high output respiratory failure: *

High Output Respiratory Failure: *

An Important Cause of Death Ascribed to Peritonitis or Ileus

JOHN F. BURKE,** M.D., HENNING PONTOPPIDAN,** M.D., CLAUDE E. WELCH,f M.D.

From the Harvard Medical School and Massachusetts General Hospital, Boston, Massachusetts

PERrrONITIS and ileus lead to pathologicchange not only in the peritoneal cavity,but throughout the body. These extraperi-toneal effects are more extensive with peri-tonitis than with ileus. They usually havebeen considered to be the systemic mani-festations of inflammatory disease so com-paratively little attention has been paid tothe function of specific organ systems andto the mechanisms of their decompensationduring this period of intense metabolic ac-tivity. It is clear, however, that peritonitisor ileus is not a primary cause of death,since it is possible to establish several morebasic pathological abnormalities. It also hasbeen established that inadequate functionof one or more specific organs may resultin death.A variable degree of interest has been

manifested in these failing systems. Theeffects of serious abnormalities of the elec-trolytes, plasma or red blood cell mass, aswell as local sepsis, have been studied thor-oughly. On the other hand, important tar-get organs32 (as Moore has called them)include the kidneys, the cardiovascular sys-

* Presented before the American Surgical Asso-ciation, Phoenix, Arizona, April 3-5, 1963.

** Clinical Associate in Surgery and Tutor inMedical Sciences, Harvard Medical School; As-sistant Surgeon, Massachusetts General Hospital.

*** Instructor in Anesthesia, Harvard MedicalSchool; Associate Anesthetist, Massachusetts Gen-eral Hospital.

f Associate Clinical Professor of Surgery, Har-vard Medical School; Visiting Surgeon, Massachu-setts General Hospital.

This work was supported in part by a grant(E-2392) from the National Institutes of Health,U.S.P.H.S.

tem, the respiratory system and the intes-tine itself. As a group these viscera have re-ceived relatively little emphasis. This isprobably because the normal respiratory,cardiovascular or renal systems have suchgreat functional reserve that little thoughtis given to them in patients with peritonitisor ileus unless there has been serious pre-existing disease involving these viscera.

In this paper the discussion will centeron the respiratory system and upon a typeof failure that occurs with initially normalor essentially normal lungs. We believethat this is a particularly dangerous andoften unrecognized syndrome that meritscareful attention.

It is clear that failure of any single organis related intimately to failure of all otherviscera. Nevertheless, failure of a singleorgan does not imply failure of all simul-taneously. There is indeed a very interest-ing temporal relationship which can be out-lined and in itself emphasizes the primaryimportance of decompensation of the re-spiratory system.

Thus, as shown in Figure 1, the type ofrespiratory failure discussed in this papertends to appear in the very early phasesof peritonitis. There is indeed a later stageof pulmonary failure that may occur aweek or more after onset of peritonitis thatis secondary to aspiration, pneumonitis orembolization, but in this paper we are notconcerned with this late type. Cardiac fail-ure in the early course of peritonitis is asequel of hypoxia that is dependent on pul-monary failure and is manifested by car-diac arrest. Late cardiac failure is usually

581

582

IPERSISTENTPERITONEAL CA4USES Of OATIr

CONTAMINATION ASCRIBED rO PERIrONrIISSEPSIS

PULMONARYFA ILURE

BLOOD, PLASMA, ELECTROLYTE ABNORMALITIES

RENAL FAILURE

EMBOLISM

INTESTINALOBSTRUCTION

O 1 2 3

POST- OPERATIVE WEEK

FIG. 1. Causes of death ascribed to peritonitis.

a result of pulmonary embolization. Renalfailure, another late manifestation, hasbeen documented fully, particularly byHagen and Merrill.21 Failure of the intes-tine itself is temporally one of the last ofthe types of organ failure. If it is not re-

lieved, as shown in a previous communica-tion,39 it essentially leads to death from in-anition.The period of intraperitoneal sepsis most

dangerous for the patient is the first week.Unless there is continued peritoneal soilageor undrained sepsis that leads to bac-teremia, fatal sepsis originating from theperitoneal cavity is not common. On theother hand, careful attention to electrolytes,plasma and red cell mass is requiredthroughout the entire illness.

It therefore appears that in the first fewdays after the onset of a severe peritonealinsult the surgeon must give priority tofluid and electrolyte replacement, to thecontrol of sepsis and to the prevention ofrespiratory failure. On the other hand, suc-

cessful treatment of one decompensatedsystem may allow the patient to live to be-come vulnerable to failure of another tar-get organ that characteristically is mani-fested at a later date.The concept of early pulmonary failure

has been considered by surgeons for many

Annals of SurgeryOctober 1963

years. A basic theory has been outlinedand a specific method of therapy has beenconsidered in many centers. However, to

test the validity of these concepts, it ishighly desirable to study a series of patientswith this specific lesion and to documenttheir course.

Material

We have been interested in these pa-

tients for the past five years. During thatperiod we have studied 21 patients withacute pulmonary failure due indirectly to

peritonitis or ileus. Their course is out-

lined briefly in Table 1. In common withall developing studies, investigation of vari-ous parameters has been more completewith the passage of time and with the de-velopment of a special respiratory unitunder the direction of one of us (H. P.).

It is obvious that criteria for inclusion insuch a study must be sharp or all typesof combined pulmonary disease and peri-tonitis could be included. The criteria usedwere as follows:

1. The patient had acute peritonitis or ileussecondary to an intraperitoneal lesion.

2. No patient had an acute pulmonary or

cardio-vascular disease or injury.3. Pulmonary failure was the only immediate

threat to life.

The exact state of the patient's pulmo-nary function preceding his illness is ofgreat importance. Since these patients were

not considered to have significant respira-tory problems prior to operation, no pre-

liminary function studies were carried outon any of them. They could, however, bedivided into three groups depending on

the presence and extent of previous pulmo-nary disease, as follows:

Group 1. No previous chest disease (7 pa-

tients).Group 2. No previous symptoms, but slight

pulmonary changes on physical or x-ray examina-tion (10 patients).

Group 3. Symptoms and signs of previousemphysema and chronic bronchitis (4 patients).

BURKE, PONTOPPIDAN AND WELCH

HIGH OUTPUT RESPIRATORY FAILURE

Clinical Data

The clinical data are summarized inTable 1. The 21 patients included 12 men

and nine women. The ages ranged from 31to 76 years.

Causes of Peritoneal Insult. Peritonitisand concomitant ileus existed prior to op-

eration in 10 patients. Four of them hadbacterial peritonitis (perforated duodenalulcer-1; perforated diverticulitis-2; per-

forated herniated colon-1); one had pan-

creatitis and five had ruptured aorticaneurysms. Eleven developed postoperativeperitonitis (following gastric resection formassive hemorrhage-4; following totalcystectomy-2; following jejunal perfora-tion-2; following aortic aneurysm resec-

tion-2; following nephrolithotomy-1).The low incidence of perforated duo-

denal ulcers in this series is worthy of com-

ment. There were indeed many patientsseen in the hospital following perforationof a peptic ulcer with severe restriction ofpulmonary function. However, with lapa-rotomy and removal of large quantities ofperitoneal exudate, particularly from thesubdiaphragmatic space, there was dra-matic improvement in pulmonary functionalmost at once. The only patient who failedto respond to this direct form of therapywas one with previous emphysema (S. H.,Table 1) who later required tracheostomyand mechanical ventilation. It is of inter-est in this respect to quote Anscome: 3

"Patients with a generalized peritonitis follow-ing perforation of a peptic ulcer showed a greaterreduction in pulmonary mechanical power thanany other group. The effect was so severe that thevital capacities were all less than one liter and inone patient it was 420 cc. and vital spirogramscould therefore not be done. Measurements on thefirst day after operation showed an increase over

the preoperative figure, a finding which did notoccur in any other type of patient or condition.Despite this severe initial reduction and in thepresence of an epigastric midline incision, therewas a surprisingly rapid recovery, the maximalfigures being reached in six days."

Our experience, as well as that of Ans-combe, indicates the serious impairment ofpulmonary function by peritoneal inflam-matory disease. If, however, the inflamma-tory stimulus can be removed, as in thecase with perforated peptic ulcer, and dis-tention relieved before fatigue, pulmonaryfunction returns to normal almost imme-diately following the relieving operativeprocedure.

It is also of interest that three of the fourpatients with peritonitis following gastricresection for massive hemorrhage were

found in Group 3 patients. The associationof severe ulcer disease with emphysemahas been noted frequently in the past.18120'28, 30, 36 Parenthetically it may be noted thatmassive gastro-intestinal hemorrhage hasbeen a frequent major complication in pa-

tients under treatment in our respiratoryunit in the past one and one-half years.

Examination of PatientPre-existing chronic bronchitis and em-

physema, obesity, malnutrition or feeblemuscles of respiration were important pre-

monitory warning signals. A history ofheavy cigarette smoking 9 33 is also a dangersign.During the development of respiratory

failure following a severe peritoneal insultall patients were severely ill. They were

usually extremely anxious, apprehensiveand were at times thrashing. Respirationswere rapid, shallow, labored and often-times grunting. They appeared exhaustedand their faces were often flushed. All were

severely dyspneic.X-rays of the chest taken at this time

showed elevated diaphragms and varyingdegrees of basal atelectasis. The absence ofany striking changes in the lung fields maylead the attending surgeon into a falsesense of security concerning the stabilityof respiratory compensation.

Vital Signs. Pulse and respiratory ratesshowed marked increases that were sus-

tained until respiratory failure was cor-

Volume 158Number 4 583

BURKE, PONTOPPIDAN AND WELCH

_0 0 0 0 CS

NO _It

4i

- t- (14 -1 t101010 010

214 IN CIN

0 0 8

°rq

8

004 00 00 t 00

010 010 010 010

0)

OU) ens

0,

010

m 00

010

00

2

t-

00

00

0-02)0

0s

00

CS


g g~os o

_ (V

If) ~ e

00 +0% 0(d C.)

0~ If)

\0 0

C U)CN CS

o on

00

t-

002)

'0a

t0 0 00

1n 10 0 10O2 t_ t1 oO

_ 0

0 0,I

0)

0cvo

0

0%

olo~

0%C IO

o

CN

~1

C,-

-t It-010

0

en

_ = = XQ ,; _ E

0A 0 0 E E

od o .¢ S .00j Sta

00

-I.

C:

0

u

C-E

.)

0

CS

6r.

0..I0't

"I.

0

u

CZ

.2

C:

NO

Q

4/)0

584

H

2i

V.v

E

CZ

0)

O

m~

0

Zolm0

N

0

00~o

oo20

0N

00

eCN

0

0

010

toOl

(N00

20

t-

0

20

a)4

02;

20C

124

04

U.

ew

bo

CZ

02

J02

0E

0

a)

r-bo

0

Volume 158 HIGH OUTPUT RESPIRATORY FAILURENumber 4

-00 > 0 0 - -5

00 0 ~ -Z Z . 0 0 0 0 0 0 O>C) C)O C) C) C) Cf CU U U C.v U

Cc

0C.-0CC)~. 0-

C - 0£ O O+ 1 CO C' 0

v rq Sh:io ioR ° f q + e

o,,~~CC Ct) C C)C CCC\

CO 1, 010°IX 0101Rt° Nj °co ual- -1o °x lo s0

Cl

,- 010 0'0 010 0 000 010, 0001 0

- - -~~~~~~~~~~~~~~~~~~~~~~~~~0 -1

uw ~~~~C) C- NCt)Co ----C) s t t

N . _

H b|1 8C ) ECO

Cq CO C)

C) -CZ ~--

8 o oo co coo o C o 0

CC)C) CO0 010 010 0)0 0 00 01 010- 0 00- 0 010

C - O -C

C)~~~~~~~~~~~~~~~~~~~~~~~~~)0

bC) - C) C) C C) C) CZ

as ~~~~~E E E ~ E E.C-OC vC - CCZ

Cd 0 a 0 00avO C)O

00 (~~~~.--e E -CZ C

cC00 CC 0aCCC. CO )0 )0

vC0C CZ)) CZC CZcv) C-i CC0.C0 CO ) 2C )) CC0 C C

COO. o > CZ m )CZ~C

CZ- d 4 Q~ CZ CZ i ACZ-CZ

0 -- - - -~~~~~~~ ~~~-n -- -

0

585

586 BURKE, PONTOP]

ADMISSION

PERITONEAL TRACHEOSTOMYINSULT ARTIFICIAL RESPIRATION

MD.B. fis4540566

PULSE RArE I10

70 _

30 _

RESP/RA 70ry 2IRATE 20

130

sYsroL/C5.P IIlo[-120

02 100

SA URA/ION 95

70

Po ¢°2 60mnm I\

50

30C0L _ _

CONP 30rtiCaVrENrT 2 S5.EqfL 20 _-

27.4

7.2E_

7.0

DAYS 2 3 4 5 6 7 8

ANURIA OLIGURIA NORMAL RENAL FUNCTION

FIG. 2. Case 1. Vital signs and arterialblood studies.

rected. The blood pressure, on the otherhand, remained in a normal range beforeand at the time of early respiratory failure.Hypotensive levels occurred only when thepatient became moribund. In some patientsa transient rise in blood pressure may beobserved during the early stages of respira-tory decompensation.

Blood Gas and pH Studies. The pH ofarterial blood varied, as expected, with theadequacy of CO2 elimination. In addition,it was also affected by the presence or ab-sence of a metabolic acidosis caused bysuch abnormalities as tissue hypoxia,29 andacidosis of renal origin.21 The adequacy ofCO2 excretion does not follow the ade-quacy of arterial oxygenation. Arterial oxy-gen desaturation almost always precedesinsufficient CO2 elimination. In fact, in theearly stages of this syndrome respiratory

PIDAN AND WELCH Annals of SurgeryOctober 1963

alkalosis was seen not infrequently as theresult of a relative hyperventilation, prob-ably because the patient attempted to oxy-genate the arterial blood adequately. Inthe later stages CO2 excretion also failedand a respiratory acidosis was the rule.The arterial blood gases showed a con-

sistent pattern only in oxygen unsaturation.The arterial CO2 tension (PaCO2) variedaccording to the adequacy of CO2 excre-tion as described above. The plasma CO2content changed with other variables suchas renal function and arterial CO2 tension(PaCO2 ) .5 Nevertheless the informationgained by these four determinations was ofgreat value in the minute to minute careof the patient.Treatment. An initial attempt to estab-

lish adequate respiratory function wasmade by relief of intra-abdominal tension(by gastro-intestinal intubation or by lapa-rotomy with decompression or drainagewhere indicated). In addition, tracheal suc-tion and chest physiotherapy was used inall cases. Bronchoscopy was done rarely.These methods were ineffective in 20 ofthe 21 patients in this study, a fact thatdoes not minimize the value of these ma-neuvers but indicates the severity of thedisease in our cases. Effective therapy wasinstituted with a tracheostomy and inser-tion of a cuffed silver tracheostomy tubein conjunction with an intermittent posi-tive pressure mechanical respirator.

Complications and Mortality. Five ofthe 21 patients died. One succumbed to afatal cardiac arrhythmia during trachealsuction two weeks following tracheostomy.Two cardiac arrests occurred just prior totracheostomy. One patient lived and theother died of pulmonary embolus threeweeks later. The other three patients diedof late renal failure two weeks to twomonths after the date of tracheostomy.

Case ReportsThree cases are included to demonstrate

specific points. Patient 1 is typical, illus-


trating this type of respiratory failure. Shewas salvaged by the treatment outlined.Patient 2 is an example of the effects offatigue. Respiratory decompensation ap-

peared ten days after operation and twodays after onset of peritonitis. Patient 3 isa representative patient in whom the impli-cations of this syndrome were not recog-

nized because of the maintenance of a nor-

mal blood pressure until cardiac arrestoccurred.

Case 1. M. B. Figure 2 gives a detailed pic-ture of the vital signs, blood gas and pH studiesof a 49-year-old woman without pre-existing pul-monary disease who was admitted 18 hours afterthe onset of peritonitis resulting from perforatedsigmoid diverticulitis. The chart illustrates theevolution of the vital signs and blood gases as

observed during the first eight days of illness. Onadmission the patient was suffering from a severe

electrolye and plasma disturbance. This was rap-

idly corrected with an improvement in the bloodpressure to a normal level, but the rapid pulse andrespirations persisted. X-ray examinations of thechest and abdomen on the day of admission (Fig.3a, 3b) showed marked elevation of the diaphragmand basilar atelectasis without other pulmonarydisease. The abdominal film shows considerablesmall bowel distention. Blood gases obtained on

the day following admission with the patientbreathing room air showed a reduced oxygen satu-ration (92%o), a markedly elevated PaCO, (78mm. Hg), a pH of 6.95 resulting from a mixedmetabolic and respiratory acidosis. The BUN atthis time was 111 mg.%. The patient had beenanuric on admission and was oliguric at the timethe arterial blood sample was drawn. She was

anxious, dyspneic and appeared exhausted, withrapid, shallow respirations.A tracheostomy was performed immediately and

a cuffed silver tracheostomy tube inserted into thetrachea. The patient was placed on artificial ven-

tilation using a mechanical respirator. The pulseand respirations promptly dropped to normal rates.Arterial blood studied at this time with the pa-tient breathing oxygen showed an oxygen satura-tion of 100%, PaCO2 in the normal range (46 mm.Hg) with a C02 content (19 mEq./L.) and pH(7.2) indicating a metabolic acidosis secondaryto renal failure. In the following four days normalrenal function resumed and the metabolic acidosiswas corrected. On the eighth day of illness, withthe patient off the respirator and breathing room

air, all studies were within normal limits. (Oxygen

saturation 96%, PaCO2-39 mm. Hg, pH-7.43,CO,-30 mEq./L., BUN-9 mg.%.)

Case 1 serves to demonstrate the rapidityand severity of respiratory failure develop-ing in a patient whose pulmonary functionhad been normal before the onset of theperitoneal insult. The case also demon-strates the rapid and complete correction ofthe abnormalities caused by respiratoryfailure by adequate ventilation as accom-

plished by the mechanical respirator.7' 8, 10,

24, 35, 37

Case 2. A. W. An examination of the case his-tory shown in Figure 4 illustrates, in our opinion,

FIG. 3. Case 1. X-ray studies on admission. A.Chest x-ray showing minimal atelectasis. B. Scoutabdominal film showing dilated bowel.


588 BURKE, PONTOPPIDAN AND WELCH

A.W d70 TOTAL WOUND TRACHEOSTOMYCYSTECTOMY DEHISCENCE RESP1169141/

0WOUND INFECTION--

RArE 30

20, e

120

RArE 10O02 St-89% 98%

// P.C02 * 28 mm 45 mm

pH 7.56

CO2 Con .25mEq/L

srsroL /C a. E ,p

LLDAYS 2 3 4 5 6 7 a 9 10 I11 12

COM0PENSATED DECOMPENSATED

INCREASING FATIGUE

FIG. 4. Case 2.

the fatiguing effect of an extensive operative pro-cedure followed by multiple complications. Thepatient was a 70-year-old man with no history ofprevious pulmonary or cardiovascular disease. Ad-mission physical examinations showed no abnor-malities in these systems. Following total cystec-tomy the patient did well for the first three post-operative days. A wound infection developed andwas drained. He became distended and the wounddehisced and was resutured on the seventh post-operative day. The following day his clinical con-dition was improved but the second day followingdehiscence (9th postoperative day) he was restlessand anxious with a respiratory rate of 40/min.Arterial blood studies with the patient breathingroom air showed °2, saturation 89%, PaCO2 28mm. Hg, C02 25 mEq./L. and a pH of 7.56,representing an early phase of respiratory failure(mild, uncompensated respiratory alkalosis). In anattempt to maintain an adequate arterial 02 satu-ration the patient was hyperventilating and blow-ing off C02. Tracheostomy and artificial ventila-tion was begun using a mechanical respirator.The patient's clinical condition improved andblood gas studies and pH returned to normal (02saturation 98%, PaCO, 45 mm. Hg ). Althoughthis patient's clinical course before respiratory fail-ure was complicated, all of the complicating fac-tors except the increasing fatigue of a long illnesswere controlled before the onset of respiratoryfailure. It seems clear that exhaustion played amajor role in creating an intolerable respiratorywork load.

Case 3. J. M. The ultimate effects of respira-tory failure lead to cardiovascular failure and car-diac arrest. Figure 5 illustrates this chain of eventsin a 62-year-old man who was admitted with aruptured abdominal aortic aneurysm. In retrospect,he had mild emphysem-a on admission. Following


operative repair he did well and was in excellentclinical condition on the first postoperative day.The pulse and respiratory rate rose steadily on thesecond postoperative day and massive distentionwas noted. On the evening of the second post-operative day the patient was disoriented andcyanotic. On the third day pulse and respirationsremained elevated but the blood pressure con-tinued at a normal level. The patient was flushed,cyanotic, restless, confused and thrashing about inbed. Cardiac arrest occurred 72 hours postopera-tively. Open cardiac massage and artificial venti-lation were immediately begun. A PaCO2 imme-diately following tracheostomy was 60 mm. Hg.The patient was resuscitated. Following mechani-cal respiration the arterial 02 saturation was 97%and the PaCO2 was 35 mm. Hg. Adequate oxygensaturation and CO2 elimination were maintainedby a period of continued artificial ventilation andthe patient left the hospital well on the 24th post-operative day. The vital signs as plotted illustratethat the patient was able to compensate for theincreased energy demanded by the severe peri-toneal insult and the decreased efficiency of themechanics of respiration on the first postoperativeday. However, as distention increased, the me-chanics of respiration became more inefficient andthe patient became exhausted. He was not ableto maintain adequate gas exchanges and cardiacarrest followed.

DiscussionIn order to understand the genesis of this

relative respiratory failure, it is necessaryto examine the demands made on the pa-

RUPTURED CYANOTICAORTIC ANEURYSM

J. M. L7" 6 OPERATIVE CARDIAC ARREST*1094269 REPAIR { TRACHEOSTOMY, ART RESP

50-

RFSPIRA TORY 4 0 -

RATE30'20

160, _

140- KPULSE 020RATE 120

100

80'-

170-

SYSTOL/CRP 140

110

DAYS 2 3 4 5 6 7L__ __ I~COMPENSATED DECOMPENSATED

FIG. 5. Case 3.


tient's respiratory system following a severe

peritoneal injury and the patient's abilityto adapt to these demands. The immediateproblem posed to the patient arises fromtwo major trends, both working to the pa-

tient's disadvantage. First, the complexprocess of inflammation, including fever,27requires a considerable increase in energy

expenditure,4'15,17, 26 forcing the patient tocompensate, using a proportional amountof his functional respiratory reserve; andsecond, the patient's total ability to com-

pensate, that is, the normal extent of hisreserve, is decreased by the intraperitonealinsult.3' 22, 31 Satisfactory compensation istherefore possible only if the decrease inreserve does not lower the patient's totalrespiratory ability below the level requiredby the increased energy demands. A third,and important, factor of time must be in-troduced. The loss of reserve tends to beprogressive as the patient becomes ex-

hausted and it is obvious that the energydemands may also be progressive. Thus, thepatient may be able to compensate duringthe first days following peritoneal insultonly to succumb to respiratory failure atfour or five days because the relationshipsof reserve and energy demand shift as timepasses.

Figure 6 is a schematic diagram illus-trating the relation between the total re-

spiratory ability and the required workload under three different circumstances.In normal activity the required work loadtakes up only a portion of the total respira-tory ability, leaving a large reserve. In vig-orous exercise the required work load in-creases but the total respiratory ability isunchanged, leaving a smaller but signifi-cant reserve. Following severe peritonealinsult the required work load increases and,in addition, the total respiratory ability de-creases. Decompensation takes place whenthe required work load is greater than thetotal respiratory ability. This diagram illus-trates two important points: 1) that thistype of respiratory failure is a combination

HIGH OUTPUT' RESPIRATORY FAILURE

NORMALACTIVIrY VIGOROUS EXERCISE (ERITo#/NITErc

FIG. 6.\Schematic diagram ofhighoutputreompeoaior

ofRansincrveased demand andadeReearvd

tota resiraorkyoa Required woknod 2)kloodth

Respirveory mechanisms, even thoughwi

FiG.e6. higheoutptidigamlure High output respr-spira~~~toryfailure,nbif s h nblt

tof sanifhincreaseddemnquira erentsedo

desptarespiatorintablityinrased respiratoryerespirator. ehnss vntog alin flow-sheraietotheinbloo adequationshi are

functioninthe majohighersthaninoratoewhati.e. havhigh eouptfiuehigh outputre-prtrsiaoyfailure,iie in briguef is The chinabiittoens atsf othiedicraethequirwsementis for

tissult oxygenationandinlaterorCxretionsdespteory iepnitiallyinceasviedrsiratoreffort.clad ehnclfatr,btAednflwshe outlninthe reatienonshipre-

spiratofryh mailure factrsicading thewhatfiammavetoryrnedinpouehigh,oturespiatorfailur is gienpraion.Fgre7AcThe2 charincu-evntaarotlnein the flow-obs,sheetfolos.a

taeshance.Thineficencts ofcrease inflan-mtrapumonresposeuntin beusedviyedhentobichemicalclapse mechanicalyt facors,perothspirthory failupre.MechInically,this amontstivlconixdealdiaphragmsency. considerable

23eftbhilh i

pano eprto.Aeetss atclal nteloe oe,sonflosMechanialineficieny is icrease byintraplmoaryhuningcausd b th me

FiG.ou6. Schemai diareamh. otput

responirat ry bme chanef is ,e ven31 fadi l-o

spirtorye efailur, inbriocef,isalytheinaility


590 BURKE, PONTOPPIDAN AND WELCH

PERITONITIS

BIOCHEMICAL - MECHANICALINFLAMMATION & DISTEPITIN

INCREASED\TISSUE DEMAND HISH DIAPHRAGG,PAIN

FOR 02

INCKASED DEND ON INCREASED -

CARDO)RESPIRTATORY MUSCULAR ATELECTASISSYSTEM WORN INTNAPULNONARY SHUNTIG

FATIGUEINIEFFKIENT REPIRATORY MECHANC

HYPOXIAMETABOLIC & RESPIRATORY ACIDOSIS'

DEATH

FIG. 7. Flow-sheet showing evolution of patho-logic events in high output respiratory failure.

spread inflammatory reaction leads to an

increased tissue utilization of oxygen.4' 15,17, 26 This increased utilization leads to an

increased demand on the cardiorespiratorysystem 13, 25 and an increased demand formuscular work."' Increased muscular workleads to a further increase in oxygen utiliza-tion,11' 34 thus establishing a vicious circlerequiring more and more oxygen transportand CO, elimination. In time this workload leads to fatigue, and fatigue, exag-

gerated by the inefficient mechanisms ofrespiration, creates a situation where theability to ventilate is not sufficient to pro-

vide the necessary amount of gas exchangefor the greatly increased amount of energy

expended. This decompensation causes tis-sue hypoxia and metabolic and respiratoryacidosis. If these are not promptly reverseddeath results.The clinical features associated with high

output respiratory failure may be dividedinto premonitory features, early signs andsigns of frank decompensation. The pre-

monitory features are those factors thatreduce the total respiratory ability of thepatient before the onset of peritoneal in-sult. These include 1) previous pulmonarydisease; 2) obesity; 3) heavy cigarettesmoking; and 4) severe debility. The earlysigns of developing respiratory failure are


1) persistent rapid pulse and respiratoryrate; 2) normal or slightly elevated bloodpressure; 3) increased arterio-venous 02difference; and 4) decreased PaCO2 dueto hyperventilation in an attempt to main-tain adequate arterial 02 saturation. Signsof frank decompensation are 1) arterialoxygen saturation below that usual for thepatient studied (since an occasional pa-tient, as for example one with severeemphysema, may function with a lowerthan normal 02 saturation, an 02 satura-tion below normal alone does not confirma diagnosis of this type of respiratory fail-ure); 2) metabolic acidosis where causesother than tissue hypoxia of respiratoryorigin have been ruled out; and 3) respira-tory acidosis. In addition to these signs,which may be documented by precisemeasurement, the clinical appearance andprogressive course of the patient are of ut-most importance. As outlined in the aboveflow-sheet, fatigue is of considerable mo-ment in the development of this syndrome.Although respiratory failure should be doc-umented by the determinations listedabove, a high degree of suspicion of im-pending decompensation should be gen-erated by the patient who has had a severeperitoneal insult, a persistent rapid pulseand respiratory rate in the face of a normalor slightly elevated blood pressure andwho is obviously tiring. His face may beflushed, he is anxious, restless and may bethrashing. Respirations are rapid, shallow,labored and, at times, grunting.Decompensation via high output respira-

tory failure seldom occurs before the sec-ond day following peritoneal insult andmost often is seen on the fourth or fifthday. In order to prevent decompensationthese patients must be followed with fre-quent evaluation of the adequacy of ar-terial oxygenation and CO2 elimination.The most exact methods 1, 2 involve theexamination of arterial blood. PaCO2 alonecan be conveniently measured by the re-breathing technic.14, 38 It is necessary to

Volume 158 HIGH OUTPUT RESPIRATORY FAILURE 591Number 4

emphasize that the standard blood gasstudies (02 saturation, PaCO2, CO2 con-tent or standard bicarbonate and pH) areneeded to confirm the presence and deter-mine the extent of decompensation.When decompensation is imminent its

progression almost always can be inter-rupted by measures that improve themechanics of respiration, i.e., relief of in-tra-abdominal tension by gastro-intestinalintubation or operative drainage and de-compression of the gastro-intestinal tractor peritoneal cavity, and with efficient pul-monary physiotherapy. Nasal 02 improvesarterial oxygen saturation but the resultingabsence of cyanosis may mask a progres-sive respiratory acidosis. If these measuresfail the intolerable work load of respirationmust be relieved by tracheostomy and arti-ficial ventilation using a mechanical respira-tor. In addition to these essentials of spe-cific therapy, it is important to maintainother adjuvant therapy that must includeantibiotics, blood, plasma and electrolytes.

SummaryHigh output respiratory failure is defined

as that type of respiratory failure charac-terized by inability to provide adequate tis-sue oxygenation and later carbon dioxideexcretion despite an initially increased gasexchange. In the past few years, 21 pa-tients have been studied and treated inthe Massachusetts General Hospital withthis type of failure associated with severeperitonitis or ileus. There were five deathsin this series including three due to renalfailure, one to pulmonary embolism, oneto ventricular arrhythmia. No death wasdue to high output respiratory failure sinceall deaths occurred from two weeks to twomonths after it had been relieved.

It is vitally important that incipient re-spiratory failure be recognized by the sur-geon who is treating a patient with peri-tonitis or ileus. He should be alerted pre-operatively by any previous lung disease,obesity, heavy smoking or severe debility.

The syndrome of high output respiratoryfailure may be suspected by persistent ele-vation of pulse and respiration with a nor-mal blood pressure in an anxious, exhaustedpatient. It is confirmed by a lower arterialoxygen saturation than is normal for thepatient. (Thus, in the presence of normallungs an arterial oxygen saturation of 95per cent or less must be regarded as evi-dence of decompensation.) Prevention ofdecompensation often can be effected byabdominal decompression or drainage, aswell as chest physiotherapy. If these failimmediate institution of tracheostomy andassisted respiration with a mechanical res-pirator will result in normal gas exchangeand relief of this type of failure.

References1. Andersen, 0. S. and K. Engel: A New Acid-

base Nomagram. An Improved Method forthe Calculation of the Relevant Blood Acid-base data. Scand. J. Clin. Lab. Invest., 12:177, 1960.

2. Andersen, 0. S., K. Engel, I. Jorgensen andP. Astrup: A Micro Method for Determina-tion of pH, Carbon Dioxide Tension, BaseExcess and Standard Bicarbonate in Capil-lary Blood. Scand. J. Clin. Lab. Invest., 12:172, 1960.

3. Anscombe, A. R.: Pulmonary Complicationsof Abdominal Surgery. Yearbook Publishers,1957.

4. Artz, C. P.: Newer Concepts of Nutrition bythe Intravenous Route. Ann. Surg., 149:841,1959.

5. Astrup, P., K. J0rgensen, 0. S. Andersen andK. Engel: The Acid-base Metabolism. ANew Approach. Lancet, 1:1035, 1960.

6. Bendixen, H. H., J. Hedley-Whyte and M. B.Laver: Impaired Oxygenation in SurgicalPatients During General Anesthesia withControlled Ventilation-a Concept of Ate-lectasis. (Submitted for Publication.)

7. Bjork, V. 0. and C. G. Engstrom: Treatmentof Ventilatory Insufficiency after PulmonaryResection with Tracheostomy and ProlongedArtificial Ventilation. J. Thoracic Surg., 30:356, 1955.

8. Bjork, V. 0. and C. G. Engstrom: Treatmentof Ventilatory Insufficiency by Tracheostomyand Artificial Ventilation: Study of 61 Tho-racic Surgical Cases. J. Thoracic Surg., 34:228, 1957.

2BURKE, PONTOPPIDAN AND WELCH Annals of Surgery592 BUK,October 1963

9. Blackburn, H., J. Brozek and H. L. Taylor:Lung Volumes in Smokers and Non-smokers.Ann. Int. Med., 51:68, 1950.

10. Boyd, A. D., R. E. Tremblay, F. C. Spencerand H. T. Bahnson: Estimation of CardiacOutput soon after Intracardiac Surgery withCardiopulmonary Bypass. Ann. Surg., 150:613, 1959.

11. Campbell, E. J. M., E. K. Westlake and R. M.Cherniack: Simple Methods of EstimatingOxygen Consumption and Efficiency ofBreathing. J. Appl. Physiol., 11:303, 1957.

12. Chemiack, R. M. and L. Cherniack: Respira-tion in Health and Disease, Philadelphia,W. B. Saunders Company, 1961.

13. Clowes, G. H. A., Jr., A. Alichniewicz, L. R.NI. DelGuercio and D. Gillespie: Relation-ship of Postoperative Acidosis to Pulmonaryand Cardiovascular Function. J. Thoracic &Cardiovas. Surg., 39:1, 1960.

14. Collier, C. R.: Determination of Mixed VenousCO2 Tension by Rebreathing. J. Appl.Physiol., 9:25, 1956.

15. Cope, O., G. L. Nardi, Nl. Quijano, R. L.Rovit, J. B. Stanbury and A. White: Meta-bolic Rate and Thyroid Function FollowingAcute Thermal Trauma in Man. Ann. Surg.,137:165, 1953.

16. Cournand, A., D. W. Richards, R. A. Bader,NI. D. Bader and A. P. Fishman: The Oxy-gen Cost of Breathing. Trans. Assoc. Amer.Physicians, 67:162, 1954.

17. DuBois, E. F.: Basal Metabolism in Healthand Disease. Lea and Febiger, 1927.

18. Flint, F. J. and A. J. N. Warrack: AcutePeptic Ulceration in Emphysema. Lancet, 2:178, 1958.

19. Fulton, R. M.: The Heart in Chronic Pulmo-nary Emphysema. Quart. J. Med., 22:43,1953.

20. Green, P. T. and J. C. Dundee: The Associa-tion of Chronic Pulmonary Emphysema withChronic Peptic Ulceration. Canad. M.A.J.,67:438, 1952.

21. Hagen, E. B. and J. P. Mlerrill: PeritonealDialysis and Acute Renal Failure. Surg. Cl.N. Amer. June, 1963.

22. Hamilton, W. K. and J. C. Devine: TheEvaluation of Respiratory Adequacy in theImmediate Postoperative Period. Surg.,Gynec. & Obst., 105:229, 1957.

23. Harris, T. A. B., R. D. Dripps, NI. V. Demingand J. F. Goodwin: Study of PostoperativeAtelectasis. Brit. J. Surg., 36:256, 1949.

24. Holmdahl, M. H.: The Respiratory Care Unit.Anesthesiology, 23:4, 559, 1962.

25. Hood, R. M. and A. C. Beall, Jr.: Hypo-ventilation, Hypoxia and Acidosis Occurringin Acute Postoperative Period. J. ThoracicSurg., 36:729, 1958.

26. Kinney, J. M.: A Consideration of EnergyExchange in Human Trauma. Bull. N. Y.Acad. Med., 36:617, 1960.

27. Kinney, J. M. and C. F. Roe: Caloric Equiva-lent of Fever. Trans. Am. Surg. Assoc., 80:282, 1962.

28. Latts, W. M., J. F. Cummins and L. Zieve:Peptic Ulcer and Pulmonary Emphysema.Arch. Int. Med., 97:576, 1956.

29. Litwin, M. S., F. G. Panico, C. Rubini, D. E.Harken and F. D. Moore: Acidosis and Lac-ticacidemia in Extracorporeal Circulation.Ann. Surg., 149:188, 1959.

30. Lowell, F. C., W. Franklin, A. L. Michelsonand I. W. Schiller: A Note on the Associa-tion of Emphysema, Peptic Ulcer and Smok-ing. New Engl. J. Med., 254:123, 1956.

31. Mastico, G. J. and F. F. Allbritten: RespiratoryFunction of the Postoperative Patient. Arch.Surg., 76:732, 1958.

32. MIoore, F. D.: Personal communication.33. Morton, H. J. V.: History of Smoking and

Pulmonary Complications after Operation.Lancet, 1:368, 1944.

34. Otis, A. B.: The Work of Breathing. Physiol.Rev., 34:449, 1957.

35. Pontoppidan, H.: Postoperative VentilatorySupport. Anesthesia and Analgesia, 41:769,1962.

36. Silen, W., W. H. Brown and B. Eiseman:Peptic Ulcer and Pulmonary Emphysema.Ann. Surg., 78:897, 1959.

37. Spencer, F. C., D. W. Benson, W. C. Liv andH. T. Bahnson: Use of Mechanical Respira-tor in Management of Respiratory Insuf-ficiency Following Trauma or Operation forCardiac or Pulmonary Disease. J. Thoracic& Cardiovas. Surg., 38:758, 1959.

38. Stein, NI. and C. R. Colp: Simple Method forthe Measurement of Alveolar Carbon Diox-ide. J.A.M.A., 173:671, 1960.

39. Welch, C. E.: Treatment of Combined In-testinal Obstruction and Peritonitis by Re-functionalization of the Intestine. Ann. Surg.,142:739, 1955.

high output respiratory failure: *

Documents