Janet A. Mulcare, PhD; Paul Webb,
MD; Thomas Mathews, MD; and Satyendra C. Gupta, MD
Janet Mulcare is a Professor in the Department of Physical Therapy at Andrews University in Dayton, Ohio. Paul Webb is a Consultant in Yellow Springs, Ohio. Thomas Mathews is Chief of the Neurology Service at the Dayton Veterans Affairs Medical Center in Ohio. Satyendra Gupta is Chief of the Cardiology Service at the Dayton Veterans Affairs Medical Center in Ohio.
The relationship between the metabolic, thermal load of exercise, sudomotor (sweat) response, and core temperature was examined in 20 individuals with multiple sclerosis (MS) and eight non-MS control subjects. Participants performed an incremental bicycle test of maximal aerobic power (VO2peak) from which 50% of VO2peak was calculated. A 50% VO2peak endurance test was performed while subject wore a full-body water garment. Initial circulating water temperature of 27°C was increased 2°C every three minutes to enhance the thermal load. Core, skin, and circulating water temperatures were monitored. Onset of sweating was measured at five sites. Results showed that 50% of the subjects with MS had an abnormally low sweat response in spite of relatively high skin temperatures (> 35°C) at several sites. The thermal load (kJ/kg) of these subjects at test termination was 89% higher than that of the MS subjects with a normal sweat response. Clinicians should be aware that sweat response may be blunted or absent in many patients with MS. Guidance for ways to enhance heat dissipation under adverse climatic conditions or during activities that increase whole body metabolism should be provided to patients with MS. Perhaps of even greater importance was the finding that increased core and skin temperature during exercise did not result in the appearance of any MS-related symptom in 95% of our sample. This finding is in direct conflict with much of the clinical literature.
Suggested citation: Sweat Response During Submaximal Aerobic Exercise in Persons With Multiple Sclerosis. Mulcare, JA, et al. Int J MS Care [Serial on-line]. December 2001;3(4).
In performing aerobic exercise, the body’s primary mechanism for eliminating metabolic heat generated from contracting muscles is the sudomotor (sweat) response. This response is initiated when skin temperature reaches approximately 35oC. Sweating is controlled from the preoptic area of the hypothalamus, which receives input from peripheral thermosensory nerves, the final portions of which are myelinated fibers in the spinal cord and the lower brain. The entire efferent path is composed of myelinated nerves, hence subject to damage by lesions such as those seen in multiple sclerosis (MS). In earlier research involving patients with MS, the sudomotor response had been reported to be abnormal in a high percentage of individuals (42% to 60%).1,2 It is unknown whether decreased sweating in some patients is due to central lesions affecting the autonomic pathways for sweating—that is, from the hypothalamus—or whether the reported inability to sweat is a matter of patients not exerting themselves sufficiently to elicit the sweat response. It also has been reported that the higher the degree of disability (eg, higher scores on the Kurtzke Expanded Disability Status Scale [EDSS]), the more extensive the loss of sudomotor responses.1
In reviewing earlier studies, there are two distinct methodological problems that need to be addressed to properly examine this phenomenon. First, sudomotor response must be examined in the context of a valid measurement of core temperature (Tc). In previous research, core temperature often has been measured orally,1,2 which is not a reliable measure of Tc for research purposes.3 Previous research has shown that a valid, minimally invasive technique that approximates true core temperature can be obtained using the rectal probe method.4,5 The second methodological concern has been the examination of sudomotor response in conjunction with conditions that would simulate a dynamic thermal phenomenon, and not merely during quiet rest (eg, light physical activity or exercise). Previous research has not examined thermoregulation in this population under exercise conditions,1,2 nor has actual skin temperature (Tsk) been measured.
One of the main problems in measuring thermoregulation in this type of population under exercise conditions is that even minimally impaired individuals with MS (patients with an EDSS score of 1 to 4), when asked to exercise at a moderate intensity (eg, 50% to 65% of maximal aerobic capacity—VO2max), would be able to exercise only at an absolute intensity equivalent to approximately 50 to 60 watts (W).6 Based upon maximal exercise power outputs reported for more severely impaired persons (with EDSS scores > 4.5),7 moderate exercise would be performed only at 35 to 50 W. These metabolic (thermal) loads are far below an absolute level and duration that would be expected to cause an increase in core temperature sufficient to elicit sweating in the presence of normal sudomotor responses. Therefore, to address this issue, the current study used a full-body, water-circulating garment to assist in imposing a controlled, ambient, thermal load during moderate exercise to simulate a greater workload. At this simulated higher workload, sweat response could then be examined.
The primary purpose of this study was to document thermoregulatory responses of sweating and core temperature during moderate aerobic exercise with an additional thermal load and to compare the response of subjects with MS with that of healthy, non-MS controls. In addition, we examined the relationship of thermoregulatory responses to the level of disability related to the MS.
Twenty persons who met the criteria
for diagnosis of MS and eight healthy, non-MS control subjects participated.
The experimental subjects (MS) were full to semi-ambulatory (EDSS scores
between 1.0 and 6.5).8 Eligibility requirements for the MS group included:
1) a diagnosis of MS, with laboratory and clinical confirmation; 2) the
disease currently in remission; 3) ambulatory with minimal use of assistive
devices (ie, cane, crutches, walker, wheelchair); and 4) no other chronic
health conditions present (ie, type 2 diabetes, coronary artery disease,
thyroid or endocrine disorder, pulmonary disease, body mass index > 25%
above recommended standards, or non-MS related orthopedic problems). A
summary of physical characteristics and neurologic status is presented
in Table 1. Eight non-MS control subjects matched for physical characteristics,
age, and sex were recruited; their data are presented in Table 2.
|Table 1. Summary of physical characteristics, neurologic status, and maximal aerobic performance variables of the experimental (MS) subjects.|
|Table 2. Summary of physical characteristics and maximal aerobic performance variables of the control (non-MS) subjects.|
All MS patients received a standard neurologic examination to ensure that they met the diagnostic criteria for MS. Based on the outcome of this examination, subjects were rated on the Functional Systems Scale and the EDSS.8 A 12-lead resting electrocardiogram was administered and interpreted to provide clearance for patients to perform moderate exercise.
Subjects were also asked to complete a short medical history form containing questions regarding past and current health. Height and weight were also measured. Prior to any exercise, the subjects were asked whether they experienced sweating under any climatic or physical exertion conditions. In addition, notation was made regarding general symptomatology experienced during increased thermal load (ie, internal or external). Informed consent approved by the local Institutional Review Board was obtained prior to all testing.
Maximal Aerobic Power (VO2peak) Exercise Test:
To generate equivalent workloads for the submaximal sweating protocol, subjects performed a test of maximal aerobic power (VO2peak) using an upright stationary bicycle (Monark, model 818e). The test began with a three-minute rest period, followed by three submaximal stages (0, 25, and 50 W), each lasting four minutes. Following these three stages, a continuous phase began with exercise resistance increased each minute until volitional fatigue. For most participants, this final stage lasted a maximum of three to four minutes. During testing, metabolic (ie, oxygen uptake [VO2], carbon dioxide production [VCO2]) and cardiopulmonary (ie, minute ventilation [VE], systolic and diastolic arterial blood pressure, and heart rate) data were collected continuously. Perceived level of exertion was solicited at the end of the test in three categories: central, peripheral, and integrated. These categories represent anchors for perceived stress related to cardiovascular, local muscle, and overall general stress, respectively.9 Test termination was based on the following criteria: 1) volitional fatigue; 2) predetermined cardiopulmonary indicators (appearance of a plateau of oxygen uptake for two consecutive 15-second samples with a concomitant increase in minute ventilation); or 3) the appearance of MS-related symptoms (eg, dizziness, double vision, spasticity). Again, this test was performed to establish an index from which to estimate a moderate level of exercise (eg, 50% VO2peak) that each subject would be able to maintain for a period of 30 minutes during the sweat response protocol.
Sweat Response During Endurance Cycling:
Subjects exercised continuously at a moderate intensity (approximately 50% of VO2peak) for a maximum of 30 minutes using the same Monark 818e bicycle ergometer. The test began with a three-minute rest period to establish baseline measurements for all physiological parameters previously described. Based upon the results of the VO2peak test, a workload that would elicit a metabolic response equal to 50% of each subject’s VO2peak was established and the subject was asked to exercise at that intensity until one of the following criteria was met: 1) volitional fatigue; 2) sweat appearance on all five sites; or 3) 30-minute time limit. Once the test was terminated, the subject remained seated while skin and core temperatures were continuously monitored. Monitoring continued until the Tc reading stabilized.
Thermal Cooling/Heating Garment:
In previous research in this laboratory, it was observed that many individuals with MS have such a low maximum work capacity that exercising at 50% would not provide sufficient internal metabolic load to elicit either an increase in core temperature or raise skin temperature to a level that would stimulate a sudomotor (sweat) response. Even if the external workload were sufficient, it would take approximately 50 to 70 minutes of exercise to increase rectal temperature even slightly.10 To reduce the time and amount of effort required by the subject, a controllable external heat load was added by using a tubing garment similar to one previously used for cooling patients.7,10 Before beginning the endurance test, the subject donned a two-piece full-body garment (Figure 1) consisting of micro-tubing (internal diameter, 3/32 inch) imbedded within stretchable lycra material, with 2 centimeters separating each line of tubing. This distance changed based upon the size of the subject and the amount of stretch imposed on the garment. The limbs and torso were wrapped with Ace® bandage to improve the contact of the suit tubing with the surface of the skin. It should be noted that the head cover depicted in this illustration was not used in the present study. To accommodate our need to access sweating sites, small openings were constructed in the garment to expose the sites for monitoring sweating. Initially during the rest period, 27°C water was circulated through the suit at a flow rate of 1.5 liters per minute. Once the exercise protocol began, the water temperature was increased 2°C every three minutes to a maximum of 41°C. The exercise test was terminated for any of the following reasons: 1) volitional fatigue; 2) appearance of MS-related symptoms; or 3) the appearance of sweating at all five sites. Using the suit in this manner “clamps” skin temperature to allow for controlled measurement of heat transfer.11-14 Once the test was terminated, 12°C water was immediately circulated through the garment to reverse the effect of the heating.
Core and Skin Temperature and Sweat Measurement:
Core temperature was measured using a rectal thermistor (YSI 400 series, Yellow Springs, Ohio) inserted 10 centimeters beyond the anal sphincter and secured with adhesive tape. Skin temperatures were monitored with small surface thermistor probes, all on the right side of the body at the following nine sites: forehead, anterior chest, abdomen, upper back, lower back, forearm, anterior thigh, posterior thigh, and calf. An area-weighted skin temperature was calculated from these sites and recorded every minute along with the individual skin temperatures. The thermistor sensors were mounted in plastic rings and worn beneath the tubing garment. This isolated the sensor from the cooling tubes and allowed air circulation over the point of contact.
Sweating was detected by the starch-iodine method.11,15,16 This commonly used method relies upon the combination of iodine and wet starch as an indication of the presence of sweat. A small area of the skin adjacent to the site of skin temperature measurement on the forehead, upper arm, upper anterior chest, abdomen, and anterior femur regions was swabbed with tincture of iodine. As long as the skin was dry (ie, free of sweat), applying a piece of starch paper to the area would elicit no chemical reaction on the paper. The onset of sweating appears as a change in color (blue or brown) at the point of contact with the surface of the skin adjacent to active sweat glands. This method was not intended to quantify sweat rate, but merely to indicate the onset of sweating. The usual areas observed for sweat onset are the forehead and chest. These two sites were observed, in addition to the arm, lower abdomen, and thigh, since neurologic damage from MS could affect central pathways for sweating in many areas.17
The amount of heat stored during exercise was derived from the temperature of the water entering the tubing suit (Twi), mean Tsk, external workload, and metabolic heat production. This method was based on previously published calibration data.18-20 In the current experiments, during the three minutes that preceded exercise, there was heat loss to the suit and to the room that exceeded metabolic heat production. Twi was lower than Tsk by 5 (± 1)°C. Heat loss continued until this gradient was reduced to 3°C, when losses equaled gains, and metabolic heat no longer escaped and was stored. The gradient from Twi to mean Tsk diminished to zero after about 10 minutes of exercise. The cumulative heat stored until this time was near zero. Thereafter, metabolic heat, plus external heating, was fully stored. As Twi rose above mean skin temperature, the heat stored was the cumulative sum of heat added by the garment, plus the metabolic heat, less external work. For each degree that Twi exceeded mean skin temperature, heat transfer into the subject was 35 W. A typical recording of Twi, mean Tsk, and heat storage for a subject during the sweating protocol is illustrated in Figure 2.
Figure 2. Typical data from an MS subject. The center panel shows how water temperature entering the suit (Twi) changes from a comfortable level (27°C) to heating, until detection of sweat onset in all five body sites at 21 minutes. This is followed by rapid cooling. The same panel shows the course of mean skin temperature (Tsk). Rectal temperature (Tre) is shown in the upper panel and cumulative body heat storage in the lower panel.
The nature of this study is descriptive research. Measures of central tendency (mean, median, mode) and variance (standard deviation) were calculated for all variables. An independent t-test comparing the experimental (MS) and control groups was used for all dependent variables, P < .05 maintained for the Type I error rate. Correlation analysis was also used to examine the relationship among several dependent variables.
The results of the test of maximal aerobic power (POmax) used to derive a submaximal workload for the sweating protocol are presented for the experimental and control groups in Tables 1 and 2, respectively. POmax for the MS subjects ranged from a low of 30 W to a high of 187 W. Seventy percent of the MS subjects were capable only of a POmax < 100 W. Based on these performance criteria, submaximal exercise at a moderate level (ie, 50% to 65% POmax) would be at an insufficient external work level to promote a metabolic drive under which to examine the sweat response.6,7 These data support earlier published findings.6,7
One of the key questions that directed this research was, “Given a sufficient metabolic (ie, thermal) load, would the sweat response in persons with MS occur in a normal manner—that is, at or before a maximum Tsk of 35°C?” The results of the endurance cycling test revealed that 50% of subjects with MS exhibited normal sweating (MS-NS) in all sites within the 30-minute maximum, and 50% showed abnormal sweat (MS-AS) patterns, with sweating occurring in at least four, but not all five sites. These latter subjects, despite relatively high skin temperatures (33.9 ± 1.10C), failed to show sweating at the fifth site before exercise was terminated. All control subjects exhibited normal sweat response within the allotted time, and the heat load for the control group was significantly lower than that for the MS-AS group (P <.05). The mean length of time required by the MS-NS group to exhibit sweating at all five sites was 13.8 minutes (range eight to 30 minutes), with 90% of the group sweating in less than 20 minutes. These results were similar to those for the non-MS control group, with a mean (± SD) time of 14.3 (± 6.0) minutes. Sixty-two percent of the control group achieved sweating at all five sites in 12 minutes or less (range, six to 12 minutes), with the remainder achieving the criterion within 21 minutes.
Figure 3 illustrates the total heat load (kJ/kg) for the MS-NS, MS-AS, and control groups during the endurance test. The total heat load for the group that did not exhibit normal sweating was significantly higher than that for the 10 MS subjects with normal sweating, as well as that for the control group (P <.05). The higher total heat load experienced by the MS-AS group was a direct reflection of heat being stored by these subjects and a “blunting” of the sweat response. Surprisingly, this higher heat storage did not result in a significantly higher mean Tc or Tsk for the MS-AS group. In addition, the magnitude of the “blunting” appears to vary greatly among individuals. Two individuals in the MS-AS group had a fairly high mean Tsk (> 35.0°C), with three of the four sites measuring higher than 35°C. This would be the upper limit at which sweating would be expected to occur.
Figure 3. Total heat load (kJ/kg) during the sweating protocol for MS subjects with a normal sweat response (MS-NS, n = 10), an abnormal sweat response (MS-AS, n = 10), and non-MS subjects (controls, n = 8) during endurance cycling. The MS-AS is significantly different from the other two groups (P < .05).
An analysis of the data in terms of disability level found a nonsignificant difference in the mean EDSS rating between the MS-NS and MS-AS groups, with 50% of the abnormal sweat response group having a rating of greater than or equal to 5.0 and only 30% of the subjects in the normal sweat response group having EDSS rating greater than or equal to 5.0. The result of a Spearman Rho correlation found only a modest relationship between EDSS and the onset of a sweat response (+ 0.44, P < .05), indicating that the EDSS scale is probably incapable of predicting whether an individual would have a normal or abnormal sweat response. This was not surprising since the EDSS rating is heavily weighted on ambulatory ability, which would be unrelated to the autonomic nervous system function of sweating.
The nature of descriptive research is to observe a phenomenon with the purpose of generating further research of either a descriptive or “experimental” design. Demonstration of the “sweat response” phenomenon in persons with MS under controlled laboratory conditions provides a foundation for future inquiry that should incorporate a larger, more diverse sample. It must be noted that, based upon the small sample of this study, the generalizability of these findings is limited.
Nearly 30 years ago, Noronha and colleagues1 reported that 58% of their MS sample demonstrated abnormal sweating. These findings are strikingly similar to the 50% incidence rate in the current study. The earlier study also suggested that people with an abnormal sweat response were probably more disabled; however, the authors offered no statistical support for their conclusion. In the present study, there were more individuals with a higher EDSS in the group exhibiting abnormal sweating; however, results of correlation analysis failed to support a relationship between disability level (ie, EDSS) and the presence of abnormal sweating.
It is interesting to note that 70% of the individuals with MS who participated in the study were able to correctly predict whether or not they would sweat. The remaining 30% were incorrect in their prediction. These findings suggest that most people with MS would be correct in predicting whether they would sweat if conditions warranted. However, there still remains the question as to why the remaining 30% have an apparent “lack of understanding” of their own physiological responses. One hypothesis is that many of these individuals avoid situations under which the sweat response would be elicited (eg, staying indoors during high humidity and temperature or avoiding exercise or strenuous physical activity). Interestingly, the subject from the control group with the lowest fitness level also predicted she would not sweat. An in-depth interview with this subject revealed a conscious and consistent avoidance of physical activities and climatic conditions that would elicit a sweat response. This behavior is similar to that often expressed by many of the MS subjects.
Noronha et al1 suggested the incidence of abnormal sweat response to be higher in patients with more severe and advanced MS, which appears to be somewhat supported by the current findings. Noronha’s “normal sweat response” group had a mean disability rating of 2.4 while the “abnormal sweat response” group had a mean disability rating of 3.8. In the current study the mean (± SD) EDSS for the normal sweat group was 3.0 (± 2), and that for the abnormal sweat group was 4.2 (± 2).
Since the onset of sweating can be altered through training and/or an individual’s aerobic fitness level, we examined the relationship between fitness level (VO2max mL/kg/min) with time to onset of sweating during the endurance test. Although the range of scores showed a fair amount of variability, the correlation between fitness level and onset of sweat was low (-0.35, P > .05). Categorization of fitness level21 for both groups is presented in Figure 4. If we remove the one outlier in the control group, it appears that the fitness levels of the majority of the MS group are similar to those of the controls. However, a more in-depth analysis of fitness level as it relates to sweat response is illustrated in Figure 5. This analysis revealed that only those subjects with abnormal sweating were significantly less fit (P < .05) than those in the control group. Part of the reason for these findings could be related to the greater percentage of subjects in the abnormal sweating group with a higher EDSS rating. Previous research has shown that VO2peak lowers as EDSS increases.22 Certainly, future research as to whether exercise training could enhance and/or normalize sweat response in our experimental population would be of great interest if it could improve exercise tolerance.
Figure 4. Fitness categories of control and MS subjects based upon performance on the test of maximal aerobic power.
Figure 5. Comparison of VO2peak among controls
(n = 8), MS subjects with normal sweat response (MS-NS, n = 10), and MS
subjects with abnormal sweat response (MS-AS, n = 10) during an incremental
bicycle exercise. The control group was statistically different from the
MS-AS group only (P = < .05).
A final, yet important question addressed
by this research concerns the relationship between sweat response, core
temperature, and perceived level of stress in persons with MS. Examination
of the relationship between the overall rating of perceived level of exertion
showed a strong positive relationship with Tc (r = +0.86). This means that
a subject’s perception of the overall level of stress during the exercise
protocol increased proportional to his or her increase in core temperature.
The peripheral rating of perceived exertion, which cues the person to consider
stress related to the musculature that is performing the exercise, showed
moderate relationships with mean skin temperature (r = +0.64) and the magnitude
of the work intensity (r = +0.66). This indicates that the “peripheral”
stress anchor is probably driven by local mechanoreceptors as well as by
peripheral sensory and temperature receptors. All other relationships among
any of the dependent variables examined fell below r = ±0.20.
It has often been reported that persons
with MS should avoid excessive heat generated through physical exertion
or from outside climatic conditions for fear of eliciting an MS symptom.
In the current investigation, participants were subjected to a large thermal
load (as great as 0.8°C in Tc and skin temperature greater than 35°C
for some individuals), greater than what most could generate simply through
physical activity or what might be absorbed from environmental surroundings.
Even so, only one person with MS reported the appearance of an MS symptom
during or immediately following the endurance exercise protocol. The single
symptom experienced by one subject was of an optic nature (cobweb-like
fuzziness in vision), which also occurred whenever this person felt hot
due to other reasons (eg, taking a hot shower). Similar symptoms of an
optic nature have been previously reported as being common.
Current guidelines for prescribing
exercise to persons with MS published by the American College of Sports
Medicine23 indicate that some individuals may have an abnormal sweat response.
The current study supports this statement. As a result, even for moderate-intensity
aerobic exercise, room temperature should be kept neutral (21° to 24°C).
Fans may also be used for added heat dissipation when normal sweating is
Based upon the percentages from the
present study, we might anticipate that 50% of any given sample of subjects
might have a problem with normal sweating. Furthermore, if individuals
with MS indicate that they do not sweat, there is a 70% likelihood that
they are correct in their observation. As such, when counseling this type
of individual regarding participation in physical activity, leisure pursuits,
exercise, and employment options, strategies for maintaining a cool environment
should be discussed. These might include the following:
Participating in outdoor activities
early in the morning or later in the day, substituting indoor activities
on days when the heat index is in the unhealthy zone
From these pilot data we can conclude
Supported by a grant from the Department
of Veterans Affairs Rehabilitation R & D.
© 2001 International Journal of MS Care
A final, yet important question addressed by this research concerns the relationship between sweat response, core temperature, and perceived level of stress in persons with MS. Examination of the relationship between the overall rating of perceived level of exertion showed a strong positive relationship with Tc (r = +0.86). This means that a subject’s perception of the overall level of stress during the exercise protocol increased proportional to his or her increase in core temperature. The peripheral rating of perceived exertion, which cues the person to consider stress related to the musculature that is performing the exercise, showed moderate relationships with mean skin temperature (r = +0.64) and the magnitude of the work intensity (r = +0.66). This indicates that the “peripheral” stress anchor is probably driven by local mechanoreceptors as well as by peripheral sensory and temperature receptors. All other relationships among any of the dependent variables examined fell below r = ±0.20.
It has often been reported that persons with MS should avoid excessive heat generated through physical exertion or from outside climatic conditions for fear of eliciting an MS symptom. In the current investigation, participants were subjected to a large thermal load (as great as 0.8°C in Tc and skin temperature greater than 35°C for some individuals), greater than what most could generate simply through physical activity or what might be absorbed from environmental surroundings. Even so, only one person with MS reported the appearance of an MS symptom during or immediately following the endurance exercise protocol. The single symptom experienced by one subject was of an optic nature (cobweb-like fuzziness in vision), which also occurred whenever this person felt hot due to other reasons (eg, taking a hot shower). Similar symptoms of an optic nature have been previously reported as being common.
Current guidelines for prescribing exercise to persons with MS published by the American College of Sports Medicine23 indicate that some individuals may have an abnormal sweat response. The current study supports this statement. As a result, even for moderate-intensity aerobic exercise, room temperature should be kept neutral (21° to 24°C). Fans may also be used for added heat dissipation when normal sweating is not present.
Based upon the percentages from the present study, we might anticipate that 50% of any given sample of subjects might have a problem with normal sweating. Furthermore, if individuals with MS indicate that they do not sweat, there is a 70% likelihood that they are correct in their observation. As such, when counseling this type of individual regarding participation in physical activity, leisure pursuits, exercise, and employment options, strategies for maintaining a cool environment should be discussed. These might include the following:
Participating in outdoor activities early in the morning or later in the day, substituting indoor activities on days when the heat index is in the unhealthy zone
From these pilot data we can conclude the following:
Supported by a grant from the Department of Veterans Affairs Rehabilitation R & D.