Body weight does not affect defibrillation, resuscitation, or survivalin patients with out-of-hospital cardiac arrest treated with anonescalating biphasic waveform defibrillator

Roger D. White, MD; Thomas H. Blackwell, MD; James K. Russell, PhD; Dawn B. Jorgenson, PhD

Biphasic waveforms were intro-duced in automated externaldefibrillators for transthoracicdefibrillation in 1996 (1). A

subsequent larger clinical study reportednonescalating biphasic truncated expo-nential waveform shocks terminated ven-tricular fibrillation in out-of-hospital car-diac arrest patients with high efficacy andhigh patient discharge rates (2). Biphasic

waveforms previously had replaced tradi-tional monophasic waveforms for im-planted defibrillators because of demon-strated advantages in shock efficacy atlower energies, i.e., lower defibrillationthresholds and a broader safety margin(3–6). The lower required energies, inparticular, facilitated the miniaturizationof implanted defibrillators. The same ad-vantage was implemented in the firsttransthoracic (external) biphasic defibril-lators, which used a fixed energy protocolat an energy far lower (150 J) than tradi-tional maximum energies in externalmonophasic devices (360 J). This innova-tion allowed for miniaturization of auto-mated external defibrillators, facilitatinggreater mobility and widespread deploy-ment for the purpose of reducing time todefibrillation. These benefits have led towidespread adoption of biphasic wave-forms by the automated external defibril-lator manufacturers.

Concern over optimal treatment forobese patients continues as obesity hasbeen demonstrated to be a risk factor forchronic heart disease. Recent data sug-gest a continued increase in the preva-

lence of obese and overweight adults inthe United States and other developedcountries (7). Although the effect of bodyweight on defibrillation success has beenstudied with monophasic waveforms andin animals, it has not been reported inhumans treated with biphasic waveforms.Historically, one rationale for energy es-calation with monophasic waveforms hasbeen the belief that obese patients orthose with high transthoracic impedance(TTI) will require more energy for defi-brillation. This study examined the influ-ence of body weight on defibrillation, re-suscitation, and survival in out-of-hospital cardiac arrest patients, in whoma nonescalating, impedance-compensat-ing, 150-J biphasic waveform defibrillatorwas used for resuscitation. The study alsoexamined the relationship between TTIand body weight.

METHODS

Out-of-hospital cardiac arrest data during a6-yr period from December 1996 through De-cember 2002 from Rochester, MN, emergencymedical systems was included in this retro-

From The Mayo Clinic College of Medicine, Roch-ester, MN (RDW); the Department of Emergency Med-icine, Carolinas Medical Center, The Mecklenburg EMSAgency, Charlotte, NC (THB); Philips Medical Systems,Seattle, WA (JKR, DBJ).

Two of the authors (DBJ and JKR) are employeesof Philips Medical Systems, the manufacturer of thedefibrillator. The other authors (RDW and THB) ac-cepted no funding or support from Philips MedicalSystems.

Presented, in part, as an abstract at the meeting ofthe American Heart Association, November 2003, Or-lando FL.

Address requests for reprints to: Roger D. White,The Mayo Clinic College of Medicine, 200 First StreetSW, Rochester, MN 55905.

Copyright © 2004 by the Society of Critical CareMedicine and Lippincott Williams & Wilkins

DOI: 10.1097/01.CCM.0000139460.25406.78

Background: This is a study of the influence of body weight ondefibrillation, resuscitation, and survival in patients with out-of-hospital cardiac arrest treated with a nonescalating impedance-compensating 150-J biphasic waveform defibrillator.

Methods: Cardiac arrest data from Rochester, MN, emergencymedical services over a 6-yr period was retrospectively analyzed.Patient weight data were available for 62 of the 68 patients whowere treated initially by basic life support personnel and whopresented with a shockable rhythm. For each defibrillation andresuscitation outcome variable, we tested for differences in bodyweight for successful vs. unsuccessful outcome.

Results: Initial shocks defibrillated 92% (83% to 97%) of pa-tients. Cumulative success with two shocks was 98% (confidenceinterval, 92% to 100%) and with three shocks was 100% (confi-dence interval, 95% to 100%). The mean shock impedance was 90� 21 ohms. The average body weight was 84 � 17 kg (minimum,

53 kg; maximum, 135 kg) and was normally distributed. Based onthe body mass index for 46 patients, approximately 41% wereclassified as overweight (body mass index, >25), 24% obese(body mass index, >30), and 4% extremely obese (body massindex, >40). The remaining 31% were classified as normal orunderweight. First-shock success, cumulative success throughtwo shocks, and cumulative success through the first-shockseries were unrelated to body weight, as were basic life supportrestoration of spontaneous circulation, prehospital restoration ofspontaneous circulation, hospital admission, and discharge.

Conclusions: Overweight patients were defibrillated by thebiphasic waveform used in this study at high rates, with a fixedenergy of 150 J, and without energy escalation. (Crit Care Med2004; 32[Suppl.]:S387–S392)

KEY WORDS: defibrillation; automatic external defibrillator; ven-tricular fibrillation; resuscitation

S387Crit Care Med 2004 Vol. 32, No. 9 (Suppl.)

spective analysis of weight effects. The studywas conducted under Institutional ReviewBoard approval of an ongoing investigationevaluating patient outcomes after defibrilla-tion with biphasic waveform automated exter-nal defibrillators. The emergency medical sys-tem uses fixed low-energy biphasic automatedexternal defibrillators (ForeRunner, PhilipsMedical Systems, Seattle, WA) used by firstresponders (police and fire rescue personnel).Presenting and pre- and postshock rhythmswere reviewed for pre-advanced life supportshocks from 68 patients. Cases with known orpresumed cardiac origin and shockable initial(presenting) rhythms were included for anal-

ysis. Both witnessed and not witnessed arrestswere included.

Rhythms were categorized as shockable(ventricular fibrillation or shockable ven-tricular tachycardia) or nonshockable.Shocks were included for analysis if the pre-shock rhythms were shockable and therhythm at 5 secs following the shock was notindeterminate. The rhythm at 5 secs is theconventional criterion for the assessment ofshock efficacy, as defined by the 2000 Amer-ican Heart Association Guidelines (8) andused by defibrillation researchers (9 –14). Ashock was considered successful if the 5-secpostshock rhythm was nonshockable, in-

cluding asystole and rhythms with QRScomplexes.

Weight and height data were obtained forall admitted patients, either at the time ofadmission to the cardiac care unit or shortlythereafter at the time of coronary angiographyand/or echocardiography. For patients whowere not admitted, height and weight datawere obtained from clinic records within amaximum of 3 months of their event. Bodymass index (BMI) was calculated for patientswith available data and used to classify sub-jects as underweight, normal, overweight,obese, or extremely obese (7). Outcome wasderived from emergency medical systems runreports and hospital records. All patients pro-vided research authorization for hospitalrecord review. Defibrillation and resuscitationoutcome variables were examined vs. patientweight. Defibrillation outcome variables in-cluded initial shock success, cumulative two-shock success, and cumulative success of thefirst “stack” of shocks (a series of up to threeshocks before an interval for cardiopulmonaryresuscitation). Resuscitation outcomes, in-cluding return of spontaneous circulation(ROSC) before the administration of cardioac-tive drugs (called basic life support ROSC),ROSC before transfer to the emergency de-partment (any ROSC), survival to hospital ad-mission (admission), and survival to hospitaldischarge (survival) were examined for anyrelationship to body weight as well.

StatXact (version 5, Cytel Software, Cam-bridge, MA) was used to calculate the 95%confidence interval using the Blyth-Still-Casella method, and the Lilliefors test wasused to test the data for a normal distribution.Statistics for continuous variables were ana-lyzed using the Student’s t-test (Statistica,version 6, StatSoft, Tulsa, OK). p � .05 wasconsidered statistically significant.

RESULTS

Patient weight data were available for62 of the 68 patients (91%) who pre-sented with a shockable rhythm. Averageage was 66 � 14 yrs (mean � SD). Fifty-one (82%) were male. The average bodyweight was 84 � 17 kg (minimum, 53 kg;maximum, 135 kg) and was normally dis-tributed. Height data were available for46 of 68 patients (68%). BMI was calcu-lated as the weight in kilograms dividedby the square of height in meters (7).Based on BMI for the 46 patients withboth height and weight data, approxi-mately 41% were classified as overweight(BMI, �25), 24% were obese (BMI, �30),and 4% were extremely obese (BMI,�40). The remaining 31% were classifiedas normal or underweight.

Figure 1. First-shock success/failure vs. patient weight.

Table 1. Effect of weight on defibrillation, resuscitation, and survival

Weight Analysis

Overall %[95% ConfidenceInterval] n � 62

Weight, kg,Mean � SD

Successful/Yes

Weight, kg,Mean � SD

Failed/No p Value

1st shock 92 [83–97] 83 � 17 88 � 14 .74n � 57 n � 5

Cumm 2 shocks 98 [92–100] 84 � 17 68 1.0n � 61 n � 1

1st stack shock 100 [95–100] 84 � 17 NAn � 62 n � 0

BLS ROSC 37 [25–50] 85 � 18 83 � 16 .53n � 23 n � 39

Any ROSC 77 [66–87] 84 � 16 84 � 21 .19n � 48 n � 14

Admission 74 [62–84] 83 � 16 85 � 20 .30n � 46 n � 16

Discharge 44 [31–57] 86 � 17 82 � 17 .89n � 27 n � 35

Cumm, cumulative; BLS, basic life support; ROSC, return of spontaneous circulation.

S388 Crit Care Med 2004 Vol. 32, No. 9 (Suppl.)

Time from emergency call receipt tofirst shock (call to shock time) was 5.9 �1.9 mins; 85% of patients had a witnessedarrest. Average shock impedance was 90� 21 ohms. First-shock success was 92%;only one patient (weight, 68 kg) failed tobe converted with two or fewer shocks. Allpatients were converted within the first se-ries of three shocks. First-shock success

was not related to weight, i.e., unsuccessfulshocks were not associated with higherweights than successful shocks (Table 1;Fig. 1). The highest and lowest weight pa-tients were defibrillated on the first shock.Cumulative efficacy through two shockswas unrelated to weight (Fig. 2). There wasno statistically significant effect of weighton basic life support ROSC (Fig. 3), any

ROSC (Fig. 4), or at admission (Fig. 5) ordischarge (Fig. 6). Figure 7 shows the rela-tionship between impedance and weight.The squared correlation coefficient was0.22, reflecting only a modest influence ofweight on TTI.

DISCUSSION

Historical Perspective. The issue ofsafe, effective, and appropriate energy andcurrent required to defibrillate patientswith an external defibrillator has a longhistory. Successful defibrillation dependson adequate myocardial current density,delivered with an effective waveform. Themanner in which this current is deliveredover time determines the energy dose ofthe waveform. Too little current densitymay result in failed defibrillation,whereas too much energy has beenshown experimentally to result in myo-cardial dysfunction (15). Unfortunately,myocardial current density and energyare very difficult to quantify, even in an-imal studies, inasmuch as external defi-brillators have been characterized interms of the less relevant energy and cur-rent delivered at the defibrillator outputs,i.e., the transthoracic current. As a result,many of the resuscitation protocols usedin clinical practice are based on the com-monsense presumption that a decrease intransthoracic current delivered by the de-fibrillator (caused by an increase in TTI)may result in a decrease in myocardialcurrent density. This presumption, how-ever, is not supported by clinical data,and counterexamples are easily imagined,e.g., decreased transthoracic current isdue to a decrease in noncardiac currentflowing around the thoracic cage, leavingmyocardial current density largely un-changed. Because direct measurementsof myocardial current density in humansremain impractical, the relationship be-tween transthoracic and myocardial doseas a function of natural human variationsin body weight and TTI remains un-known. Consequently, it is appropriate toquestion conventional dose protocols, es-pecially when a therapy that compensatesfor the effects produced by impedance-induced waveform changes is used.

Escalating energy protocols are typicalfor monophasic defibrillation (16–20). Byincreasing the energy setting for a patientwith high TTI, the transthoracic current,and presumably the myocardial currentdensity, would be restored to the leveldelivered to a patient with low TTI. Thisbelief was also extended to obese patients,

Figure 2. Cumulative (CUMM) two-shock success/failure vs. patient weight.

Figure 3. Basic life support (BLS) return of spontaneous circulation (ROSC) success/failure vs. patientweight.

S389Crit Care Med 2004 Vol. 32, No. 9 (Suppl.)

based in part on the assumption that theywould also have increased TTI. In 1974,Tacker et al. (21) proposed a dose-response curve based on body weight inan effort to quantify the electrical energyneeded for defibrillation. They suggesteda starting level between 3.5 and 6.0 J/kgfor patients weighing �50 kg, and forheavier patients, the “full output of thedefibrillator” should be used. However,

Gascho and others (16, 20, 22–25) dis-puted their analysis, reporting that thedefibrillation rate did not correlate withweight. Gascho et al. (22, 23) reportedthat lower energy levels (under 3 J/kg)defibrillated more often than higher en-ergy levels. Gascho et al. further pointedout that animal studies were the basis ofmuch of the energy dose to body weightresearch, yet animals needed four to

eight times as much energy to defibrillateas humans of the same weight.

Predicting a patient’s TTI based onphysical characteristics has proven diffi-cult. Because TTI is the parameter thatdetermines the current output for a par-ticular monophasic energy setting, Ker-ber et al. (26) examined the relationshipbetween TTI and body weight. They re-ported a weak correlation (r2 � .20) be-tween body weight and TTI and a stron-ger correlation (r2 � .64) between TTIand chest width. Geuze and de Feijter(27) reported no correlation betweentransthoracic resistance and body weight.Our data are consistent with Geuze andde Feijter and Kerber et al., showing aweak correlation (r2 � .22) between TTIand body weight (Fig. 7).

In 1980, the American Heart Associa-tion reduced the recommended startingenergy for monophasic defibrillators for a70-kg subject from 420 J to 200 J (28).Questions regarding optimal energy andescalation protocols continued. An inves-tigation by Weaver et al. (29) in 1982supported even lower energy levels. Theystudied initial monophasic shocks of 175J and 320 J in 249 patients and concludedthat shocks using 175 J were as effectiveas the higher energy shocks. Further-more, the lower energy shocks were as-sociated with less postshock heart block.Currently, monophasic defibrillators fol-low an escalating energy protocol, despitethe lack of definitive data supporting theefficacy of additional energy levels for“hard to defibrillate” patients. This maybe due, in part, to the relatively low first-shock efficacy for monophasic waveforms(11). In 1998, an American Heart Associ-ation Guideline on waveforms stated, “Areview of previous American Heart Asso-ciation guidelines for the energy se-quence 200J–300J–360J reveals that theevidence supporting this reputed “goldstandard” is largely speculative and basedon commonsense extrapolation from an-imal data and human case series.” (30).

Factors Determining DefibrillationSuccess and Resuscitation Outcome.Concern over optimal treatment forobese patients has continued. Obesity hasbeen demonstrated to be a risk factor forchronic heart disease, and recent datasuggest a continued increase in the prev-alence of obese and overweight adults inthe United States and other developedcountries (7). We used BMI to classifypatients enrolled in the weight analysisand found that 41% were classified asoverweight and 28% as obese or ex-

Figure 5. Admission (ADMIT) success/failure vs. patient weight.

Figure 4. Any return of spontaneous circulation (ROSC) success/failure vs. patient weight.

S390 Crit Care Med 2004 Vol. 32, No. 9 (Suppl.)

tremely obese. Thus, many patients inthis study were overweight.

Out-of-hospital cardiac arrest studieshave reported high shock efficacy for bi-phasic waveforms (10, 11). However, theeffect of body weight on biphasic wave-form defibrillation in humans has notbeen studied previously. Zhang et al. (13)studied body weight as a predictor of bi-phasic shock success for defibrillation inswine and found that when energy levelsof 150 J or more were used, shock successdid not vary with body weight or TTI.Recent counter-arguments have cited astudy by Caffrey et al. (31) that describedthe use of 150-J biphasic waveform auto-mated external defibrillators in airportpassenger terminals. Of the 18 ventricu-lar fibrillation patients in the study, three

who were described as diabetic and obesewere reported to have “remained in fibril-lation despite a rapid response.” Caffreyclarified that ventricular fibrillation wassuccessfully terminated by 150-J biphasicshocks in all three patients but that ven-tricular fibrillation recurred frequentlyand was the final recorded rhythm aftermultiple shocks (S, Caffrey-Villari, per-sonal communication, December 26,2002). Furthermore, 16 of 19 shocks de-livered to these patients by first respond-ers successfully terminated ventricular fi-brillation for periods ranging from 5 secsto �10 mins. Thus, poor outcome forthese patients is not attributable to fail-ure of the defibrillation shock but is likelyrelated to the underlying disease and con-dition of the patient.

In this study, the lack of impact ofbody weight on defibrillation efficacy andresuscitation outcome indicates thatthere is no evident need for weight-basedenergy escalation and the associated pos-sible risk of shock-induced cardiac dys-function (29, 32–34). Neither defibrilla-tion nor patient outcome depended on(or were related to) patient weight. Thegreater efficacy of biphasic waveforms atlower energies compared with monopha-sic waveforms at higher energies illus-trates the shortcoming of applying his-torical dose regimens to new therapies.

This study is limited in that it is retro-spective and involves only one site. Height

data used to calculate BMI was not availablefor all patients. Nonetheless, the percentageof overweight patients, based on the BMIsof available patients, was large (69% wereoverweight or obese). Furthermore, a sin-gle biphasic waveform was evaluated, andthe conclusions cannot necessarily be gen-eralized to other biphasic waveforms.

CONCLUSIONS

First-shock efficacy and subsequentshock success, resuscitation, and survivalwere not related to body weight for pa-tients in this analysis. The fixed-energyprotocol used by this automated externaldefibrillator incorporating a 150-J imped-ance-compensating biphasic truncatedexponential waveform in this basic lifesupport use setting appears appropriateand effective.

ACKNOWLEDGMENTS

The authors thank Dom Maniace forcontributions in auditing the databaseand assisting in literature searches.

REFERENCES

1. White RD: Early out-of-hospital experiencewith an impedance-compensating low-energy biphasic waveform automatic exter-nal defibrillator. J Interv Card Electrophysiol1997; 1:203–208

2. White RD, Hankins DG, Atkinson EJ: Patientoutcomes following defibrillation with a lowenergy biphasic truncated exponential wave-form in out-of-hospital cardiac arrest. Resus-citation 2001; 49:9–14

3. Jones JL, Swartz JF, Jones RE, et al: Increas-ing fibrillation duration enhances relativeasymmetrical biphasic versus monophasicdefibrillator waveform efficacy. Circ Res1990; 67:376–384

4. Jones JL, Jones RE: Decreased defibrillator-induced dysfunction with biphasic rectangu-

Figure 7. Correlation between patient weight (kg)and impedance (ohms).

Figure 6. Discharge success/failure vs. patient weight.

F irst-shock efficacy

and subsequent

shock success, re-

suscitation, and survival

were not related to body

weight for patients in this

analysis.

S391Crit Care Med 2004 Vol. 32, No. 9 (Suppl.)

lar waveforms. Am J Physiol 1984; 247:H792–H796

5. Jones JL, Jones RE: Improved defibrillatorwaveform safety factor with biphasic wave-forms. Am J Physiol 1983; 245:H60–H65

6. Winkle RA, Mead RH, Ruder MA, et al: Im-proved low energy defibrillation efficacy inman with the use of a biphasic truncatedexponential waveform. Am Heart J 1989;117:122–127

7. Flegal KM, Carroll MD, Ogden CL, et al: Prev-alence and trends in obesity among US adults,1999–2000. JAMA 2002; 288:1723–1727

8. AHA Guidelines 2000 for CardiopulmonaryResuscitation and Emergency CardiovascularCare: An international consensus on science.Circulation 2000; 102(Suppl)

9. Gliner BE, White RD: Electrocardiographicevaluation of defibrillation shocks deliveredto out-of-hospital sudden cardiac arrest pa-tients. Resuscitation 1999; 41:133–144

10. van Alem AP, Chapman FW, Lank P, et al: Aprospective, randomized and blinded com-parison of first shock success of monophasicand biphasic waveforms in out-of-hospitalcardiac arrest. Resuscitation 2003; 58:17–24

11. Schneider T, Martens PR, Paschen H, et al:Multicenter, randomized, controlled trial of150-J biphasic shocks compared with 200- to360-J monophasic shocks in the resuscita-tion of out-of-hospital cardiac arrest victims.Circulation 2000; 102:1780–1787

12. Clark CB, Zhang Y, Davies LR, et al: Trans-thoracic biphasic waveform defibrillation atvery high and very low energies: A compari-son with monophasic waveforms in an ani-mal model of ventricular fibrillation. Resus-citation 2002; 54:183–186

13. Zhang Y, Clark CB, Davies LR, et al: Bodyweight is a predictor of biphasic shock suc-

cess for low energy transthoracic defibrilla-tion. Resuscitation 2002; 54:281–287

14. Calle PA, Monsieurs KG, Buylaert WA: Equiv-alence of the standard monophasic waveformshocks delivered by automated external defi-brillators? Resuscitation 2002; 53:41–46

15. Tang W, Weil MH, Sun S, et al: The effects ofbiphasic waveform design on post-resuscita-tion myocardial function. J Am Coll Cardiol2004; 43:1228–1235

16. Kerber RE, Jensen SR, Gascho JA, et al: De-terminants of defibrillation: Prospectiveanalysis of 183 patients. Am J Cardiol 1983;52:739–745

17. Moulton C, Dreyer C, Dodds D, et al: Place-ment of electrodes for defibrillation: A reviewof the evidence. Eur J Emerg Med 2000;7:135–143

18. Adgey AAJ, Patton JN, Campbell NPS, et al:Ventricular defibrillation: Appropriate en-ergy levels. Circulation 1979; 60:219–223

19. Lown B: The energy for ventricular defibril-lation: Too little or too much? N Engl J Med1978; 1252–1253

20. Crampton R: Accepted, controversial, andspeculative aspects of ventricular defibrilla-tion. Prog Cardiovasc Dis 1980; 23:167–186

21. Tacker WA Jr, Galioto FM Jr, Giuliani E, et al:Energy dosage for human trans-chest elec-trical ventricular defibrillation. N Engl J Med1974; 290:214–215

22. Gascho JA, Crampton RS, Sipes JN, et al: En-ergy levels and patient weight in ventriculardefibrillation. JAMA 1979; 242:1380–1384

23. Gascho JA, Crampton RS, Cherwek ML, et al:Determinants of ventricular defibrillation inadults. Circulation 1979; 60:231–240

24. Patton JN, Pantridge JF: Current required forventricular defibrillation. BMJ 1979;1:513–514

25. Adgey AAJ, Campbell NPS, Webb SW, et al:Transthoracic defibrillation in the adult. MedInstr 1978; 17–19

26. Kerber RE, Grayzel J, Hoyt R, et al: Trans-thoracic resistance in human defibrillation:Influence of body weight, chest size, serialshocks, paddle size and paddle contact pres-sure. Circulation 1981; 63:676–682

27. Geuze RH, de Feijter PJ: Evaluation of trans-thoracic countershock with initial energylevels up to 200 J in a coronary care unit.J Electrocardiol 1985; 18:251–258

28. Standards and guidelines for cardiopulmo-nary resuscitation (CPR) and emergency car-diac care (ECC). JAMA 1980; 244:453–509

29. Weaver WD, Cobb LA, Copass MK, et al: Ven-tricular defibrillation: A comparative trial us-ing 175-J and 320-J shocks. N Engl J Med1982; 307):1101–1106

30. Cummins RO, Hazinski MF, Kerber RE, et al:Low-energy biphasic waveform defibrillation:Evidence-based review applied to emergencycardiovascular care guidelines. Circulation1998; 97:1654–1667

31. Caffrey SL, Willoughby PJ, Pepe PE, et al:Public use of automated external defibrilla-tors. N Engl J Med 2002; 347:1242–1247

32. Reddy RK, Gleva MJ, Gliner BE, et al: Bipha-sic transthoracic defibrillation causes fewerECG ST-segment changes after shock. AnnEmerg Med 1997; 30:127–134

33. Tang W, Weil MH, Sun S: Low-energy bipha-sic waveform defibrillation reduces the sever-ity of postresuscitation myocardial dysfunc-tion. Crit Care Med 2000; 28(11 Suppl):N222–N224

34. Xie J, Weil MH, Sun S, et al: High-energydefibrillation increases the severity of pos-tresuscitation myocardial dysfunction. Cir-culation 1997; 96:683–688

S392 Crit Care Med 2004 Vol. 32, No. 9 (Suppl.)