Sudden Death During the Triathlon

To the Editor: Triathlon is among the most vigorous amateur athletic disciplines, requiring expertise in swimming, biking, and running.^1-3 Although sudden death risk has been assessed for the amateur marathon,⁴ it has not been systematically investigated for triathlon.

Participants who completed 2971 USA Triathlon (USAT) sanctioned events from January 2006 through September 2008 were tabulated using online race results (approximately 95% of event results).⁵ Participants in nonsanctioned races, relay races, or triathlons without full swim-bike-run sequence were excluded. Deaths were identified in the US Registry of Sudden Deaths in Athletes⁶ and USAT records,⁵ which have tabulated these events over 30 and 5 years, respectively; autopsy reports were obtained from medical examiners. The Abbott Northwestern Hospital institutional review board determined this study was exempt. Confidence intervals (CIs) were calculated using Poisson analysis (JMP version 7; SAS Institute Inc, Cary, North Carolina).

A total of 959 214 participants were analyzed (mean [SD], 323 [444] per race); 59% were men. Forty-five percent competed in short (swim <750 m), 40% in intermediate (swim 750-1500 m), and 15% in long (swim >1500 m) triathlon races (Table).

Fourteen participants died during 14 triathlons (rate, 1.5 per 100 000 participants; 95% CI, 0.9-2.5), including 13 while swimming and 1 biking (Table). Athletes who died were 28 to 65 years old (mean [SD] age, 44 [10] years). Triathlons with deaths included more participants (n = 1319; 95% CI, 1084-1584) than races without deaths (n = 318; 95% CI, 302-334). Of the swimming deaths, 11 were men and 2 were women.

Six deaths occurred in short, 4 in intermediate, and 3 in long races (2 in an Ironman triathlon). Eight swimmers were in distress and called for assistance, and 5 were found motionless on the water. Deaths occurred in the open ocean (n = 6), lakes (n = 4), reservoirs (n = 2), or a river (n = 1). The bicycle fatality resulted from a fall causing cervical injuries.

Drowning was the declared cause of each swimming death, but 7 of 9 athletes with autopsy had cardiovascular abnormalities identified. Six had mild left ventricular hypertrophy with maximum wall thickness of 15 to 17 mm and mean (SD) heart weight of 403 (77) g, including 1 with a clinical history of Wolff-Parkinson-White syndrome. One other athlete had a congenital coronary arterial anomaly, and 2 had structurally normal hearts.

Although the contribution of cardiovascular abnormalities cannot be definitively excluded in some cases,² logistical factors and adverse environmental conditions may have been responsible for these events, given that about 95% of triathlon fatalities occurred during the swimming segment. Furthermore, deaths were more common in triathlons involving greater numbers of competitors. Because triathlons begin with chaotic, highly dense mass starts, involving up to 2000 largely novice competitors entering the water simultaneously, there is opportunity for bodily contact and exposure to cold turbulent water.³ Triathlons also pose inherent obstacles to identifying distressed athletes and initiating timely resuscitation on open water. Compared with these triathlon findings, marathon racing analyzed for more than 3 million runners over 30 years reported a mortality rate of 0.8 per 100 000 participants (95% CI, 0.5-1.1).⁴

Study limitations include the possibility that all sudden deaths may not have been identified, as neither of the registries is based on mandatory reporting. Although it is not possible to determine the precise number of US triathlons annually, USAT events likely represent a large proportion. This study was designed to explore risk per participation; an unknown number of athletes competed more than once within the data set.

Although mass screening before competition may be impractical, awareness of cardiovascular risks may motivate athletes to seek preparticipation evaluations on an individual basis. Efforts to improve triathlon safety could include establishing minimum achievement standards for participation, including swimming proficiency.

Author Contributions: Dr Harris had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Harris, Maron.

Acquisition of data: Harris, Henry, Rohman, Haas, Maron.

Analysis and interpretation of data: Harris, Maron.

Drafting of the manuscript: Harris, Henry, Rohman, Haas, Maron.

Critical revision of the manuscript for important intellectual content: Harris, Maron.

Statistical analysis: Henry.

Obtained funding: Harris, Maron.

Administrative, technical, or material support: Harris, Henry, Rohman, Haas, Maron.

Funding/Support: This study was supported by the Hearst Foundations, San Francisco, California.

Role of the Sponsor: The sponsor had no role in the design and conduct of the study; in the collection, analysis, and interpretation of the data; or in the preparation, review, or approval of the manuscript.

Additional Contributions: James S. Hodges, PhD, Division of Biostatistics, University of Minnesota, provided statistical advice and analysis as a contract employee of the Minneapolis Heart Institute Foundation.