| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
HIGHLIGHTED TOPICS
A Physiological Systems Approach to Human and Mammalian Thermoregulation
1Environmental Ergonomics Laboratory, School of Health and Human Performance, Dalhousie University, Halifax, Nova Scotia; and 2PacificSport, Canadian Sport Centre, Victoria, British Columbia, Canada
Submitted 2 August 2005 ; accepted in final form 6 December 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
maximal voluntary contraction; hyperthermia; fatigue; neuromuscular activation; triceps surae
40°C) despite alterations in either starting core temperature or the rate of heat storage. This suggests that a critical internal temperature, rather than circulatory failure, may directly elicit exhaustion (5, 24, 39). Recent research has focused on the direct effects of hyperthermia on the central nervous system, including alterations in brain arousal (25), cerebral blood flow (27, 29), brain heat storage (31), perceived exertion (30), and muscular activation (23, 28, 35, 37). Both central and peripheral mechanisms can contribute to neuromuscular fatigue (for review see Ref. 8) (14, 16, 26), and the extent of central vs. peripheral mechanisms that influence neuromuscular impairment during hyperthermia remains unclear. Nybo and Nielsen (28) demonstrated a decrease in voluntary isometric force in both the exercised leg and nonexercised arm muscles after exercise-induced hyperthermia, which they attributed to a lower central voluntary activation. However, a similar study by Saboisky et al. (35) reported a reduction in the central activation ratio, the ratio of maximal voluntary contraction force to superimposed force, of the exercised leg muscles but no decrease in the central activation ratio of the nonexercised arm muscles after exhaustive exercise in the heat. However, the use of exercise-induced hyperthermia, although of practical relevance, typically leads to high cardiovascular strain and dehydration (7), which, coupled with metabolic changes from exercise (33), can confound direct interpretation of the role of hyperthermia per se.
To target the central effects of hyperthermia, Morrison et al. (23) passively heated and then cooled subjects using a liquid-conditioning garment while maintaining euhydration and relatively low (55% heart rate reserve) cardiovascular strain, permitting the comparison of the same core temperatures with both hot and cool skin. Both maximal voluntary force and central voluntary activation during brief (10 s) maximal isometric knee extension gradually decreased with an increase in core temperature. Both subsequently returned to baseline values when cooled to normothermia, thus suggesting a centrally mediated impairment of neuromuscular activation (23). Using transcranial magnetic stimulation, Todd et al. (37) reported impairment of voluntary torque during brief and sustained maximal isometric elbow flexion during passive hyperthermia. Coupled with a decrease in half relaxation time while hyperthermic, this suggests that descending voluntary drive was not able to compensate for local muscular changes. However, the whole body passive heating model used in these studies resulted in increases in both core and muscle temperature. Consequently, it was not possible to isolate whether the effects were due to central and/or peripheral mechanisms.
To further isolate the role of hyperthermia on central impairment of neuromuscular activation, the present experiment passively raised core temperature while maintaining thermoneutrality in one leg and allowing the other leg to change temperature with core body temperature. Therefore, the purpose of the study was to test the role of local skin and muscle temperature during hyperthermia on the ability to perform maximal isometric voluntary contractions. The hypothesis was that hyperthermia would reduce voluntary drive to the muscle, resulting in equivalent impairment of force production and voluntary activation in both the experimental and thermoneutral leg during brief maximal isometric plantar flexion.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Subject characterization. In an initial familiarization session, both height (cm) and weight (kg) were taken, and body fatness was calculated using the seven-site skinfold method (13). Subjects then performed a maximal aerobic power test using an incremental cycle ergometer (model 824E, Monark, Varberg, Sweden) protocol. Subjects warmed up for 4 min at a power output of 60 W followed by 30-W increases each minute until they could no longer maintain the work rate. The subjects breathed into an oronasal mask during the maximal aerobic power test to obtain respiratory measures through open-circuit spirometry. Gas samples were collected at 30-s intervals using a metabolic gas analyzer (model VO2000, Aerosport, Ann Arbor, MI) Before the exercise test, the gas analyzers and the volume transducers were calibrated. Verbal encouragement was provided during the later stages of the test. Heart rate was recorded during each stage of the test using a transmitter-receiver telemetry unit (Polar Vantage XL, Polar Electro, Kempele, Finland). Subjects were also introduced to a plantar flexor torque-measuring apparatus and practiced performing maximal isometric voluntary contractions.
Research design.
Subjects were passively heated from a resting core temperature of 37.0–39.5°C. On reaching 39.5°C, subjects were cooled back down to 38.0°C. Local muscle and skin temperature of the right calf (thermoneutral leg) was kept constant throughout the protocol by wearing a sleeve with ice packs in it, while the left calf (heated leg) was allowed to heat up and cool down during the protocol. At multiple core temperature increments, subjects performed neuromuscular tests and cardiovascular and psychophysical measures were collected. Muscle temperature of the soleus was also recorded before maximal voluntary contractions. See Fig. 1 for time line.
|
Core body temperature control.
The subject was seated in an environmental chamber kept at 32°C (SD 1.1) and 15% (SD 1.0) humidity. Subjects wore a liquid-conditioning garment (Med-Eng, Pembroke, ON, Canada) that consisted of close-fitting stretchable material with Tygon tubing sewn throughout the suit. The liquid-conditioning garment covered the arms, upper legs, lower left leg, head, neck, and torso. Subjects wore shorts, and women also wore a bra under the liquid-conditioning garment to promote maximum heat transfer between the skin and the suit. Water at 50°C was pumped through the suit tubing at 1 l/min to heat the subject from resting core temperature up to 39.5°C. On reaching core temperature of 39.5°C, 10°C water was then pumped through the liquid-conditioning garment tubing to cool the core temperature down to 38.0°C. The subjects were removed from the environmental chamber during cooling and were seated in a temperate (21°C, 17% humidity) environment. Local muscle and skin temperature of the right calf was kept constant throughout both the heating and cooling phases using a custom-made sleeve filled with ice packs.
Thermal measurements. Core temperature was measured using both rectal and esophageal thermistors (Mon-A-Therm Core, Mallinckrodt Medical, St. Louis, MO). The rectal thermistor was inserted 15 cm beyond the anal sphincter when the subject first arrived in the laboratory. The esophageal probe was inserted to the depth of the xiphoid process based on Mekjavic and Rempel's regression equations (20). Subjects were then instrumented with skin thermistors to monitor skin temperature (MA-100, Thermometrics, Edison, NJ). Skin temperature was recorded at four sites on the left side of the body, and mean skin temperature was calculated using an area-weighted equation (0.3 chest, 0.3 upper arm, 0.2 thigh, 0.2 calf) (34). Skin temperature was also recorded on the right calf. Mean body temperature was calculated using a core and skin temperature weighting of 0.65 and 0.35, respectively (34). Core temperature and skin temperature were recorded every 8 s using a portable data unit and stored on a computer with accompanying software (SmartReader 8 Plus, ACR Systems, Surrey, BC, Canada).
A needle microprobe (model MT-26/2, Physitemp, Clifton, NJ) was used to measure intramuscular temperature of the soleus in both legs. The thermocouple was connected to a portable thermometer (Thermalert TH-5, Physitemp), which digitally displayed temperature. The intramuscular temperature probe was inserted 2 cm into the soleus on the lateral side of the leg, posterior to the fibula and 2 cm below the base of the gastrocnemius. At the posterior aspect of the calf, subcutaneous fat thickness averaged 4.2 mm for healthy subjects (22). The probe was inserted into the same site each time and remained in the muscle until the temperature reached a stable value (5 s) and then was immediately removed. The intramuscular temperature of the soleus was recorded before the neuromuscular tests were performed.
Neuromuscular function. Strength of the plantar flexors was evaluated using a custom-made apparatus described below (Fig. 2). The neuromuscular testing protocol consisted of measuring two supramaximal single twitches, followed by two maximum isometric voluntary contractions of 3- to 5-s duration to determine their maximum isometric voluntary contraction. Subjects then performed two additional isometric maximal voluntary contractions with the interpolated twitch technique to assess voluntary activation. Subjects received visual feedback during each trial, and they were provided with a target torque that was based on prior measurement during the familiarization session. Moderate encouragement was given to the subject during each maximum isometric voluntary contraction attempt.
|
Electrically evoked muscle twitches were first recorded while the subject sat in position. The posterior tibial nerve of both legs was stimulated in the popliteal fossa using a 3.5 x 2.5-cm cathode lead electrode with a 11.5 x 16.5-cm anode placed under the posterior surface of the thigh. The position on the anode was adjusted to ensure it was only activating the tibial nerve and not the common peroneal nerve. A supramaximal 1-ms-duration square-wave pulse was delivered to the tibial nerve using a constant voltage stimulator (Grass S48 stimulator and Grass S1U5A stimulus isolation unit, West Warwick, RI). The intensity of the supramaximal pulse was determined at the start of the experiment by increasing the voltage in 5- and 10-V increments (beginning at 50 V) until no further increase in twitch torque was observed. The twitch duration was checked to ensure it was within normal range of 110–130 ms (6). Two supramaximal twitches delivered 5 s apart were recorded at each time point. Contractile characteristics of the single twitches, which included peak twitch torque, contraction time, and half relaxation time, were averaged at each test point. A rest period of 1 min was given between the twitches and in between the maximal voluntary contractions.
Maximal voluntary plantar flexor activation was assessed using the interpolated twitch technique (21). Two supramaximal (150% of control twitch stimulus intensity) square-wave 1-ms pulses were delivered at a frequency of 200 Hz at the point when subjects achieved maximal torque. The experimenter watched the torque signal on two oscilloscopes, one set at a low and the other at a high sensitivity. The nerve was stimulated when no further increment in torque was observed on the oscilloscope set at a higher sensitivity. If the stimulus was not given at the point of maximal torque, the trial was rejected, and the subject was asked to perform an additional trial. A second stimulus was given 2 s after the subject completed the contraction for comparison with superimposed stimulus. The trial with the highest activation was used to prevent underestimation of voluntary activation. Percent voluntary activation (%VA) was calculated using the following formula:
 |
Cardiovascular response. Heart rate and blood pressure were recorded before recording of the intramuscular temperature of the soleus. Heart rate was recorded using a Polar heart rate monitor (Polar Vantage XL, Polar Electro). Heart rate is expressed as heart rate reserve, a percentage of the difference between maximal heart rate and resting heart rate. Blood pressure was recorded manually using a sphygmomanometer on the arm.
Hydration. To prevent dehydration subjects were encouraged to drink lukewarm Gatorade throughout the protocol. Subjects' nude body weight was measured before beginning the session and before each neuromuscular function test. If there was a decrease in body weight, subjects were encouraged to drink more fluid.
Data analysis. For all descriptive variables (e.g., maximal isometric voluntary contraction, heart rate, blood pressure), mean and standard deviation values are reported. Repeated-measures analysis of variance (core temperature) was performed to determine the effect of core temperature on all repeated variables. Normality of the data was confirmed using Mauchly's test sphericity, if the assumption of sphericity was not met the P value was Huynh-Feldt corrected. If significant main effects were found for maximal isometric voluntary contraction or voluntary activation, post hoc trend analysis of within-subject contrasts was performed to determine the best fit of the data. Trend analysis was done rather than pairwise comparisons to encompass the changes seen over the entire range of core temperatures. For all other data, significant main effects were further investigated using Bonferroni post hoc tests. Because of the possible number of pairwise comparisons, only values at initial, peak core temperatures, and final time points were compared. A t-test was performed to test whether there were any differences between relative changes in maximal isometric voluntary contraction and voluntary activation from respective baseline values to values at peak core temperatures for respective legs. This comparison was done twice, once using the peak values with warm skin and another using the peak values with cool skin. Significance was accepted at P < 0.05 level. Statistical analysis was performed using SPSS 11.5.1 software (SPSS, Chicago, IL).
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
|
Neuromuscular function. There was no significant difference found in change in maximal isometric voluntary contraction or voluntary activation between thermoneutral and heated (P > 0.05). Maximal isometric voluntary contraction of the thermoneutral leg was significantly affected by the heating and cooling protocols (P = 0.04). Post hoc trend analysis revealed a significant quadratic trend (P = 0.03) with torque initially at 177.8 N·m (SD 37) and then decreasing to 165.4 N·m (SD 38) N·m at the end of passive heating. There was no significant change in torque [161.3 N·m (SD 38)] with the initiation of cooling; however, at the completion of the cooling phase, torque had returned to baseline values [171.6 N·m (SD 39)] (Fig. 5). Voluntary activation of the thermoneutral leg was also significantly affected by the heating and cooling protocols (P = 0.004) beginning at 97.0% (SD 2) and decreasing to 94.0% (SD 5) at peak core temperature. With initiation of cooling, voluntary activation remained depressed at 93.8% (SD 5), but it returned to baseline [95.2% (SD 3)] on completion of the protocol (Fig. 5).
|
There were some changes seen in the contractile properties of the plantar flexors in both the heated and thermoneutral leg (Fig. 6). Peak twitch torque did not change in either leg (P > 0.05) throughout the protocol. Contraction time was significantly decreased in both the heated leg (P < 0.001) and the thermoneutral leg (P < 0.001) during the protocol. Half relaxation time was also significantly decreased in the heated leg (P < 0.001) and the thermoneutral leg (P < 0.001) throughout the protocol. All values had returned to baseline levels at the completion of the protocol.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
This study found small but significant changes in voluntary activation as determined by the interpolated twitch technique. Although the interpolated twitch technique is the most commonly used method to assess voluntary activation, the sensitivity, especially at high contraction levels, has been questioned (2). During maximal isometric voluntary contractions, there is increased fluctuation of voluntary torque, which can affect discrimination of the interpolated twitch. We collected torque data at a high-resolution level in an attempt to maximize the resolution of the twitch. A double pulse was also used to increase the sensitivity of the interpolated twitch technique (9). Use of a double pulse to compare the interpolated twitch has been suggested to minimize any potentiation effects on variability in voluntary activation (32). Allen et al. (1) found that there was considerable variability in voluntary activation between subjects, however, within-subject variability was consistent and reproducible.
Neuromuscular fatigue during exercise can be caused by either a central failure to maximally activate muscles, alterations in local muscle characteristics, or a combination of both (16, 26). Other studies investigating fatigue of the plantar flexors in a thermoneutral environment used repeated and prolonged maximal contractions and found contributions to fatigue from both central and peripheral mechanisms (14, 26). The strong correlation between peripheral fatigue and decreases in plantar flexor strength (14, 26) suggests that the cumulative fatigue from repeated muscle contractions, such as changes in the intramuscular metabolic environment, can contribute significantly to fatigue beyond a failure of neuromuscular transmission (16). By removing sustained exercise and choosing a brief 5-s isometric plantar flexion that did not alter force production or voluntary activation when repeated periodically over 90 min of thermoneutral exposure in our protocol, the impairment with our heating protocol would not be due to peripheral fatigue or boredom. Rather, our finding of equivalent magnitude and similar patterns of decrease in torque and voluntary activation in both the thermoneutral and heated leg during passive heating to 39.5°C, despite differences in soleus temperature, supports the earlier finding that both exercised (leg) and nonexercised (arm) muscle exhibited a decrease in maximal isometric voluntary contraction and voluntary activation after exercise-induced hyperthermia even with differences in arm and leg muscle temperatures (28). Together, these studies suggest that elevated core temperatures directly affect the ability to maximally activate a muscle, and changes in muscle contractile characteristics due to peripheral muscle temperature alterations minimally influenced voluntary activation. Although none of our subjects were initially able to achieve full activation of the plantar flexors, the voluntary activation levels are in agreement with other studies (14). It is interesting to note that the change in muscle temperature in our heated leg was less than the thigh muscle temperatures at the point of volitional exhaustion during exercise in the heat (10, 28). Therefore, it remains possible that a higher soleus temperature in our heated leg may ultimately have resulted in a greater impairment in torque than observed in the thermoneutral leg.
A central dominance of muscle activation failure during hyperthermia supports the primary role of core temperature afferents in mediating responses to the muscle from a point at or above the level of the motor axons. In a review by Hensel (12), it was established that sweating, metabolic, and vascular responses in humans are controlled by both central and peripheral thermal factors. However, at core temperatures above 40.3°C panting as a means of cooling in dogs could not be prevented by reductions in skin temperature, whereas at lower core temperatures skin cooling had an effect (12), suggesting an increased importance of core temperature vs. peripheral temperature in thermoregulation at high core temperatures. By both heating and cooling the subjects, we were able to differentiate the effects of skin temperature on maximal isometric voluntary contraction and voluntary activation. At the onset of the cooling process, mean skin temperature immediately decreased, whereas core temperature remained elevated. Despite this change in skin temperature, there was no increase in plantar flexor torque or voluntary activation, in agreement with the findings of Morrison et al. (23). Skin temperature also does not appear to be a significant factor in voluntary exhaustion during exercise in the heat. When the rate of heat storage and trunk skin temperature was manipulated by wearing a water-perfused jacket while cycling, highly trained subjects still fatigued at the same core and thigh temperature (40°C) despite a difference of 3°C in skin temperature (10).
Increases in muscle temperature have been shown to increase the speed of contraction and relaxation twitch times of the triceps surae while having no effect on peak twitch tension or maximal isometric voluntary contraction force (6). The changes in twitch characteristics noted in this study are consistent with previous findings during passive hyperthermia studies (23, 37), with changes seen only in speed of the twitch and not the peak tension. Surprisingly, both the heated and the thermoneutral leg exhibited changes in twitch characteristics, despite the soleus temperature of the thermoneutral leg remaining similar to resting values throughout the heating and cooling protocol. The soleus muscle was selected for muscle temperature measurement because in a seated flexed knee position the contribution of gastrocnemius is reduced (36), and plantar flexor torque would have been primarily determined by the soleus in our protocol. During the evoked twitches, however, both the gastrocnemius and the soleus were stimulated, and it is possible that the gastrocnemius temperature was increased enough to affect the overall twitch properties of the calf during plantar flexion. Hansen et al. (11) have shown that increases in catecholamines can prevent force loss due to high levels of K+ and activation of glycogen phosphorylase (17). Passive heating has been shown to produce small increases in catecholamines, which may have played a role in altered twitch characteristics in the thermoneutral leg.
Care must be taken in extrapolating results from isolated muscle movements to exhaustion during whole body exercise in the heat, because discrepancies exist in neuromuscular responses to hyperthermia depending on duration and type of contraction. Although this study found significant decreases in maximal isometric voluntary contraction and voluntary activation using a relatively brief isometric contraction of 3–5 s, others only found impairment with longer duration contractions (28, 37). Nybo and Nielsen (28) observed that at voluntary exhaustion after active hyperthermia, maximal isometric voluntary contraction did not become significantly less than the thermoneutral MVC until after 30 s of maximal contraction. These differences may have been confounded by effects of exercise itself on the metabolic state of exercising and nonexercising muscles. Todd et al. (37) used passive methods to induce hyperthermia and found that only maximal isometric voluntary contraction was decreased during brief (2 s) contractions and not voluntary activation; however, subjects were only heated to 38.5°C. They suggested that, during hyperthermia, motor unit firing rate can be increased to overcome the faster contractile speeds of muscle and maintain voluntary activation in contractions up to 2–3 s (37). Isokinetic contractions after hyperthermia have also shown a similar time-influenced response pattern. Cheung and Sleivert (4) found that isokinetic torque was not affected by passive hyperthermia during two brief contractions. After exercise-induced hyperthermia, no changes were seen in torque or voluntary activation during initial contractions, although a decrement in both torque and voluntary activation occurred with multiple contractions (19). Thus it is possible that the body is able to overcome the effects of increased core temperature for very brief contractions but that the body may respond differently to prolonged contractions.
Although a critical core temperature of 40°C has been hypothesized to directly elicit voluntary exhaustion (7, 10, 28, 30), an alternative model has also emerged suggesting that a central thermal anticipatory mechanism inhibits exercise capacity before the attainment of a critical temperature (18). Neuromuscular support for this model can be found during whole body self-paced exercise, with decreases in power output and/or electromyogram before significant increases in core temperature, heart rate, or perception of effort occurred (15, 38). The significant quadratic trend shown by the gradual decrease in torque seen with the progressive increase in core temperature in our study, combined with similar findings by Morrison et al. (23), also suggests that there is no critical internal temperature at which force and voluntary activation suddenly drop off. In addition, decreases in force and voluntary activation during isometric contractions of both arm and leg muscle after either passive (37) or active (35) heating was evident even at relatively low core temperature of 38.5–38.8°C, respectively. Clearly, further work is required to directly determine whether the presence and nature of a central anticipatory process regulating exercise drive involve thermal or other inputs.
In summary, the present study demonstrates that voluntary isometric force production and activation is reduced with hyperthermia, occurring independently of peripheral muscle and skin temperature. Furthermore, this reduction is progressive with increases in core temperature, gradually decreasing with rising core temperature and progressively increasing back to baseline levels as core temperature returns to thermoneutral levels. Therefore, we conclude that 1) neuromuscular impairment during hyperthermia is primarily due to a central failure to fully activate the muscle and can occur independent of local thermal alterations within the muscle, and 2) neuromuscular function does not become critically impaired on attainment of one particular temperature.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Racinais, N. Gaoua, and J. Grantham Hyperthermia impairs short-term memory and peripheral motor drive transmission J. Physiol., October 1, 2008; 586(19): 4751 - 4762. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Nybo Hyperthermia and fatigue J Appl Physiol, March 1, 2008; 104(3): 871 - 878. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
), body temperature (
), and soleus muscle temperature of H (
) and TN (
) legs during passive heating from rectal temperature of 37.2–39.5°C and then passive cooling from 39.4–37.9°C. Values are means (SD). Matching letters indicate significant differences for mean skin and body temperature (P < 0.001). *Muscle temperature different than all other time points (P < 0.001).
), and diastolic blood pressure (
