Original Contribution

Survival in out-of-hospital cardiac arrest before and after use of advanced postresuscitation care

A survey focusing on incidence, patient characteristics, survival, and estimated cerebral function after postresuscitation care^?

Louise Martinell MD ^a, Malena Larsson MD ^a, Angela Bang MD ^b, Thomas Karlsson MSc ^c,

Jonny Lindqvist ^c, Ann-Britt Thoren MD ^c, Johan Herlitz MD, PhD ^c,?

^aMedicine Clinic, NU Hospital Organisation, NAL, 461 85 Trollhattan, Sweden

^bUniversity College of Boras, School of Health Sciences, 501 90 Boras, Sweden

^cInstitute of Medicine, Department of Molecular and Clinical Medicine, Sahlgrenska University Hospital, SE-413 45 Goteborg, Sweden

Received 22 December 2008; revised 28 January 2009; accepted 29 January 2009

Abstract

Background: Knowledge of the epidemiology of postresuscitation care is insufficient. We describe the epidemiology of postresuscitation care in a community from a 26-year perspective, focusing on incidence, patient characteristics, survival, and estimated cerebral function in relation to intensified postresuscitation care and initial arrhythmia.

Methods: The study included patients with out-of-hospital cardiac arrest (OHCA) who were brought alive to a hospital ward in Goteborg, Sweden, between 1980 and 2006. Two periods (1980-2002 and 2003-2006) were compared.

Results: In all, 1603 patients were included. For age, sex, and history, no significant differences between the 2 periods were seen. There was a significant multiple increase in bystander cardiopulmonary resuscitation, the use of coronary angiography, coronary revascularization, and therapeutic hypothermia. The number of patients found in ventricular fibrillation (VF) decreased (P =

.011).For all patients, 1-Year survival did not change significantly (27% vs 32%; P = .14). Among patients found in VF, an increase in 1-year survival was found (37% vs 57%; P b .0001), whereas no significant change was seen in nonshockable rhythm (10% vs 7%; P = .38). Survivors to discharge displaying low cerebral function (ie, Cerebral Performance Categories score >=3) decreased from 28% to 6% (P = .0006) among all patients.

^? This study was supported by grants from The Laerdal Foundation.

* Corresponding author. Fax: +46 31 82 73 75.

E-mail address: [email protected] (J. Herlitz).

0735-6757/$ – see front matter (C) 2010 doi:10.1016/j.ajem.2009.01.042

Conclusion: After the introduction of a more intensified postresuscitation care, there was no overall improvement in survival but signs of an improved cerebral function among survivors. There was a marked increase in survival among patients found in a shockable rhythm but not among those found in a nonshockable rhythm.

(C) 2010

Introduction

Sudden cardiac death is one of the most common causes of death, particularly in the industrialized world [1]. During the past few decades, large-scale efforts have been made to improve outcome after out-of-hospital cardiac arrest (OHCA) [2]. Various links in the Chain of survival have been defined [3]. Recent guidelines define the last link in the chain of survival as postresuscitation care [4]. This link has been particularly highlighted as a result of studies suggesting that hypothermia has a beneficial effect on outcome after OHCA [5].

Our knowledge of the outcome in postresuscitation care is still limited. The first large multicenter report on patients treated for cardiac arrest was published in 1953. The in- hospital mortality rate for 672 adults and children was 50% [6]. In a recent study of 24 132 patients in United Kingdom who were admitted to the critical care unit after cardiac arrest, the in-hospital mortality was 71% [7]. Recent data suggest that the introduction of mechanical revasculariza- tion and hypothermia has been associated with improved outcome among patients brought alive to hospital after OHCA [8,9].

The aim of this survey was to describe the epidemio- logy of postresuscitation care in a well-defined community with half a million inhabitants and 3 hospitals from a 26- year perspective. Among patients brought alive to a hospital ward after OHCA, we focus on the following parameters: (1) incidence, (2) patient characteristics, (3) treatment, (4) survival, and (5) estimated cerebral function among survivors.

Particular attention will be paid to possible changes in outcome in relation to the introduction of therapeutic hypothermia and coronary revascularization among patients found in a shockable and nonshockable rhythm. We have previously reported from this database [10,11].

Patients and methods

Patients

The study included all patients who had an OHCA in the Municipality of Goteborg, Sweden, in whom cardiopulmon- ary resuscitation (CPR) was attempted between October 1, 1980, and June 17, 2006 (excluding the latter part of 2002

and the first part of 2003, as we did not have the capability to evaluate cases at that time for logistical reasons) and who were then brought alive to a hospital ward were included in the survey.

Information on Cerebral performance categories score was gathered by a careful evaluation of the hospital records of the patients’ condition at hospital discharge. Information relating to factors at resuscitation, age, sex, history, and various Diagnostic and therapeutic procedures including therapeutic hypothermia was gathered from prehospital and hospital records. Information relating to factors at resuscitation was collected prospectively, whereas information relating to history and postresuscita- tion care was collected retrospectively a couple of months after hospital discharge (sometimes there was an even longer delay).

Target population

Since 1974, the Municipality of Goteborg has had an area of 445 km², excluding areas of water. The population in 2004 comprised 481 410 inhabitants. Of the total population, 49% were men.

Organization and equipment

All the ambulances were dispatched by one emergency medical service (EMS) dispatch center according to a 2- tier system, that is, for each call judged to involve a cardiac arrest, an advanced cardiac life support unit, if available, and the nearest basic life support unit were dispatched simultaneously. The percentage of patients reached by both tiers was almost 100%. All patients having an OHCA arrived at 1 of the 3 city hospitals in Goteborg. The advanced cardiac life support unit was staffed with a nurse with increasing frequency during the first period (100% in the latter part). During the second period, a nurse was on board in 100% of calls. The basic cardiac life support unit was staffed with ambulance personnel with a 6-week medical training in medicine during major part of the first period. During the second period, the unit was staffed with a nurse with a varying degree of experiences of emergency medicine.

The median interval between call for ambulance and arrival of ambulance during the first period was 5 minutes as compared with 6 minutes during the second period (P b .0001).

Fig. 1 Number of patients exposed to postresuscitation care during 26 years in the community of Goteborg.

Data collection

Patients having an OHCA between October 1, 1980, and June 17, 2006, were included in the survey, regardless of the cause of the arrest and the patient’s age. Patients were divided into 2 periods. The 2 periods included the following times: period 1 = October 1, 1980, to June 30, 2002, and period 2 = May 22, 2003, to June 17, 2006. Data on the entire cardiac arrest cohort were obtained from the Goteborg EMS system. Patients were excluded if the EMS did not attempt resuscitation. Further medical data on patients admitted to hospital were obtained from hospital records and general practitioners‘ records.

Being hospitalized alive was defined as being brought alive to a hospital ward.

Treatment and investigations

There were no strict guidelines for treatment and investigations at any of the 3 hospitals. However, the main indication for therapeutic hypothermia (latter part of the survey) was that the cardiac arrest should be witnessed, the patient should be found in ventricular fibrillation, and the patient should be unconscious on admission to hospital.

According to guidelines in Goteborg, even more patients were included; for example, patients found with an initial rhythm of asystole or pulseless electrical activity while simultaneously considering age and comorbidity. The hope was that they would also benefit from the treatment. Patients receiving therapeutic hypothermia were cooled to a tem- perature of 32?C to 34?C for 12 to 24 hours. The treatment should preferably be started within 2 hours, but it could start as much as 4 hours after the arrest. Cold intravenous saline (4?C, 30 mL/kg) was administered as a rapid infusion. The temperature was maintained by applying ice packs, and two of the hospitals also used external cooling devices.

The main indication for coronary angiography was ST elevation on the electrocardiogram on admission to hospital

or a strong suspicion of myocardial infarction as the Underlying etiology based on clinical and electrocardio- graphic findings.

Definition of CPC score (3)

CPC 1: Good cerebral function. Conscious and alert. Can work and live a normal life. There might be minor psychologic or neurologic defects such as mild dysphasia.

CPC 2: Moderate cerebral dysfunction. Conscious. Cerebral function is good enough to allow part-time work in sheltered environment, using public transport, and handle activities of daily living. However, there might be a more severe cerebral sequelae involving hemiplegia and dysarthria.

CPC 3: Severe cerebral dysfunction. Conscious. Dependent on others for activities of daily living because of severe cerebral dysfunction (at an institution or at home with exceptional support from family members or others).

CPC 4: Coma. No verbal or psychologic communication with others.

CPC 5: Brain dead.

Statistical methods

For comparisons between the 2 periods, Fisher permuta- tion test was used for continuous/ordered variables, whereas Fisher exact test was used for dichotomous variables.

All P values are 2-tailed, and because of the large number of comparisons, the significance level was set at .01.

Results

In all, 1603 patients took part in the survey (1415 and 188 patients, respectively, in the 2 periods). The number of patients who were brought alive to a hospital ward after OHCA varied from 41 to 99 patients a year (Fig 1). However, the mean number of patients receiving postresuscitation care was the same during period 1 compared with period 2 (five a

Plate 1 General results.

Prehospital factors“>Plate 2 Not witnessed cardiac arrest.

Plate 4 emergency medical service personnel-witnessed cardiac arrest.

month). The mean number of patients with OHCA per year in whom CPR was attempted was 222 for period 1 and 242 in period 2. In Plates 1 to 4 is the background population described according to an Utstein template. Note that the overall number of survivors in this template is lower than the number of survivors in Table 1. The Utstein template describes the outcome among patients who had a cardiac arrest judged to be of a cardiac etiology. In this survey (Table 1-3), however, we describe all patients who were brought alive to a hospital ward, regardless of etiology.

Patient characteristics

There was no significant change in age, sex distribution, or history of cardiovascular disease either among all patients

Plate 3 Bystander-witnessed cardiac arrest.

or among patients found in a shockable and a nonshockable rhythm (Tables 1-3).

Prehospital factors

The percentage of patients found in ventricular fibrillation tended to decrease, whereas the percentage of patients experiencing a witnessed cardiac arrest did not change. Cardiopulmonary resuscitation before the arrival of the rescue team increased markedly and significantly in all patients, as well as in the subgroups of patients found in a shockable and a nonshockable rhythm. In a subset of patients (from 1993 and thereafter, regardless of initial arrhythmia), there was a slight increase in the percentage of patients who were crew witnessed from 15% in period 1 to 19% in period 2 (Tables 1-3).

Status on admission to hospital

The percentage of patients who were unconscious and the percentage of patients who had a sinus rhythm did not change significantly (Tables 1-3).

Investigation and treatment

The use of coronary angiography and percutaneous coronary intervention (PCI) increased markedly but not significantly in the subgroup of patients found in a nonshockable rhythm (Table 3).

Treatment with Coronary artery bypass grafting did not change significantly. Treatment with fibrinolysis tended to decrease among patients found in a shockable rhythm and to increase among patients found in a nonshockable rhythm. Therapeutic hypothermia was not used in Goteborg until 2003, but it has since been used increasingly, particularly among patients found in a shockable rhythm. During the first

Period 1, n = 1415			Period 2, n = 188	P	Odds ratio a (95% confidence interval)
Age (y), mean +- SD (3.0) b	67	+- 15	66 +- 16	.63	0.86 c (0.63-1.16)
Sex (%) (1.0)
Female	30		29	.87	0.97 (0.69-1.35)
Factors at resuscitation (%)
Ventricular fibrillation (30.0)	62		52	.01	0.67 (0.49-0.92)
Witnessed cardiac arrest (160.0)	84		81	.39	0.84 (0.56-1.26)
Bystander CPR (99.53)	23		67	b.0001	6.60 (4.51-9.65)
Status on admission to hospital (%)
Unconscious (30.0)	91		92	.49	1.27 (0.70-2.31)
Sinus rhythm (30.11)	56		63	.05	1.38 (1.00-1.90)
History (%)
Myocardial infarction (35.0)	37		31	.16	0.78 (0.56-1.09)
Heart failure (38.0)	30		26	.34	0.83 (0.59-1.18)
Diabetes (38.4)	14		16	.42	1.19 (0.78-1.82)
Investigation and treatment (%)
Coronary angiography (31.0)	11		33	b.001	3.85 (2.72-5.44)
PCI (26.0)	3		20	b.001	7.67 (4.79-12.28)
Coronary artery bypass grafting (29.0)	4		5	.42	1.32 (0.64-2.72)
Fibrinolysis (31.0)	4		3	.54	0.69 (0.27-1.74)
Hypothermia (0.0)	0		41	b.001	Not defined
Survival (%)
30 d (22.0)	33		35	.68	1.07 (0.78-1.48)
1 y (22.0)	27		32	.14	1.29 (0.93-1.80)
CPC score among survivors at discharge (4.0)				.0006	0.17 d (0.06-0.48)
1	52		71
2	19		23
3	25		6
4	3		0
5	0		0
a Odds ratio for period 2 vs period 1 with corresponding 95% confidence interval. b Number of survivors with missing information in the 2 groups respectively. c The percentage of patients above median age was used for odds ratio calculations. d The percentage with a CPC score of 3 or more was used for odds ratio calculations.

2 years of therapeutic hypothermia in Goteborg (June 2003- January 2005), therapeutic hypothermia was given to 33% of patients who were comatose on admission to hospital. This figure had increased toward the end of the period (January 2005-June 2006), where 48% of the comatose patients were treated with therapeutic hypothermia.

Table 1 Changes in various characteristics and outcome in all patients

Between 2005 and 2006, 93% of patients who received therapeutic hypothermia reached the target temperature of 34?C compared with only 52% of the patients during the previous period. The duration of the cooling period had also increased, from a median of 16 hours to a median of 26 hours (Tables 1-3).

Survival

Survival to 30 days

In the all-patients group, there was no significant change in 30-day survival (Table 1). However, in the subgroup of patients found in a shockable rhythm, there was a trend toward a significant increase (Table 2).

Survival to 1 year

Among all the patients, there was no significant change in 1-year survival (Table 1). Among patients found in a shockable rhythm, however, there was a highly significant increase (Table 2).

Cerebral performance at hospital discharge

The degree of CPC among all the patients who were discharged alive from hospital changed significantly. The percentage with a CPC score of 3 or more (poor cerebral function) was reduced from 28% in period 1 to 6% in period 2 (Table 1).

Discussion

This survey covers postresuscitation care within a well- defined community for a period of 26 years (with the exception of 11 months) and therefore provides long-term

Period 1, n = 861			Period 2, n = 85	P	Odds ratio a (95% confidence interval)
Age (y), mean +- SD (3.0) b	69	+- 12	66 +- 15	.23	0.76 c (0.50-1.17)
Sex (%) (1.0)
Female	24		20	.45	0.79 (0.47-1.35)
Factors at resuscitation (%)
Witnessed cardiac arrest (79.2)	91		89	.57	0.82 (0.40-1.64)
Bystander CPR (43.24)	28		79	b.0001	9.37 (5.20-16.90)
Status on admission to hospital (%)
Unconscious (12.0)	87		85	.50	0.82 (0.44-1.53)
Sinus rhythm (10.6)	57		63	.37	1.26 (0.80-1.98)
History (%)
Myocardial infarction (11.0)	46		41	.44	0.82 (0.53-1.27)
Heart failure (14.1)	36		32	.50	0.85 (0.54-1.33)
Diabetes (14.1)	13		18	.21	1.43 (0.82-2.51)
Investigation and treatment (%)
Coronary angiography (10.0)	16		58	b.0001	7.17 (4.59-11.2)
PCI (7.0)	4		34	b.0001	12.64 (7.29-21.9)
Coronary artery bypass grafting (9.0)	6		7	.49	1.33 (0.59-3.03)
Fibrinolysis (9.0)	6		1	.052	0.18 (0.02-1.31)
Hypothermia (0.0)	0		54	b.0001	Not defined
Survival (%)
30 d (15.0)	45		59	.01	1.74 (1.13-2.67)
1 y (15.0)	37		57	b.0001	2.23 (1.45-3.43)
CPC score among survivors at discharge (4.2)				.0006	0.17 d (0.05-0.56)
1	54		76
2	20		19
3	24		6
4	2		0
5	0
a Odds ratio for period 2 vs period 1 with corresponding 95% confidence interval. b Number of survivors with missing information in the 2 groups respectively. c The percentage of patients above median age was used for odds ratio calculations. d The percentage with a CPC score of 3 or more was used for odds ratio calculations.

surveillance of chosen end points in this population. We found that the mean number of patients receiving post- resuscitation care per month was similar during the 2 periods. Our results indicate that about 1 per 100 000 inhabitants in a month receives postresuscitation care after initial survival of OHCA.

Table 2 Changes in characteristics and outcome in patients with ventricular fibrillation

One important observation was that, during the time when an intensification of postresuscitation care, including an increase in coronary revascularization and the introduction of therapeutic hypothermia, was used, the percentage of patients admitted alive to hospital after OHCA who were found in ventricular fibrillation had decreased.

Although this is in agreement with previous observations [12,13], it is a paradoxical finding because the increase in bystander CPR and the increase in crew-witnessed cases that was found are both expected to increase the proportion of patients found in ventricular fibrillation. The mechanism behind the decrease in ventricular fibrillation can only be speculated upon. One hypothesis is that new treatments such as coronary artery bypass grafting, PCI, ?-blockers, lipid-lowering drugs, angiotensin-converting enzyme inhi-

bitors, and aspirin have changed the epidemiology in coronary artery disease. When cardiac arrest occurs today, many patients might have reached an end-stage heart disease that might more frequently be reflected in asystole rather than ventricular fibrillation as the first recorded rhythm. Another possible contributing factor is the prolonged Ambulance response time, which has been found in Goteborg and is expected to be reflected in a lower occurrence of ventricular fibrillation.

Despite the marked increase in bystander CPR, the slight increase in crew-witnessed cardiac arrest, and the introduc- tion of therapeutic hypothermia and coronary revasculariza- tion in this patient population, the overall survival did not improve over time. The expected increase in survival was most probably counteracted by a decrease in the percentage of patients found in ventricular fibrillation, as most survivors have been found among patients found in this type of arrhythmia [14].

This overall negative result could be interpreted as a complete lack of effect of therapeutic hypothermia and coronary revascularization.

Period 1, n = 524			Period 2, n = 86	P	Odds ratio a (95% confidence interval)
Age (y) (2,0) b	64	+- 20	66 +- 17	.40	1.04 c (0.66-1.64)
Sex (%) (1.0)
Female	39		41	.81	1.06 (0.67-1.69)
Factors at resuscitation (%)
Witnessed cardiac arrest (61.6)	71		74	.69	1.13 (0.66-1.94)
Bystander CPR (27.24)	15		55	b.0001	7.17 (4.10-12.54)
Status on admission to hospital (%)
Unconscious (15.4)	97		100	.24
Sinus rhythm (17.4)	53		63	.07	1.56 (0.96-2.52)
History (%)
Myocardial infarction (21.5)	23		21	.78	0.89 (0.50-1.57)
Heart failure (22.2)	19		19	1.00	1.00 (0.55-1.79)
Diabetes (21.3)	14		13	1.00	0.94 (0.48-1.87)
Investigation and treatment (%)
Coronary angiography (17.0)	3		7	.10	2.46 (0.93-6.53)
PCI (15.0)	1		6	.02	4.43 (1.37-14.28)
Coronary artery bypass grafting (16.0)	0.6		1	.47	1.98 (0.20-19.26)
Fibrinolysis (16.0)	0.6		5	.01	8.21 (1.80-37.36)
Hypothermia (0.0)	0		29	b.0001	Not defined
Survival (%)
30 d (6.0)	13		9	.38	0.67 (0.31-1.46)
1 y (6.0)	10		7	.44	0.64 (0.27-1.54)
CPC score among survivors at discharge (0.1)				.36	0.20 d (0.02-1.72)
1	42		43
2	12		43
3	36		14
4	10		0
5	0		0
a Odds ratio for period 2 vs period 1 with corresponding 95% confidence interval. b Number of survivors with missing information in the 2 groups respectively. c The percentage of patients above median age was used for odds ratio calculations. d The percentage with a CPC score of 3 or more was used for odds ratio calculations.

Signs of cerebral damage were estimated retrospectively using information from medical records. This could be regarded as a rough estimate and should therefore be interpreted with caution. However, we have previously shown that a measurement of this kind is predictive of the risk of death after hospital discharge [15].

Table 3 Changes in characteristics and outcome in patients found in a Non-shockable rhythm

We found a marked decrease in the percentage of patients with indirect signs of poor cerebral function among all survivors of OHCA. This could only be addressed satisfactorily among patients found in a shock- able rhythm, as the number of survivors among those found in a nonshockable rhythm was very small. The mechanism behind this observation is unclear and could depend on increasing bystander CPR, crew-witnessed cases, therapeutic hypothermia, or coronary revasculariza- tion. One might argue that advanced directives in recent years have more frequently indicated that life support should be stopped in the event of serious brain dysfunc- tion. However, such an explanation should be expected to be associated with a decrease in overall survival that was not found in this survey.

Among patients found in ventricular fibrillation, we observed a marked increase in survival. This was not seen among patients found in a nonshockable rhythm.

It is only possible to speculate about the possible mechanisms behind this finding. The increase in coronary revascularization and therapeutic hypothermia was far more marked among patients found in ventricular fibrillation. To date, therapeutic hypothermia has only been shown to increase survival among patients found in ventricular fibrillation [5]. The use of PCI among patients having OHCA has never been studied in a randomized trial. Observational studies suggest that this treatment might increase survival [16].

On theoretical grounds, it seems reasonable to assume that a coronary occlusion as the mechanism behind a cardiac arrest is more common among patients who are found in ventricular fibrillation. It can therefore be assumed that coronary revascularization might be successful in a higher percentage of patients found in ventricular fibrillation.

On the other hand, bystander CPR, which was observed to increase among patients found in both a shockable and a

nonshockable rhythm, has been shown to be associated with an increase in survival after OHCA in both types of arrhythmia [17-19].

Parallel to the increase in bystander CPR among non- crew-witnessed cardiac arrests, we observed a slight increase in the percentage of crew-witnessed cases. This may indicate that people tend to call earlier when they experience warning symptoms of a threatened myocardial infarction such as chest pain. This may have contributed to the increase in survival among patients found in a shockable rhythm.

Finally, patients found in ventricular fibrillation conform to the subset in which the use of an internal cardioverter defibrillator is often helpful. The percentage of patients who received an internal cardioverter defibrillator in this survey was not reported.

One might argue that the proportion of patients who were found in ventricular fibrillation and the proportion of patients who had a witnessed cardiac arrest were high in this survey as compared with other surveys. This is explained by our survey that is selected by definition because it only includes patients who were brought alive to a hospital ward and thus were initially successfully resuscitated. Such patients are expected to more frequently be found in ventricular fibrillation and to have a witnessed cardiac arrest.

Future aspects

Although the observation among patients found in ventricular fibrillation is encouraging as well as the results indicating an overall decrease in signs of cerebral dysfunc- tion, we must admit that the overall survival did not change. A pessimistic approach would state that given the lack of improvement in overall survival over time, the cost of the medical resources directed to interventions such as therapeutic hypothermia and coronary revascularization for OHCA patients needs to be quantitated and considera- tion be given to the potential value of shifting these intervention resources to primary coronary artery disease

prevention programs.

An alternative approach would be to highlight the need for further randomized trials evaluating the impact of therapeutic hypothermia and coronary revascularization in subsets of patients surviving the initial phase of OHCA.

Finally, while awaiting such study, registry data from other communities should either confirm or reject the hypothesis raised from a subset analysis in the survey, ie, an increasing use of therapeutic hypothermia and coronary revascularization in a community, is associated with an increased survival among patients found in ventricular fibrillation.

Limitation

There are a number of facts that limit the interpretation of our data as follows:

1. Some information was missing for most of the variables that were evaluated.
2. We lack information from a 6-month period in 2002 and 5 months in 2003 due to logistical reasons.
3. It was only possible to perform PCI at 1 of the 3 hospitals at the time of the survey. However, patients could be quickly transported to this hospital from one of the others (10 minutes’ transport time).
4. Information on the CPC score was collected retrospectively and should therefore be interpreted with some caution.

Conclusion

After the introduction of a more intensified postresuscita- tion care, there was no overall improvement in survival, but there were signs of an improved cerebral function among survivors. There was a marked increase in survival among patients found in a shockable rhythm but not among those found in a nonshockable rhythm.

References

1. Zheng Z, Croft J, Giles H, et al. Sudden cardiac death in the United States, 1989-1996. Circulation 2000;102(18):841 [abstract].
2. Eisenberg MS, Mengert T. Cardiac resuscitation (review). N Engl J Med 2001;344:1304-13.
3. Cummins RO, Ornato JP, Thies WH, et al. Improving survival from sudden cardiac arrest: the “chain of survival” concept. Circulation 1991;83:1832-47.
4. American Heart Association. 2005. guidelines for CPR and ECC. Postresuscitation support. Circulation 2005;112:IV-84-IV-88.
5. The Hypothermia after Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med 2002;346:549-56.
6. Stephenson Jr HE, Reid LC, Hinton JW. Some common denominators in 1200 cases of cardiac arrest. Ann Surg 1953;137:731-44.
7. Nolan JP, Laver SR, Welch CA, et al. Outcome following admission to UK intensive care units after cardiac arrest: a secondary analysis of the ICNARC Case Mix Programme Database. Anaesthesia 2007;62: 1207-16.
8. Sunde K, Pytte M, Jacobsen D, et al. Implementation of a standardised treatment protocol for post resuscitation care after out-of-hospital cardiac arrest. Resuscitation 2007;73:29-39.
9. Busch M, Soreide E, Lossius HM, et al. Rapid implementation of therapeutic hypothermia in comatose out-of-hospital cardiac arrest survivors. Acta Anaesthesiol Scand 2006;50:1277-83.
10. Herlitz J, Bang A, Gunnarsson J, et al. Factors associated with survival to hospital discharge among patients hospitalised alive after out-of- hospital cardiac arrest: change in outcome over 20 years in the community of Goteborg, Sweden. Heart 2003;89:25-30.
11. Werling M, Thoren AB, Axelsson C, et al. Treatment and outcome in post-resuscitation care after out-of-hospital cardiac arrest when a modern therapeutic approach was introduced. Resuscitation 2007;73: 40-5.
12. Kuisma M, Repo J, Alaspaa A. The incidence of out of hospital ventricular fibrillation in Helsinki Finland from 1994 to 1999. Lancet 2001;358:473-4.
13. Herlitz J, Engdahl J, Svensson L, et al. Decrease in the occurrence of ventricular fibrillation as the initially observed arrhythmia after out-of- hospital cardiac arrest during 11 years in Sweden. Resuscitation 2004; 60:283-90.
14. Herlitz J, Engdahl J, Svensson L, et al. Factors associated with an increased chance of survival among patients suffering from an out-of- hospital cardiac arrest in a national perspective in Sweden. Am Heart J 2005;149:61-6.
15. Herlitz J, Ekstrom L, Wennerblom B, et al. Prognosis among survivors of prehospital cardiac arrest. Annals Emerg Med 1995;25: 58-63.
16. Spaulding CM, Joly LM, Rosenberg A, et al. Immediate coronary angiography in survivors of out-of-hospital cardiac arrest. N Engl J Med 1997;336:1629-33.
17. Stiell IG, Wells GA, DeMaio VJ, et al. for the OPALS Study Group. Modifiable factors associated with improved Cardiac arrest survival in a multicenter basic life support defibrillation system: OPALS study phase I results. Ann Emerg Med 1999;33:44-50.
18. Holmberg M, Holmberg S, Herlitz J. Effect of bystander cardiopul- monary resuscitation in out-of-hospital cardiac arrest patients in Sweden. Resuscitation 2000;47:59-70.
19. Herlitz J, Svensson L, Engdahl J, et al. Characteristics and outcome in out-of-hospital cardiac arrest when patients are found in a non- shockable rhythm. Resuscitation 2008;76:31-6.