Truncated Ex Sinusoida P

onential Versus Damped Waveform Shocks for

Transthoracic Defibrillation Joseph C. Behr, BBA, Lori L. Hartley, RN, Douglas K. York, EMT-P, Donald D. Brown, MD,

and Richard E. Kerber, MD

Currently available transthoracic defibrillators use either a damped sinusoidal or truncated exponential (TE) waveform. Truncated exponential waveforms deliver a long pulse if the transthoracic impedance is high; it has been suggested that such a long pulse may be less ef- fective for defibrillation. Our objective was to compare the ability of damped sinusoidal (DS) waveform shocks versus TE waveform shocks to terminate ventricular fib- rillation (VF) and achieve survival from witnessed car- diac arrest. We retrospectively reviewed field-recorded electrocardiograms from 86 patients with witnessed VF, treated by prehospital personnel equipped with DS or TE waveform defibrillators. Forty-four patients received 130 shocks from TE defibrillators; 42 patients received 108 shocks from DS defibrillators. There were no signif-

icant differences in time from arrest to first shock (8.0 vs 8.1 minutes), nor were there any differences in the size of the communities involved. The shocks resulted in the following rhythms: organized rhythm: TE: 15 of 130 (12%), DS: 24 of 108 (22%), p = 0.10 (NS); persistent VF: TE: 85 of 130 (65%), DS: 45 of 108 (42%), p <O.Ol; asystole: TE: 30 of 130 (23%), DS: 39 of 108 (360/o), p = NS; and survival to hospital discharge: TE: 5 of 44 (11 o/o), DS: 8 of 42 (19%~)~ p = NS. We conclude that DS waveforms terminated VF more frequently than TE, but there was no significant difference in resumption of an organized rhythm or survival. A prospective comparison of these 2 waveforms is needed. 0 I996 by Excerpta Medica, Inc.

(Am J Cardiol 1996;78: 1242- 1245)

E arly transthoracic defibrillation has been shown to be a major component of successful cardiac

resuscitation.l,* It is crucial to deliver the first shock as rapidly as possible. The widespread application of this principle relies heavily on the automated exter- nal defibrillator (AED)3-5 used by nontraditional re- sponders (“public access defibrillation”6). Shock waveform may be an important determinant of shock success. Commercially available defibrillators, in- cluding AEDs, utilize 2 basic categories of electric waveforms. Most machines employ a damped sinu- soidal (DS) waveform, illustrated in Figure 1. These waveforms operate on the principle of damped ca- pacitor discharge.7 The other category is the trun- cated exponential (TE) or trapezoidal waveform (Figure 1) that operates on the principle of direct discharge of a capacitor which is electrically trun- cated after a discrete period of time in which a pre- cise amount of energy is delivered.*,9 Direct com- parisons of the effectiveness of DS versus TE waveforms have not previously been presented. In- direct comparisons can be made from several clinical reports.rO,ll The present study compares the ability of DS and TE waveforms to terminate ventricular fibrillation (VF) and achieve survival from prehos- pita1 cardiac arrest.

From the Department of Internal Medicine and the Emergency Medi- cal Services Learning Resource Center, University of Iowa Hospitals and Clinics, Iowa City, Iowa. Manuscript received March 1 2, 1996; revised manuscript received and acceptedjune IO, 1996.

Address for reprints: Richard E. Kerber, MD, Department of Inter- nal Medicine, University of Iowa Hospital, 200 Hawkins Drive, Iowa City, Iowa 52242

1242 0 1996 by Excerpta Medica, Inc All rights reserved.

METHODS This study was performed by retrospectively re-

viewing cardiac arrest data reports filed with the Uni- versity of Iowa Emergency Medical Services Learn- ing Resource Center. This database includes all prehospital cardiac arrests in the state of Iowa be- tween March 1988 and June 1994. Field-recorded preshock and postshock electrocardiograms (ECGs) were recorded on casette tapes integrated into the defibrillators and submitted; a hardcopy ECG print- out was generated for review. Patients were not iden- tified by name on these tapes, and their names were not available to or known to the investigators. Pa- tients received shocks from external defibrillators us- ing either the TE waveform: Heartstart 1000, Heart- start 3000 (HSlOOO, HS3000, Laerdal, At-monk, New York) and MRL 450-SL (Medical Research Labs, Buffalo, Illinois), or the DS waveform: Heartstart 2000 (HS2000, Laerdal, Armonk, New York), and Lifepak 5 (LP5, PhysioControl, Redmond, Washing- ton). Forty-four cases and 42 controls met the fol- lowing inclusion criteria: witnessed cardiac arrests, with initial rhythm of VF documented by electro- cardiographic recordings at the scene, and time from collapse to first defibrillatory shock < 15 minutes (re- cording times as defined by the Utstein style’*). Shocks given for VF were defined as “successful” if the shock resulted in its termination. Resulting rhythms were categorized as an organized rhythm, asystole, or persistent VF. Shocks were administered in accordance with the American Heart Associa- tion’s Advanced Cardiac Life Support guidelines.

Additional information obtained included the age and sex of each patient; the distance (miles) from the ambulance service base to the cardiac arrest site; the

0002.9 149/96/s 15.00 PII SOOO2-9149(96)00603-O

population of the ambulance service community, the I time from collapse to first defibrillatory shock and the ultimate survival (i.e., hospital discharge). Some of the defibrillators used have the capability to mea- sure and record transthoracic impedance prior to shock . However, this particular information was not recorded by the individual ambulance services and was not available at the time of our review. The height and weight of the patients were also not re- ported.

After completion of the data collection, a staff cardiologist who had no knowledge of the nature of the study or of the previous diagnoses was asked to make a rhythm diagnosis of each preshock and postshock ECG rhythm strip. Based upon this inde- pendent review, 4 individual TE waveform shocks and 9 individual DS waveform shocks were elimi- nated from the data set because of technically un- satisfactory recordings containing excessive artifact, which precluded an accurate rhythm diagnosis.

Statistical analysis: We used Fisher’s 2-tail ex- act tests to compare the success rates of shocks given from the truncated exponential defibrillators (HSlOOO, HS3000, and MRL450-SL) with the suc- cess rate of the DS defibrillators (HS2000, LP5). Testing was performed separately for all shocks, for

100

80

60

3 z 40

20

0

-20

Damped Sinusoidal

:

I I I I I

5 10 15 20 25

Duration (milliseconds)

the first shock and for the first 3 shocks. Also, shocks were analyzed separately for whether the operators

FIGURE 1. Typical waveforms using damped sinusoidal and trun-

used self-adhesive pad electrodes or hand-held pad- cad exponential puIses.

dle electrodes. After these subgroups were identified, Fisher’s 2-tail exact tests were again performed to compare all shocks, first shock, and the first 3 shocks for the 2 waveforms.

Two-sample t tests were used to compare the av- erage age of the patients, as well as the time from collapse to the first shock, in the TE group with those of the DS group. Fisher’s 2-tail exact tests were again used to compare the percent of women, the percent of cases with 53 shocks, and the percent of victims discharged alive from hospital in the TE group with those of the DS group. Wilcoxon rank sum tests were used to compare the distance (miles) from ambu- lance service base to the cardiac arrest site, the pop- ulation of the ambulance service community, and the average number of shocks per case in the TE group with those of the DS group. These statistics were compiled for all shocks and were also separately compiled for shocks given from self-adhesive elec- trode pads and from hand-held paddle electrodes.

RESULTS A total of 86 patients receiving 238 shocks were

reviewed. The TE waveform group consisted of 44 patients receiving 130 shocks. The DS waveform group consisted of 42 patients receiving 108 shocks. No patient received shocks from 2 different defibril- lators. Of the 44 TE patients, 15 were treated by paramedic services, 25 by emergency medical tech- nician services, and 4 by first responder services. Of the 42 DS patients, 14 were treated by paramedic services, 25 by emergency medical technician ser- vices, and 3 by first responder services.

TABLE I Patient Demographics: Truncated Exponential Versus Damped Sinusoidal Waveform Shocks

Truncated Damped Exponential Sinusoidal

(n = 44) [n = 42)

Average age (yr) Women/men Population of community Distance from ambulance

base to arrest site (miles) Time from collapse to first

shock (min) Total shocks Patients receiving 53 shocks Average no. of shocks/arrest Patients discharged alive from

hospital

66 k 2.1 73 ” 1.6* 8/36 lo/32

8,755 ? 1,705 7,079 2 1,714 1.4 2 0.3 1.4 t 0.3

8.0 + 0.4 8.4 t 0.4

130 108 21 (48%) 17 (40%)

3.0 L 0.3 2.6 k 0.3 5 (11%) 8 (19%)

All values are expressed CIO mecln 2 SEM.

l p <O.Ol.

Table I compares the demographic characteristics of the patients in the TE category with those in the DS category. The percent of women, median size of the community, and miles from the ambulance ser- vice to the arrest site were not significantly different. The average age of the patients receiving defibrilla- tory shocks from DS waveforms was 73 years versus 66 years for the TE waveforms, which was statisti- cally significant, p <O.Ol. The 2 categories were not significantly different with respect to time from col- lapse to first shock, number of patients receiving 53 shocks, average number of shocks per arrest, and number of patients discharged alive from hospital.

ARRHYTHMIAS AND CONDUCTION DISTURBANCES/WAVEFORMS FOR TRANSTHORACIC DEFlBRlllATlON 1243

Table II compares the same parameters as Table I subdivided by self-adhesive pad electrodes and hand-held paddle electrodes. There were no statisti- cally significant differences in these subgroups with respect to demographics, with the exception of the average ages (see Table II).

TABLE II Patient Demographics: Truncated Exponential Versus Damped Sinusoidal Waveform Shocks, Subdivided by Use of Self-Adhesive Pad Electrodes and Hand-Held Paddle Electrodes

Truncated Damped Exponential Sinusoidal

(n = 29) (n = 28)

Self-adhesive pads Average age (yr) 64 + 3 71 k 2* Women/men 6/23 7/2 1

Population of community 2,745 + 460 1,840 + 330 Distance from ambulance 1.5 + 0.3 1.5 + 0.4

base to arrest site (miles) Time from collapse to first a.4 + 0.5 a.6 k 0.6

shack (min)

Total shocks 91 Patients receiving 53 I 4 (48%) 11:329%) shocks

Average no. of shocks/ 3.1 2 0.4 2.6 k 0.3 arrest

Patients discharged alive 2 (7%) 5 (18%) from hospital

Hand-held paddles (n = 15) (n = 14) Average age (yr) 70 2 3.5 77 ? 2.6 Women/men 2/13 3/l 1 Population of community 20,376 + 3,276 17,559 + 3,563 Distance from ambulance 1.4 + 0.4 1.4 + 0.3

base to arrest (miles) Time from collapse to first 7.2 k 0.6 a.1 k 0.6

shock (min) Total shocks 39 36 Patients receiving 53 7 (47%) 6 (43%)

shocks Average no. of shocks/ 2.6 k 0.6 2.6 k 0.5

arrest Patients discharged alive 3 (20%) 3 (21%)

from hospital

* p <0.05.

TABLE III Rhythms Resulting from Truncated Exponential Versus Damped Sinusoidal Waveform Shocks: All Shocks

All shocks (n = 238 shacks) Organized Persistent VF Asystole

All self-adhesive pad shocks (n = 163)

Organized Persistent VF Asystole

All hand-held paddle shocks (n = 75)

Organized Persistent VF Asystole

*p<0.01;+p=0.10.

VF = ventricular fibrillation.

15 (12%) 24 (22%)+ a5 (65%) 45 (42%)* 30 (23%) 39 (36%)

12 (13%) 16 (22%) 56 (62%) 31 (43%) 23 (25%) 25 (35%)

3 (a%) a (22%) 29 (74%) 14 (39%)*

7 (18%) 14 (39%)

Table III compares the resultant rhythms of those patients defibrillated with TE AEDs with those re- ceiving shocks from DS AEDs for all shocks. DS waveform AEDs had resultant rhythms of persistent VF 42% of the time versus 65% of the time for TE waveforms (p <O.Ol). This difference was also seen when comparing all hand-held paddle shocks: DS had persistent VF 39% of the time; TE had persisted ventricular fibrillation 74% (p <O.Ol). Although the percentage of shocks resulting in an organized rhythm and the percentage of shocks resulting in asystole was greater for DS then TE, these differ- ences were not statistically significant (for shocks resulting in organized rhythms, p = 0.10).

Tables IV and V examine resultant rhythms for first shocks only and for the first 3 shocks, respec- tively. With smaller sample sizes there are no statis- tically significant differences in Table IV. Table V shows that the resultant rhythm of persistent VF was more common in TE versus DS (54% vs 35%; p

TABLE IV Rhythms Resulting from Truncated Exponential Versus Damped Sinusoidal Waveform Shocks: First Shocks Only

Truncated Damped Resultant Rhythm Exponential Sinusoidal

All first shocks (n = 84)* Organized 3 (7%) 11 (27%) Persistent VF 24 (56%) 14 (34%) Asystole 16 (37%) 16 (39%)

All first shocks with self-adhesive pad electrodes (n = 56)

Organized 2 (7%) a (30%) Persistent VF 19 (66%) a (30%) Asystole 8 (27%) 11 (40%)

All first shocks with hand-held paddle electrodes (n = 28)

Organized 1 (7%) 3 (21%) Persistent VF 5 (36%) 6 (43%) Asystole a (57%) 5 (36%)

l 86 total arrests; in 2 arrests resultant rhythm of first shock not interpretable

from rhythm strip, therefore it was deleted.

VF = ventricular fibrillation.

TABLE V Rhythms Resulting from Truncated Exponential Versus Damped Sinusoidal Waveform Shocks: First Three Shocks Only

Truncated Damped Resultant Rhythm Exponential Sinusoidal

First 3 shacks (n = 175) Organized 10 (11%) 1 a (22%) Persistent VF 50 (54%) 29~ (35%)* Asystole 32 (35%) 36 (43%)

First 3 shocks with selfadhesive pad electrodes (n = 122)

Organized 9 (14%) 12 (21%) Persistent VF 38 (58%) 19 (34%)* Asystole I 9 (28%) 25 (45%)

First 3 shocks with hand-held paddle electrodes (n = 53)

Organized 1 (4%) 6 (22%) Persistent VF 12 (46%) 10 (37%) Asystole 13 (50%) 11 (41%)

* p <0.05. VF = ventricular fibrillation.

1244 THE AMERICAN JOURNAL OF CARDIOLOGY@ VOL. 78 DECEMBER 1, 1996

<0.05) when the analysis included only the first 3 shocks. The same result is true for the first 3 shocks using self-adhesive pads (58% vs 34%; p <0.05).

DISCUSSION The major finding of this study is that DS wave-

form shocks terminated VF more frequently than TE waveform shocks. There was no significant differ- ence between the 2 waveforms in the resumption of an organized rhythm, although there was a trend in favor of DS waveforms (p = 0.10). There was no significant difference in survival after shocks using the 2 waveforms.

Two previous studies provide data allowing in- direct comparison of DS versus TE waveforms. Bocka and Swor” compared fully automatic versus semi-automatic defibrillators. Since the semi-auto- matic device used was the Laerdal Heartstart 2000, which employs a DS waveform, and the fully auto- mated device was the Laerdal Heartstart 1000, which uses a TE waveform, the 2 waveforms can be com- pared. Although that study did not examine resultant rhythms, it did examine ultimate survival and found that in witnessed VF arrests those treated with fully automatic defibrillators (TE waveform) had a sur- vival rate of 35% (6 of 17) versus 0% (0 of 6) for those treated with semi-automatic defibrillators (DS waveform) (p = 0.12). However, the trend toward higher survival with TE waveform defibrillators may in part be attributable to the shorter time-to-shock after proper lead placement and activation in fully automated defibrillators (TE waveform) versus semi- automatic defibrillators (DS waveform) (16.6 vs 44.3 seconds; p <O.OOl).

Mols et al,’ ’ while examining early defibrillation in Brussels, Belgium, also indirectly compared DS versus TE waveforms.They collected data from “an- nual results in the outcome of out-of-hospital pri- mary cardiac arrests” for approximately 3 years. During the first 2 years (April 1989 to June 1991) their data came solely from Heartstart 2000 devices (DS waveform), while from July 1991 through 1992, their data came from Heartstart 3000 devices (TE waveform). Since they reported results on a calendar year basis, one could compare DS with TE by com- paring results from 1989 to 1990 versus results from 1992 (excluding 1991 since it combines the 2 wave- forms together). In doing this we find the ratio of postshock ‘ ‘restoration of systemic circulation’ ’ to persistent ventricular fibrillation to be 34% (15 of 44) for DS and 57% (27 of 47) for TE (p <0.05). Again, this result favored the TE waveform. It is unclear what confounders may be present in this study, since it involves emergency medical services much different than the United States, including phy- sician-manned ambulances in some instances. Also, time from collapse to first shock for these 2 groups is not reported for comparison.

The shock waveform duration varies with TE waveforms. In patients with high transthoracic im-

pedance, the TE waveform duration increases in or- der to deliver the required energy dose. DS wave- forms vary less with impedance. Jones and Jonest in an experimental study, suggested that waveforms with durations >20 ms have declining safety mar- gins between the currents necessary to defibrillate and those associated with postshock dysfunction. If this is applicable to humans, it might explain the better performance of the DS waveform shocks for VF termination. A study combining the variables we have examined in this paper with transthoracic im- pedance of the patients receiving shocks would be helpful in understanding the relation between im- pedance, shock duration, and VF termination when using TE waveforms.

The main limitation of this study is that it is a retrospective comparison using data initially col- lected for a different purpose-quality assurance and protocol reviews of emergency medical service personnel and performance. The study was not ran- domized, and we have not excluded the possibility that there are differences between the TE and DS groups in weight, nature of heart disease, or other characteristics that might alter the response to shocks. There is a significant difference in the ages of the 2 groups, but since both are elderly the bio- logic significance of this difference is uncertain. We need a prospective randomized clinical trial compar- ing the effectiveness of TE versus DS waveform shocks for transthoracic defibrillation.

Acknowledgment: We thank M. Bridget Zimmer- man for statistical assistance and Diane Phillips for secretarial assistance.

1. Hargarten KM, Stueven HA, Waite EM, Olson DW, Mateer JR, Aufderheide TP, Dtin JC. Prehospital experience with defibrillation of coarse ventricular fibrillation: a ten-year review. Ann Emerg Med 1990;19:157- 162 2. Kerber RE. Statement on early defibrillation from the Emergency Cardiac Care Committee, American Heart Association. Circulation 1991;83:2233. 3. Cummins RO, Thies WH. Encouraging early defibrillation: The American Heart Association and automated external defibrillators. Ann Emerg Med 1990;19:1245-1248. 4. Cummins RO, Thies WH. Automated external defibrillators and the Ad- vanced Cardiac Life Support Program: a new initiative from the American Heart Association. Am J Emerg Med 1991;9:91-93. 5. Defibrillation. In: Cummins RO, ed, Textbook of Advanced Cardiac Life Support. Dallas: American Heart Association 1994;4-l-4-20. 6. Weisfeldt ML, Kerber RE, McGoldrick RP, Moss AJ, Nichol G, Omato JP, Palmer DG, Reigel B, Smith SC. American Heart Association Report on Public Access Defibrillation Conference, Dec. R-10, 1994. Circulation 1995; 92:2740- 2147. 7. Lown B, Neuman J, Amarasingham R and Berkovits BV. Comparison of altematine current with direct electroshock across the closed chest. Am J Cardiol 1962~2231233. 8. Anderson GJ, Suelzer I. The efficiency of trapezoidal wave forms for ven- tricular defibrillation. Chest 1976;70:298-300. 9. Schuder JC, Gold JH, Stoeckle H, Granberg TA, Dettmer JC, Larwill MH. Transthoracic ventricular defibrillation in the 100 kg calf with untmncated and tmncated exponential stimuli. IEEE Trans Eiomed Eng. 1980;27:37-43. 10. Bocka JJ, Swor R. In-field comparison between fully automatic and semi- automatic defibrillators. Prehospital Disaster Med 1991;6:415-420. 11. Mols P, Beaucame E, Bruyninz I, Labmyere JP, De Myttenaere L, Naeiji N, Watteeuwg, Verset D, Flamand JP. Early defibrillation by EMTs: The Bms- sels experience. Resuscitation 1994;27:129- 136. 12. Cummins RO. The Utstein style for uniform reporting of data from out-of- hospital cardiac arrest. Ann Emerg Med 1993;22:49-52. 13. Jones JL, Jones RE. Postshock arrhythmias-a possible cause of unsuc- cessful defibrillation. Crit Care Med 1980;8:167- 171.

ARRHYTHMIAS AND CONDUCTION DISTURBANCES/WAVEFORMS FOR TRANSTHORACIC DEFlBRlllATlON 1245