Practice Points
- Acute aerobic exercise using various modalities resulted in temporary worsening of sensory function, with increased pain and physical exhaustion for up to 24 hours for people with advanced multiple sclerosis (MS).
- Acute aerobic exercise improved selective attention and executive control and was enjoyed overall by individuals with advanced MS, despite sensory and symptomatic changes.
- Exercise with an arm cycle ergometer or a recumbent stepper caused more pronounced symptoms than cycling with functional electrical stimulation.
Approximately 30% of people living with multiple sclerosis (MS) rely on a wheelchair for mobility.1 The accumulation of mobility disability in MS is often accompanied by additional disease burden, including symptoms of fatigue, greater cognitive impairment, depression and anxiety, and reduced participation in daily activities.2-5 Nonpharmacological approaches, such as exercise, are essential components of comprehensive disease management for people with advanced MS.
Despite the benefits of exercise for people with MS with mild to moderate disability (ie, Expanded Disability Status Scale [EDSS] scores of 1.0-6.5),6 the role of exercise for people with advanced/severe MS (ie, EDSS score ≥ 7.0) is largely unknown.7,8 Indeed, a recent systematic review reported that 88% of exercise studies that reported EDSS (N = 118) included participants with EDSS scores of 2.0 to 6.5, with only 1 randomized controlled trial (RCT) reporting a mean EDSS of 7.0 or above.7 Similarly, a review of exercise safety reported that of 40 RCTs published since 2013, only 8% included participants who had severe MS (EDSS score of 6.5-9.5).9 Accordingly, current exercise recommendations for people with EDSS scores ranging from 7.0 to 9.0 are primarily derived from expert opinion.10 One challenge to exercise prescription and delivery for this group of people with MS is the need for specialized or adapted exercise modalities that can accommodate various limitations. A few studies have explored the potential of exercise (eg, arm cycle ergometer, recumbent stepper) in people with advanced MS8,11,12; however, this literature is scarce and generally of low quality.
An investigation that considers the functional, symptomatic, physiological, and safety profiles of different adapted exercise modalities would help inform exercise recommendations for people with advanced MS. Additionally, understanding user experiences with different exercise modalities would provide information to optimize exercise prescription. Therefore, characterization of experiences with different exercise modalities could provide direction to promote exercise adoption and adherence at an individual level and to direct resource allocation in community settings.
The objectives of this cross-sectional feasibility study were to characterize the profile of 3 aerobic exercise modalities in people with advanced MS by (1) describing changes in neurological function, cognitive performance, symptoms of fatigue and pain, and adverse events; and (2) describing participant experiences and satisfaction in response to acute exercise.
Methods
Procedures for participant recruitment and baseline data collection have previously been published13 and are briefly summarized below.
Participants
Inclusion criteria stipulated that participants must (1) be between the ages of 18 and 75 years, (2) self-report a diagnosis of MS, (3) have an EDSS score of between 7.0 and 8.0, (4) be relapse free for the past 30 days, (5) be on a stable course of disease-modifying therapy over the past 6 months, and (6) be asymptomatic (ie, present/report no signs or symptoms of acute or uncontrolled cardiovascular, metabolic, or renal diseases or other conditions that would make exercise unsafe) based on the Canadian Society for Exercise Physiology’s Get Active Questionnaire (GAQ)14 screening. Exclusion criteria were (1) other neurological conditions; (2) contraindications to functional electrical stimulation (FES) cycling (eg, pacemakers, implanted devices, unstable fractures); and (3) pregnancy. The sample size was based on recommendations for pilot and feasibility studies15 and was informed by previous exercise studies in advanced MS.16-18
Baseline Outcome Measures
Anthropometrics, Demographics, and Clinical Characteristics
Participants’ height and body mass were measured in the laboratory to the nearest 0.1 cm and 0.1 kg, respectively. Participant height was measured in a supine position. Body mass was measured using a platform or wheelchair scale (Sartorius AG; Global Industrial). Sociodemographic and clinical characteristics were recorded from participants using a questionnaire developed for this study.
Neurological Function
Neurological function was assessed using the EDSS19 by a Neurostatus-certified assessor. The EDSS is a standard measure of disability for people with MS that ranges from 0, no disability, to 10, death.
Cognitive Performance
Baseline cognitive performance was assessed with the oral Symbol Digit Modalities Test (SDMT)20 and the Stroop Color and Word Test (SCWT).21,22 The SDMT was scored based on the total number of correct responses in 90 seconds. The SCWT was scored based on the total number of correct responses in 45 seconds on the incongruent trial (ie, color-word interference score), after completing 2 congruent trials (ie, word reading and color naming).
Symptoms of Fatigue and Pain
Baseline fatigue was measured with the Fatigue Severity Scale (FSS)23 and the Modified Fatigue Impact Scale (MFIS).24 Baseline pain was assessed using the short version of the McGill Pain Questionnaire (SF-MPQ).25
Submaximal Exercise Outcome Measures
Neurological Function
Change in neurological function was assessed through the evaluation of the visual, brainstem, cerebellar, and sensory systems within the EDSS,19 which was administered by the same assessor as at baseline.
Cognitive Performance
Change in cognitive performance was assessed with the SDMT and the SCWT using the same protocols as at baseline. To minimize practice effects, different SDMT versions were administered at each time point.
Symptoms of Fatigue and Pain
Change in acute fatigue was assessed using the Daily Fatigue Impact Scale (D-FIS), an 8-item scale used to capture momentary fatigue.26 Total scores on the D-FIS are calculated as a sum of the 8 items (range, 0-32). Change in acute pain was assessed with 3 items used in previous studies of acute exercise in people with advanced MS16,17 that were developed from the Brief Pain Inventory.27 Participants were asked to rate how much shoulder pain, body pain, and physical discomfort they experienced when using each piece of exercise equipment. Each item was scored on a 7-point scale ranging from 1, not at all, to 7, a lot. Total pain scores were calculated as the sum of the 3 items (range, 3-21).
Adverse Events
Adverse events (AE) during and/or after exercise were characterized using the Common Terminology Criteria for Adverse Events.28 The severity of AEs was graded from grade 1, mild, to grade 5, death.28
Participant Experience and Satisfaction
The affective response to exercise was captured using the 12-item Exercise-Induced Feeling Inventory (EFI)29 and the single-item Feeling Scale (FS).30 The EFI measures 4 feeling states: revitalization, tranquility, positive engagement, and physical exhaustion. Each feeling state is calculated as a mean of 4 items. The single-item FS is an 11-point bipolar scale that asks respondents to rate how they feel at the current moment; responses range from –5, very bad to +5, very good. Participant experience and satisfaction with each modality were captured using a 6-item scale.16 Scores on the satisfaction scale were calculated as a mean of the 6 individual item scores.
Exercise Modalities
The submaximal exercise bouts were performed on an arm-cycle ergometer (ACE; PRO2 Total Body Exerciser; SCIFIT Systems, Inc), a recumbent stepper (RS; StepOne Recumbent Stepper; SCIFIT Systems, Inc), and an FES leg-cycle ergometer (RT300 leg cycle; Restorative Therapies, Inc). The 3 modalities were selected based on use in previous studies with people with MS with mobility impairment.17,18 Prior to exercise, the equipment was adjusted to fit the participant, and participants received instructions on how to use it. Participants used the ACE by propelling their arms around a central axis from a seated position. Participants exercised on the RS using both their arms and legs in a bilateral, reciprocal motion. Participants used the FES while seated and were encouraged to voluntarily cycle with their legs.18 Additional tools (eg, gloves) were available to stabilize and secure participants during exercise as
needed. Participants received automated, superficial stimulation during FES cycling via self-adhering surface electrodes placed over the quadriceps, hamstrings, and calves (PALS Electrodes; Axelgaard Manufacturing Co, Ltd). The intensity of stimulation was adjusted per muscle group according to each participant’s tolerance and was determined prior to the bout of FES cycling. The stimulation parameters were waveform symmetric biphasic, phase duration of 250 µs, and pulse rate of 50 pulses per second.
Procedures
The study protocol was approved by the Health Sciences and Science Research Ethics Board at the University of Ottawa (REB H03-19-3436). Eligible participants completed a baseline session (session 1) and 3 subsequent submaximal exercise sessions (sessions 2-4; Figure 1). At session 1, participants completed the consent process, baseline anthropometrics, neurological function testing, physiological fitness testing (reported elsewhere),13 and questionnaires. Participants then completed sessions 2, 3, and 4 on the 3 exercise modalities in a random order. Sessions 2, 3, and 4 were separated by approximately 7 days and were held at about the same time of day. Participants were asked to maintain a consistent medication schedule and refrain from exercise, caffeine, alcohol, or tobacco for 12 hours prior to each session. Before submaximal exercise (T0), participants underwent an initial assessment of neurological function (visual, brainstem, cerebellar, and sensory functional systems), fatigue, pain (3-item version), cognition (SDMT and SCWT), and affect (EFI and FS). Participants were then familiarized with the Borg Rating of Perceived Exertion (RPE)31 scale and asked to complete 15 minutes of submaximal exercise at an intensity of 12 to 13 RPE (ie, somewhat hard).31 During exercise, RPE was monitored every minute and resistance was adjusted throughout the sessions to achieve the prescribed RPE. AEs were monitored during and after each bout. Neurological function, fatigue, pain, cognition, and affect were measured immediately after exercise (T1; ie, assessments began within about 2 minutes of exercise cessation) and again at 30 minutes post exercise (T2). Participant satisfaction with equipment was measured at T1. A phone call was conducted 24 hours after the session to measure fatigue, pain, and affect, and to capture any AEs within 24 hours after exercise (T3).
Data Analysis
Data were analyzed using IBM SPSS Statistics Version 28. Descriptive statistics were used to characterize the sample at baseline. Differences in acute responses (neurological function, cognitive performance, fatigue, pain, and affect) across the modalities were tested using a repeated measures analysis of variance (ANOVA) with exercise modality (3 levels: ACE, RS, and FES) and time as within-subjects factors (3-4 levels: T0-T2/T3). The frequency of AEs experienced during/within 30 minutes post exercise (T1/T2) and within 24 hours (T3) of exercise was compared between modalities with χ2 tests. Differences in participant satisfaction across the modalities were compared using a 1-way ANOVA. Significant effects were evaluated using Bonferroni post hoc comparisons. Statistical significance was set at P < .05. In-text values are reported within as mean (SD), unless specified otherwise.
Results
Participants
Participant flow through the study is presented in Figure 2. Baseline characteristics of the 18 participants have been reported elsewhere.13 The mean age of the sample was 60.7 (11.1) years, mean disease duration was 21.4 (8.0) years, and the median (IQR) EDSS was 7.5 (0.5).13 Most participants were women (83.3%) and had diagnoses of secondary (55.6%) or primary (33.3%) progressive MS.13 Additional sociodemographic characteristics and baseline cognitive and symptomatic variables are presented in Table S1.
Submaximal Exercise Sessions
All 18 participants completed 15 minutes of submaximal exercise using the FES cycle. Seventeen (94%) participants completed the 15-minute session using the RS and 15 (83%) participants completed the 15-minute ACE session. Regarding exercise intensity, the median (IQR) RPE achieved for ACE, RS, and FES exercise was 12.4 (1.6), 11.9 (2.1), and 11.5 (2.9), respectively. There was no significant difference in RPE reported across the modalities (F(2,34) = 1.19; P= .32; ηp2 = 0.07). A detailed comparison of the cardiorespiratory response to the different modalities will be reported elsewhere.
Acute Response
Neurological function. Changes in acute neurological function are presented in Figure S1. Time had a significant impact on sensory function (F(2,34) = 8.14, P = .001, ηp2 = 0.32). Greater sensory impairment was observed at T1 compared with T0 (P = .03) and T2 (P = .005). There were no other time, modality, or interaction effects on neurological function (all P > .05).
Cognitive performance.Changes in cognitive performance are presented in Figure S2. Neither time (F(2,32) = 2.18, P = .13, ηp2 = 0.12) nor modality (F(1.39,22.25) = 1.21, P = .30, ηp2 = 0.07) had an effect on SDMT performance. There was no time by modality interaction (F(4,64) = 1.03, P = .40, ηp2 = 0.06) on SDMT performance. Time had a significant effect on SCWT performance (F(2,34) = 7.01, P = .003, ηp2 = 0.29), as participants performed better at T2 compared with T0 (P = .008). Neither modality (F(2,34) = 1.90, P = .17, ηp2 = 0.10) nor interaction (F(4,68) = 1.87, P = .13, ηp2 = 0.10) affected SCWT performance.
Symptoms of fatigue and pain. Changes in acute fatigue and pain are presented in Figure S2. Time did not have an effect on acute fatigue (F(1.81,30.81) = 3.02, P = .07, ηp2 = 0.15). Modality had a significant effect on acute fatigue (F(2,34) = 7.81, P = .002, ηp2 = 0.32), as participants reported more fatigue with ACE compared with FES (P = .006). There was no modality by time interaction on acute fatigue (F(3.59,60.99) = 1.50, P = .22, ηp2 = 0.08). Time had a significant effect on acute pain (F(3,51) = 7.81, P < .001, ηp2 = 0.32). Participants reported an increase in acute pain at T1 (P = .018) and T3 (P = .003) compared with T0. Modality had a significant effect on acute pain (F(2,34) = 8.50, P = .001, ηp2 = 0.33), as participants reported more pain in response to ACE (P = .004) and RS (P = .019) than in response to FES. There was a significant modality by time interaction on pain symptoms (F(6,102) = 7.47, P < .001, ηp2 = 0.31). At T1, T2, and T3, participants reported more pain in response to ACE compared with FES (P = .02; P = .03; P < .001, respectively), and at T1 and T3, participants reported more pain in response to RS compared with FES (P = .01; P = .007, respectively).
Adverse events. Table 1 summarizes the reported AEs. Most AEs (82%) involved mild to moderate pain, stiffness, or extremity discomfort (grade 1 or 2). Five AEs were reported during or immediately after the 54 exercise sessions (ACE = 3; RS = 2; FES = 0). Twelve AEs were reported within 24 hours of the exercise sessions (ACE = 6; RS = 3; FES = 3). There was no significant difference in the number of AEs reported during or immediately after (χ2 = 3.09, P = .21) or 24 hours post exercise (χ2 = 1.93, P = .38) across modalities.
Participant Experience
Affective response. Changes in affect as measured by the EFI are presented in Figure S3. Time had a significant effect on tranquility (F(3,51) = 7.88, P < 0.01, ηp2 = 0.32), as participants reported lower values at T1 compared with T0 (P = .005), and T3 (P = .008). Modality affected positive engagement (F(2,34) = 4.41, P = .03, ηp2 = 0.20), such that participants reported lower values (less positive engagement) with ACE compared with FES exercise (P = .04). Time affected physical exhaustion (F(1.75,29.79) = 12.65, P < 0.001, ηp2 = 0.43), as participants reported more physical exhaustion at T1 (P = .008), T2 (P = .02), and T3 (P = .002) compared with T0. Modality had a significant effect on physical exhaustion (F(2,34) = 13.30, P < .01, ηp2 = 0.44), as participants reported being more exhausted with both ACE (P = .004) and RS (P < .001) compared with FES. There was a significant time by modality interaction on physical exhaustion (F(6,102) = 3.57, P = .003, ηp2 = 0.17). At T1, T2, and T3, participants reported more exhaustion with RS compared with FES (P = .03; P = .008; P = .005, respectively). At T2 and T3, participants also reported more exhaustion with ACE compared with FES (P = .006; P = .001, respectively). There were no other time, modality, or interaction effects on other EFI domains
(all P > .05).
Time had a significant effect on FS scores (F(2.28,38.82) = 4.92, P = .01, ηp2 = 0.22), such that participants reported feeling better at T2 (P < .001) and T3 (P = .01) compared with T1. Modality also had an effect on FS scores (F(2,34) = 9.07, P < .001, ηp2 = 0.35), as participants reported feeling better in response to FES compared with ACE (P = .002). There was a significant modality by time interaction on FS scores (F(3.36,57.06) = 3.73, P = .01, ηp2 = 0.18). At T1, T2, and T3, participants reported feeling better in response to FES compared with ACE (P = .01; P = .03; P < .001, respectively). Participants also reported feeling better in response to FES compared with RS at T3 (P = .02). There were no other time, modality, or interaction effects on FS scores (all P > .05).
Participant experience and satisfaction. Table 2 presents a summary of participant experiences and satisfaction with each modality. There were no statistically significant differences in satisfaction across the modalities (F(2,34) = 3.22, P = .052, ηp2 = 0.16).
Discussion
We identified a gap in the MS exercise literature: understanding the role of exercise for people living with advanced MS. Recent calls to action have highlighted the inability to provide evidence-based exercise recommendations and the need for exercise research in understudied MS groups.7,32 We have conducted the first study to characterize the functional, symptomatic, affective, and safety profile of acute aerobic exercise in people with advanced MS (EDSS ≥ 7.0). Findings from this characterization can inform expected responses to exercise and guide exercise interventions in advanced MS.
We observed temporary worsening of sensory symptoms and tranquility, accompanied by an increase in acute pain symptoms and physical exhaustion in response to aerobic exercise. We also observed increases in selective attention and control and overall positive feeling. No changes were noted in other neurological systems, SDMT performance, or symptoms of fatigue or revitalization, suggesting no worsening of these outcomes in response to acute aerobic exercise. The results of a study of 34 people with MS (mode Disease Steps score = 1.0) reported an increase in the number and intensity of sensory symptoms immediately after an acute bout of moderate-intensity (median RPE = 12) combined exercise, without changes in perceived fatigue or physical functioning.33 Importantly, sensory symptoms were alleviated within 24 hours, reinforcing temporary sensory effects of acute exercise in MS.33 Another study examining acute exercise in people with MS (n = 8; EDSS = 5-6) and chronic fatigue syndrome (n = 8) reported a similar increase in pain symptoms in response to 15 minutes of moderate-intensity aerobic exercise, without significant time effects on fatigue symptoms.34 Together, this suggests that people with advanced MS and those prescribing exercise for this population might expect similar symptomatic changes in response to acute aerobic exercise. Of note, acute pain and fatigue symptoms reported herein were relatively low across all modalities with mean scores ranging between 4.2 and 4.7 for pain (scale range, 3-21), and 8.4 and 9.6 for fatigue (scale range, 0-32); this interpretation is important to address potential concerns about symptoms as perceived barriers to engaging in exercise with MS.35 It would be prudent to examine changes in functional and symptomatic responses with repeated exercise exposure and in response to different exercise prescriptions in advanced MS.
Participants reported more fatigue symptoms and a less positive experience with ACE compared with FES and experienced more pain symptoms and physical exhaustion with both ACE and RS compared with FES. All participants completed the FES exercise session, but 3 participants and 1 participant were unable to complete the submaximal exercise session using the ACE and RS, respectively. While there were no significant differences in AE frequency across modalities, 53% of AEs were reported in response to ACE. These differences may partly be explained by greater fatigability of upper limb musculature engaged in ACE and RS. It has been reported that muscles exercising in a state of exhaustion are prone to damage and acute injury.36 Overexertion of the upper body during ACE and RS may have contributed to a greater symptomatic response, physiological fatigue, and, consequently, a less favorable experience. Other studies with people with MS with mobility impairment have not reported differences across different adapted exercise modalities on pain (n = 10; EDSS = 4.0-6.5)17 or fatigue (n = 9; EDSS = 6.5-8.5)37 symptoms in response to acute exercise. Similar AEs were reported in an exercise intervention involving untrained people with spinal cord injury (SCI); temporary increases in shoulder pain were common after ACE exercise.38 That study reported that poor technique and low fitness likely contributed to AEs, and the frequency of AEs decreased with intervention progression. Taken together, ACE may be less favorable for people with advanced MS who are new to exercise and/or physiologically deconditioned, and other modalities for improving physiological fitness might be considered in such scenarios.
Overall, participants reported positive experiences (FS range: +3.0 to +3.6) and expressed high satisfaction across all modalities. Such findings are encouraging for exercise prescription and promotion in this population, as equipment enjoyment is associated with exercise adherence.39 Previous investigations in people with moderate to advanced MS and SCI have reported similar levels of enjoyment when using different types of adapted exercise equipment.17 While there were no significant differences in satisfaction across modalities, participants expressed reservations about using the FES cycle independently, suggesting ACE and RS exercise might be more appropriate in environments without direct supervision/assistance.
Limitations and Future Directions
Importantly, our design included only 3 exercise modalities and single bouts of exercise. Future investigations should examine other types of exercise (eg, resistance) over multiple sessions to examine potential differences in responses and long-term adaptations. We did not measure physical activity levels or prior exercise experience within the current study, and this may have impacted participant perceptions and responses. Further, we did not include separate visits for equipment familiarization to limit testing burden on participants. While there were no significant differences in perceived intensity across the exercise modalities based on RPE, we acknowledge the potential for differences in exercise volume and physiological variables that may have impacted the outcomes reported herein. We further note the potential for discrepancies between perceived (eg, RPE) and objective (eg, heart rate) indicators of exercise intensity.
Conclusions
Single bouts of aerobic exercise resulted in temporary sensory worsening with increased pain and experiences of physical exhaustion in people with advanced MS. These changes were accompanied by improved cognitive performance in some domains. Greater symptomatic responses, more physical exhaustion, and a less positive experience were noted with ACE and RS compared with FES exercise. Commonly reported AEs were mild to moderate extremity pain or discomfort. Importantly, people with advanced MS enjoyed engaging in aerobic exercise, despite sensory and symptomatic changes, reinforcing the feasibility of exercise for this group.
