Does sports activity enhance the risk of sudden death in adolescents and young adults?

Population

The Veneto Region of Northeast Italy covers an area of 18,368 km². From 1979 to 1999, the population was stable and averaged 4,379,900, according to the Italian Census Bureau. The vast majority of residents were white, comprising an ethnically homogeneous population.

During the study period, the entire population age 12 to 35 years averaged 1,386,600, consisting of 692,100 males and 694,500 females. This population for the 21 years of observation provided 29,118,600 age-specific person-years, consisting of 14,534,100 males and 14,584,500 females, respectively. According to the Sports Medicine Data Base of the Veneto Region of Italy, the athletic population averaged 112,790 (90,690 males and 22,100 females), providing 2,368,590 athlete-years of observation (1,904,490 males and 464,100 females).

Morphologic protocol

The protocol of investigation has been reported in detail elsewhere (2,6). In brief, macroscopic examination included measurement of heart weight and wall thickness, inspection of the coronary arteries and valves, and identification of any myocardial scar or dilation. The origin and course of the coronary arteries were examined, and the patency of the major epicardial coronary trunks was analyzed by taking transverse section at 3-mm intervals. Coronary artery segments and several transmural blocks of ordinary myocardium from the right and left ventricular walls and from the interventricular septum were processed for histologic examination; 7-μm-thick sections were stained with the hematoxylin-eosin, Weigert-van Gieson, and trichrome Heidenhain (azan) techniques. The specialized conduction system was studied by a serial section technique, as previously reported (7). Arrhythmogenic right ventricular cardiomyopathy (ARVC) was diagnosed in the presence of gross and/or histologic evidence of regional or diffuse transmural fibrofatty replacement of the right ventricular free wall and in the absence of other known cardiac or non-cardiac causes of death (8).

Statistical analysis

Continuous data are expressed as the mean ± standard deviation. The chi-square or Fisher exact test was used to assess the significance of differences between subgroups. The relative risk (RR) of SD (the ratio of the risk of SD among competitive athletes to the risk among non-athletes) and the corresponding 95% confidence interval (CI) were calculated with the Stata version 5.0 (1997) statistical package (Stata, College Station, Texas). Poisson multivariate regression was used to simultaneously test sports exposure, gender, and the interaction between the two. A two-tailed p value <0.05 was considered statistically significant.

Incidence of SD

In the present study, the incidence of SD by all causes was 2.3 (2.6 in males and 1.1 in females) in 100,000 athletes per year, and that of SD from cardiovascular diseases was 2.1 in 100,000 athletes per year. In previous studies, the assessment of the precise frequency with which SD occurs in young athletes during organized competitive sports was encumbered by a number of limitations, mostly related to retrospective analysis. Such studies probably resulted in an underestimation of the true prevalence of sports-related SD, because they relied on reporting from individual schools and institutions or media accounts. Van Camp et al. (9), in a nationally based survey, estimated the prevalence of SD in high school and college athletes from the U.S. to be 0.4 in 100,000 athletes per year; estimated rates in male athletes (age 13 to 24 years, mean 16.9 ± 2.0) were 0.66 in 100,000 high school athletes per year and 1.45 in 100,000 college athletes per year; and estimated rates in female athletes (age 14 to 22 years, mean 16.2 ± 2.4) were 0.12 in 100,000 high school athletes per year and 0.28 in 100,000 college athletes per year. However, the methodology employed in that study was largely dependent on news media accounts, with the inherent limitations. Maron et al. (10)estimated the prevalence of cardiovascular SD in competitive high school athletes (age range 13 to 19 years, mean = 16) from Minnesota to be 0.35 in 100,000 sports participants and 0.46 in 100,000 individual participants per year (0.77 in 100,000 male athletes). The reasons for the higher mortality rates found in our investigation, compared with those reported by Maron et al. (10), may include: 1) a prospective versus retrospective study design; 2) the broader spectrum and higher mean value of age in our series of athletes (12 to 35 years, mean 23 ± 12); 3) different underlying pathologic substrates, which partly reflect differences in ethnic and genetic factors; and 4) participation at a higher level of intensity among our competitive athletes than among U.S. high school and college participants.

In the present study, the athletic-field SD rate showed a clear gender predilection, with striking male predominance (male/female ratio of 10:1). This predominance of fatal events in male athletes is consistent with the findings of previous surveys of athletic-field deaths (1,2,9–12)and has been explained by the lower number of participating females than males in competitive sports programs. Accordingly, the prevalence of sports participation in the young female population in our study was only 25% of that of the male individuals. However, our results indicate that male gender is in itself a risk factor for sports-related SD. It is noteworthy that males are generally exposed to more intensive training and higher levels of intensity during athletic sports and have a greater prevalence of cardiac diseases at risk of arrhythmic cardiac arrest in the age range of competitive sports than females (2,3,6). Unlike male athletes, the RR of SD between athletes and non-athletes did not reach statistical significance in females, mostly because of the relatively small number of fatal events in females.

Causes of SD

The present study showed that ARVC and CAD, either congenital or atherosclerotic, were the pathologic substrates associated with the greatest risk of SD in athletes. Previous studies in the U.S. showed a higher prevalence of other pathologic substrates such as hypertrophic cardiomyopathy (HCM), anomalous coronary arteries, and myocarditis (1,9,11,12). This discrepancy may be explained by several factors. There have been no previous studies like the present one that have prospectively investigated a consecutive series of young people with SD occurring in a well-defined geographic area with a homogeneous ethnic group. Moreover, morphologic examination of all hearts was performed by the same team of experienced cardiovascular pathologists according to a standard protocol. The high incidence of ARVC in our series may be due to a genetic factor in the population of the Veneto region of Italy (13,14). However, ARVC is rarely associated with cardiomegaly and usually spares the left ventricle, so that affected hearts may be erroneously diagnosed as normal hearts (3,8,15). In the past, therefore, a number of SDs in young people and athletes, in which the routine pathologic examination disclosed a normal heart, may, in fact, have been due to an unrecognized ARVC. Finally, systematic preparticipation screening of young competitive athletes, which has been in practice in Italy for more than 20 years, has successfully prevented SD from HCM by identification and disqualification of affected athletes (4).

Implications for preparticipation screening

The present study showed that sports was not in itself the cause of enhanced mortality, but it triggered cardiac arrest in those athletes who were affected by cardiovascular conditions predisposing to life-threatening ventricular arrhythmias during physical exercise. A corollary is that every effort should be made to recognize, by preparticipation screening, such diseases implicated in SD during sports, because disqualification of affected athletes makes the prevention of athletic-field death feasible. We previously reported that HCM was successfully detected at preparticipation evaluation (4). On the contrary, ARVC, premature CAD, and CCA (16), which formed the underlying substrate in most cases of SD in young competitive athletes, in the present study were missed at screening. These results do not dispute the usefulness of systematic evaluation of adolescents and young individuals embarking in sports activity; rather, they indicate those cardiovascular diseases, other than HCM, for which cardiovascular screening has to improve in the future.