Journal Logo

CLINICAL SCIENCES: Clinical Investigations

An Ultracyclist with Pulmonary Edema during the Bicycle Race Across America


Author Information
Medicine & Science in Sports & Exercise: January 2007 - Volume 39 - Issue 1 - p 8-12
doi: 10.1249/01.mss.0000235885.79110.79
  • Free


Multiday ultraendurance athletic events stress the body's ability to maintain sodium and water homeostasis. Despite the growth of this type of athletic competition, there are few data to guide adequate and safe repletion strategies. A recent, widely cited report (1) documented a 13% incidence of hyponatremia among nonelite marathon runners. Speedy et al. (26) reported an 18% incidence of hyponatremia among a cohort of participants in an ultradistance triathlon. Because 8 of the 11 athletes (73%) with severe hyponatremia (plasma sodium < 130 mmol·L−1) either gained or maintained weight during the course of the race, they concluded that fluid overload was the predominant cause of the hyponatremia. An implication of these studies was that excessive water intake without sufficient sodium repletion can predispose to this condition. However, excessive sodium intake also poses problems. We present the case of a 38-yr-old participant in the Bicycle Race Across America who undertook an aggressive program of sodium repletion during the first 4 d of the race and subsequently developed pulmonary edema while riding at an elevation of 2380 m. This case demonstrates the pitfalls of overly aggressive sodium intake, particularly with heavy exertion at high altitude.


This extremely fit, well-trained 38-yr-old man participated in the Bicycle Race Across America, a 4920-km continuous race from San Diego, CA to Atlantic City, NJ. The race included more than 33,500 m of vertical climbing throughout the course of the event. Before the race, his only medical problem was mild hypertension, for which he was treated with irbesartan, an angiotensin receptor antagonist. Born and raised in Switzerland, the patient was a formerly competitive cross-country skier whose only history of problems at high altitude involved repeated episodes of mild acute mountain sickness at elevations above 3000 m. The patient planned to ride 21 out of every 24 h to finish the race in 10-11 d. His dietary regimen included consumption of 23 g of sodium on day 1 of the race and 25 g on day 2. The dietary logs from subsequent days were lost, but the patient reported that he had maintained similar dietary sodium intake during the remainder of his race. The patient had no difficulties until the evening of the third day, when he developed increased dyspnea at his fixed race pace. As he climbed towards 2380 m the next morning, his dyspnea worsened, causing him to slow his pace and take frequent rest stops. He also developed blurry vision and a cough that produced white sputum. On two separate occasions, he stopped riding and used supplemental oxygen for 5 min. The supplemental oxygen, permitted under race rules as long as the participant is off the bicycle, did not improve his symptoms.

Given his worsening clinical condition, he and his support crew decided to seek medical attention. On arrival at a local medical facility in Pagosa Springs, CO (elevation 2290 m), he was noted to be severely dyspneic, cyanotic, and diaphoretic. He had a temperature of 39.2°C, blood pressure of 170/100 mm Hg, heart rate of 128 bpm, respiratory rate of 60 breaths per minute, and arterial oxygen saturation of 42-48% on room air. His body weight was 2.7 kg higher than his prerace weight. On exam, he had crackles throughout the right lung and at the left base. An electrocardiogram revealed sinus tachycardia without ischemic changes. Portable chest x-ray demonstrated opacities in the right middle, right lower, and left lower lobes. Laboratory studies (Table 1) revealed a total white blood cell count of 7.7 × 103 mm3 with a normal differential and a hematocrit of 38.2%. The chemistry panel included sodium of 139 mmol·L−1; urea nitrogen of 25 mg·dL−1; creatinine of 0.8 mg·dL−1; albumin of 3.5 g·dL−1; and total protein of 6.1 g·dL−1. Of note, on laboratory studies conducted 3 months before the race, the patient had a hematocrit of 46%, albumin of 4.9 g·dL−1, and total protein of 7.4 g·dL−1 (Table 1). Blood cultures were negative. He was treated with 500 mg of intravenous methylprednisolone and two separate 20-mg doses of intravenous furosemide and oxygen before being transferred to a larger medical facility at a lower elevation in Albuquerque, NM (elevation of 1490 m).

Comparison of selected laboratory values drawn 3 months before the Bicycle Race Across America and on initial presentation to the emergency room.

On transfer, the patient underwent a radionuclide ventilation-perfusion scan that was interpreted as low probability for pulmonary embolism. A CT pulmonary angiogram showed no evidence of pulmonary embolism or pleural effusions, but it did reveal bilateral patchy ground glass opacities and a dense right middle-lobe infiltrate (Fig. 1). A lower-extremity venous duplex was negative for deep venous thrombosis. A transthoracic echocardiogram demonstrated a left ventricular ejection fraction of 65%, normal right and left ventricular size and function, no valvular abnormalities, and a pulmonary artery systolic pressure of 45-50 mm Hg. He was continued on intravenous furosemide and started on inhaled salmeterol, moxifloxacin, and nifedipine. He was taken off the methylprednisolone that had been started at the original medical facility. Forty-eight hours after admission, the patient's oxygen saturation had improved to 90% on room air, and he no longer had dyspnea at rest or with exertion. He was subsequently discharged from the hospital without further complications.

Single slice from a CT pulmonary angiogram obtained on transfer to the second hospital, demonstrating a dense right middle-lobe infiltrate and patchy ground glass opacities in the left and right lung fields.

Several weeks after the patient's symptoms had resolved and he had returned home, he underwent further evaluation. A repeat chest x-ray was normal. On pulmonary function testing, the patient had a forced vital capacity (FVC) of 6.19 L (114% predicted), forced expiratory volume in 1 s (FEV1) of 4.41 L (100% predicted), FEV1/FVC ratio of 71%, total lung capacity of 8.30 L (116% predicted), and a diffusion capacity for carbon monoxide of 33.6 mL·min−1·mm Hg−1 (98% predicted). He then underwent a progressive cardiopulmonary exercise test, reaching a power output of 500 W (234% predicted) and a V˙O2max of 63.9 mL·kg−1·min−1 (156% predicted). The latter value was essentially unchanged from a prior cardiopulmonary exercise test performed 15 months before the bicycle race. At peak exercise, his heart rate reached 179 bpm (98% predicted), and his minute ventilation rose to 186 L·min−1, representing 100% of his maximum voluntary ventilation. He also demonstrated a clear anaerobic threshold, a small fall in his oxygen saturation from 100 to 96%, and a fall in his dead-space fraction from 0.29 at rest to 0.17 during exercise. Echocardiography revealed a pulmonary artery systolic pressure of 25-30 mm Hg at rest, which did not change significantly after a 45-min exposure to an inspired oxygen concentration (FiO2) of 0.12 at rest. An attempt to measure the pulmonary artery systolic pressure after exercising on a treadmill to 80% of his age-predicted maximum heart rate was unsuccessful because of difficulty identifying a clear tricuspid regurgitant jet at the end of exercise.


Acute pulmonary edema has been infrequently described in athletes engaged in ultraendurance events. McKechnie et al. (18) described two athletes who developed acute pulmonary edema during a 90-km running race; these outcomes were hypothesized to have been caused by changes in myocardial compliance with prolonged exercise. In addition to this case report, there is a large body of evidence that athletes develop interstitial edema during sustained heavy exercise. McKenzie et al. (19), for example, used magnetic resonance imaging to demonstrate a 9.4% increase in extravascular lung water in highly trained cyclists who had just completed 45 min of cycling at 76% V˙O2max. Similarly, Anholm et al. (2) had well-trained cyclists ride between 5 and 131 km and used conventional radiography to demonstrate an increase in the lung edema score between the pre- and postride radiographs. It is possible that our patient's case of acute pulmonary edema represents an extreme form of the processes described in these studies, exacerbated by the exceedingly long duration of cycling during this particular race.

The fact that our patient's problem developed at high elevation, albeit only a modestly high one, also raises the question of whether this represents a typical case of high-altitude pulmonary edema (HAPE). HAPE is a noncardiogenic form of pulmonary edema seen in unacclimatized lowlanders who ascend to high elevations. Symptoms develop anywhere from 2 to 5 d after ascent and include dyspnea on exertion, decreasing exercise performance, and a dry cough. As the illness worsens, patients develop dyspnea at rest, cyanosis, and a cough producing pink, frothy sputum; they can die if the disorder is not treated in time (6). The incidence of HAPE varies between 0.2 and 15%, depending on the altitude reached and the rate of ascent (5,23); the main risk factors are an overly fast ascent to too high an elevation, overexertion, and individual susceptibility. Although our patient decompensated at high altitude, his case is atypical for HAPE in several respects. First, he developed pulmonary edema at a relatively low elevation of 2300 m. HAPE is significantly more common at elevations above 3000 m (6), although cases have been described in Summit County, CO (24) (base elevation 2743-3354 m), and a recent report described a series of patients with HAPE at elevations below 2400 m (11). Second, the patient does not fit the phenotype demonstrated by HAPE-susceptible individuals. Multiple studies have demonstrated that those with HAPE susceptibility have abnormal pulmonary vascular reactivity. In particular, while breathing air with a FiO2 of 0.12 at rest, exercising in normoxia, or exercising while breathing an FiO2 of 0.12, HAPE-susceptible individuals demonstrate significantly greater increases in their pulmonary artery pressure than non-HAPE-susceptible individuals, with systolic pressures rising in many cases to more than 60 mm Hg (9,13). Our patient's hypoxic pulmonary vascular response was found to be minimal; while breathing a FiO2 of 0.12 for 45 min, there was little change in his systolic pulmonary artery systolic pressure (< 5 mm Hg). These observations fit well with the patient's previous experience with exercise at altitude. Given this modest hypoxic vasoconstrictor response, we would argue that he does not have the characteristic phenotype of HAPE susceptibility and, as a result, would not have been expected to develop pulmonary edema at an elevation of 2300 m in the absence of another critical factor.

In this case, we believe that the other factor was his aggressive sodium-loading regimen. Several studies suggest that high salt intake can lead to marked plasma volume expansion. Luetkemeier (16) performed a retrospective analysis of sodium intake and plasma volume expansion in cyclists engaged in a 5-d training regimen and found a mean increase in plasma volume of 4.5 mL·kg−1 of body weight as well as a strong positive correlation (r = 0.81) between total estimated sodium intake and plasma volume expansion. Individuals who ingested between 750 and 850 mEq of sodium during the course of 3 d experienced a 9- to 14-mL·kg−1 increase in plasma volume. Armstrong et al. (3) randomized subjects to either a high-sodium intake regimen (397 mEq·d−1) or a low-sodium intake regimen (97 mEq·d−1) during an 8-d training regimen (90 min·d−1 in a warm environment) and found that on day 4, the high-sodium intake group had a statistically significant increase in plasma volume compared with the low-intake group. By the end of the training period, there was no longer a difference in plasma volume expansion between the high- and low-sodium intake groups, but both groups still had increases in plasma volume compared with their pretraining values. Even in the absence of extensive salt loading, there is evidence to suggest that ultraendurance athletic activity itself leads to plasma volume expansion. For example, Neumayr et al. (20) studied 16 ultraendurance cyclists during the 525-km Race Across the Alps and found that plasma volume expanded by 8% at the end of the race and by 22% after 1 d of recovery.

The upper limit for maximal sodium excretion and defense of extracellular fluid volume in humans is not known. In resting conditions, the ability of the kidney to excrete sodium in the face of increased intake depends on the magnitude of the change in intake; in general, progressively more extracellular fluid expansion occurs with greater increases in sodium ingestion. In humans, an increase in dietary sodium from 10 to 150 mEq·d−1 is accompanied by 2-kg weight gain reflecting a 2-L expansion of the extracellular fluid (21). With heat and exercise, sodium losses occur in sweat as well, and thus the cumulative extracellular fluid gain with added sodium intake is less. In addition, renal blood flow decreases in proportion to the level of exertion (22), and with 21 h·d−1 of exertion, our subject could be considered to have been in a mild form of functional renal failure. The data in exercising humans are limited but instructive. In a study employing a similar design as the one cited above, Armstrong et al. (4) demonstrated that despite increased urine and sweat sodium excretion, subjects randomized to the high-sodium intake regimen (397 mEq·d−1) had a net daily sodium gain of 114.5 ± 12.3 mEq and an average gain of 916 ± 98.9 mEq during the 8-d training period. We have no measure of our patient's urinary and sweat sodium losses, and thus we cannot calculate his net sodium retention. However, because the volume of distribution of sodium is limited to the extracellular space, the patient's 2.7-kg weight gain suggests about a 3-L expansion of extracellular fluid. On the basis of our patient's ideal body weight (70 kg), this represents a 21% increase in the volume of the extracellular fluid. Given that plasma volume represents roughly 20% of extracellular fluid, we would predict that his plasma volume expanded by 600 mL. This represents about a 21% increase over his estimated prerace plasma volume of 2.8 L, which should, in turn, lead to a similar fall in his hematocrit and the concentration of plasma proteins. As reflected in Table 1, the actual changes in his hematocrit, total protein, and albumin varied slightly from these predicted values, but all followed the expected trend.

An excessive gain in plasma volume is problematic with exercise at high altitude. All individuals who ascend to high altitude experience alveolar hypoxia, which leads to hypoxic pulmonary vasoconstriction and, in turn, elevated pulmonary artery pressure. Exercise leads to a further rise in these pressures. In non-HAPE-susceptible individuals the rise in pulmonary arterial pressure usually does not cause a rise in capillary hydrostatic pressure sufficient to trigger frank pulmonary edema, although several studies suggest that some individuals develop asymptomatic mild interstitial fluid accumulation (8,17). Our patient used a very aggressive sodium-loading regimen (1000 mEq·d−1), which likely increased his plasma volume to a point where hydrostatic pressure was increased and oncotic pressures was decreased. In conjunction with the increased cardiac output and pulmonary blood flow associated with continuous submaximal exercise, these changes in the Starling forces caused fluid transit out of the vascular space to a degree that exceeded the normal active and passive reabsorptive processes in the lung. Although the altitude alone was probably not sufficient to cause pulmonary edema, the sodium-loading regimen and the subsequent changes in hydrostatic and oncotic pressure provided the necessary conditions for his respiratory decompensation.

Although appropriate sodium replacement is important to maintain adequate intravascular volume and ensure optimal performance, the literature contains little evidence to guide fluid and sodium replacement in ultraendurance athletes. Most studies focus on the incidence of hyponatremia and identify excessive water intake as a major risk factor for its development (1,25,27). Inadequate sodium intake during heavy exercise with associated sweating loss can also contribute to the development of hyponatremia, but information is lacking on how much sodium must be ingested to avoid this problem. Several case reports have documented sodium intake in individual athletes who completed multiday ultraendurance races (10,15). Lindeman (15), for example, described an individual who finished the Bicycle Race Across America and consumed an average of 10.99 g (478 mEq) of sodium per day, and Eden and Abernethy (10) describe an individual who consumed only 5.8 g (252 mEq) of sodium per day while finishing the Sydney to Melbourne Footrace. In one of the few systematic studies on this issue, Glace et al. (12) examined 26 participants in a 160-km ultramarathon and found that the finishers had consumed an average of 1.6 ± 0.7 mg·kg−1·km−1 of sodium during the race, whereas nonfinishers had consumed only 0.94 ± 0.6 mg·kg−1·km−1. There were no statistically significant body weight changes between the two groups, but the finishers were noted to have a 3% increase in plasma volume after 90 km of racing, whereas nonfinishers lost an average of 5% of their plasma volume. Published recommendations (7,14) stipulate that ultraendurance athletes should ingest 1 g·h−1 of sodium, but these recommendations are not based on systematic studies examining the adequacy of this regimen.

Our case report adds to this small body of available literature and suggests that, although sodium repletion is necessary to ensure adequate performance, prevent hyponatremia, and maintain adequate plasma volume, overly aggressive replacement regimens, particularly during exercise at high altitude, can lead to edema formation and may predispose to development of acute pulmonary edema.


1. Almond, C. S., A. Y. Shin, E. B. Fortescue, et al. Hyponatremia among runners in the Boston Marathon. N. Engl. J. Med. 252:1550-1556, 2005.
2. Anholm, J. D., E. N. Milne, P. Stark, J. C. Bourne, and P. Friedman. Radiographic evidence of interstitial pulmonary edema after exercise at altitude. J. Appl. Physiol. 86:503-509, 1999.
3. Armstrong, L. E., D. L. Costill, and W. J. Fink. Changes in body water and electrolytes during heat acclimation: effects of dietary sodium. Aviat. Space Environ. Med. 58:143-148, 1987.
4. Armstrong, L. E., D. L. Costill, W. J. Fink, et al. Effects of dietary sodium on body and muscle potassium content during heat acclimation. Eur. J. Appl. Physiol. Occup. Physiol. 54:391-397, 1985.
5. Bartsch, P., M. Maggiorini, H. Mairbaurl, P. Vock, and E. R. Swenson. Pulmonary extravascular fluid accumulation in climbers. Lancet 360:571, 2002.
6. Bartsch, P., H. Mairbaurl, M. Maggiorini, and E. R. Swenson. Physiological aspects of high-altitude pulmonary edema. J. Appl. Physiol. 98:1101-1110, 2005.
7. Clark, N., J. Tobin Jr., and C. Ellis. Feeding the ultraendurance athlete: practical tips and a case study. J. Am. Diet Assoc. 92:1258-1262, 1992.
8. Cremona, G., R. Asnaghi, P. Baderna, et al. Pulmonary extravascular fluid accumulation in recreational climbers: a prospective study. Lancet 359:303-309, 2002.
9. Dehnert, C., E. Grunig, D. Mereles, N. von Lennep, and P. Bartsch. Identification of individuals susceptible to high-altitude pulmonary oedema at low altitude. Eur. Respir. J. 25:545-551, 2005.
10. Eden, B. D., and P. J. Abernethy. Nutritional intake during an ultraendurance running race. Int. J. Sport Nutr. 4:166-174, 1994.
11. Gabry, A. L., X. Ledoux, M. Mozziconacci, and C. Martin. High-altitude pulmonary edema at moderate altitude (< 2,400 m; 7,870 feet): a series of 52 patients. Chest 123:49-53, 2003.
12. Glace, B. W., C. A. Murphy, and M. P. McHugh. Food intake and electrolyte status of ultramarathoners competing in extreme heat. J. Am. Coll. Nutr. 21:553-559, 2002.
13. Grunig, E., D. Mereles, W. Hildebrandt, et al. Stress Doppler echocardiography for identification of susceptibility to high altitude pulmonary edema. J. Am. Coll. Cardiol. 35:980-987, 2000.
14. Hiller, W. D. B. Dehydration and hyponatremia during triathlons. Med. Sci. Sports Exerc. 21:S219-S221, 1989.
15. Lindeman, A. K. Nutrient intake of an ultraendurance cyclist. Int. J. Sport Nutr. 1:79-85, 1991.
16. Luetkemeier, M. J. Dietary sodium intake and changes in plasma volume during short-term exercise training. Int. J. Sports Med. 16:435-438, 1995.
17. Mason, N. P., M. Petersen, C. Melot, et al. Serial changes in nasal potential difference and lung electrical impedance tomography at high altitude. J. Appl. Physiol. 94:2043-2050, 2003.
18. McKechnie, J. K., W. P. Leary, T. D. Noakes, J. C. Kallmeyer, E. T. MacSearraigh, and L. R. Olivier. Acute pulmonary oedema in two athletes during a 90-km running race. S. Afr. Med. J. 56:261-265, 1979.
19. McKenzie, D. C., T. J. O'Hare, and J. Mayo. The effect of sustained heavy exercise on the development of pulmonary edema in trained male cyclists. Respir. Physiol. Neurobiol. 145:209-218, 2005.
20. Neumayr, G., R. Pfister, H. Hoertnagl, G. Mitterbauer, W. Prokop, and M. Joannidis. Renal function and plasma volume following ultramarathon cycling. Int. J. Sports Med. 26:2-8, 2005.
21. Reineck, H. J., J. H. Stein, and D. W. Seldin. Integrated responses of the kidney to alterations in extracellular fluid volume. In: The Kidney: Physiology and Pathophysiology, D. W. Seldin and G. Giebisch (Eds.). New York: Raven Press, 1985, pp. 1137-1161.
22. Rowell, L. B. Human cardiovascular adjustments to exercise and thermal stress. Physiol. Rev. 54:75-159, 1974.
23. Singh, I., C. C. Kapila, P. K. Khanna, R. B. Nanda, and B. D. Rao. High-altitude pulmonary oedema. Lancet 191:229-234, 1965.
24. Sophocles, A. M. Jr., and J. Bachman. High-altitude pulmonary edema among visitors to Summit County, Colorado. J. Fam. Pract. 17:1015-1017, 1983.
25. Speedy, D. B., T. D. Noakes, I. R. Rogers, et al. A prospective study of exercise-associated hyponatremia in two ultradistance triathletes. Clin. J. Sport Med. 10:136-141, 2000.
26. Speedy, D. B., T. D. Noakes, I. R. Rogers, et al. Hyponatremia in ultradistance triathletes. Med. Sci. Sports Exerc. 31:809-815, 1999.
27. Stuempfle, K. J., D. R. Lehmann, H. S. Case, et al. Hyponatremia in a cold weather ultraendurance race. Alaska Med. 44:51-55, 2002.


©2007The American College of Sports Medicine