American Journal of Emergency Medicine (2012) 30, 1712-1718

Original Contribution

In-hospital cardiac arrest characteristics and outcome after defibrillator implementation and education: from 1 single hospital in Sweden^?

Marie-Louise Sodersved Kallestedt PhD ^a,?, Anders Berglund PhD ^b,

Mats Enlund MD ^a, Johan Herlitz MD ^c

^aUppsala University, Centre for Clinical Research, Vasteras, Sweden

^bDepartment of Medical Epidemiology and Biostatistics, Karolinska Institute, Stockholm, Sweden

^cCentre for Pre-hospital research, Western Sweden University College of Boras and Sahlgrenska University Hospital, Gothenburg, Sweden

Received 28 October 2011; revised 25 January 2012; accepted 26 January 2012

Abstract

Background: Survival after in-hospital cardiac arrest has been reported to be surprisingly low without any major improvement during the last decade.

Aims: The aim of this study is to evaluate the clinical impact (delay to defibrillation and survival after CA) of an intervention within 1 single hospital (Vasteras, Sweden), including (1) a systematic education of all health care professionals in cardiopulmonary resuscitation and (2) the implementation of 18 automated external defibrillators.

Methods: Information was retrieved from the Swedish National Register of Cardiopulmonary Resuscitation. The differences between the 2 calendar periods were evaluated by ?² and Fisher exact tests. Logistic regression was used to control for potential confounders.

Results: In total, there were 73 in-hospital CAs before (12 months) and 133 after (18 months) the intervention. The overall delay to defibrillation was not reduced after the intervention, and the proportion of survivors to hospital discharge was 26% before and 32% after the intervention (P =.51). cerebral function, however, was improved after the intervention (as judged by the Cerebral Performance Categories score; P b .001). Thus, the proportion of survivors among all CA patients discharged with a cerebral performance scale score of 1 or 2 (good or acceptable cerebral function) increased from 20% to 32%.

Conclusion: An intervention within 1 single hospital (systematic training of all health care professionals in cardiopulmonary resuscitation and implementation of automated external defibrilla- tors) did not reduce treatment delay or increase overall survival. Our results, however, suggest indirect signs of an improved cerebral function among survivors.

(C) 2012

^? Confticts of interests: All the authors declare no confticts of interest and no financial interests in the publishing of the manuscript.

* Corresponding author. Centre for Clinical Research, Central Hospital, Vasteras, S-721 89 Vasteras, Sweden. Tel.: +46 21 17 58 02; fax: +46 21 17 37 33.

E-mail addresses: [email protected] (M.-L.S. Kallestedt), [email protected] (A. Berglund), [email protected] (M. Enlund), [email protected] (J. Herlitz).

0735-6757/$ - see front matter (C) 2012 http://dx.doi.org/10.1016/j.ajem.2012.01.026

Introduction

During the last 2 decades, action has been taken to improve the prognosis for patients having Cardiac arrest outside hospital [1]. Survival after in-hospital CA has been reported to vary considerably in some reports [2-4], whereas other studies report an average survival rate of approximately 15%, despite the development and widespread implementa- tion of basic life support (BLS) and advanced cardiac life support guidelines and training [5-7]. Cardiac arrest is a complication that can occur in any hospital ward [8]. There has been contradictory evidence presented on the use of AEDs in hospital concerning the effects on survival rates [9,10]. Patients with in-hospital CA have more concomitant diseases than patients with out-of-hospital CA, which may inftuence outcome [2]. Studies have shown a variation in functional status among survivors after CA [11,12]. A systematic review indicated that the quality of life after CA is good [13].

The Swedish National Registry of Cardiopulmonary Resuscitation (NRCR) was established in 2005 to monitor cardiopulmonary resuscitation (CPR) care. Vasteras Central Hospital joined the registry on May 1, 2006. Over a 10-year period, the organization of CPR training has improved in Sweden [14]. There is, however, a lack of knowledge about the inftuence of CPR training on outcome after in-hospital CA. Previously, no hospital in Sweden has studied their CA survival rates before and after reorganization (implementa- tion of AEDs and CPR training among the health care professionals). The aim of the present study was to describe the clinical impact of a systematic CPR education and training for all hospital staff in addition to the implementa- tion of AEDs throughout the hospital, focusing on outcome after in-hospital CA. In more detail, we sought to describe if there were any changes from before to after intervention, considering: time to first defibrillation, alive at discharge from hospital, alive after 30 days, functional status among survivors, and the proportion of victims who received chest compressions and defibrillation before the arrival of the rescue team, particularly in nonintensive care areas.

Patients and methods

Patient recruitment was carried out during the period from May 1, 2006, to April 30, 2007 (12 months), from now on, called calendar period 1 (before intervention; education and AED implementation), and April 1, 2009, to December 31,

2010 (18 months), called calendar period 2 (after interven- tion; education and AED implementation). Hypothermia was introduced and used in patient care from year 2005. The data collected prospectively were all patients who had a suspected CA and for whom the rescue team was called. Patients with a CA before admission to hospital were excluded and assigned to the out-of-hospital CA registry. A case record form was

filled in for each CA by a nurse on the ward where the arrest took place and by a nurse from the department of anesthesia. Furthermore, clinical data were collected retrospectively from medical case records. These data were reported to and taken together by the NRCR of in-hospital CA. Data accuracy was ensured by an external controller who checked the data for completeness and accuracy. This person worked for the NRCR as a professional data controller with the aim to secure that the data given to the NRCR was in accordance with the medical patient case records and alarm system at the hospital.

The Swedish NRCR

The NRCR [15] is an ongoing national survey of all CAs in Sweden, regardless of whether they took place inside or outside a hospital. The section of the register, which covers all CAs inside hospital, was initiated in 2005. Today, 54 (74%) of 73 hospitals in Sweden, which have an organization for treatment of in-hospital CA, are included in the register.

Organization

Before reorganization

Vasteras Central Hospital includes 30 wards and approx- imately 420 beds. The rescue team was previously organized so that when a CA occurred, the hospital staff dialed a specific emergency number. A rescue team consisting of an anesthetist, anesthetic nurse, and an internal medicine physician was then immediately alerted. Before reorganization, the wards had no standardized emergency equipment, and the ward staff possessed varying degrees of knowledge about performing CPR. Most of the staff had not been trained in BLS for several years. No area for resuscitation training was available, and there were only approximately 20 instructors able to teach CPR and defibrillation. These instructors were mainly working at the intensive, emergency, and coronary care units. manual defibrillators were available at the emergency, intensive, and coronary care units; the angiography laboratory; and at 1 regular ward. The nurses in these wards were trained and delegated to perform defibrillation with a manual defibrillator.

When a CA occurred in a regular ward, an assistant nurse or a nurse arrived with a manual defibrillator from the emergency unit. The health care professional with respon- sibility of care for the patient had to wait until the arrival of someone with authority to use the manual defibrillator. At the end of 2006, the coronary care unit and the angiography laboratory received AEDs with electrocardiographic displays and vocal instructions to guide the health care professionals through the resuscitation program.

Reorganization and intervention

From May 2007, courses in CPR with AEDs started at the regular wards. A room for resuscitation training was arranged, and 30 new instructors were educated. Any

instructor from any ward was given the opportunity to make a reservation for the training room. The training in CPR followed the guidelines from the Swedish Resuscitation Council. In Sweden, the use of an AED is taught in a 4-hour course that also includes theory and practical training in the importance of AED early use and basic CPR. It also includes the use of oxygen and ventilation with mouth-to-mask technique and the use of suction devices for the clearing of airways [16]. From May 1, 2008, 18 AEDs were strategically placed around the hospital. A mobile intensive group was started in May 2008.

Table 1 Demographic data (n) (% if not otherwise stated)

Calendar period P

Before

intervention ^a intervention ^b

After

radiology Other Etiology ^d

Myocardial ischemia/

n (%)			n (%)
Age in years c	73 (15)	72 (13)
Total no. of CAs	73	133		.54
Sex, women Previous history	27 (37)	60 (45)		.30
Heart failure	35 (53)	33 (25)		b.001
Diabetes	21 (31)	31 (23)		.31
Myocardial infarction	23 (36)	30 (23)		.06
Respiratory insufficiency Stroke	9 (14) 7 (11)	27 (21) 16 (12)		.33 .82
Cancer	6 (10)	28 (21)		.07
Cancer with metastases	2 (100)	11 (41)		.19
Ongoing myocardial infarction Ongoing stroke	20 (45) 2 (3)	19 (15) 2 (1)		b.001 .59
Witnessed CA Place of CA	54 (74)	114 (86)		.02
Coronary care unit 20 (27)		25 (19)		.16
Intensive care unit 8 (11)		9 (7)		.30
Emergency department 3 (4)		6 (5)		-
Angiography laboratory 4 (5)		14 (11)		.30
Operation ward 0 (0)		1 (0.8)		-
General wards 35 (48)		73 (55)		.38
Clinic, laboratory, 3 (4)		3 (2)		.67

The health care professionals were tested for their theoretical knowledge and attitudes to CPR before and after the educational intervention, resulting in an overall improvement in their theoretical knowledge [17] as well as their attitudes to CPR (Sodersved Kallestedt et al, submitted for publication).

Statistical methods

To evaluate the proportion of the independent variables between the 2 calendar periods, the ?2 and Fisher exact tests were used. P b .05 was considered significant. In a subsequent step, a binary logistic regression with estimated odds ratios (ORs) and 95% confidence interval (CI) was used to study the effect of calendar period in the evaluation of return of spontaneous circulation (ROSC) and survival immediately after termination of CPR, at discharge from hospital, and at 1 month. All logistic regression models were adjusted for witnessed CA, shockable rhythm, and for where the arrest took place. All tests were 2 sided, and SAS 9.2 (SAS, Cary, NC) was used for statistical analysis. Analyses of survival to hospital discharge and to 30 days included only the first CA at each hospitalization during the study periods.

infarction
Arrhythmia	17 (37)	45 (49)	.20
Hypotension/	0 (0)	1 (1)	-
hypoperfusion
acute pulmonary edema	0 (0)	2 (2)	.55
Acute respiratory	2 (4)	5 (6)	-
insufficiency
Other	8 (17)	23 (25)	.39
a Calendar period before intervention, 12 months. b Calendar period after intervention, 18 months. c Data are presented as means (SD within parenthesis). d Missing information, n = 27, before intervention and, n = 42, after intervention.

0 (0)	2 (2)	.54
19 (41)	16 (16)	.01

Results

The present intervention study included 190 patients who, together, had 206 CAs. Data were missing on the first registered rhythm for 7 of 87 cases, which required defibrillation, and for 1 patient for the variable “alive after 30 days.” The rescue team was called a median of 6 times a month for CA and 26 times a month for any cause. This did not change after the intervention.

Age, sex, previous history, and etiology

There were no significant differences between the 2 calendar periods regarding patient age and previous history of sickness (Table 1). myocardial infarction/ischemia was a more frequent cause of CA before the intervention (P = .01). However, the number of patients with missing information on this variable was relatively high.

Treatment at resuscitation and various aspects of delay times

There was no significant difference between the 2 calendar periods with regard to the various treatments being used (Table 2). The delay time from CA to call for the rescue team and the delay time until start of CPR did not differ. The overall delay from collapse to defibrillation increased in the period after the intervention. The number of

patients found in ventricular fibrillation doubled in regular wards in the period after the intervention, whereas the opposite was found in intensive care units. Overall, there were a lower proportion of patients found in ventricular fibrillation after the intervention. In wards outside intensive care, defibrillation before arrival of the rescue team took place in 3 cases before and in 10 cases after the intervention. Overall, there were no significant differences in the proportion of patients who received chest compressions before the arrival of the rescue team. We found a significant increase, though, in the proportion of patients who received chest compressions in regular wards before the arrival of the

Table 2 Treatment at resuscitation and various aspects of delay times

Calendar period P

Before

intervention ^a intervention ^b

After

n (%)

n (%)

ECG monitored: Yes 35 (48) 59 (44) .66 Initial rhythm Shockable rhythm c 25 (58) 31 (29) .001 If found in VF, defibrillation before arrival of rescue team: Yes (all patients) 21/25 (84) 20/28 (71) .34 Yes, intensive care d 18/20 (90) 10/11 (91) 1.0 Yes, regular wards e 3/5 (60) 10/17 (59) 1.0 CPR before arrival of the rescue team: Yes (all patients) 59 (81) 108 (86) .68 If yes: chest compression 54 (92) 105 (97) .13 (all patients) Chest compression 23/24 (96) 27/30 (90) .62 intensive care Chest compression 31/35 (89) 78/78 (100) .008 regular wards Ventilation (all patients) 39 (66) 59 (55) .19 Treatment Adrenaline 40 (57) 70 (57) - Other vasopressor 10 (15) 15 (13) .82 Antiarrhythmics 19 (28) 28 (24) .60 Treatment of acidosis 11 (17) 29 (25) .26 Mechanical chest 0 (0) 1 (0.9) - compression Delay between CA and: Median Median (q1, q3) (q1, q3) P

Call for rescue team 0 (0, 1) 0 (0, 1) .37 (all patients) Start of CPR (all patients) 0 (0, 1) 0 (0, 1) .63 First defibrillation f 1 (0, 2) 2 (1, 5) .005 (all patients) First defibrillation g 1 (0, 2) 1 (0, 2) 1.0 (intensive care) First defibrillation h W 5 (2.5, 6) .13 (regular wards) Arrival of rescue team 2 (2, 4) 2 (2, 5) .29 (all patients) First ECG recording 0 (0, 0) 0 (0, 1) .85 (all patients)

W indicates information only on 1 patient; VF, ventricular fibrillation. a Calendar period before intervention, 12 months. b Calendar period after intervention, 18 months. c Missing information, n = 30, before intervention and, n = 26, after intervention. d Intensive care, coronary care, or angiography laboratory. e Remaining wards. f Only patients found in ventricular fibrillation included. g Intensive care, coronary care, or angiography laboratory. h Remaining wards.

rescue team (from 89% to 100%, P = .008).

Outcome

The proportion of patients with a ROSC did not differ between the calendar periods (Tables 3 and 4). However, the likelihood of patients being alive immediately after resusci- tation was greater after the intervention (OR, 2.75; 95% CI, 1.15-6.59). Neither the proportion of patients who were alive at hospital discharge nor at 30 days after CA differed between the 2 periods.

In one third of cases, both before and after the intervention, patients had a palpable pulse on admission of

Table 3 Outcome

Calendar Period

P

Before

intervention ^a intervention ^a

After

n (%)

n (%)

Status on admission of the rescue team
Conscious	10 (14)	26 (21)	.25
Breathing	23 (33)	42 (34)	.88
Palpable pulse	23 (33)	44 (36)	.75
Initial results of CPR ROSC at any time	42 (58)	94 (72)	.06
Alive after termination of CPR	33 (45)	80 (60)	.04
Long-term results Alive at discharge from 18 (26) 39 (32) .51 hospital Alive after 30 d 18 (25) 43 (32) .27 CPC score b among survivors
1	9 (50)	37 (95)	b.0001
2	5 (28)	2 (5)
3	3 (17)	0 (0)
4	1 (6)	0 (0)
5	0 (0)	0 (0)
a Data are presented as numbers (percentage within parenthesis). b Cerebral performance scale score definition; 1, conscious and alert with normal function or only slight disability; 2, conscious and alert with moderate disability; 3, conscious with severe disability; 4, comatose or Persistent vegetative state; 5, brain dead or death from other causes [22].

Time Calendar period (2 vs 1)	0.89 a	(0.13-6.01)	2.12	(0.92-4.90)	2.75	(1.15-6.59)	1.62	(0.63-4.14)	1.84	(0.69-4.93)
Witnessed status
Witnessed vs nonwitnessed	0.97	(0.08-11.16)	2.77	(0.86-8.93)	2.14	(0.61-7.48)	2.04	(0.39-10.84)	1.66	(0.31-8.90)
Place
ICU, coronary care,	25.07	(5.41-187.18)	2.88	(1.35-6.45)	3.28	(1.53-7.05)	3.08	(1.42-6.87)	3.31	(1.49-7.61)
angiolab vs remaining wards
Initial rhythm
Shockable vs nonshockable rhythm	-	-	2.10	(0.81-4.83)	3.33	(1.46-7.59)	3.12	(1.42-6.99)	3.16	(1.41-7.24)
Delay to defibrillation is defined as the proportion of cases that were defibrillated within 3 minutes after CA. The place of CA predicted delay to defibrillation and ROSC; thus, patients were defibrillated more rapidly, and ROSC occurred more frequently, if CA took place in intensive care. Survival after termination of resuscitation was predicted by place (more frequent in intensive care), initial rhythm (more frequent in ventricular fibrillation), and period (more frequent after intervention). Survival to discharge and survival to 1 month was predicted by place and initial rhythm. ICU indicates intensive care unit. a Odds ratio (95% CI).

the rescue team due to prior successful resuscitation by the ward staff. All these patients had a confirmed CA.

Table 4 Predictors of delay to defibrillation and outcome after in-hospital CA

Independent variables Delay from ROSC at CA to defibrillation any time

OR (95% CI)

OR (95% CI)

Alive after termination of CPR

OR (95% CI)

Alive at discharge Alive at 30 d from hospital

OR (95% CI)

OR (95% CI)

Estimated cerebral function among patients discharged from hospital was markedly improved after the intervention: 78% of those who were discharged alive had a cerebral performance scale (CPC) score of 1 or 2 before the intervention compared with 100% afterwards (Table 3). The proportion of patients who were discharged with an acceptable cerebral function (CPC score 1 or 2) was 20% before the intervention and increased to 32% afterwards.

The proportion of patients who were discharged with a good cerebral function (CPC score 1) increased from 13% to 30% (among survivors, 50% had a CPC score of 1 before the intervention vs 95% afterwards) (Table 3).

The location of CA predicted the delay to defibrillation and ROSC. Thus, if CA occurred in intensive care, patients were defibrillated more rapidly, and ROSC occurred more frequently.

Survival after termination of resuscitation was predicted by location (more frequent in intensive care), initial rhythm (more frequent in ventricular fibrillation), and period (more frequent after intervention).

Survival to discharge and survival to 1 month was predicted by location and initial rhythm (Table 4).

Survival with good cerebral function in Vasteras and the rest of Sweden

Fig. 1 shows the proportion of patients who survived with a CPC score of 1 and 2 before and after the intervention in Vasteras and in all the other Swedish hospitals, which reported to the register during the 2 calendar periods. In terms of CPC score 1, there was an increase in Vasteras from 13% to 30%, whereas the corresponding figures for the rest of Sweden were 22% and 22%, respectively.

Discussion

The major findings in the present intervention study were that systematic training and education in CPR and the use of an AED for all the health care providers within 1 single hospital did not result in a shortening of the time to start of treatment, neither did it result in a significant increase in survival. However, survival with an estimated good cerebral function was more common after the intervention.

Time to first defibrillation

Priori et al [18] stated that time to defibrillation would be delayed if the patient had to wait for a rescue team. This is one of the reasons to advocate the deployment of AEDs at various different locations within hospitals. This is also a

Survival and CPC 1 o 2

%

50

32

25

25

20

69

122

513

874

40

30

20

10

0

Period 1 Period 2 Period 1 Period 2

Vasteras Remaining Sweden

Similar proportions were observed when analyzing only the cases with a CPC of 1.

Fig. 1 Proportion of patients who survived with a CPC score of 1 or 2 before and after intervention in Vasteras and the rest of Sweden.

reason to train all medical and nonmedical health care professionals to defibrillate and perform CPR.

Our study did not demonstrate any reduction in time to defibrillation. In fact, there was an increase after the intervention. There was, however, a change in the number of patients found in ventricular fibrillation between the 2 calendar periods in relation to the location of CA. Thus, there was a marked increase in the number of patients found in ventricular fibrillation in regular wards, whereas there was a marked decrease in the number of patients found in ventricular fibrillation in intensive care units. Time to defibrillation was, as expected, shorter in the intensive care units, but its impact became relatively smaller with declining numbers in the second calendar period. Our results regarding delay to defibrillation will, therefore, be difficult to interpret. Thus, the impact on delay to defibrillation by the type of intervention could not be adequately assessed.

Another approach is to follow the critical variable delay time from collapse to first defibrillation in large national registers and try to relate possible changes to the implemen- tation of AED throughout hospitals. However, it is probably more difficult to adjust for confounders in a national registry than in 1 single hospital.

Automated external defibrillators have a voice prompt and require 30 to 45 seconds to analyze the rhythm of a heart, which may delay fast defibrillation. A study by Chan et al [9] indicated that time to defibrillation increases with the introduction of in-hospital AEDs. Several studies have tried to ascertain if in-hospital AED implementation improves survival; 2 studies showed improvement [10,19], but 3 did not [9,20,21].

The finding of a much higher rate of observed ventricular fibrillations outside the intensive care unit in calendar period 2 and that many of these patients were defibrillated before the arrival of the rescue team should be related to the implementation of AEDs in these hospital areas.

Outcome

There are several factors that may inftuence the chance of survival after in-hospital CA. There was some dissimilarity between the 2 periods, which might have inftuenced the results. Thus, overall, patients with CA were less frequently found in ventricular fibrillation after the intervention, and they had less frequently an ongoing myocardial infarction at the time of CA. This might favor the cohort before intervention. On the other hand, patients with CA had more frequently a witnessed CA after the intervention, and these patients had less frequently from heart failure before the arrest. These factors might favor the cohort after the intervention. The overall survival to hospital discharge and to 1 month did not increase after the intervention, but the estimated cerebral function among survivors improved.

The mechanisms behind this observation remain specu-

lative. If the intervention played a major role, other aspects than time to start of treatment may be important. Three,

possibly linked, aspects of CPR ought to be considered: (1) theoretical knowledge, (2) practical knowledge, and (3) attitudes. All 3 might theoretically result in an improvement in the Quality of resuscitation, which might result in a better cerebral function among survivors of a CA. The association between these aspects of CPR and improved outcome has not been confirmed in the previous literature.

Strengths and limitations

Information was received from the Swedish National Register of in-hospital CA. The information from Vasteras was controlled by an external audit, in which the data from the register was compared with the medical case records. A shortcoming was that this study followed a limited number of patients. Many factors may, as previously discussed, affect survival rates after in-hospital CA, not just the training given to health care professionals and the placement of AEDs at the hospital. The CPC score could be criticized for inaccuracy. To the best of our knowledge, no studies have been performed with regard to the validity and reliability of CPC, as a measure of cerebral function after CA, when assessed retrospectively from medical case records. Furthermore, CPC score was measured at hospital discharge, whereas some patients could have improved during the subsequent weeks [22,23].

Cerebral performance scale score has, however, been the recommended method of estimating cerebral function among survivors of a CA [23].

There was a difference between calendar periods with regard to myocardial infarction/ischemia as the Underlying etiology. However, as shown in a footnote to Table 1, there were a large number of patients with missing information regarding this variable. Therefore, adjustment for this variable in the analyses became less meaningful.

Conclusion

A systematic CPR training for all health care pro- fessionals within 1 single hospital, in combination with the implementation of a certain number of AEDs outside intensive care units, did not shorten the delay time to delivery of treatment, including defibrillation. There was, however, a shift of ventricular fibrillation and defibrillation from intensive to nonintensive care. Survival to hospital discharge was not affected. Estimated cerebral function among survivors, however, did improve. The mechanisms behind this finding remain speculative, but improved quality of resuscitation might be a contributing factor.

Acknowledgments

The authors would like to acknowledge the valuable help of Jonny Lindqvist at the Swedish National Register of in-

hospital CA. This study was supported by Vastmanland County Council. The Laerdal foundation and the Swedish Association of Local Authorities and Regions support the national register of in-hospital CA.

References

1. Herlitz J, Bang A, Ekstrom L, et al. A comparison between patients suffering in-hospital and out-of-hospital cardiac arrest in terms of treatment and outcome. J Intern Med 2000;248:53-60.
2. Andreasson AC, Herlitz J, Bang A, et al. Characteristics and outcome among patients with a suspected in-hospital cardiac arrest. Resusci- tation 1998;39:23-31.
3. Herlitz J, Aune S, Bang A, et al. Very high survival among patients defibrillated at an early stage after in-hospital ventricular fibrillation on wards with and without monitoring facilities. Resuscitation 2005;66: 159-66.
4. Jastremski MS. In-hospital cardiac arrest. Ann Emerg Med 1993;22: 113-7.
5. Taffet GE, Teasdale TA, Luchi RJ. In-hospital cardiopulmonary resuscitation. JAMA 1988;260:2069-72.
6. Tortolani AJ, Risucci DA, Rosati RJ, et al. In-hospital cardiopulmo- nary resuscitation: patient, arrest and resuscitation factors associated with survival. Resuscitation 1990;20:115-28.
7. Tunstall-Pedoe H, Bailey L, Chamberlain DA, et al. Survey of 3765 cardiopulmonary resuscitations in British hospitals (the BRESUS Study): methods and overall results. BMJ 1992;304:1347-51.
8. Herlitz J, Bang A, Aune S, et al. Characteristics and outcome among patients suffering in-hospital cardiac arrest in monitored and non- monitored areas. Resuscitation 2001;48:125-35.
9. Chan PS, Krumholz HM, Spertus JA, et al. Automated external defibrillators and survival after in-hospital cardiac arrest. JAMA 2010;304:2129-36.
10. Hanefeld C, Lichte C, Mentges-Schroter I, et al. Hospital-wide first- responder automated external defibrillator programme: 1 year experience. Resuscitation 2005;66:167-70.
11. FitzGerald JD, Wenger NS, Califf RM, et al. Functional status among survivors of in-hospital cardiopulmonary resuscitation. SUPPORT Investigators Study to Understand Progress and Preferences for Outcomes and Risks of Treatment. Arch Intern Med 1997;157:72-6.
12. Herlitz J, Andreasson AC, Bang A, et al. long-term prognosis among survivors after in-hospital cardiac arrest. Resuscitation 2000;45: 167-71.
13. Elliott VJ, Rodgers DL, Brett SJ. Systematic review of quality of life and other patient-centred outcomes after Cardiac arrest survival. Resuscitation 2011.
14. Aune S, Eldh M, Engdahl J, et al. Improvement in the hospital organisation of CPR training and outcome after cardiac arrest in Sweden during a 10-year period. Resuscitation 2011.
15. Herlitz J. Nationellt register for hjartstopp, arsrapport 2010. Goteborg: National Register for Cardiac Arrest (in swedish); 2010.
16. S-HLR, HLR for sjukvardspersonal. Instruktorsbok. CPR for health- care professionales. Instruction manual, (in swedish). 1:a upplagan ed. Goteborg: Stiftelsen for HLR, Svenska radet for hjart-lungraddning; 2006.
17. Kallestedt ML, Rosenblad A, Leppert J, et al. Hospital employees’ theoretical knowledge on what to do in an in-hospital cardiac arrest. Scand J Trauma Resusc Emerg Med 2010;18:43.
18. Priori SG, Bossaert LL, Chamberlain DA, et al. Policy statement: ESC- ERC recommendations for the use of Automated external defibrillators in Europe. Resuscitation 2004;60:245-52.
19. Zafari AM, Zarter SK, Heggen V, et al. A program encouraging Early defibrillation results in improved in-hospital resuscitation efficacy. J Am Coll Cardiol 2004;44:846-52.
20. Forcina MS, Farhat AY, O’Neill WW, et al. Cardiac arrest survival after implementation of automated external defibrillator technology in the in-hospital setting. Critical Care Medicine 2009;37:1229-36.
21. Smith RJ, Hickey BB, Santamaria JD. Automated external de- fibrillators and survival after in-hospital cardiac arrest: early experience at an Australian teaching hospital. Crit Care Resusc 2009;11:261-5.
22. A randomized clinical study of cardiopulmonary-cerebral resuscita- tion: design, methods, and patient characteristics. Brain Resuscitation Clinical Trial I Study Group. Am J Emerg Med 1986;4:72-86.
23. Jacobs I, Nadkarni V, Bahr J, et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: update and simplification of the Utstein templates for resuscitation registries. A statement for healthcare professionals from a task force of the international liaison committee on resuscitation (American Heart Association, European Resuscitation Council, Australian Resuscitation Council, New Zealand Resuscitation Council, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Council of Southern Africa). Resuscitation 2004;63:233-49.