Skeletal muscle cramps that occur during or shortly after exercise in healthy individuals are termed exercise-associated muscle cramps (EAMC). They are highly prevalent in competitive athletes (3,10,39) and physically active populations (29). These cramps often occur 0.5-1 h after exercise (8), although symptoms of EAMC have also been reported occurring up to 8 h after activity (6,11). Moreover, EAMC seem to be chronic in some individuals (2,8) and have a high possibility of recurrence during exercise (33).
Despite the commonality and prevalence of EAMC, controversy exists regarding their etiology. Some scientists postulate that EAMC are due to dehydration and electrolyte losses (4,14). These scientists propose that dehydration causes a contracture of the interstitial space, which increases the mechanical pressure on nerve terminals, thus causing cramp (4,22). To support this hypothesis, some scientists cite the loss of substantial amounts of fluid and electrolytes in athletes who develop EAMC (3,4,35). Others, however, believe that EAMC are due to neuromuscular fatigue, creating an imbalance between excitatory and inhibitory stimuli acting at the alpha motor neuron (31,32). Support for this theory comes from observations of athletes developing EAMC while contracting their muscles in already-shortened positions (32) and after exercising for long periods (3,19,33). Moreover, athletes who develop EAMC often have similar hematological characteristics to noncrampers (24,33,39), and EAMC can be relieved by moderate static stretching of cramping muscles (23,33,38) or activation of Golgi tendon organs (21). Neither of these treatments directly affects the interstitial compartment or hydration level of the individual, yet they adequately relieve a skeletal muscle cramp. The disparity in findings has led some researchers to propose that there may be different types of muscle cramps (i.e., those that occur locally vs those that are generalized) (4). Overall, neither theory accounts for every observation surrounding the occurrence of EAMC.
The lack of controlled randomized experimental research is one reason for the diversity of theories regarding the etiology of EAMC. Moreover, the unpredictability and spontaneity of EAMC have made studying their etiology difficult. To help solve these problems, various models to induce cramps in laboratory settings have been used (9,18,37). Percutaneous electrical stimulation of the tibial nerve is one method of inducing skeletal muscle cramps and is reliable (25,37), generally well tolerated (25,37), and highly associated with the occurrence of EAMC (26). With this model, the tibial nerve is stimulated with low-frequency trains of electrical stimuli until the flexor hallucis brevis muscle cramps. The electrical stimulation frequency at which the muscle cramps is termed "threshold frequency." Because lower cramp threshold frequencies are associated with the occurrence of EAMC (26), threshold frequency may be used as a quantitative measurement of cramp susceptibility (i.e., lower cramp threshold frequencies indicate an increased predisposition and vice versa) (20,34).
Finally, little research has been done exploring the best method for inducing the longest-lasting, most intense electrically induced muscle cramps. Previous investigations of electrically induced muscle cramp duration indicate a duration ranging from 12 to 40 s (37,41). Such short durations limit the investigation of cramp treatment efficacy because it would be unclear if the cramp resolved spontaneously or because of the treatment. Because of this limitation, research that uses the electrically induced cramp model to investigate the effectiveness of cramp treatments usually uses a pre-post experimental design, with threshold frequency as the dependent variable (20,34). The limitation with this design is that researchers can only examine an intervention's preventative effect rather than its treatment effect. Therefore, we questioned if cramp duration could be lengthened and used as the dependent variable to determine treatment efficacy (i.e., shorter cramp durations would be associated with a treatment effect). Previous research in our laboratory suggests that stimulating the tibial nerve above an individual's threshold frequency is one method of eliciting longer and more intense electrically induced muscle cramps (41). If hypohydration increases the excitability of nerve terminals, as some scientists hypothesize (4,22), then dehydrating subjects in future experiments may also be used to improve the length and intensity of cramps examined. However, no data that explore the relationship between hypohydration and electrically induced muscle cramp duration or intensity have been reported.
Therefore, the purpose of our study was twofold. First, we sought to determine whether mild hypohydration (∼3%) and electrolyte losses decrease cramp threshold frequency. To help isolate the effects of dehydration from fatigue on cramp susceptibility, we dehydrated subjects but induced cramps in a rested, unexercised muscle. Second, we wanted to know if mild hypohydration induces longer and more intense electrically induced muscle cramps. Increasing cramp duration would allow scientists to better study the effectiveness of cramp interventions. We hypothesized that mild hypohydration would decrease cramp threshold frequency and increase the intensity and duration of electrically induced muscle cramps.
MATERIALS AND METHODS
Sample size was estimated a priori using the following equation (40):
where n is the number of subjects needed, SD is the standard deviation of the groups (assuming equal variance), Zα is the z-score of α level 0.05 (i.e., 1.96), Zβ is the z-score for 80% power (i.e., 0.84), and Δ is the hypothesized difference between euhydrated and hypohydrated conditions. For this experiment, it was hypothesized that there would be a difference of 8 Hz between cramp threshold frequencies and that there would be an SD of 6 Hz (on the basis of SD of a previous experiment using threshold frequency) (25). On the basis of the above information, it was determined that nine subjects would be needed to achieve 80% power.
Twelve healthy, college-aged males volunteered for this study. Two subjects were excluded from participating because we could not induce a cramp in the flexor hallucis brevis during the familiarization session. Therefore, 10 males (mean ± SE; age = 23.5 ± 1.0 yr, height = 177.8 ± 1.8 cm, mass = 73.9 ± 2.8 kg) completed this study. Subjects were excluded from participating if they 1) had experienced any lower extremity injury or surgery within the 6 months before the study; 2) did not have a history of EAMC in the 6 months before the study; 3) could not be cramped during the familiarization session; or 4) self-reported any neurological, cardiovascular, or blood-borne diseases. The procedures were approved by our university's institutional review board, and subjects provided written informed consent before participating.
Subjects reported for a familiarization session and 2 d of testing. No data were collected from the familiarization session; it was simply used to screen potential subjects to ensure that an electrically induced cramp could be induced in the flexor hallucis brevis and subjects could tolerate the electrical stimuli. The familiarization session occurred 24 h before the first experimental day for all subjects.
For the familiarization session, standard EMG preparatory procedures (37) were performed on the medial plantar aspect of the foot, area around the medial malleolus, and ipsilateral tibial tuberosity. Two EMG measurement electrodes were placed 2 cm apart over the midbelly of the flexor hallucis brevis with a single ground measurement electrode over the ipsilateral tibial tuberosity. Subjects were taught how to perform a maximum voluntary isometric contraction (MVIC) with their dominant leg's flexor hallucis brevis by performing 15 practice MVIC. Each 2-s MVIC was separated by 1 min of rest.
After the last practice MVIC, subjects rested for 15 min. We then attempted to induce a cramp in the flexor hallucis brevis via low-frequency percutaneous electrical stimulation of the tibial nerve. An 8-mm Ag/AgCl-shielded electrode was placed over the medial ankle, and an 8-cm square dispersive electrode was placed over the lateral malleolus. The tibial nerve was then submaximally stimulated two to four times with 1-ms electrical stimuli at 80 V to determine the site around the medial malleolus, which caused the greatest hallux flexion. On achieving the locations of the electrodes to elicit hallux flexion, the electrodes were secured with medical tape and an elastic wrap at these locations.
The tibial nerve was then stimulated with two consecutive trains of electrical stimuli (1 train per second; no rest intervals between trains) beginning at a train frequency of 4 Hz (eight total stimuli on the first trial). If a cramp did not occur at 4 Hz, subjects rested for 1 min, and train frequency was increased by 2 Hz. This process continued until the flexor hallucis brevis cramped.
A muscle cramp was defined as an involuntary, painful contraction of the flexor hallucis brevis immediately after electrical stimulation and was verified by involuntary, sustained hallux flexion, subject perception that a cramp had occurred, and an average EMG root mean square amplitude >2 SD above the 1-s baseline EMG average root mean square amplitude (37). To ensure that the induced cramps were moderately intense, cramp duration must have been ≥90 s and have an intensity ≥50% of MVIC EMG activity. The electrical stimulation frequency that induced cramps meeting these criteria was considered the subject's threshold frequency.
On cramp identification, EMG and stimulating electrode sites were marked for replication, and subjects were invited back the following day for the first experimental day of testing. Subjects were instructed to fast for 12 h before the experimental days, to drink water consistently throughout the evening and morning before these days, and to avoid exercising for 24 h before testing.
On the first experimental day, subjects reported to the laboratory, voided their bladders, and were weighed. A sterile single-use venous catheter was inserted into a superficial vein in the forearm, and subjects ingested 5 mL·kg−1 body weight of tap water within 5 min to help ensure hydration. Subjects lay supine for 30 min to allow fluid absorption and body compartment equilibration. During this time, a portion of the right midforearm was shaved in preparation for sweat patch placement. Subjects voided their bladders, were weighed, and had their dominant leg prepared for EMG analysis (37). They lay supine for an additional 30 min after which they performed fifteen 2-s-long practice MVIC with 1 min of rest between contractions. After the last practice MVIC, subjects rested for 5 min and then performed three consecutive 2-s MVIC. The mean EMG activity (V) of these MVIC was recorded and averaged for statistical analysis. After 15 min of rest, subjects' toes were removed from the harness, and a 5-mL blood sample was collected. Subjects then voided their bladders and were weighed.
Subjects then lay supine with their dominant ankle hanging off a table and were prepared for muscle cramp induction. Stimulating and EMG electrodes were placed over the areas that were marked on the familiarization session. Subjects were instructed to remain relaxed for the duration of the muscle cramp and to let the cramp proceed for as long as possible. A flexor hallucis brevis cramp was induced per the procedures described in the previous paragraphs. The electrical stimulation frequency used to induce the cramp was recorded and was considered the euhydrated cramp threshold frequency. Euhydrated cramp duration and intensity was also determined at this time (explained below).
After muscle cramp induction, an HR monitor was placed over the chest (Polar Electro, Inc., Lake Success, NY). The right forearm was washed with distilled water and dried. A sterile sweat patch was placed on the forearm, and subjects were weighed again. Subjects then began a 30-min bout of one-leg (nondominant) semirecumbent cycle ergometer exercise at 41°C and 15% relative humidity. Subjects exercised at a moderate intensity that kept their HR around 145-150 bpm. After 30 min of nondominant leg exercise, subjects rested for 5 min, were towel dried, and weighed. This process of 30-min exercise/5-min rest continued until subjects lost ∼3% of their body mass (2.1 ± 0.1 h). Three percent hypohydration was chosen as a benchmark for hypohydration because it seems to be an adequate amount of hypohydration to induce cramping in athletic populations (24,33,35). Fluid ingestion was prohibited during and after exercise. None of the subjects developed an EAMC in their exercised leg during the exercise protocol.
On ∼3% body mass loss, subjects exited the heat chamber, voided their bladders, were weighed, and laid supine on a treatment table for 30 min. Subjects' MVIC EMG activity was then reassessed using the same procedures as reported in the previous paragraphs. We induced a second cramp in the dominant leg's flexor hallucis brevis using the same procedures described in the previous paragraphs to determine the subjects' hypohydrated cramp threshold frequency, cramp duration, and intensity. After cramp cessation, a 5-mL blood sample was collected, and subjects voided their bladders. Subjects were then weighed, the catheter assembly was removed, and subjects were excused.
Because prior studies have shown acclimatization to the cramp induction protocol over time (25), subjects returned to the laboratory 1 wk later to repeat the experiment (i.e., experimental day 2). Subjects were instructed to not drastically alter their diet or activity level the week between experimental sessions and to remark the locations of the EMG and stimulating electrodes if they noticed that the marks were fading. Compliance with instructions was assessed via a diet/exercise log on the final day of testing.
Muscle cramp induction instruments.
Flexor hallucis brevis muscle cramps induced with this low-frequency percutaneous electrical stimulation model have both high intrasession (intraclass correlation coefficient (ICC) (3,1) = 0.844) (37) and intersession reliability (ICC (3,1) > 0.963) (25,37). Scientists have used this model to investigate the effects of various interventions (20,34) as well as their association with EAMC (26,36).
The compound muscle action potentials of the flexor hallucis brevis were sampled at 2000 Hz and were filtered (band-pass, low frequency = 10 Hz; high frequency = 500 Hz) using the MP150 analog-to-digital system operated by AcqKnowledge v3.7.3 software (Biopac Systems, Santa Barbara, CA). Disposable long-term recording electrodes (EL502-10; Biopac Systems) were used to collect EMG data from the subjects' dominant limb. The total EMG recording consisted of baseline (1 s), stimulation (2 s), and poststimulus activity (5 min).
A Grass S88 stimulator with SIU5 Stimulus Isolation Unit (Astro-Med, Inc., West Warwick, RI) with an 8-mm Ag/AgCl-shielded active electrode (EL258S; Biopac Systems) and an 8-cm square dispersive electrode was used to deliver the train of electrical stimuli. Stimulus intensity was set at 80 V because this intensity has been shown to induce muscle cramps in healthy subjects (25).
MVIC EMG activity determination.
The dominant leg's big toe was placed in a toe harness that was attached to a strain gauge rated for loads <11 kg and calibrated with a 4-kg weight. Four-centimeter nylon straps were tightened over the subject's midthigh and shin to prevent movement of the hip and knee. The subject's dominant ankle was placed in a foam block with a footpad at 120° to keep the ankle in slight plantarflexion and to prevent the ankle from extreme inversion and eversion. The subject was instructed to keep the plantar aspect of their foot against this foam block when performing their MVIC. The compound muscle action potentials of the flexor hallucis brevis during MVIC were sampled using similar parameters as described above for flexor hallucis brevis cramps.
Gastrocnemius muscle activity was monitored with a biofeedback unit (Pathway TR-10C; Prometheus Group, Dover, NH) to ensure that subjects were not producing force by using the incorrect muscles. Gastrocnemius EMG activity exceeding 8 mV constituted a failed MVIC attempt. If subjects performed an MVIC incorrectly, they rested for 1 min and then reattempted the contraction. These settings have been used successfully in previous experiments with high intratester reliability (ICC (3,3) = 0.92, unpublished observations).
Cramp duration and intensity determination.
Subjects were instructed to allow the induced cramps to proceed for as long as possible. The flexor hallucis brevis cramp EMG activity was recorded until it seemed to return to resting activity. The filtered and rectified EMG measurements were saved and placed into an algorithm that calculated cramp duration. Cramps were considered alleviated when the cramp EMG activity was <2 SD above resting (1 s) EMG activity. Cramp intensity (%) was calculated by dividing the 2 s of cramp EMG activity immediately after the cessation of the electrical stimulation by the mean 2-s MVIC EMG activity and multiplying by 100.
Blood analysis procedures.
Five-milliliter blood samples were collected before and after hypohydration. One milliliter of blood from each sample was used to analyze hematocrit (Hct) and hemoglobin (0.5 mL for each), then 4 mL of blood remaining was sealed and stored in a 6.0-mL lithium heparin vacutainer (BD, Franklin Lakes, NJ) and placed into an ice bath until the last blood sample was collected.
Blood for Hct analysis was drawn into heparinized microcapillary tubes and centrifuged at 3000 rpm (IEC Micro-MB; International Equipment Co., Needham Heights, MA) for 5 min and read using a microcapillary reader (model IEC 2201; Damon/IEC, Needham Heights, MA). Hemoglobin concentration ([Hb]) was measured by mixing 20 μL of whole blood with 5 mL of cyanomethemoglobin reagent, and the absorbance was read at 540 nm on a spectrophotometer (Smartspec 3000; Bio-Rad, Hercules, CA). [Hb] (g·dL−1) was then calculated from the [Hb] absorbance data using a hemoglobin standard curve. Hct and [Hb] were measured in triplicate immediately after sampling and were averaged for each blood sample for statistical analysis and calculations. Hct and [Hb] measurements were used to calculate changes in plasma volume (PV) per the Dill and Costill equation (12). To determine plasma electrolyte concentrations, plasma was removed from the packed red blood cells, and plasma sodium concentration ([Na+]p), plasma potassium concentration ([K+]p), plasma magnesium concentration ([Mg2+]p), and plasma calcium concentration ([Ca2+]p) were determined using an ion-selective electrode system (NOVA 8 electrolyte analyzer; Nova Biomedical, Waltham, MA). Plasma osmolality (OSMp) was determined using freezing-point depression osmometry (model 3D3 Osmometer; Advanced Instruments, Inc., Norwood, MA). An OSMp of ≤290 mOsmol·kg−1 H2O was used as the benchmark for euhydration (30).
Sweat analysis procedures.
Sweat patches were collected after 45 min of exercise, placed into a clean test tube, and centrifuged for 5 min at 5000 rpm. Sweat samples were analyzed in duplicate with an ion-selective electrode analyzer (NOVA 8 electrolyte analyzer; Nova Biomedical) for sweat sodium concentration (Sw[Na]), sweat potassium concentration (Sw[K]), sweat magnesium concentration (Sw[Mg]), and sweat calcium concentration (Sw[Ca]). Sweat volume (L) was calculated by subtracting postexercise body weight from preexercise body weight with the assumption that 1 kg of body mass lost represented 1 L of fluid lost.
Statistical analysis procedures.
Differences between euhydrated and hypohydrated cramp threshold frequency during experimental days were analyzed with a 2 × 2 factorial repeated-measures ANOVA. Independent variables were experimental day (days 1 and 2) and hydration status (euhydrated or hypohydrated).
To assess the effects of hypohydration on cramp intensity and duration, separate ANCOVA were used. For cramp duration, cramp intensity and cramp threshold frequency were the covariates; for cramp intensity, cramp duration and cramp threshold frequency were the covariates. Because both ANCOVA covariates were insignificant, we removed the covariates from the analysis and reanalyzed cramp duration and intensity with a 2 × 2 factorial (experimental day and hydration status) repeated-measures ANOVA. Tukey-Kramer post hoc multiple comparison tests were used to confirm differences between days and hydration status after significant main effects.
Blood data (OSMp, plasma electrolyte concentrations, Hct, [Hb], and PV) and changes in body weight were analyzed with separate 2 × 2 factorial repeated-measures ANOVA (hydration status and experimental day as the independent variables). Sweat volume and electrolyte concentration were compared between days with repeated-measures ANOVA. All statistical analyses were performed with Number Cruncher Statistical Software (NCSS 2007, Kaysville, UT). Significance was accepted when P < 0.05.
Subjects self-reported compliance with testing instructions between experimental days. Some data (e.g., urine data) have been previously reported (27). Data are presented as means ± SE.
Effects of the dehydration protocol on hematological variables and body weight.
There was no interaction between hydration status and experimental day for percent changes in body weight (F1,9 = 0.31, P = 0.58, power = 0.1) or PV (F1,9 = 1.49, P = 0.25, power = 0.2), Hct (F1,9 = 2.9, P = 0.13, power = 0.33), [Hb] (F1,9 = 0.9, P = 0.37, power = 0.14), OSMp (F1,9 = 0.74, P = 0.41, power = 0.12), [Na+]p (F1,9 = 1.45, P = 0.26, power = 0.2), [K+]p (F1,9 = 0.12, P = 0.74, power = 0.1), [Mg2+]p (F1,9 = 3.69, P = 0.09, power = 0.4), or [Ca2+]p (F1,9 = 2.98, P = 0.12, power = 0.34). There were also no differences between experimental days for any of these variables (F1,9 < 4.61, P > 0.06, power > 0.1). However, exercise caused a significant reduction in body weight (F1,9 = 2563.42, P < 0.001, power = 1) and PV (F1,9 = 131.64, P < 0.001, power = 1), which corresponded with significant increases in OSMp (F1,9 = 215.61, P < 0.001, power = 1), Hct (F1,9 = 42.8, P < 0.001, power = 0.99), [Hb] (F1,9 = 146.1, P < 0.001, power = 1.0), [Na+]p (F1,9 = 230.48, P < 0.001, power = 1), [K+]p (F1,9 = 21.35, P = 0.001, power = 0.98), and [Ca2+]p (F1,9 = 24.4, P < 0.001, power = 0.99). Plasma [Mg2+] showed a trend toward increasing after exercise (F1,9 = 4.85, P = 0.06, power = 0.5). On the basis of similar results on each experimental day, data were combined, and these are reported descriptively in Table 1.
Cramp threshold frequency, duration, and intensity.
Cramp duration, intensity, and threshold frequency data can be found in Table 2. We observed no interaction between hydration status and experimental day for cramp threshold frequency (F1,9 = 1.61, P = 0.24, power = 0.21). Moreover, hypohydration did not decrease cramp threshold frequency (F1,9 = 2.81, P = 0.12, power = 0.32). Finally, we observed no change in threshold frequency between experimental days 1 and 2 (F1,9 = 1.05, P = 0.33, power = 0.2).
No interaction was observed between hydration status and experimental day for cramp duration (F1,9 = 1.42, P = 0.26, power = 0.2). Hypohydration also did not significantly increase muscle cramp duration (F1,9 = 2.91, P = 0.12, power = 0.33). However, cramp duration was significantly longer on experimental day 2 than day 1 (F1,9 = 8.82, P = 0.02, power = 0.8).
Like cramp duration and threshold frequency, no interaction between hydration status and experimental day was observed for cramp intensity (F1,9 = 0.1, P = 0.76, power = 0.1). Moreover, hypohydration did not significantly increase the intensity of electrically induced muscle cramps (F1,9 = 0.31, P = 0.59, power = 0.1). Also, cramp intensity did not differ between experimental days (F1,9 = 2.61, P = 0.14, power = 0.3).
Sweat losses and composition.
Subjects lost similar volumes of sweat (F1,9 = 0.89, P = 0.37, power = 0.14) and had similar Sw[Na] (F1,9 = 1.78, P = 0.22, power = 0.22), Sw[K] (F1,9 = 2.47, P = 0.15, power = 0.3), Sw[Mg] (F1,9 = 1.43, P = 0.27, power = 0.2), and Sw[Ca] (F1,9 = 0.76, P = 0.41, power = 0.1) on each experimental day. Therefore, the sweat data from each day were combined (Table 3). Overall, subjects lost a total of 144.9 ± 9.8 mmol of Na+, 11.2 ± 0.4 mmol of K+, 3.3 ± 0.3 mmol of Mg2+, and 3.1 ± 0.1 mmol of Ca2+ via exercise-induced sweating. The high Sw[Na] of our subjects indicated that our subjects were unacclimated to exercise in the heat (1,7).
Mild hypohydration with minimal muscle fatigue did not decrease flexor hallucis brevis cramp threshold frequency, a quantitative measure of cramp susceptibility (26), as hypothesized. Cramp threshold frequency was unchanged despite our subjects having substantial losses in body weight (∼3%), PV (∼15%), Na+ (∼145 mmol), K+ (∼11 mmol), Mg2+ (∼3 mmol), and Ca2+ (∼3 mmol). Our sweat electrolyte concentrations, plasma data, and body weight losses were similar to or higher than those of other authors who reported these values in athletes with EAMC (2,3,15,33,35). For example, Schwellnus et al. (33) reported marathoners who developed EAMC had a 2.9% decrease in body weight, an OSMp of 280 ± 6 mOsm·kg−1 H2O, a [Na+]p of 139.8 ± 3.1 mmol·L−1, a [K+]p of 4.9 ± 0.6 mmol·L−1, a [Mg2+]p of 0.73 ± 0.1 mmol·L−1, and a [Ca2+]p of 2.3 ± 0.2 mmol·L−1 immediately after a 56-km ultramarathon. Similarly, Maughan (24) observed PV losses of 5.2% ± 8.4%, blood volume losses of 1.7% ± 4.2%, body weight changes ranging from 1% to 6%, a [Na+]p of 147 ± 2 mmol·L−1, a [K+]p of 4.19 ± 0.33 mmol·L−1, and a [Ca2+]p of 2.47 ± 0.13 mmol·L−1 after a 42.2-km race in athletes who developed EAMC. Our [Na+]p and PV values were similar to or higher than these studies (24,33) because we restricted fluid ingestion during exercise. Athletes observed in the study of Schwellnus et al. (33) ingested fluids ad libitum (volume ingested was not measured) throughout the race, whereas in Maughan's study (24), athletes were allowed 200 mL of fluid at seven checkpoints. By restricting fluid intake during exercise, we observed a high amount of hemoconcentration (∼15%), which, according to the dehydration theory, would have increased the interstitial fluid pressure because there would have been a greater osmotic drive for fluid to shift from the intracellular to the intravascular space. In contrast, the gross losses of electrolytes observed in our study were slightly less than those of others (15,35). For example, Stofan et al. (35) observed that American football athletes (mean body mass = 112.9 kg) who had experienced an EAMC had a Sw[Na] and Sw[K] of 55 and 4.6 mmol·L−1, respectively, and a gross sweat loss of 4 L. Thus, these athletes lost 220 mmol of Na+, ∼18 mmol of K+, and approximately 3.5% (4/112.9 kg) of their body mass (assuming 1 L of sweat is equivalent to 1 kg of body mass lost). The discrepancy between our fluid and electrolyte loss data and these data is likely due to the differences in experimental design. We used a single-leg exercise model at a moderate intensity to minimize the amount of fatigue the subjects experienced. The other authors who have reported slightly higher total sweat electrolyte and fluid losses are observational studies in which the athletes performed sport-specific activities at various (and potentially) higher exercise intensities with no predetermined hypohydration level to achieve (2,3,15,35). Regardless of these differences, if dehydration actually causes a contracture of the interstitial space with a concurrent excitation of nerve terminals as some health professionals theorize (4,22), a 3% reduction in total body fluid does not seem to be a great-enough stimulus to elicit this effect. The effect of larger fluid and electrolyte losses on flexor hallucis brevis cramp threshold frequency is unknown and warrants further investigation before dehydration may be ruled out as a cause of EAMC.
Our data support the hypothesis that muscle cramps, specifically those that occur when individuals are mildly hypohydrated, may be more associated with a neurological mechanism than dehydration (32). Investigations of small-diameter muscle afferent activation after fatigue (17,28) and painful (34) and electrical stimuli (21) further support this hypothesis. Serrao et al. (34) observed that injection of hypertonic saline into the flexor hallucis brevis muscle decreased cramp threshold frequency. They postulated that the pain induced by the hypertonic saline increased small-diameter muscle afferent activation and led to increased cramp susceptibility (as indicated by lower flexor hallucis brevis cramp threshold frequencies). Similarly, injection of glutamate, an algesic agent, into latent myofascial trigger points increases resting gastrocnemius EMG activity and causes muscle cramps (16). In contrast, rapid cramp cessation occurs after activation of Golgi tendon organ afferents (21), further supporting a neurological mechanism of cramp. Therefore, increased activation of excitatory small muscle afferents or disinhibition of others may promote enhanced alpha motor neuron excitability via a positive feedback loop and lead to cramp (31). Because we did not observe lower cramp threshold frequencies after our dehydration protocol, we can only assume that these conditions did not contribute to increased small-diameter muscle afferent activation.
Our results are inconsistent with the conclusions from observational research, indicating that dehydration and electrolyte losses cause EAMC. This discrepancy is likely due to the differences in experimental design. Often, the evidence used to support or refute the dehydration/electrolyte loss theory comes from field studies (24,33,35,39) or single-subject case studies (2). Much of this research compares fluid and electrolyte losses before and after exercise in athletes who do and do not develop EAMC (2,3,24,35). Although these studies offer valid descriptive data, scientists cannot make inferences of cause and effect from observational data. Moreover, conclusions made from research comparing athletes who develop EAMC and those that do not develop EAMC after exercise may be misleading because exercise introduces several confounding variables that make causal relationships difficult to demonstrate (e.g., accumulation of lactate and ammonia, intensity of exercise between groups, fatigue, etc). The American College of Sports Medicine recognized these problems when it concluded that the evidence for EAMC was based on minimal scientific evidence (30). We attempted to minimize the confounding effects of fatigue and muscle metabolite accumulation by inducing cramps before and after dehydration in a rested, unexercised muscle. Controlling fatigue and metabolite production is essential because both have been implicated in the development of EAMC (5,13,32).
Our second hypothesis that 3% hypohydration would increase cramp intensity and duration was also rejected. Currently, scientists use a pretest-posttest experimental design with cramp threshold frequency as the dependent variable to determine the effect of an intervention (20,34) where a lower threshold frequency indicates an increased susceptibility to cramp and vice versa (26). However, cramps induced at threshold frequency are moderately intense (∼60% of MVIC EMG activity) and of short duration (12-40 s) (37,41). In a preliminary study, we observed that cramp duration could be significantly improved by increasing the stimulation frequency (41). On the basis of these results, we also questioned if cramp duration could be enhanced if subjects were hypohydrated. Enhancing cramp duration would allow for the investigation of cramp treatments in real time. However, because hypohydration does not seem to increase cramp duration or intensity, scientists can save considerable time by not having to dehydrate subjects if they wish to examine the effect of cramp interventions on cramp duration. However, if inducing muscle cramps in subjects who have similar hematological characteristics as athletes who develop EAMC is a concern, our protocol is more than adequate. Thus, it seems that increasing the electrical stimulation frequency above threshold remains the best method for improving electrically induced muscle cramp duration and intensity (41).
Although hypohydration does not increase cramp duration, subject familiarity with the cramp induction protocol does seem to have an effect. Cramp duration was the longest on experimental day 2. This may be due to a learning effect associated with decreased muscle guarding and apprehension. Several research studies using this cramp induction model have advocated using a familiarization session to decrease these effects (25,26,37). Although a familiarization session was included in this study, it seems that more than one familiarization session is required to eliminate learning. More familiarization sessions may be necessary if scientists wish to investigate the effect of cramp treatments that may not have an immediate (i.e., <2 min) effect on cramp alleviation. The effect of more treatment sessions on cramp duration is unknown. Future research is required to clarify this relationship further.
We must mention two limitations of our study. First, our results are applicable only to males with a previous history of EAMC; future research should replicate this study with female participants. Second, we induced muscle cramps with percutaneous electrical stimulation rather than exercise. This was done to minimize the confounding effects of fatigue and its metabolites, and this attempted to isolate the effects of dehydration on muscle cramp susceptibility. Although the two methods of inducing cramps differ, flexor hallucis brevis cramp threshold frequency has been shown an objective measure of cramp susceptibility, and it seems to be highly associated with the occurrence of EAMC (26). Scientists have frequently used cramp threshold frequency to indicate changes in cramp susceptibility after miscellaneous treatments and experimental interventions (e.g., pain induction, cryotherapy, etc.) (20,34). Continued research on the validity of this cramp induction model and EAMC is necessary.
In conclusion, mild hypohydration with minimal muscle fatigue does not seem to increase individuals' susceptibility to muscle cramps or the duration and intensity of electrically induced muscle cramps. Thus, muscle cramps occurring to individuals who are mildly hypohydrated may be more associated with neuromuscular fatigue than dehydration or electrolyte loss. Health professionals may have more success preventing EAMC by focusing on strategies that minimize neuromuscular fatigue (e.g., endurance or plyometric training) than dehydration. Further clarification is required to elucidate the relationship between cramp duration and familiarity as well as the effect of greater amounts of hypohydration on cramp susceptibility.
The authors thank Samantha R. Miller for her help with data collection and the Brigham Young University Graduate Studies for partially funding this research.
The results of the present study do not constitute endorsement by the American College of Sports Medicine.