Journal Logo

Clinical Sciences: Clinical Investigations

Serum Electrolytes in Ironman Triathletes with Exercise-Associated Muscle Cramping


Author Information
Medicine & Science in Sports & Exercise: July 2005 - Volume 37 - Issue 7 - p 1081-1085
doi: 10.1249/
  • Free


The ironman triathlon is an ultraendurance event, which includes a 3.9-km swim, 180.2-km bicycle ride, and 42.2-km run. Among the more common medical problems experienced by athletes participating in such races are skeletal muscle cramps (4), accounting for between 6% and 20% of the medical diagnoses encountered at ironman triathlon events (3,5). Exercise-associated muscle cramping (EAMC) can be defined as a “painful spasmodic involuntary contraction of skeletal muscle that occurs during or immediately after exercise” (13). Despite the high incidence of EAMC at these events, there has only been one published study of muscle cramping in triathletes (4), and no published studies have yet identified the cause of EAMC in triathletes.

It is still a popular belief that EAMC is caused by dehydration and imbalances in serum electrolyte concentrations (2,7,8,10,17). Recent studies fail to show an association between EAMC and dehydration and abnormalities in serum electrolyte concentrations in marathon runners (6,14). To our knowledge, no study has investigated whether alterations in serum electrolyte concentrations and pre- and postrace changes in body mass are associated with EAMC in cramping ironman triathletes.

Although the electromyographic (EMG) activity of cramping muscles has often been recorded in a laboratory setting, only one study has recorded the “baseline” EMG activity (the electrical activity in a muscle between bouts of acute cramping) of cramping and noncramping runners after a marathon. In this study from our laboratory, we have shown that baseline EMG activity was significantly increased in the cramping group compared with a noncramping control group after exercise (Nicol J., MPhil Sports Medicine thesis, 1996, University of Cape Town). There are no data on the baseline EMG activity of cramping ironman triathletes after exercise.

The principle aim of this study was to compare the serum electrolyte concentrations and pre- and postrace body mass changes in cramping and noncramping control athletes after an ironman triathlon. A secondary aim was to document the EMG activity of a cramping and control muscle from cramping ironman triathletes during recovery from an ironman triathlon.



All triathletes who registered for the 2000 South African Ironman Triathlon were considered as potential subjects. All triathletes (irrespective of their cramping history) were informed of the study and the testing procedures at registration. Triathletes who volunteered to participate were required to sign a written informed consent form. The Research and Ethics Committee of the Faculty of Health Sciences, University of Cape Town, South Africa, approved the study.

Subjects were weighed on the morning of the race before the start of the swimming leg (not more than 60 min before race start) using calibrated Adamlab JPS electronic scales (Scales Inc, Brackenfell, SA) that were placed on hard, flat surfaces. All the subjects were weighed in swimming costumes and without shoes. Body mass was corrected for this clothing (200 g) and calculated as net body mass (15). All other data were collected at the race finish.

Race finish.

All race finishers, irrespective of whether they were research subjects or not, were required to walk to the medical facility, which was situated 10 m from the race finish. All the subjects who signed the informed consent could then volunteer for further testing and a medical evaluation. Eleven triathletes suffered from acute EAMC at the time of entering the medical facility volunteered to take part in the study and were escorted to a specific area of the medical tent for investigation and medical treatment if necessary. These athletes formed the “cramping group” (CR, N = 11). The diagnostic criteria for EAMC were: 1) acute muscle pain without any history of an acute muscle tear, 2) a visibly contracted muscle (with or without fasciculation), and 3) symptoms present at the time of entry into the medical tent but that could have developed during the race. Noncramping athletes who volunteered for the study were matched for gender, race finishing time, and prerace body mass. They formed the control group (CON, N = 9). Gender and prerace body mass data were available on a computer spreadsheet at the finish of the race for the purposes of rapidly matching the control to the cramping athletes. However, due to constraints in matching suitable controls with the same gender, similar prerace body mass, and similar finishing times as the cramping athletes, two cramping subjects did not have matched control subjects.

On race day, the average temperature was 20.5°C and the maximum temperature was 23.9°C. The relative humidity averaged 68% for the day, with the average wind speed being 4.6 m·s−1. Sea temperature was recorded as 16°C.

All subjects were weighed on entering the medical tent (Adamlab JPS electronic scales, Scales Inc, Brackenfell, SA), and their race times were recorded. Triathletes were weighed in their running shorts but without shoes. Body mass was corrected for running kit (200 g) (16). Percent body mass loss was calculated as the difference between the prerace body mass and final body mass divided by the prerace mass and expressed as a percentage. This calculation was based on the assumption that percentage body mass change would reflect hydration status (15).

Cramping triathletes were asked to lie supine on an examination plinth and were examined by a medical doctor. EMG testing began immediately on all crampers unless acute medical treatment was required. Crampers requiring medical attention were first treated (mainly by passive stretching) until they were cramp- and pain-free, before they were tested. The control subjects were also asked to lie supine on examination plinths until their blood sample had been taken.

Surface electromyography (EMG).

EMG testing of the CR group consisted of placing two disposable pregelled EMG electrodes (The Prometheus Group ™, U.S.) onto the muscle bellies of the right triceps brachii muscle (the noncramping control muscle) and the most severely cramping lower limb muscle groups (either the quadriceps, hamstring, or calf muscle groups). The quadriceps electrode was placed obliquely over the right vastus medialis oblique muscle belly 5 cm above the superior pole of the patella, the hamstring electrode was placed over the right biceps femoris muscle at the site of greatest thigh girth and the calf electrode was placed over the right gastrocnemius muscle at the site of greatest calf circumference.

The skin over each site was prepared by shaving the hair with a disposable razor (Dahlhausen) in a 3-cm radius around each site. Any oil and dirt was removed with alcoholic swabs and cotton gauze swabs (Naturil, Smith and Nephew). The two electrodes were attached to a portable EMG machine (Pathway MR 20, The Prometheus Group ™, U.S.), which displayed the electrical activity of the muscle as a millivolt (mV) reading.

EMG data previously collected in this laboratory on noncramping control subjects during recovery from a 2-h treadmill run showed that recovery EMG activity was minimal, ranging between 0 and 2 mV (MSc thesis, N Sulzer, University of Cape Town 2003 and MPhil thesis, J. Nicol, University of Cape Town 1996). Based on this information, the present pilot EMG study did not measure any noncramping control subjects but did measure both a cramping and noncramping control muscle from cramping subjects during recovery. Surface EMG data from both the cramping and noncramping control muscles of the CR group were read from the millivolt reading on the display and recorded manually onto a prepared data collection form. Subjects were asked to lie as still as possible for the duration of the recording, which was started as soon as possible after subjects were supine on the examination plinths. EMG activity was monitored continuously over a 10-min period, but a reading was done for 1 s at the beginning of every minute for 10 min.

Postrace serum electrolyte, glucose, and hemoglobin concentrations and hematocrit.

A 4.5-mL postrace venous blood sample was drawn into lithium heparin Vacutainer tubes for hemoglobin, hematocrit, and serum magnesium and serum electrolyte determination. In addition, 2-mL venous blood was also drawn into sodium fluoride and potassium oxalate Vacutainer tubes for serum glucose analysis. A medical doctor drew all the blood samples from the right forearm antecubital vein of the subjects in both the CR and CON groups while subjects were lying in the supine position. All samples were drawn within 10 min of completing the race and entering the testing area. The CON group’s blood sample was taken as soon as they were supine on the examination plinths in the testing area of the medical tent, whereas the CR group’s blood sample was taken during the first break in EMG recording to prevent any interference in the triceps EMG recording. Two of the cramping subjects refused to have their blood taken; thus, the CR group’s sample size for the blood analysis was only nine.

A small volume (a few microliters) of the heparinized blood was used to measure the hematocrit immediately on site, in triplicate, as the packed cell volume. The micro capillary tubes were centrifuged in a Hawksley microhematocrit centrifuge for 5 min and read with a Hawksley microhematocrit reader. For the hemoglobin determination, 100 μL of the whole blood was diluted on site and stored at −20°C for further analysis. Thereafter, a standard cyanmethemoglobin procedure using a 1 in 500 dilution of heparinized whole blood with Drabkins solution was used to measure blood hemoglobin concentrations (1).

After removing small volumes of the whole blood for the determination of hematocrit and hemoglobin, the remaining whole blood samples were centrifuged on site at 3000 rpm for 10 min and the serum was stored at −20°C until later analysis of serum magnesium and electrolyte concentrations. Serum magnesium concentration was measured by atomic absorption spectrophotometry (Varian AA1275 atomic absorption spectrophotometer, Varian Techtron, Melbourne, Australia) after dilution of the serum 1:50 with 0.1% Lanthanum chloride solution containing 60 mmol·L−1 HCl. Serum sodium, potassium, and chloride concentrations were analyzed using an EasyLyte PLUS Na/K/Cl Analyzer (Medica Corporation, Bedford, MA). Blood glucose concentrations were analyzed from the blood collected in the sodium fluoride and potassium oxalate Vacutainer tubes, using a Beckman Glucose Analyzer 2 (Beckman Instruments, Inc., CA).

Medical discharge.

A medical doctor discharged those subjects with no further symptoms once all testing had been completed. Discharge criteria were no further cramps and no pain at rest or with movement.

Statistical analysis.

All statistical analyses were performed using Statistica Version 6.0 (© StatSoft, Inc., U.S.) on a personal computer. All data are represented by the mean ± standard deviation. Independent t-tests were used to evaluate differences between groups (CR and CON) and muscles (cramping and control). A two-way analysis of variance (ANOVA) for repeated measures was performed on all data to determine whether there was an interaction effect between group and time. Significant differences (P < 0.05) between groups and over time were further analyzed by means of a Tukey post hoc analysis to determine the site of significance. P values less than 0.05 were considered significant.


Descriptive data.

The average age (yr), pre- and postrace body mass (kg), percent body mass loss, and total race time (min) for the CR and CON group are presented in Table 1; there were no significant differences in any measures between the two groups.

Age, pre- and postrace body mass, percent body mass loss, and total race time in the CR (N = 11) and CON (N = 9) groups (mean ± standard deviation).

Surface electromyographic (EMG) data.

EMG (mV) activity (N = 11) for both the noncramping control (triceps) and cramping (calf, quadriceps, or hamstring) muscles of the CR group recorded from immediate recovery (Imm.Rec.) for 10 min during recovery is presented in Figure 1.

FIGURE 1—Surface EMG (mV) for both the noncramping control (
FIGURE 1—Surface EMG (mV) for both the noncramping control (:
N =11) and cramping ( N =11) muscles of the CR group during 10 min of recovery. Data presented as mean ± standard deviation. * Indicates a statistical significance ( P < 0.05) between cramping and control muscles of the CR group.

EMG activity was significantly greater (P = 0.04, 0.04, 0.002, 0.05, respectively) in the cramping compared with the control muscle at 0, 3, 4, and 5 min. There were no significant differences over time.

Postrace serum electrolyte, glucose, and hemoglobin concentrations and hematocrit.

The serum magnesium, glucose, sodium, potassium, and chloride concentrations (mmol·L−1) as well as blood hemoglobin concentration (g·dL−1) and hematocrit (%) for both the CR (N = 9) and CON (N = 9) groups are presented in Table 2.

Postrace serum magnesium, glucose, sodium, potassium, chloride concentrations, and blood hemoglobin concentration and hematocrit in the CR (N = 9) and CON (N = 9) groups (mean ± standard deviation).

The postrace serum sodium concentration was significantly higher (P = 0.01) in the CON group compared with the CR group. There were no other significant differences between the two groups.


The main findings of this study are that Ironman triathletes who suffer from EAMC did not suffer a significantly greater body mass loss during the race than did control triathletes who completed the race without developing EAMC. Similarly, affected triathletes did not have clinically significant differences in serum electrolyte concentrations compared with triathletes in the control group. Furthermore, a pilot study indicated that the EMG activity of cramping muscles in the cramping group is higher and more variable than that of noncramping muscles in the same individuals during recovery from fatiguing exercise.

There were no significant differences in percent body mass loss between the CR and CON groups during the race. Changes in prerace compared with postrace body mass were similar in both groups and were also similar to those previously reported for the total group of triathletes (reported in Sharwood et al.; 15). We, therefore, conclude that crampers did not experience a greater body mass loss than the control athletes and that this is indicative of similar levels of dehydration during the race in both groups. Thus, these results do not support the widely publicized hypothesis that EAMC is related to dehydration (2,8). Rather, the findings confirm the results of others showing that EAMC is not associated with dehydration in marathon runners (6,14) or in ironman triathletes (4). The results of the present study therefore add to the growing body of literature, which shows dissociation between dehydration and EAMC in athletes, irrespective of the type or duration of the activity.

One of the most commonly accepted hypotheses for the etiology of EAMC is an abnormality in serum electrolyte concentrations, in particular serum magnesium, sodium, potassium, and chloride concentrations, due to an inadequate replacement of electrolytes during exercise (2,7,10). However, two studies failed to find any differences in postrace serum electrolyte concentrations (sodium, potassium, calcium, phosphate, and serum magnesium) in cramping and control runners both before and after exercise (6,14).

In our study, the CR group had a statistically significant lower postrace serum sodium concentration. However, this difference is not clinically significant as the serum sodium concentrations in both groups fell within the standard norms (16,17). Fluid intake was not measured in our study, but we (15,16), and others (3), have previously shown that those athletes who increase their fluid intake during exercise are more likely to have lower serum sodium concentrations postexercise. Although speculative, the lower serum sodium concentration in the CR group may have been associated with increased fluid intake during exercise.

There were no significant differences in postrace serum magnesium, potassium, or chloride concentrations between the CR and CON group. There were also no significant intergroup differences in postrace glucose and hemoglobin concentrations or hematocrit. Thus, these findings do not support an association between EAMC and abnormalities in serum electrolyte concentrations and conflicts with the “serum electrolyte theory” of EAMC (6,14).

An increased EMG activity during an acute bout of induced and spontaneous cramping has been reported (9,11). In one study conducted in our laboratory, the baseline EMG activity (the electrical activity in a muscle between bouts of acute cramping) in cramping runners was increased immediately after intensive exercise (Nicol J., MPhil Sports Medicine thesis, 1996, University of Cape Town). In that study, there was a significant decrease in the cramping runners’ baseline EMG activity compared with the control runners over a 60-min recovery period from exercise. In our pilot study, there were no significant differences in either the cramping and noncramping control muscles over time, but higher EMG activity, reflecting increased neuromuscular excitability, was recorded in the cramping muscles at certain time intervals. Any longer term changes in EMG activity in the cramping and control muscles could not be assessed, because of the relatively short period during which EMG activity could be recorded. This short recording time was because subjects were reluctant to be tested for much longer than 10 min after the race.

A novel hypothesis suggests that the etiology of EAMC may be associated with muscle fatigue that alters the alpha motor neuron control at the spinal level, resulting in a heightened motor neuron activity that is reflected by an increase in the EMG activity of cramping muscles between bouts of acute cramping (13). The results of our pilot study showed that the EMG activity in the cramping muscle was consistently higher and more variable (as reflected by the large standard deviation) than that of the noncramping control muscle throughout the 10-min period and was significantly higher than the noncramping control muscle at 0, 3, 4, and 5 min. Although the findings have to be interpreted with caution, it would appear that they support the hypothesis that EAMC may be accompanied by a heightened muscle activity possibly associated with muscle fatigue. This however, requires further study.

The present study was unique in that EMG activity from both a cramping and noncramping control muscle from the same CR group was recorded. Of importance is that only those muscles directly involved in force generation during the triathlon developed muscle cramping. Altered serum electrolyte concentrations caused by systemic abnormalities can result in generalized skeletal muscle cramping (12,13). EAMC, however, occurs only in the localized muscle groups specifically involved in force-generation during exercise (12,13). The results from our study indicate a regional rather than a systemic cause for muscle cramping, a point that logically conflicts with both the “dehydration” and the “electrolyte depletion” theories of muscle cramping.

This case control study has some limitations. The sample size was small as a result of the small number of cramping triathletes on race day. The subjects for this study were part of a larger group of athletes where, as previously reported, and body mass was measured at the time of registration (24–72 h before the race) as well as on the morning of the race (prerace) (15). Because body mass was shown to increase from the time of registration to prerace in that study (15), we elected to use only the prerace body mass data and not the registration body mass in our study. A limitation of our study was that we did not record any causes for changes in body mass from the prerace body mass to the finish of the race. Causes for body mass changes during the race would include food or fluid intake, as well as fluid loss from sweat, urine, feces or respiration, or body mass loss from metabolic fuel. We, as others (15), made the assumption that percentage body mass change would approximate postrace hydration status. However, more to the point, body mass loss was similar in both groups and was relatively mild (about 3 kg). If dehydration was a significant factor in the etiology of EAMC, we would have expected much larger changes in body mass in the cramps group.

A further limitation of the study was the insufficient duration of recording the EMG data. The relatively short 10-min period limited the scope and nature of deductions that could be made from the pilot study of the EMG data. Finally, the EMG equipment used in this study, although portable, did not allow for the more detailed analyses of power frequency shifts and spectral changes that provide important insight into the relative fatigue state of the muscle.

In conclusion, the results of this study show that EAMC is not associated with higher levels of dehydration or clinically significant differences in serum electrolyte concentrations in ironman triathletes. In addition, results of the pilot study showed that increased EMG amplitudes recorded in the force-generating cramping muscles appear to support the hypothesis that EAMC may be accompanied by heightened neuromuscular activity possibly associated with muscle fatigue (13).


1. Drabkin, D. L., and J. H. Austin. Spectrophotometricstudies II: preparations from washed cells; nitric oxide, hemoglobin and sulfhemoglobin. J.Biol. Chem. 112:51, 1935.
2. Eichner, E. R. Heat cramps: salt is simplest, most effective antidote. Sports Med. Dig. 21:88, 1999.
3. Hiller, W. D., M. L. O’Toole, E. E. Fortess, R. H. Laird P. C. Imbert, and T. D. Sisk. Medical and physiological considerations in triathlons. Am. J. Sports Med. 15:164–167, 1987.
4. Kantorowski, P. G., W. D. Hiller, W. E. Garrett, P. S. Douglas, R. Smith, and M. L. O’Toole. Cramping studies in 2 600 endurance athletes. Med. Sci. Sports Exerc. 22:S104, 1990.
5. Laird, R. H. Medical care at ultra endurance triathlons. Med. Sci. Sports Exerc. 21:S222–S225, 1989.
6. Maughan, R. J. Exercise-induced muscle cramp: a prospective biochemical study in marathon runners. J. Sports Sci. 4:31–34, 1986.
7. McArdle, W. D., F. I. Katch, and V. L. Katch. Exercise Physiology: Energy, Nutrition, and Human Performance. Philadelphia: Lea & Febiger, 1991, p. 58.
8. Micheli, L. J., and M. Jenkins, The Sports Medicine Bible. New York: HarperCollins, 1995, p. 117.
9. Norris, F. H., E. L. Gasteiger, and P. O. Chatfield. An electromyographic study of induced and spontaneous muscle cramps. EEG Clin. Neurophysiol. 9:139–147, 1957.
10. Read, M., and P. Wade. Sports Injuries: A Unique Guide to Self Diagnosis and Rehabilitation. Cape Town, South Africa: Vlaeberg Publishers, 1994, p. 32.
11. Ross, B. H., and C. K. Thomas. Human motor unit activity during induced muscle cramp. Brain 118:983–993, 1995.
12. Schwellnus, M. P. Skeletal muscle cramps during exercise. Physician Sportsmed. 27:109–115, 1999.
13. Schwellnus, M. P., E. W. Derman, and T. D. Noakes. Aetiology of skeletal muscle “cramps” during exercise: a novel hypothesis. J. Sports Sci. 15:277–285, 1997.
14. Schwellnus, M. P., J. Nicol, and R. Laubscher. Serum electrolyte concentrations and hydration status are not associated with Exercise Associated Muscle Cramping (EAMC) in distance runners. BJSM 38:488–492, 2004.
15. Sharwood, K., M. Collins, J. Goedecke, G. Wilson, and T. D. Noakes. Weight changes, sodium levels, and performance in the South African Ironman Triathlon. Clin. J. Sports Med. 12:391–399, 2002.
16. Speedy, D. B., T. D. Noakes, I. R. Rogers, J. M. Thompson, R. G. Campbell, and J. A. Kuttner. Hyponatraemia in ultra distance triathletes. Med. Sci. Sports Exerc. 31:809–815, 1999.
17. Tortora, G. J., and S. R. Grabowski. Principles of Anatomy and Physiology. New York: HarperCollins, 1993, pp. 908–911.


©2005The American College of Sports Medicine