7

Click here to load reader

Simultaneous sternothoracic cardiopulmonary resuscitation: A new method of cardiopulmonary resuscitation

Embed Size (px)

Citation preview

Page 1: Simultaneous sternothoracic cardiopulmonary resuscitation: A new method of cardiopulmonary resuscitation

Resuscitation 48 (2001) 293–299

Simultaneous sternothoracic cardiopulmonary resuscitation:A new method of cardiopulmonary resuscitation

Sung Oh Hwang a,*, Kang Hyun Lee a, Jun Hwi Cho a, Bum Jin Oh a,Deepak S. Gupta b, Joseph P. Ornato c, Seung Hwan Lee d, Junghan Yoon d,

Kyung Hoon Choe d

a Department of Emergency Medicine, Wonju College of Medicine, Yonsei Uni6ersity, 162 IIsandong Wonju, South Koreab Department of Biomedical Engineering, Medical College of Virginia, Virginia Commonwealth Uni6ersity, MCV Station Box 525, Richmond,

VA 23298, USAc Department of Emergency Medicine, Medical College of Virginia, Virginia Commonwealth Uni6ersity, MCV Station Box 525, Richmond,

VA 23298, USAd Department of Internal Medicine, Wonju College of Medicine, Yonsei Uni6ersity, 162 IIsandong Wonju, South Korea

Received 7 January 2000; received in revised form 26 June 2000; accepted 29 June 2000

Abstract

No existing device for cardiopulmonary resuscitation (CPR) is designed to exploit both the ‘cardiac pump’ and the ‘thoracicpump’ effect simultaneously. The purpose of this study was to measure the haemodynamic effect of a new simultaneoussternothoracic cardiopulmonary resuscitation (SST-CPR) device that could compress the sternum and constrict the thoracic cavitysimultaneously in a canine cardiac arrest model. After 4 min of ventricular fibrillation, 24 mongrel dogs were randomized toreceive standard CPR (n=12) or SST-CPR (n=12). SST-CPR generated a new pattern of the aortic pressure curve presumed tobe the result of both sternal compression and thoracic constriction. SST-CPR resulted in significantly higher mean arterialpressure than standard CPR (68.9916.1 vs. 30.5910.0 mmHg, PB0.01). SST-CPR generated higher coronary perfusionpressure than standard CPR (47.0911.4 vs. 17.398.9 mmHg, PB0.01). End tidal CO2 tension was also higher during SST-CPRthan standard CPR (11.696.1 vs. 2.1793.3 mmHg, PB0.01). In this preliminary animal model study, simultaneous sternotho-racic cardiopulmonary resuscitation generated better haemodynamic effects than standard, closed chest cardiopulmonaryresuscitation. © 2001 Elsevier Science Ireland Ltd. All rights reserved.

Keywords: Cardiopulmonary Resuscitation; Cardiac arrest; Resuscitation

Resumo

Nao esta disponıvel no mercado qualquer equipamento capaz de, no contexto da reanimacao cardiorespiratoria (RCR),permitir a associacao simultanea do efeito de ‘‘bomba cardıaca’’ e ‘‘bomba toracica’’. O objectivo deste estudo foi medir, nummodelo canino de paragem cardıaca, o impacto hemodinamico dum novo aparelho para reanimacao cardio-respiratoriaesterno-toracica simultanea (SST-CPR), capaz de comprimir o esterno e em simultaneo permitir a constriccao da cavidadetoracica. Randomizamoa 24 caes para, apos 4 minutos de fibrilhacao ventricular, receberem uma de duas abordagens: RCRpadrao (n = 12) ou SST-CPR (n = 12). O SST-CPR criou um novo padrao de pressao intra-aortica presumivelmentedependente da associacao entre compressao do externo e constricao da cavidade toracica. A utilizacao do SST-CPR provocou umaumento significativo da pressao media intra-aortica quando comparada com a RCR padrao (68.9916.1 vs. 30.5910.0 mmHg,PB0.01). O SST-CPR permitiu um aumento da pressao de perfusao coronaria por comparacao com a RCR padrao (47.0911.4vs. 17.398.9 mmHg, PB0.01). O CO2 end tidal foi igualmente superior com a SST-CPR comparativamente a RCR padrao(16.196.1 vs. 2.1793.3 mmHg, PB0.01). Neste estudo preliminar em modelo animal, a reanimacao cardiorespiratoria comcompressao esternotoracica simultanea gerou um comportamento hemodinamico melhor comparativamente a RCR padrao,utilizando compressao cardıaca externa. © 2001 Elsevier Science Ireland Ltd. Todos os direitos reservados.

Pala6ras cha6e: Reanimaçao Cardiorespiratoria; Paragem Cardıaca; Ressuscitaçao

www.elsevier.com/locate/resuscitation

* Corresponding author. Fax: +82-0371-742-3030.

0300-9572/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved.PII: S 0 3 0 0 -9572 (00 )00250 -1

Page 2: Simultaneous sternothoracic cardiopulmonary resuscitation: A new method of cardiopulmonary resuscitation

S.O. Hwang et al. / Resuscitation 48 (2001) 293–299294

1. Introduction

Maintenance of vital organ perfusion is criticalduring resuscitation. Coronary perfusion pressureneeeds to be maintained above 20 mmHg toachieve return of spontaneous circulation duringCPR [1]. Standard CPR can usually generate only15–20% of normal cardiac output [2,3], which isinadequate to restore spontaneous circulation inthe majority of patients. Cardiac output (pump)and vascular resistance (load) are the main factorsaffecting perfusion pressure during resuscitation.Although vascular resistance can be optimized byadministration of vasopressors such as epinephrine[4] or vasopressin [5], an increase in vascular resis-tance without an adequate cardiac output could bedetrimental to tissue perfusion. Therefore, en-hancement of cardiac output is needed to optimizetissue perfusion during cardiac arrest.

Although a number of investigators have devel-oped techniques and/or devices that may augmentblood flow during CPR, only a few non-invasiveCPR techniques [6] [7] [8] [9] have been testedclinically in cardiac arrest patients. Most of thesehave tried to exploit either the cardiac or thoracicpump mechanism of blood flow during CPR. Wehypothesized that simultaneous sternal compres-sion and chest constriction (simultaneous ster-nothoracic cardiopulmonary resuscitation, orSST-CPR) might produce an additive haemody-namic effect. The purpose of this study was tomeasure the haemodynamic effect of a novel pro-totype SST-CPR device that could compress the

sternum and constrict the thoracic cavity simulta-neously in a canine cardiac arrest model.

2. Materials and methods

The experimental protocol was reviewed andapproved by the Animal Research Committee ofthe Wonju College of Medicine, YonseiUniversity.

2.1. De6ice description

The prototype SST-CPR device is a simple me-chanical device consisting of two main elements: acentral piston and a circumferential thoracic strap.The central main piston depresses the sternum atthe same time as the thoracic strap constricts thethorax circumferentially (Fig. 1). The prototypeSST-CPR device is shown in (Fig. 2).

2.2. Animal experiment

2.2.1. Experimental preparationTwenty-four adult mongrel dogs weighing 19–

30 kg were anaesthetized with intramuscular ke-tamine (20 mg/kg for induction) and intermittentintravenous sodium pentobarbital (20 mg/kg formaintenance). After tracheal intubation, the dogswere ventilated with a volume-cycled respirator(Companion 2800 portable ventilator, Puritan-Bennett Corporation, USA). End-tidal CO2 wasmeasured with a rapid response mainstream cap-

Fig. 1. Schematic presentation of simultaneous sternothoracic cardiopulmonary resuscitation. During compression systole, thestrap constricts the thorax while the piston depresses the sternum. During compression diastole, the piston returns to it’s originalposition by recoiling force of the thorax of itself.

Page 3: Simultaneous sternothoracic cardiopulmonary resuscitation: A new method of cardiopulmonary resuscitation

S.O. Hwang et al. / Resuscitation 48 (2001) 293–299 295

Fig. 2. Prototype of the device performing simultaneoussternothoracic cardiopulmonary resuscitation. The deviceconsists of two main elements. Piston (A) in the center is todepress the sternum. Strap (B) is to constrict the thoraxcircumferentially. Strap is attached to both sides of the piston.When the piston is pushed down, it depresses the sternum andpulls on the thoracic strap.

haemodynamically after catheters were placed.Baseline data (including blood pressure, rightatrial pressure, pulmonary artery pressure, car-diac output, and end tidal CO2) were measuredbefore induction of ventricular fibrillation. Car-diac output was measured by thermodilutionmethod in triplicate. Ventricular fibrillation wasinduced by passing an AC current of 30–60 mAthrough the right ventricular pacing catheter for3–5 s. Ventricular fibrillation was confirmed byits characteristic pattern on the ECG monitorand disappearance of pulsatile aortic pressure.The ventilator was disconnected from the endo-tracheal tube. Ventricular fibrillation was allowedto continue for 4 min without chest compres-sions or ventilation.

2.2.2. Experimental procedureAfter an interval of 4 min without interven-

tion, either standard CPR or SST-CPR was ini-tiated. Standard CPR was performed on a groupof 12 randomly assigned dogs with an automaticmechanical resuscitator (Thumper®, Michigan In-struments, USA). The chest was compressed at arate of 80 cycles /min; the force of chest com-pression was adjusted to depress the sternum toa depth of 5 cm. The compression to relaxationratio was set at 1. SST-CPR was performed onanother group of 12 randomly assigned dogswith a SST-CPR device. SST-CPR was achievedby compressing the piston of the SST-CPRdevice overlying the sternum with the same auto-matic mechanical resuscitator as that used forstandard CPR. The rate and depth of chest com-pression during SST-CPR was the same as withstandard CPR. A ventilation was delivered afterevery fifth compression by a ventilation deviceequipped with an automatic mechanical resusci-tator that provided an inspired oxygen concen-tration of 100%. Epinephrine was injected as a 1mg bolus into the right atrium at the beginning,and 3 min after the initiation, of CPR. CPR wasperformed for 6 min. Right atrial pressure, aor-tic pressure, and end tidal CO2 were recordedand cardiac output was measured at 1.5 min and4.5 min after initiation of CPR. Defibrillationwas not attempted after completion of the proto-col. An autopsy was performed to verify thecatheter position and to document any complica-tions associated with resuscitation.

nograph (Tidal wave Novametrix capnography,Novametrix, USA). The initial tidal volume was12 ml/kg with a respiratory rate of 14 breaths/min and an inspired oxygen fraction of 0.4. Res-piratory rate was adjusted to maintain theend-tidal CO2 concentration at approximately 35mmHg. After shaving the thorax, limbs, andboth cervical areas, the cardiac rhythm wasmonitored continuously. Micromanometer-tippedcatheters (Microtip catheter transducer SPC-350,5 Fr, 120 cm, Millar Instruments, USA) zeroedand calibrated at 37°C were placed into the rightatrium and ascending aorta from a femoral cut-down approach. A Swan-Ganz catheter (7 Fr.Swan-Ganz Catheter, Arrow international Inc.,USA) was introduced to the pulmonary arteryfrom a jugular venous approach to measure car-diac output. Catheter position was verified bythe presence of typical pressure waves. A bipolarpacing catheter was placed into the right ventri-cle to induce ventricular fibrillation. All dogswere kept in the supine position during the ex-periment. The mean right atrial pressure wasmaintained at 591 mmHg by infusing normalsaline intravenously as necessary. Data (includingthe cardiac rhythm, aortic pressure, right atrialpressure, and pulmonary artery pressure) wererecorded with a computerized data acquisitionsystem (MacLab/4S data acquisition system,ADI Instruments, USA) using a sampling rate of200 Hz. Animals were allowed to stabilize

Page 4: Simultaneous sternothoracic cardiopulmonary resuscitation: A new method of cardiopulmonary resuscitation

S.O. Hwang et al. / Resuscitation 48 (2001) 293–299296

Table 1Baseline measurements of experimental animalsa

S-CPR group SST-CPRgroup (n=12)(n=12)

Body weight (kg) 23.695.124.394.0Chest circumference (cm) 59.196.5 57.198.5Systolic blood pressure 153918 148920

(mmHg)115916Diastolic blood pressure 112914

(mmHg)Mean arterial pressure 134916 130915

(mmHg)493Right atrial pressure 493

(mmHg)Cardiac output (L/min) 3.5191.123.4491.03

31.096.6End tidal CO2 (mmHg) 34.097.6

a Data are given as mean9S.D. No statistical significancebetween two groups. S-CPR: standard cardiopulmonary re-suscitation; SST-CPR: simultaneous sterno-thoracic car-diopulmonary resuscitation.

2.3. Analysis

Haemodynamic data were analyzed using anaveraged sample of five consecutive compressions inthe data-sampling period. Coronary perfusion pres-sure was calculated by subtracting the right atrialpressure from the aortic pressure during diastole.All data are presented as mean9S.D. Differencesin aortic pressure, coronary perfusion pressure, endtidal CO2 tension, and cardiac output generated bystandard CPR and SST-CPR were tested with theStudent’s t-test. Statistical analysis was performedusing a SPSS statistical software package (SPSS forwindows 8.0, SPSS Inc., Chicago, IL). A P-valueB0.05 was considered to be statistically significant.

3. Results

Baseline data were not significantly differentbetween the two group (Table 1). In the aorticpressure curve produced by SST-CPR, the pressureincreased rapidly similar to the pattern noted duringstandard CPR early in compression systole; thepressure then reached a peak and plateau as theaortic pressure increased during continued chestcompression (Fig. 3). Mean aortic pressure was68.9916.1 mmHg with SST-CPR and 30.5910.0mmHg with standard CPR (PB0.01). Systolicaortic pressure was 99.7918.9 mmHg with SST-CPR, which was higher than the 42.4912.4 mmHgnoted with standard CPR (PB0.01). Diastolicaortic pressure was 51.3910.4 mmHg with SST-CPR, which was higher than the 23.298.0 mmHgnoted with standard CPR (PB0.01). SST-CPRgenerated higher coronary perfusion pressures thandid standard CPR (47.0911.4 mmHg with SST-CPR, 17.398.9 mmHg with standard CPR (PB0.01). End tidal CO2 tension was recorded at11.696.1 mmHg with SST-CPR, which was higherthan 2.1793.3 mmHg with standard CPR (PB0.01). Cardiac output was 0.6490.59 L/min withSST-CPR and 0.3090.42 L/min with standardCPR (P\0.05). Another interesting finding wasthat right atrial pressure was below zero mmHg inthe early relaxation period during SST-CPR (Fig.4). Negative right atrial pressure may facilitatethoracic filling of blood and coronary perfusion. Atautopsy, SST-CPR did not increase the frequencyof complications compared with standard CPR(standard CPR complications included two animalswith isolated rib fracture and one animal with a lung

Fig. 3. An illustration of aortic pressure curve with simulta-neous sternothoracic cardiopulmonary resuscitation (SST-CPR, thick line) compared with standard cardiopulmonaryresuscitation (S-CPR, thin line). SST-CPR generated a dis-tinct pressure waveform with a peak and plateau patternduring the systole.

Fig. 4. Right atrial pressure from simultaneous sternothoraciccardiopulmonary resuscitation. Negative right atrial pressurewas observed at early relaxation phase, which may facilitatethoracic filling.

Page 5: Simultaneous sternothoracic cardiopulmonary resuscitation: A new method of cardiopulmonary resuscitation

S.O. Hwang et al. / Resuscitation 48 (2001) 293–299 297

contusion accompanied by a rib fracture; SST-CPR complications included two dogs with iso-lated rib fracture, one dog with a lung contusion,and one dog with a lung contusion accompaniedby a rib fracture). However, one of the lungcontusions in the SST-CPR group was severe.

4. Discussion

Even though there has been debate as towhether precordial compression can compress theheart directly, clinical investigators have observeddirect cardiac compression on transesophagealechocardiography during CPR [10] [11]. Changesin the intrathoracic pressure without direct cardiaccompression can also generate blood flow [12] andimprove survival from prolonged cardiac arrest[13]. The SST-CPR device was designed to exploitboth mechanisms of blood flow during CPR. Infact, SST-CPR generated a different aortic pres-sure curve pattern compared to that seen withconventional CPR, which may reflect a combina-tion of both cardiac and thoracic mechanisms ofblood flow.

Addition of a thoracic strap to standard sternalcompression theoretically could yield certainbenefits over the use of standard CPR. In thisstudy, a significant increase in systolic aortic pres-sure was observed at the same compression depthwith SST-CPR compared to standard CPR. Thissystolic aortic pressure increase is likely to be dueto further enchancement of intrathoracic pressureand/or prolongation of the duration of the rise inintrathoracic pressure. We also found that theright atrial pressure dropped below 0 mmHg at avery early phase of the relaxation period duringSST-CPR. This finding suggests that a negativeintrathoracic pressure is generated when thoracicconstriction is released. Such a negative intratho-racic pressure during diastole facilitates venousthoracic filling, thus potentially improving bloodflow.

Mathematical models [14] and studies in hu-mans [15] suggest that use of a higher than normalcompression force or depth could increase bloodflow by increasing the fluctuation in intrathoracicpressure during standard CPR. However, increas-ing the sternal compression force is likely to resultin a higher incidence of internal organ injury.With SST-CPR, comparable or better haemody-

namic effects are expected without an increase incompression depth and resultant internal organinjury. In this study, incidence of internal organinjury with SST-CPR was not higher than stan-dard CPR.

SST-CPR generated a higher mean aortic andcoronary perfusion pressure, and improved pul-monary perfusion, compared to standard CPR inthis canine cardiac arrest model. SST-CPR gener-ated an average coronary perfusion pressure of47911 mmHg, which was significantly higherthan that generated by standard CPR (1799mmHg). The higher coronary perfusion pressurenoted with SST-CPR could provide better myocar-dial blood flow than standard CPR, because my-ocardial blood flow is pressure-dependent duringcardiac arrest unless critical coronary obstructionor occlusion is present [16]. End tidal CO2 tension,which is known to correlate with blood flow,coronary perfusion pressure, and survival, [17,18]was also significantly higher with SST-CPR com-pared to standard CPR. These results suggest thatSST-CPR might be a more effective techniquethan conventional CPR.

Several forms of noninvasive experimental CPRincluding vest CPR, ACD-CPR, and IAC CPRhave been developed to augment blood flow car-diac arrest. Vest CPR, which is performed byrhythmical inflation and deflation with a circum-ferential pneumatic vest, produces sufficient coro-nary perfusion pressure to increase survivalcompared to the use of conventional CPR in thedog. In addition to haemodynamic efficacy, vestCPR produces less chest wall trauma than stan-dard CPR because the pneumatic vest displacesthe sternum B0.8 cm [19]. Active compression–decompression CPR using a hand-held suctiondevice also increased systolic blood pressure, car-diac output, and end tidal CO2 tension in humans[20]. Active decompression results in negative in-trathoracic pressure, improved venous return with-out increasing right atrial pressure, and anincreased pressure difference between the aortaand the right atrium, which facilitates coronaryperfusion. Interposed abdominal compressionCPR is a variant of standard CPR that simulatesthe haemodynamic effects of intra-aortic ballooncounterpulsation [21].

SST-CPR has several potential advantages com-pared with these alternative CPR techniques. SST-CPR is a very simple variation of standard CPR.

Page 6: Simultaneous sternothoracic cardiopulmonary resuscitation: A new method of cardiopulmonary resuscitation

S.O. Hwang et al. / Resuscitation 48 (2001) 293–299298

The technique requires minimal training. Wespeculate that SST-CPR not only augments fluc-tuation of intrathoracic pressure by additionaland prolonged constriction of the thorax, butalso includes precordial compression to take ad-vantage of any ‘cardiac pump’ effect that maybe occurring in an individual patient. Generationof negative right atrial pressure during early re-laxation also increases the aortic-right atrialpressure gradient, which facilitates coronary per-fusion and venous return to the right heart. Inaddition to the haemodynamic advantages, SST-CPR might be expected to have an effect onventilation because constriction and relaxation ofthe thorax with an encircling strap might affectchest compliance, airway pressures, and tidalvolumes. The ventilatory effects of SST-CPR areunder investigation.

There were a number of limiations of thepresent study. The device used was a prototypemade out of aluminium. The strap used for cir-cumferential chest constriction was a polyesterband. The width of the band was chosen basedon a small number of pilot studies in the animalmodel prior to this experiment. However, furtherinvestigation is needed to optimize the designcharacteristics of the device and the strap to op-timize energy transfer.

Our results cannot be transferred to humansat this time because of the obvious differences inchest wall configuration between the two species.However, these results suggest that synergism be-tween the thoracic and cardiac pump mecha-nisms may be possible. Further studies areneeded to assess the potential clinical value ofthis finding.

Acknowledgements

This study was presented in part at the Ameri-can Heart Association 71st Scientific Sessions,Dallas, TX, November 1998. This study wassupported by a grant (c HMP 97-E-4-0013) ofthe Good Health R&D Project, Ministry ofHealth and Welfare, R.O.K

References

[1] Paradis NA, Martin GB, Rivers EP, Goetting MG,Appleton TJ, Feingold M, Nowak RM. Coronary perfu-

sion pressure and the return of spontaneous circulationin human cardiopulmonary resuscitation. J Am MedAssoc 1990;23:1106–13.

[2] Fitzgerald KR, Babbs CF, Frissora HA, Davis RW,Silver DI. Cardiac output during cardiopulmonary re-suscitation at various compression rates and durations.Am J Physiol 1981; H442–H448.

[3] Voorhees WD, Babbs CF, Tacker WA, Jr. Regionalblood flow during cardiopulmonary resuscitation indogs. Crit Care Med 1980;8:134–6.

[4] Rubertsson S, Grenvik A, Zemgulis V, Wiklund L.Systemic perfusion pressure and blood flow before andafter administration of epinephrine during experimentalcardiopulmonary resuscitation. Crit Care Med1995;23:1984–96.

[5] Lindner KH, Prengel AW, Pfenninger EG, Lindner IM,Strohmenger HU, Georgieff M, Lurie KG. Vasopressinimproves vital organ blood flow during closed-chestcardiopulmonary resuscitation in pigs. Circulation1995;91:215–21.

[6] Lurie KG, Shultz JJ, Callaham ML, Schwab TM, GischT, Rector T, Frascone RJ, Long L. Evaluation of activecompression–decompression CPR in victims of out-of-hospital cardiac arrest. J Am Med Assoc1994;271:1405–11.

[7] Mauer D, Schneider T, Dick W, Withelm A, Elich D,Mauer M. Active compression–decompression resuscita-tion: a prospective, randomized study in a two-tieredEMS system with physicians in the field. Resuscitation1996;33:125–34.

[8] Sack JB, Kesselbrenner MB, Bregman D. Survival fromin-hospital cardiac arrest with interposed abdominalcounterpulsation during cardiopulmonary resuscitation.J Am Med Assoc 1992;267(15):379–85.

[9] Halperin HR, Tsitlik JE, Gelfand M, Weisfeldt ML,Gruben KG, Levin HR. A preliminary study of car-diopulmonary resuscitation by circumferential compres-sion of the chest with use of a pneumatic vest. N Eng JMed 1993;329:762–8.

[10] Higano ST, Oh JK, Ewy GA. The mechanism of bloodflow during closed chest cardiac massage in humans:transesophageal echocardiographic observations. MayoClin Proc 1990;65:1432–40.

[11] Ma MH, Hwang JJ, Lai LP, Wang SM, Huang GT,Shyu KG, Ko YL, Lin JL, Chen WJ, Hsu KL. Trans-esophageal echocardiographic assessment of mitral valveposition and pulmonary venous flow during cardiopul-monary resuscitation in humans. Circulation1995;92:854–61.

[12] Criley JM, Blaufuss AN, Kissel GL. Cough inducedcardiac compression. J Am Med Assoc 1976;13:1246–50.

[13] Halperin HR, Guerci AD, Chandra N, Herskowitz A,Tsitlik JE, Niskanen RA, Wurmb E, Weisfeldt ML. Vestinflation without simultaneous ventilation during cardiacarrest in dogs: improved survival from prolonged car-diopulmonary resuscitation. Circulation 1986;74:1407–15.

[14] Talley DB, Ornato JP, Clarke AM. Computer-aidedcharacterization and optimization of the Thumper com-pression waveform in closed-chest CPR. Biomed In-strum Technol 1990;24:283–8.

Page 7: Simultaneous sternothoracic cardiopulmonary resuscitation: A new method of cardiopulmonary resuscitation

S.O. Hwang et al. / Resuscitation 48 (2001) 293–299 299

[15] Ornato JP, Levine RL, Young DS, Racht EM, Gar-nett AR, Gonzalez ER. The effect of applied chestcompression force on systemic arterial pressure andend-tidal carbon dioxide concentration during CPR inhuman beings. Ann Emerg Med 1989;18:732–7.

[16] Klocke FJ, Ellis AK. Control of coronary blood flow.Annu Rev Med 1980;31:489–508.

[17] Gudipati CV, Weil MH, Bisera J, Deshmukh HG,Rackow EC. Expired carbon dioxide: a noninvasivemonitor of cardiopulmonary resuscitation. Circulation1988;77:234–9.

[18] Levine RL, Wayne MA, Miller CC. End-tidal carbondioxide and outcome of out-of-hospital cardiac arrest.N Engl J Med 1997;31:301–6.

[19] Halperin H, Tsitlik J, Guerci A, Mellits ED, LevinH, Shi AY, Chandra N, Weisfeldt ML. The deter-minants of blood flow to vital organs during car-diopulmonary resuscitation in dogs. Circulation1986;73:539.

[20] Cohen TJ, Tucker KJ, Lurie KG, Redberg RF, Dut-ton JP, Dwyer KA, Scgwab RM, Chinn MC, GelbAM, Scheinman MM, Schiller NB, Callaham ML. Ac-tive compression–decompression. A new method ofcardiopulmonary resuscitation. J Am Med Assoc1992;267:2916–41.

[21] Babbs CF, Tacker WA, Jr. Cardiopulmonary resusci-tation with interposed abdominal compression. Circu-lation 1986;74:IV37–41.

.