- Email: [email protected]
Rescue shock outcomes during out-of-hospital cardiac arrest
Resuscitation (2007) 75, 469—475
CLINICAL PAPER
Rescue shock outcomes during out-of-hospital cardiac arrest夽 Bruce M. Lo b, Stephen M. Quinn c, David Hostler a, Clifton W. Callaway a,∗ a
Department of Emergency Medicine, University of Pittsburgh, Pittsburgh, PA, United States Department of Emergency Medicine, Eastern Virgina Medical School, Norfolk, VA 23501, United States c Department of Emergency Medicine, Sentara CarePlex Hospital, Hampton VA 23666, United States b
Received 20 October 2006 ; received in revised form 27 May 2007; accepted 4 June 2007 KEYWORDS Ventricular fibrillation; Defibrillation; Emergency medical services; Monophasic
Summary Objective: Questions remain about the optimal timing and method for treating ventricular fibrillation (VF) during out-of-hospital cardiac arrest, and a variety of treatment protocols are used. Detailed description of rescue shock outcomes during actual patient care under different protocols would allow rational comparison of treatment strategies. The purpose of this study is to describe rescue shock outcomes in a single system using a specific defibrillation protocol. Methods: Patient care records were examined for all adult (age ≥ 18 years) outof-hospital cardiac arrest cases treated by an urban paramedic system during a 52-month interval. The immediate outcomes of monophasic rescue shocks were determined from the record and were classified as asystole, VF, restoration of organized electrical activity (ROEA), or restoration of spontaneous circulation (ROSC). Results: Among 1496 cases of cardiac arrest, 654 received a median of 3 (IQR 1,5) rescue shocks. Of these cases, 408 (28%) had an initial rhythm of VF whereas VF developed later during resuscitation in the remainder. For an initial series of three escalating rescue shocks, most cases of ROSC (9%) and ROEA (12%) occurred after the first shock. The likelihood that a rescue shock would result in ROSC or ROEA increased with witnessed collapse, and rescue shock number. An initial rhythm of asystole was associated with a decreased likelihood that a rescue shock would result in ROEA. Conclusions: Witnessed collapse and an initial rhythm other than asystole were associated with an increased likelihood of rescue shock success. There is a small but real incremental gain in ROSC and ROEA from delivering three rescue shocks in rapid succession. The greater incidence of rescue shock success with later rescue shocks suggests that VF responds favorably to ongoing resuscitation. © 2007 Elsevier Ireland Ltd. All rights reserved.
夽 A Spanish translated version of the summary of this article appears as Appendix in the final online version at 10.1016/j.resuscitation.2007.06.003. ∗ Corresponding author at: Department of Emergency Medicine, 230 McKee Place, Suite 400, Pittsburgh, PA 15213, United States. Tel.: +1 412 647 9047; fax: +1 412 647 6999. E-mail address: [email protected] (C.W. Callaway).
0300-9572/$ — see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.resuscitation.2007.06.003
470 Out-of-hospital cardiac arrest remains a major public health problem for which current therapy is inadequate, resulting in survival to hospital discharge rates between 6% and 7%.1 The highest rates of survival are observed for patients with dysrhythmias that can be treated with rescue shocks (ventricular tachycardia or ventricular fibrillation). For this group, survival to hospital discharge approaches 30% in some exemplary systems.2 However, survival declines with increasing delays prior to delivery of rescue shocks.3 As a consequence of these data, there has been tremendous emphasis on designing systems to accomplish rapid defibrillation,4—6 as well as on changes to defibrillator technology.7 There are controversies about optimal timing and pace of rescue shocks. For example, while CPR prior to initial rescue shocks has been advocated in clinical2,8 and animal studies,9—11 this finding was not confirmed in another study.12 Current treatment protocols allow both immediate and delayed rescue shocks.13 In addition, it is unknown when best to time rescue shocks for patients where VF develops from another initial rhythm. Some data suggest that outcome is worse for this secondary VF.14 Finally, the most recent clinical guidelines have eliminated repeated shocks without intervening chest compressions.13 A better description of the effectiveness of rescue shocks during clinical care and the influence of clinical features on rescue shock success would facilitate efforts to assess different rescue shock protocols. There are few detailed descriptions of individual shock outcomes during actual clinical practice. Rescue shock outcomes have been reported for subsets of patients during out-ofhospital cardiac arrest,15 and for patients treated with automated external defibrillators,16 In order to describe the response to rescue shocks in the total population treated by paramedics more completely, we report the rescue shock success rate for one municipal emergency medical service (EMS) system over 4 years. Survival data for this population have been reported previously.17 As a secondary objective, this study examines the association between clinical features and rescue shock success.
Methods The City of Pittsburgh Emergency Medical System provides prehospital care to approximately 335,000 permanent residents and a daytime population of over 600,000. This municipal EMS system responds to approximately 65,000 calls annually. Calls to the
B.M. Lo et al. 911 center with the potential to be OOHCA initiate a first-responder dispatch composed of firefighters trained in basic life support skills and equipped with an automated external defibrillator (AED). Two EMS units, each manned with two full-time paramedics, are also dispatched and an emergency medicine physician provides on-line or personal medical direction for the scene. A retrospective review was performed with patients undergoing OOHCA from paramedic patient care records (PCR) from January 1998 to April 2002 (52 months). The Institutional Review Board of the University of Pittsburgh approved this retrospective review. Inclusion criteria were all adult (≥18 years), non-traumatic cardiac arrests treated by city of Pittsburgh EMS that received at least one rescue shock during the resuscitation. Variables of interest were reviewed and entered into a database separately by two investigators. These variables included age, sex, bystander, or EMS witnessed collapse, presence of bystander CPR, initial ECG rhythm, time intervals, and use of drugs. All discrepancies were discussed between the two investigators and resolved to allow for consistency in reporting. Rescue shock (RS) outcome was determined by the first post-shock assessment by the paramedic. In this EMS system, this interpretation is typically obtained during the first 20 s after the RS. The rhythm recorded in the patient care record was determined from paper summary strips that were available to the paramedics at the time that the record was prepared. However, ECG tracings were not available for over-reading at the time of research data abstraction. The recorded interpretation of the rhythm would accurately represent the interpretation that guided clinical care. Electrical success of a RS was defined as any return of organized electrical activity (ROEA), with or without pulses. Unsuccessful RS resulted in persistent VF or asystole. Mechanical success of a RS was defined as return of spontaneous circulation (ROSC), recognized by palpable pulses. Pulseless ventricular tachycardia was rare and was not distinguished from ventricular fibrillation during data analysis. Outcomes for subjects were determined from review of paramedic records and inspection of local public records of deaths. All rescue shocks were delivered by LIFEPAK 12 monitor/defibrillators (Medtronic ERS, Redmond, WA) using monophasic waveforms and hands-free defibrillation pads. These devices were replaced with biphasic defibrillators in 2003, after this study was concluded. Relationships between clinical data and RS outcome were evaluated with a chi-square test for
Rescue shock outcomes
471
parametric data or t-test for continuous data. Multivariable relationships were determined using binary logistic regression. Significance for statistical tests was set at an alpha error rate of 0.05. All variables with significant univariable relationships to RS outcome were entered into the analysis, and then variables that were not significant in the multivariable analysis were eliminated. Missing values were not imputed, but cases were excluded on a case-to-case basis. Because shock outcomes might be correlated within individual subjects, final confidence intervals for the multivariable model were calculated considering each case as a cluster. Fit of the logistic model was assessed by the Hosmer—Lemeshow statistic.
Results During the 52-month study period, there were 1496 cases of adult OOHCA (Table 1). OOHCA was slightly more common in men. The most common rhythm upon EMS arrival was asystole. A total of 45% of patients were pronounced dead at the scene after failed resuscitative efforts. A total of 32% of patients regained pulses at some point during outof-hospital care, but only 25% of patients had pulses at the time of arrival at the hospital A date of death could be determined for 95.6 % of subjects, and the survival curves for these subjects have been presented separately.17
Table 1
Rescue shocks were administered in 654 cases. All subjects with VF received rescue shocks, and 108/389 (28%) of subjects with an initial rhythm of PEA and 126/638 (20%) of subjects with an initial rhythm of asystole developed VF during resuscitation and also received rescue shocks. A firstresponder AED was deployed and delivered one or more shocks before paramedic arrival in 101 cases. In these cases the initial rhythm for paramedics was more often VF (64%, 65/101), than asystole (17%, 17/101) or PEA (19%, 19/101). However, there were no univariable relationships between receipt of shocks by first-responder AED and outcomes for subsequent rescue shocks delivered by paramedics. Subjects who received shocks required a median of three rescue shocks (IQR 1, 5) (Figure 1). For nonEMS witnessed cardiac arrest cases for whom VF was the initial rhythm, rescue shock were administered a median of two (IQR 1, 4) minutes after arrival at the scene. Consistent with local protocols, the energy level for the first three rescue shocks escalated (200, 300 then 360 J) in the majority of cases, although there were exceptions (Table 2). Outcomes for each rescue shock were available for the majority of cases (Table 2). Because few subjects received more than six rescue shocks, outcomes for rescue shocks 7—19 were collapsed together. In general, for shocks beyond the sixth, outcomes were more often persistent VF (Tables 2 and 3). These cases of shock-resistant
Demographic features of 1496 adult, out-of-hospital cardiac arrest cases Number/valid data (%)
Male sex Witnessed by bystander Witnessed by EMS Bystander CPR
849/1492 646/1222 120/1342 439/1097
Initial rhythm VF VT PEA Asystole
408/1444 (28) 394/1444 (27) 642/1444 (45)
Transported to hospital Pulses ever Pulse at hospital arrival Alive after 1 year
816/1495 (55) 475/1487 (32) 367/1496 (25) 81/1496 (5.4) Mean ± S.D.
Response intervals (excludes EMS witnessed cases) Dispatch to scene 5.8 ± 3.1 Scene to first shock 3.0 ± 3.1 Scene to hospital 27.4 ± 9.0 Epinephrine dose
4.2 ± 2.4 mg
(57) (53) (9) (40) (excludes EMS witnessed cases)
Median (intraquartile range) 5 (IQR 4, 7.25) 2 (IQR 1, 4) (335 VF cases only) (536 transported cases only) (1360 cases where epinephrine used)
472
B.M. Lo et al.
Figure 1 Total number of rescue shocks delivered per person. The number of subjects who received a given number of rescue shocks is depicted for a total of 659 subjects who received rescue shocks.
Figure 2 Outcomes for initial three (‘‘stacked’’) rescue shocks delivered to 408 subjects with a presenting rhythm of VF. If a subject remained in VF after each rescue shock, another rescue shock was delivered immediately.
VF are a unique subset of the population for which only few observations per shock are available and are not considered in the multivariable analysis. Later rescue shocks were less likely to result in asystole and more likely to result in ROEA or ROSC (2 = 113.4, d.f. = 18, p < 0.0001). The same pattern of outcomes was observed when only those cases where VF was the initial rhythm were considered (Table 3) (2 = 72.4; d.f. = 15; p < 0.0001). For cases in which the first recorded rhythm was VF, up to three rescue shocks usually occurred in rapid succession (‘‘stacked shocks’’). Characteristics and outcomes of these stacked rescue
shocks were not different from those of the whole population of shocks (Table 4, Figure 2). Return of pulses was achieved for 46/378 (12.2%) subjects after three successive shocks, with the first shock contributing 33/46 (72%), the second shock contributing 7/46 (15%) and the third shock contributing 6/46 (13%) to the overall success. Return of electrical activity (including those cases with pulses) was achieved for 107/378 (28%) after three successive shocks, with the first shock contributing 71/107 (66%), the second shock contributing 19/107 (18%) and the third shock contributing 17/107 (16%) to the overall success. Several variables were associated with rescues shock outcome for all of the first six rescue shocks
Table 2
Details and outcomes for rescue shocks in 654 subjects receiving rescue shocks Valid data
Shock 1 Shock 2 Shock 3 Shock 4 Shock 5 Shock 6 Shocks 7—19
Shock 1 Shock 2 Shock 3 Shock 4 Shock 5 Shock 6 Shocks 7—19
Energy level (J) ≤100
200
300
360
647 450 361 244 169 110 240
8 (1%) 2 (1%) 2 (1%) 0 (0%) 0 (0%) 1 (1%) 0 (0%)
540(84%) 41 (9%) 9 (3%) 6 (3%) 4 (2%) 2 (2%) 12 (5%)
35 (5%) 330 (73%) 30 (8%) 9 (4%) 4 (2%) 4 (4%) 6 (3%)
64 (10%) 76 (17%) 320 (89%) 229 (94%) 161 (95%) 103 (94%) 222 (93%)
Valid data
Outcome ROSC
ROEA
VF
Asystole
592/647 399/450 299/361 189/244 140/169 88/110 240
54 (9%) 27 (7%) 25 (8%) 17 (9%) 16 (11%) 18 (20%) 20 (11%)
69 (12%) 44 (11%) 54 (18%) 42 (22%) 20 (14%) 16 (18%) 19 (11%)
281 (48%) 236 (59%) 151 (51%) 102 (54%) 79 (56%) 47 (53%) 130 (72%)
188 (32%) 92 (23%) 69 (23%) 28 (15%) 25 (18%) 7 (8%) 11 (6%)
Rescue shock outcomes Table 3
473
Outcomes stratified by initial rhythm recorded by EMS Valid data
Outcome ROSC
ROEA
VF
Asystole
38 (10%) 27 (10%) 41 (20%) 37 (27%) 13 (13%) 12 (17%) 18 (11%)
184 (49%) 152 (57%) 99 (48%) 66 (49%) 59 (57%) 36 (52%) 114 (71%)
123 (33%) 69 (26%) 48 (23%) 21 (15%) 17 (16%) 6 (9%) 9 (6%)
Outcomes for 108 cases where PEA was initial rhythm recorded by EMS Shock 1 97/108 13 (13%) 24 (25%) Shock 2 64 3 (5%) 14 (22%) Shock 3 48 5 (10%) 9 (19%) Shock 4 25 2 (8%) 3 (12%) Shock 5 18 0 (0%) 4 (22%) Shock 6 11 2 (18%) 2 (18%) Shock 7—19 13 0 (0%) 1 (8%)
46 (47%) 38 (59%) 23 (48%) 15 (60%) 12 (67%) 7 (64%) 12 (92%)
13 (13%) 8 (13%) 11 (23%) 5 (20%) 2 (11%) 0 (0%) 0 (0%)
Outcomes for 128 cases where Asystole was initial rhythm recorded by EMS Shock 1 112/128 8 (7%) 7 (6%) Shock 2 64 5 (8%) 3 (5%) Shock 3 45 3 (7%) 4 (9%) Shock 4 27 3 (11%) 3 (11%) Shock 5 16 1 (6%) 2 (13%) Shock 6 8 1 (13%) 1 (13%) Shocks 7—19 14 1 (14%) 0 (0%)
48 (43%) 43 (67%) 29 (64%) 19 (70%) 8 (50%) 4 (50%) 11 (79%)
49 (44%) 13 (20%) 9 (20%) 2 (7%) 5 (31%) 2 (25%) 2 (14%)
Outcomes for 408 cases where VF was initial rhythm recorded by EMS Shock 1 378/408 33 (9%) Shock 2 267 19 (7%) Shock 3 205 17 (8%) Shock 4 136 12 (9%) Shock 5 104 15 (14%) Shock 6 69 15 (22%) Shocks 7—19 160 19 (12%)
delivered to subjects with valid data (n = 1707 shocks). Bystander witnessed collapse, cardiac arrest witnessed by EMS, and rescue shock number were positively associated with the likelihood of a rescue shock resulting in organized electrical activity or pulses (Table 5). For subjects who developed VF from an initial rhythm of PEA, rescue shocks were more likely to restore organized electrical activity. Conversely, for subjects who developed VF from an initial rhythm of asystole, rescue shocks were less likely to result in organized electrical activity The likelihood that a rescue shock would restore pulses did not differ for either of these groups of subjects (Table 3). There was no apparent relationship between the outcomes of the first rescue shock and the interval
Table 4 of VF
Shock 1 Shock 2 Shock 3
from dispatch to arrival at the scene. In those cases where collapse was not witnessed by EMS and where VF was the first recorded rhythm, there also was no relationship between first rescue shock outcome and the interval from scene arrival to first-shock delivery. However, the distribution of these intervals was not wide (Table 1). Logistic regression was used to assess the multivariable associations between clinical variables and rescue shock outcome (Table 6). Witnessed collapse and EMS witnessed collapse were both independently associated with rescue shock success. Rescue shock number had an independent positive association, and an initial rhythm of asystole had an independent negative association with restoration of organized electrical activity.
Energy levels for three initial stacked rescue shocks delivered to 408 subjects with a presenting rhythm Valid data
Energy level (J) ≤100
200
300
360
405/408 182/184 118/118
4 (1%) 0 (0%) 0 (0%)
350 (87%) 11 (6%) 3 (3%)
16 (4%) 142 (78%) 5 (4%)
35 (9%) 29 (15%) 110 (93%)
474
B.M. Lo et al. ferential treatment of primary and secondary VF. These data will provide a useful benchmark for comparison with EMS systems using biphasic defibrillation and more recent guidelines.13 An unexpected observation was that electrical success was associated with increasing RS number during resuscitation (Tables 2, 3, 5 and 6). Previous animal and human data suggest that CPR prior to defibrillation is more likely to result in ROEA or ROSC while immediate RS for prolonged VF often results in asystole and subsequent poor outcome.2,8,10—12 One possible explanation of the present data is that the subgroup of subjects who received higher numbers of rescue shocks benefited from ongoing CPR. The medical protocols in use by paramedics during this time dictated that patients in VF receive up to three rescue shocks, after which epinephrine (adrenaline) was given intravenously or via a tracheal tube. Subsequent drugs were given according to the advanced cardiac life support guidelines. Additional rescue shocks were to be given whenever VF was recognized. Thus, increasing RS number also is associated with increased likelihood of receiving epinephrine and other drugs. Alternatively, those subjects who remain in VF after multiple rescue shocks may represent a more salvageable subgroup. For example, subjects whose cardiac rhythm was converted permanently into asystole by early rescue shocks did not go on to receive later rescue shocks. It is unlikely that the greater success of later rescue shocks is related to increasing RS energy. First, energy levels typically escalated during the first three rescue shocks (Table 2). Second, the increased rates of ROEA and ROSC were most evident for the fourth through sixth rescue shocks for which the majority were 360 J. Previous analyses from this same dataset indicated that the number of rescue shocks was not associated with duration of survival when adjusted for other clinical variables.17 There are several limitation of these data. One is the use of monophasic rescue shocks. Several studies suggest that biphasic rescue shocks are more effective for terminating VF.7,18—20 The overall first-
Table 5 Univariable relationships between clinical variables and favorable RS outcomes for the first six rescue shocks in 654 subjects Variable
ROEA
ROSC
Total (n = 1707) Male sex (n = 1143) Witnessed (n = 1029) Bystander CPR (n = 490) EMS witnessed (n = 121)
402 259 281 120 38
157 (9.2%) 104 (9.1) 104 (10.1%)* 46 (9.4%) 21 (17.4%)*
Initial rhythm VF/VT (n = 1160) PEA (n = 261) Asystole (n = 271)
280 (24.1%) 80 (30.7%)* 41 (15.1%)*
111 (9.6%) 25 (9.6%) 21 (7.7%)
Shock number RS 1 (n = 592) RS 2 (n = 399) RS 3 (n = 299) RS 4 (n = 189) RS 5 (n = 140) RS 6 (n = 88)
123 71 79 59 36 34
54 (9.1%) 27 (6.8%) 25 (8.4%) 17 (9.0%) 16 (11.4%) 18 (20.5%)
(23.6%) (22.7%) (27.3%)* (24.5%) (21.4%)*
(20.8%) (17.8%) (26.4%) (31.2%) (25.7%) (38.6%)
Electrical success was defined as return of organized electrical activity (ROEA), and mechanical success was defined as ROEA accompanied by return of pulses (ROSC) (*p < 0.05 for univariable test).
Discussion This study describes the outcomes for rescue shocks administered by paramedics during attempted resuscitation of out-of-hospital cardiac arrest. The population and characteristics of the EMS systems are similar to many urban settings. The significant findings are that rescue shocks are more likely to restore organized electrical activity and pulses when collapse was witnessed or when collapse occurred in front of the EMS providers. Furthermore, rescue shocks are less likely to result in organized electrical activity or pulses when VF develops in a subject initially found in asystole. These observations suggest that the initial ECG rhythm might be a surrogate for the responsiveness of the heart to rescue shocks when VF does develop. However, these data do not suggest dif-
Table 6
Multivariable relationships between clinical variables and RS outcomes presented as odds ratio [95% CI] ROEA [1.66, [1.44, [1.05, [0.42,
ROSC
EMS witnessed collapse Witnessed collapse Shock number Asystole
2.87 2.12 1.13 0.66
4.97] 3.12] 1.23] 1.03]
Hosmer—Lemeshow
9.67, p = 0.208
Model goodness-of-fit was assessed using the Hosmer—Lemeshow statistic.
4.41 2.25 1.11 0.99
[2.07, [1.29, [0.98, [0.54,
9.35] 3.91] 1.25] 1.81]
3.04, p = 0.881
Rescue shock outcomes shock success for ROEA in this study (21%, Table 2) was similar to the 12%7 and 15% ROSC19 reported previously for monophasic shocks. Another limitation of this study is that the ECG tracings were interpreted only by the treating paramedics, and were not available to the investigators for overreading. However, the readings that are recorded are in fact the readings that were used for clinical decision-making, perhaps making them more relevant. These data provide some quantitative confirmation for abandoning the practice of administering three rescue shocks for the initial treatment of VF.13 Certainly, the incidence of ROSC is low (4—6%) for each subsequent rescue shock (Table 2). The fact that our data do not reveal a relationship between the outcomes of the first rescue shock and the interval from dispatch to arrival at the scene differs from prior studies.3 One possible explanation for our failure to detect an influence of response time is that our data are derived from a municipal EMS system without a large variation in response intervals. Dispatch to scene arrival intervals were 7 min or less for 75% and 10 min or less for 90% of cases. Thus, the effects of other patient characteristics may overshadow any effect of dispatch to arrival interval. In summary, these data provide an estimate of the probability of rescue shock success during paramedic practice in an urban setting. These data supplement previous reports about rescue shock outcomes for subjects with witnessed VF.15 Future data will allow comparison of outcomes in EMS systems using biphasic defibrillators or different energy escalation protocols.
Conflict of interest Dr. Callaway is co-inventor of methods to analyze the VF waveform and to predict rescue shock success that have been licensed by the University of Pittsburgh to Medtronic ERS, Inc., a manufacturer of defibrillators. Other authors have no conflicts.
References 1. Nichol G, Stiell IG, Laupacis A, Pham B, De Maio VJ, Wells GA. A cumulative meta-analysis of the effectiveness of defibrillator-capable emergency medical services for victims of out-of-hospital cardiac arrest. Ann Emerg Med 1999;34:517—25. 2. Cobb LA, Fahrenbruch CE, Walsh TR, et al. Influence of cardiopulmonary resuscitation prior to defibrillation in patients with out-of-hospital ventricular fibrillation. JAMA 1999;281:1182—8.
475 3. De Maio VJ, Stiell IG, Wells GA, Spaite DW, Ontario Prehospital Advanced Life Support Study Group. Optimal defibrillation response intervals for maximum out-ofhospital cardiac arrest survival rates. Ann Emerg Med 2003;42:242—50. 4. Valenzuela TD, Roe DJ, Nichol G, Clark LL, Spaite DW, Hardman RG. Outcomes of rapid defibrillation by security officers after cardiac arrests in casinos. N Engl J Med 2000;343:1206—9. 5. White RD, Asplin BR, Bugliosi TF, Hankins DG. High discharge survival rate after out-of-hospital ventricular defibrillation with rapid defibrillation by police and paramedics. Ann Emerg Med 1996;28:480—5. 6. Caffrey SL, Willoughby PJ, Pepe PE, Becker LB. Public use of automated external defibrillators. N Engl J Med 2002; 347:1242—7. 7. Morrison LJ, Dorian P, Long J, et al. Lerman BB; Steering Committee, Central Validation Committee Safety and Efficacy Committee Out-of-hospital cardiac arrest rectilinear biphasic to monophasic damped sine defibrillation waveforms with advanced life support intervention trial (ORBIT). Resuscitation 2005;66:149—57. 8. Wik L, Hansen TB, Fylling F, et al. Delaying defibrillation to give basic cardiopulmonary resuscitation to patients with out-of-hospital ventricular fibrillation: a randomized trial. Jama 2003;289:1389—95. 9. Yakaitis RW, Ewy GA, Otto CW, Taren DL, Moon TE. Influence of time and therapy on ventricular defibrillation in dogs. Crit Care Med 1980;8:157—63. 10. Niemann JT, Cairns CB, Sharma J, Lewis RJ. Treatment of prolonged ventricular fibrillation Immediate countershock versus high-dose epinephrine and CPR preceding countershock. Circulation 1992;85:281—7. 11. Menegazzi JJ, Davis EA, Yealy DM, et al. An experimental algorithm versus standard advanced cardiac life support in a swine model of out-of-hospital cardiac arrest. Ann Emerg Med 1993;22:235—9. 12. Jacobs IG, Finn JC, Oxer HF, Jelinek GA. CPR before defibrillation in out-of-hospital cardiac arrest: a randomized trial. Emerg Med Australas 2005;17:39—45. 13. American Heart, Association. Part 5. Electrical therapies: automated external defibrillators, defibrillation, cardioversion, and pacing. Circulation 2005;112(Suppl I):IV-35—46. 14. Kerber RE, Jensen SR, Gascho JA, Grayzel J, Hoyt R, Kennedy J. Determinants of defibrillation: prospective analysis of 183 patients. Am J Cardiol 1983;52:739—45. 15. Hargarten KM, Stueven HA, Waite EM, et al. Prehospital experience with defibrillation of coarse ventricular fibrillation: a ten-year review. Ann Emerg Med 1990;19:157—62. 16. White RD, Hankins DJ, Bugliosi TG, Seven Years. Experience with Early Defibrillation by Police and Paramedics in an Emergency Medical Services System. Resuscitation 1998;39:145—51. 17. Wang HE, Min A, Hostler D, Chang CC, Callaway CW. Differential effects of out-of-hospital interventions on short- and long-term survival after cardiopulmonary arrest. Resuscitation 2005;67:69—74. 18. Martens PR, Russell JK, Wolcke B, et al. Optimal Response to Cardiac Arrest study: defibrillation waveform effects. Resuscitation 2001;49:233—43. 19. Stothert JC, Hatcher TS, Gupton CL, Love JE, Brewer JE. Rectilinear biphasic waveform defibrillation of out-of-hospital cardiac arrest. Prehosp Emerg Care 2004;8:388—92. 20. Van Alem AP, Chapman FW, Lank P, Hart AA, Koster RW. A prospective, randomised and blinded comparison of firstshock success of monophasic and biphasic waveforms in out-of-hospital cardiac arrest. Resuscitation 2003;58:17—24.
CLINICAL PAPER
Rescue shock outcomes during out-of-hospital cardiac arrest夽 Bruce M. Lo b, Stephen M. Quinn c, David Hostler a, Clifton W. Callaway a,∗ a
Department of Emergency Medicine, University of Pittsburgh, Pittsburgh, PA, United States Department of Emergency Medicine, Eastern Virgina Medical School, Norfolk, VA 23501, United States c Department of Emergency Medicine, Sentara CarePlex Hospital, Hampton VA 23666, United States b
Received 20 October 2006 ; received in revised form 27 May 2007; accepted 4 June 2007 KEYWORDS Ventricular fibrillation; Defibrillation; Emergency medical services; Monophasic
Summary Objective: Questions remain about the optimal timing and method for treating ventricular fibrillation (VF) during out-of-hospital cardiac arrest, and a variety of treatment protocols are used. Detailed description of rescue shock outcomes during actual patient care under different protocols would allow rational comparison of treatment strategies. The purpose of this study is to describe rescue shock outcomes in a single system using a specific defibrillation protocol. Methods: Patient care records were examined for all adult (age ≥ 18 years) outof-hospital cardiac arrest cases treated by an urban paramedic system during a 52-month interval. The immediate outcomes of monophasic rescue shocks were determined from the record and were classified as asystole, VF, restoration of organized electrical activity (ROEA), or restoration of spontaneous circulation (ROSC). Results: Among 1496 cases of cardiac arrest, 654 received a median of 3 (IQR 1,5) rescue shocks. Of these cases, 408 (28%) had an initial rhythm of VF whereas VF developed later during resuscitation in the remainder. For an initial series of three escalating rescue shocks, most cases of ROSC (9%) and ROEA (12%) occurred after the first shock. The likelihood that a rescue shock would result in ROSC or ROEA increased with witnessed collapse, and rescue shock number. An initial rhythm of asystole was associated with a decreased likelihood that a rescue shock would result in ROEA. Conclusions: Witnessed collapse and an initial rhythm other than asystole were associated with an increased likelihood of rescue shock success. There is a small but real incremental gain in ROSC and ROEA from delivering three rescue shocks in rapid succession. The greater incidence of rescue shock success with later rescue shocks suggests that VF responds favorably to ongoing resuscitation. © 2007 Elsevier Ireland Ltd. All rights reserved.
夽 A Spanish translated version of the summary of this article appears as Appendix in the final online version at 10.1016/j.resuscitation.2007.06.003. ∗ Corresponding author at: Department of Emergency Medicine, 230 McKee Place, Suite 400, Pittsburgh, PA 15213, United States. Tel.: +1 412 647 9047; fax: +1 412 647 6999. E-mail address: [email protected] (C.W. Callaway).
0300-9572/$ — see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.resuscitation.2007.06.003
470 Out-of-hospital cardiac arrest remains a major public health problem for which current therapy is inadequate, resulting in survival to hospital discharge rates between 6% and 7%.1 The highest rates of survival are observed for patients with dysrhythmias that can be treated with rescue shocks (ventricular tachycardia or ventricular fibrillation). For this group, survival to hospital discharge approaches 30% in some exemplary systems.2 However, survival declines with increasing delays prior to delivery of rescue shocks.3 As a consequence of these data, there has been tremendous emphasis on designing systems to accomplish rapid defibrillation,4—6 as well as on changes to defibrillator technology.7 There are controversies about optimal timing and pace of rescue shocks. For example, while CPR prior to initial rescue shocks has been advocated in clinical2,8 and animal studies,9—11 this finding was not confirmed in another study.12 Current treatment protocols allow both immediate and delayed rescue shocks.13 In addition, it is unknown when best to time rescue shocks for patients where VF develops from another initial rhythm. Some data suggest that outcome is worse for this secondary VF.14 Finally, the most recent clinical guidelines have eliminated repeated shocks without intervening chest compressions.13 A better description of the effectiveness of rescue shocks during clinical care and the influence of clinical features on rescue shock success would facilitate efforts to assess different rescue shock protocols. There are few detailed descriptions of individual shock outcomes during actual clinical practice. Rescue shock outcomes have been reported for subsets of patients during out-ofhospital cardiac arrest,15 and for patients treated with automated external defibrillators,16 In order to describe the response to rescue shocks in the total population treated by paramedics more completely, we report the rescue shock success rate for one municipal emergency medical service (EMS) system over 4 years. Survival data for this population have been reported previously.17 As a secondary objective, this study examines the association between clinical features and rescue shock success.
Methods The City of Pittsburgh Emergency Medical System provides prehospital care to approximately 335,000 permanent residents and a daytime population of over 600,000. This municipal EMS system responds to approximately 65,000 calls annually. Calls to the
B.M. Lo et al. 911 center with the potential to be OOHCA initiate a first-responder dispatch composed of firefighters trained in basic life support skills and equipped with an automated external defibrillator (AED). Two EMS units, each manned with two full-time paramedics, are also dispatched and an emergency medicine physician provides on-line or personal medical direction for the scene. A retrospective review was performed with patients undergoing OOHCA from paramedic patient care records (PCR) from January 1998 to April 2002 (52 months). The Institutional Review Board of the University of Pittsburgh approved this retrospective review. Inclusion criteria were all adult (≥18 years), non-traumatic cardiac arrests treated by city of Pittsburgh EMS that received at least one rescue shock during the resuscitation. Variables of interest were reviewed and entered into a database separately by two investigators. These variables included age, sex, bystander, or EMS witnessed collapse, presence of bystander CPR, initial ECG rhythm, time intervals, and use of drugs. All discrepancies were discussed between the two investigators and resolved to allow for consistency in reporting. Rescue shock (RS) outcome was determined by the first post-shock assessment by the paramedic. In this EMS system, this interpretation is typically obtained during the first 20 s after the RS. The rhythm recorded in the patient care record was determined from paper summary strips that were available to the paramedics at the time that the record was prepared. However, ECG tracings were not available for over-reading at the time of research data abstraction. The recorded interpretation of the rhythm would accurately represent the interpretation that guided clinical care. Electrical success of a RS was defined as any return of organized electrical activity (ROEA), with or without pulses. Unsuccessful RS resulted in persistent VF or asystole. Mechanical success of a RS was defined as return of spontaneous circulation (ROSC), recognized by palpable pulses. Pulseless ventricular tachycardia was rare and was not distinguished from ventricular fibrillation during data analysis. Outcomes for subjects were determined from review of paramedic records and inspection of local public records of deaths. All rescue shocks were delivered by LIFEPAK 12 monitor/defibrillators (Medtronic ERS, Redmond, WA) using monophasic waveforms and hands-free defibrillation pads. These devices were replaced with biphasic defibrillators in 2003, after this study was concluded. Relationships between clinical data and RS outcome were evaluated with a chi-square test for
Rescue shock outcomes
471
parametric data or t-test for continuous data. Multivariable relationships were determined using binary logistic regression. Significance for statistical tests was set at an alpha error rate of 0.05. All variables with significant univariable relationships to RS outcome were entered into the analysis, and then variables that were not significant in the multivariable analysis were eliminated. Missing values were not imputed, but cases were excluded on a case-to-case basis. Because shock outcomes might be correlated within individual subjects, final confidence intervals for the multivariable model were calculated considering each case as a cluster. Fit of the logistic model was assessed by the Hosmer—Lemeshow statistic.
Results During the 52-month study period, there were 1496 cases of adult OOHCA (Table 1). OOHCA was slightly more common in men. The most common rhythm upon EMS arrival was asystole. A total of 45% of patients were pronounced dead at the scene after failed resuscitative efforts. A total of 32% of patients regained pulses at some point during outof-hospital care, but only 25% of patients had pulses at the time of arrival at the hospital A date of death could be determined for 95.6 % of subjects, and the survival curves for these subjects have been presented separately.17
Table 1
Rescue shocks were administered in 654 cases. All subjects with VF received rescue shocks, and 108/389 (28%) of subjects with an initial rhythm of PEA and 126/638 (20%) of subjects with an initial rhythm of asystole developed VF during resuscitation and also received rescue shocks. A firstresponder AED was deployed and delivered one or more shocks before paramedic arrival in 101 cases. In these cases the initial rhythm for paramedics was more often VF (64%, 65/101), than asystole (17%, 17/101) or PEA (19%, 19/101). However, there were no univariable relationships between receipt of shocks by first-responder AED and outcomes for subsequent rescue shocks delivered by paramedics. Subjects who received shocks required a median of three rescue shocks (IQR 1, 5) (Figure 1). For nonEMS witnessed cardiac arrest cases for whom VF was the initial rhythm, rescue shock were administered a median of two (IQR 1, 4) minutes after arrival at the scene. Consistent with local protocols, the energy level for the first three rescue shocks escalated (200, 300 then 360 J) in the majority of cases, although there were exceptions (Table 2). Outcomes for each rescue shock were available for the majority of cases (Table 2). Because few subjects received more than six rescue shocks, outcomes for rescue shocks 7—19 were collapsed together. In general, for shocks beyond the sixth, outcomes were more often persistent VF (Tables 2 and 3). These cases of shock-resistant
Demographic features of 1496 adult, out-of-hospital cardiac arrest cases Number/valid data (%)
Male sex Witnessed by bystander Witnessed by EMS Bystander CPR
849/1492 646/1222 120/1342 439/1097
Initial rhythm VF VT PEA Asystole
408/1444 (28) 394/1444 (27) 642/1444 (45)
Transported to hospital Pulses ever Pulse at hospital arrival Alive after 1 year
816/1495 (55) 475/1487 (32) 367/1496 (25) 81/1496 (5.4) Mean ± S.D.
Response intervals (excludes EMS witnessed cases) Dispatch to scene 5.8 ± 3.1 Scene to first shock 3.0 ± 3.1 Scene to hospital 27.4 ± 9.0 Epinephrine dose
4.2 ± 2.4 mg
(57) (53) (9) (40) (excludes EMS witnessed cases)
Median (intraquartile range) 5 (IQR 4, 7.25) 2 (IQR 1, 4) (335 VF cases only) (536 transported cases only) (1360 cases where epinephrine used)
472
B.M. Lo et al.
Figure 1 Total number of rescue shocks delivered per person. The number of subjects who received a given number of rescue shocks is depicted for a total of 659 subjects who received rescue shocks.
Figure 2 Outcomes for initial three (‘‘stacked’’) rescue shocks delivered to 408 subjects with a presenting rhythm of VF. If a subject remained in VF after each rescue shock, another rescue shock was delivered immediately.
VF are a unique subset of the population for which only few observations per shock are available and are not considered in the multivariable analysis. Later rescue shocks were less likely to result in asystole and more likely to result in ROEA or ROSC (2 = 113.4, d.f. = 18, p < 0.0001). The same pattern of outcomes was observed when only those cases where VF was the initial rhythm were considered (Table 3) (2 = 72.4; d.f. = 15; p < 0.0001). For cases in which the first recorded rhythm was VF, up to three rescue shocks usually occurred in rapid succession (‘‘stacked shocks’’). Characteristics and outcomes of these stacked rescue
shocks were not different from those of the whole population of shocks (Table 4, Figure 2). Return of pulses was achieved for 46/378 (12.2%) subjects after three successive shocks, with the first shock contributing 33/46 (72%), the second shock contributing 7/46 (15%) and the third shock contributing 6/46 (13%) to the overall success. Return of electrical activity (including those cases with pulses) was achieved for 107/378 (28%) after three successive shocks, with the first shock contributing 71/107 (66%), the second shock contributing 19/107 (18%) and the third shock contributing 17/107 (16%) to the overall success. Several variables were associated with rescues shock outcome for all of the first six rescue shocks
Table 2
Details and outcomes for rescue shocks in 654 subjects receiving rescue shocks Valid data
Shock 1 Shock 2 Shock 3 Shock 4 Shock 5 Shock 6 Shocks 7—19
Shock 1 Shock 2 Shock 3 Shock 4 Shock 5 Shock 6 Shocks 7—19
Energy level (J) ≤100
200
300
360
647 450 361 244 169 110 240
8 (1%) 2 (1%) 2 (1%) 0 (0%) 0 (0%) 1 (1%) 0 (0%)
540(84%) 41 (9%) 9 (3%) 6 (3%) 4 (2%) 2 (2%) 12 (5%)
35 (5%) 330 (73%) 30 (8%) 9 (4%) 4 (2%) 4 (4%) 6 (3%)
64 (10%) 76 (17%) 320 (89%) 229 (94%) 161 (95%) 103 (94%) 222 (93%)
Valid data
Outcome ROSC
ROEA
VF
Asystole
592/647 399/450 299/361 189/244 140/169 88/110 240
54 (9%) 27 (7%) 25 (8%) 17 (9%) 16 (11%) 18 (20%) 20 (11%)
69 (12%) 44 (11%) 54 (18%) 42 (22%) 20 (14%) 16 (18%) 19 (11%)
281 (48%) 236 (59%) 151 (51%) 102 (54%) 79 (56%) 47 (53%) 130 (72%)
188 (32%) 92 (23%) 69 (23%) 28 (15%) 25 (18%) 7 (8%) 11 (6%)
Rescue shock outcomes Table 3
473
Outcomes stratified by initial rhythm recorded by EMS Valid data
Outcome ROSC
ROEA
VF
Asystole
38 (10%) 27 (10%) 41 (20%) 37 (27%) 13 (13%) 12 (17%) 18 (11%)
184 (49%) 152 (57%) 99 (48%) 66 (49%) 59 (57%) 36 (52%) 114 (71%)
123 (33%) 69 (26%) 48 (23%) 21 (15%) 17 (16%) 6 (9%) 9 (6%)
Outcomes for 108 cases where PEA was initial rhythm recorded by EMS Shock 1 97/108 13 (13%) 24 (25%) Shock 2 64 3 (5%) 14 (22%) Shock 3 48 5 (10%) 9 (19%) Shock 4 25 2 (8%) 3 (12%) Shock 5 18 0 (0%) 4 (22%) Shock 6 11 2 (18%) 2 (18%) Shock 7—19 13 0 (0%) 1 (8%)
46 (47%) 38 (59%) 23 (48%) 15 (60%) 12 (67%) 7 (64%) 12 (92%)
13 (13%) 8 (13%) 11 (23%) 5 (20%) 2 (11%) 0 (0%) 0 (0%)
Outcomes for 128 cases where Asystole was initial rhythm recorded by EMS Shock 1 112/128 8 (7%) 7 (6%) Shock 2 64 5 (8%) 3 (5%) Shock 3 45 3 (7%) 4 (9%) Shock 4 27 3 (11%) 3 (11%) Shock 5 16 1 (6%) 2 (13%) Shock 6 8 1 (13%) 1 (13%) Shocks 7—19 14 1 (14%) 0 (0%)
48 (43%) 43 (67%) 29 (64%) 19 (70%) 8 (50%) 4 (50%) 11 (79%)
49 (44%) 13 (20%) 9 (20%) 2 (7%) 5 (31%) 2 (25%) 2 (14%)
Outcomes for 408 cases where VF was initial rhythm recorded by EMS Shock 1 378/408 33 (9%) Shock 2 267 19 (7%) Shock 3 205 17 (8%) Shock 4 136 12 (9%) Shock 5 104 15 (14%) Shock 6 69 15 (22%) Shocks 7—19 160 19 (12%)
delivered to subjects with valid data (n = 1707 shocks). Bystander witnessed collapse, cardiac arrest witnessed by EMS, and rescue shock number were positively associated with the likelihood of a rescue shock resulting in organized electrical activity or pulses (Table 5). For subjects who developed VF from an initial rhythm of PEA, rescue shocks were more likely to restore organized electrical activity. Conversely, for subjects who developed VF from an initial rhythm of asystole, rescue shocks were less likely to result in organized electrical activity The likelihood that a rescue shock would restore pulses did not differ for either of these groups of subjects (Table 3). There was no apparent relationship between the outcomes of the first rescue shock and the interval
Table 4 of VF
Shock 1 Shock 2 Shock 3
from dispatch to arrival at the scene. In those cases where collapse was not witnessed by EMS and where VF was the first recorded rhythm, there also was no relationship between first rescue shock outcome and the interval from scene arrival to first-shock delivery. However, the distribution of these intervals was not wide (Table 1). Logistic regression was used to assess the multivariable associations between clinical variables and rescue shock outcome (Table 6). Witnessed collapse and EMS witnessed collapse were both independently associated with rescue shock success. Rescue shock number had an independent positive association, and an initial rhythm of asystole had an independent negative association with restoration of organized electrical activity.
Energy levels for three initial stacked rescue shocks delivered to 408 subjects with a presenting rhythm Valid data
Energy level (J) ≤100
200
300
360
405/408 182/184 118/118
4 (1%) 0 (0%) 0 (0%)
350 (87%) 11 (6%) 3 (3%)
16 (4%) 142 (78%) 5 (4%)
35 (9%) 29 (15%) 110 (93%)
474
B.M. Lo et al. ferential treatment of primary and secondary VF. These data will provide a useful benchmark for comparison with EMS systems using biphasic defibrillation and more recent guidelines.13 An unexpected observation was that electrical success was associated with increasing RS number during resuscitation (Tables 2, 3, 5 and 6). Previous animal and human data suggest that CPR prior to defibrillation is more likely to result in ROEA or ROSC while immediate RS for prolonged VF often results in asystole and subsequent poor outcome.2,8,10—12 One possible explanation of the present data is that the subgroup of subjects who received higher numbers of rescue shocks benefited from ongoing CPR. The medical protocols in use by paramedics during this time dictated that patients in VF receive up to three rescue shocks, after which epinephrine (adrenaline) was given intravenously or via a tracheal tube. Subsequent drugs were given according to the advanced cardiac life support guidelines. Additional rescue shocks were to be given whenever VF was recognized. Thus, increasing RS number also is associated with increased likelihood of receiving epinephrine and other drugs. Alternatively, those subjects who remain in VF after multiple rescue shocks may represent a more salvageable subgroup. For example, subjects whose cardiac rhythm was converted permanently into asystole by early rescue shocks did not go on to receive later rescue shocks. It is unlikely that the greater success of later rescue shocks is related to increasing RS energy. First, energy levels typically escalated during the first three rescue shocks (Table 2). Second, the increased rates of ROEA and ROSC were most evident for the fourth through sixth rescue shocks for which the majority were 360 J. Previous analyses from this same dataset indicated that the number of rescue shocks was not associated with duration of survival when adjusted for other clinical variables.17 There are several limitation of these data. One is the use of monophasic rescue shocks. Several studies suggest that biphasic rescue shocks are more effective for terminating VF.7,18—20 The overall first-
Table 5 Univariable relationships between clinical variables and favorable RS outcomes for the first six rescue shocks in 654 subjects Variable
ROEA
ROSC
Total (n = 1707) Male sex (n = 1143) Witnessed (n = 1029) Bystander CPR (n = 490) EMS witnessed (n = 121)
402 259 281 120 38
157 (9.2%) 104 (9.1) 104 (10.1%)* 46 (9.4%) 21 (17.4%)*
Initial rhythm VF/VT (n = 1160) PEA (n = 261) Asystole (n = 271)
280 (24.1%) 80 (30.7%)* 41 (15.1%)*
111 (9.6%) 25 (9.6%) 21 (7.7%)
Shock number RS 1 (n = 592) RS 2 (n = 399) RS 3 (n = 299) RS 4 (n = 189) RS 5 (n = 140) RS 6 (n = 88)
123 71 79 59 36 34
54 (9.1%) 27 (6.8%) 25 (8.4%) 17 (9.0%) 16 (11.4%) 18 (20.5%)
(23.6%) (22.7%) (27.3%)* (24.5%) (21.4%)*
(20.8%) (17.8%) (26.4%) (31.2%) (25.7%) (38.6%)
Electrical success was defined as return of organized electrical activity (ROEA), and mechanical success was defined as ROEA accompanied by return of pulses (ROSC) (*p < 0.05 for univariable test).
Discussion This study describes the outcomes for rescue shocks administered by paramedics during attempted resuscitation of out-of-hospital cardiac arrest. The population and characteristics of the EMS systems are similar to many urban settings. The significant findings are that rescue shocks are more likely to restore organized electrical activity and pulses when collapse was witnessed or when collapse occurred in front of the EMS providers. Furthermore, rescue shocks are less likely to result in organized electrical activity or pulses when VF develops in a subject initially found in asystole. These observations suggest that the initial ECG rhythm might be a surrogate for the responsiveness of the heart to rescue shocks when VF does develop. However, these data do not suggest dif-
Table 6
Multivariable relationships between clinical variables and RS outcomes presented as odds ratio [95% CI] ROEA [1.66, [1.44, [1.05, [0.42,
ROSC
EMS witnessed collapse Witnessed collapse Shock number Asystole
2.87 2.12 1.13 0.66
4.97] 3.12] 1.23] 1.03]
Hosmer—Lemeshow
9.67, p = 0.208
Model goodness-of-fit was assessed using the Hosmer—Lemeshow statistic.
4.41 2.25 1.11 0.99
[2.07, [1.29, [0.98, [0.54,
9.35] 3.91] 1.25] 1.81]
3.04, p = 0.881
Rescue shock outcomes shock success for ROEA in this study (21%, Table 2) was similar to the 12%7 and 15% ROSC19 reported previously for monophasic shocks. Another limitation of this study is that the ECG tracings were interpreted only by the treating paramedics, and were not available to the investigators for overreading. However, the readings that are recorded are in fact the readings that were used for clinical decision-making, perhaps making them more relevant. These data provide some quantitative confirmation for abandoning the practice of administering three rescue shocks for the initial treatment of VF.13 Certainly, the incidence of ROSC is low (4—6%) for each subsequent rescue shock (Table 2). The fact that our data do not reveal a relationship between the outcomes of the first rescue shock and the interval from dispatch to arrival at the scene differs from prior studies.3 One possible explanation for our failure to detect an influence of response time is that our data are derived from a municipal EMS system without a large variation in response intervals. Dispatch to scene arrival intervals were 7 min or less for 75% and 10 min or less for 90% of cases. Thus, the effects of other patient characteristics may overshadow any effect of dispatch to arrival interval. In summary, these data provide an estimate of the probability of rescue shock success during paramedic practice in an urban setting. These data supplement previous reports about rescue shock outcomes for subjects with witnessed VF.15 Future data will allow comparison of outcomes in EMS systems using biphasic defibrillators or different energy escalation protocols.
Conflict of interest Dr. Callaway is co-inventor of methods to analyze the VF waveform and to predict rescue shock success that have been licensed by the University of Pittsburgh to Medtronic ERS, Inc., a manufacturer of defibrillators. Other authors have no conflicts.
References 1. Nichol G, Stiell IG, Laupacis A, Pham B, De Maio VJ, Wells GA. A cumulative meta-analysis of the effectiveness of defibrillator-capable emergency medical services for victims of out-of-hospital cardiac arrest. Ann Emerg Med 1999;34:517—25. 2. Cobb LA, Fahrenbruch CE, Walsh TR, et al. Influence of cardiopulmonary resuscitation prior to defibrillation in patients with out-of-hospital ventricular fibrillation. JAMA 1999;281:1182—8.
475 3. De Maio VJ, Stiell IG, Wells GA, Spaite DW, Ontario Prehospital Advanced Life Support Study Group. Optimal defibrillation response intervals for maximum out-ofhospital cardiac arrest survival rates. Ann Emerg Med 2003;42:242—50. 4. Valenzuela TD, Roe DJ, Nichol G, Clark LL, Spaite DW, Hardman RG. Outcomes of rapid defibrillation by security officers after cardiac arrests in casinos. N Engl J Med 2000;343:1206—9. 5. White RD, Asplin BR, Bugliosi TF, Hankins DG. High discharge survival rate after out-of-hospital ventricular defibrillation with rapid defibrillation by police and paramedics. Ann Emerg Med 1996;28:480—5. 6. Caffrey SL, Willoughby PJ, Pepe PE, Becker LB. Public use of automated external defibrillators. N Engl J Med 2002; 347:1242—7. 7. Morrison LJ, Dorian P, Long J, et al. Lerman BB; Steering Committee, Central Validation Committee Safety and Efficacy Committee Out-of-hospital cardiac arrest rectilinear biphasic to monophasic damped sine defibrillation waveforms with advanced life support intervention trial (ORBIT). Resuscitation 2005;66:149—57. 8. Wik L, Hansen TB, Fylling F, et al. Delaying defibrillation to give basic cardiopulmonary resuscitation to patients with out-of-hospital ventricular fibrillation: a randomized trial. Jama 2003;289:1389—95. 9. Yakaitis RW, Ewy GA, Otto CW, Taren DL, Moon TE. Influence of time and therapy on ventricular defibrillation in dogs. Crit Care Med 1980;8:157—63. 10. Niemann JT, Cairns CB, Sharma J, Lewis RJ. Treatment of prolonged ventricular fibrillation Immediate countershock versus high-dose epinephrine and CPR preceding countershock. Circulation 1992;85:281—7. 11. Menegazzi JJ, Davis EA, Yealy DM, et al. An experimental algorithm versus standard advanced cardiac life support in a swine model of out-of-hospital cardiac arrest. Ann Emerg Med 1993;22:235—9. 12. Jacobs IG, Finn JC, Oxer HF, Jelinek GA. CPR before defibrillation in out-of-hospital cardiac arrest: a randomized trial. Emerg Med Australas 2005;17:39—45. 13. American Heart, Association. Part 5. Electrical therapies: automated external defibrillators, defibrillation, cardioversion, and pacing. Circulation 2005;112(Suppl I):IV-35—46. 14. Kerber RE, Jensen SR, Gascho JA, Grayzel J, Hoyt R, Kennedy J. Determinants of defibrillation: prospective analysis of 183 patients. Am J Cardiol 1983;52:739—45. 15. Hargarten KM, Stueven HA, Waite EM, et al. Prehospital experience with defibrillation of coarse ventricular fibrillation: a ten-year review. Ann Emerg Med 1990;19:157—62. 16. White RD, Hankins DJ, Bugliosi TG, Seven Years. Experience with Early Defibrillation by Police and Paramedics in an Emergency Medical Services System. Resuscitation 1998;39:145—51. 17. Wang HE, Min A, Hostler D, Chang CC, Callaway CW. Differential effects of out-of-hospital interventions on short- and long-term survival after cardiopulmonary arrest. Resuscitation 2005;67:69—74. 18. Martens PR, Russell JK, Wolcke B, et al. Optimal Response to Cardiac Arrest study: defibrillation waveform effects. Resuscitation 2001;49:233—43. 19. Stothert JC, Hatcher TS, Gupton CL, Love JE, Brewer JE. Rectilinear biphasic waveform defibrillation of out-of-hospital cardiac arrest. Prehosp Emerg Care 2004;8:388—92. 20. Van Alem AP, Chapman FW, Lank P, Hart AA, Koster RW. A prospective, randomised and blinded comparison of firstshock success of monophasic and biphasic waveforms in out-of-hospital cardiac arrest. Resuscitation 2003;58:17—24.