Ratings of Perceived Exertion During Aerobic Exercise in Multiple Sclerosis

. Morrison EH, Cooper DM, White LJ, Larson J, Leu S-Y, Zaldivar F, Ng AV. Ratings of perceived exertion during aerobic exercise in multiple sclerosis. Arch Phys Med Rehabil 2008;89:1570-4. Objective: To compare ratings of perceived exertion (RPEs) during aerobic exercise in people with multiple sclerosis (MS) and control participants. Design: Prospective experimental study. Setting: An exercise testing laboratory.

M ULTIPLE SCLEROSIS IS an autoimmune demyelinat- ing disease of the central nervous system and a frequent cause of nontraumatic disability.In MS, exercise is known to provide many preventive and therapeutic benefits. 1Randomized controlled trials have demonstrated that aerobic exercise training can improve fitness and quality of life in MS 2,3 across a range of disability levels. 4,5Like many others with disabling conditions, 6 people with MS tend to be less physically active than the general population, 7,8 even when their MS has caused minimal disability. 9Rehabilitation and exercise professionals would therefore benefit from additional information for helping people with MS exercise safely and effectively.Past research has not fully clarified why people with MS tend to exercise less, apart from decreased conditioning. 7We question whether people with MS may perceive greater exercise effort.
RPEs are practical and cost-effective tools for assessing exercise effort among people with and without disabling conditions. 10Using RPE data generated during exercise testing, rehabilitation professionals can tailor symptom-specific exercise prescriptions.Yet our collective knowledge base lacks data on how best to use RPE in people with MS.2][13] This abnormal fatigue likely stems from multiple factors, 14,15 may be debilitating even early in the disease course, and may unduly increase effort sense during exercise as it does during daily activities.Furthermore, cardiovascular autonomic dysfunction, which has been reported in 25% to 75% of people with MS, 16,17 could potentially prohibit using heart rate to gauge exertion levels accurately. 18Finally, central motor impairment during exercise is common in MS 19 and could increase perceived exertion during exercise, as could heat sensitivity. 20Any of these common factors could affect RPE in MS, especially if symptomatic fatigue and effort sense share a common neural substrate, as recent data suggest. 21espite this theoretical rationale for why MS might result in altered RPE during exercise, at least 2 research groups have found that people with mild-to-moderate MS rate their isometric exercise effort similarly to controls when using the modified Borg 10-point RPE scale. 22,23 study has yet been published that systematically investigates the use of RPE during dynamic or aerobic exercise in MS.
To begin addressing these knowledge gaps, we undertook the present study to determine how the RPE response to aerobic exercise in a group of ambulatory adults with MS would compare with the RPE response of control participants without MS.Our primary objective was to compare psycho-physiologic and physiologic responses to graded endurance or aerobic exercise testing (particularly responses on the Borg 10-point RPE scale) between ambulatory people with mild MS and matched control participants who were equally sedentary.We hypothesized that in the participants with MS, RPE-at any relative work intensity-would exceed RPE of healthy control participants during graded cycling exercise.

Participants
We screened a total of 59 potential study participants who responded to recruitment advertisements (on bulletin boards and web sites at our urban university, and in National Multiple Sclerosis Society publications) or received referrals from their MS physicians.We excluded 11 respondents because of MSrelated disability or other medical problems.To make fitness levels as equivalent as possible between the study groups, we excluded 5 other candidates (including 4 with MS) because they were too physically active to meet the inclusion criterion of Modified Baecke Questionnaire of Habitual Physical Activity 24,25 score of 8.5 or less.Nineteen candidates did not enroll for various other reasons, such as work schedules or failure to follow up after the initial telephone screening.
Of the 59 initial candidates, we enrolled a total of 24 sedentary adults (4 men, 20 women) between 30 and 45 years of age.Twelve participants had mild MS as defined by an EDSS 26 score of 3 or less, with the EDSS administered by an investigator certified for the Neurostatus e-Test (http:// www.neurostatus.net).We matched each MS participant by age (2-year difference or less) and sex to 1 of 12 control participants without MS.All completed the Baecke questionnaire and a medical history review to verify fit with the study's inclusion criteria.We also screened the participants' BMI, blood pressure, heart rate, body temperature, and electrocardiogram before they began to exercise.
Participants with MS met additional inclusion criteria, including current adherence to one of the approved diseasemodifying therapies for MS.Exclusion criteria for both study groups included pregnancy, morbid obesity as defined by BMI of 40kg/m 2 or more, substantial cognitive impairment, cardiopulmonary disease or other condition prohibiting safe exercise testing, MS exacerbation within 3 months of enrollment, and an MAS 27 score greater than 3.We obtained approval from the University of California, Irvine, Institutional Review Board, and all participants gave written informed consent.

Aerobic Exercise Testing
During a single 90-minute study visit, each participant underwent a standardized, graded exercise test using an Ergometrics 800 cycle ergometer with VMax Spectra metabolic analysis system, a with breath-by-breath measures of ventilation and gas exchange and monitoring of heart rate, blood pressure, and 12-lead electrocardiogram.All testing occurred between 7:00 AM and 12:00 M in a temperature-controlled laboratory (21°C).Each test began with a 3-minute unloaded warm-up period, followed by a continuous ramp-type increase in workload of 5 to 20W/min to ensure a test lasting 8 to 12 minutes or until the participant reached a symptom-limited maximum. 28o protect participants' safety, we followed the American College of Sports Medicine's indications for terminating testing if excessive fatigue or any other concerning symptoms occurred. 29We asked all participants afterward to state the sensations that caused them to stop the test.At the same study visit, we collected blood lactate samples immediately before and after exercise from an indwelling catheter placed in the antecubital vein.

Descriptive and Outcome Measures
We obtained each participant's height and mass using standard, calibrated scales and stadiometers, and calculated the BMI.Participants in both study groups underwent whole-body dual-energy x-ray absorptiometry for lean body mass with a Hologic QDR 4500W densitometer.b During the exercise testing, we determined work rate, heart rate, and ventilatory threshold using the V-slope method. 30We collected breath-bybreath data for V ˙O2 and carbon dioxide output.Every 30 seconds during the exercise testing, participants rated their effort sense using the modified 10-point Borg visual analog scale. 31From the individual exercise data, we also calculated each participant's oxygen cost slope (in V ˙O2 •min Ϫ1 •W Ϫ1 ).Because participants provided an RPE only every 30 seconds, we obtained the corresponding heart rate and relative work level (percentage of VO 2 peak) for each reported RPE.Other outcome measures included the 21-item MFIS (which reflects baseline fatigue over the preceding month), 32 which we administered once before exercise and for which we calculated summary scores as well as physical, cognitive, and psychosocial subscale scores.We measured serum lactate with a YSI lactate analyzer c with a sensitivity of .01mmol/L.

Data Analyses
Using 2-sample t tests and (when variables were not normally distributed) Wilcoxon-Mann-Whitney rank-sum tests, we evaluated the differences between the MS and control groups for baseline characteristics, fitness measures (ventilatory threshold, peak work rate, VO 2 peak, oxygen cost slope), RPE, and heart rate at relative work levels, and posttest serum lactate levels.We then applied mixed-model analysis, a statistical method for repeated measurements, to evaluate whether the 2 groups had different patterns of work rate, V ˙O2 , heart rate, and RPE through the exercise protocol, adjusting for covariates such as sex, weight, and work rate.All statistical results were obtained from SAS, d and the statistical significance level was set at .05.

Participants
Table 1 summarizes the baseline characteristics of the MS and control participants.All 24 participants completed the study without difficulty.The groups did not differ appreciably on baseline characteristics except for fatigue (MFIS summary scores and physical subscale scores).MS and control participants scored similarly on BMI, Baecke scores, and pretest lactate levels.Among the MS participants, the median EDSS score was 2.75 (range, 0 -3).The mean MAS score Ϯ SD was .50Ϯ.67 (range, 0 -2).Seven (58%) of the MS participants were using interferons for disease-modifying therapy, and 5 (42%) were using glatirimer acetate.

Responses to Aerobic Exercise Testing
We found no significant differences between the MS and control groups for any physiologic or psychophysiologic characteristic (table 2), including oxygen cost slope (in V ˙O2 •min Ϫ1 •W Ϫ1 ) and the patterns of work rate, RPE, heart rate, and V ˙O2 during the exercise protocol.Postexercise lactate increased 6.80Ϯ1.85mmol/L in the MS group and 8.96Ϯ3.64mmol/L in the control group, not a significant difference between the 2 groups.Neither the mean RPE nor the mean heart rate measured at 25%, 50%, 75%, and 100% of VO 2 peak differed significantly between controls and participants with MS (fig 1).The participants' stated reasons for stopping the testing varied little between the study groups, with 8 of 12 participants with MS and 10 of 12 controls giving leg fatigue as the primary reason.The remainder cited other causes: overall fatigue (2 MS, 1 control), "breathing got hard" (MS), "light-headed" (control), and discomfort with the mouthpiece (MS).

DISCUSSION
To our knowledge, ours is the first study that systematically uses RPE to assess perceived exertion in MS and control participants during incremental exercise testing.We achieved our goal of enrolling study groups comparable for age, sex, body composition, and sedentary levels of physical activity.Contrary to our hypothesis and despite higher baseline fatigue scores in our participants with MS, both groups yielded similar ratings of perceived exertion during graded exercise testing on a cycle ergometer.Our observations suggest that symptomatic fatigue as assessed by the MFIS may not be linked to effort sense during physical exertion, possibly reflecting stimulation of different neural pathways.Because our MS and control groups exhibited similar VO 2 peak, there were no appreciable  We obtained serum samples just before and just after each participant's exercise test from an indwelling catheter placed in the antecubital vein 20 to 30 minutes before the first blood sampling.Lactate showed a significant change from pretest to posttest (PϽ.001) within both study groups, but the change did not differ between the 2 groups.
differences in the absolute work performed that could confound comparisons at relative work intensities.Despite the exercise test's short duration, lactate levels after the testing rose significantly within both study groups, demonstrating that participants did achieve the expected metabolic response to the heavy exercise.
Overall, our results extend those of previous investigations exploring the Borg 10-point scale 22,33 and 6 to 20 scale 20 for isometric exercise testing in MS.For dynamic exercise, Petajan et al 2 measured mean VO 2 peak levels Ϯ SE between 24.2Ϯ1.4 and 26.0Ϯ1.3mL•minϪ1 •kg Ϫ1 for their ambulatory subjects with MS with EDSS score less than 6, using an arm-leg ergometer for graded exercise testing.Although direct comparison with this previous study is not possible because of differences in testing ergometers and EDSS scores, it would appear that our MS group's mean VO 2 peak results of 22.9Ϯ6.2mL•minϪ1 •kg Ϫ1 during leg-only cycling are at least somewhat comparable.This finding suggests these studies are representative of the ambulatory MS population.Cohen et al 34 likewise reported similar mean peak RPE and VO 2 peak among 5 subjects with mild MS (mean EDSS score, 1.7) and 11 controls who completed graded aerobic exercise testing on a cycle ergometer.Our control and MS groups demonstrated oxygen costs (mean VO 2 peak Ϯ SE, 9.1Ϯ1.0mL•minϪ1 •W Ϫ1 and -9.3Ϯ1.3mL•minϪ1 •W Ϫ1 ) resembling those that Wasserman and Whipp 35 reported for healthy but sedentary volunteers (10.1mL•minϪ1 •W Ϫ1 ), suggesting similar energy economy.
7][38] Our results extend the use of RPE to people with mild MS who want to gain the benefits of aerobic conditioning.RPE in fact might indicate effort more accurately than a prescribed heart rate could for any condition in which baroreflexes would tend to slow the pulse, such as aquatic or recumbent exercise (ie, cycling in a recumbent or semirecumbent position).

Study Limitations
Readers should note limitations to our study design.Ours was a preliminary study in an urban university setting that included participants with MS with only mild disability, limiting its generalizability to other groups.We recruited a sedentary control sample (with percentage of predicted VO 2 peak means of 66.4% in the MS group and 74.2% in the control group), providing the advantage of well-matched study groups but conferring the potential drawback that our controls may not have closely resembled the general young adult population.We did not attempt to screen eligible participants with MS for heat sensitivity, the presence or absence of which may have affected the study's results.

CONCLUSIONS
Our data add to the evidence-supported knowledge base about exercise in MS.We hope this study will assist MS and rehabilitation professionals with exercise testing and prescription by highlighting a simple means for people with MS to calibrate their effort sense to physiologic parameters for appropriate exercise intensity and duration.Despite reporting greater baseline fatigue, our participants with MS showed no significant differences from controls in exercise RPE, suggesting that the overall perceptions of symptomatic fatigue and effort sense might be modulated by different neural structures or pathways.It remains to be seen how RPE interacts with aerobic exercise among people with a wider range of MS disability.

Fig 1 .
Fig 1. Mean ratings of perceived exertion and heart rate (HR) during an incremental aerobic exercise test, MS, and control groups.Twelve sedentary participants with minimal to mild disability from MS (EDSS score <3) and 12 sedentary, age-matched and sex-matched control participants rated their perceived exertion on the Borg 10-point visual analog scale every 30 seconds throughout an incremental aerobic exercise test on a cycle ergometer.Two-sample, 2-tailed t tests and mixed-model analysis revealed no significant differences between study groups for RPE or heart rate at any relative work load.NOTE.Values are group means ؎ SDs.
Other investigators have studied another Borg scale (the 6 -20 version) in aerobic arm-leg ergometry exercise for people with MS 20 but not compared with controls.Although MS exercise programs commonly use RPE for exercise prescriptions, to our knowledge no controlled From the Program in Geriatrics, University of California, Irvine, School of Medicine, Orange, CA (Morrison); General Clinical Research Center, School of Medicine, University of California, Irvine, Irvine, CA (Cooper, Larson, Leu, Zaldivar); Department of Kinesiology, University of Georgia, Athens, GA (White); and Department of Exercise Science, Marquette University, Milwaukee, WI (Ng).

Table 2 : Comparison of Exercise Testing Data for the MS and Control Participants
Predicted VO 2 peak was calculated for men as 60 -(0.55 ϫ age), and for women as 48 -(0.37 ϫ age).39    † NOTE.Values are means Ϯ SDs or medians (ranges).Abbreviations: LBM, lean body mass; TBW, total body weight.*