Reprinted from ANNALS OF INTERNAL MEDICINE, Vol. 67, No . 3, Part 1, September, 1967Printed in U . S . A .

Controlled Ventilation with Intermittent Positive-

Pressure Breathing in the Management o Acute

Ventilatoov Failure Associated with Chronic

Obstructive Pulmonary Disease

EARLS B. WEISS, M .D., and MAURICIO J. DULFANO, M.D.Boston, Massachusetts

C

ONTROLLED MECHANICAL VENTILATIONmay be employed for ventilatory sup-

port in patients with respiratory failurewhen assisted ventilation fails . The purposeof controlled ventilation is to reduce pa-tient respiratory control and effort and tosubstitute adequate artificial ventilation,thereby providing time for appropriatemedical therapy . With improvement the pa-tient's own ventilatory effort may be capa-ble of maintaining normal or clinicallystable balance . These principles have beenadequately demonstrated in acute ventila-tory failure with volume-cycled respirators(1, 2). It was the purpose of this prospectivestudy to evaluate in patients with chronicobstructive lung disease, a group usuallydifficult to manage during acute ventilatoryfailure, the following : [1] whether conven-tional pressure-cycled intermittent positive-

Received April 10, 1967 ; revision accepted May29, 1967 .

From the Department of Medicine, Tufts Univer-sity School of Medicine, and the Lung Station(Tufts) and 1 and 3 Medical Services (Tufts), Bos-ton City Hospital, Boston, Mass .

This study was supported in part by grants fromthe Pittsfield Anti-Tuberculosis Association, Pitts-field, Mass., and from The Council for TobaccoResearch, New York, N. Y.Dr. Weiss was supported in this study by post-

doctoral fellowship 5-F2-HE-23, 300, NationalHeart Institute, National Institutes of Health,Bethesda, Md .Requests for reprints should be addressed to

Earle B . Weiss, M .D., Lung Station (Tufts), BostonCity Hospital, 818 Harrison Ave ., Boston, Mass .02118 .

556

pressure breathing (IPPB) could achieve ef-fective controlled ventilation under bedsideconditions ; [2] at what point this procedureshould be substituted for conventional as-sisted IPPB; and [3] the sequential stepsnecessary to achieve this goal .

MATERIALS AND METHODSThis report is based on 21 ventilatory

studies performed at the bedside on 19 pa-tients with chronic obstructive lung diseasein acute ventilatory failure. Major pulmo-nary diagnoses (Table 1) included asthmaticbronchitis, chronic bronchitis, cor pulmo-nale, obstructive emphysema, and tubercu-losis . In addition, medical complications ofpneumonia, cerebrovascular problems, car-diac disease, obesity, bacteremia, gastro-intestinal abnormalities, and shock charac-terized the type of patient studied .The presumptive causes for ventilatory fail-

ure are included in Table 1 . Acute respiratoryinsufficiency was judged clinically and con-firmed by arterial blood gas measurements(Pacoz > 55 mm Hg, pH < 7 .25, and Pao2 < 60mm Hg, as acute changes) . Constant conditionswere maintained during the trial period .Tracheal suctioning was permitted, but theaddition of mucolytic, bronchodilator, or wet-ting agents was deferred until the best controlcycles were established . Periodic brief deepbreathing was permitted .Serial arterial blood gases were drawn via a

Cournand needle at 15- to 30-min intervalsuntil optimal ventilation was established andfrequently thereafter to prevent hyperventila-tory alkalosis . Once relative stabilization of

557

Pco2 and pH occurred, serial gases were drawnat I- to 2-hr intervals to facilitate adjustmentof the respirator. All tidal volumes were mea-sured at the exhalation port with a MonaghanVentilation Meter . Initial tidal volumes werearbitrarily selected to be 500 cc to 700 cc .Arterial blood was obtained anaerobically

and immediately analyzed in the IL instru-ment (Model 102) for Poe, Pco2, and pHat 37 C. Blood exposed to known gas concen-trations (Scholander) was tonometered in awater bath at 37 C for Po e calibration . Drygas was used for the Pco2 calibration slope .All duplicate samples were required to checkwithin 5 mm for Poe (50- to 100-mm range),2 mm for Pco2, and 0.005 units for pH. Oxy-gen saturation was calculated from the Poeand the pH using the standard dissociationcurve for oxyhemoglobin at 37 C and 7 .40 pH .Plasma CO 2 content and bicarbonate were cal-culated from the measured Pco2 and pH bymeans of a standard nomogram based on theHenderson-Hasselbalch equation .The sequence of approach is presented be-

low. A period of 1 hr in which Pco2 failed tofall more than 5 mm Hg was accepted as venti-latory failure. Failure of one method was theindication to move to the next .

IMPROVED ASSISTED VENTILATION

Every attempt was made to improve con-ventional assisted respiration. In all cases, atracheostomy or an endotracheal tube was estab-lished and ventilation performed by the BirdMark 7 or 8 intermittent positive-pressure res-pirator with 40%0 oxygen setting and, occa-sionally, 100% oxygen. It should be noted thatthe 40% dial value represents the mixture dialon the IPPB unit only and does not representsuch concentrations. Actual concentrations of in-spired oxygen were not measured . The reasonsfor failure of assisted IPPB were judged to be[1] progressive clinical obtundation despite as-sisted IPPB therapy ; [2] failure to reduce or toprevent a rise in Paco2 on assisted IPPB ; [3]patient uncooperative or agitated, refusing toaccept or phase with respirator; [4] persistenttachypnea, physical exhaustion, obvious exces-sive work of breathing; and [5] patient becom-ing obtunded on oxygen therapy .

CONTROLLED VENTILATION

Controlled ventilation was instituted in thefollowing sequence when assisted ventilationfailed .Machine Cycle : With the sensitivity at 40 -}-,

the IPPB was set for automatic cycling of 10

E . B . WEISS AND M. J . DULFANO Annals ofInternal Medicine

to 15/min, a flow dial setting of 15, and pres-sure of 15 cm H2O. A period of 5 to 7 min waspermitted for synchronization to occur. If suc-cessful, the pressure, flow, and rate dials werereset for effective ventilation, confirmed byarterial blood .100% Oxygen Suppression with Machine

Automatic Cycle : Since ventilation in patientswith chronic hypercapnia may be depressed byoxygen administration, 100% oxygen was givenby assisted IPPB or tracheostomy box untilhypoventilation ensued . A limit of 5 to 7 minwas set for this to occur. If successful, presetautomatic machine cycle was then instituted asin Machine Cycle, using only air mixtures .Drug Suppression and Machine Automatic

Cycle : Before drug use the airway was checked,the IPPB was preset and ready to ventilate,and a bag respirator was at the bedside incase of mechanical failure . Either morphine orsuccinylcholine or both were used arbitrarily :morphine sulfate-3 to 5 mg, intravenously,with total dose up to 10 to 20 mg as neededto reduce agitation and produce muscle re-laxation ; then machine cycle as describedabove, with 2 to 4 mg at intervals as required ;succinylcholine-20 to 40 mg, intravenously,as a stat dose, repeated as needed to desiredlevel and followed immediately with machinecycle .

Two phases of controlled ventilation aftersynchronization were delineated .

Hyperventilation Period : This was the phaseof acute hyperventilation required to lowerPaco2 and [H*] and improve Pao2. The cri-teria for improvement were [1] patient nolonger agitated or uncooperative and able tocycle with ventilator; [2] adequate tidal vol-ume on the respirator, even with oxygen ad-ministration ; [3] improvement in blood gasparameters; [4] reduction in gross ventilatoryeffort in those with severe obstruction andobvious excessive work of breathing; and [5]disappearance of cyanosis .

Maintenance Period : Herein was that phaseadministered supportively to maintain ade-quate alveolar ventilation (as defined byPco2 = 40 to 60 mm Hg; pH = 7.30 to 7 .45 ;Poe > 70 mm Hg) while associated medicaltherapy was continued.

RESULTS

ANALYSIS OF CONTROLLED VENTILATION AP-PROACH

Of the 21 instances requiring ventilatorysupport 4 were managed by assisted venti-

Volume 67, No . 3September 1967

T-7, chronic pyelonephritis,hypertension

MANAGEMENT OF ACUTE VENTILATORY FAILURE

TABLE 1 . Clinical Data*

558

* CB = chronic bronchitis ; CHF = congestive heart failure ; CVA = cerebrovascular accident ; CAD = coro-nary artery disease ; CPE = chronic pulmonary emphysema ; TBC = tuberculosis ; PVCs = premature ventricularcontractions ; IPPB = intermittent positive-pressure breathing ; AMI = acute myocardial infarction; RHD = rheu-matic heart disease ; MS = mitral stenosis ; PAT = paroxysmal auricular tachycardia .

t Hyperventilation phase only .

Patient Age Sex Diagnosis Cause of Acute VentilatoryFailure

Paco2-pH-Pao2t

Initial End

Assisted ventilationyr

M. H . 47 M Asthmatic bronchitis, CO2 retention, broncho- 110-7 .18-283 67-7 .42-142

M. K . 80 F

bronchiectasis, CO2narcosis, alcoholism

CB, pulmonary fibrosis, pleural

spasm, and secretions

Pneumonia, pleural effusion, 90-7,24-64 68-7 .36-67

F. Bu .

effusion, pneumococcal pneu-monia, CHF secondary to corpulmonale, CVA, hyperten-sion, CAD

CB, CPE, TBC (inactive),

CHF

Pneumonia, secretions, 60-7 .30-99 46-7.39-9762 M

E. P . 2 56 F

pneumococcal pneumonia,CO2 narcosis, cirrhosis,duodenal ulcer, CAD

CB, CPE, bronchiectasis, cor

Demerol®

Pneumonia, secretions 129-7 .12-178 65-7 .33-310pulmonale, atelectasis, pneu-monitis (Klebsiella-Aero-bacler), CO2 narcosis, CVA,acute myocardial infarction,septicemia

Controlled ventilationA. Machine cycleF. Ca . 65 M CB, pneumonia, obesity, CO2 Secretions, poor cough, oh- 110-7 .18-154 78-7 .31-180

E. P ., 56 F

narcosis, cor pulmonale,CAD, CHF and PVCs,gastric dilatation, ? pulmo-nary emboli

See Patient E . P .2, acute

tunded, tachypnea, failureto cycle with IPPB

Cardiac arrest, apnea, 63-7 .34-90 43-7 .52-288

P. Ha. 87 M

bronchitis, CO2 narcosis,CAD and AMI, cardiacarrest, shock, CVA

CB and pulmonary fibrosis,

secretions, exhaustion

Cardiac arrest, secretions,

(receivedNaHCO3)

98-7.05-81 37-7 .34-311

F. C .

acute bronchitis, CAD andCHF, old CVA, ? emboli,hypotension, cardiac arrest

Pulmonary TBC, CB, CPE, cor

shallow and rapidrespiration

Tachypnea, secretions, 86-7.24-76 49-7.48-21265 M

P. Hu . 65 M

pulmonale, pneumonitis,CO2 narcosis

CB-CPE-staphylococcal pneu-

somnolent

Secretion, drugs Not available 56-7 .41-500

A. M ., 59 M

monia, sedation (Librium®,barbiturates), CO2 narcosis,cor pulmonale, dehydration

Asthmatic bronchitis, CPE, Pseudomonas septicemia and 92-7 .24-50 54-7 .46-260TBC (?activity), CO2 nar-cosis, adrenal insufficiency,osteoporotic fracture T-6 and

pneumonia, secretions,tachypnea, exhaustion

559 E . B . WEISS AND M. J . DULFANO

TABLE 1-(Continued)

Annals ofInternal Medicine

Patient Age Sex Diagnosis Cause of Acute VentilatoryFailure

Paco2-pH-Pao2t

Initial End

yrA. Mu. 66 F CB, pneumococcal pneumonia, Cardiac arrest, secretions, 33-7.44-57 40-7 .35-220

B .

cardiac arrest, obesity, shock,renal failure, CHF, CVA,? lactic acidosis

and machine cyclePneumonia, CB and CPE, TBC

hypoxemia

Pneumonitis, secretions, 02 99-7.16-34 57-7.28-380Oxygen

J. 0 .depression68 M

;. G . 58 M

(active), pleuritis, bronchi-ectasis, CO2 narcosis, gastriculcer, RHD and MS, PATwith block, gastrointestinalbleeding, cirrhosis,bacteremia, dehydration

CB and CPE, TBC (active),

sensitive, obtunded

Secretions, tachypnea, CHF 91-7.26-59 86-7.31-132

C. S . 75 M

pneumococcal pneumonia,CO2 narcosis, CAD, AMIand CHF, pulmonary emboli,cardiac arrest, alcoholism

Asthmatic bronchitis, pneumo- Inspissated secretions, 67-7.24-33 58-7.33-141

C . MorphineJ. R .

sulfate55 M

coccal pneumonia, corpulmonale, old TBC, CO2narcosis, dehydration,oliguria, CHF, alcoholism

and succinylcholineAsthmatic bronchitis, pneumo-

agitation, bronchospasm

Fatigue, secretions, agita- 88-7.12-173 58-7.26-101

D . MorJ. L .

coccal pneumonia, asbesto-sis, cirrhosis, dehydration

tion, bronchospasm,tachypnea

Intrinsic disease, secretions, 77-7 .24-93 39-7.44-69ine sulfate70 M CB and CPE, pneumococcal

M. S . 60 F

pneumonia, cor pulmonale,CAD, CHF, alcoholism,dehydration

Asthmatic bronchitis, CO 2

agitation, tachypnea,uncooperative

Pneumonia, secretions, 118-7.21-62 62-7.34-80

A. M .2 59 M

narcosis, pseudomonas pneu-monia, obesity, pulmonaryemboli, chronic pyelone-phritis, sigmoid-vesiclefistulae, bacteremia, shock

See A . M.,, lung abscess

agitation, tachypnea,uncooperative

Asthma, inspissated secre- 76-7.31-71 49-7.38-400

R. M . 49 F Asthmatic bronchitis, pneumo-

tions, exhaustion, CO2narcosis, agitation

Agitated, uncooperative, 110-7.18-275 56-7.35-200

E .T. W .Succinylcholine

40 Fand

coccal pneumonia, CO 2narcosis, obesity, CHF, corpulmonale

machine cycleAsthmatic bronchitis, pseudo-

secretions

Exhaustion, secretions, 140-6.99-57 62-7.25-57

S. W . 17 F

monas pneumonia, CO2narcosis, duodenal ulcer,cirrhosis, bacteremia, shock,anuria

Doriden® ingestion,

coma, poor synchroniza-tion with IPPB

Coma, tachypnea, poor 88-7.22-55 52-7.38-90pneumonia, CB synchronization with IPPB

Volume 67, No. 3September 1967

lation alone. In the remaining 17 instances7 (41%) could be cycled simply by machinesynchronization. In the remaining 10 cases,where oxygen control was attempted, only3 (18%) suppressed to the point where ma-chine cycle could be employed. This lim-ited response may be due to the brief timeallotted for depression with 100% oxygen .The remaining 7 (410/,,), after failing to re-spond to machine cycle or oxygen suppres-

required drugs to achieve optimalcoordinated respirator effect: 4 with mor-phine sulfate, 2 with suc ,,-inylcholine, and1 employing both .

BLOOD GAS DATA-HYPERVENTILATION PHASE

Blood gas data are presented in Table 2 .Initial data represent the period where con-

s, assisted IPPB ventilation failedt before controlled ven

begun. End values represent thetion of purposeful hyperventithe establishment of the maintenance pe-riod. During the hyperventilation phase ofcontinuous assisted ventilation, the Paco2was lowered by a mean value of 35 .7 mmHg, and pH was improved by 0.17 pHunits. The mean blood gas changes in eachsubgroup of controlled ventilation revealeda general reduction in Paco2 and a rise inpH and Pao, by each modality attemptedfor each group. In the machine cycle groupone patient required correction of hypoxiaonly. One patient in the oxygen suppres-sion group failed to demonstrate any re-duction in Paco2. In cases with severeacidosis, end pH values are influenced byuse of NaHCO3 . Adequate oxygenationcould be provided by all methods. Analysisof mean falls in Paco 2 in mm Hg/min foreach modality suggests that the method ofsynchronization was unimportant.

FEATURES OF THE IPPB AND VENTILATORY EF-FECTS

In Table 3 pressure, flow, and respiratoryrate data with the tidal volumes before and

g the hyperventilation phase of the

MANAGEMENT OF ACUTE VENTILATORY FAILURE

termina-nd

560

particular group are presented . In general,a 67 to 2117, increase in tidal volume and a34 to 50% fall in respiratory rate (main-tained at 16 to 20/min) were necessary toobtain effective alveolar hyperventilation .However, minute ventilation changes were

iable, despite reduction in Paco 2 . Meand inspiratory pressures varied from 24

settings, from 7 .5to 22.8. The wide range of pressuresto 40 cm H20) and flow settings (5 to(relative dial settings, no : _') . N

,7 ute val de :indicates that great in ,I q.

'. "aria ionwas present and stresse

for peri-o( 7 ,'.c and personalized att. .

assisted groupr, with less end inspiratory driv-

pressures required .

tients were controlled only untileffective spontaneous respirations in con-

tion with assisted ventilation were pos-sible. In all groups hyperventilation wasterminated within 2 to 6 hr, being briefestwith the machine cycle and longest in thosesuppressed with succinylcholine .nance periods extendingwere employed until assisted ventilcould be resumed . The total time on con-

cycle ranged from 16 hr to 36 hr .

TRANSITION FROM CONTROLLED TO ASSISTEDCYCLE

Controlled ventilation was continued un-til clinical and blood gas data indicatedthat the patient could be managed by as-sisted IPPB. The transition was madeabruptly with blood gas monitoring. Allsurvivors returned first to continuous andthen to intermittent assisted ventilationuntil they could ventilate on their own .Generally, most patients could be placedon assisted ventilation on the first trial .Of the 16 patients on controlled ventila-tion 5 died during the procedure, leaving11 (69%), who were transferred to assisted

Improved Assisted

Pco2/min,mm Hg/min

* Numbers in parentheses indicate the range of values .

TABLE 2. Mean Arterial Blood Gas Data--Hyperventilation Phase*

Type of Ventilation

Machine Cycle Oxygen Suppression

Controlled

Morphine Sulfate (MS)

Succinylcholine (SC) MS andSC

Drug Control

C.n

Number of cases 7 2

Paco 2, mm HgInitial 97 .2 (60-129) 80.3 (33-110) 96.6 (67-99) 95 (76-118) 88114 (88-140)End 61 .5 (46-68) 50 .2 (37-78) 67,0 (57-86) 51 .5 (39-62) 57 (52-62)1 58Difference -35.7 (-14 to -64) -30.1 (+7 to -61) -29.6 (-5 to -42) -43.5 (-27 to -56) -570 (-36 to -78) -30

pHInitial 7 .25 (7 .05-7 .44) 7.22 (7,16-7 .26) 7 .23 (7 .18-7 .31) 7 .12

z

7 .21 (7 .12-7 .30) 7 .11 (6 .99-7 .22)End 7 .38 (7 .33-7 .42) 7 .41 (7 .31-7 .52) 7 .31 (7 .28-7 .33) 7 .38 (7 .34-7 .44) 7 .32 (7 .25-738) 7 .26Difference -0.17 (-0.09 to -0 .14) -0.16 (+0.09 to -0.29) -0.09 (-0.05 to -0 .12) -0.15 (-0.07 to'-0 .20) -0.21 (-0.16 to -0 .26) -0.14

P.* MM HgInitial 156 (64-283) 83 .6 (50-154) 42 (33-59) 125 (62-275) 56 (55-57) 173End 154 (67-310) 282 (180-500) 218 (132-380) 187 (69-400) 73 .5 (57-90) 0101Difference -2 (-141 to +182) 198,4 (26 to 230) 176 (73 to 346) +62 (-75 to +400) 17 .5 (0 to 35) -72

Hour to decrease 4 :15 2 :30 4:00 3:15 4 :30Paco2

Mean fall 0.14 0.16 0 .12 0 .21 0,15 0,11

Volume 67, No . 3September 1967

Type Ventilation

* Relative dial settings only .

ventilation and were .ned there-after. From this latter group 3 died variableperiods later while on a continuous orintermittent assisted schedule, leaving 8(50%Jo ) survivors, who were entirely weanedoff the respirator. In the 5 patients whodied, no attempt was made to switch toassisted cycle because of the lack of clinicalrecovery despite improvement of Po e andPco2-pH parameters .

COMPLICATIONS

Complications (Table 4) caused by theprocedure are difficult to evaluate becauseof associated diseases and therapy . A com-mon problem was the fall in arterial bloodpressure, temporarily related to high cy-cling pressures and often associated withgastrointestinal bleeding or shock . Arrhyth-mias and gastrointestinal bleeding werecommon. Falls in urinary output were ob-served in a few instances; but actual mea-surements were limited, and the true fre-quency is unknown .Comparison of initial Paco 2 and pH with

mortality rates revealed no significant dif-ferences for predicting outcome, and anal-ysis of mortality versus time on controlledcycle indicated that survivors were on con-trolled ventilation for more prolongedperiods .

MANAGEMENT OF ACUTE VENTILATORY FAILURE

TABLE 3 . Intermittent Positive-Pressure Breathing (IPPB) Dynamics During Hyperventilation : TidalVolume, Respiratory Rate, Minute Volume Changes, and IPPB Pressure-Flow

Observations (Mean Values)

Tidal Volume Mean Respiratory Rate Mean Change in Mean

IPPB SettiIn- Fall

Minute

Paco,Initial End

crease Initial End

Volume

Change Pressure Flow*

562

TABLE 4 . Complications During ControlledVentilation

Group

Complications*

Improved assisted Reversible hypotension (1), needledislodged into trachea (1) .

Controlled :

Machine cycle Arrhythmia ultifocal PVCs (1),atrial fibril ' (1), cardiac ar-rest (1), ven ricular tachycardia(1)], oliguria (1), gastric dilata-tion (1), transient cerebrovascu-larinsufficiency (1), hypotension[myocardial damage (2), res-pirator (2)], gastrointestinalbleeding (2), shock (1) .

Oxygen

Respirator hypotension (1), shocksuppression with ventricular tachycardia

(1), acute myocardial infarction(1).

Morphine Reversible respirator hypotension(4), arrhythmia (PAT withblock secondary to digitalisadministration) (1), gastroin-testinal bleeding (2), trachealbleeding (1), transient oliguria(1), aspiration pneumonia (1),hypokalemia (1) .

Succinylcholine Reversible respirator hypotension(1), gastrointestinal bleeding (1) .

Succinylcholine Reversible respirator hypotensionand morphine

(1), oliguria (1) .sulfate

* Numbers in parentheses = numbers of patiPVCs = premature ventricular contractions ;= paroxysmal auricular tachycardia .

cc/grain /o breaths/-vain ofo% mm 119 cm 1120

Improved assisted 300 500 67 33 19 .3 42 -2.5 -35.7 24.3 20ControlledMachine cycle 375 658 75 29 16 .3 44 -1.3 -30 .1 28 .5 22Oxygen suppression 260 433 67 32 21 .0 34 +9.3 -29.6 30 .7 21Drugs :Morphine 281 687 .5 144 32 .8 17 48 +26.0 -43.5 28 .3 77 .5Succinylcholine 225 700 211 34 17 50 +56.0 -57.0 40 8Morphine and 300 500 67 34 18 47 -12.0 -30.0 28 20

succinylcholine

563

MORTALITY DATA

There was no immediate mortalitd group. In the control ventilation

group the mortality was 50% with 31%dying during the procedure and 19% dur-ing variable periods after the procedure .Of the 5 who died during controlled venti-

4 deaths occurred during the main-tenance period and I during acute hyper-ventilation. Thus, of the total 19 pation ventilatory assistance, the overall mor-tality was 41%.

Of

Many patients in acute venture will require some form of ventilatoryassistance. Both intermittent and conous assisted PPB have been used to improvealveolar ventilation, prevent oxygen respim

depression, allow delivery of aerosolbronchodilators, improve intrapulair distribution, and reduce the mechanicalwork of breathing (3). Conversely, IPPB isreported to be of no benefit or detrimentalto patients in the chronic stable diseasestate or to those with superimposed acuteventilatory failure (4-7) .

Thus, clinical situations arise whereure of assisted IPPB is associated with detri-mental ranges of hypoxemia and acidosis .Possible causes for failure may be [1] in-spiratory effort too weak to trigger the res-pirator; [2] patient too confused or agitatedto accept or coordinate his breathing withthe rate of the respirator ; [3] increases inrespiratory work (intrinsic or machine-induced); [4] tachypnea ; [5] failure of bloodgases to improve, despite what appears tobe adequate cycling between patient andrespirator ; and [6] depression of ventilationsecondary to oxygen administration. Con-trolled ventilation should be considered insuch cases since prognosis cannot be deter-mined and immediate support is necessary .The body tank respirator for controlling

ventilation aroused interest in 1951 as aresult of the work of Boutourline-Youngand Whittenberger (8) in which two mori-

E . W WEBS AND M. J. DULFANO

bund patients with emphysema receivedoxygen and PaC02 was reduced. Favorablereports of the Engstr6m and the Mi5rchvolume-cycled ventilators document that ef-fective hyperventilation and prolonged con-

ion are feasible in patientsobstructive lung disease. To

to large series of patientswith chronic obstructive lung disease has

reported in which conventional pres-sure-cycled respirators (IPPB) were em-ployed, for controlled ventilation (2, 9) .

Some reports describing volume-cycled ortank respirators do not stress the problem ofsynchronization . This phenomenon may berare, or most patients quickly learn to cyclesmoothly. However, Marchand and VanHasselt (10) described controlled ventilationas "last-resort treatment" for status asth-maticus in advanced ventilatory failure .Despite supportive care their

fined restless and confused and would notsynchronize with the Engstrom unless mor-phine (20 to 40 mg) was administered .Others (11, 12) also indicate the problemof synchronization and the need for para-lytic or depressant agents with tank orvolume-cycled respirators .

Studies with IPPB indicate that simplecycling of the patient with the respirator(machine cycle) may reduce hypercapnia(13) . Automatic machine cycle is effective insevere crush thoracic injuries, yet anesthesiaor d-tubocurarine was necessary for unco-operative patients (14) . Neuromuscularblockade was necessary for successful venti-lation in status asthmaticus (15) . The useof depressants or sedatives has been discour-aged (16). However, it may be physicallyimpossible to manage certain individualsunless such agents are used .Employing simple machine cycle, the con-

tinuous automatic Mark 7 or 8 IPPB unitwas most successful in the group of patientswho were already somnolent or depressedby intrinsic acidosis, hypoxemia, cerebralinsults, drugs, or prior cardiac arrest . Theremaining patients (597, of this series)

Annals ofInternal Medidne

Volume 67, No. 3September 1967

were agitated, confused, and uncooperative,and either 100%a oxygen, succinylcholine,morphine, or all three were given toproduce relaxation or muscle paralysis . Onceone of these methods was selected, adequatereduction in Paco 2 and [H'] and improve-ment in Pao 2 were possible . Furthermore,the reduction in Paco 2 is rate-controllable,depending upon the IPPB features selected .It should be emphasized that, since effectivemachine synchronization frequently occursand obviates the need for drugs, it shouldbe attempted first. If synchronization fails,agent selection thereafter will depend uponthe existing clinical situation . Succinylcho-line in the severely agitated patient willcause rapid paralysis and must be followedimmediately by automatic cycle . Those un-cooperative patients sensitive to 100%Q oxy-gen may depress sufficiently so that no otherdrug is required. Morphine (or other de-pressant agents) may be titrated in smalldoses to reduce agitation . It was our generalexperience that, once effective control cyclewas established, the patient remained co-ordinated and did not require very frequentor large doses for maintenance .

Care must be employed to avoid reduc-ing the Paco2 too rapidly (17) . In only oneinstance did alkalosis of 7 .52 occur, butwith no side effects . Rapid correction ofacidosis may reduce cerebral blood flow,shift the 0 2 dissociation curve to the left,or produce hypotension or apnea . Meantimes to reduce Paco 2 and [H-] ranged from21 hr to 64, hr. Whether such intervals aretoo rapid is difficult to document . Blockand Ball (9) reported little difficulty inlowering Paco 2 over a 1-hr period . Thefour deaths in this study during hyperven-tilation may or may not be related to thisphenomenon .

The fairly uniform reduction in Paco 2during the hyperventilation period was ac-companied by variable increases in minuteventilation. The mean falls in Paco2 wereassociated with 34 to 50% reduction in res-piratory rate and a 67 to 2111y,, increase in

MANAGEMENT OF ACUTE VENTILATORY FAILURE 564

tidal volume. Thus, appropriate manage-ment should focus on tidal volumes andrespiratory rates and not on the minutevolume alone. Factors responsible for theimproved ventilation may include im-proved inspired gas distribution, reducedphysiologic dead space, improved ventila-tion/perfusion relationships, more completeemptying of lung compartments (18), andless metabolic demands related to reducedwork of breathing (19) .

High inspiratory pressures required foroptimal tidal volumes may reduce venousreturn to the left atrium and cardiac out-put, particularly in hypovolemia or shock,creating the difficulty of maintaining ap-propriate Pao2 and Paco 2 levels and ade-quate cardiac output (20) . Rapid reversi-bility of hypotension was noted in ourpatients once high pressures (40 cm 1420)were reduced, and prolonged periods re-quiring high inspiratory pressures were notobserved. This hypotension was not relatedto rapid reductions in Paco2 since Pco2 waslowered relatively gradually. Thus, duringthe procedure temporary reduction in ve-nous return must be balanced by the needfor immediate ventilation ; once underlyingbronchopulmonary and cardiac factors arecontrolled, reduction in high inspiratorypressures is possible . Similarly, an elevationof central venous pressure and a temporaryreduction in urinary output may occur.These changes are reversed with lowerIPPB driving pressures, and they shouldnot be interpreted as ischemic acute renalfailure .

The advantage of maintaining controlledventilation after the hyperventilation phasehas the following rationale: [1] furthertime for medical therapy and [2] reductionin work of breathing . The improvement inblood gases by mechanical ventilation willnot concurrently improve the precipitatingand underlying pathophysiologic factors .Controlled cycle may be continued untilbronchospasm, secretions, infections, car-diac failure, etc . are controlled, and then

565

assisted IPPB will be effective . In this studymaintenance periods varied with existingmedical conditions, with 12 to 30 hr repre-senting the minimum necessary for clinicalstablization (total time, 16 to 36 hr) .Whether such intervals are safe cannot bedefined at present; however, artificial ven-tilation is possible for prolonged periodswith no definable deleterious effects, provid-ing oxygen toxicity is avoided (21, 9) .

The total mortality in this series was410/,. The mortality during controlled ven-tilation was 50 0]0 , 310j% dying during theprocedure and 19 0]0 after the controlledcycling was terminated. Complications en-countered were bacteremia (Pseudomonas-two cases), acute myocardial infarction (twocases), arrhythmia (three cases), gastrointes-tinal bleeding (one case), shock (four cases),and suspected pulmonary emboli (twocases) . The experiences of others vary fromcomplete success (22) to 500]0 (23) to 30 to35 0]0 mortality (24) . Variations in age, ex-tent and type of pulmonary disease, meth-ods of ventilation, and complicating medi-cal conditions must affect such differences,as we have previously emphasized (25) . Theneed for control cycle may influence mor-tality by [1] selection of more critically illpatients ; [2] complications of prolongedcontrol ventilation and immobilization(parenteral therapy, venous stasis with pul-monary emboli, infection, etc .) ; [3] effectsof reduction in Paco2 and [H'] and expo-sure to high respirator pressures ; and [4]oxygen toxicity .

In the context of the type of patient wehave described, this procedure is associatedwith a significant mortality . While a vigor-ous attitude is emphasized, controlled ven-tilation should be employed with care andawareness of the potential dangers and maybe performed at the bedside only if con-stant physiological monitoring and medicaland nursing care are available .

SUMMARY

Controlled mechanical ventilation withpressure-cycled intermittent positive-pres-

E . B . WEISS AND M . J . DULFANO Annals ofInternal Medicine

sure breathing may be used to maintainadequate ventilation in patients withchronic obstructive lung disease in acuteventilatory insufficiency when optimal as-sisted intermittent positive-pressure venti-lation fails to perform this function. Inthis series this modality was substituted forassisted ventilation because of poor patientcooperation, inability to reduce Paco 2 (or[H']) or improve Pao2 , oxygen depressionof respiration, or progressive deteriorationdespite conventional assisted ventilation .

In 19 patients with 21 episodes of acuteventilatory failure, only 4 trials could bemanaged by assisted intermittent positive-pressure breathing. The remaining 17 epi-sodes were managed by controlled ventila-tion in the following sequence, employingeither endotracheal intubation or trache-ostomy: automatic machine cycle, oxygendepression, and drug suppression. Simplemachine cycle was effective in 41% of cases,particularly in the obtunded, comatose pa-tient. With agitated, uncooperative patientsoxygen depression in 1870 and drug sup-pression (morphine or succinylcholine orboth) in 41% were effective in establishingcontrol .

Proper synchronization with the respira-tor was a prerequisite for controlled ven-tilation . Thereafter, effective ventilationcould be achieved . Mean falls of 30 to 57mm Hg in Paco 2 , improvement in pH of0.09 to 0.21 units, and adequate Pao 2levels were observed in contrast to the lackof improvement during assisted ventilation .Such improvements were often associatedwith reduced respiratory rates and in-creased tidal volumes rather than with grossincreases in minute ventilation .

High respirator pressures with moderateinspiratory and slow expiratory flow rateswere employed with relative safety exceptfor reversible hypotension encountered withhigh inspiratory pressures (25 to 40 cmH2O) . Complications such as gastrointes-tinal bleeding, cardiac arrest, arrhythmia,pulmonary emboli, and shock (cardiogenic

Volume 67, No . 3September 1967

or bacteremic) had an adverse effect uponsurvival and may have been in part a con-sequence of the procedure . The overallmortality rate was 41%Jo .

A hyperventilation period (2 to 6 hr), di-rected to eliminate dangerous hypoxemiaand acidosis, and a maintenance period (12to 30 hr), where control ventilation wascontinued, permitted full medical therapyto correct the acute precipitating factors .Thereafter, conventional assisted intermit-tent positive-pressure breathing and non-assisted breathing were possible . Whetherventilation is performed at the bedside orin an intensive care unit, constant physio-logical monitoring is mandatory .

REFERENCES

1. SLUITER, H . J ., KOOLHAAS, B ., LENDE, R . VAN DER,TAMMELING, G. J ., BLOKZIJL, E., VAN DIJL,W., ORIE, N. G. M . : Conservative and aggres-sive treatment of acute severe respiratory in-sufficiency in patients with chronic non-specific lung disease (C.N .S .L .D .) . Med . Tho-rac . 21 : 335, 1964 .

2. NORLANDER, 0 . P., BJSRK, V . 0., CRAFOORD, C .,FRIBERG, 0 ., HOLMDAHL, M., SWENSSON, A .,WIDMAN, B . : Controlled ventilation in medi-cal practice . Anaesthesia 16: 285, 1961 .

3 . SHELDON, G. P . : Pressure breathing in chronicobstructive lung disease . Medicine (Bait.) 42 :197, 1963 .

4. SUKUMALCHANTRA, Y., PARK, S . S., WILLIAMS,M. H ., JR . : The effect of intermittent positivepressure breathing (IPPB) in acute ventila-tory failure . Amer. Rev. Resp . Dis . 92: 885,1965 .

5 . KAMAT, S. R., DULFANO, M . J ., SEGAL, M. S . :The effects of intermittent positive pressurebreathing (IPPB/I) with compressed air inpatients with severe chronic nonspecific ob-structive pulmonary disease . Amer. Rev. Resp .Dis . 86 : 360, 1962.

6. CULLEN, J. H., BRUM, V. C., REIDT, W. U . : Anevaluation of the ability of intermittent posi-tive pressure breathing to produce effectivehyperventilation in severe pulmonary em-physema . Amer. Rev. Resp . Dis. 76: 33, 1957.

7 . JONES, R . H., MACNAMARA, J ., GAENSLER, E . A . :The effects of intermittent positive pressurebreathing in simulated pulmonary obstruc-tion. Amer. Rev. Resp. Dis. 82: 164, 1960 .

8 . BOUTOURLINE-YOUNG, H . J ., WHITTENBERCER, J .L.: The use of artificial respiration in pul-monary emphysema accompanied by highcarbon dioxide levels. J . Clin . Invest . 30 : 838,1951 .

MANAGEMENT OF ACUTE VENTILATORY FAILURE 566

9 . BLOCK, A. J ., BALL, W. C., JR. : Acute respiratoryfailure : observations on the use of the Mbrchpiston respirator . Ann. Intern . Med . 65: 957,1966.

10. MARCHAND, P., VAN HASSELT, H . : Last-resorttreatment of status asthmaticus . Lancet 1 :227, 1966.

11 . WESTLAKE, E . K., SIMPSON, T ., KAYE, M.: Carbondioxide narcosis in emphysema . Quart. J.Med . 24 : 155, 1966.

12 . RONCORONI, A. J . : Resuscitation in severe res-piratory insufficiency, in Bronchitis, editedby ORIE, N. G. M., SLUITER, H. J. CharksThomas Co., The Netherlands, 1964, p . 105 .

13 . ADDINGTON, W. W ., KETTEL, L. J ., CUGELL, D . W . :Alkalosis due to mechanical hyperventilationin patients with chronic hypercapnia . Amer .Rev. Resp. Dis . 93 : 736, 1966 .

14. AMBIAVAGAR, M ., ROBINSON, J . S., MORRISON,I. M., SHERWOOD JONES, E.: Intermittent posi-tive pressure ventilation in the treatment ofsevere crushing injuries of the chest. Thorax21: 359, 1966.

15. BEAM, L . R., MARCY, J. H., MANSMANN, H . C.,JR . : Medically irreversible status asthmaticusin children . Report of three cases treatedwith paralysis and controlled respiration .JAMA 194: 968, 1965.

16 . GAENSLER, E. A., SMITH, C., MCCLEMENT, J. H.,NORLANDER, 0. P., HERZOG, H., COURNAND,A., ENGSTR6M, C . G ., LASSEN, H . C . A ., SADOUL,P., CAMPBELL, E . J. M ., WOLFF, E. S., CLOWES,H. A ., JR., DULFANO, M.: Methods of therapy :mechanical ventilation. Panel discussion .Ann. N. Y. Acad. Sci . 121 : 793, 1965 .

17 . ROBIN, E. D . : Abnormalities of acid-base regu-lation in chronic pulmonary disease, withspecial reference to hypercapnia and extra-cellular alkalosis . New Eng. J. Med . 268 : 917,1963 .

18 . EMMANUEL, G. E ., SMITH, W. M., BRISCOE, W. A. :The effect of intermittent positive pressurebreathing and voluntary hyperventilationupon the distribution of ventilation and pul-monary blood flow to the lung in chronicobstructive lung disease . J. Clin. Invest . 45 :1, 221, 1966 .

19. AYRES, S. M ., GIANNELLI, S ., JR . : Oxygen con-sumption and alveolar ventilation during in-termittent positive pressure breathing. Dis.Chest 50 : 409, 1966.

20 . HEDLEY-WHYTE, J ., PONTOPPIDAN, H ., JOCELYNMORRIS, M . : The response of patients withrespiratory failure and cardiopulmonary dis-ease to different levels of constant volumeventilation . J . Clin. Invest . 45 : 1, 543, 1966 .

21. PONTOPPIDAN, H ., HEDLEY-WHYTE, J ., BENDIZIN,H. H., LAVER, M. B., RADFORD, E. P., JR. :Ventilation and oxygen requirements duringprolonged artificial ventilation in patients

567

with respiratory failure . New Eng. J. Med.273: 401, 1965 .

22. LOVEJOY, F. W., Yu, P. N. G., NYE, R . E., JR.,Joos, H. A., SIMPSON, J . H.: Pulmonary hy-pertension . 111 . Physiologic studies in threecases of carbon dioxide narcosis treated byartificial respiration . Amer. J. Med. 16 : 4,1954.

23 . SEEKER, H. 0., HICKAM, J . B.: Carbon dioxideintoxication: the clinical syndrome, its etiol-ogy and management with particular refer-

SS AND M. J. DULFANO Annals ofInternal Medicine

ence to the use of mechanical respiration .Medicine (Balt.) 35 : 389, 1956.

24. MUNCK, O ., KRISTENSEN, H . S ., LASSEN, H . C . A . :Mechanical ventilation for acute respiratoryfailure in diffuse chronic lung disease . LancetI : 66, 1961 .

25. DULFANO, M. J., ISHIKAWA, S . : Hypercapnia:mental changes and extrapulmonary compli-cations. An expanded concept of the "CO,intoxication" syndrome. Ann. Intern. Med .63: 829, 1965 .