Incidence, Causes, and Survival Trends From Cardiovascular-Related Sudden Cardiac Arrest in Children and Young Adults 0 to 35 Years of AgeClinical Perspective
A 30-Year Review
Background—Sudden cardiac arrest is a leading cause of death in children and young adults. This study determined the incidence, cause, and outcomes of cardiovascular-related out-of-hospital cardiac arrest (OHCA) in individuals <35 years of age.
Methods and Results—A retrospective cohort of OHCA in children and young adults from 1980 through 2009 was identified from the King County (Washington) Division of Emergency Medical Services' Cardiac Arrest Database. Incidence was calculated from population census data and causes of arrest determined by review of autopsy reports and all available medical records. A total of 361 cases (26 cases 0–2 years of age, 30 cases 3–13 years of age, 60 cases 14–24 years of age, and 245 cases 25–35 years of age) of OHCA were treated by emergency medical services responders, for an overall incidence of 2.28 per 100 000 person-years (2.1 in those 0–2 years of age, 0.61 in those 3–13 years of age, 1.44 in those 14–24 years of age, and 4.40 in those 25–35 years of age). The most common causes of OHCA were congenital abnormalities in those 0 to 2 years of age (84.0%) and 3 to 13 years of age (21%), presumed primary arrhythmia in those 14 to 24 of age (23.5%), and coronary artery disease in those 25 to 35 years of age (42.9%). The overall survival rate was 26.9% (3.8% in those 0–2 years of age, 40.0% in those 3–13 years of age, 36.7% in those 14–24 years of age, and 27.8% in those 25–35 years of age). Survival increased throughout the study period from 13.0% in 1980 to 1989 to 40.2% in 2000 to 2009 (P<0.001).
Conclusions—The incidence of OHCA in children and young adults is higher than previously reported, and a more specific understanding of the causes should guide future prevention programs. Survival trends support contemporary resuscitation protocols for OHCA in the young.
Sudden cardiac arrest (SCA) in children and young adults is a devastating event and a leading cause of death in this population.1,2 Sudden death from cardiovascular disease is also the principal cause of death in young athletes during exercise and represents 75% of all fatalities during sport.2,3 The incidence of SCA in the young is widely debated, ranging from 0.5 to 20 per 100 000 person-years.3–11 Prior estimates have used highly variable methodology for case ascertainment, from search of public media reports to hospital reporting systems, making comparison across studies and age ranges difficult. Many studies also include all causes of SCA such as trauma, respiratory failure, drowning, and overdose.5,6,10 Limited data are available characterizing only cardiovascular causes of SCA that may be preventable through targeted screening programs.
Editorial see p 1325
Clinical Perspective on p 1372
The origin of cardiovascular-related SCA in the young is broad, involving congenital cardiac disorders, inheritable cardiomyopathies, primary electric diseases, and premature atherosclerosis. The leading cause of SCA in young athletes in the United States is reported to be hypertrophic cardiomyopathy (HCM),3,12 yet few studies have detailed the cause of SCA in the general pediatric population or compared cardiovascular abnormalities in athletes and nonathletes. SCA in the young has a low survival rate, with survival to hospital discharge <15% in many communities.5,6,10,13–17 Uncertainty about the frequency of SCA in the young, variable origin, and poor outcomes have raised several issues concerning the indication, feasibility, and cost-effectiveness of primary and secondary prevention strategies.18–23 An accurate estimate of the frequency of SCA in the young, combined with a precise determination of the specific causes, will assist the development of more effective prevention strategies. This study examines the incidence, specific cardiac causes, and outcomes of cardiovascular-related out-of-hospital cardiac arrest (OHCA) in the pediatric and young adult population in King County, Washington, over a 30-year period.
Study Setting and Population
The study area was King County, Washington, excluding Seattle, which covers 2000 square miles, has a population of 1.23 million, and includes urban, suburban, and rural areas. Approximately 50% of the population (620 000) is ≤35 years of age. The emergency medical services (EMS) system in the study community is a 2-tiered response system and has maintained an ongoing registry of each cardiac arrest treated since 1976.
The study population consisted of patients 0 to 35 years of age with treated OHCA between 1980 and 2009. All EMS-reported cases were reviewed to determine the likely cause of cardiac arrest. Only cases with confirmed or likely cardiovascular causes were included in this study to calculate incidence and survival rates. Noncardiac causes such as trauma, respiratory failure, drowning, and overdose were excluded. Cases resulting from sudden infant death syndrome were also excluded because a cardiac origin could not be confirmed.
Data Collection and Classification
The EMS medical incident reports, the electronic defibrillator recording when available, and the dispatch tape were reviewed to determine patient demographics (age and sex), event circumstances (witness status, location, citizen cardiopulmonary resuscitation [CPR] status, and arrest before EMS arrival), EMS response intervals, presenting rhythm, and immediate outcome (admission to hospital versus death). In cases when the EMS response interval could not be determined, the average EMS response time across all cases was assumed.
Origin was determined from all available information, including EMS incident reports, autopsy reports, death certificates, and hospital records when available, and a specific cause of SCA was determined for each case. Causes were categorized into subgroups of like cardiac disorders to compare survival and cause based on age of occurrence: (1) all primary electric diseases, (2) all cardiomyopathies, (3) atherosclerotic coronary artery disease (CAD), (4) congenital anomalies, (5) other cardiac causes, and (6) cardiac cause unspecified. Primary electric diseases included long-QT syndrome, Wolff-Parkinson-White syndrome, and presumed primary arrhythmia. Cases thought to be cardiac in origin on the basis of the details of the event and resuscitation but in which a specific cause could not be determined by autopsy (autopsy-negative sudden unexplained death [SUD]) were classified as a presumed primary arrhythmia. Cases of survival in which medical records identified no structural cardiac disease or other specific disorder were also classified as a presumed primary arrhythmia. The cardiomyopathic group included HCM, arrhythmogenic right ventricular cardiomyopathy, dilated cardiomyopathy, and left ventricular noncompaction. Congenital anomalies included complex heart anomalies such as tetralogy of Fallot, hypoplastic left heart, transposition of the great vessels, chromosomal alterations, and ventricular septal defects. Various other origins that could not be classified as primary electric disease, cardiomyopathy, CAD, or congenital anomaly included aortic dissection/Marfan syndrome, commotio cordis, bicuspid aortic valve/stenosis, pericarditis, coronary arteritis, and mitral valve prolapse. It is possible that some of these findings were incidental and not causal; however, no other cause could be confirmed. Cases resulting from multiple abnormalities in which a single cause could not be determined and cases in individuals who expired without autopsy were considered to be of unspecified cardiac origin.
Cases were also classified on the basis of their relationship to exercise. Exercise-related SCA was defined as cardiac arrest during or within 1 hour of physical activity. Differences in origin and survival were compared between exercise-related and non–exercise-related SCA. Survival was measured as survival to hospital discharge.
Incident reports and hospital records from cases occurring between 1986 and 1989 were not available for review. Therefore, cases from this time period were not included in the analysis of outcome by type of cardiac cause or the relation of SCA to exercise. The event and resuscitation characteristics for these cases were available in the cardiac arrest registry, so calculations related to incidence, resuscitation details, and survival rates included data from these years.
Population-based incidence rates per 100 000 person-years were calculated from census data broken into three 10-year intervals (1980–1989, 1990–1999, and 2000–2009). Population data were determined at the midway point in each interval, and this number was used as the average population over the 10-year time period to determine incidence rates across each interval. Overall and age-specific incidence rates across the 30-year study were calculated by finding the mean of the three 10-year periods, and 95% confidence intervals (CIs) were calculated with the Newcombe-Wilson method. Incidence, survival rates, and cases were analyzed according to the age ranges of 0 to 2, 3 to 13, 14 to 24, and 25 to 35 years.
SPSS 14.0 (Chicago, IL) was used for statistical analysis. Descriptive statistics are reported as means and standard deviation or medians and interquartile range. Continuous variables were analyzed with ANOVA. Categorical variables were analyzed by use of the Mantel-Haenszel χ2 test. A univariate logistic regression was used to determine the odds ratio of survival by age, sex, origin, witnessed arrest, location (public versus private), bystander CPR, arrest before EMS arrival, EMS response time, and first rhythm of ventricular fibrillation/ventricular tachycardia. A multivariate logistic regression was used to model survival and potential predictors of outcome.
An analysis of survival from 2000 to 2004 versus 2005 to 2009 was completed to evaluate the effect of the CPR protocol change that occurred in January 2005.24 A multivariate logistic regression was used to model survival and potential predictors of outcome to determine the effect of the protocol change on survival.
Overall, a total of 361 cases of cardiovascular-related OHCA in individuals 0 to 35 years of age across the 30-year interval were included in the study (Figure 1). EMS incident reports and hospital records were not available for 47 cases between 1986 and 1989. Arrest and resuscitation details were available for all 361 cases and used to calculate incidence and event characterization.
Incidence and Event Characteristics
Table 1 shows subject and event characteristics. Cases were divided into 3 time intervals to compare event characteristics and survival across the 30-year time period. The average annual incidence rate of OHCA for the overall study population was 2.28 (95% CI, 2.06–2.52) per 100 000 persons. Age-stratified incidence rates were 2.1 (95% CI, 1.44–3.06) in those 0 to 2 years of age, 0.61 (95% CI, 0.43–0.88) in those 3 to 13 years of age, 1.44 (95% CI, 1.12–1.85) in those 14 to 24 years of age, and 4.40 (95% CI, 3.90–4.99) in those 25 to 35 years of age.
The mean±SD age of OHCA victims was 25.1±10.1 years and was not statistically different across the period. The majority of arrests occurred in male patients (70.6%), occurred before EMS arrival (92.2%), and were witnessed (66.7%). The mean EMS response time (from 9-1-1 call to arrival on scene) was 4.9±2.5 minutes after the call was received.
The overall survival rate to hospital discharge was 26.9% (97 of 361) for the 30-year period, with a significant increase from 1980 to 1989 (13.0%) to 2000 to 2009 (40.2%; P<0.001; Table 2). In a univariate logistic regression analysis, the main factors associated with survival were witnessed arrest (odds ratio [OR],13.42; 95% CI, 5.48–32.88), initial rhythm of ventricular fibrillation/ventricular tachycardia (OR, 9.37; 95% CI, 4.91–17.88), bystander CPR (OR, 3.15; 95% CI, 1.82–5.44), and public location of arrest (OR, 1.97; 95% CI, 1.2–3.24). In a multivariate logistic regression analysis, independent predictors of survival were witnessed arrest (adjusted OR, 6.29; 95% CI, 4.68–30.81), first rhythm of ventricular fibrillation/ventricular tachycardia (OR, 6.50; 95% CI, 3.09–13.66), and bystander CPR (OR, 2.28; 95% CI 1.18–4.41).
When the analysis was based on origin of arrest subgroups, there was no difference in survival over the 30-year time period among primary electric and cardiomyopathic causes, although both groups had higher survival rates compared with CAD and congenital abnormalities. In the 302 cases in which the relation of OHCA to exercise could be determined, a total of 77 cases (25%) occurred during or within 1 hour of exercise with a 37.7% survival, although exercise was not statistically associated with survival.
An analysis of survival rates between 2000 to 2004 and 2005 to 2009 was also conducted, given the CPR protocol change in January 2005 that minimized interruptions in chest compressions. The overall survival rate in the first 5-year period was 25.4% compared with 58.2% in the second 5-year period (P<0.01). Univariate and multivariate logistic regression analysis showed no significant difference in EMS response time, witnessed arrest, bystander CPR, first rhythm, or location of arrest between the two 5-year periods.
Detailed Causes of Arrest
The most specific cause of cardiac arrest was determined for each case after review of all available information, including autopsy reports, death certificates, EMS incident reports, and hospital or emergency room records. Records were available for 314 of the 361 cases (records from 1986–1989 could not be located to conduct this review). Of the 314 cases, there were 221 deaths. The cause of death in those who died was determined by review of the EMS Medical Incident Report Form and death certificate in all cases and the autopsy report if performed (161 of 221). In cases of survival, cause was determined by review of the EMS Medical Incident Report Form and emergency room and hospital records. Causes were placed into subgroups to compare like cardiac origins, including (1) primary electric diseases, (2) cardiomyopathies, (3) CAD, (4) congenital anomalies, (5) other cardiac causes, and (6) cardiac origin unspecified (Table 3).
The cause of arrest subgroups were examined on the basis of survival, sex, age, and association with exercise (Table 4). Primary electric disease, cardiomyopathy, and CAD all occurred most frequently in the 25- to 35-year-old group. Congenital abnormalities were the predominant cause of arrest in those 0 to 2 years of age. CAD was found only in the oldest age range (25–35 years of age), and 89% occurred in males. The highest survival rates were in the primary electric (51.4%) and cardiomyopathic subgroups (41.7%), and the lowest survival was in the congenital anomalies (6.4%).
Figure 2 shows the specific causes that were most prevalent in each age group. The main cause of SCA in those 0 to 2 years of age (84.0%) and those 3 to 13 years of age (21.0%) was congenital anomalies. The leading cause of OHCA in those 14 to 24 years of age was presumed primary arrhythmia (23.5%). All cases of CAD occurred in the 25- to 35-year-old population and accounted for nearly half (42.9%) of the cardiac events in this age group.
A comparison of the specific origins for exercise-related versus non–exercise-related OHCA in those 14 to 35 years of age showed no significant differences (Figure 3). The leading cause of cardiac arrest in both the exercise- and non–exercise-related events was CAD at 39% and 33%, respectively. HCM represented <4% of cases in persons 14 to 35 years of age in both groups.
The incidence of OHCA in children and young adults is widely debated and largely unknown. In the United States, there is no mandatory reporting system for juvenile sudden death, making accurate estimation of the magnitude of this problem difficult. This study provides a unique 30-year perspective in which all cases of OHCA in King County, Washington, with EMS response were entered into a single cardiac arrest database.
This study provides a unique estimate of the risk of OHCA for persons 0 to 35 years of age. Other studies of pediatric OHCA have included all causes of nontraumatic cardiac arrest such as drowning, respiratory failure, and drug poisoning, making estimates of cardiovascular-related SCA challenging and thus making it difficult to draw conclusions that affect primary prevention.10 Many cardiac screening programs, especially those in young athletes, have focused on adolescents and young adults. Our study identified the risk of OHCA to be 1:69 000 persons per year for those 14 to 24 years of age and 1:23 000 persons per year in those 25 to 35 years of age. These results are comparable to a 3-year prospective population-based study in Oregon that used similar methodology and reported an incidence of 1:59 000 for OHCA in children 10 to 14 years of age.25 Eckart et al11 also reported a high incidence of cardiovascular-related sudden death in military personnel 18 to 35 of age in whom the incidence of sudden cardiac death (SCD) was ≈1 in 25 000.
The incidence of OHCA in adolescents and young adults in this study is substantially higher than the initial estimates of SCD in young competitive athletes, ranging from 1:160 000 to 1:300 000 deaths per year in athletes 12 to 35 years of age.3,8,9 However, these studies underestimate the incidence of SCD in athletes owing to the lack of a systematic reporting system, incomplete identification of all cases, and heavy reliance of case ascertainment through search of public media reports, catastrophic insurance claims, and other electronic databases. In contrast, a recent study of the incidence of SCD in National Collegiate Athletic Association (NCAA) athletes demonstrated that the incidence of SCD in college-aged athletes is 4 to 5 times higher than traditional estimates.2 The overall risk of SCD in NCAA athletes was 1:43 000, with higher risks found in male athletes (1:33 000), black male athletes (1:13 000), and male basketball players (1:7000).2 The NCAA study used an internal reporting system and demonstrated that reliance on media reports, even in this high-profile population, missed 44% of cases.2
Questions exist about the relative risk of SCA in competitive athletes versus the general population and whether this risk justifies a separate, more advanced cardiovascular screening program in athletes. It is generally accepted that exercise and intense physical exertion through athletic participation increase the likelihood of sudden death for many disorders predisposing to SCA. An Italian study identified a 2.5-times relative risk for SCD in adolescents and young adults engaged in competitive sports compared with an age-matched nonathletic population.26 Marijon et al27 also reported that the relative risk of sports-related sudden death was 4.5 times higher in competitive young athletes 10 to 35 years of age compared with noncompetitive sports participants of the same age in France. Although this study could not determine a rate of OHCA in young athletes, it is unlikely to be significantly less than what was found for the overall adolescent and young adult population. In addition, the high rate of such events determined in this study may justify the development of more aggressive screening programs for the general adolescent and young adult population, not just competitive athletes.
Accurate characterization of the cause of pediatric OHCA allows screening programs to target the causes posing the greatest risk. This study details the various cardiac disorders leading to OHCA in the young stratified by age. A comprehensive evaluation of each case was performed by review of all available medical records to determine the circumstances and specific cause of death. This is necessary because a previous report found that determination of SCA from death certificates alone was inaccurate.2
Several studies have reported HCM as the leading cause of SCD in young competitive athletes 12 to 35 years of age in the United States, representing up to 36% of cases.3,12 In this study, HCM represented only 18% of cases for those 3 to 13 years of age, with just 1 case of HCM occurring in a child <10 years old. In addition, this study found that <3% of OHCA in the general population 14 to 35 years of age was due to HCM.
This study found a higher proportion of autopsy-negative SUD in the general population than prior reports of sudden death in same-aged competitive athletes. Autopsy-negative SUD is reported in only 3% of sudden death cases in young competitive athletes 12 to 35 years of age in the United States,3 whereas it accounted for 10% of OHCA in persons 14 to 35 years of age in this study. Autopsy-negative SUD also represents a substantially larger proportion of SCD in the young in other study populations. Autopsy-negative SUD represents ≈30% of SCD in Australia28,29 and 36% of SCD in Denmark30 in individuals <35 years old. In US military personnel 18 to 35 years of age, autopsy-negative SUD accounts for 41% of cardiovascular-related sudden death.11 Autopsy-negative SUD may be due to inherited arrhythmia syndromes and ion channel disorders such as long-QT syndrome, short-QT syndrome, Brugada syndrome, and familial catecholaminergic polymorphic ventricular tachycardia or other primary electric diseases such as Wolff-Parkinson-White syndrome.12 The accurate diagnosis of ion channelopathies postmortem is still limited; however, postmortem genetic testing (so-called molecular autopsy), if performed, identifies a pathogenic cardiac ion channel mutation in more than one third of cases.13,31,32 A better understanding of the cause of arrest in survivors with a negative evaluation after SCA is also needed.
We analyzed those cases of OHCA that occurred during or within 1 hour of exercise to determine whether causation was different in exercise-related events. Although this group is not a direct surrogate for competitive athletes, there were no significant differences in causes of arrest between exercise- and non–exercise-related events. HCM represented only 4% of cases of exercise-related OHCA in this study compared with one third of cases reported in competitive athletes.3 Presumed primary arrhythmia was the leading cause of exercise-related OHCA (23%) in adolescents and young adults 14 to 25 years of age found in this study. More research is needed to understand these differences and to determine whether they are real or influenced by methodology, ascertainment bias, or study demographics.
The reason for the difference in origins of SCA found in this study compared with prior reports in young competitive athletes is uncertain. Disparities may be due to differences in methodology and case identification, differences in the demographics and ethnicity of young competitive athletes compared with the general population, or perhaps variability in the likelihood of SCA with vigorous exercise depending on the specific cause. More research is needed to understand the specific causes of SCA in the young general population and in competitive athletes to guide the development of appropriate screening procedures.
As expected, CAD becomes the main risk factor for SCA in young adults 25 to 35 years of age. In children 0 to 13 years of age, congenital anomalies were the most common cause of OHCA, suggesting that enhanced postnatal diagnosis and management of congenital cardiac disorders may provide the largest impact on the prevention of young pediatric SCD.
This study found an important increase in survival from 13% in the 1980s to 40% in the past decade. Outcomes from pediatric OHCA are generally poor as reported in past studies.5,6,10,13–17 This study demonstrates that improved survival rates are possible with the use of a robust community-based EMS system and modern resuscitation protocols. It also suggests that the CPR protocol change on January 1, 2005, was associated with a significant increase in survival The 2005 guidelines called for a single shock without rhythm reanalysis, stacked shocks, or postshock pulse check, and they extended the period of CPR between shocks to 2 minutes.24,33 A 2006 report suggested the new resuscitation guidelines may improve outcomes for adults with out-of-hospital ventricular fibrillation arrest.33 To the best of our knowledge, this is the first study to compare outcomes of pediatric OHCA before and after implementation of the new guidelines. In this study, survival in the 5 years before this protocol change was 25.4%, which is consistent with a survival of 25.4% in the 1990s. In the 5 years after the protocol change and after controlling for the covariates that affect outcomes, survival increased dramatically to 58.2%.
This study is limited by its retrospective design and lack of a mandatory reporting system for OHCA. It is possible that cases of OHCA were not captured by the EMS system and that the actual incidence of OHCA in this population could be higher. However, we believe this to be unlikely because all EMS run reports flow to the King County EMS, and thus all EMS-treated cardiac arrests are captured. To further verify the capture of all cardiac arrests, redundant sources of information such as dispatch codes are also screened. Census data can also be imprecise but are the only means of calculating incidence in population-based studies. In addition, although a comprehensive review of all available medical reports was performed, inherent limitations in current autopsy protocols and coroner diagnoses and the lack of postmortem genetic testing in cases of a negative autopsy complicate the determination of a specific disorder as the cause of cardiac arrest.
This study indicates that cardiovascular-related OHCA in children and young adults is more common than reported in prior estimates. The favorable survival trends identified over the last 3 decades provide compelling support of contemporary resuscitation protocols for OHCA in the young. A detailed characterization of the causes affecting different age groups should guide the development of targeted screening programs and more effective strategies for prevention.
Sources of Funding
This work was supported by the University of Washington School of Medicine, 2010 Medical Student Research Training Program.
We wish to acknowledge and thank the team at the King County EMS office for their assistance in the record and data collection for this study.
- Received October 28, 2011.
- Accepted July 6, 2012.
- © 2012 American Heart Association, Inc.
- Harmon KG,
- Asif IM,
- Klossner D,
- Drezner JA
- Maron BJ,
- Doerer JJ,
- Haas TS,
- Tierney DM,
- Mueller FO
- Atkins DL,
- Everson-Stewart S,
- Sears GK,
- Daya M,
- Osmond MH,
- Warden CR,
- Berg RA
- Eckart RE,
- Shry EA,
- Burke AP,
- McNear JA,
- Appel DA,
- Castillo-Rojas LM,
- Avedissian L,
- Pearse LA,
- Potter RN,
- Tremaine L,
- Gentlesk PJ,
- Huffer L,
- Reich SS,
- Stevenson WG
- Young KD,
- Gausche-Hill M,
- McClung CD,
- Lewis RJ
- Drezner JA,
- Courson RW,
- Roberts WO,
- Mosesso VN Jr.,
- Link MS,
- Maron BJ
- Drezner JA,
- Rao AL,
- Heistand J,
- Bloomingdale MK,
- Harmon KG
- Maron BJ,
- Thompson PD,
- Ackerman MJ,
- Balady G,
- Berger S,
- Cohen D,
- Dimeff R,
- Douglas PS,
- Glover DW,
- Hutter AM,
- Krauss MD,
- Maron MS,
- Mitten MJ,
- Roberts WO,
- Puffer JC
ECC Committee, Subcommittees and Task Forces of the American Heart Association. 2005 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation. 2005;112(suppl):IV-1–IV-203.
- Marijon E,
- Tafflet M,
- Celermajer DS,
- Dumas F,
- Perier MC,
- Mustafic H,
- Toussaint JF,
- Desnos M,
- Rieu M,
- Benameur N,
- Le Heuzey JY,
- Empana JP,
- Jouven X
- Rea TD,
- Helbock M,
- Perry S,
- Garcia M,
- Becker L,
- Eisenberg M
Limited epidemiological data are available on cardiovascular-related sudden cardiac arrest (SCA) in children and young adults. This study conducted a 30-year review of SCA cases in persons 0 to 35 years of age treated by emergency medical services in King County, Washington. The study findings indicate that SCA in children and young adults is more common than reported in prior estimates. The study also provides a detailed characterization of the causes affecting different age groups. The most common causes of SCA were congenital abnormalities in individuals 0 to 13 years of age, presumed primary arrhythmia in those 14 to 24 years of age, and coronary artery disease in individuals 25 to 35 years of age. Survival data were also presented, and a favorable survival trend over the last 3 decades lends additional support to contemporary resuscitation protocols for SCA in the young. Although more epidemiological research is needed, these findings may influence the development of more effective screening and targeted strategies for prevention of SCA in children and young adults.