Author + information
- Received October 28, 1996
- Revision received April 7, 1997
- Accepted August 14, 1997
- Published online November 15, 1997.
- Jacqueline J.M de Vreede-Swagemakers, PhDA,*,
- Anton P.M Gorgels, MDA,
- Willy I Dubois-Arbouw, MScA,
- Jan W van Ree, MDB,
- Mat J.A.P Daemen, MDC,
- Leon G.E Houben, RND and
- Hein J.J Wellens, MD, FACCA
- ↵*Dr. Jacqueline J. M. de Vreede-Swagemakers, Department of Cardiology, University Hospital Maastricht, P.O. Box 5800, 6202 AZ Maastricht, The Netherlands.
Objectives. We sought to describe the incidence, characteristics and survival of out-of-hospital sudden cardiac arrest (SCA) in the Maastricht area of The Netherlands.
Background. Incidence and survival rates of out-of-hospital SCA in different communities are often based on the number of victims resuscitated by the emergency medical services. Our population-based study in the Maastricht area allows information on all victims of witnessed and unwitnessed SCA occurring outside the hospital.
Methods. Incidence, patient characteristics and survival rates were determined by prospectively collecting information on all cases of SCA occurring in the age group 20 to 75 years between January 1, 1991 and December 31, 1994. Survival rates were related to the site of the event (at home vs. outside the home) and the presence or absence of a witness and rhythm at the time of the resuscitation attempt in out-of-hospital SCA.
Results. Five hundred fifteen patients were included (72% men, 28% women). In 44% of men and 53% of women, SCA was most likely the first manifestation of heart disease. In patients known to have had a previous myocardial infarction (MI), the mean interval between the MI and SCA was 6.5 years, with >50% having a left ventricular ejection fraction >30%. The mean yearly incidence of SCA was 1 in 1,000 inhabitants. Of all deaths in the age groups studied, 18.5% were sudden. Nearly 80% of SCAs occurred at home. In 60% of all cases of SCA a witness was present. Cardiac resuscitation, which was attempted in 51% of all subjects, resulted overall in 32 (6%) of 515 patients being discharged alive from the hospital. Survival rates for witnessed SCA were 8% (16 of 208 subjects) at home and 18% (15 of 85 subjects) outside the home (95% confidence interval 1% to 18.8%).
Conclusions. The majority of victims of SCA cannot be identified before the event. Sudden cardiac arrest usually occurs at home, and the survival of those with a witnessed SCA at home was low compared with that outside the home, indicating the necessity of optimizing out-of-hospital resuscitation, especially in the at-home situation.
In industrialized countries many people die of sudden cardiac arrest (SCA), with coronary heart disease as the most common cause . To develop and evaluate preventive strategies, such as selecting target groups for cardiopulmonary resuscitation (CPR) training, it is important to know the yearly incidence of SCA in the community, the circumstances surrounding SCA, the patient’s profile and the factors determining survival. The incidence of SCA ranges from 36 to 128 per 100,000 inhabitants per year in different communities [2–5]. However, in these studies only victims with witnessed cardiac arrest seen or resuscitated by emergency medical services (EMS) were included. Furthermore, as pointed out by Siscovick and Becker et al. , true incidence information has been a neglected factor in evaluating survival rates.
In our study region there is only one hospital, one EMS and a network of cooperative general practitioners (GPs), which makes the region suitable for population-based studies. In this study we report on 1) the yearly incidence during a 4-year period of unexpected witnessed and unwitnessed cases of SCA in the Maastricht area of the Netherlands; 2) the medical characteristics of the victims; and 3) the relation between survival and site of SCA.
1.1 Study Group
During a 4-year period (January 1, 1991 to December 31, 1994), all victims of unexpected out-of-hospital SCA between 20 and 75 years of age and living in the Maastricht region of the Netherlands were registered. The area encloses 203 km2and has ∼181,500 inhabitants, ∼133,000 (73%) of whom are between 20 and 75 years of age. In this area there is only one EMS service, which has seven ambulances. All seven ambulances are equipped with defibrillators. The EMS can be contacted 24 h a day by calling 112. The crew of each ambulance consists of one nurse and one driver.
1.2 Inclusion and Exclusion Criteria
Included in this study were all witnessed and unwitnessed victims of SCA living in the study region. Unwitnessed cases were included when circumstances were pointing to an unexpected SCA (for example, those who died unexpectedly during sleep). Excluded were patients with a circulatory arrest after a traumatic event or intoxication or SCA occurring in the terminal phase of a chronic disease. Patients with severe heart failure who were in the terminal phase of the disease and totally confined to their home were not included.
The age limit of 75 years was chosen because inhabitants over this age often live alone, limiting the possibility to obtain information on circumstances of death and complaints preceding the event.
1.3 Data Collection
All victims who were found dead or who had no ambulance present at the time of death, were reported by the GPs in the region. All victims who had an ambulance present were reported by the ambulance personnel. The ambulance service was contacted daily. All 84 GPs located in the study region were phoned weekly.
For all subjects information was collected on age, gender, circumstances and whether and by whom the SCA was witnessed. This information was obtained from the ambulance personnel, the GP or family members or witnesses who were interviewed later, or by a combination of these. Information on the performance of CPR and by whom, the cardiac rhythm at the moment of arrival of the ambulance, the estimated interval between the moment of collapse and the start of resuscitation and the ambulance delay time (time between the moment of the emergency call and the moment of arrival) was obtained from a questionnaire that was filled out immediately after the event by the ambulance personnel.
Information on the medical history of victims was gathered by collecting information from the GPs and by examining the records of the only hospital in Maastricht. Data on the overall mortality in our study group were obtained from the Central Statistical Office. These data will be presented according to gender and age (range 25 to 75 years).
1.4.1 Incidence of SCA
The yearly incidence of SCA was assessed for 1991, 1992, 1993 and 1994 separately by dividing the total number of cases registered each year by the total number of inhabitants in the same age range standardized for 10,000 inhabitants.
Data on the total number of residents in the region (calculated on January 1 of each year) and on the gender and age distribution of the population in the region were obtained from the Central Statistical Office . Also determined was the mean yearly incidence of SCA over the total study period for inhabitants with and without known cardiac disease. Incidence rates were calculated similarly as described earlier. The prevalence of cardiac disease in the community was estimated by using information from the regional registration network of GPs . This network consists of 42 participating GPs from 15 practices in the district of Limburg. The patient group of the network resembles the general population of the Netherlands with respect to age, gender, marital status, type of household, insurance status and level of education . The data base contains information on the medical history of a representative group of 12,061 inhabitants of the Maastricht area in Limburg. The age- and gender-adjusted prevalence of cardiac disease in this representative sample was extrapolated to the study group. The presence of cardiac disease was defined as known with angina pectoris, myocardial infarction (MI), heart failure, valvular disease, arrhythmias or other heart diseases.
1.4.2 Sudden Cardiac Arrest
Sudden cardiac arrest was defined as an unexpected, nontraumatic loss of vital signs, such as consciousness, arterial pulse, blood pressure and respiration, without preceding complaints or within 24 h of the onset of complaints. We used the 24-h definition to include also those victims who were found dead but seen alive within 24 h of the event.
Witnessed SCAsare arrests occurring in the presence of a bystander or emergency medical personnel. Unwitnessed SCAwas defined as a SCA that occurred when a person was alone at the moment of the event and who was found unconscious or dead by a family member, neighbor, friend, etc.
The “Utstein Style” definitions were used for basic CPR, advanced CPR and bystander CPR. The Utstein recommendations for reporting on the outcome of out-of-hospital SCA focus on arrests in which the EMS was involved. However, as Siscovick mentions, when limiting data collection to information available through EMS data bases, comparison of outcomes across communities or over time will be biased. Therefore, we included all witnessed and unwitnessed SCAs seen or not seen by the EMS.
Survival ratewas defined as the number of victims of SCA who were discharged alive from the hospital, divided by the total number of SCAs (witnessed and unwitnessed) registered. Resuscitation attempt ratewas defined as the number of resuscitation attempts, divided by the total number of victims of SCA. Resuscitation success ratewas defined as the total number of hospital survivors, divided by the total number of resuscitation attempts.
1.5 Statistical Methods
All data were entered into the SPSS statistical program. Statistical significance for differences was determined by the Mann-Whitney Utest for continuous variables. Furthermore, 95% confidence intervals (CIs) for proportions and their differences and relative risk were calculated.
From January 1, 1991 until December 31, 1994, a total of 515 subjects were included. Of these, 369 were men (72%) and 146 were women (28%) (mean age 62.2 ± 8.9 and 62.7 ± 12.2 years, respectively). Information on the victims’ medical history was obtained from the GPs or from hospital records, or both, in all 515 subjects. A history of cardiac disease was present in 277 subjects (53.8%). Interestingly, in 53% of women and 44% of men, SCA was most likely the first manifestation of heart disease. A review of the GP and hospital records of all 277 subjects known to have had a previous cardiac history showed that 177 victims (64%) were known to have had one or more previous MIs. The interval between the last MI and SCA was >2 years in 66% of the victims (mean 6.5 years, median 5 years). In 131 of 177 victims known to have had a previous MI, a left ventricular ejection fraction (LVEF) was determined (after MI) by two-dimensional echocardiography.
The mean (±SD) interval between the last echocardiogram and SCA was 18.8 ± 21.2 months. In 56.5% of the victims, a LVEF >30% was present and in 20% of the victims a LVEF >50% was present.
2.1 Yearly Incidence of SCA
The mean yearly incidence of SCA in the 20- to 75-year age group was 129 SCAs per 132,762 mean population of the Maastricht region or 9.7 per 10,000 inhabitants. The yearly incidence of SCA was 9.8 per 10,000 in 1991; 9.9 per 10,000 in 1992; 10.1 per 10,000 in 1993; and 9.0 per 10,000 in 1994.
The prevalence of heart disease in the GPs’ network representative sample was 9.15%. When extrapolating this figure to our study group, the prevalence rate of heart disease was 12,148/132,762 inhabitants. So the mean yearly incidence rate of SCA in inhabitants with a known cardiac disease was 53.8% of 129 (69.4 per 12,148 or 57 per 10,000). A total of 120,615 inhabitants never had cardiac complaints or never sought medical help for possible cardiac complaints. The mean yearly incidence of SCA in this group was 46.2% of 129 (59.6 per 120,615 or 5 per 10,000). Therefore, the relative risk of SCA for persons with cardiac disease was ∼11 (95% CI 8 to 16).
In the age group 50 to 59 years, men with a previous cardiac history had a three times higher mean yearly incidence of SCA compared with women (60 per 10,000 vs. 18 per 10,000). In men without a previous cardiac history between the age of 50 and 59 years, the mean yearly incidence of SCA was 11 per 10,000 compared with 2 per 10,000 in women of the same age group.
2.2 Contribution of Sudden Death To Total Mortality
During the period January 1991 until January 1994, a total of 2,030 inhabitants between 25 and 75 years of age died. In 375 inhabitants (18.5%), the mode of death was sudden. As shown in Table 1this was 27% in men in the age group 55 to 64 years. In women, the highest percentage of SCA (16%) was found in the age categories 25 to 44 and 65 to 74 years.
2.3 Site at Time of SCA
In 501 of 515 patients information on the location of the event was available. Three hundred ninety-nine (79.6%) events occurred at home and 102 (20.4%) outside the home (Table 2).
2.4 Number of Witnessed and Unwitnessed SCAs
It was not clear whether 30 victims had witnessed or unwitnessed events, and so these patients were excluded. Of 485 victims, 293 (60.4%) had witnessed events, 208 (71%) had events at home and 85 (29%) had events outside the home. Of those with SCA at home, the witness was a family member in 183 cases (88%). In 15 cases (7%) a friend and in 10 cases (5%) the GP was present. Of the 85 witnessed SCAs outside the home, 57 (67%) were witnessed by a bystander, 23 (27%) by a family member and 5 (6%) by a medical doctor or nurse. One hundred ninety-two subjects (39.6%) had unwitnessed SCAs, with 176 (91.7%) occurring at home.
2.5 Complaints Before SCA
In 227 of the 293 patients with a witnessed SCA, the victims had spoken to the witness before the event. The complaints of the victims are given in Table 3. As indicated in Table 3the majority of victims with witnessed events had complaints before the SCA. Complaints suggesting an ischemic cause of SCA were present in 43% of the victims, and signs of a fast downhill course of congestive heart failure were present in 7%. Absence of complaints or complaints of dizziness and palpitations shortly before SCA suggest a primary arrhythmic mechanism in 29% of the victims. Finally, there is a group of victims who indicated they did not feel well or that they felt nauseated. Some victims may not have communicated their complaints to the witness.
Fig. 1is a flow diagram of the 515 cases of SCA outside the hospital, indicating the site of SCA, presence of a witness, incidence of a resuscitation attempt and by whom and the number of patients discharged alive from the hospital.
2.6 Resuscitation Attempt, Success and Survival Rates
In 16 of 485 cases of SCA it was not known whether resuscitation was attempted. Resuscitation was done by the ambulance personnel or a bystander, or both, in 237 subjects (50.5%). Of these, 32 (13.5%) were discharged alive from the hospital, giving an overall survival rate of out-of-hospital SCA of 6.2% (32 of 515 subjects).
2.6.2 Witnessed SCA
Of 11 victims with witnessed SCA (9 at home, 2 outside the home), data on resuscitation were missing. In 210 (74.5%) of 282 witnessed cases resuscitation was attempted. Of these, 31 (14.8%) were discharged alive from the hospital. The survival rate of witnessed out-of-hospital SCA was therefore 10.6% (31 of 293 subjects).
126.96.36.199 Witnessed SCA at Home Versus Outside the Home
In 133 (66.8%) of 199 subjects with witnessed events resuscitation was attempted at home and outside the home in 77 (92.7%) of 83 subjects (95% CI 17.4% to 34.5%). Between these two groups, there were no significant differences in terms of mean age and gender distribution. Sixteen (12%) of 133 subjects with a resuscitation attempt at home and 15 (19.5%) of 77 resuscitated outside the home were discharged alive from the hospital (95% CI −2.9% to 17.9%). Therefore, survival rates of witnessed SCA were 7.7% (16 of 208 subjects) and 17.6% (15 of 85 subjects) (95% CI 1% to 19.9%), respectively.
188.8.131.52 Bystander CPR
At home, only 56 (28.1%) of 199 victims with witnessed events received bystander CPR. Outside the home, this occurred in 52 (62.6%) of 83 subjects (95% CI 22.4% to 46.6%). Success rates of CPR initiated by a bystander were 10.7% (6 of 56 subjects) at home and 25.5% (13 of 51 subjects) outside the home (95% CI 0.3% to 29.3%).
184.108.40.206 Advanced CPR
Of 77 (38.7%) of 199 victims who were at home at the time of the event and who received advanced CPR only, 10 (13%) were discharged alive from the hospital. Of 25 (30.1%) of 83 victims who had their SCA outside the home with advanced CPR only, 2 (8%) survived hospital admission (95% CIs −3.4% to 20.6% and −8.0% to 18.0%, respectively).
2.6.3 Unwitnessed SCA
In five subjects with an unwitnessed SCA at home, data on resuscitation were missing. In 27 (14.4%) of 187 victims with unwitnessed events resuscitation was attempted—at home in 21 (12.3%) of 171 subjects and outside the home in 6 (37.5%) of 16 victims (95% CI 7% to 43%). Only one victim (resuscitated at home) survived to hospital discharge. Therefore, the survival rates of unwitnessed SCA were 0.6% (1 of 176 subjects) at home and 0% (0 of 16 subjects) outside the home.
2.7 Delays and Cardiac Rhythm at Time of Resuscitation Attempt
The overall mean estimated interval between the moment of collapse and the start of resuscitation was 5.3 ± 5.2 min at home and 3.9 ± 5.4 min outside the home (p < 0.05). When CPR was started by a bystander, these values were 3 ± 3.2 min at home and 2.9 ± 2.7 min outside the home. The mean ambulance delay time was 6.4 ± 2.8 min at home and 6.5 ± 3.1 min outside the home.
In 310 (60.2%) of 515 victims an ambulance arrived at the scene. In 90 victims the ambulance personnel considered a resuscitation attempt no longer indicated. In 220 victims resuscitation was performed by the ambulance crew. In this latter group, the first documented rhythm was ventricular fibrillation (VF) in 120 subjects, ventricular tachycardia (VT) in 8, bradycardia in 30 and asystole in 62. Of 128 victims who were found in VT or VF, 29 (22.7%) and 3 (3.3%) of 92 (95% CI 11.6% to 27%) victims with bradycardia or asystole survived to hospital discharge.
Although the yearly incidence of out-of-hospital sudden coronary deaths seems to be declining , SCA continues to be a common cause of death. Our study, which is the first study on the incidence of out-of-hospital SCA performed in the Netherlands, shows that the mean yearly incidence of unexpected SCA was 1 in 1,000 persons between 20 and 75 years of age and demonstrates that its contribution to total mortality in people between 25 and 75 years of age was nearly 19%. In men in the age group 55 to 64 years, 27% of all deaths occurred suddenly.
3.1 Study Limitations
By restricting our cohort of interest to individuals <75 years, we are aware of the fact that exclusion of a segment of the population that is prone to cardiac arrest may produce selection bias. However, we adjusted our denominator (all inhabitants between 20 and 75 years of age) in concordance with our numerator (all those between 20 and 75 years of age), which is what Becker et al. stated they were unable to do, because of unavailable data, in their incidence study of 20 communities. Furthermore, we included not only victims who had emergency service, but also those victims who were found dead. This may give better insight in the overall incidence and survival of SCA. However, the use of the “24-h” definition will lead to the inclusion of unexpected SCAs from various causes such as ruptured abdominal aortic aneurysm and pulmonary embolism. A study by Pratt et al. showed that in a number of deaths that would have been classified as sudden cardiac based solely on temporal criteria, the cause of death at autopsy was noncardiac. The yearly incidence of SCA with an underlying cardiac cause is therefore lower. Based on our autopsy data of 127 SCA victims included in the study, it was estimated that ∼20% of the total number of victims had a noncardiac cause of SCA.
Like other investigators [12–15], we found that 80% of the total number of SCAs occurred at home. Only 60% of the victims had witnessed cardiac arrest. The majority of victims with witnessed events had complaints before the SCA. Complaints suggesting an ischemic cause of SCA were present in most of these victims. Two-thirds of the witnessed arrests took place at home, usually with a family member as the witness. Of the unwitnessed arrests, 9 of 10 occurred at home. Only 6% of all subjects with out-of-hospital SCAs were discharged alive from the hospital.
In our study, resuscitation was attempted in nearly 75% of subjects with witnessed events, with a higher incidence of CPR on the street and in public places than at home. In witnessed cases, the resuscitation success rate was 12% at home and nearly 20% outside the home. These differences in success rates at home versus outside the home have also been reported by Ritter et al. and Litwin et al. (6% vs. 13% and 13% vs. 27%, respectively). Most studies described improved outcome when the resuscitation attempt was started immediately by a witness before the arrival of the ambulance [16–24]. Also in our study, the mean interval between collapse and the start of the resuscitation attempt was shorter when SCA occurred outside the home compared with at home [13, 14]. The ambulance delay time was not different between witnessed cardiac arrest at home and that outside the home. To our surprise, we found that the at-home outcome of the resuscitation attempt was not different between CPR started by a bystander and that by ambulance personnel. This is compatible with the hypothesis that the resuscitation performance level is better among bystanders in the street than at home, perhaps because a far greater number of potential resuscitators are available in public places. As shown by Gallagher et al. , effective bystander CPR is independently associated with a significant improvement in survival.
As previously reported , a rapid rhythm (VT or VF) at the time of the resuscitation attempt was associated with a better prognosis than asystole or a slow rhythm.
Our study shows that the incidence of SCA is 11 times higher in people with known cardiac disease as compared with asymptomatic men and women. Although this figure was calculated on the basis of an extrapolation, it is very similar to the nine times higher risk reported by Kannel and Schatzkin . Furthermore, most of these victims had a previous MI. The mean interval between MI and SCA was 6.5 years, indicating that SCA frequently occurs late after MI. A low LVEF (<30%), which is known to be an important risk factor for SCA after MI [27–29], was present in less than half of the victims. The latter finding in victims known to have had a previous MI, together with the observation of the absence of a previous cardiac history in 53% of the women and 44% of the men, supports the idea that even in the 1990s the majority of the victims of SCA cannot be identified before the event.
To improve overall outcome of out-of-hospital SCA, thus decreasing the incidence of sudden death, emphasis should be placed on increasing both the number of resuscitation attempts and resuscitation success rate. In our study most witnessed sudden deaths occurred at home and the resuscitation attempt rate and success rate were low compared with those outside the home, indicating that emphasis should be placed on optimizing resuscitation efforts in the at-home situation. Because our study shows that people with a history of heart disease had a higher chance of dying suddenly, we confirm the statement of Dracup et al. that family members of patients with cardiac disease are an important target group for training in CPR. Furthermore, we should also search for better telecommunication methods by developing small, portable devices that transmit the occurrence of cardiac arrest to people who can help with basic and advanced cardiac life support.
To evaluate the effect of these measures and the value of primary and secondary preventions on the problem of SCA in the community, it is essential to continue to record all witnessed and unwitnessed SCAs in the total population with and without a cardiac history. Furthermore, for comparison across communities it is important that studies on the incidence and overall outcome of SCA include all victims of SCA and not only those who had the EMS involved.
We gratefully acknowledge all general practitioners in the Maastricht area, the personnel from the Maastricht ambulance service, family members of cardiac arrest victims and members from the departments of pathology, general practice and cardiology of the University of Maastricht for their help in collecting the data for this study.
☆ This study was supported by the Wijnand M. Pon Foundation, Leusden and the Research in Cardiology Foundation, Maastricht, The Netherlands.
☆☆ To discuss this article on-line, visit the ACC Home Page at www.acc.org/membersand click on the JACC Forum
- confidence interval
- cardiopulmonary resuscitation
- emergency medical service
- general practitioner
- left ventricular ejection fraction
- myocardial infarction
- sudden cardiac arrest
- ventricular fibrillation
- ventricular tachycardia
- Received October 28, 1996.
- Revision received April 7, 1997.
- Accepted August 14, 1997.
- The American College of Cardiology
- Vertesi L
- ↵Composition of the population of the southern part of Limburg, The Netherlands, 1991–1995. Voorburg, The Netherlands: Central Statistical Office.
- Metsemakers JFM,
- Hoppener P,
- Knottnerus JA,
- Kocken RJJ,
- Limonard CBG
- Task force of representatives from the European Resuscitation Council, American Heart Association, Heart and Stroke Foundation of Canada, and Australian Resuscitation Council
- Goldberg RJ
- Pratt CM,
- Greenway PS,
- Schoenfeld MH,
- Hibben ML,
- Reiffel JA
- Fitzpatrick B,
- Watt GCM,
- Tunstall-Pedoe H
- Copley DP,
- Mantle JA,
- Rogers WJ,
- Russel RO,
- Rackley CE
- Kannel WB,
- Schatzkin A
- the Multicenter Postinfarction Research Group,
- Bigger JT,
- Fleiss JL,
- Kleiger R,
- Miller JP,
- Rolnitzky LM