The new england journal of medicine original article Hyponatremia among Runners in the Boston Marathon Christopher S.D. Almond, M.D., M.P.H., Andrew Y. Shin, M.D., Elizabeth B. Fortescue, M.D., Rebekah C. Mannix, M.D., David Wypij, Ph.D., Bryce A. Binstadt, M.D., Ph.D., Christine N. Duncan, M.D., David P. Olson, M.D., Ph.D., Ann E. Salerno, M.D., Jane W. Newburger, M.D., M.P.H., and David S. Greenes, M.D. abstract background From the Departments of Medicine Hyponatremia has emerged as an important cause of race-related death and life-threat- (C.S.D.A., A.Y.S., E.B.F., R.C.M., B.A.B., ening illness among marathon runners. We studied a cohort of marathon runners to C.N.D., D.P.O., A.E.S., D.S.G.) and Cardi- ology (D.W., J.W.N.) and the Clinical Re- estimate the incidence of hyponatremia and to identify the principal risk factors. search Program (D.W.), Children’s Hospi- tal; the Department of Pediatrics, Harvard methods Medical School (C.S.D.A., A.Y.S., E.B.F., R.C.M., D.W., B.A.B., C.N.D., D.P.O., A.E.S., Participants in the 2002 Boston Marathon were recruited one or two days before the J.W.N., D.S.G.); and the Department of race. Subjects completed a survey describing demographic information and training Biostatistics, Harvard School of Public history. After the race, runners provided a blood sample and completed a questionnaire Health (D.W.) — all in Boston. Address reprint requests to Dr. Almond at the De- detailing their fluid consumption and urine output during the race. Prerace and post- partment of Cardiology, Children’s Hospi- race weights were recorded. Multivariate regression analyses were performed to iden- tal, Bader 2, 300 Longwood Ave., Boston, tify risk factors associated with hyponatremia. MA 02115, or at christopher.almond@ childrens.harvard.edu. results N Engl J Med 2005;352:1550-6. Of 766 runners enrolled, 488 runners (64 percent) provided a usable blood sample at Copyright © 2005 Massachusetts Medical Society. the finish line. Thirteen percent had hyponatremia (a serum sodium concentration of 135 mmol per liter or less); 0.6 percent had critical hyponatremia (120 mmol per liter or less). On univariate analyses, hyponatremia was associated with substantial weight gain, consumption of more than 3 liters of fluids during the race, consumption of flu- ids every mile, a racing time of >4:00 hours, female sex, and low body-mass index. On multivariate analysis, hyponatremia was associated with weight gain (odds ratio, 4.2; 95 percent confidence interval, 2.2 to 8.2), a racing time of >4:00 hours (odds ratio for the comparison with a time of <3:30 hours, 7.4; 95 percent confidence interval, 2.9 to 23.1), and body-mass-index extremes. conclusions Hyponatremia occurs in a substantial fraction of nonelite marathon runners and can be severe. Considerable weight gain while running, a long racing time, and body- mass-index extremes were associated with hyponatremia, whereas female sex, compo- sition of fluids ingested, and use of nonsteroidal antiinflammatory drugs were not. 1550 n engl j med 352;15 www.nejm.org april 14, 2005 Downloaded from www.nejm.org at KAROLINSKA INSTITUTE UNIVERSITY LIBRARY on May 28, 2008 . Copyright © 2005 Massachusetts Medical Society. All rights reserved. hyponatremia among runners in the boston marathon s marathon running has surged outcome measures in popularity during the past quarter-cen- The primary hypothesis of the study was that exces- a 1 tury, reports have emerged of serious ill- sive consumption of hypotonic fluids is associated ness and death from hyponatremia,2-8 as in the case with hyponatremia in marathon runners. Hypona- of a 28-year-old woman who died after the 2002 tremia was defined as a serum sodium concentra- Boston Marathon.2 The incidence of hyponatremia tion of 135 mmol per liter or less. Severe hypona- among marathon runners is unknown, since previ- tremia and critical hyponatremia were defined as ous studies have been small and limited to runners serum sodium concentrations of 130 and 120 mmol presenting for medical attention.4,5,7,9-11 per liter or less, respectively. Independent variables Excessive fluid intake is believed to be the pri- analyzed for association with hyponatremia includ- mary risk factor for hyponatremia, on the basis of ed weight change during the race and self-reported observations of marathon runners who have col- fluid intake including volume, frequency, and type. lapsed2-5,7,11,12 and studies of elite athletes.13-17 Both water and a sports drink containing electro- However, other risk factors have also been suggest- lytes were offered at each milepost, and runners ed, including the composition of fluids consumed were asked to estimate the proportion of their in- (e.g., plain water, rather than sports drinks that con- take from each. Other predictors that we considered tain electrolytes), relatively low body-mass index, included sex (a dichotomous variable), body-mass long racing time, lack of marathon experience, use index (the weight in kilograms divided by the square of nonsteroidal antiinflammatory drugs (NSAIDs), of the height in meters), training pace, number of and female sex.4,5,9,18 We undertook the present previous marathons (dichotomized at a median of study to estimate prospectively the incidence of hy- five), duration of the marathon in hours and min- ponatremia among marathon runners and to iden- utes, use or nonuse of NSAIDs in the past week (a di- tify the principal risk factors involved. chotomous variable), age, and race (a dichotomous variable [white vs. nonwhite]). Race was self-report- methods ed by the runners. study population statistical analysis Marathon runners were recruited prospectively at Descriptive statistics were used to estimate the in- an exposition one or two days before the Boston cidence of hyponatremia and to characterize the Marathon, in April 2002. All registered participants demographic information supplied by the runners. 18 years of age or older were eligible, regardless of Unless otherwise specified, t-tests and Fisher’s ex- whether they registered for the marathon on the ba- act test were used to identify univariate predictors sis of a competitive qualifying time or on behalf of a associated with hyponatremia, at a level of statisti- charitable organization — a mechanism for which cal significance of P≤0.05. Logistic regression (SAS no previous marathon experience was required. software, version 9.0) and generalized additive 19 Subjects were approached at random in an area ad- models (S-Plus software, version 6.1 for Win- jacent to race registration and invited to participate. dows) were used in the multivariate analysis to iden- Written informed consent was obtained from all tify independent predictors of hyponatremia. subjects. The study protocol was approved by the Committee on Clinical Investigation at Children’s results Hospital in Boston. Table 1 summarizes the baseline demographic and study design training characteristics of the study population. Of Before running the marathon, subjects completed 766 runners enrolled, 511 (67 percent) reported to a survey describing baseline demographic and train- the finish-line research station. Of these, 489 pro- ing information, medical history, and anticipated vided a blood sample (constraints such as plane hydration strategies for the race. At the finish line, flights precluded 22 runners from providing a sam- runners provided a blood sample and completed a ple). One sample was considered of insufficient questionnaire detailing their fluid consumption and quantity, leaving a total of 488 subjects for analysis. urine output during the race. Blood samples were Overall, among all 766 runners enrolled, female centrifuged on site and frozen at ¡70°C until ana- runners were younger than male runners (mean lyzed. With the use of a digital balance, the prerace [±SD] age, 36.1±8.8 vs. 40.4±9.7 years; P<0.001) and postrace weights were recorded for each runner. and had a lower prerace weight (58.8±6.8 vs. n engl j med 352;15 www.nejm.org april 14, 2005 1551 Downloaded from www.nejm.org at KAROLINSKA INSTITUTE UNIVERSITY LIBRARY on May 28, 2008 . Copyright © 2005 Massachusetts Medical Society. All rights reserved. The new england journal of medicine Table 1. Baseline Characteristics of the 2002 Boston Marathon Study Population.* Characteristic Male Runners (N=473) Female Runners (N=293) Reporting Not Reporting Reporting Not Reporting at Finish Line at Finish Line at Finish Line at Finish Line (N=336) (N=137) (N=175) (N=118) Age — yr 40.4±9.6 40.4±10.0 36.3±8.8 35.7±8.8 Nonwhite race — % 9 10 6 6 Prerace weight — kg 74.6±9.5 76.6±10.7 58.9±6.7 58.7±7.1 Body-mass index† 23.7±2.6 24.5±2.7 21.4±2.0 21.4±2.1 Training pace — min:sec/mi 7:53±1:02 8:04±1:09 8:40±1:01 8:41±1:02 Previous marathons — median no. 5 (2–12) 4 (1–12) 4 (2–8) 3 (1–6) (interquartile range) Self-reported water loading — %‡ 75 79 70 85 Self-reported use of NSAIDs — %§ 51 54 60 61 Race duration — hr:min¶ 3:37±0:42 3:46±0:40 4:02±0:36 4:02±0:32 * Plus–minus values are means ±SD. The temperature and humidity at noon, at the start of the race, were 53°F (12°C) and 96 percent, respectively; at 2 p.m. at the finish line, they were 55°F (13°C) and 83 percent.