You’re working the medical tent at the finish line of a marathon. It’s a cool day, and you’ve been treating blisters and minor injuries. Six hours after the start of the race, a petite, 28-year-old woman is brought to the tent by two family members. She’s confused, stumbling and slurring her speech. Her family members tell you she said she felt OK at the finish but has slowly worsened, in spite of drinking a lot of Gatorade, both on the course and after. What should you be concerned about, and how should this patient be managed?
Endurance-based athletic events, such as marathons and triathlons, are increasing in popularity throughout the world. EMS personnel are often called on to provide medical support for these events. Although these endurance sports attract healthy participants and make them healthier, a growing body of literature demonstrates the association between these events and the risk of sudden cardiac death, gastrointestinal disorders and hyponatremia.
History of Endurance Sports
The first modern marathon was held during the 1896 Olympic Games in Athens. Only 17 men ran that first Olympic marathon, and the winner finished in two hours, 58 minutes. The distance and course recreated the route of Pheidippides, a Greek messenger, who ran 24.8 miles to Athens after the defeat of the Persians at the Plain of Marathon. According to legend, he arrived, announced, ˙We are victorious,Ó and then died on the spot.
The first Boston Marathon took place in 1897, with just 15 entrants. The current marathon distance of 26 miles and 385 yards was established at the 1908 London Olympics so the race could start at Windsor Castle and end at the royal box in the stadium in London. The Boston Marathon broke the 1,000-runner mark in 1968 and allowed women as participants in 1972. In 1996, the Boston Marathon passed the 10,000-runner mark.
By 2005, there were 314 marathons in the U.S. with 382,000 finishers, 40% of whom were women. The New York City Marathon is the world_s largest, with 37,000 runners. As fields have grown in size, winning times have shrunk; the winning time in the 2006 Boston Marathon was two hours, seven minutes.
The average male marathoner is 40 years old and finishes in 4.5 hours; the average woman is 36 years old and finishes in five hours. Overall, finishing times are increasing as the event becomes more popular and crowded. This popularity has been spurred by celebrity participants, including George W. Bush, Al Gore, Sean ˙P. DiddyÓ Combs and Oprah Winfrey.
Triathlons are also increasing in popularity. The first event titled a ˙triathlonÓ consisted of a long jump, shot put and 100-yard dash at the 1904 Olympic Games in St. Louis. Modern triathlon was born in San Diego in 1974. In 1978, naval personnel who had transferred from San Diego to Hawaii began the signature triathlon event, the Ironman Triathlon World Championship, when they combined three classic Hawaiian distance racesƒthe Waikiki Roughwater Swim, the Around-Oahu Bike Race and the Honolulu Marathon.„
Twelve of the 15 starters finished the inaugural Ironman_s 2.4-mile ocean swim, 112-mile bike ride and 26.4-mile run. Several events then increased its appeal: It was featured on ABC_s ˙Wide World of SportsÓ in 1980, was a central focus of an episode on the ˙Magnum P.I.Ó TV series in 1981, and featured a dramatic ending in 1982, with the women_s leader crawling the last 100 yards to the finish line as the eventual women_s winner ran by her. Triathlon re-entered the Olympics in 2000 as a 1.5-km (0.93 mile) swim, 40-km (24.9 mile) bike ride and a 10-km (6.2 mile) run.
In 2006, approximately 125,000 people participated in triathlons in the United States. Standard distances include Olympic, half Ironman and full Ironman. Professionals finish Olympic, half Ironman and Ironman races in two, four and nine hours, respectively; amateurs typically take 50Ï100% more time to finish.„
The Chicago Triathlon, an Olympic distance race, is the world_s largest, with 7,600 entrants in 2005; the largest Ironman event is Ironman Wisconsin, with 2,500 athletes. These smaller numbers compared with those of marathons reflect the greater logistical complexity and physical demands of triathlons.
Sudden Cardiac Death
Jim Fixx was a recreational runner in the 1970s whose rise to celebrity mirrored the ˙running boomÓ of the time. He began running, quit smoking, lost 60 lbs. and authored a New York Times bestseller in 1977 titled, ˙The Complete Book of Running.Ó In 1984, at the age of 52, Fixx died of a myocardial infarction (MI) while running. The untimely deaths of runners, such as Fixx, are an irony that cynics still use as justification for not exercising.
Indeed, modern medicine’s view of the risk/benefit balance of endurance athletics has gone through phases. In 1867, a London surgeon named F.C. Skey wrote of competitive rowers, ˙young men Ú cannot perform a feat which compels them to put forth the absolute strength of their physical frame for 20 minutes without injury, often of a permanent nature.Ó1 A 1903 JAMA editorial asserted, ˙That repeated exercise, such as prolonged training in successive years, may lead to permanent injury is unquestionable.Ó
The tide had begun to swing back when the New England Journal of Medicine published an autopsy report in 1958 of Clarence DeMar. Known as ˙Mr. Marathon,Ó DeMar was a 32-time Boston Marathon finisher and seven-time winner. The autopsy noted his coronary arteries to be ˙unusually large and were estimated to be two or three times normal size,Ó speculating that running had a protective effect against coronary artery disease. During the beginning of the running boom, the Annals of the New York Academy of Science published an article in 1977 titled ˙Marathon running and immunity to atherosclerosis.Ó
Current evidence points to a reality that lies somewhere between the extremes. Early reports published in 1979 and 1980 demonstrated that marathoners were, in fact, not immune to occlusive coronary artery disease. Autopsy reports of male marathoners in their 30s and 40s revealed atherosclerosis in several who died accidentally and showed several others died of massive MIs while jogging.„
A 1982 study examined all deaths that occurred while the victim was jogging during a five-year period from 1975Ï1980 in the state of Rhode Island. Twelve men died (seven of them age 44 or younger), 11 of the 12 from heart disease. Using a telephone survey to determine the amount of jogging in the population, the authors calculated that one sudden death occurred for every 396,000 hours spent running. This was much higher than the baseline risk of one sudden death per three million hours for men the same age. A similar study in King County, Wash., utilized interviews with widows of men who died from sudden cardiac arrest to determine several important points.„
First, the risk of sudden death was increased while exercising (in terms of sudden cardiac deaths per minutes exercising compared with per minute at rest). The risk was 56 times higher for men who rarely exercised, 13 times higher for those who exercised moderately, and five times higher for those who exercised the most.„
Second, researchers found the more the men exercised, the lower their risk of sudden cardiac death at any time. The vast majority of these deaths occurred at rest, because the time spent exercising represented such a small proportion of overall time.
Two studies, one of 22,000 physicians and one of 70,000 female nurses, repeated the finding that the risk of sudden cardiac death was increased during exercise; that is, one_s risk of death during an hour of exercise was higher than during an hour of sedentary activity. Men who exercised less than once a week had a risk with exertion that was 74 times their risk at rest, whereas men who exercised more than five times per week had a risk with exertion only 10 times their risk at rest.„
These statistics support the stereotype of the sedentary man who collapses with a heart attack while shoveling snow. These studies also demonstrated lower all-cause mortality in subjects who exercised compared with those who did not. Additionally, the more time per week spent exercising, the more pronounced this effect was.„
The risk of death during a marathon was quantified in a report on all deaths occurring in 440,000 finishers in the Marine Corps Marathon from 1976Ï2004 and the Twin Cities Marathon from 1982Ï2004. In the report, there were nine cardiac arrestsƒtwo in runners with congenital heart defects, and the other seven in men aged 32 to 58 who had occlusive coronary artery disease. Roughly half of the cardiac arrests occurred between 1976 and 1994 (out of approximately 220,000 finishers). Of these first five events, four were fatal. From 1995Ï2004, out of an additional 220,000 finishers, four events occurred, only one fatal.„
Although these numbers are too small to reach statistical significance, the authors conjectured that rapid EMS attention with early defibrillation accounted for this decrease. The overall arrest rate of one per approximately 221,000 racing hours is comparable to the 1982 Rhode Island rate of one per 396,000 jogging hours.„
A recent New York Times article highlighted concerns that endurance events may be detrimental to cardiac health. It_s true that slightly elevated cardiac bio-markers and changes in cardiac relaxation patterns have been found after marathons, triathlons and cycling events; however, no human study has shown long-term damage.
In a laboratory study, rats were forced to swim for 3.5 hours, five hours or five hours after one week of daily swim training. Examination of their hearts following this exhausting exercise revealed microscopic damage and scarring in the untrained five-hour group, but not in the trained five-hour group. This suggests that although exhaustive exercise is potentially dangerous, proper training can prevent the negative effects.
Overall, current medical evidence demonstrates that habitual exercise is beneficial. The modestly increased risk of death while exercising is more than compensated for by the overall health benefits. However, regular exercise is not a panacea against heart disease. EMS providers at endurance events should anticipate providing ACLS care with early defibrillation, because sudden death that does occur is likely related to coronary artery disease.
Most runners are personally acquainted with side-stitches and ˙runner_s trots,Ó and as distance and effort increase, this gastrointestinal distress occasionally becomes more serious. Three-time Boston Marathon winner Uta Pippig broke the tape in 1996 with blood dripping down her legs from ischemic colitis, a condition usually found in elderly people with occlusive vascular disease that prevents adequate perfusion of the gut.„
Non-occlusive ischemic colitis has been reported in young victims of hemorrhagic shock; ischemic colitis from prolonged exercise may occur by a similar mechanism, with shunting of blood to the working muscles leading to hypoperfusion of the gut.„
Within 24 hours of the 2003 and 2004 Boston marathons, three women aged 26 to 39 presented to a local emergency department (ED) with abdominal pain, nausea, vomiting and bloody diarrhea. CT scans showed large bowel wall thickening consistent with runner_s colitis, without any occlusive vessel disease. Each patient improved with supportive care.„
A more dramatic example is that of Chris Legh, a professional triathlete whose collapse within sight of the finish line at the 1997 Ironman Hawaii was featured in a recent Gatorade commercial. Just 25 years old and in peak physical condition, Legh was unable to complete the race and received IV fluids at the aid station. The next day, he sought care for continued pain and fever and had an emergent hemicolectomy to resect a segment of ischemic colon.
Frank colitis and ischemia are uncommon, but other studies have demonstrated that 30Ï80% of runners experience gastrointestinal symptoms (e.g., pain, cramping, diarrhea) during long runs or races, and 25% of marathoners have trace blood in their stool following a race, suggesting that some degree of gut ischemia is common in endurance athletics. Gut ischemia compromises the protective mucosal barrier in the intestines and can lead to translocation of gut bacteria into the bloodstream.„
This, in turn, may lead to a state of diffuse inflammation commonly associated with sepsis. Endurance athletes exhibit elevated white cell counts (usually associated with infection) and creatine kinase levels (a marker of muscle breakdown), both of which are consistent with a generalized inflammatory state. It_s not clear if bacteremia alone, muscle breakdown from prolonged exercise, or a combination of the two leads to these biochemical findings.
Exercise-associated hyponatremia occurs when athletes drink excess water, diluting their serum sodium. This condition is exacerbated by sodium loss in sweat. Normal serum sodium concentration is approximately 140 mEq/L; sweat ranges from 20Ï90 mEq/L. Sweat rates in well-conditioned athletes range between 0.5Ï1.75 L per hour.„
The sodium concentrations of common sports drinks range from 8Ï20 mEq/L varying by brand. Athletes must ingest large amounts of fluids to maintain hydration during prolonged exercise, and even specialized sports drinks will dilute serum sodium.„
Exercise-associated hyponatremia is dangerous because its symptoms mimic those of dehydration: weakness, cramping, nausea, vomiting, fatigue, headache and irritability. A common reaction to these symptoms is increased fluid consumption in an ill-conceived attempt to correct perceived dehydration. Hyponatremia diverges symptomatically from dehydration only when symptoms become severe, with altered mental status followed by seizures.„
This phenomenon gained widespread attention in the media following a study on the 2002 Boston Marathon published in 2005 in the New England Journal of Medicine. However, prior to this media attention, significant work had already been published on the subject and race organizers had taken specific prevention measures to minimize exercise-associated hyponatremia during endurance events.„
For example, a study of athletes in the 1997 Ironman New Zealand showed that weight gain during the race correlated with hyponatremia, and that women gained more weight, had lower average sodium levels and were more likely to become hyponatremic. Measurement of fluid balance in 18 participants showed that the average racer drank 716 mL of fluid (24 oz.) per hour for a total of 9 L (2.4 gallons) over the course of the race. In spite of this, the average weight loss was 2.5 kg (5.5 lbs.). These racers burned 800 g (1.7 lbs.) of their fat and carbohydrate stores, liberating 1.4 kg (3.1 lbs.) of water, which explained almost the entire weight loss. Therefore, one could lose 2.2 kg (4.9 lbs.) of weight during an Ironman race and not change one_s total body water at all, because water content is created by aerobic metabolism.
Sports drink manufacturers have championed the notion that 2% dehydration is detrimental to performance and thermoregulation. The data supporting this assertion were derived from trials of short-term maximal intensity anaerobic activity, but studies of endurance races paint a very different picture. Data from the 2000 Ironman South Africa demonstrated a correlation between greater weight loss and faster finishing times, and no increase in core body temperature with weight loss.„
Analysis of 330 finishers at the 2004 Ironman New Zealand also showed that the faster racers lost more weight. At the 2004 Ironman Western Australia, 10 men swallowed temperature-sensing pills; their average increase in core temperature was only 1_ C (34_ F) in spite of weight loss averaging 2.3 kg (5 lbs.) or 3% of pre-race weight.
The Boston Marathon study raised awareness of exercise-induced hyponatremia and provided a comprehensive risk-factor analysis. Blood was drawn from 488 finishers, including 175 women. Sixty-two (13%) of these runners were hyponatremic (sodium less than 135 mEq/L), 37 (22%) of them women, and 25 (8%) of them men. Three (0.6%) had critical hyponatremia (less than 120 mEq/L).„
Factors associated with hyponatremia included longer race time, weight gain during the race, and low body mass index (BMI) (less than 20). Factors not associated with hyponatremia include gender, choice of fluid (water versus sports drink), age and NSAID use.
Previous studies identified women as being at higher risk, but after controlling their higher BMI and slower finishing times, women were not at increased risk. BMI is believed to be a risk factor because, for the same amount of fluid consumed, the dilutional effects on sodium will be greater for a smaller person. The 13% incidence of hyponatremia was surprising and showed that exercise-associated hyponatremia was not a rare condition.„
The Boston Marathon study may not reflect the incidence of hyponatremia at other marathons because the Boston Marathon is the only marathon in the U.S. to require its runners to meet strict qualifying times at a previous marathon. Therefore Boston runners don_t represent the populations at marathons with a large percentage of first-time participants, whose inexperience and slower speed place them at higher risk for hyponatremia.
Patients may also present with exercise-associated hyponatremia in a delayed fashion, after going home and treating their symptoms with hydration. In the 24 hours after the 1998 San Diego Rock Ân_ Roll Marathon, 53 participants presented to area EDs. Sixteen of these patients had orthopedic injuries or other benign processes. Of the remaining 37 patients, 21 (58%) were hyponatremic (sodium less than 125mEq/L). Five of these patients were admitted, three requiring intubation following seizures. Eighteen of these 21 runners were female and took four to 6.5 hours to finish the race.„
Participants who present in a delayed fashion after an endurance event are at high risk for severe hyponatremia, and caution should be exercised in giving these patients fluids. Physical examination and symptoms are unreliable in differentiating hyponatremia from dehydration, so on-site race medical staff should consider having rapid point-of-care testing available to accurately triage hyponatremic patients to EDs for hypertonic saline. Other reversible causes of patients_ symptoms should be evaluated, such as hypoxia, hypotension or hypoglycemia.
Strategies have been implemented to prevent exercise-associated hyponatremia during these events. The researchers of the 2004 Ironman New Zealand found only six racers with sodium levels less than 135mEq/L, compared with the 1997 event when 58 racers had sodium levels less than 135 and 11 had less than 130 (both studies drew blood from 330 finishers).28„
This improvement was attributed to education (warnings against excessive water intake in entrants_ race packets) and decreased exposure to the causative agent (water stations on the run were decreased from every 1.8 km in 1997 to every 2.4 km in 2004).„
With weight gain during racing linked to hyponatremia in multiple studies, prospective participants should calculate their sweat rates during training (by pre- and post-training weights) and then calculate the amount of fluid they would need to maintain a slightly negative weight loss during the race. Data from the 2002 Boston Marathon study supports this strategyƒno runner who lost more than 1 kg had severe hyponatremia, and the risk of severe hyponatremia increased dramatically with weight gain. (Editor_s note: For more on this condition, read the bonus article„˙Hormonal Imbalance Found to Contribute to Exercise-Associated Hyponatremia,Ó at„www.jems.com.)
Going back to the introductory case presentation, dehydration and hyponatremia are both possible causes for this 28-year-old runner_s symptoms. Her fluid intake, small body size and delayed presentation point toward hyponatremia. Altered mental status is an alarming finding, and she should be transported to an ED for serum sodium testing and appropriate resuscitation. Although normal saline administration in the field would not necessarily harm this woman, she needs prompt lab testing and access to hypertonic saline in the event her hyponatremia worsens and she begins seizing. Hypotonic saline should be avoided.
Endurance athletic events, such as marathons and triathlons, are growing sports. Although the health benefits of habitual exercise far outweigh the negative effects, some risks remain, including sudden cardiac death, intestinal ischemia and exercise-associated hyponatremia. Health-care providers at endurance events should be prepared to provide rapid defibrillation anywhere on the race course and recognize frank peritonitis as„
a surgical emergency. Additionally, they should be cautious in the use of fluids to resuscitate an athlete with undifferentiated symptoms and an unknown serum sodium who may be hyponatremicƒa common, potentially lethal, and difficult-to-diagnose condition.„
Daniel K. Vining, MD, and„David F. Gaieski, MD, are emergency physicians in the Department of Emergency Medicine at the Hospital of the University of Pennsylvania, Philadelphia. Contact Dr. Vining at [email protected]
- Thompson PD, Apple FS, Wu A: ˙Historical concepts of the athlete_s heart.Ó„Medicine and Science in Sports and Exercise. 36(3):363Ï370, 2004.
- Currens JH, White PD: ˙Half a century of running. Clinical, physiologic and autopsy findings in the case of Clarence DeMar (ÂMr. Marathon_).Ó„New England Journal of Medicine. 265:988Ï993, 1961.
- Bassler TJ: ˙Marathon running and immunity to atherosclerosis.Ó„Annals of the New York Academy of Sciences. 301:579Ï592, 1977.
- Noakes TD, Opie LH, Rose AG, et al: ˙Autopsy-proved coronary atherosclerosis in marathon runners.Ó„New England Journal of Medicine. 301(2):86_Ï89, 1979.
- Waller BF, Roberts WC: ˙Sudden death while running in conditioned runners aged 40 years or over.Ó„American Journal of Cardiology. 45(6):1292Ï1300, 1980.
- Thompson PD, Funk EJ, Carleton RA, et al: ˙Incidence of death during jogging in Rhode Island from 1975 through 1980.Ó„JAMA. 247(18):2535Ï2538, 1982.
- Siscovick DS, Weiss NS, Fletcher RH, et al: ˙The incidence of primary cardiac arrest during vigorous exercise.Ó„New England Journal of Medicine. 311(14):874Ï877, 1984.
- Albert CM, Mittleman MA, Chae CU, et al: ˙Triggering of sudden death from cardiac causes by vigorous exertion.Ó„New England Journal of Medicine. 343(19):1355Ï1361, 2000.
- Whang W, Manson JE, Hu FB, et al: ˙Physical exertion, exercise, and sudden cardiac death in women.Ó„JAMA. 295(12):1399Ï1403, 2006.
- Roberts WO, Maron BJ: ˙Evidence for decreasing occurrence of sudden cardiac death associated with the marathon.Ó„Journal of the American College of Cardiology. 46(7):1373Ï1374, 2005.
- Reynolds G: ˙Is marathoning too much of a good thing for your heart?Ó„New York Times. Dec 7, 2006. www.nytimes.com.
- Whyte G, George K, Shave R, et al: ˙Impact of marathon running on cardiac structure and function in recreational runners.Ó„Clinical Science. 108(1):73Ï80, 2005.
- George K, Whyte G, Stephenson C, et al: ˙Postexercise left ventricular function and cTnT in recreational marathon runners.Ó„Medicine and Science in Sports and Exercise. 36(10):1709Ï1715, 2004.
- Rifai N, Douglas PS, O_Toole M, et al: ˙Cardiac troponin T and I, echocardiographic wall motion analyses, and ejection fractions in athletes participating in the Hawaii Ironman Triathlon.Ó„American Journal of Cardiology. 83(7):1085Ï1089, 1999.
- La Gerche A, Boyle A, Wilson AM, et al: ˙No evidence of sustained myocardial injury following an Ironman distance triathlon.Ó„International Journal of Sports Medicine. 25(1):45Ï49, 2004.
- Kalliokoski KK, Laaksonen MS, Luotolahti M, et al: ˙Myocardial perfusion after marathon running.Ó„Scandinavian Journal of Medicine & Science in Sports.„ 14(4):208Ï214, 2004.
- Chen Y, Serfass RC, Mackey-Bojack SM, et al: ˙Cardiac troponin T alterations in myocardium and serum of rats after stressful, prolonged intense exercise.Ó„Journal of Applied Physiology. 88(5):1749Ï1755, 2000.
- Lucas W, Schroy PC: ˙Reversible ischemic colitis in a high endurance athlete.Ó„American Journal of Gastroenterology. 93(11):2231Ï2234, 1998.
- Kyriakos R, Siewert B, Kato E, et al: ˙CT findings in runner_s colitis.Ó„Abdominal Imaging. 31(1):54Ï56, 2006.
- McMahon LF, Ryan MJ, Larson D, et al: ˙Occult gastrointestinal blood loss in marathon runners.Ó„Annals of Internal Medicine. 100(6):846Ï847, 1984.
- Bosenberg AT, Brock-Utne JG, Gaffin SL, et al: ˙Strenuous exercise causes systemic endotoxemia.Ó„Journal of Applied Physiology. 65(1):106Ï108, 1988.
- Smith JE, Garbutt G, Lopes P, et al: ˙Effects of prolonged strenuous exercise (marathon running) on biochemical and haematological markers used in the investigation of patients in the emergency department.Ó„British Journal of Sports Medicine.„ 38(3):292Ï294, 2004.
- Jeukendrup AE, Vet-Joop K, Sturk A, et al: ˙Relationship between gastro-intestinal complaints and endotoxaemia, cytokine release and the acute-phase reaction during and after a long-distance triathlon in highly trained men.Ó„Clinical Science. 98(1):47Ï55, 2000.
- Almond CS, Shin AY, Fortescue EB, et al: ˙Hyponatremia among runners in the Boston Marathon.Ó„New England Journal of Medicine. 352(15):1550Ï1556, 2005.
- Speedy DB, Noakes TD, Rogers IR, et al: ˙Hyponatremia in ultradistance triathletes.Ó„Medicine and Science in Sports and Exercise. 31(6):809Ï815, 1999.
- Speedy DB, Noakes TD, Kimber NE, et al: ˙Fluid balance during and after an ironman triathlon.Ó„Clinical Journal of Sport Medicine.„ 11(1):44Ï50, 2001.
- Sharwood KA, Collins M, Goedecke JH, et al: ˙Weight changes, medical complications, and performance during an Ironman triathlon.Ó„British Journal of Sports Medicine. 38(6):718Ï724, 2004.
- Wharam PC, Speedy DB, Noakes TD, et al: ˙NSAID use increases the risk of developing hyponatremia during an Ironman triathlon.Ó„Medicine and Science in Sports and Exercise. 38(4):618Ï622, 2006.
- Laursen PB, Suriano R, Quod MJ, et al: ˙Core temperature and hydration status during an Ironman triathlon.Ó„British Journal of Sports Medicine.„ 40(4):320Ï325, 2006.
- Davis DP, Videen JS, Marino A, et al: ˙Exercise-
associated hyponatremia in marathon runners: a two-year experience.Ó„Journal of Emergency Medicine. 21(1):47Ï57, 2001.
- Almond CS, Shin AY, Fortescue EB, et al: ˙Hyponatremia among runners in the Boston Marathon.Ó Author reply.„New England Journal of Medicine. 353(4):427Ï428, 2005.„„„