Share this article on:

Preliminary Results: Effect of Whole-Body Cooling in Patients with Myasthenia Gravis


Medicine & Science in Sports & Exercise: January 2006 - Volume 38 - Issue 1 - p 13-20
doi: 10.1249/01.mss.0000180887.33650.0f
Clinical Sciences: Clinically Relevant

Introduction: Local muscle cooling has been shown to reduce symptoms in some neuromuscular diseases.

Purpose: To determine whether lowering body temperature using a cooling vest would result in improvement in muscle strength, subjective perceptions of weakness and fatigue, and pulmonary function in patients with generalized myasthenia gravis (MG).

Methods: Patients (five female, one male) with generalized MG, aged 29–58 yr, were studied when mean body temperature was lowered (34.9 ± 0.62°C) using a cooling vest (CT) and at normal body temperature (35.8 ± 0.58°C) (NT). Isokinetic and isometric muscle strength and isokinetic endurance were measured using upper body muscle groups. The myasthenic muscle score (MMS) was performed to assess muscle impairment specific to MG patients. Pulmonary function measurements included forced vital capacity (FVC), maximal inspiratory pressure (MIP), and maximal expiratory pressure (MEP). The fatigue impact scale (FIS) was used to track subjective perceptions of fatigue. Core body and skin temperatures were measured throughout each NT and CT trial.

Results: Mean body temperature (0.65core + 0.35skin) was reduced significantly following 30–45 min of cooling (NT = 35.76 ± 0.58, CT= 34.96 ± 0.62°C). The MMS and MIP increased significantly (NT = 91.9 vs CT = 96.3; and NT = 69.3 vs CT = 79.5 cm/H2O) with cooling. Although all other variables were not statistically different between temperature conditions, individual results for grip strength of the right hand, fatigue ratio for wrist extensors and shoulder internal and external rotators, FVC, and MEP showed improvement for the majority of the subjects.

Conclusion: Cooling in patients with MG shows promise to decrease symptoms of weakness and fatigue, thus allowing increased muscle strength and endurance in some patients.

1Exercise Physiology Laboratories, Division of Physical Performance and Development, University of New Mexico, Albuquerque, NM; and 2Physical Therapy Department, School of Medicine, University of New Mexico, Albuquerque, NM2006

Address for correspondence: Dr. Christine Mermier, Exercise Physiology Laboratories, Johnson Center, University of New Mexico, Albuquerque, NM 87131; E-mail:

Submitted for publication November 2004.

Accepted for publication July 2005.

Supported by NIH NCRR GCRC GRANT 5 MO1 RROO997

The authors wish to acknowledge Lou Stephenson, Ph.D., and Brad Laprise of Natick for their equipment loans and technical assistance, and the nursing staff of the Clinical Research Center at the University of New Mexico Health Sciences Center.

Myasthenia gravis (MG) is an acquired autoimmune disorder caused by antibodies that disrupt the acetylcholine receptors (AChRs) at the neuromuscular junction of skeletal muscle. Loss and dysfunction of these receptors leads to a defect in neuromuscular transmission causing muscle weakness and fatigue. Weakness can remain localized in the eye muscles, termed ocular myasthenia, or affect other skeletal muscles, termed generalized MG. The most common presenting feature of generalized MG is weakness and fatigability. Weakness becomes more evident with physical exertion and improves with rest. Respiratory muscle weakness is a common cause of hospitalization in patients with MG. Death from MG is usually due to failure of the muscles of respiration and cough (8). The annual incidence is approximately 1 in 300,000 and prevalence is one per 17,000; therefore, there are approximately 36,000 cases in the United States.

It has been proposed that symptom severity of affected muscles in MG may be influenced by the mean muscle temperature (10). Several studies have demonstrated an influence of temperature on neuromuscular transmission in MG (2,3,9,10,20,25). Local muscle cooling generally produces improvement in manifestations of the disease, such as decreased weakness and fatigue, whereas local heating worsens symptoms (13). Borenstein and Desmedt (2) also have suggested that ambient temperature and weather may have clinical implications for MG patients. Several mechanisms have been proposed to explain cooling effects on neuromuscular transmission, including temperature effects on the release of acetylcholine (ACh), endplate sensitivity, and differences in the activity of acetylcholinesterase (AChE) (24).

Several investigators have shown that some patients with other neuromuscular disorders such as multiple sclerosis (MS) and Lambert-Eaton syndrome exhibit clinical improvement using short-term cooling therapy. Cooling garments have been used successfully with these patients to help decrease symptoms of fatigue and weakness during physical activity (12). One hypothesis is that higher temperatures increase conduction block in demyelinated pathways in MS, and lowered temperatures would then have the opposite effect. Others have suggested that cold exposure may induce metabolic changes that influence conduction in the axon (1). MS shares some symptoms with MG, such as exercise-induced weakness and fatigability that improves with rest, and heat sensitivity, although the mechanism behind such symptoms may be different.

Muscles of MG patients tested by repetitive nerve stimulation show a reduced decrement of the electrical responses and decreased neuromuscular block by cooling and a worsening with warmer temperatures (2,3,13,20,24,25). Because of this evidence, we hypothesized that a cooler body temperature may allow MG patients to exercise for longer or at a higher intensity before the onset of fatigue and weakness. The fact that many patients with MG are sedentary and often take medication that further increases their risk for the development of cardiovascular disease, diabetes, and osteoporosis underscores the importance of developing a method to increase safe participation in regular exercise. Because exercise tends to exacerbate symptoms in MG patients, only 7% are involved in regular physical activity (8). However, lack of exercise has been shown to cause muscle atrophy, fatigue, and weakness, and thus could worsen the ability of MG patients to perform their activities of daily living.

Decreased levels of fatigue and other benefits of exercise have been observed in patients with various diseases such as lupus (22), and MS (17) when they increase their physical activity. There have been few studies examining the acute effect of exercise training with MG patients, although the literature does show generally positive results. There is evidence that individuals with MG not only have the ability to perform exercise, but have the potential to improve with training (15). Identification of a tool to allow patients to sustain exercise with less weakness and fatigue could enhance their health and quality of life. Thus, our aim was to determine if there were differences in several measurements of strength, endurance, and neurological function, as well as subjective perceptions of fatigue in patients with generalized MG when body temperature was lowered.

Back to Top | Article Outline



Five women and one man with diagnosed generalized MG, ages 29–58 yr (mean age 47.8) participated in this study (Table 1). The sample was of mixed ethnicity with one Hispanic, one African American, one Native American, and three Caucasians. Before study entry, participants were screened for uncontrolled or severe depression, hypertension, cardiovascular and metabolic disease, swallowing difficulties, and diverticulitis. All eligible subjects provided written informed consent using the University of New Mexico's institutional review board–approved forms. For descriptive purposes, each participant completed questionnaires to obtain their health history, level of physical activity, and a fatigue impact scale. Five out of the six subjects had a transsternal thymectomy performed within 5 months of diagnosis. All subjects were on oral AChE inhibitor medication (Mestinon™) and four subjects were also taking oral immune suppressant medications (prednisone, azathiaprine, or mycofenolate). One subject was also receiving biweekly plasmapheresis treatments. A complete blood count (CBC) and urine analysis (UA) were done at the beginning of each trial to ensure that patients were free of infection that would affect their body temperature. All female subjects were either postmenopausal or were tested at the follicular phase of their menstrual cycle to control for fluctuations in body temperature.



Back to Top | Article Outline

Experimental Design

The study protocol was approved by the university's institutional review board. This study used a single-case experimental design using subjects as their own control. Each subject completed a screening appointment and a visit for familiarization of the isokinetic strength and endurance test procedures. They then were scheduled for three experimental trials, two of which were performed at ambient temperature (noncooled condition (NT)), and one with a cooling garment (CT) to lower mean body temperature. The order of the first two trials was controlled such that every other subject did the CT trial first. The third visit was a follow-up NT intervention used to assess reproducibility of the measurements. The order of specific tests within each temperature condition also was controlled. Test order was the same for all subjects for all trials.

Subjects were instructed to take their prescribed medications as usual. To control for the intraindividual effects of AChE inhibitor and bronchodialator medications that could confound the measurements, all testing sessions were done at the same time of day for all subjects with the same medication regimen before each trial. In addition, subjects did not perform any physical exercise for at least 24 h before each testing session to ameliorate differences in symptoms of weakness and fatigue between trials. Subjects waited at least 3 d between testing sessions to ensure that previous testing did not influence weakness or fatigue scores for subsequent data collection. Participants stayed overnight at the university hospital's general clinical research center (GCRC) after each exercise testing session for safety and to allow symptom monitoring and posttest data collection.

Back to Top | Article Outline


Myasthenic muscle score.

The myasthenic muscle score (MMS), developed by Gajados et al. (7), was performed to assess muscle impairment symptoms. Possible total score ranged from 0 to 100 for nine items. Timed items included the number of seconds subjects could maintain their arms outstretched horizontally (1 point per 10 s), maintain their lower legs above the bed (horizontal plane) while lying on their back (1 point per 5 s), and how long they could raise their head above the bed (horizontal plane) while lying on their back (scoring depended on whether subjects could raise head with or without resistance, or not at all). Strength of extrinsic ocular muscles and eyelid occlusion were assessed, and speech was rated as normal, nasal, or slurred. Asking the patient to drink a glass of water and to bite a tongue depressor, respectively, assessed swallowing and chewing. These items were rated as normal, weak or impaired, or impossible. The sequence of test items was constant for all subjects. The same physical therapist trained in muscle testing scored all of the tests.

Back to Top | Article Outline

Isometric hand-grip strength.

Grip strength was measured on both hands with subjects in a seated position with the elbow at 90° and the shoulder neutral. A grip dynamometer (Model 1201, Takei & C., Tokyo, Japan) was used for all measurements and adjusted so that the middle phalanx lined up with the handle. Subjects performed two trials of 3–5 s each to attain maximum isometric grip strength, and the highest reading for each hand was recorded.

Back to Top | Article Outline

Peak torque and fatigue index.

Peak torque and fatigue index of isokinetic shoulder internal/external rotation and wrist extension/flexion of the right side of the body were measured using a Cybex II (Lumex, Ronkonkoma, NY) isokinetic dynamometer for both CT and NT conditions. These muscle groups were chosen to test both proximal and distal muscles of the upper extremity.

Subjects were positioned on the Cybex II with the tested joint being lined up with the machine's axis of rotation for each movement. Subjects were tested in a standing position for the shoulder movement and tested in a seated position straddling the bench for the wrist movements. All settings were recorded to ensure that they were the same for subsequent visits. Up to five familiarization repetitions were allowed for each muscle group and test speed. The test speeds were 60 and 120°·s−1 for the wrist for peak torque and fatigue index, respectively, and 60 and 180°·s−1 for the shoulder for peak torque and fatigue index, respectively. The best of five repetitions was used for peak torque, and fatigue index (also known as endurance ratio) was measured over 25 repetitions. Fatigue index is defined as the percent decline in torque over a certain number of repetitions derived from the following: 100 * [1 – (final torque/initial torque)].

The same trained technician completed subject instruction, dynamometer calibration, and operation for all tests. All measurements were made on the right side of the body. The order of condition (CT vs NT) and muscle group tested were randomly assigned. Strength was always measured before fatigue index. Subjects rested for at least 15 min between peak torque and fatigue index measurements for each muscle group.

Back to Top | Article Outline

Fatigue impact scale.

Subjective physical fatigue was assessed using the physical dimension subscale of the fatigue impact scale (6) to assess perceived fatigue on physical functioning. This scale requires subjects to rate their current fatigue on a scale from 0 to 4 (0 = no problem and 4 = extreme problem) by answering questions relating to weakness, coordination, motivation, ability to do physical activity, need for rest, and physical discomfort (40 points possible). A lower score reflects a lower perception of fatigue.

For the present study, the FIS was used not only to examine differences between temperature conditions, but also to monitor safety of the subjects. This was done by administration of the questionnaire at baseline, before and after each exercise bout, and at 4, 8, 12, and 18 h postexercise. This helped researchers and nurses at the GCRC assess any worsening of fatigue or weakness that may have resulted from the exercise and lung function testing protocols.

Back to Top | Article Outline

Pulmonary function tests.

Measurement of forced vital capacity (FVC), maximal inspiratory pressure (MIP), and maximal expiratory pressure (MEP) were done in a seated position using a calibrated pulmonary function system (Collins GS, Warren E. Collins, Braintree, MA). FVC was measured first. The order of testing of MIP and MEP was randomly assigned. Subjects were given standardized instructions regarding the performance of the maneuvers. FVC testing required subjects to exhale as forcefully and rapidly as possible after a maximal inspiration. MIP and MEP required inspiring/expiring maximally against an occluded airway. Subjects performed a minimum of three trials for FVC, MIP, and MEP to assure consistent and optimal performance. The highest value was reported for each test.

Back to Top | Article Outline

Dual-energy x-ray absorptiometry.

Body composition (% body fat (%BF) and bone mineral density (BMD)) was assessed for descriptive purposes using dual-energy x-ray absorptiometry (DXA) (Lunar DPX, software version 3.5, Lunar Radiation Corp., Madison, WI). A whole-body scan was performed with participants positioned in a supine position and scanned head to foot using a medium speed mode. The software provided an estimate of %BF based on an extrapolation of fatness from the ratio of soft tissues attenuation of two x-ray energies in bone-free pixels.

Back to Top | Article Outline

Temperature recording and pill calibration.

Two hours before both cooled (CT) and ambient temperature (NT) exercise testing trials, subjects ingested a CorTemp® thermometer pill, 2 cm in length by 1.2 cm in diameter (Cor-100, renamed HT 150002, HQ Inc. Technologies, Palmetto, FL). An ambulatory recorder/data logger (BCTM3, PED Inc., v1.08, Southborough, MA) was carried by the subjects in a small waist-belt bag for the continuous measurement and recording of core body temperature transmitted by the temperature sensor pill. Before data collection, the ingestible pills were calibrated in a beaker of heated water with a calibrated mercury thermometer. Individual calibration curves were developed for each pill versus a calibrated thermometer and were applied to the data collected during exercise and pulmonary function testing. To assure accuracy of the pill and calibration procedures, the methods of Lee et al. (14) were used.

Back to Top | Article Outline

Skin temperature.

Skin temperatures were recorded using calibrated thermistors connected to a thermometer/recorder (YSI, model 461UC, Yellow Springs, OH). Calibration of each probe was done before each trial. Probes were attached on the right side of the body at the midthigh, lateral calf, lateral upper arm, and the chest on the upper sternum area. Mean skin temperature (Tsk) was calculated using the method of Ramanathan (21): °C = [0.3 (chest + upper arm)] + [0.2 (thigh + calf)]. Burton's (4) equation was used to estimate the mean body temperature (Tb) using a weighted sum of the core temperature and mean skin temperature (Tb = (0.35Tsk) + (0.65 Tcore)).

Back to Top | Article Outline

Cooling system.

Upon arrival at the laboratory for each CT, subjects were fitted with a cooling garment (Personal Ice Cooling System, developed by the U.S. Army Soldier and Biological Chemical Command, Natick, MA). This garment consisted of a network of small flexible tubing sewn into a lightweight long-sleeved shirt with a high collar with Velcro closure. Ice-cooled water was pumped through the tubing network in the heat transfer garment by a battery/pump unit. Metabolic heat from the body was transferred to the chilled water as it flowed through the network of tubing and the water flowed back to the ice bag where the heat was released. The water temperature was lowered to approximately 10°C, the maximal cooling capacity of the system. Testing was started when core temperature dropped 0.5°C or 45 min had elapsed, whichever came first. Core temperature was monitored continuously and a thermal scale was administered every 5 min during the seated cooling to ensure subject safety and comfort level during each CT.

Back to Top | Article Outline

Statistical Analysis

Following descriptive analysis of the subjects, nonparametric analysis (Wilcoxon signed rank test) was used to analyze differences between the two control conditions. There were no significant differences between the two NT conditions, thus NT data were averaged for a single NT score. Subsequently, a Wilcoxon signed rank test was used to analyze differences between the NT and CT conditions. A simple regression analysis was used to examine the relationship between Tb and %BF for both CT and NT. An SPSS statistical package (version 11.0, SPSS, Inc., Chicago, IL) was used for all analyses. A P < 0.05 (two-tailed) was used to determine significance.

Back to Top | Article Outline


The physical and demographic characteristics of the subjects (N = 6) are presented in Table 1. There were no significant differences between the two NT trials for any of the variables; therefore the data were combined and averaged for subsequent analyses. The individual temperature data are shown in Figure 1. The results of pulmonary function testing are presented in Table 2 and Figure 2. There also were no significant differences between the NT and CT trials for the isometric and isokinetic testing results. All measured variables for the complete blood count and urine (UA) testing were within normal limits for all trials. Binding AchR antibody levels ranged from 0.0 to 18.6 nmol·L−1 (normal values = 0.0–0.4 nmol·L−1), and blocking AchR antibodies from 0 to 59% (normal = 0–15%). One subject was diagnosed with antibody negative MG, and thus would not be expected to have measurable antibodies (approximately 15% of cases of MG are antibody negative).

FIGURE 1—Changes in temperature variables for ambient temperature (NT) and cold temperature (CT) trials.

FIGURE 1—Changes in temperature variables for ambient temperature (NT) and cold temperature (CT) trials.



FIGURE 2—Pulmonary function data for ambient temperature (NT) and cold temperature (CT) trials.

FIGURE 2—Pulmonary function data for ambient temperature (NT) and cold temperature (CT) trials.

Back to Top | Article Outline

Comparison of CT and NT Conditions.

Tb was significantly lower during CT trials compared with NT (34.96 ± 0.62 and 35.76 ± 0.58°C, respectively). The MMS and MIP were significantly higher during CT compared with NT (96.3 ± 5.7 vs 91.9 ± 9.7 for MMS, and 79.5 ± 19.4 vs 64.6 ± 19.4 cm/H2O for MIP). All other variables did not differ significantly between NT and CT trials/conditions.

Though the mean data were not significantly different between NT and CT trials (P = 0.115, right hand; P = 0.463, left hand), isometric grip strength for the right (dominant) hand was higher in five out of six subjects during the CT trial, for a mean improvement of 3.0 kg. However, with the left (nondominant) hand, only two subjects had higher grip strength values during CT, for a mean improvement of less than 0.5 kg.

Observed differences between conditions for isokinetic strength measurements were inconclusive. Peak torque values for the right wrist extensor muscles for individual subjects showed no difference between the two conditions for two subjects, whereas three subjects showed a slight decrease (1.0–2.0 ft·lb−1) and one subject showed a 1.0 ft·lb−1 increase with the CT. Four subjects decreased their fatigue ratio with cooling (showing better endurance) for the wrist extensors, and the mean fatigue ratio showed an improvement of 18.9% for the cooled condition. Three subjects improved peak torque and fatigue ratio of the wrist flexors with the CT, and three had better scores with the NT. Shoulder internal rotation strength improved for only two subjects with CT, whereas five out of six had decreased fatigue for that muscle group with cooling. Results for isokinetic strength of the external rotators of the shoulder demonstrated that half the subjects improved their peak torque with CT, and four of six had improved endurance. Similarly, when individual subject data for shoulder internal and external rotation were analyzed, there were no significant differences between conditions for peak torque, but the fatigue ratio decreased for five out of six subjects for internal rotation and four out of six for external rotation with cooling.

Back to Top | Article Outline

Effects of temperature on pulmonary function.

Mean (±SD) data are presented in Table 2. MIP was significantly higher for CT compared with NT. MEP and FVC showed no significant differences between the two conditions.

Back to Top | Article Outline

Effects of temperature on myasthenic muscle score and fatigue impact scale.

Mean MMS scores were 91.9 ± 9.7 for NT and 96.3 ± 5.7 for CT (P < 0.05), out of a total possible score of 100. All six subjects improved during the CT for the MMS with differences between 1 and 14.5 points. Mean scores for the FIS were 13.3 ± 10.8 points for NT and 12.1 ± 11.4 points for CT out of a possible 40 points. Individual subject scores showed four subjects improving with CT, one with no change between conditions, and one subject who reported increased fatigue with the cold temperature.

Back to Top | Article Outline


The major finding of this study is that whole-body cooling shows promise as a method to alleviate symptoms in patients with generalized MG. Several researchers (2,3,13,20,24,25) have shown promising results using local muscle cooling to decrease neuromuscular decrements during repetitive nerve stimulation in patients with MG. Lowering of muscle temperature in MG by only a few degrees results in pronounced amelioration in local muscle weakness, whereas application of local heat augments myasthenic symptoms (2,3,9). MG patients often avoid physical activity, perhaps because the concurrent rise in body temperature may result in increased symptoms. Attenuation of the rise in body temperature appears to decrease symptoms and improve the performance of physical activity in some patients. Quality of life could be enhanced with the ability to engage in more regular physical activity, and through improvement in patients' ability to perform their activities of daily living. Patients' ability to increase their level of regular physical activity could also help prevent side effects of common drugs used to treat MG such as steroids, and lower their risk for development of other chronic diseases such as osteoporosis, hypertension, obesity, diabetes, and cardiovascular disease. This study is unique in that it is the first to address body core cooling in MG patients. Previous experimental cooling investigations in this population examined temperature effects only at the level of individual muscles using repetitive nerve stimulation, and did not address changes that affect functional activities. Body core cooling has been effective in lessening symptoms and increasing the ability to tolerate activity in patient populations with other neuromuscular disorders, such as those with MS (12) and Lambert-Eaton syndrome (16).

Back to Top | Article Outline

Comparison of NT and CT Conditions


Mean body temperature (Tb) was reduced significantly by an average of 0.6°C by using the cooling garment. Our subjects experienced varying levels of shivering and subjective discomfort during the cooling phase. The small drop in mean temperature is not surprising as the body's natural thermoregulatory response to cold exposure produces peripheral vasoconstriction and shivering, thereby minimizing heat loss from the central regions of the body (12,26).

It has been suggested that an important determinant governing the lowering of body temperature using external cooling is an individual's %BF (5). Those with a greater amount of adipose tissue are able to prevent heat loss and defend a constant internal temperature. Our results showed that the %BF was not a significant predictor of the body temperature changes during the CT in this study. Therefore, in the range of adiposity found for subjects with generalized MG in this study (range= 25.5–46.6%, mean = 38.8% ± 9.8 body fat) the amount of Tb or Tc cooling was unaffected by the level of body fatness. Interestingly, subject 5, who had very little change in Tb with cooling, had the highest relative body fat (46.6%). However three other subjects had similar levels of body fat (43.9, 44.6, and 45.4%), and subject 4 had the biggest drop in Tb with cooling and the second highest %BF. It appears that other factors are involved beyond adiposity in regulation of body temperature in these subjects. The inability of this cooling device to adequately lower body temperature in some patients may limit its effectiveness for use during physical activity, so physicians need to take this into consideration when advising patients.

Back to Top | Article Outline

Isometric strength and isokinetic strength and endurance.

Strength and endurance measurements were not statistically affected by the changes in Tb in this study. Handgrip dynamometry was used in this study because several validated disability scales developed for MG use grip strength as a variable (29,30). Handgrip dynamometry is a simple, reliable, quick test that leads to fewer problems with fatigue and learning effects than isokinetic and isotonic testing. Grip strength of the right (dominant) hand in the current study showed a mean difference of 3.0 kg with the CT (27.3 ± 6.2 for NT, 30.3 ± 7.9 for CT), resulting in more than a 10% increase. Only subject 4 decreased with CT, resulting in grip strength measured at 20.7 kg for NT and 18.0 kg with CT. However this subject wore a brace on her right (dominant) hand due to a wrist injury, and this may have affected the results. The individual strength data for this sample support the possibility that these differences could be important, and that it could be meaningful to repeat the measurements with more subjects.

Though precautions were taken to control for learning effects, including familiarization trials and assigning every other subject to the NT trial first, some of the subjects had a hard time with consistency of effort during the isokinetic tests. Subject 3 in the present study had extensive experience being tested on the Cybex II equipment, and this subject's results showed the largest improvement during the CT as compared with the others. This suggests that perhaps learning effect was a factor compromising the results of the isokinetic testing. This unusually large variation in isokinetic measurements has been previously shown with MG patients. Lohi et al. (15) found large test–retest variability in fatigue testing of elbow extension/flexion and knee extension (coefficient of variation (CV)= 23–43%), but smaller variation in maximal voluntary contraction of the same muscle groups (CV = 10–15%). In the present study, the CV was higher for isokinetic compared with isometric testing for the same muscle group (CV = 26–39% for isokinetic wrist extension/flexion, 18–27% for isometric grip strength), again suggesting that a learning effect may influence the variability in isokinetic testing compared with isometric measurements. Steiner et al. (28) did a study on the reliability of isokinetic measurements and concluded that there may be an intrasession “learning effect” that could influence the reliability of the test. Perhaps most importantly, they state that reliability for isokinetic testing could be more variable among patient populations than in healthy subjects. This suggests that isokinetic testing may not be as effective as isometric testing as a tool for measurement of strength and endurance in MG patients, and this should be considered in future studies.

Back to Top | Article Outline

Effect of temperature on pulmonary function.

In generalized MG, it is unusual for the respiratory muscles to be spared. Severe respiratory muscle weakness can cause shortness of breath and dyspnea. Weakness of bulbar and expiratory muscles impair cough and can predispose patients to aspiration and pneumonia. Mier-Jedrzejowicz et al. (18) among others, have found that expiratory muscle strength is slightly less affected than inspiratory muscle strength. Our group of MG patients had lower mean MIP and MEP compared with age- and sex-matched predicted values, but surprisingly, MEP results showed a larger deficit than MIP. Mean values were 87.8 and 50.7% of age- and gender-predicted values for MIP and MEP, respectively.

MIP was significantly higher during the CT, whereas mean FVC and mean MEP showed no significant differences between conditions (Table 2). According to Rochester (23), there is a much greater degree of intrasubject variability in maximal respiratory pressures than in other pulmonary function tests (mean CV for FVC = 15%, CV for respiratory pressures = 25%), and the lack of agreement in results between MIP and MEP may be, in part, due to this. It is hard to speculate why the effect of the colder temperature would have a significant effect on the inspiratory muscles, which are relatively stronger compared with predicted values, and not the expiratory muscles. Because external intercostals are contracted during inspiration, and the internal intercostals are used for forced expiration, perhaps the more superficial external intercostals were more effectively cooled than the deeper internal intercostals. Because respiratory muscle weakness is still a significant problem in these patients, even with advances in treatment, further study of temperature and its effect on the respiratory pressures with more subjects is warranted to clarify these findings.

Back to Top | Article Outline

Effects of Temperature on MMS and FIS


A valid, reliable, and sensitive scale to assess impairment in MG patients is especially important because symptoms and signs of MG can fluctuate. Sharshar et al. (27) assessed the reliability of the MMS and another commonly used scale (30) and concluded that both scales were highly reliable, had a high interrater reliability (r = 0.906), and a good capacity to detect subtle changes in MG impairment.

MMS items include testing for symptoms commonly reported by patients with generalized MG. Other measurements done in this study were specific to a single muscle or muscle group. The MMS includes combined motions that result in purposeful functional activities. This test set was designed to assess the everyday capacities of MG patients to perform independent activities of daily living. The fact that there was an increase in MMS in these patients with the CT shows that the use of this cooling technique has promise in increasing the everyday functioning of those with generalized MG. Based on statistical significance, ease of administration, and reliability and sensitivity of the MMS, this appears to be a good way to assess changes in weakness with alteration of body temperature in MG.

Back to Top | Article Outline


There were no adequate published scales available to assess the subjective fatigue of MG, therefore a scale developed for the use in MS was employed (6). This scale has been used in at least one other study to assess fatigue in patients with MG (19). A scale more specific to these patients should be developed for future studies. Several limitations of the current study are worth considering. Due to the relatively low prevalence of generalized MG, the sample size was limited to six participants. Recruitment was confounded by the ability of larger patients to fit into the cooling garments and the prevalence of many of the excluding conditions such as depression and diabetes in potential volunteers. Another potential limitation is that because MG patients characteristically have fluctuations in symptom severity, there was concern as to the possibility that measured performance and symptom changes could be reflecting these fluctuations rather than changes related to body temperature. To address this limitation, two control trials were performed to assess the reliability of the measurements. Also, because weakness is typically less severe in the morning and worsens as the day progresses (11,22), subjects in this study were scheduled early in the morning for all trials, as this is when most patients usually feel best.

In conclusion, these results show that whole-body cooling in patients with generalized MG shows promise as a method to decrease symptoms of weakness and fatigue, and therefore allow increased muscle strength and endurance in some patients, particularly with activities that target muscle groups that are especially weak. These data show that there may be a subset of thermosensitive MG patients (as in the case in MS) who may have enhanced symptom relief from use of a cooling garment, whereas others may experience less or little benefit. The cooling procedure was well tolerated in this group of volunteers. Also, none of the subjects reported serious symptoms of weakness or fatigue stemming from the moderate physical exertion required for the protocol, suggesting that moderate levels of strength training may be safe in this population and could help lower risks of diseases caused by inactivity.

Future research on the effect of cooling on the ability to perform physical activity in patients with MG should consider measuring differences in symptoms of weakness and fatigue with cooling during performance of aerobic activities such as walking, stair-climbing, or other functional and locomotive activities. Additional research is needed on the usefulness of training the specific muscles affected in generalized MG, such as the expiratory and inspiratory muscles, and whether simple devices easily available to patients may cause functional changes and enhance the comfort of the training. Additionally, questionnaires and testing could be used to determine the muscles most affected by MG for individual patients. Local cooling could then be applied to individual muscles or muscle groups, and strength and endurance data could be compared to the same muscles without cooling.

If beneficial effects of cooling through one or more of these techniques is confirmed with additional research, training studies using cooling garments or precooling should be done to assess the ability of MG patients to safely increase their strength, endurance, and aerobic fitness without significant increases in symptoms. If evidence is found that cooling proves to be useful in decreasing weakness and fatigue during aerobic exercise, then health care providers could modify exercise recommendations for this population, and patients could feel more comfortable engaging in regular aerobic physical activity. This, in turn, could help increase quality of life and lower cardiovascular risk in patients with MG.

Back to Top | Article Outline


1. Beenaker, E. A. C., T. I. Oparina, A. Hartgring, A. Teelken, A. V. Arutjunyan, and J. De Keyser. Cooling garment treatment in MS: Clinical improvement and decrease in leukocyte NO production. Neurology 57:892–894, 2001.
2. Borenstein, S., and J. E. Desmedt. Temperature and weather correlates of myasthenic fatigue. Lancet ii:63–66, 1974.
3. Borenstein, S., and J. E. Desmedt. Local cooling in myasthenia. Arch. Neurol. 32:152–157, 1975.
4. Burton, A. C. Human calorimetry: The average temperature of the body. J. Nutr. 9:261–280, 1935.
5. Cannon, P., and W. R. Keatinge. The metabolic rate and heat loss of fat and thin men in heat balance in cold and warm water. J. Physiol. 154:329–344, 1960.
6. Fisk, J. D., A. Pontefract, P. G. Ritvo, C. J. Archibald, and T. Murray. The impact of fatigue on patients with multiple sclerosis. Can. J. Neurol. Sci. 21:9–14, 1994.
7. Gajados, P., N. Simon, P. de Rohan-Chabot, and M. Goulon. Long-term effects of plasma exchanges in myasthenia. Results of a randomized study. Presse Méd. 12:939–942, 1983.
8. Grob, D. Natural history of myasthenia gravis. In: Myasthenia Gravis and Myasthenic Disorders, A. G. Engel (Ed.). Contemporary Neurology Series. Oxford: Oxford University Press, 1999, pp. 131–145.
9. Gutmann, L. Heat exacerbation of myasthenia gravis. Neurology 28:398, 1978.
10. Jablecki, C., and A. Benton. The frequency of muscle involvement in myasthenia gravis correlates with mean muscle temperature. Muscle Nerve 5:491–493, 1982.
11. Jaretzki, A., R. J. Barohn, R. M. Ernstoff, ET AL. Myasthenia gravis: recommendations for clinical research standards. Ann. Thorac. Surg. 70:327–334, 2000.
12. Ku, Y. E., L. D. Montgomery, H. C. Lee, B. Luna, and B. W. Webbon. Physiologic and functional responses of MS patients to body cooling. Am. J. Phys. Med. Rehabil. 79:427–734, 2000.
13. Lange, D. J. Electrophysiologic testing of neuromuscular transmission. Neurology 48(Suppl 5):S18–S22, 1997.
14. Lee, S. M. C., W. J. Williams, and S. M. Fortney-Schneider. Core temperature measurement during supine exercise: Esophageal, rectal and intestinal temperatures. Aviat. Space Environ. Med. 71:939–945, 2000.
15. Lohi, E., C. Lindberg, and O. Andersen. Physical training effects in myasthenia gravis. Arch. Phys. Med. Rehabil. 74:1178–1180, 1993.
16. Maddison, P., J. Newsom-Davis, and K. R. Mills. Decay of postexercise augmentation in the Lambert-Eaton Myasthenic syndrome: Effect of cooling. Neurology 50:1083–1087, 1998.
17. McCartney, N., D. Moroz, S. H. Garner, and A. J. McComas. The effects of strength training in patients with selected neuromuscular disorders. Med. Sci. Sports Exerc. 20:362–368, 1988.
18. Mier-Jedrzejowicz, A. K., C. Brophy, and M. Green. Respiratory muscle function in myasthenia gravis. Am. Rev. Respir. Dis. 138:867–873, 1988.
19. Paul, R. H., R. A. Cohen, J. M. Goldstein, and J. M. Gilchrist. Fatigue and its impact on patients with myasthenia gravis. Muscle Nerve 23:1402–1406, 2000.
20. Ricker, K., G. Hertel, and S. Stodieck. Influence of temperature on neuromuscular transmission in myasthenia gravis. J. Neurol. 216:273–282, 1977.
21. Ramanathan, N. L. A new weighting system for mean surface temperature of the human body. J. Appl. Physiol. 19:531–533, 1963.
22. Robb-Nicholson, L. C., L. Daltroy, H. Eaton, ET AL. Effects of aerobic conditioning in lupus fatigue: A pilot study. Br. J. Rheumatol. 28:500–505, 1989.
23. Rochester, D. F. Tests of respiratory muscle function. Clin. Chest Med. 9:249–261, 1988.
24. Rutkove, S. B. Effects of temperature on neuromuscular electrophysiology. Muscle Nerve 24:867–882, 2001.
25. Rutkove, S. B., J. M. Shefner, A. K. Wang, M. Ronthal, and E. M. Raynor. High-temperature repetitive nerve stimulation in myasthenia gravis. Muscle Nerve 21:1414–1418, 1998.
26. Schmidt, V., and K. Brück. Effect of a precooling maneuver on body temperature and exercise performance. J. Appl. Physiol: Respir. Environ. Exerc. Phys. 50:772–778, 1981.
27. Sharshar, T., S. Chevret, M. Mazighi, ET AL. Validity and reliability of two muscle strength scores commonly used as endpoints in assessing treatment of myasthenia gravis. J. Neurol. 247:286–290, 2000.
28. Steiner, L. A., B. A. Harris, and D. E. Krebs. Reliability of isokinetic knee flexion and extension measurements. Arch. Phys. Med. Rehabil. 74:1327–1335, 1993.
29. Tindall, R. S. A., J. T. Phillips, J. A. Rollins, L. Wells, and K. Hall. A clinical therapeutic trial of cyclosporine in myasthenia gravis. Ann. NY Acad. Sci. 681:539–551, 1993.
30. Tindall, R. S. A., J. A. Rollins, J. T. Phillips, R. G. Greenlee, L. Wells, and G. Belendiuk. Preliminary results of a double-blind, randomized, placebo-controlled trial of cyclosporine in myasthenia gravis. N. Engl. J. Med. 316:719–724, 1987.


©2006The American College of Sports Medicine