- Email: [email protected]
Who survives from out-of-hospital pulseless electrical activity?
Resuscitation (2008) 76, 207—213
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/resuscitation
CLINICAL PAPER
Who survives from out-of-hospital pulseless electrical activity?夽 a¨ att¨ a, James Boyd Taneli V¨ ayrynen ∗, Markku Kuisma, Teuvo M¨ Helsinki Emergency Medical Services (EMS), Department of Anaesthesiology and Intensive Care Medicine, Helsinki University Central Hospital, P.O. Box 112, FIN-00099 Helsingin Kaupunki, Finland Received 9 March 2007; received in revised form 9 July 2007; accepted 19 July 2007
KEYWORDS Pulseless electrical activity (PEA); Out-of-hospital CPR; Outcome; Bystander-CPR
Summary Aim of the study: To study factors associated with short-term and long-term survival after out-of-hospital cardiac arrest presenting with pulseless electrical activity (PEA). Materials and methods: This was a retrospective observational study. All out-of-hospital cardiac arrests in Helsinki, Finland during 1 January1997—31 December 2005 were prospectively registered in the cardiac arrest database. Of 3291 arrests 984 had PEA as the first registered rhythm. Results: The use of adrenaline was the only factor associated with long-term survival, by increasing mortality. Increasing delay to the return of spontaneous circulation (ROSC) was the only factor associated with survival among patients that survived to admission, also by increasing mortality. There were no survivors that were discharged in overall performance category (OPC) 1—2 after a bystander-witnessed arrest (excluding cases of hypothermia and/or near-drowning) with first responding unit (FRU)-delay over 14 min, or that were resuscitated for more than 20 min. There were no survivors who were discharged in OPC 1—2 after an unwitnessed arrest with the duration of advanced life support (ALS) exceeding 5.5 min. Conclusions: The use of adrenaline during resuscitation was the only significant factor which was found to decrease the long-term survival. Among admitted patients, short delay to ROSC was the only factor associated with increased survival. Bystander-CPR and delays to the arrival of the FRU or to the initiation of ALS were not associated with survival. Therefore, it seems difficult to increase survival rates of PEA by improving the chain of survival. More effort should be put to education of the public to call for an ambulance before the cardiac arrest occurs. © 2007 Elsevier Ireland Ltd. All rights reserved.
Introduction 夽 A Spanish translated version of the summary of this article appears as Appendix in the final online version at 10.1016/j.resuscitation.2007.07.023. ∗ Corresponding author. Fax: +358 931030189. E-mail address: Taneli.vayrynen@hus.fi (T. V¨ ayrynen).
Pulseless electrical activity (PEA) is defined as (organised) cardiac electrical activity in the absence of any palpable pulses. Out-of-hospital PEA has poor prognosis, with 2.4—6.5% of patients surviving to hospital discharge.1—3
0300-9572/$ — see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.resuscitation.2007.07.023
208 It has been shown that the presence of cardiac contractions (verified by cardiac ultrasound examination during the arrest) is associated with survival.4—6 Due to the subjective nature of the diagnostic criteria for PEA, the circulatory status of patients judged to have PEA may range from total cardiac standstill to profound shock (pseudo-PEA) without a palpable pulse due to severe hypotension. Palpation of carotid pulses may be difficult due to anatomical reasons (e.g. obesity), circumstances of the arrest (e.g. in the street in cold weather) or inexperience of the person palpating the pulse. Recognition of pulselessness by palpation of the carotid pulse performed by the first responders is inaccurate.7 In asystole and ventricular fibrillation (VF) the diagnosis can be made from the ECG-tracing, and once the diagnosis is established it is assured that the patient is in true cardiac arrest. Poor prognosis after out-of-hospital PEA has been reported after blunt trauma,8 unwitnessed arrests (especially in the geriatric population)2,3,9 and if bystander-CPR is not provided.2 Our hypothesis is that due to the diversity of haemodynamic states judged to be PEA, there are few factors (compared to VF and asystole) that are associated with survival. Therefore it may be even more difficult to predict the prognosis in individual cases. We suspect that the marginal perfusion provided by the cardiac contractions may cause all delays to be less important in PEA than in the other initial rhythms of cardiac arrest. The purpose of this study was to describe factors associated with short- and long-term survival among 984 patients judged as having PEA as the initial rhythm of out-of-hospital cardiac arrest. Short-term survival was defined as survival to hospital admission and long-term survival was defined as survival to 30 days after the arrest.
Materials and methods Study setting Helsinki is a middle-sized urban city with 559,000 inhabitants. It is served by a three-tiered Emergency Medical Services (EMS) consisting of seven to eight BLS-units depending on the time of the day, five advanced life support (ALS)-units (of which three are ambulances, one is a medical supervisor unit and one is a fire engine) and a physician staffed mobile intensive care unit (MICU). Units are distributed to eight regional rescue stations. According to dispatching protocols, in suspected cardiac arrest at least two units (of which one is an ALS-unit) are dispatched. If a fire engine can reach the scene faster than an ambulance it is used as a first responding unit (FRU). There are seven fire engines that provide the same level of care as BLS ambulances. If cardiac arrest is not suspected and the patients’ risk is considered to be low or intermediate, an ALS-unit is dispatched only if the scene is within the unit’s own area. If a BLS-unit finds a patient in unsuspected cardiac arrest, an ALS-unit is requested from the dispatching centre and resuscitation is initiated. The time intervals are calculated from the beginning of the emergency call (time zero) to the arrival of BLS and ALS units at the patients’ side. In EMS witnessed arrests the actual time of collapse is used as time zero.
T. V¨ ayrynen et al. In Helsinki EMS all BLS-units are capable of defibrillation, intubation of an adult cardiac arrest victim and basic cardiopulmonary resuscitation. ALS units are equipped additionally with i.v. medications, e.g. adrenaline and amiodarone. Guidelines used in cardiopulmonary resuscitation of PEA follow international recommendations, with adrenaline [epinephrine] (1 mg at 3—5 min intervals) as the vasopressor. No other drugs such as atropine are given routinely during CPR. Drugs are given either into the external jugular or the antecubital vein. Intubation precedes i.v. access in priority. Patients are intubated routinely as soon as possible. If intubation by an EMT or a paramedic cannot be done with two attempts bag-valve mask ventilation is carried out until the arrival of MICU. According to Helsinki EMS do-not-attempt resuscitation (DNAR) guidelines resuscitation is not attempted in PEA (excluding cases of hypothermia and/or near-drowning) if the patient has a living will or terminal illness, traumatic arrest or when resuscitation (including bystander-CPR) has not been initiated within 15 min. Resuscitation is terminated according to DNAR guidelines after 20 min of resuscitation in PEA. The decision to withhold resuscitation or to terminate ongoing resuscitation is always made by a physician. DNAR guidelines were introduced by the end of 1997.
Data collection The Institutional Review Board of Helsinki University Hospital approved the study. The study plan was retrospective but the data were collected prospectively by EMS-physicians into the cardiac arrest database. The on-call physician completed a data collection form based on Utstein recommendations immediately after arrest. Information regarding the cause of arrest, survival to discharge and overall performance category (OPC) on discharge were achieved later from hospital files by a physician responsible for the maintenance of the database. Patients were assigned to OPC 1—4 according to hospital records at the time of hospital discharge. OPC 1 includes patients with full functional recovery, OPC 2 includes patients with moderate disability (but still able to perform independently in daily activities), OPC 3 includes patients with severe disability (in need of assistance in daily activities) and patients in OPC 4 are in a coma. The definition for discharged alive was that after hospital care was not needed any more. Patients were discharged from hospital either to their homes or in some cases to nursing homes. No arrested patient with secondary sign of death (rigor mortis, dependent lividity) were included for analysis.
Statistical methods For normally distributed continuous variables mean and standard deviation (S.D.) are presented and independent samples T-test is used as a test of significance. Median and interquartile range (IQR) are presented for nonparametrically distributed data and Mann—Whitney U-test is used as a test of significance. All categorical variables were analysed using chi-square test or Fischer’s exact test. All significance tests were two-tailed with p less than 0.05 considered significant. Logistic regression was used to calculate
Who survives from out-of-hospital PEA?
209
adjusted odd ratios (OR) for short- and long-term mortality using variables that were univariately associated with shortand long-term survival with significance of p < 0.5. In the univariate analysis continuous variables were dichotomized according to mean or median values. In the multivariate analysis continuous variables were used and FRU and ALS delays were transformed to natural logarithmic scale due to their skewed distribution. Hosmer—Lemeshow-test was used to assess the goodness of fit of these models. Reasonable fit was assumed since the results were not significant. A sample size calculation was not performed since this was a retrospective study in which all patients in the Helsinki EMS cardiac arrest database were included.
Results During the study period (1 January1997—31 December 2005) there were 3291 out-of-hospital cardiac arrests. The initial rhythm was PEA in 984 (29.9%) cases. These 984 arrests form the study population. The cardiac arrest was bystander-witnessed in 55.2% (n = 543), witnessed by EMS-personnel in 30.9% (n = 304) and unwitnessed in 13.8% (n = 136) of cases. Data regarding the witness status of one patient was missing. Resuscitation was attempted in 80.2% (n = 789) of arrests. The majority (68.2%, n = 133) of decisions to withhold resuscitation were based on Helsinki EMS DNAR guidelines. Survival to hospital admission rate was 25.7% (n = 253). In-hospital mortality rate was 75.1%. Survival to discharge rate was 6.4% (n = 63), and 5.8% (n = 57) were alive 30 days after the arrest. Seventy-one percent (n = 45) of discharged patients were in OPC 1—2. Long-term survival rate was 5.7% in bystanderwitnessed arrests, 3.7% in unwitnessed arrests and 6.9% in EMS-witnessed arrests. Characteristics of long-term survivors and those who died within 30 days of the arrest are presented in Table 1. Cases in which resuscitation was not attempted are excluded. Longterm survivors had significantly shorter ROSC delays and had received adrenaline less frequently and in smaller total
Table 1
amounts. There were non-significant trends (with p < 0.2) towards survivors being younger, having received bystanderCPR less frequently, having shorter FRU delays and also being defibrillated less often. Total incidence of bystander-CPR in PEA (excluding EMS-witnessed arrests) was 29.5% in arrests in which resuscitation was attempted and 32.6% if all arrests were included. Results of the univariate analysis are presented in Table 2. Arrests in which resuscitation was not attempted were excluded. Cardiac cause and defibrillation (in case of conversion of PEA into a shockable rhythm) were associated with decreased short-term survival rate and use of adrenaline was associated with decreased long-term survival rate. All factors that were associated with survival with p < 0.5 were included in multivariate analysis. Results of the multivariate analysis are shown in Table 3. The use of adrenaline in resuscitation was the only significant factor found and was associated with decreased long-term survival. Results of the univariate analysis of factors associated with 30-day survival among admitted patients are presented in Table 4. ROSC-delay under 19 min and the use of adrenaline were the only factors univariately associated with survival. Shorter delays to ROSC increased the survival rate and use of adrenaline decreased the survival rate. All factors that were associated with survival (with p < 0.5) were included in multivariate analysis. Increasing ROSC delay was the only significant factor, which will decrease the chance of survival. Adjusted OR for survival for increasing ROSC delay (per minute) was 0.92, 95% CI 0.85—0.995 (p = 0.037). Survival to different end-points according to confirmed diagnosis is presented in Table 5. Cardiac cause was the most common diagnosis among long-term survivors. Of the 25 long-term survivors with cardiac aetiology, 6 had suffered a myocardial infarction. In-hospital mortality rate after arrests due to a cardiac cause was 72.1%. There were no survivors to discharge in OPC 1—2 after a bystander-witnessed arrest (excluding cases of hypothermia and/or near-drowning) with an FRU-delay over 14 min, or who were resuscitated for more than 20 min. There was no survivors to discharge from hospital in OPC 1—2 after an
Characteristics of long-term survivors and non-survivors
Male (gender) Age (years) Arrest at home Bystander-witnessed arrest B-CPRa (data missing: 0, 9) FRU-delay (data missing: 1, 3) ALS-delay (data missing: 1, 8) ROSC-delay (data missing: 3, 11) Adrenaline used Amount of adrenaline (mg) Defibrillated Cardiac cause (data missing: 0, 98)
n (%) Mean (S.D.) n (%) n (%) n (%) Median (IQR) Median (IQR) Median (IQR) n (%) Median (IQR) n (%) n (%)
Long-term survivors (n = 57)
Non-survivors (n = 732)
p
40 (70.2) 63 (17) 31 (54.4) 31 (54.4) 8 (16.7) 5 (0—8.5) 9 (4.5—13.5) 16 (7—21) 39 (68.4) 2 (0—3) 9 (15.8) 25 (43.9)
448 (61.2) 67 (17) 458 (62.6) 412 (56.3) 174 (28.1) 7 (0—10) 10 (5—14) 20 (14—26) 664 (90.7) 4 (2—5) 189 (25.8) 336 (47.9)
0.204 0.097 0.257 0.784 0.094 0.166 0.221 <0.0001 <0.0001 0.006 0.112 0.213
Arrests in which resuscitation was not attempted are excluded. B-CPR, bystander-cardiopulmonary resuscitation; FRU, first responding unit; ALS, advanced life support; ROSC, return of spontaneous circulation. a Arrests witnessed by the ambulance personnel are excluded. Number of cases with missing data are given for both groups, respectively.
210
Table 2
Results of the univariate analysis Short-term survival rate (%)
OR
95% CI
p
Long-term survival rate (%)
OR
95% CI
p
Male gender
Yes No
30.5 34.2
0.845
0.622—1.148
0.307
8.2 5.6
2.492
0.830—2.682
0.204
Age under 66 years
Yes No
34.8 30.1
1.244
0.917—1.687
0.161
7.7 6.9
1.134
0.657—1.959
0.674
Arrest at home
Yes No
29.9 35.3
0.779
0.574—1.058
0.116
6.3 8.7
0.713
0.415—1.227
0.257
Cardiac cause
Yes No
28.5 44.8
0.491
0.358—0.673
<0.0001
6.9 9.7
0.693
0.401—1.196
0.213
Bystander-witnessed arrest
Yes No
32.5 31.2
1.061
0.785—1.436
0.758
7.0 7.5
0.926
0.539—1.591
0.784
Bystander-CPR
Yes No
29.1 33.8
0.804
0.555—1.165
0.266
4.4 8.2
0.511
0.235—1.115
0.094
FRU-delay under 7 min
Yes No
31.2 32.2
0.952
0.705—1.287
0.76
8.0 6.3
1.292
0.748—2.232
0.406
ALS-delay under 9.5 min
Yes No
31.3 32.0
0.969
0.716—1.310
0.878
7.6 6.8
1.141
0.662—1.967
0.678
Adrenaline used
Yes No
28.2 62.8
0.232
0.146—0.371
<0.0001
5.5 20.9
0.222
0.120—0.409
<0.0001
Defibrillated
Yes No
22.7 35.0
0.547
0.376—0.792
0.001
4.5 8.1
0.539
0.259—1.119
0.112
Continuous variables were dichotomised according to their mean or median values. CPR, cardiopulmonary resuscitation; FRU, first responding unit; ALS, advanced life support; OR, odds ratio and CI, confidence interval. OR value above 1 indicates increased chance of survival.
T. V¨ ayrynen et al.
Who survives from out-of-hospital PEA? Table 3
211
Results of the multivariate analysis Short-term survival
Cardiac cause Adrenaline used Defibrillated
Long-term survival
Adjusted OR
95% CI
p
Adjusted OR
95% CI
p
0.55 0.23 0.65
0.38—0.81 0.13—0.41 0.42—0.99
0.002 <0.0001 0.046
0.16
0.07—0.35
<0.0001
Short-term mortality is mortality before hospital admission, long-term mortality is mortality within 30 days of the cardiac arrest. Only significant (p < 0.05) associations are shown. OR, odds ratio; CI, confidence interval. OR value above 1 indicates increased chance of mortality.
Table 4
Univariate analysis of long-term survival among admitted patients Survival rate (%)
OR
95% CI
p
Gender Male
Yes No
26.7 16.5
1.84
0.976—3.467
0.067
Age under 64 years
Yes No
21.6 23.2
0.911
0.498—1.668
0.878
Arrest at home
Yes No
21.1 24.5
0.882
0.454—1.489
0.544
Cardiac cause
Yes No
24.0 21.6
1.147
0.632—2.082
0.650
Bystander-witnessed arrest
Yes No
21.5 23.9
0.876
0.484—1.585
0.761
Bystander-CPR
Yes No
15.1 24.2
0.556
0.242—1.277
0.186
FRU-delay under 7 min
Yes No
25.6 19.4
1.433
0.788—2.605
0.288
ALS-delay under 10 min
Yes No
25.4 19.8
1.376
0.757—2.503
0.362
ROSC-delay under 19 min
Yes No
30.1 15.4
2.367
1.269—4.415
0.008
Adrenaline used
Yes No
19.7 32.7
0.504
0.260—0.979
0.046
Defibrillated
Yes No
20.0 23.1
0.833
0.375—1.852
0.844
Continuous variables were dichotomised according to their mean or median values. CPR, cardiopulmonary resuscitation; FRU, first responding unit; ALS, advanced life support; ROSC, return of spontaneous circulation; OR, odds ratio and CI, confidence interval. OR value above 1 indicates increased chance of survival.
unwitnessed arrest with duration of resuscitation exceeding 14 min or duration of ALS exceeding 5.5 min.
Discussion The use of adrenaline was the only factor associated with long-term mortality among arrests in which resuscitation was attempted. Survival to 30 days was reduced by over six times if adrenaline was used during the resuscitation. When the subgroup of patients who survived to hospital admission was analysed, a short delay to ROSC was the only factor associated with survival. Both the use of adrenaline and the ROSC delay are factors that are associated with the duration
of resuscitation, i.e. the patients’ responsiveness to resuscitative measures. It was possible that majority of survivors were in fact in pseudo-PEA and therefore responded rapidly to resuscitative measures. Since bystander-CPR and delays to the arrival of the FRU or to the initiation of ALS were not associated with survival, it seems difficult to increase the survival rate from PEA by further improving the chain of survival. Since EMS-witnessed arrests have the highest survival rate, the public should be educated to recognise the warning signs and to call for help before the cardiac arrest occurs. It is probable that the wide spectrum of haemodynamic states that can be judged to be PEA is a major confounding factor in the prognostication of PEA. Some of the patients are in true cardiac arrest while others are merely in profound shock.
212 Table 5 (%)
T. V¨ ayrynen et al. Confirmed diagnosis of patients that survived to hospital admission, hospital discharge or 30 days after the arrest, n Admitted to hospital (n = 253)
Cardiac cause Pulmonary embolism Trauma Intoxication Bleeding (without trauma) Suffocation Drowning Seizure Pneumonia Hanging Asthma Cerebrovascular accident/stroke Sepsis Malignancy Carbon monoxide poisoning Pancreatitis Other obvious cause Data missing
104 (41.1) 18 (7.1) 14 (5.5) 12 (4.7) 10 (4.0) 13 (5.1) 6 (2.4) 5 (2.0) 11 (4.3) 2 (0.8) 4 (1.6) 32 (12.6) 2 (0.8) 1 (0.4) 1 (0.4) 1 (0.4) 16 (6.3) 1 (0.4)
Discharged from hospital (n = 63) 29 7 6 4 3 3 3 2 2 1 1 0 0 0 0 0 2 0
(46) (11) (10) (6) (5) (5) (5) (3) (3) (2) (2) (0) (0) (0) (0) (0) (3) (0)
Long-term survival (n = 57) 25 7 6 4 3 3 2 2 2 1 1 0 0 0 0 0 1 0
(44) (12) (11) (7) (5) (5) (4) (4) (4) (2) (2) (0) (0) (0) (0) (0) (2) (0)
Hypothermia was the cause of the arrest of the only long-term survivor in the ‘‘other obvious cause’’-category.
Cardiac ultrasound examination during the arrest could aid in the differentiation of these states and the evaluation of the prognosis of individual patients. It is also difficult to recognise subgroups that do not benefit from resuscitation. There was no survivor discharged from hospital in OPC 1—2 after bystander-witnessed arrest if the FRU-delay was over 15 min, and in the previous study by Engdahl et al. there were no survivors at all if the FRU-delay exceeded 15 min. Therefore it seems that resuscitation is not beneficial and should probably be withheld if the patient is not reached within 15 min after bystander-witnessed arrest. However, it should be remembered that only a small minority of patients in an urban environment experience such a long delay. There were no long-term survivors who were discharged in OPC 1—2 after bystander-witnessed arrest who were resuscitated for more than 20 min. Prognosis was poorer in unwitnessed than witnessed arrests, but still 3.7% survived after an unwitnessed arrest. As all survivors from unwitnessed PEA responded to ALS very rapidly, prolonged efforts may be futile in this subgroup. Contradicting the previous findings by Engdahl et al., bystander-CPR was not beneficial in this study. In our study the incidence of bystander-CPR was high compared to the previous study by Engdahl et al. (29.5% versus 6.6—10.2%). The incidence of bystander-CPR was lower among survivors in our study, but not reaching statistical significance. In the previous study the incidence of bystander-CPR was significantly higher among survivors than in non-survivors (28% versus 8%, p = 0.008). It is possible that in the previous study bystanders initiated resuscitation in cases that had a better prognosis per se, since the incidence was significantly lower. It is also possible that bystander-CPR is associated with shorter delay to call-for-help, but since the delay to the arrival of the FRU was not associated with survival it should
not alter survival rates significantly. Quality of bystanderCPR may also vary. It is highly speculative whether the marginal perfusion provided by the cardiac contractions in pseudo-PEA diminishes the benefit from bystander-CPR, and the survivors will ultimately survive regardless of bystanderCPR. There were also differences in survival, compared to previous results by Engdahl et al. In our study survival rate to hospital discharge was higher (6.4% versus 2.4%) and the proportion of survivors in OPC 1—2 was also higher (over 70% versus almost 50%). More arrests were witnessed in the present study, which may explain the difference in the survival rates at least partly. In particular the number of EMS-witnessed arrests was high in the present study. The incidence of bystander-CPR was also higher in the present study, but in our multivariate analysis it was not associated with survival. Patients in our study were younger, which may explain at least partly both the higher survival rate and lower incidence of cardiac aetiology. It is also possible that the execution of resuscitation is different, and that intra-resuscitation delays to different interventions (i.e. intubation, intravenous access, and administration of medications) may differ. There are some limitations to this study. Due to the nature of the diagnostic criteria for PEA, there may be differences in the incidence of PEA and pseudo-PEA between EMS systems. In some cases, PEA with a low frequency of electrical complexes may be judged to be asystole. Survival rates and factors associated with survival in different EMSsystems may vary more in PEA than in asystole or VF. There are also major differences between different EMS systems, in terms of structure of the EMS (BLS, ALS or a multi-tiered system) and time delays to the initiation of the treatment. Therefore these results may not be applicable to all EMS systems.
Who survives from out-of-hospital PEA?
Conclusions The only factor associated with long-term survival was the use of adrenaline during resuscitation. The chance of death was over six times higher if adrenaline was used. Among patients who were admitted, a short delay to ROSC was the only factor associated with improved survival. BystanderCPR and delays to the arrival of the FRU or to the initiation of ALS were not associated with survival. Therefore it seems difficult to increase survival rates from PEA by improving the chain of survival, besides by educating the public to call for help before the cardiac arrest occurs. Prognosis is poor if the FRU-delay is over 15 min in bystander-witnessed arrests or if the duration of ALS exceeds 5.5 min in EMS-witnessed arrests, excluding cases of hypothermia and near-drowning.
Conflicts of interest There authors hereby state that there is no conflict of interest.
References 1. Ontario Prehospital Advanced Life Support Study Group. Advanced cardiac life support in out-of-hospital cardiac arrest. N Engl J Med 2004;351:647—56.
213 2. Engdahl J, B˚ ang A, Lindqvist J, Herlitz J. Factors affecting short- and long-term prognosis among 1069 patients with out-of-hospital cardiac arrest and pulseless electrical activity. Resuscitation 2001;51:17—25. 3. Pepe PE, Levine RL, Fromm Jr RE, Curka PA, Clark PS, Zachrahiah BS. Cardiac arrest presenting with rhythms other than ventricular fibrillation: contribution of resuscitative efforts toward total survivorship. Crit Care Med 1993;21:143—50. 4. Salen P, O’Connor R, Sierzenski P, et al. Can cardiac sonography be used independently and in combination to predict resuscitation outcomes? Acad Emerg Med 2001;8:610—5. 5. Blaivas M, Fox JC. Outcome in cardiac arrest patients found to have cardiac standstill on the bedside emergency department echocardiogram. Acad Emerg Med 2001;8:616—21. 6. Salen P, Melniker L, Chooljian C, et al. Does the presence or absence of sonographically identified cardiac activity predict resuscitation outcomes of cardiac arrest patients? Am J Emerg Med 2005;23:459—62. 7. Dick WF, Balthasar E, Wisser G, Schneider T. The carotid pulse check revisited: what if there is no pulse? Crit Care Med 2000;28(S):183—5. 8. Martin SK, Shatney CH, Sherck JP, Ho C-C, Homan SJ, Neff J. Blunt trauma patients with prehospital pulseless electrical activity (PEA): poor ending assured. J Trauma 2002;53:876—81. 9. Stratton SJ, Niemann JT. Outcome from out-of-hospital cardiac arrest caused by nonventricular arrhythmias: contribution of successful resuscitation to overall survivorship supports the current practice of initiating out-of-hospital ACLS. Ann Emerg Med 1998;32:448—53.
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/resuscitation
CLINICAL PAPER
Who survives from out-of-hospital pulseless electrical activity?夽 a¨ att¨ a, James Boyd Taneli V¨ ayrynen ∗, Markku Kuisma, Teuvo M¨ Helsinki Emergency Medical Services (EMS), Department of Anaesthesiology and Intensive Care Medicine, Helsinki University Central Hospital, P.O. Box 112, FIN-00099 Helsingin Kaupunki, Finland Received 9 March 2007; received in revised form 9 July 2007; accepted 19 July 2007
KEYWORDS Pulseless electrical activity (PEA); Out-of-hospital CPR; Outcome; Bystander-CPR
Summary Aim of the study: To study factors associated with short-term and long-term survival after out-of-hospital cardiac arrest presenting with pulseless electrical activity (PEA). Materials and methods: This was a retrospective observational study. All out-of-hospital cardiac arrests in Helsinki, Finland during 1 January1997—31 December 2005 were prospectively registered in the cardiac arrest database. Of 3291 arrests 984 had PEA as the first registered rhythm. Results: The use of adrenaline was the only factor associated with long-term survival, by increasing mortality. Increasing delay to the return of spontaneous circulation (ROSC) was the only factor associated with survival among patients that survived to admission, also by increasing mortality. There were no survivors that were discharged in overall performance category (OPC) 1—2 after a bystander-witnessed arrest (excluding cases of hypothermia and/or near-drowning) with first responding unit (FRU)-delay over 14 min, or that were resuscitated for more than 20 min. There were no survivors who were discharged in OPC 1—2 after an unwitnessed arrest with the duration of advanced life support (ALS) exceeding 5.5 min. Conclusions: The use of adrenaline during resuscitation was the only significant factor which was found to decrease the long-term survival. Among admitted patients, short delay to ROSC was the only factor associated with increased survival. Bystander-CPR and delays to the arrival of the FRU or to the initiation of ALS were not associated with survival. Therefore, it seems difficult to increase survival rates of PEA by improving the chain of survival. More effort should be put to education of the public to call for an ambulance before the cardiac arrest occurs. © 2007 Elsevier Ireland Ltd. All rights reserved.
Introduction 夽 A Spanish translated version of the summary of this article appears as Appendix in the final online version at 10.1016/j.resuscitation.2007.07.023. ∗ Corresponding author. Fax: +358 931030189. E-mail address: Taneli.vayrynen@hus.fi (T. V¨ ayrynen).
Pulseless electrical activity (PEA) is defined as (organised) cardiac electrical activity in the absence of any palpable pulses. Out-of-hospital PEA has poor prognosis, with 2.4—6.5% of patients surviving to hospital discharge.1—3
0300-9572/$ — see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.resuscitation.2007.07.023
208 It has been shown that the presence of cardiac contractions (verified by cardiac ultrasound examination during the arrest) is associated with survival.4—6 Due to the subjective nature of the diagnostic criteria for PEA, the circulatory status of patients judged to have PEA may range from total cardiac standstill to profound shock (pseudo-PEA) without a palpable pulse due to severe hypotension. Palpation of carotid pulses may be difficult due to anatomical reasons (e.g. obesity), circumstances of the arrest (e.g. in the street in cold weather) or inexperience of the person palpating the pulse. Recognition of pulselessness by palpation of the carotid pulse performed by the first responders is inaccurate.7 In asystole and ventricular fibrillation (VF) the diagnosis can be made from the ECG-tracing, and once the diagnosis is established it is assured that the patient is in true cardiac arrest. Poor prognosis after out-of-hospital PEA has been reported after blunt trauma,8 unwitnessed arrests (especially in the geriatric population)2,3,9 and if bystander-CPR is not provided.2 Our hypothesis is that due to the diversity of haemodynamic states judged to be PEA, there are few factors (compared to VF and asystole) that are associated with survival. Therefore it may be even more difficult to predict the prognosis in individual cases. We suspect that the marginal perfusion provided by the cardiac contractions may cause all delays to be less important in PEA than in the other initial rhythms of cardiac arrest. The purpose of this study was to describe factors associated with short- and long-term survival among 984 patients judged as having PEA as the initial rhythm of out-of-hospital cardiac arrest. Short-term survival was defined as survival to hospital admission and long-term survival was defined as survival to 30 days after the arrest.
Materials and methods Study setting Helsinki is a middle-sized urban city with 559,000 inhabitants. It is served by a three-tiered Emergency Medical Services (EMS) consisting of seven to eight BLS-units depending on the time of the day, five advanced life support (ALS)-units (of which three are ambulances, one is a medical supervisor unit and one is a fire engine) and a physician staffed mobile intensive care unit (MICU). Units are distributed to eight regional rescue stations. According to dispatching protocols, in suspected cardiac arrest at least two units (of which one is an ALS-unit) are dispatched. If a fire engine can reach the scene faster than an ambulance it is used as a first responding unit (FRU). There are seven fire engines that provide the same level of care as BLS ambulances. If cardiac arrest is not suspected and the patients’ risk is considered to be low or intermediate, an ALS-unit is dispatched only if the scene is within the unit’s own area. If a BLS-unit finds a patient in unsuspected cardiac arrest, an ALS-unit is requested from the dispatching centre and resuscitation is initiated. The time intervals are calculated from the beginning of the emergency call (time zero) to the arrival of BLS and ALS units at the patients’ side. In EMS witnessed arrests the actual time of collapse is used as time zero.
T. V¨ ayrynen et al. In Helsinki EMS all BLS-units are capable of defibrillation, intubation of an adult cardiac arrest victim and basic cardiopulmonary resuscitation. ALS units are equipped additionally with i.v. medications, e.g. adrenaline and amiodarone. Guidelines used in cardiopulmonary resuscitation of PEA follow international recommendations, with adrenaline [epinephrine] (1 mg at 3—5 min intervals) as the vasopressor. No other drugs such as atropine are given routinely during CPR. Drugs are given either into the external jugular or the antecubital vein. Intubation precedes i.v. access in priority. Patients are intubated routinely as soon as possible. If intubation by an EMT or a paramedic cannot be done with two attempts bag-valve mask ventilation is carried out until the arrival of MICU. According to Helsinki EMS do-not-attempt resuscitation (DNAR) guidelines resuscitation is not attempted in PEA (excluding cases of hypothermia and/or near-drowning) if the patient has a living will or terminal illness, traumatic arrest or when resuscitation (including bystander-CPR) has not been initiated within 15 min. Resuscitation is terminated according to DNAR guidelines after 20 min of resuscitation in PEA. The decision to withhold resuscitation or to terminate ongoing resuscitation is always made by a physician. DNAR guidelines were introduced by the end of 1997.
Data collection The Institutional Review Board of Helsinki University Hospital approved the study. The study plan was retrospective but the data were collected prospectively by EMS-physicians into the cardiac arrest database. The on-call physician completed a data collection form based on Utstein recommendations immediately after arrest. Information regarding the cause of arrest, survival to discharge and overall performance category (OPC) on discharge were achieved later from hospital files by a physician responsible for the maintenance of the database. Patients were assigned to OPC 1—4 according to hospital records at the time of hospital discharge. OPC 1 includes patients with full functional recovery, OPC 2 includes patients with moderate disability (but still able to perform independently in daily activities), OPC 3 includes patients with severe disability (in need of assistance in daily activities) and patients in OPC 4 are in a coma. The definition for discharged alive was that after hospital care was not needed any more. Patients were discharged from hospital either to their homes or in some cases to nursing homes. No arrested patient with secondary sign of death (rigor mortis, dependent lividity) were included for analysis.
Statistical methods For normally distributed continuous variables mean and standard deviation (S.D.) are presented and independent samples T-test is used as a test of significance. Median and interquartile range (IQR) are presented for nonparametrically distributed data and Mann—Whitney U-test is used as a test of significance. All categorical variables were analysed using chi-square test or Fischer’s exact test. All significance tests were two-tailed with p less than 0.05 considered significant. Logistic regression was used to calculate
Who survives from out-of-hospital PEA?
209
adjusted odd ratios (OR) for short- and long-term mortality using variables that were univariately associated with shortand long-term survival with significance of p < 0.5. In the univariate analysis continuous variables were dichotomized according to mean or median values. In the multivariate analysis continuous variables were used and FRU and ALS delays were transformed to natural logarithmic scale due to their skewed distribution. Hosmer—Lemeshow-test was used to assess the goodness of fit of these models. Reasonable fit was assumed since the results were not significant. A sample size calculation was not performed since this was a retrospective study in which all patients in the Helsinki EMS cardiac arrest database were included.
Results During the study period (1 January1997—31 December 2005) there were 3291 out-of-hospital cardiac arrests. The initial rhythm was PEA in 984 (29.9%) cases. These 984 arrests form the study population. The cardiac arrest was bystander-witnessed in 55.2% (n = 543), witnessed by EMS-personnel in 30.9% (n = 304) and unwitnessed in 13.8% (n = 136) of cases. Data regarding the witness status of one patient was missing. Resuscitation was attempted in 80.2% (n = 789) of arrests. The majority (68.2%, n = 133) of decisions to withhold resuscitation were based on Helsinki EMS DNAR guidelines. Survival to hospital admission rate was 25.7% (n = 253). In-hospital mortality rate was 75.1%. Survival to discharge rate was 6.4% (n = 63), and 5.8% (n = 57) were alive 30 days after the arrest. Seventy-one percent (n = 45) of discharged patients were in OPC 1—2. Long-term survival rate was 5.7% in bystanderwitnessed arrests, 3.7% in unwitnessed arrests and 6.9% in EMS-witnessed arrests. Characteristics of long-term survivors and those who died within 30 days of the arrest are presented in Table 1. Cases in which resuscitation was not attempted are excluded. Longterm survivors had significantly shorter ROSC delays and had received adrenaline less frequently and in smaller total
Table 1
amounts. There were non-significant trends (with p < 0.2) towards survivors being younger, having received bystanderCPR less frequently, having shorter FRU delays and also being defibrillated less often. Total incidence of bystander-CPR in PEA (excluding EMS-witnessed arrests) was 29.5% in arrests in which resuscitation was attempted and 32.6% if all arrests were included. Results of the univariate analysis are presented in Table 2. Arrests in which resuscitation was not attempted were excluded. Cardiac cause and defibrillation (in case of conversion of PEA into a shockable rhythm) were associated with decreased short-term survival rate and use of adrenaline was associated with decreased long-term survival rate. All factors that were associated with survival with p < 0.5 were included in multivariate analysis. Results of the multivariate analysis are shown in Table 3. The use of adrenaline in resuscitation was the only significant factor found and was associated with decreased long-term survival. Results of the univariate analysis of factors associated with 30-day survival among admitted patients are presented in Table 4. ROSC-delay under 19 min and the use of adrenaline were the only factors univariately associated with survival. Shorter delays to ROSC increased the survival rate and use of adrenaline decreased the survival rate. All factors that were associated with survival (with p < 0.5) were included in multivariate analysis. Increasing ROSC delay was the only significant factor, which will decrease the chance of survival. Adjusted OR for survival for increasing ROSC delay (per minute) was 0.92, 95% CI 0.85—0.995 (p = 0.037). Survival to different end-points according to confirmed diagnosis is presented in Table 5. Cardiac cause was the most common diagnosis among long-term survivors. Of the 25 long-term survivors with cardiac aetiology, 6 had suffered a myocardial infarction. In-hospital mortality rate after arrests due to a cardiac cause was 72.1%. There were no survivors to discharge in OPC 1—2 after a bystander-witnessed arrest (excluding cases of hypothermia and/or near-drowning) with an FRU-delay over 14 min, or who were resuscitated for more than 20 min. There was no survivors to discharge from hospital in OPC 1—2 after an
Characteristics of long-term survivors and non-survivors
Male (gender) Age (years) Arrest at home Bystander-witnessed arrest B-CPRa (data missing: 0, 9) FRU-delay (data missing: 1, 3) ALS-delay (data missing: 1, 8) ROSC-delay (data missing: 3, 11) Adrenaline used Amount of adrenaline (mg) Defibrillated Cardiac cause (data missing: 0, 98)
n (%) Mean (S.D.) n (%) n (%) n (%) Median (IQR) Median (IQR) Median (IQR) n (%) Median (IQR) n (%) n (%)
Long-term survivors (n = 57)
Non-survivors (n = 732)
p
40 (70.2) 63 (17) 31 (54.4) 31 (54.4) 8 (16.7) 5 (0—8.5) 9 (4.5—13.5) 16 (7—21) 39 (68.4) 2 (0—3) 9 (15.8) 25 (43.9)
448 (61.2) 67 (17) 458 (62.6) 412 (56.3) 174 (28.1) 7 (0—10) 10 (5—14) 20 (14—26) 664 (90.7) 4 (2—5) 189 (25.8) 336 (47.9)
0.204 0.097 0.257 0.784 0.094 0.166 0.221 <0.0001 <0.0001 0.006 0.112 0.213
Arrests in which resuscitation was not attempted are excluded. B-CPR, bystander-cardiopulmonary resuscitation; FRU, first responding unit; ALS, advanced life support; ROSC, return of spontaneous circulation. a Arrests witnessed by the ambulance personnel are excluded. Number of cases with missing data are given for both groups, respectively.
210
Table 2
Results of the univariate analysis Short-term survival rate (%)
OR
95% CI
p
Long-term survival rate (%)
OR
95% CI
p
Male gender
Yes No
30.5 34.2
0.845
0.622—1.148
0.307
8.2 5.6
2.492
0.830—2.682
0.204
Age under 66 years
Yes No
34.8 30.1
1.244
0.917—1.687
0.161
7.7 6.9
1.134
0.657—1.959
0.674
Arrest at home
Yes No
29.9 35.3
0.779
0.574—1.058
0.116
6.3 8.7
0.713
0.415—1.227
0.257
Cardiac cause
Yes No
28.5 44.8
0.491
0.358—0.673
<0.0001
6.9 9.7
0.693
0.401—1.196
0.213
Bystander-witnessed arrest
Yes No
32.5 31.2
1.061
0.785—1.436
0.758
7.0 7.5
0.926
0.539—1.591
0.784
Bystander-CPR
Yes No
29.1 33.8
0.804
0.555—1.165
0.266
4.4 8.2
0.511
0.235—1.115
0.094
FRU-delay under 7 min
Yes No
31.2 32.2
0.952
0.705—1.287
0.76
8.0 6.3
1.292
0.748—2.232
0.406
ALS-delay under 9.5 min
Yes No
31.3 32.0
0.969
0.716—1.310
0.878
7.6 6.8
1.141
0.662—1.967
0.678
Adrenaline used
Yes No
28.2 62.8
0.232
0.146—0.371
<0.0001
5.5 20.9
0.222
0.120—0.409
<0.0001
Defibrillated
Yes No
22.7 35.0
0.547
0.376—0.792
0.001
4.5 8.1
0.539
0.259—1.119
0.112
Continuous variables were dichotomised according to their mean or median values. CPR, cardiopulmonary resuscitation; FRU, first responding unit; ALS, advanced life support; OR, odds ratio and CI, confidence interval. OR value above 1 indicates increased chance of survival.
T. V¨ ayrynen et al.
Who survives from out-of-hospital PEA? Table 3
211
Results of the multivariate analysis Short-term survival
Cardiac cause Adrenaline used Defibrillated
Long-term survival
Adjusted OR
95% CI
p
Adjusted OR
95% CI
p
0.55 0.23 0.65
0.38—0.81 0.13—0.41 0.42—0.99
0.002 <0.0001 0.046
0.16
0.07—0.35
<0.0001
Short-term mortality is mortality before hospital admission, long-term mortality is mortality within 30 days of the cardiac arrest. Only significant (p < 0.05) associations are shown. OR, odds ratio; CI, confidence interval. OR value above 1 indicates increased chance of mortality.
Table 4
Univariate analysis of long-term survival among admitted patients Survival rate (%)
OR
95% CI
p
Gender Male
Yes No
26.7 16.5
1.84
0.976—3.467
0.067
Age under 64 years
Yes No
21.6 23.2
0.911
0.498—1.668
0.878
Arrest at home
Yes No
21.1 24.5
0.882
0.454—1.489
0.544
Cardiac cause
Yes No
24.0 21.6
1.147
0.632—2.082
0.650
Bystander-witnessed arrest
Yes No
21.5 23.9
0.876
0.484—1.585
0.761
Bystander-CPR
Yes No
15.1 24.2
0.556
0.242—1.277
0.186
FRU-delay under 7 min
Yes No
25.6 19.4
1.433
0.788—2.605
0.288
ALS-delay under 10 min
Yes No
25.4 19.8
1.376
0.757—2.503
0.362
ROSC-delay under 19 min
Yes No
30.1 15.4
2.367
1.269—4.415
0.008
Adrenaline used
Yes No
19.7 32.7
0.504
0.260—0.979
0.046
Defibrillated
Yes No
20.0 23.1
0.833
0.375—1.852
0.844
Continuous variables were dichotomised according to their mean or median values. CPR, cardiopulmonary resuscitation; FRU, first responding unit; ALS, advanced life support; ROSC, return of spontaneous circulation; OR, odds ratio and CI, confidence interval. OR value above 1 indicates increased chance of survival.
unwitnessed arrest with duration of resuscitation exceeding 14 min or duration of ALS exceeding 5.5 min.
Discussion The use of adrenaline was the only factor associated with long-term mortality among arrests in which resuscitation was attempted. Survival to 30 days was reduced by over six times if adrenaline was used during the resuscitation. When the subgroup of patients who survived to hospital admission was analysed, a short delay to ROSC was the only factor associated with survival. Both the use of adrenaline and the ROSC delay are factors that are associated with the duration
of resuscitation, i.e. the patients’ responsiveness to resuscitative measures. It was possible that majority of survivors were in fact in pseudo-PEA and therefore responded rapidly to resuscitative measures. Since bystander-CPR and delays to the arrival of the FRU or to the initiation of ALS were not associated with survival, it seems difficult to increase the survival rate from PEA by further improving the chain of survival. Since EMS-witnessed arrests have the highest survival rate, the public should be educated to recognise the warning signs and to call for help before the cardiac arrest occurs. It is probable that the wide spectrum of haemodynamic states that can be judged to be PEA is a major confounding factor in the prognostication of PEA. Some of the patients are in true cardiac arrest while others are merely in profound shock.
212 Table 5 (%)
T. V¨ ayrynen et al. Confirmed diagnosis of patients that survived to hospital admission, hospital discharge or 30 days after the arrest, n Admitted to hospital (n = 253)
Cardiac cause Pulmonary embolism Trauma Intoxication Bleeding (without trauma) Suffocation Drowning Seizure Pneumonia Hanging Asthma Cerebrovascular accident/stroke Sepsis Malignancy Carbon monoxide poisoning Pancreatitis Other obvious cause Data missing
104 (41.1) 18 (7.1) 14 (5.5) 12 (4.7) 10 (4.0) 13 (5.1) 6 (2.4) 5 (2.0) 11 (4.3) 2 (0.8) 4 (1.6) 32 (12.6) 2 (0.8) 1 (0.4) 1 (0.4) 1 (0.4) 16 (6.3) 1 (0.4)
Discharged from hospital (n = 63) 29 7 6 4 3 3 3 2 2 1 1 0 0 0 0 0 2 0
(46) (11) (10) (6) (5) (5) (5) (3) (3) (2) (2) (0) (0) (0) (0) (0) (3) (0)
Long-term survival (n = 57) 25 7 6 4 3 3 2 2 2 1 1 0 0 0 0 0 1 0
(44) (12) (11) (7) (5) (5) (4) (4) (4) (2) (2) (0) (0) (0) (0) (0) (2) (0)
Hypothermia was the cause of the arrest of the only long-term survivor in the ‘‘other obvious cause’’-category.
Cardiac ultrasound examination during the arrest could aid in the differentiation of these states and the evaluation of the prognosis of individual patients. It is also difficult to recognise subgroups that do not benefit from resuscitation. There was no survivor discharged from hospital in OPC 1—2 after bystander-witnessed arrest if the FRU-delay was over 15 min, and in the previous study by Engdahl et al. there were no survivors at all if the FRU-delay exceeded 15 min. Therefore it seems that resuscitation is not beneficial and should probably be withheld if the patient is not reached within 15 min after bystander-witnessed arrest. However, it should be remembered that only a small minority of patients in an urban environment experience such a long delay. There were no long-term survivors who were discharged in OPC 1—2 after bystander-witnessed arrest who were resuscitated for more than 20 min. Prognosis was poorer in unwitnessed than witnessed arrests, but still 3.7% survived after an unwitnessed arrest. As all survivors from unwitnessed PEA responded to ALS very rapidly, prolonged efforts may be futile in this subgroup. Contradicting the previous findings by Engdahl et al., bystander-CPR was not beneficial in this study. In our study the incidence of bystander-CPR was high compared to the previous study by Engdahl et al. (29.5% versus 6.6—10.2%). The incidence of bystander-CPR was lower among survivors in our study, but not reaching statistical significance. In the previous study the incidence of bystander-CPR was significantly higher among survivors than in non-survivors (28% versus 8%, p = 0.008). It is possible that in the previous study bystanders initiated resuscitation in cases that had a better prognosis per se, since the incidence was significantly lower. It is also possible that bystander-CPR is associated with shorter delay to call-for-help, but since the delay to the arrival of the FRU was not associated with survival it should
not alter survival rates significantly. Quality of bystanderCPR may also vary. It is highly speculative whether the marginal perfusion provided by the cardiac contractions in pseudo-PEA diminishes the benefit from bystander-CPR, and the survivors will ultimately survive regardless of bystanderCPR. There were also differences in survival, compared to previous results by Engdahl et al. In our study survival rate to hospital discharge was higher (6.4% versus 2.4%) and the proportion of survivors in OPC 1—2 was also higher (over 70% versus almost 50%). More arrests were witnessed in the present study, which may explain the difference in the survival rates at least partly. In particular the number of EMS-witnessed arrests was high in the present study. The incidence of bystander-CPR was also higher in the present study, but in our multivariate analysis it was not associated with survival. Patients in our study were younger, which may explain at least partly both the higher survival rate and lower incidence of cardiac aetiology. It is also possible that the execution of resuscitation is different, and that intra-resuscitation delays to different interventions (i.e. intubation, intravenous access, and administration of medications) may differ. There are some limitations to this study. Due to the nature of the diagnostic criteria for PEA, there may be differences in the incidence of PEA and pseudo-PEA between EMS systems. In some cases, PEA with a low frequency of electrical complexes may be judged to be asystole. Survival rates and factors associated with survival in different EMSsystems may vary more in PEA than in asystole or VF. There are also major differences between different EMS systems, in terms of structure of the EMS (BLS, ALS or a multi-tiered system) and time delays to the initiation of the treatment. Therefore these results may not be applicable to all EMS systems.
Who survives from out-of-hospital PEA?
Conclusions The only factor associated with long-term survival was the use of adrenaline during resuscitation. The chance of death was over six times higher if adrenaline was used. Among patients who were admitted, a short delay to ROSC was the only factor associated with improved survival. BystanderCPR and delays to the arrival of the FRU or to the initiation of ALS were not associated with survival. Therefore it seems difficult to increase survival rates from PEA by improving the chain of survival, besides by educating the public to call for help before the cardiac arrest occurs. Prognosis is poor if the FRU-delay is over 15 min in bystander-witnessed arrests or if the duration of ALS exceeds 5.5 min in EMS-witnessed arrests, excluding cases of hypothermia and near-drowning.
Conflicts of interest There authors hereby state that there is no conflict of interest.
References 1. Ontario Prehospital Advanced Life Support Study Group. Advanced cardiac life support in out-of-hospital cardiac arrest. N Engl J Med 2004;351:647—56.
213 2. Engdahl J, B˚ ang A, Lindqvist J, Herlitz J. Factors affecting short- and long-term prognosis among 1069 patients with out-of-hospital cardiac arrest and pulseless electrical activity. Resuscitation 2001;51:17—25. 3. Pepe PE, Levine RL, Fromm Jr RE, Curka PA, Clark PS, Zachrahiah BS. Cardiac arrest presenting with rhythms other than ventricular fibrillation: contribution of resuscitative efforts toward total survivorship. Crit Care Med 1993;21:143—50. 4. Salen P, O’Connor R, Sierzenski P, et al. Can cardiac sonography be used independently and in combination to predict resuscitation outcomes? Acad Emerg Med 2001;8:610—5. 5. Blaivas M, Fox JC. Outcome in cardiac arrest patients found to have cardiac standstill on the bedside emergency department echocardiogram. Acad Emerg Med 2001;8:616—21. 6. Salen P, Melniker L, Chooljian C, et al. Does the presence or absence of sonographically identified cardiac activity predict resuscitation outcomes of cardiac arrest patients? Am J Emerg Med 2005;23:459—62. 7. Dick WF, Balthasar E, Wisser G, Schneider T. The carotid pulse check revisited: what if there is no pulse? Crit Care Med 2000;28(S):183—5. 8. Martin SK, Shatney CH, Sherck JP, Ho C-C, Homan SJ, Neff J. Blunt trauma patients with prehospital pulseless electrical activity (PEA): poor ending assured. J Trauma 2002;53:876—81. 9. Stratton SJ, Niemann JT. Outcome from out-of-hospital cardiac arrest caused by nonventricular arrhythmias: contribution of successful resuscitation to overall survivorship supports the current practice of initiating out-of-hospital ACLS. Ann Emerg Med 1998;32:448—53.