The aims of this study were to report normative data on regional sweat-sodium concentrations of professional male team-sport athletes, to compare sweat-sodium concentrations among sports, and to examine the relationships between sweat-sodium concentrations and various self-reported measures. The novel findings are that professional athletes competing in American football, baseball and basketball exhibit significantly greater regional sweat-sodium concentrations than athletes competing in soccer or rugby. Furthermore, we report strong positive correlations between sweat-sodium concentrations and self-reported sodium losses across all sports.
An earlier retrospective analysis of team-sport athletes found that the sweat-rates of baseball, basketball and American football players were greater than that reported for soccer , although the sweat-sodium concentrations of each independent sport were not reported. The present study is the first, therefore, to report significant differences in the sweat-sodium concentrations of professional athletes training in North America (American football, basketball, baseball) relative to those training in the United Kingdom (soccer and rugby). The differences observed are unlikely due to the greater heat-acclimatisation status of athletes in North American sports, since acclimatisation has been shown to reduce the sweat-sodium concentration , due primarily to the reabsorption of sodium through the sweat duct . Moreover, the sports that were evaluated comprise athletes from a myriad of ethnic backgrounds, which discounts ethnicity as a primary explanation for the differences observed between groups. Other factors including age, body-composition, training status, and sweat rate have no association with sweat-electrolyte content , although this latter marker has been contested . Aldosterone is supressed in response to a sodium-rich diet, which can account for higher sweat-sodium concentrations [26, 27]. Given that dietary sodium, as assessed via urinary excretion, is higher in the United States compared to the United Kingdom , the sweat-sodium concentration differences between groups might be related to dietary sodium, although, this was not currently measured.
In 696 professional male team-sport athletes, we report regional mean sweat-sodium concentrations of 46.4 ± 13.6 mmol.L−1, which is in accordance with values previously reported for elite male athletes (48.2 ± 18.1 mmol.L−1) . Furthermore, the values we report are congruent with values observed in soccer (43.2 versus 42.5 mmol.L−1, ref. ), rugby (44.0 versus 41.9 mmol.L−1 ref. ), and American Football (54.0 versus 50.3 mmol.L−1, ref. ). Although Baker et al.  report mean data encompassing several sports (including baseball and basketball), data regarding regional sweat-sodium concentrations for individual sports were not reported. As such, to our knowledge, we are the first to report normative values for professional baseball and basketball.
We also observed a strong positive correlation between sweat-sodium concentration and self-reported sodium loss. This suggests that self-reported measures of sodium loss might be a good predictor of actual sodium loss. The relationship is likely to be attributable, at least in part, to greater accuracy in the reporting of these subjective measures due to the visual prompting (e.g., salt residue on training apparel) and the recall of sensory experiences (e.g., stinging eyes and salt-taste in sweat) utilised in the questionnaire. Although we did not measure sweat rates in our study, others have found significant correlations for self-reported sweat rates with actual sweat rates in male basketball players . Collectively, these findings suggest that both self-reported sodium losses and self-reported sweat rates might serve as an effective surrogate in the absence of more direct measures.
In contrast, we found that only the rugby sample exhibited a weak positive correlation between sweat-sodium concentration and self-reported cramping-frequency (r s = 0.184, p = 0.015). Given the poor correlation coefficient, and the fact that no such correlations were observed in soccer, American football, or basketball, the relationship between sweat-sodium concentration and self-reported cramping-frequency is unlikely to be meaningful. Moreover, the causes of cramp are multi-factorial  and, therefore, evidence for a causative relationship between electrolyte depletion and cramping remains equivocal.
Based on the data collected in this study, we report large inter-sport and inter-athlete variations in sweat-sodium concentration (range, 17–92 mmol.L−1). Moreover, athletes competing in baseball, American football, and basketball, respectively, exhibit higher sweat-sodium concentrations relative to rugby and soccer. Since the sodium content of many commercial sports drinks ranges from 10 to 25 mmol.L−1 [31, 32], it is likely that the drinks habitually consumed by competitive athletes contain sodium concentrations that are insufficient to replace sodium losses. Our findings, therefore, suggest that sodium concentrations substantially greater than 25 mmol.L−1should be consumed, but concentrations greater than 50 mmol.L−1 should be avoided as this will compromise drink palatability . Moreover, special consideration should be given to the fluid volume and sodium concentration required to facilitate recovery following high-intensity or prolonged exercise. Many team-sport athletes have dense fixture lists and daily training and, therefore, normal eating and drinking practices are likely to be insufficient to replenish fluid and electrolyte balance for full recovery . Therefore, aggressive fluid and electrolyte intake should be encouraged, particularly for athletes exhibiting the greatest sweat-sodium losses.
It also appears that self-reported sodium losses might be useful in estimating sweat-sodium concentrations when the latter cannot be measured directly. Therefore, athletes are encouraged to report their perceived sodium losses during exercise, making use of appropriate visual prompts (e.g., identifying sodium residue on training apparel) and sensory recall (e.g., experiencing the discomfort of stinging eyes, tasting salt in sweat).
There are several important considerations that should be made when interpreting the data presented in this study. First, it is well-established that there exist regional differences in sweat-electrolyte content , and regional sweat collection techniques have been shown to overestimate sweat-sodium concentrations when compared to whole-body values [36–38]. It is, therefore, common to collect sweat samples from various body locations (e.g., forearm, upper-back, chest, thigh) and combine measures into an arithmetic mean, or use a weighting-factor to account for regional differences. Regression equations have also been used to adjust raw values for regional sweat-sodium data; however, such correction equations were derived from sweat-patch tests, whereas the published literature has employed a range of different techniques in the collection of regional sweat-sodium concentrations, which might compromise the validity of their reported measures. Since the techniques employed presently are reliable and have been adequately validated (see Methods), we opted to report data for regional concentrations. Care should be taken, however, when extrapolating our findings to studies that have used different techniques. Second, the purpose of this paper was to report normative data on regional sweat-sodium concentrations in a large cohort of professional team-sport athletes. In addition to data on sweat-sodium concentrations, it should be noted that effective fluid and electrolyte replacement strategies cannot be determined without additional measures of whole-body sweat-rate; however, this can be estimated from pre-to-post-exercise changes in body mass. Third, heat acclimation status has been shown to influence sweat-sodium concentration , but could not be accounted for in the present observational study. Finally, this study induced the sweat response via pilocarpine iontophoretic discs applied to the skin, which has been shown to reflect heat- and exercise-induced sweating capacity, but is yet to be validated against the maximal and whole-body sweat response .