cardiac arrest in the critically ill -...

Cardiac Arrest in the Critically Ill

I. A Study of Predisposing Causes in 132 Patients

By SAMUEL J. CAMARATA, M.D., MAX HARRY WEIL, M.D., PH.D.,PAUL K. HANASHIRO, M.D., AND HERBERT SHUBIN, M.D.

SUMMARYThe events preceding cardiac arrest were investigated in 132 critically ill patients

on the basis of the hemodynamic, respiratory, and metabolic status prior to cardiacarrest. Approximately one-half of the patients had respiratory acidosis prior to cardiacarrest and only one of these patients ultimately survived. An additional one-fourth hadmetabolic acidosis due to progressive perfusion failure (shock), and none survived.In the remaining one-fourth, an immediate catastrophic and potentially preventableevent accounted for cardiac arrest, and all but one of the long-term survivors wereincluded in this group. In thirteen patients, obstruction or dislodgment of endotrachealtracheostomy tubes, or interruption of ventilation, was recognized as the immediatecause. In an additional 20 patients, cardiac arrest was associated with an adverseeffect of a drug. Twelve instances of cardiac arrest were associated with central venousinjection of aminophylline.The initial cardiac resuscitation attempt was effective in 38% of patients. However,

the majority of patients who survived the first episode succumbed after a second or

third episode. The cumulative survival was disappointingly small. Only six patients,or 5% of the total group of 132 patients, were long-term survivors. These data mayprovide the basis for more restricted use of routine and repetitive resuscitative measures

particularly for elderly patients in whom

circulatory failure.

acidemia is due to progressive ventilatory or

Additional Indexing Words:Acidemia Acidosis Aminophylline Cardiac resuscitation Inotropic agents

W HEN CARDIAC arrest occurs as a

catastrophic event, current techniquesof cardiac resuscitation are remarkably effec-tive. Prompt action by well-trained personnelmay restore more than one-half of otherwisefatal cases of airway obstruction, apnea,drowning, electric shock, or patients who havesustained an acute coronary episode. On theother hand, when cardiac arrest occurs in pa-

From the Shock Research Unit and the Departmentof Medicine, University of Southern California Schoolof Medicine, and Los Angeles County/University ofSouthern California Medical Center, Los Angeles,California.

Supported by the John A. Hartford Foundation,Inc., and by Public Health Service grants HS-05570from the National Heart and Lung Institute, GM16462 from the National Institute of General MedicalSciences and HS-00238 from the National Center for

688

tients who are already critically ill, resuscita-tive measures are much less effective. Fewerthan 5% of such patients are successfully resus-

citated.1 In the present study, we searched formore objective understanding of mechanismsof cardiac arrest and the effectiveness ofresuscitation in a large group of critically illpatients. Our efforts were in part promoted bypractical and ethical considerations.A unique opportunity for investigation was

provided by the special monitoring facilities of

Health Services Research and Development, HealthServices of the Mental Health Association.

Address for reprints: Max Harry Weil, M.D.,University of Southern California School of Medicine,1322 North Vermont Avenue, Los Angeles, Califomia90027.

Received October 7, 1970; revision accepted forpublication June 19, 1971.

Circulation, Volume XLIV, October 1971

by guest on June 20, 2018http://circ.ahajournals.org/

Dow

nloaded from

http://circ.ahajournals.org/

CARDIAC ARREST

our Shock Research Unit.2 Comprehensivefacilities were used for monitoring alarms andrecording of hemodynamic, pulmonary, andmetabolic changes for periods as long as 10days. The episode of cardiac arrest was

immediately recognized since the electrocar-diogram, intraarterial pressure pulses, andcentral venous pressure were prominentlydisplayed on an oscilloscopic screen above thepatient's bed and as a numerical output on an

adjacent video screen. Detailed data were

stored in part by use of a digital-computersystem.

Method of StudyPatientsThe patient group included 78 men and 54

women, ranging from 4 to 84 years of age, whowere referred from medical and/or surgicalservices of the Los Angeles County/University ofSouthern California Medical Center to the shockward for management of shock or other life-threatening cardiopulmonary crises. Primary dis-eases were due to infection (37%), cardiac cause

(22%), blood or fluid loss (13%), and miscellaneousor multiple causes in the remaining patients(28%).A total of 193 episodes of cardiac arrest were

documented in the 132 patients. Each patientwas under supervision of a nurse or physicianconstantly at the bedside. The electrocardiogram,intraarterial pressure pulse, and the centralvenous pressure were prominently displayed on a

24-inch oscilloscopic screen suspended from theceiling above the patient's head and as a

numerical output on an adjacent video screen.These three measurements were also preserved forretrospective analysis by use of a four-channelrecorder and in part by acquisition of datautilizing a digital-computer system.2

Cardiac arrest was defined as the suddencessation of effective cardiac action with theintraarterial pulse pressure in each instancereduced to levels of less than 6 mm Hg, absentheart sounds, and loss of consciousness. Patientsin whom cardiac arrest occurred as a consequenceof progressive blood loss, uncontrolled congestiveheart failure, or protracted cardiac arrhythmiawere excluded. Standard resuscitation maneuvers

were employed including closed-chest cardiacmassage, mechanical ventilation, endotrachealintubation, and direct-current defibrillation.Drugs were injected through a central venous

catheter which had previously been inserted on a

routine basis at the time of admission. Adrenergicamines, especially epinephrine and isoproterenol,Ctifculation, Volume XLIV, October 1971

and 10% solutions of calcium chloride or calciumgluconate were the primary inotropic agents.Hypertonic sodium lactate or bicarbonate wasadministered routinely in amounts ranging from44 to 220 mEq.

Procedures

An inventory of hemodynamic and metabolicmeasurements was routinely obtained within 4hours after admission of a patient to the ShockResearch Unit. For this purpose, a polyethylenecatheter was introduced into the right atriumthrough a surgically exposed basilic vein. Correctposition was established by observation ofintracardiac pressure pulses during advancementof the catheter into the right ventricle and itssubsequent withdrawal to a position just proximalto the tricuspid valve.3 A red Odman catheter wasinserted into the brachial, radial, or femoral arteryover a guide wire by percutaneous technique orafter surgical exposure and advanced for adistance of 10 cm. The catheters were connectedto Statham 23 AA strain-gauge pressure transduc-ers and flushed intermittently under pressure withnormal saline solution containing 5 units/ml ofheparin. Arterial and central venous pressureswere recorded on a direct-writing multichannelrecorder. The electrocardiogram (lead II) wasmonitored continuously, and heart rate wasmeasured from it.

Blood was mwthdrawn from the arterial catheterat regular intervals of between 2 and 4 hours, andas frequently as every 11 minutes at time ofclinical crises, for measurement of blood gases.Ventilation was assisted in a majority of patientswith the use of a self-cycling positive-pressureventilator (Bird Mark VIII) or a volume-controlled ventilator (Engstrom Respirator model200). The inspired gas contained between 40 and100% oxygen. A Beckman 160 analyzer, with aflow-through pH electrode and subsequently aRadiometer system, was used for measurement ofblood pH, pCO2, and P02. Arterial oxygensaturation was measured by use of a Watersdouble-scale cuvette oximeter, the calibration ofwhich was periodically confirmed by Van Slykeanalyses.

ResultsSurvivalAn initial episode of cardiac arrest was

successfully reversed in 50 of the 132 patients(38%), but only five of these survived to leavethe hospital. A second episode of cardiacarrest occurred in 45 of the surviving patients,and 17 (38%) were successfully resuscitated.Only one of these 45 patients ultimately

689


Dow

nloaded from


CAMARATA ET AL.

survived. A third attempt at cardiac resuscita-tion was made in 16 patients. Cardiac arrestwas successfully reversed in seven (44%), butall of these patients ultimately died. Twopatients had more than three episodes ofcardiac arrest and both died during asubsequent episode. These data are summar-ized in table 1.The percentages of successful resuscitations

during each of the three successive resuscita-tive efforts were remarkably similar, i.e., 39,38,and 44%. Assuming independence of episodeoutcome, the probability of successfully resus-citating a patient from three episodes ofcardiac arrest, therefore, was 0.39 x 0.38 x 0.44= 0.06. However, survival to the time of hos-pital discharge was even less, for only six pa-tients (5%) left the hospital alive.

Predisposing CausesFor purposes of analysis, three groups of

patients were identified. In 98 patients, 74% ofthe total group, acidosis with arterial bloodpH of less than 7.35 was detected prior toonset of cardiac arrest. Group I included 65

Table 1Survival following Cardiac Resuscitation

First Second Third* Total

Attempted resuscitations 132 45 16 193No. survivors 50 17 7 6Success of each attempt(%) 38 37 43 38t

Cumulative survival (%) 38 14 6 5No. long-term survivors 5 1 0 6

*More then three resuscitation efforts were made inonly two patients.

tWeighted average of individual resuscitation at-tempts.

patients with predominant respiratory acidosisin whom arterial blood pH was less than 7.35,and concurrent pCO2 was 41 mm Hg or more;pH averaged 7.16 and pCO2, 67 mm Hg.Group II included 34 patients with predom-inantly metabolic acidosis with pH less than7.35 and pCO2 equal to or less than 39 mmHg. Arterial blood pH averaged 7.21 andpCO2, 28 mm Hg. A majority of these patientshad clinical and hemodynamic changes char-acteristic of circulatory shock.The remaining 33 patients constituted

group III and accounted for 26% of the totalgroup. Arterial blood pH prior to cardiacarrest was 7.37 or greater. In each of thesecases, however, an unexpected and catastroph-ic event accounted for cardiac arrest. Themean value of pH and pCO2 and the ranges ofvalues for each of the three categories ofpatients are summarized in table 2.The mechanism of cardiac arrest for one of

the patients in group I is illustrated in figure 1.

A 34-year-old woman was admitted afteringestion of an overdose of barbiturates.Because of aspiration of oral and gastriccontents prior to admission, the clinical course

was complicated by bronchopneumonia. Aprogressive fall in pH and rise in pC02preceded cardiac arrest. The patient was

successfully resuscitated, maintained on a

mechanical ventilator with dedicated airwaycare for a protracted period, and ultimatelysurvived. This case is specifically cited becauseamong the 99 patients included in groups Iand II this is the only patient who was a long-term survivor.

Illustrative of patients in group II is thecase of a 63-year-old man who had massive

Table 2Arterial Blood pH and pCO, within Four Hours preceding Cardiac Arrest

No. ofPatient group patients pH pC02

I: Respiratory acidosis 65 7.16 67(pH < 7.35; pCO2 > 40 mm Hg) (6.61-7.35) (41-120)

II: Metabolic acidosis 34 7.21 28(-pH < 7.35; pC02 < 40) (6.96-7.35) (11-39)

III: Catastrophic event 33 7.45 37(7.37-7.59) (19-54)


690


Dow

nloaded from


CARDIAC ARREST

Table 3

Presumptive Causes of "Catastrophic) CardiacArrest (Group III)

No. of CatastrophicEvent patients events (%)

Drug therapy:Aminophylline 12 36Epinephrine :3 9Digoxin (Lanoxin) 2 6Potassium penicillin G 2 6Chlorpromazine (Thorazine) 1 3

20 60Airway manipulation - anoxia:

Endotracheal manipulation 9 27Dislodgment of tracheostomy

tube 2 6Accidental interruption of

ventilation 2 613 40

pulmonary infarction due to embolic occlusionof the left pulmonary artery. Cardiac arrestfollowed a progressive fall in pH from 7.29 to7.11 and pCO2 from 25 to 18 mm Hg. Aconcomitant decline in blood pressure from150/78 to 88/49 mm Hg and an increase inlactate content of arterial blood from 14.8 to23.0 mm were observed. Cardiac arrest wasreversed shortly after admission to the shockward, but progressive decline in cardiacoutput and arterial pressure ensued, terminat-ing in irreversible cardiac arrest within 5hours after admission.The catastrophic events which accounted

for the 33 patients in group III were related toan adverse reaction to drugs (60%) or acuteanoxia (40%). Specific causes are summarizedin table 3. The cardiac catastrophies in 13 or

ARTERIAL BLOODpH units (pHa)

62 PC02 mmHg (P(C042 l_

22

100 2 Sot % (San0)±~~~~~95 - 4!4

90 -1E70 I P02 mmHg (P )

t1 a°2

__---

ARREST

_M

_ow~*

2100 0100 0500 0900 1300

TIME -hours

1530 160U ZZOO 0600SRU *317 34 9Seconal overdoseThrombus Rt. Fem. ort.

1100

Figure 1

A progressive fall in pH and rise in PCO2 preceding cardiac arrest.


a a A

691

am

a 1


Dow

nloaded from


CAMARATA ET AL.

40% of the patients in this group vereassociated with interruption of mechanicalventilation, most commonly while the nurse orphysician was manipulating the airway. Acci-dental dislodgment of a tracheostomy cannulaand temporary interruption of mechaniicalventilation accounted for four episodes ofcardiac arrest. In each instance, these inter-ruptions occurred while the patient's bodyposition was changed during nursing proce-dures and especially for washing or changingbed sheets.An example of ventricular standstill folloxv-

ing temporary interruption of mechanicalventilation for readjustment of a ventilator isdemonstrated in figure 2. The patient was a68-year-old man in whom mesenteric arteryocclusion was demonstrated on surgical ex-ploration.

Outstanding among the drugs implicated insudden cardiac arrest was aminophylline, atheophylline ethylenediamine compound,which was slowly injected through the centralvenous catheter in doses ranging from 250 to750 mg. Injection of epinephrine and digoxinthrough the central venous catheter also wasassociated with catastrophic cardiac arrest. Intwo instances, cardiac arrest followed intra-venous administration of potassium penicillin

ARTERIAL PRESSURE mmHg100- E X ::

} . ~~~~~~4 . ;.-lo

0o,

VENOUS PRESSURE mmHg

ELECTROCARDIOGRAM

mv

G in an amount of 2 million and 5 millionunits over a period of 15-30 min. The adverseeffect may have been due to the potassiumion, but an undetected sensitivity to penicillinis not excluded.

Illustrative is the case of a 68-year-oldpatient in whom surgical repair of anesophageal diverticulum was complicated bymediastinitis and shock. Ventricular fibrilla-tion followed rapid infusion of 5 million unitsof potassium penicillin G containing 9 mEq ofpotassium ion into the right atrium over aperiod of 15-30 min (fig. 3).The possibility that the route of administra-

tion or the speed of infusion of aminophyllinein relation to the underlying cardiopulmonarydefects accounted for cardiac arrest is notexcluded. What does emerge is that theincidence of cardiac arrest after aminophyllineunder these conditions was selectively high. Inthe shock ward, aminophylline was used lessthan one-tenth as often as digoxin or peni-cillin, yet the incidence of cardiac arrest within5 min after its administration was six timesas great. This estimated 60-fold greater inci-dence of cardiac arrest with aminophyllinethan with the other drugs is presumptiveevidence of a specific reaction to aminophyl-line.

gr t-,, ?, _ Ye -@f iS .......................... ' o............',..

CARDIAC RESUSCITATION

SRU# 704 Z/68MECHANICAL VENTILATOR FAILURE

Figure 2Ventricular standstill following teemporary interruption of mechanical ventilation for re-adjustment of a ventilator.


692


Dow

nloaded from


CARDIAC ARREST

200 ARTERIAL PRESSURE mmHg:rF K PENICILLIN 5M units200

ol~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~7 .....:

VENOUS PRESSURE mmHg

201

,-,<~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~... ... ....* S

ELECTROCARDIOGRAM

mI

8:00 8:05 1 minTIME

< s t | | ts os X tt~~~~~~~~~~~~~PA+- | * *s ~~~~~~.. ... ......... .eS. -4 _ ;,::'. ..7.z;'i'5>C,.A.:&$ e .4w3.v . 'i...,.kX#]S- 'it;.

VENTRICULAR FIBRILLATION

r~~~ ~ ~~~~~~~~~~J

fi E( *- .i *S ............................................ ............

8:10 8.14 1 sec

SRU * 691KCARREST

Figure 3

Ventricular fibrillation after infusion of 5 million units of potassium penicillin G containing9 mEq of potassium ion into the right atrium over a period of 15-30 min. Arterial andcentral venous pressure pulses reflect closed-chest cardiac massage.

Electrical Mechanism of Cardiac Arrest

The relative incidence of ventricular asys-

tole and ventricular fibrillation was analyzedfor the 132 initial episodes of cardiac arrest.Pertinent data are summarized in table 4. In76 patients (57%), ventricular asystole was theprimary arrhythmia, and in 56 patients (43%)cardiac arrest was due to ventricular fibrilla-tion. There were no significant differences inthe relative incidence of cardiac standstill andventricular fibrillation among the three cate-gories of patients included in this study.

Discussion

The patients on whom this report is basedwere selectively the most critically ill patientsin a large general hospital. The mortality ofpatients admitted to the shock ward was 61%.Even when patients are excluded in whomcardiac arrest occurred as a terminal eventcomplicating exsanguination, protracted ar-

rhythmias, or congestive heart failure whichfails to improve promptly on medical manage-

ment, this patient population still represents a

selected group of very ill persons. This

Table 4

Electrical Mechanism of Cardiac Arrest

Ventricular standstill Ventricular fibrillationSuccessfully Successfully

No. of Incidence reverted Incidence revertedPatient group patients No. % No. % No. % No. %

I: Respiratory acidosis 65 34 52 7 20 31 48 15 5011: Metabolic acidosis 34 22 65 16 72 12 35 2 7

III: Catastrophic event 33 20 60 0 0 13 40 5 30Total 132 76 57 23 30 56 43 22 40

Ci.rculation, Volume XLIV, October 1971

693


Dow

nloaded from


CAMARATA ET AL.

precludes extrapolation of the informationgathered during this study to the considerableinformation now available on previouslyhealthy patients in whom sudden obstructionof the airway, immediate reactions to medica-tion, electric shock, or sudden cardiac failurefollowing myocardial infarction is the cause ofcardiac arrest.On the other hand, useful information

emerges from the analysis which has bothclinical and ethical implications. For practicalpurposes, there was no salvage of patients inwhom progressive respiratory or circulatoryfailure accounted for acidosis prior to onset ofcardiac arrest. This failure to ultimatelyreverse cardiac arrest in this large group ofpatients is of interest in the light of currenttraditions to attempt resuscitation on allexcept patients with terminal neoplasms orcerebrovascular accidents.4 Successful resusci-tations were therefore primarily limited tothose patients in whom cardiac arrest wascatastrophic and potentially preventable.The success of initial resuscitation is consid-

erably greater in coronary care units. How-ever, our resuscitative experience is morecomparable to that reported by mobileresuscitation teams in teaching hospitals.5 Thedisappointingly small number of long-termsurvivors reflects the very high incidence ofrecurrence of cardiac arrest and the ultimatelethality of the primary disease which ac-counted for protracted acidosis prior to theonset of cardiac arrest.Ventricular fibrillation was not the predom-

inant cause of cardiac arrest. A majority of thepatients initially manifested ventricular stand-still. We would therefore not favor defibrilla-tion on an empirical basis prior to electrocar-diographic confirmation.

In comparison to previous studies, thepresent investigation provides special advan-tages which stem from the routine monitoringof patients employed in the shock ward.Continuous recording of the intraarterialpressure in addition to the electrocardiogramprovides an immediate indication of mechani-cal cardiac arrest. This is particularly impor-tant since alterations in electrocardiographic

complexes in instances of cardiac arrest inpatients with cardiogenic shock may followrather than precede cessation of an effectivearterial pulse.

Ventilatory failure appears to be the singlemost common cause of death in critical carefacilities including shock units. This alsoemerges from the present data. Approximatelyone-half of the patients were in respiratoryacidosis at the time of cardiac arrest. Mechan-ical ventilators were used in almost all of thepatients who demonstrated respiratory acido-sis. While anoxemia was initially reversed inmost instances by increasing the oxygencontent of the inspired gas, progressiveincreases in the airway pressures, a decline inarterial oxygen tension and saturation, andincreases in pCO2 terminated in cardiacarrest.6 A reduction in pulmonary complianceand progressive pulmonary arteriovenousshunting characterize the syndrome of pul-monary failure associated with shock. Theincidence of this syndrome has rapidly in-creased but its mechanism is as yet poorlyunderstood.7

In approximately one-fourth of the patientgroup, cardiac arrest occurred in patients withmetabolic acidosis. This was usually associatedwith a concomitant increase in blood lactateand occurred in patients with advancedclinical signs of shock. The carbon dioxidetension was reduced and oxygen tension inarterial blood was normal in the majority ofthese patients. These observations suggestthat, normal arterial oxygen tensions notwith-standing, perfusion failure, anoxic metabolism,and consequent metabolic acidosis predisposeto cardiac arrest. Our observations confirmprevious experimental and clinical studies inwhich cardiac output was observed to increasewith acidosis.8 9, 10 It is therefore more likelythat acidosis itself rather than a decline incardiac output accounted for cardiac arrest.

In 33 patients or approximately 25% of thetotal 132 patients (group III), neither respi-ratory nor metabolic acidosis was implicated.In each instance, however, cardiac arrestoccurred as an unexpected event. The hazardsof airway obstruction in patients who have


694


Dow

nloaded from


CARDIAC ARREST

endotracheal tubes or tracheostomy, particu-larly in the ventilator-dependent patient, isnow generally appreciated.7 This was thecause of cardiac arrest in 13 patients includedin group III. The seriousness of this problemis underscored in the present study in whichmechanical failure accounted for interruptionof ventilation and consequent cardiac arrest intwo instances. In each of these cases, effectiveventilation had been established and arterialblood gases had previously reflected normalpulmonary gas exchange while the patientswere maintained on the ventilator. Thesecatastrophies represent human error andprovide incentive for more intense training ofbedside personnel in respiratory management.

In 20 patients, catastrophic causes wereattributable to adverse effects of drugs. Therelative incidence of such adverse reactions isundoubtedly related to the selective practicesin various centers. In our own unit, it was onlyon retrospective analysis that we recognizedan inordinately high incidence of cardiac arrestwhich followed the intravenous administrationof aminophylline. The hazards of this agenthave been previously suggested."' 12 Thedoses which were used and protracted injec-tion over a period of between 3 and 5 minwere in accordance with established practice.The only additional variable was the site ofinjection. The possibility that the catastrophicreaction to this drug would have been avoidedby more peripheral injections at a slower rateand in greater dilution is not excluded. For thepresent, we have discarded the intravenoususe of aminophylline in favor of alternateroutes of administration or we have used otherbronchodilator or diuretic agents for treat-ment of the critically ill.

Finally, we are brought face-to-face withthe ethical issues that stem from long-termsurvival of only one 34-year-old woman fromamong the 99 cases of cardiac arrest whichoccurred in patients who had progressiveacidosis due either to ventilatory or tocirculatory failure. The extraordinary de-

mands on staff and the need for equipmentand supplies during resuscitation inevitablytax the resources of the hospital and, moreimportant, detract from intensive care of otherpatients. These data may therefore provide anobjective basis for more restricted use ofroutine and repetitive cardiac resuscitation,particularly of elderly patients in whomacidosis reflects progressive ventilatory andcirculatory failure.

References

1. JUDE JR, KOUWENHOVEN WB, KNiCKERBOCKERGG: Essentials of External CardiopulmonaryResuscitation. Baltimore, The Johns HopkinsMedical Institutions Monograph, 1968

2. JENSEN RE, SHUBIN H, MEAGHER PF, WEILMH: On-line computer monitoring of theseriously ill patient. Med Biol Engin 4: 265,1966

3. WEIL MH, SIUBIN H, ROSOFF L: Fluidrepletion in circulatory shock. JAMA 192: 668,1965

4. STEPHENSON HE: Cardiac arrest and resuscita-tion. St. Louis, C. V. Mosby Co., 1964

5. KIRBY BJ: Cardiac resuscitation: Clinical andmetabolic studies. Proc Roy Soc Med 60: 47,1967

6. LEWIN I, WEIL MH, SHUBIN H, SHERWIN R:Pulmonary failure associated with clinicalshock states. J Trauma 11: 22, 1971

7. MooRE FD, LYONS JH, PIERCE EC JR, MORGANAP JR, DRINKER PA, MACARTHUR JD,DAMMIN GJ: Posttraumatic pulmonary insuffi-ciency. Philadelphia, W. B. Sanders Co.,1969

8. DOWNING SE, TALNER NS, CARDNER TH:Influences of hypoxemia and acidemia on leftventricular function. Fed Proc 24: 157,1965

9. FEINS NR, DEL GUERCIO LRM: Increasedcardiovascular function in clinical metabolicacidosis. Surg Forum 17: 39, 1966

10. ANDERSEN MN, MOURITZEN C: Effect of acuterespiratory and metabolic acidosis on cardiacoutput and peripheral resistance. Ann Surg163: 161, 1966

11. BRESNICK E, WOODARD WK, SAGEMAN CB: Fatalreactions to intravenous administration ofaminophylline. JAMA 136: 397, 1948

12. MERRILL GA: Aminophylline deaths. JAMA123: 1115, 1943


695


Dow

nloaded from


HERBERT SHUBINSAMUEL J. CAMARATA, MAX HARRY WEIL, PAUL K. HANASHIRO and

PatientsCardiac Arrest in the Critically Ill: I. A Study of Predisposing Causes in 132

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1971 American Heart Association, Inc. All rights reserved.

75231is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TXCirculation

doi: 10.1161/01.CIR.44.4.6881971;44:688-695Circulation.

http://circ.ahajournals.org/content/44/4/688located on the World Wide Web at:

The online version of this article, along with updated information and services, is

http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer

of the Web page under Services. Further information about this process is available in thewhich permission is being requested is located, click Request Permissions in the middle columnClearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of the CopyrightCirculationoriginally published in Requests for permissions to reproduce figures, tables, or portions of articlesPermissions:


Dow

nloaded from

http://circ.ahajournals.org/content/44/4/688

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circ.ahajournals.org//subscriptions/


cardiac arrest in the critically ill -...

Documents