Since its first description in 1985 (19), hyponatremia (plasma sodium concentration <135 mmol·L−1) has increasingly been recognized as a potentially serious complication of prolonged exercise, especially ultradistance triathlon and running races.
Estimates of the incidence of hyponatremia following such ultradistance races vary considerably. Speedy et al. (24) reported hyponatremia to be present in 9% of athletes requiring medical care after the 1996 New Zealand Ironman Triathlon. In the Hawaiian Ironman Triathlon, hyponatremia was reported in up to 29% of race finishers (8,9).
Exercise associated hyponatremia is a life-threatening condition (3,4,6,19,20,25,27,28). Mild symptoms of hyponatremia are nonspecific and include malaise, confusion, nausea, and fatigue. More specific symptoms include seizures, coma, and even death (3,4-7,19). However, not all athletes with chemically diagnosed hyponatremia present for medical care; this condition has been termed "asymptomatic hyponatremia" and may account for more than 50% of cases (9,21).
Despite the potential seriousness of exercise associated hyponatremia and an incidence of between 9 and 29% after ultradistance exercise, the etiology of the condition remains uncertain. Hiller (7) has proposed that hyponatremia is caused by large, unreplaced salt losses in sweat associated with dehydration. In contrast, others have documented weight gains in hyponatremic athletes after ultradistance races (1,4,19,21,24) or a marked water loss in recovery (12). Others have reported an inverse relationship between the postrace plasma sodium concentration and weight change after ultradistance races; the greater the weight loss the higher the plasma sodium concentration (12,21,23,24), indicating that dehydration causes hypernatremia. Hyponatremia was more likely in athletes who either gained or did not lose weight during these races. These observations suggest that fluid overload is the cause of the exercise associated hyponatremia. The reason why normal renal mechanisms would not appropriately cope with this fluid overload is uncertain: a decreased glomerular filtration rate (GFR) during ultradistance exercise has been postulated (12), as have inappropriately high arginine vasopressin (AVP) levels (4,12,28).
Accordingly, the aims of this study were to determine prospectively changes in body weight and plasma sodium concentrations and, hence, the incidence of hyponatremia in a large group of athletes competing in the New Zealand Ironman triathlon. We wished to relate changes in plasma sodium concentrations with weight changes during the race and the development of specific symptoms.
Ethical approval for this study was obtained from the North Health Ethics Committee. All 660 entrants in the 1997 New Zealand Ironman ultradistance triathlon were invited to participate in this study. Written informed consent was obtained from 605 athletes. Each athlete swam 3.8 km, cycled 180 km, and ran 42.2 km consecutively. The race began at 0700 h. The race takes between 8.5 to 16.5 h to complete. Ambient air temperature at 12 a.m. on race day was 21°C with a relative humidity of 91%. Water temperature was 20.7°C. Food and drink were freely available at support stations every 12 km on the cycle course and every 1.8 km on the run. Fluids available included water, Coca-Cola (Atlanta, GA), and a sports drink (Powerade).
Because of time constraints, it was not possible to weigh all athletes on the morning of the race. Instead all participating subjects were weighed wearing minimal clothing and without shoes at race registration 2 d before the race on calibrated Seca scales (Hamburg, Germany). In addition, 323 athletes were weighed again immediately before the race to establish the extent of any change in weight in the 2 d before the race.
Athletes were re-weighed either at the finish line, or on arrival at the medical tent adjacent to the finish line after finishing the race, wearing their running clothes and shoes. An adjustment for the weight of running shoes was made by weighing 18 athletes with and without running shoes at the end of the race and calculating the mean weight of a pair of running shoes after the race (0.78 kg, range 0.5-1.5, SD ± 0.3).
Blood was collected by routine venipuncture from all consenting athletes within 15 min of their finishing the race. Athletes were either in the sitting or supine postures for venipuncture. Urgent assays for plasma sodium concentration were carried out on-site on samples from 89 athletes presenting for medical care; samples collected from 284 well athletes were analyzed the following morning. The urgent assays were performed on-site using a Nova Ion Selective Electrode analyzer (Waltham, MA) on lithium-heparin anticoagulated samples. Assays performed the next day were carried out with a Hitachi 747 analyzer (Boehringer Mannheim, Mannheim, Germany), using standard methods and the manufacturer's reagents, on serum that had been collected into silicone-gel separator tubes and stored at 4°C after centrifugation within 1 h of collection. When the 89 samples analyzed on-site were re-analyzed the following morning using the Hitachi 747 analyzer, a calibration difference of 2 mmol·L−1 was detected between the on-site and the Hitachi analyzers. The results from the Hitachi analyzer were therefore scaled down by 2 mmol·L−1 to ensure comparability with the specimens analyzed on-site. Routine hematological assays for measurement of hemoglobin (Hb) and hematocrit (Hct) were carried out the next day with a Technicon H3 analyzer on EDTA-anticoagulated samples that had been stored at 4°C. Athletes diagnosed with symptomatic hyponatremia immediately had another blood specimen collected by venipuncture for measurement of AVP and aldosterone levels: these assays were performed on EDTA-anticoagulated plasma that was obtained by centrifugation, frozen on dry ice within 1 h of venipuncture, and stored frozen at −20°C or below until thawed for assay. Specimens were assayed for AVP concentration according to the method described by Inder et al. (11). Assays for aldosterone concentration were performed according to the method described by Lun et al. (14). Reference ranges for AVP were <9 pmol/L and for aldosterone were 100-800 pmol/L.
For the purposes of this study, mild hyponatremia was defined as a plasma sodium concentration of 130-134 mmol·L−1. Severe hyponatremia was defined as a plasma sodium concentration below 130 mmol·L−1. Athletes who presented to the medical tent with symptoms of hyponatremia and a plasma sodium concentration below 135 mmol·L−1 were categorized as "symptomatic" for hyponatremia. Hyponatremic athletes who did not present for medical care after the race were defined as "asymptomatic" for hyponatremia.
Tests for differences between groups with the use of categorical variables were carried out using Chi-squared tests; similarly continuous variables were tested by Student's t-tests. Relationships between variables were modeled using linear regression. Significance was defined as P < 0.05.
One hundred and fifteen athletes (including finishers and nonfinishers) needed medical care after the race. Twenty-six of these athletes (23%) were hyponatremic. Symptoms in the hyponatremic athletes included light-headedness, nausea and vomiting, malaise, exhaustion, altered mental status, tonic clonic seizures, and headache. Fourteen hyponatremic athletes needed subsequent transfer to a hospital emergency department for further observation and medical care; two were admitted to an intensive care unit, critically ill with hyponatremia.
There were 660 race entrants: of these, 605 (90%) consented to participate in the study. There was no difference in the age or gender of the consenting and nonconsenting athletes. Of the consenting athletes, 540 (89%) successfully completed the race compared with 35 of the 55 (64%) nonconsenters (P < 0.0001). The nonconsenting athletes completed the event more quickly (681 min vs 734 min, P < 0.001).
Registration weights were successfully collected in 579 subjects (96%) and postrace weights in 558 (92%). Three hundred and twenty-three athletes were weighed at race registration and again on the morning of the race: the difference in mean weight between race registration and race morning was 0.1 kg (70.5 ± 10.0 kg at race registration vs 70.6 ± 10.1 kg on race morning). Data on both registration and postrace weights were available for 543 subjects. Weights were significantly decreased at the end of the race (N = 543, mean % Δ weight = −4.1%, P < 0.0001). The range of percentage weight change was large (−12.6 to +5.9%); 27 athletes (5%) either maintained their weight or gained weight. Of these 27 athletes who maintained or gained weight, 18 had blood samples taken: 13 (72%) were hyponatremic (median plasma sodium concentration, 129 mmol·L−1; range, 116-134 mmol·L−1) and five (28%) were normonatremic (median plasma sodium concentration, 137 mmol·L−1; range, 135-140 mmol·L−1).
Blood samples for measurement of plasma sodium concentration were successfully collected in 373 athletes (61%). Sixty-five of these 373 (17%) athletes were hyponatremic. Full data on registration weights, postrace weights, and postrace plasma sodium concentration were available on 330 race finishers (55%) (292 men and 38 women); 58 (18%) of these race finishers were hyponatremic. The following analysis was carried out on these 330 race finishers with full data unless otherwise stated.
There was no significant difference in age, sex, and race times between those athletes in whom plasma sodium concentrations were or were not measured. The subjects who did not consent to venesection were significantly heavier (73.1 vs 70.7 kg; P = 0.007) but had the same relative weight loss during the race (4.1% for each group).
Table 1 lists the characteristics of athletes whose plasma sodium concentrations were measured and who either were or were not treated in the medical facility at the race finish. Female athletes were more likely to attend the medical tent than males (37% vs 16%, P = 0.002). There was no significant difference in weight loss between the two groups, but postrace plasma sodium concentrations were significantly higher in the well athletes (138 vs 136 mmol·L−1; P = 0.007).
Plasma sodium concentrations were significantly related to both absolute weight loss (β = −1.17, P = 0.0001) and relative weight loss (β = −76.6, P = 0.0001): athletes with the lower plasma sodium concentrations either gained weight or had the smallest weight loss (Fig. 1). There was no significant relationship between race or split times and plasma sodium concentration at the univariate level. Likewise there was no significant relationship between age and plasma sodium concentration.
Relative weight change over the race in athletes with severe hyponatremia (plasma sodium concentration <130 mmol·L−1) ranged from −2.4% to +5% (N = 11); eight of these athletes (73%) either maintained their weight or gained weight over the race. However, the relative weight change in athletes with mild hyponatremia (plasma sodium concentration 130-134 mmol·L−1) ranged from −9.2% to +2.2% (N = 47).
Analysis by gender showed females (N = 38) had significantly lower postrace plasma sodium concentrations (134 ± 5 vs 137 ± 3 mmol·L−1; P = 0.0001) than males (N = 292). Females also lost significantly less absolute (−1.6 ± 1.8 vs −3.0 ± 1.5 kg; P = 0.0001) and relative (−2.7 ± 3.1 vs −4.3 ± 2.3%; P = 0.0002) weight during the race. Seventeen of the 38 (45%) female race finishers were hyponatremic compared with 41 of the 292 (14%) male race finishers.
Of the 264 race finishers who did not seek medical care after the race and in whom plasma sodium concentrations were measured, 40 (15%) were hyponatremic, although only 4 (1.5%) had a plasma sodium concentration below 130 mmol·L−1. The symptomatic hyponatremic athletes (N = 18) were more likely than the asymptomatic hyponatremic athletes to be female (53% of those tested) than male (20% of those tested) (χ2 = 6.48, P = 0.01). However, no significant difference was found between registration weight, relative weight change, or Hct in symptomatic compared with asymptomatic hyponatremic athletes. Figure 2 shows the number of athletes who were either symptomatic or asymptomatic at different plasma sodium concentrations. All athletes with plasma sodium concentrations below 125 mmol·L−1 were symptomatic whereas the majority with plasma sodium concentrations >130 mmol·L−1 were asymptomatic.
Weight loss, both relative and absolute, during the race was significantly related to total race time (P = 0.02), and to swim (P = 0.0002), cycle (P = 0.001), and run (P = 0.001) times, but not to transition times (P = 0.0001). Weight loss was greatest in the slower athletes. Relative weight loss was also significantly related to the registration weight (P = 0.004).
Absolute weight change was predicted by a multivariate model including the athlete's sex, time taken on the cycling leg, and the time for the transition between swim and bike. Percentage weight change was predicted by the same variables as absolute weight change with the addition of the competitor's prerace registration weight. The coefficients for these models are shown in Table 2.
The best multivariate model to describe postrace sodium concentrations contained the variables for age, sex, percentage weight loss, and running time, even though times did not show any significant effect at the univariate level. The terms for absolute weight loss and percentage weight loss were not significantly different in their predicting power; hence either could be used in the above model. The coefficients for both these models are shown in Table 3. Subgroup analysis by sex was not carried out because of the relatively small number of female athletes.
The postrace plasma sodium concentration was significantly related to Hct (P = 0.0001), so that for each unit increase in Hct there was a 0.38 mmol·L−1 increase in plasma sodium concentration (Fig. 3). This association is probably overestimated because of the artifact in Hct measurement introduced by the suspension of red cells in fluid of standard osmolality for volume measurement in vitro. The size of this effect is not precisely known, but the change in Hct exceeds its theoretical size by an order of magnitude, and we can be confident that the Hct change truly reflects hemodilution. Hct was inversely related to both absolute weight loss (P = 0.0001) and % weight loss (P = 0.0001).
Specimens were available for assay of AVP and aldosterone concentrations on 18 athletes with symptomatic hyponatremia. The relationship of AVP and aldosterone concentrations to plasma sodium concentration are shown in Figures 4 and 5, respectively. The relationship of AVP concentration to relative weight change is shown in Figure 6. AVP concentrations in hyponatremic athletes were all below 6.5 pmol·L−1 except for one athlete: this athlete was a 23-yr-old male who completed the race in 9.8 h, losing 3 kg of weight (5.6% loss of body weight) and with a postrace plasma sodium concentration of 131 mmol·L−1. His AVP level was markedly elevated at 59 pmol·L−1. (He was subsequently diagnosed with a nonfunctioning anterior pituitary microadenoma.)
This study represents the first report of changes in plasma sodium concentrations and in body weight in a large number of endurance athletes. It therefore allows the more accurate determination of the prevalence of changes in plasma sodium concentrations, especially of hyponatremia, and the relationship of hyponatremia to changes in body weight during the ultradistance triathlon. Previous reports have been either case descriptions of persons with symptomatic hyponatremia or have studied smaller numbers of athletes (9,19,21,24).
This study confirms the high incidence (18%) of hyponatremia at an ultradistance triathlon. In addition, symptomatic hyponatremia was present in 23% of athletes seeking medical care; a similar figure has been reported from the Hawaiian Ironman (5). The incidence of asymptomatic hyponatremia in our study (15%) is comparable with other reported incidences of between 10 and 44% (9,10,21). But only 1.5% of asymptomatic athletes in our study had a plasma sodium concentration of less than 130 mmol·L−1, suggesting that whereas mild hyponatremia (plasma sodium concentration between 130-134 mmol·L−1) may often be asymptomatic, more severe hyponatremia (plasma sodium concentration <130 mmol·L−1) is usually associated with significant symptoms.
An important aim of this paper was to investigate the etiology of hyponatremia. Postrace plasma sodium concentration was inversely related to relative weight change during the race (Fig. 1) as repeatedly shown by other investigators (12,17,21,23,24), confirming that athletes with hypernatremia are dehydrated whereas athletes with severe hyponatremia are usually overhydrated. The relationship between postrace plasma sodium concentration and relative change in body weight in our study was highly significant at the univariate level and in a multivariate model (Table 3). The majority (73%) of athletes with severe hyponatremia either gained weight or maintained their prerace weight during the race. Thus, it is clear that athletes who develop severe symptomatic hyponatremia are invariably suffering from fluid overload, confirming previous findings (1,4,19,24).
If severe hyponatremia is indeed a result of fluid overload, an unresolved issue is the location of the retained fluid. There are two possibilities: unabsorbed fluid may pool in the gut acting as a third space with movement of sodium ions from the extracellular fluid into this third space (1,15); alternatively, the fluid may be retained in the extracellular space because of altered renal function, causing a dilutional hyponatremia. Our data demonstrate that the postrace plasma sodium concentration was significantly related to Hct (Fig. 3); athletes with the lower plasma sodium concentrations had the lower Hct, indicating that the hyponatremia was a result of an expansion of the extracellular space. Hence these data support others (4,20,28) suggesting that exercise associated hyponatremia is probably a dilutional hyponatremia and not primarily from pooling of fluid in the gut (1,15).
Less well defined is the relationship of mild hyponatremia to fluid status. There was a wide range of relative weight changes in this group; some athletes were clearly overhydrated, while others were significantly dehydrated (Fig. 1). This raises the possibility that there may be two separate etiological mechanisms for mild hyponatremia: fluid overload, on the one hand, and large salt losses in the sweat, on the other, or indeed a combination of the two mechanisms (21).
Another important novel finding of this study was that females were at significantly increased risk for the development of symptomatic hyponatremia. Women had a significantly lower plasma sodium concentration than male athletes, lost less weight than men, and were more likely to be symptomatic for hyponatremia. Noakes (17) has suggested that the smaller size of women may explain their greater risk for the development of hyponatremia because they are more likely to maintain body weight during the race. Ayus has reported that 97% of patients who develop brain damage or die from postoperative hyponatremia are female (2); the relative risk for severe hyponatremic sequelae being greater in menstruant than in postmenopausal women.
Previous investigators (18,19,27) have noted that slower runners rather than faster runners develop hyponatremia. Noakes has postulated that slower runners exercise at a lower intensity with a lower fluid requirement, have greater opportunity to consume fluid, and thus are at risk of "over-drinking" and hyponatremia (18,19). Although postrace plasma sodium concentration was not significantly related to race times at the univariate level in our study, running time was a significant variable in the multivariate analysis for plasma sodium concentration, confirming other investigators' observations that a slow running time is a risk factor for the development of hyponatremia.
Despite the finding that some athletes (4.8%) either did not lose or indeed gained weight during the race, the mean relative weight loss over the race was large (4.1%) but similar to reports from other ultradistance races (13,16,21,23). Weight loss was significantly related to total race time, and each of the race "split" times, the slower athletes losing the most weight. Factors predicting the extent of both absolute and relative weight loss included a slow cycling time and male gender. However, the range of weight loss was large (Fig. 1).
Of particular interest in Figure 1 is the prediction that athletes who retained their body mass during the race would finish the race with borderline hyponatremia (plasma sodium concentration of 135 mmol·L−1). Furthermore, maintenance of a normal plasma sodium concentration requires that the athlete lose about 4% of body mass during the race. This is similar to the prediction from another study of ultradistance triathletes (21).
This surprising finding could be explained by the following possibilities: (i) the unreplaced loss of sufficient sodium during the race causes a dilutional hyponatremia even in dehydrated athletes with a contracted extracellular space. However, as argued elsewhere (16), dilutional hyponatremia of whatever cause indicates that the primary abnormality is in the regulation of the extracellular fluid volume; or (ii) the absolute weight loss during the triathlon is not an exact measure of changes in body fluid status. For example, Rogers et al. (22) have argued that a weight loss of 4% of body weight can occur in an ultradistance triathlon without the development of dehydration. This weight loss includes the loss of metabolic fuel oxidized during the race, as well as the release of a substantial volume of water stored with glycogen. Neither of these sources of weight loss during the race contribute to any dehydration during the race. In addition, metabolic water is generated during the race.
The accuracy of these predictions is highly dependent on the postulated presence of a large volume of fluid stored with glycogen and released during the race. While the evidence for that store of fluid is less than completely convincing, our finding that athletes are unable to maintain their plasma sodium concentrations unless they lose a substantial amount of weight during the race does suggest that weight changes during ultradistance triathlons are not an exact measure of the changes in fluid status that have occurred.
A further important finding of our study is that plasma AVP concentrations were not elevated in the athletes with symptomatic hyponatremia, including those with weight gains. There was no apparent relationship between plasma aldosterone concentration and plasma sodium concentration. As discussed above, our data suggest that exercise associated hyponatremia is a dilutional hyponatremia. The reason why the body does not excrete the free water load in hyponatremic athletes has not been determined. Wolfson (26) has suggested that the answer is either that such large amounts of water have been ingested that "even normally functioning kidneys cannot excrete it rapidly enough (the equivalent of psychogenic polydipsia), or that renal free-water excretion has been limited by one or more factors (decreased glomerular filtration rate, impaired electrolyte transport in the loop of Henle or more distal diluting sites, or increased AVP effect)." Other investigators have also proposed that inappropriately high AVP levels are the cause of the failure of hyponatremic athletes to excrete this excess ingested fluid (4,12,27). However, we know of no studies that have prospectively investigated the role of AVP in exercise associated hyponatremia. Armstrong et al. (1) reported a 460% increase in AVP levels in a hyponatremic runner, despite marked fluid overload. Holtzhausen et al. (10) found no differences in AVP levels between collapsed runners and normonatremic controls; however, the two groups had similar numbers of hyponatremic athletes. Our data do not support the postulate that inappropriately high AVP concentrations are involved in the etiology on hyponatremia.
Hyponatremia is a common (18% of race finishers) and potentially serious medical complication of an ultradistance triathlon. Mild levels of hyponatremia (plasma sodium concentration 130-134 mmol·L−1) are usually asymptomatic, but most athletes with severe hyponatremia (plasma sodium concentration <130 mmol·L−1) seek medical care. Females are at significantly increased risk for hyponatremia. Fluid overload is the likely etiology of severe hyponatremia, and the hyponatremia appears to be dilutional. Inappropriately high AVP levels are not responsible for this dilutional hyponatremia. The etiology of mild hyponatremia is less clear since athletes can be either dehydrated or overhydrated. Ultradistance athletes need to be aware of the risks of overdrinking and hyponatremia as well as the risks of dehydration. Further research is needed to determine the etiology of mild hyponatremia, to determine why women are more at risk for this condition than men, and to determine appropriate fluid intakes and fluid composition during ultradistance exercise.
1. Armstrong, L. E., W. C. Curtis, R. W. Hubbard, R. P. Francesconi, R. Moore, and E. W. Askew. Symptomatic hyponatremia
during prolonged exercise in heat. Med. Sci. Sports Exerc.
2. Ayus, J. C. Postoperative hyponatremic encephalopathy in menstruant women. Ann. Intern. Med.
3. Backer, H., E. Shopes, and S. L. Collins. Hyponatremia
in recreational hikers in Grand Canyon National Park. J. Wilderness Med.
4. Clark, J. M. and F. J. Gennari. Encephalopathy due to severe hyponatremia
in an ultramarathon runner. West. J. Med.
5. Eichner, E. R., R. Laird, D. Hiller, E. Nadel, and T. D. Noakes. Hyponatremia
in sport: symptoms and prevention. Sports Sci. Exchange: Roundtable.
Gatorade Sports Sci. Institute 4:1-4, 1993.
6. Frizzell, R. T., G. H. Lang, D. C. Lowance, and R. Lathan. Hyponatremia
and ultramarathon running. JAMA
7. Hiller, W. D. B. Dehydration and hyponatremia
during triathlons. Med. Sci. Sports Exerc.
8. Hiller, W. D. B. The United States Triathlon Series. Medical Coverage of Endurance Athletic Events.
Columbus, OH: Ross Laboratories, 1988, pp. 80-81.
9. Hiller, W. D. B., M. L. O'Toole, F. Massimino, R. E. Hiller, and R. H. Laird. Plasma electrolyte and glucose changes during the Hawaiian Ironman Triathlon
. Med. Sci. Sports Exerc.
10. Holtzhausen, L. M., T. D. Noakes, B. Kroning, M. De Klerk, M. Roberts, and R. Emsley. Clinical and biochemical characteristics of collapsed ultramarathon runners. Med. Sci. Sports Exerc.
11. Inder, W. J., J. Hellemans, M. J. Ellis, M. J. Evans, J. H. Livesey, and R. A. Donald. Elevated basal adrenocorticotropin and evidence for increased central opiod tone in highly trained male athletes. J. Clin. Endocrinol. Metab.
12. Irving, R. A., T. D. Noakes, R. Buck, et al. Evaluation of renal function and fluid homeostasis during recovery from exercise-induced hyponatremia
. J. Appl. Physiol.
13. Laird, R. H. Medical care at ultraendurance triathlons. Med. Sci. Sports Exerc.
14. Lun, S., E. A. Espiner, M. G. Nicholls, and T. G. Yandle. A direct radioimmunoassay for aldosterone in plasma. Clin. Chem.
15. Noakes, T. D. Dehydration during exercise: what are the real dangers? Clin. J. Sport Med.
16. Noakes, T. D. Fluid replacement during exercise. Exerc. Sports Sci. Rev.
17. Noakes, T. D. Hyponatremia
of exercise. Insider News Sports Nutr.
18. Noakes, T. D. The hyponatremia
of exercise. Int. J. Sport Nutr.
19. Noakes, T. D., N. Goodwin, B. L. Rayner, T. Branken, and R. K. N. Taylor. Water intoxication: a possible complication during endurance exercise. Med. Sci. Sports Exerc.
20. Noakes, T. D., R. J. Norman, R. H. Buck, J. Godlonton, K. Stevenson, and D. Pittaway. The incidence of hyponatremia
during prolonged ultraendurance exercise. Med. Sci. Sports Exerc.
21. O'Toole, M. L., P. S. Douglas, R. H. Laird, and D. B. Hiller. Fluid and electrolyte status in athletes receiving medical care at an ultradistance triathlon
. Clin. J. Sport Med.
22. Rogers, G., C. Goodman, and C. Rosen. Water budget during ultra-endurance exercise. Med. Sci. Sports Exerc.
23. Speedy, D. B., R. G. D. Campbell, G. Mulligan, et al. Weight changes and serum sodium concentrations after an ultradistance
. Clin. J. Sport Med.
24. Speedy, D. B., J. G. Faris, M. Hamlin, P. G. Gallagher, and R. G. D. Campbell. Hyponatremia
and weight changes in an ultradistance triathlon
. Clin. J. Sport Med.
25. Surgenor, S. and R. Uphold. Acute hyponatremia
in ultraendurance-athletes. Am. J. Emerg. Med.
26. Wolfson, A. B. Acute hyponatremia
-endurance athletes. Am. J. Emerg. Med.
27. Young, M., F. Sciurba, and J. Rinaldo. Delirium and pulmonary edema after completing a marathon. Am. Rev. Respir. Dis.
28. Zelingher, J., C. Putterman, Y. Ilan, et al. Case series: hyponatremia
associated with moderate exercise. Am. J. Med. Sci.