered breathing comesin 3 varieties: obstructive sleep apnea with hypopneas,obstructive hypoventilation due to upper-airway resistancesyndrome, and central hypoventilation.49 A treatment al-gorithm for obesity hypoventilation syndrome has beenproposed, based on the frequency of occurrence of thedisordered sleep patterns (Fig. 8). Obstructive sleep apneais the most common sleep disorder, so CPAP is tried first.If the oxyhemoglobin saturation cannot be brought above90%, NIV is tried, with inspiratory pressure increased abovethe last expiratory pressure setting by at least 8–10 cm H2Oto bring the saturation above 90%. If that is not effective,then supplemental oxygen is added to bring the saturation

above 90%. If none of those interventions are successful,then weight-loss surgery or even tracheostomy and inva-sive ventilation are considered.

A recent prospective trial of CPAP versus NIV included23 patients with obesity hypoventilation syndrome, whowere compared to 23 matched patients who had eucapnicobstructive sleep apnea.48 In 57% of the subjects withobesity hypoventilation syndrome, CPAP alone relievedsleep-disordered breathing and hypoxemia, with an opti-mal mean CPAP pressure of 13.9 � 3.1 cm H2O. Theremaining 43% required various levels of NIV and sup-plemental oxygen to maintain saturation above 90%.

In another retrospective study, Perez et al studied 54patients with obesity hypoventilation syndrome who un-derwent NIV.43 Outcomes measured included survival, clin-ical status, and arterial blood gas values. At the end of thefollow-up period (mean duration 50 months), PaO2

hadincreased by 24 mm Hg from baseline (95% CI 21–28 mm Hg, P � .001) and PaCO2

had decreased by 17 mm Hg(95% CI 13–20 mm Hg, P � .001). NIV improved sub-jective sleepiness (mean Epworth sleepiness scale scoredecreased from 16 � 5 to 6 � 2, P � .001), and dyspneadecreased in all but 4 patients. During follow-up, 3 pa-tients died (one of them due to the progression of respi-ratory failure). NIV was withdrawn in 5 patients who hadachieved sufficient weight loss, and the condition of 16patients was maintained without respiratory failure by theuse of simple therapy with CPAP.

It is clear that for the majority of patients with obesityhypoventilation syndrome CPAP is generally adequate,though NIV and supplemental oxygen may be needed if

Table 1. Evidence About Novel Uses of CPAP and NIV

Evidence Grade*

Application CPAP NIV

Prevention of postoperative reintubation A AHigh-risk bronchoscopy B BPercutaneous endoscopic gastrostomy tube

placementB B

Pandemic/disaster management I IObesity hyponventilation syndrome A A†

* A: Good scientific evidence suggests that the potential benefits substantially outweigh therisks. Discuss the treatment with eligible patients.B: At least fair scientific evidence suggests that the potential benefits outweigh the risks.Discuss the treatment with eligible patients.C: At least fair scientific evidence suggests that there are benefits, but the balance betweenbenefits and risks is too close to make a general recommendation. Do not offer the treatmentunless there are individual considerations that suggest it.D: At least fair scientific evidence suggests that the risks outweigh the potential benefits. Donot routinely offer the treatment.I: Evidence is insufficient to recommend for or against the treatment.† Noninvasive ventilation (NIV) should be used if continuous positive airway pressure(CPAP) is found ineffective.(Based on data in Reference 50.)

Fig. 8. Treatment algorithm for obesity hypoventilation syndrome.CPAP � continuous positive airway pressure. BiPAP � bi-levelpositive airway pressure. IPAP � inspiratory positive airway pres-sure. EPAP � expiratory positive airway pressure. (Based on datain Reference 42.)



CPAP is inadequate; this therapy improves daytime bloodgas values, quality of life, and probably survival.

Summary

Table 1 summarizes the strength of the evidence forcurrently novel CPAP and NIV applications. The use ofCPAP and NIV will probably continue to grow as im-proved technology and creative clinicians attempt to treatmore complicated conditions.

REFERENCES

1. Sullivan CE, Ellis ER. Nocturnal positive pressure ventilation via anasal mask [letter]. Am Rev Respir Dis 1987;136(3):791-792.

2. Ellis ER, Bye PT, Bruderer JW, Sullivan CE. Treatment of respira-tory failure during sleep in patients with neuromuscular disease.Positive-pressure ventilation through a nose mask. Am Rev RespirDis 1987;135(1):148-152.

3. Gay PC, Patel AM, Viggiano RW, Hubmayr RD. Nocturnal nasalventilation for treatment of patients with hypercapnic respiratoryfailure. Mayo Clin Proc 1991;66(7):695-703.

4. Hill NS, Eveloff SE, Carlisle CC, Goff SG. Efficacy of nocturnalnasal ventilation in patients with restrictive thoracic disease. Am RevRespir Dis 1992;145(2 Pt 1):365-371.

5. Hill NS. Noninvasive ventilation for chronic obstructive pulmonarydisease. Respir Care 2004;49(1):72-87; discussion 87-89.

6. Wijkstra PJ, Lacasse Y, Guyatt GH, Casanova C, Gay PC, MeechamJones J, Goldstein RS. A meta-analysis of nocturnal noninvasivepositive pressure ventilation in patients with stable COPD. Chest2003;124(1):337-343.

7. Thompson JS, Baxter BT, Allison JG, Johnson FE, Lee KK, ParkWY. Temporal patterns of postoperative complications. Arch Surg2003;138(6):596-602.

8. Lawrence VA, Dhanda R, Hilsenbeck SG, Page CP. Risk of pulmo-nary complications after elective abdominal surgery. Chest 1996;110(3):744-750.

9. Siafakas NM, Mitrouska I, Argiana E, Bouros D. Effects of surgeryon the function of the respiratory muscles. Monaldi Arch Chest Dis1999;54(6):526-531.

10. Celli BR, Rodriguez KS, Snider GL. A controlled trial of intermittentpositive pressure breathing, incentive spirometry, and deep breathingexercises in preventing pulmonary complications after abdominalsurgery. Am Rev Respir Dis 1984;130(1):12-15.

11. Freitas ER, Soares BG, Cardoso JR, Atallah AN. Incentive spirom-etry for preventing pulmonary complications after coronary arterybypass graft. Cochrane Database Syst Rev 2007;(3):CD004466.

12. Hall JC, Tarala R, Harris J, Tapper J, Christiansen K. Incentivespirometry versus routine chest physiotherapy for prevention of pul-monary complications after abdominal surgery. Lancet 1991;337(8747):953-956.

13. Pfenninger J, Roth F. Intermittent positive pressure breathing (IPPB)versus incentive spirometer (IS) therapy in the postoperative period.Intensive Care Med 1977;3(4):279-281.

14. Kindgen-Milles D, Buhl R, Gabriel A, Bohner H, Muller E. Nasalcontinuous positive airway pressure: a method to avoid endotrachealreintubation in postoperative high-risk patients with severe nonhy-percapnic oxygenation failure. Chest 2000;117(4):1106-1111.

15. Pennock BE, Kaplan PD, Carlin BW, Sabangan JS, Magovern JA.Pressure support ventilation with a simplified ventilatory supportsystem administered with a nasal mask in patients with respiratoryfailure. Chest 1991;100(5):1371-1376.

16. Squadrone V, Coha M, Cerutti E, Schellino MM, Biolino P, Occella,et al. Continuous positive airway pressure for treatment of postop-erative hypoxemia: a randomized controlled trial. JAMA 2005;293(5):589-595.

17. Kindgen-Milles D, Muller E, Buhl R, Bohner H, Ritter D, SandmannW, Tarnow J. Nasal-continuous positive airway pressure reducespulmonary morbidity and length of hospital stay following thoraco-abdominal aortic surgery. Chest 2005;128(2):821-828.

18. Auriant I, Jallot A, Herve P, Cerrina J, Le Roy Ladurie F, FournierJL, et al. Noninvasive ventilation reduces mortality in acute respi-ratory failure following lung resection. Am J Respir Crit Care Med2001;164(7):1231-1235.

19. Goldstein RA, Rohatgi PK, Bergofsky EH, Block ER, Daniele RP,Dantzker DR, et al. Clinical role of bronchoalveolar lavage in adultswith pulmonary disease. Am Rev Respir Dis 1990;142(2):481-486.

20. Levine S, Stover D. Bronchosopy and related techniques. In: LevineS, editor. Respiratory disease in the immunosuppressed host. Phila-delphia, PA: Lippincott Williams & Wilkins; 1991:73-93.

21. 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(3):724-728.

22. Antonelli M, Pennisi MA, Conti G, Bello G, Maggiore SM, MichettiV, et al. Fiberoptic bronchoscopy during noninvasive positive pres-sure ventilation delivered by helmet. Intensive Care Med 2003;29(1):126-129.

23. Soroker D, Ezri T, Szmuk P. Fiberoptic bronchoscopy in a patientrequiring continuous positive airway pressure. Anesthesiology 1995;82(3):797-798.

24. Maitre B, Jaber S, Maggiore SM, Bergot E, Richard JC, Bakthiari H,et al. Continuous positive airway pressure during fiberoptic bron-choscopy in hypoxemic patients: a randomized double-blind studyusing a new device. Am J Respir Crit Care Med 2000;162(3 Pt1):1063-1067.

25. Antonelli M, Conti G, Rocco M, Arcangeli A, Cavaliere F, ProjettiR, Meduri GU. Noninvasive positive-pressure ventilation vs. con-ventional oxygen supplementation in hypoxemic patients undergoingdiagnostic bronchoscopy. Chest 2002;121(4):1149-1154.

26. Pope JF, Birnkrant DJ. Noninvasive ventilation to facilitate extuba-tion in a pediatric intensive care unit. J Intensive Care Med 2000;15:99-103.

27. Miller RG, Rosenberg JA, Gelinas DF, Mitsumoto H, Newman D,Sufit R, et al. Practice parameter: the care of the patient with amyo-trophic lateral sclerosis (an evidence-based review): report of theQuality Standards Subcommittee of the American Academy of Neu-rology: ALS Practice Parameters Task Force. Neurology 1999;52(7):1311-1323.

28. Boitano LJ, Jordan T, Benditt JO. Noninvasive ventilation allowsgastrostomy tube placement in patients with advanced ALS. Neu-rology 2001;56(3):413-414.

29. Birnkrant DJ, Ferguson RD, Martin JE, Gordon GJ. Noninvasiveventilation during gastrostomy tube placement in patients with se-vere Duchenne muscular dystrophy: case reports and review of theliterature. Pediatr Pulmonol 2006;41(2):188-193.

30. Branson RD, Rubinson L. Mechanical ventilation in mass casualtyscenarios. Respir Care 2008;53(1):38-39.

31. Caves ND, Irwin MG. Attitudes to basic life support among medicalstudents following the 2003 SARS outbreak in Hong Kong. Resus-citation 2006;68(1):93-100.

32. Lau AC, Yam LY, So LK. Management of critically ill patients withsevere acute respiratory syndrome (SARS). Int J Med Sci 2004;1(1):1-10.

33. Li TS, Gomersall CD, Joynt GM, Chan DP, Leung P, Hui DS.Long-term outcome of acute respiratory distress syndrome caused by



severe acute respiratory syndrome (SARS): an observational study.Crit Care Resusc 2006;8(4):302-308.

34. Li Y, Leung GM, Tang JW, Yang X, Chao CY, Lin JZ, et al. Roleof ventilation in airborne transmission of infectious agents in thebuilt environment - a multidisciplinary systematic review. Indoor Air2007;17(1):2-18.

35. Thomas PA. Severe acute respiratory syndrome. Indian J Med Mi-crobiol 2003;21(3):152-160.

36. Cheung TM, Yam LY, So LK, Lau AC, Poon E, Kong BM, YungRW. Effectiveness of noninvasive positive pressure ventilation in thetreatment of acute respiratory failure in severe acute respiratory syn-drome. Chest 2004;126(3):845-850.

37. Yam LY, Chan AY, Cheung TM, Tsui EL, Chan JC, Wong VC.Non-invasive versus invasive mechanical ventilation for respiratoryfailure in severe acute respiratory syndrome. Chin Med J (Engl)2005;118(17):1413-1421.

38. Fowler RA, Guest CB, Lapinsky SE, Sibbald WJ, Louie M, Tang P,et al. Transmission of severe acute respiratory syndrome during in-tubation and mechanical ventilation. Am J Respir Crit Care Med2004;169(11):1198-1202.

39. Scales D. SARS exposure in an ICU (presentation). http://microbiology.mtsinai.on.ca/sarswatch/presentations/scales/pres_scales.swf. Accessed December 4, 2008.

40. Hui DS, Hall SD, Chan MT, Chow BK, Tsou JY, Joynt GM, et al.Noninvasive positive-pressure ventilation: An experimental model toassess air and particle dispersion. Chest 2006;130(3):730-740.

41. American Association for Respiratory Care. AARC SARS guidancedocument. 2003. http://www.aarc.org/resources/sars/sars%20aarc%20guidance%20document.pdf. Accessed December 4, 2008.

42. Mokhlesi B, Tulaimat A. Recent advances in obesity hypoventilationsyndrome. Chest 2007;132(4):1322-1336.

43. Perez de Llano LA, Golpe R, Ortiz Piquer M, Veres Racamonde A,Vazquez Caruncho M, Caballero Muinelos O, Alvarez Carro C.Short-term and long-term effects of nasal intermittent positive pres-sure ventilation in patients with obesity-hypoventilation syndrome.Chest 2005;128(2):587-594.

44. Hida W, Okabe S, Tatsumi K, Kimura H, Akasiba T, Chin K, et al.Nasal continuous positive airway pressure improves quality of life inobesity hypoventilation syndrome. Sleep Breath 2003;7(1):3-12.

45. Norman RG, Goldring RM, Clain JM, Oppenheimer BW, CharneyAN, Rapoport DM, Berger KI. Transition from acute to chronichypercapnia in patients with periodic breathing: predictions from acomputer model. J Appl Physiol May 2006;100(5):1733-1741.

46. Goldring RM, Heinemann HO, Turino GM. Regulation of alveolarventilation in respiratory failure. Am J Med Sci 1975;269(2):160-170.

47. Goldring RM, Turino GM, Heinemann HO. Respiratory-renal ad-justments in chronic hypercapnia in man; extracellular bicarbonateconcentration and the regulation of ventilation. Am J Med Dec 1971;51(6):772-784.

48. Kessler R, Chaouat A, Schinkewitch P, Faller M, Casel S, Krieger J,Weitzenblum E. The obesity-hypoventilation syndrome revisited: aprospective study of 34 consecutive cases. Chest 2001;120(2):369-376.

49. Banerjee D, Yee BJ, Piper AJ, Zwillich CW, Grunstein RR. Obesityhypoventilation syndrome: hypoxemia during continuous positiveairway pressure. Chest 2007;131(6):1678-1684.

50. Agency for Healthcare Research Quality. Grade definitions. http://www.ahrq.gov/clinic/3rduspstf/ratings.htm. Accessed December 4,2008.

Discussion

Hill: Upper-abdominal surgery hasfor years been on many lists of con-traindications to NIV, on the reason-ing that the sutures might be stressedby the pressure. Was there any com-ment on that in the literature you re-viewed? Is there good reason for thatconcern?

Benditt: I did not see any articlesthat stressed that point, though theydid list complications related to sur-gical mishaps from overinflation. Atmy hospital a lot of surgeons go toCPAP or BiPAP [bi-level positive air-way pressure] ventilation very quickly,especially after thoracic surgery—right after extubation.

Hill: I’m not aware of any literaturethat substantiates that concern. We userelatively low NIV pressures, ofcourse, and studies that have usedpostoperative CPAP or NIV have notreported any problems with suture in-

tegrity,1,2 but this concern seems to beuniversal. Many people have told me,“Our surgeons are really hesitant touse NIV.” Has anybody else encoun-tered that?

1. Joris JL, Sottiaux TM, Chiche JD, DesaiveCJ, Lamy ML. Effect of bi-level positiveairway pressure (BiPAP) nasal ventilationon the postoperative pulmonary restrictivesyndrome in obese patients undergoing gas-troplasty. Chest 1997;111(3):665-670.

2. Kindgen-Milles D, Muller E, Buhl R,Bohner H, Ritter D, Sandman W, TamowJ. Nasal-continuous positive airway pres-sure reduces pulmonary morbidity andlength of hospital stay following thoraco-abdominal aortic surgery. Chest 2005;128(2):821-828.

Nava: Yes. Think about what youdo with these patients. You get thephysiotherapist to do some exercisesright after surgery, because you wantto avoid atelectasis, and they some-times apply the positive pressure. So Ithink we are a little bit misled by thatpaper, because it tells us that you cando physiotherapy without a human be-

ing there, because CPAP is a physio-therapy approach; it’s not ventilatorysupport.

Hill: You’re saying that you need toexpand the lungs?

Nava: Exactly. Instead of having aphysiotherapist come twice a day, youapply the helmet and you do physio-therapy. When you expand the lungs,even voluntarily, you generate an “ab-normal” pressure, so it’s the same con-cept.

Hill: But the pressure that surgeonsare concerned about is on the uppergastrointestinal tract. With physiother-apy, you wouldn’t be applying pres-sure there. So maybe it would be apreferable therapy?

Benditt: In one study1 they gave allpatients physiotherapy and acetylcys-teine to try to control for that. I thinkthe intervention had an effect beyondthe physiotherapy. But I agree that it



has to do with maintaining patency ofairways and alveoli.

1. Kindgen-Milles D, Muller E, Buhl R,Bohner H, Ritter D, Sandmann W, TarnowJ. Nasal-continuous positive airway pres-sure reduces pulmonary morbidity andlength of hospital stay following thoraco-abdominal aortic surgery. Chest 2005;128(2):821-828.

Gay: Probably 4 or 5 times in thepast couple of years I’ve seen patientswith acromegaly who needed pituitaryresection. They’ve got severe apneaand obviously they’re going to havepotential cerebrospinal-fluid disrup-tion, so surgeons aren’t going to wantany positive pressure near that at all.There is debate whether one shouldget a tracheostomy. There are someanecdotal reports, but nothing that pro-vides a good guide, and I’ve seen bothextremes. On the one hand there arepatients with relatively minor OSA[obstructive sleep apnea] with theirnoses packed, and they’re miserablepostoperatively and getting narcotics;they don’t get CPAP and they get intoa lot of trouble. Then there are thoseyou anticipate and tracheotomize, butthey might have just flown throughthis. Most of the surgeons urge no post-operative positive airway pressure atall, so we do end up tracheotomizingsome of the more serious OSA pa-tients now.

Benditt: We do not see those kindsof patients.

Doyle:* At the last AARC [Ameri-can Association for Respiratory Care]meeting I saw an abstract on auto-titrating CPAP with undiagnosed OSApatients postoperatively.1 They used amodified Berlin questionnaire beforeintubation in the operating room, andfound significant postoperative im-provement with CPAP. Did you seeany papers on that application?

1. MacAulay MJ, Bramlett M. 30 days of postsurgical OSA identification and treatment(abstract). Respir Care 2007;52(11):1577.

Benditt: I did not.

Gay: Anesthesiologist BhargaviGali, at Mayo Clinic, does a preoper-ative assessment of higher-risk OSApatients,1 with the Calgary score.2

They identify higher-risk patients, andif the patient desaturates for an ex-tended period in the recovery room,they go to a more heavily monitoredarea. They’re not really determining atreatment approach; they’re determin-ing who needs to be monitored morecarefully. When we try to do just-in-time rescue therapy with CPAP, thenit is difficult. Think about it: these pa-tients are miserable; they’re just get-ting out of surgery; the mask is on forabout 10 minutes and then it’s on thefloor. So trying to figure out wherebest to monitor these people is the bet-ter mindset. Perhaps the more impor-tant thing is making a decision onwhere to monitor them without swal-lowing up every ICU bed.

1. Gali B, Whalen FX Jr, Gay PC, Olson EJ,Schroeder DR, Plevak DJ, Morgenthaler TI.Management plan to reduce risks in peri-operative care of patients with presumedobstructive sleep apnea syndrome. J ClinSleep Med 2007;3(6):582-588.

2. Flemons WW, Whitelaw WA, Brant R,Remmers JE. Likelihood ratios for a sleepapnea clinical prediction rule. Am J RespirCrit Care Med 1994;150(5 Pt 1):1279–1285.

Mehta: Baillard et al1 used NIV priorto semi-urgent intubation, and NIV re-duced the nadir of oxygen saturationduring the intubation. Did you reviewthat paper?

1. Baillard C, Fosse JP, Sebbane M, ChanquesG, Vincent F, Courouble P, et al. Nonin-vasive ventilation improves preoxygenationbefore intubation of hypoxic patients. Am JRespir Crit Care Med 2006;174(2):171-177.

Benditt: No, I was concentrating onthe postoperative period, but I amaware of that article.

Hill: They found that pre-intubationNIV avoided oxygen desaturations inpatients with hypoxemic respiratoryfailure.

Mehta: Yes, the saturation washigher.

Hill: They had such a good responseto oxygen, I wondered why they intu-bated all of them afterwards.

Doyle: Regarding concern aboutNIV with infectious-disease patients,clearly there’s a plume of aerosol com-ing out of the opening of a single-limb NIV system, but the exhaled gasfrom an intubated patient also ends upin the ambient air unless the exhaledgas is filtered. However, many of theinvasive ventilators do not have exha-lation filters, so clinicians may have afalse sense of security with invasiveventilation. Near the gas output portthey could be exposed to an infectiousaerosol. In those studies was the gasfiltered?

Kacmarek: In every one that I’maware of the recommendation is to puta filter on the expiratory port, whichis easy with an intubated patient, andvery difficult with NIV.

Hill: I have a problem with theAARC’s statement that NIV shouldnot be used for SARS,.1 I think that’sgoing too far on the available data. Idon’t think we know what happensunder these circumstances, and I don’tknow that David Hui’s findings2 tellus exactly what’s going to happen. Acouple other studies found very lowrates of disease transmission whenhealth-care workers were exposed topatients with SARS on NIV.3,4 So I’mnot convinced that NIV is a poorchoice.

There are ways to filter expired airfrom BiPAP, and you can use a dual-limb circuit during NIV if you’re reallyconcerned about infectious aerosol. AsGeeta [Mehta] pointed out, in the To-ronto epidemic it was during intuba-

* Peter Doyle RRT, Respironics, Carlsbad, Cal-ifornia.



tion that there seemed to be a lot oftrouble. I don’t think we have enoughevidence to make a firm statement thatNIV is contraindicated. If I get pneu-monia from bird flu, I’ll try NIV first,and you can intubate me if it doesn’twork.

1. American Association for Respiratory Care.AARC SARS guidance document. 2003.http://www.aarc.org/resources/sars/sars%20aarc%20guidance%20document.pdf. Ac-cessed August 13, 2008.

2. Hui DS, Hall SD, Chan MTV, Chow BK,Tsou JY, Joynt g, et al. Noninvasive posi-tive-pressure ventilation: an experimentalmodel to assess air and particle dispersion.Chest 2006;130(3):730-740.

3. Cheung TM, Yam LY, So LK, Lau AC,Poon E, Kong BM, Yung RW. Effective-ness of noninvasive positive-pressure ven-tilation in the treatment of acute respiratoryfailure in severe acute respiratory syndrome.Chest 2004;126(3):845-850.

4. Han F, Jiang YY, Zheng JH, Gao ZC, HeQY. Noninvasive positive-pressure venti-lation treatment for acute respiratory fail-ure in SARS. Sleep Breath 2004;8(2):97-106.

Benditt: Actually I think the AARCstatement didn’t address the infectiousaerosol. Rather, the statement wasbased on the prediction that many ofpandemic patients will progress toARDS [acute respiratory distress syn-drome] (though the evidence on thatis debatable) and NIV is not the stan-dard of care for ARDS. And in a mass-casualty situation the time required forNIV initiation would be prohibitive.

Hill: Now wait a damn minute! Iknow you’re just describing what theysaid, but how much time does it taketo intubate someone? How much skilldoes it take? In an epidemic are yougoing to have hundreds of health-careworkers greasing up endotrachealtubes to intubate all these people?Come on: NIV is so much easier toapply, and with a cooperative patientwho has some dyspnea it doesn’t nec-essarily take a lot of additional time. Idon’t buy that.

Kacmarek: I respectfully disagree.In that situation it’s much easier to

deal with an intubated patient. You’vedone it; it takes a long time to accli-mate a patient to NIV. If any of thepredictions are near correct about thenumber of health-care workers thatwill be available, we will not be ableto do NIV. If 40% of the population issick and you’ve cut down your staffby 40% and you’ve doubled or tripledor quadrupled the number of patientswith acute respiratory failure, you willnot be able to spend an hour or two toinitiate NIV. These patients can be in-tubated in minutes and started on me-chanical ventilation. From the workperspective, I agree 100% that NIVwould be difficult to do. I don’t thinkthere is as much concern about trans-mitting the organism via NIV as israised in some of the discussions, butfrom a work perspective I think intu-bation clearly is the way to go in thatsetting.

Keenan: Where I work we simplywould not be able to provide enoughinvasive ventilation during an epi-demic, because we don’t have enoughventilators, so we would be forced touse everything available to us, includ-ing NIV. The most important thing isto know how to make NIV safe in thatsetting, and it appears that if you’reproperly gowned and protected, youcan probably use NIV. I understandthe concern that patients will developARDS [acute respiratory distress syn-drome], and some of them will de-compensate and not be adequately sup-ported by NIV, and will requireintubation. However, even today wehave difficulty coping with our cur-rent patient needs for ventilatory sup-port. I don’t know how we would copewith a superimposed epidemic. I imag-ine we would have to pull out all avail-able ventilators, including NIV. So itmay not be ideal, but we need to knowhow to do it safely.

Kacmarek: You can use noninva-sive ventilators invasively. There’s noreason to relegate them only to NIV.

Most NIV ventilators that are designedfor use in acute care are approved forinvasive use, and those that are notcan be safely used if monitors areadded.

Mehta: I respectfully disagree withNick. With a patient with influenza oranother highly contagious respiratorydisease, if I had the choice of NIV orintubation, I would intubate. We hadSARS at my hospital, and I think therisk of transmitting infection is a lothigher with NIV than with intubation,as your image of the NIV exhaled aero-sol plume showed. And we don’t haveenough negative-pressure rooms,which would be even more importantfor patients treated with NIV.

Hill: I’ll have to take a minority po-sition here. With regard to preventingtransmission, protection is the mostimportant thing. Regarding clinicianwork load, we found that it takes a bitmore time to initiate NIV,1 but I don’tthink it requires 90 to 120 minutes ofdedicated time at the bedside. It’s pos-sible to initiate multiple masks if thepatients are near to each other.

1. Kramer N, Meyer TJ, Meharg J, Cece RD,Hill NS. Randomized, prospective trial ofnoninvasive positive-pressure ventilation inacute respiratory failure. Am J Respir CritCare Med 1995;151(6):1799-1806.

Kallet: I side with Bob on this. I thinkit’s going to be a crap-shoot! Patientswith fulminate ARDS will get intu-bated straight away.

Hill: I don’t argue with that, but Ithink there are people who are goingto be on the edge, who are dyspneicand hypoxemic but otherwise stable,and you’re going to have to use whatyou have. I’m not saying everybodyshould go on NIV. My objection isthat if you say NIV should not be used,that’s going too far. There may be pa-tients in whom NIV is appropriate, andI think the AARC statement shouldsay that patients with fulminant ARDS



should definitely be intubated, but thatyou can consider NIV in dyspneic, hy-poxemic patients who are less ill.

Kallet: That’s very reasonable.When we were talking about prepar-ing for an avian flu pandemic, oneidea that came up was to maybe finda way to cheaply mass-produce some-thing like the old Down’s flow gen-erators for mask CPAP. A lot of peo-ple may be able to get that and we’lljust have to cross our fingers that it’llbe sufficient. We’re going to have totriage our equipment in a pandemicsituation. At my hospital we’re not go-ing to have the luxury of spendingmuch time with any one patient. Eachrespiratory therapist might have tocover 20 or more ventilators, whetherinvasive or NIV. So I think the situa-tion will dictate what actually happens:whether we do the therapy the way wewant to do it or we’re just trying tostabilize oxygenation and save asmany people as possible.

Kacmarek: In a full pandemic wemay be using very different types ofventilators. Many ventilators I canteach a non-respiratory person to usein 15 minutes. Instead of having 15therapists on a shift, I may only have6 therapists on a shift, and instead of80 or 90 ventilated patients there maybe 300. Even if we had the opportu-nity to start NIV, with inexperiencedhands operating the ventilator, I don’tsee how somebody could initiate NIVcompared to being told to set dials X,Y, and Z on a simple straightforwardventilator after the patient had beenintubated. I think you are right, Nick,that some patients will benefit fromNIV, but I don’t think the practicali-ties will allow us to do NIV.

Nava: Would you use an air-entrainment mask, or a reservoir mask?The aerosol contamination risk isabout the same with supplemental ox-ygen or NIV, I suspect. I agree withNick that not all hypoxic patients need

to be intubated, and I don’t think wecan afford to intubate all the patients.NIV is for a subset of less-sick pa-tients. This may be a situation in whicha helmet is useful, because it wouldprotect clinicians from aerosol infec-tion risk, because everything is en-closed inside it. And the risk of put-ting on or removing a helmet shouldbe the same as that of extubating orsuctioning a patient with an endotra-cheal tube.

Benditt: A bench study1 found thatthe spread of aerosol particles wasclearly related to the pressure in themask, and at lower pressure there wasa much lower dispersion, so I don’tthink an air-entrainment mask has thesame risk as BiPAP at, say, 18/4 cm H2O. I think they showed thatpretty clearly.

1. Hui DS, Hall SD, Chan MTV, Chow BK,Tsou JY, Joynt GM, et al. Noninvasive pos-itive-pressure ventilation: an experimentalmodel to assess air and particle dispersion.Chest 2006;130(3):730-740.



novel uses of noninvasive ventilation - respiratory care · novel uses of noninvasive ventilation...

Documents