Effects of exercise in people with multiple sclerosis: a systematic review and meta-analysis

Background A growing body of studies have examined the effect of exercise in people with multiple sclerosis (MS), while findings of available studies were conflicting. This meta-analysis aimed to explore the effects of exercise on balance, walking ability, walking endurance, fatigue, and quality of life in people with MS. Methods We searched PubMed, Web of Science, Scopus, and Cochrane databases, through March 1, 2024. Inclusion criteria were: (1) RCTs; (2) included an intervention and control group; (3) had people with MS as study subjects; (4) had balance, walking ability, walking endurance, fatigue, or quality of life as the outcome measures. Exclusion criteria were: (1) non-English publications; (2) animal model publications; (3) review articles; and (4) conference articles. A meta-analysis was conducted to calculate weighted mean difference (WMD) and 95% confidence interval (CI). Cochrane risk assessment tool and Physiotherapy Evidence Database (PEDro) scale were used to evaluate the methodological quality of the included studies. Results Forty studies with a total of 56 exercise groups (n = 1,300) and 40 control groups (n = 827) were eligible for meta-analysis. Exercise significantly improved BBS (WMD, 3.77; 95% CI, 3.01 to 4.53, P < 0.00001), TUG (WMD, −1.33; 95% CI, −1.57 to −1.08, P < 0.00001), MSWS-12 (WMD, −2.57; 95% CI, −3.99 to −1.15, P = 0.0004), 6MWT (WMD, 25.56; 95% CI, 16.34 to 34.79, P < 0.00001), fatigue (WMD, −4.34; 95% CI, −5.83 to −2.84, P < 0.00001), and MSQOL-54 in people with MS (WMD, 11.80; 95% CI, 5.70 to 17.90, P = 0.0002) in people with MS. Subgroup analyses showed that aerobic exercise, resistance exercise, and multicomponent training were all effective in improving fatigue in people with MS, with resistance exercise being the most effective intervention type. In addition, a younger age was associated with a larger improvement in fatigue. Furthermore, aerobic exercise and multicomponent training were all effective in improving quality of life in people with MS, with aerobic exercise being the most effective intervention type. Conclusion Exercise had beneficial effects in improving balance, walking ability, walking endurance, fatigue, and quality of life in people with MS. Resistance exercise and aerobic exercise are the most effective interventions for improving fatigue and quality of life in people with MS, respectively. The effect of exercise on improving fatigue was associated with the age of the participants, with the younger age of the participants, the greater the improvement in fatigue. Systematic review registration https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=371056, identifier: CRD42022371056.

Results: Forty studies with a total of exercise groups (n = , ) and control groups (n = ) were eligible for meta-analysis.Exercise significantly improved BBS (WMD, .; % CI, .to ., P < .
), and MSQOLin people with MS (WMD, .; % CI, .to ., P = . ) in people with MS.Subgroup analyses showed that aerobic exercise, resistance exercise, and multicomponent training were all e ective in improving fatigue in people with MS, with resistance exercise being the most e ective intervention type.In addition, a younger age was associated with a larger improvement in fatigue.Furthermore, aerobic exercise and multicomponent training were all e ective in improving quality of life in people with MS, with aerobic exercise being the most e ective intervention type.

Conclusion:
Exercise had beneficial e ects in improving balance, walking ability, walking endurance, fatigue, and quality of life in people with MS.Resistance exercise and aerobic exercise are the most e ective interventions for improving fatigue and quality of life in people with MS, respectively.The e ect of exercise on improving fatigue was associated with the age of the participants, with the younger age of the participants, the greater the improvement in fatigue.

Introduction
Multiple sclerosis (MS) is a disabling neurological disease common in young and middle-aged adults with a mean age of onset of 29 years (1,2).The manifestations of people with MS include physical symptoms such as muscle weakness, muscle spasms, decreased mobility and balance, and increased sensitivity to pain, with psychiatric episodes and fatigue leading to severe disability and deterioration of physical condition, mobility, cognition, and quality of life (3)(4)(5)(6)(7).In fact, 50-80% of people with MS, even in its mild stages, will result in impaired walking performance, further reducing their quality of life as the disease progresses (8).
People with MS usually use pharmacologic strategies that down-regulate immune activation to halt disease progression, prevent relapse, or partially reverse disability (9).However, pharmacologic treatments are often accompanied by adverse effects such as infection, headache, and diarrhea (10).In recent years, exercise has been found to be beneficial in improving aerobic capacity, muscle strength, flexibility, balance, fatigue, and cognitive function in people with MS (11).
A growing body of studies have examined the effect of exercise in people with MS, while findings of available studies were conflicting.Kubsik et al. (12) showed that exercise not only contributes to the physical abilities of people with MS, but also to their mood and attitude toward exercise.In addition, Grazioli et al. (13) reported that multicomponent training was effective in improving quality of life, walking ability, and balance, as well as reducing depression, fatigue, and disease severity in people with MS.Furthermore, Feys et al. (14) showed that running improved aerobic capacity, functional mobility, spatial memory, fatigue, and quality of life in people with MS.However, a meta-analysis showed no significant differences in step count and moderate to vigorous physical activity among individuals with MS, both within and between groups receiving physical activity interventions (14).To the best of our knowledge, Arntzen et al. (15) included only eight randomized controlled trials (RCTs) and the number of included studies was quite small, and the authors included one study in which participants in the control group also received exercise intervention.Another study evaluated the effects of Pilates on balance in people with MS, which included only seven RCTs (16).However, the authors included studies in which control group participants also received exercise interventions such as home exercises (two studies), relaxation exercises (one study), aerobic exercises (one study), and traditional exercises (one study), which may have had some impact on their findings.Therefore, we conducted a comprehensive systematic review and meta-analysis of RCTs to explore the effects of exercise on balance, walking ability, walking endurance, fatigue, and quality of life in people with MS.

Methodological quality assessment
The methodological quality for the included studies was independently assessed by two authors (L.D. and H.X.) based on the Cochrane risk of bias tool (RoB2) (19) and Physiotherapy Evidence Database (PEDro) scale (20,21).If there was disagreement between the two authors, a third author (LY) would join the discussion until the three reach a consensus.RoB2 was assessed mainly from seven items: random sequence generation (selection bias), allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), selective reporting (reporting bias), and other biases.PEDro scale is an 11-item scale used to evaluate the quality of the RCTs of the physical therapy studies, where studies scoring <4, 4-5, 6-8, and >9 points are considered poor quality, average, good, and excellent, respectively (21).

Statistical analysis
We extracted the mean and SD values reflecting changes in timed up and go test (TUG), Berg balance scale (BBS), multiple sclerosis walking scale-12 (MSWS-12), 6-minute walk test (6MWT), fatigue severity scale (FSS), modified fatigue impact scale (MFIS), and multiple sclerosis quality of life-54 (MSQOL-54) from baseline to post-intervention from each study for pooling effects.Weighted mean difference (WMD) and 95% confidence interval (CI) were used to estimate the effects of exercise on balance, walking ability, walking endurance, fatigue, and quality of life in people with MS.For studies reporting standard error (SE) or 95% confidence interval (CI), SD was calculated using the previously described formula.Otherwise, PlotDigitizer online software (www.plotdigitizer.com)was used (22).The I 2 static was used to assess heterogeneity, where I 2 < 25% indicates no significant heterogeneity, 25% < I 2 < 50% indicates low heterogeneity, 50% < I 2 < 75% indicates moderate heterogeneity, and I 2 > 75% indicates high heterogeneity (23).If there was a high heterogeneity (I 2 > 60%), meta-regression analysis, subgroup analysis, and sensitivity analysis were used to interpret the results (19).
For subgroup analyses, we examined the effects of intervention type (aerobic exercise, resistance exercise, and multicomponent exercise), participants' age (young, <45 years old; and middleaged and older adult, ≥45 years old), and type of fatigue detection (FSS and MFIS) on fatigue and intervention type (aerobic exercise and multicomponent exercise) on quality of life in people with MS.Meta-regressions were conducted based on the participants' age, disease duration, duration of intervention, session duration, and weekly time.The analysis result, funnel plot, and forest plot were generated using RevMan 5.2 software.Statistical significance was considered for outcomes with a P < 0.05.

Study selection
As shown in Figure 1, 5,432 records were initially identified from the databases and 11 records from other sources.Three thousand nine hundred and twenty-five studies remained after excluding duplicates and 130 potentially eligible studies remained after the title and abstract screening.Ninety studies were excluded by reading the full text: (1)

E ects of exercise on balance in people with MS
The balance of people with MS was detected by BBS and TUG, with 20 studies providing BBS data and 20 studies providing TUG data.Our results showed that exercise had a significant effect on improving BBS (WMD, 3.77; 95% CI, 3.01 to 4.53, P < 0.00001, I 2 = 50%, Figure 2) and TUG (WMD, −1.33; 95% CI, −1.57 to −1.08, P < 0.00001, I 2 = 34%, Figure 3) in people with MS.

E ects of exercise on walking ability and walking endurance in people with MS
MSWS-12 was used to test walking ability and 6MWT was used to test walking endurance of people with MS.It was found that exercise had a significant effect on improving MSWS-12 (WMD, −2.57; 95% CI, −3.99 to −1.15, P = 0.0004, I 2 = 19%, Figure 4) and 6MWT (WMD, 25.56; 95% CI, 16.34 to 34.79, P < 0.00001, I 2 = 47%, Figure 5) in people with MS.

E ects of exercise on fatigue in people with MS
The fatigue of people with MS was detected by FSS and MFIS.As shown in Figure 6, exercise had a significant effect on improving fatigue in people with MS (WMD, −4.34; 95% CI, −5.83 to −2.84, P < 0.00001, I 2 = 79%).

E ects of exercise on quality of life in people with MS
The fatigue of people with MS was detected by MSQOL-54.As shown in Figure 7, exercise had a significant effect on improving MSQOL-54 in people with MS (WMD, 11.80; 95% CI, 5.70 to 17.90, P = 0.0002, I 2 = 66%).
Our meta-analysis results showed high heterogeneity in fatigue (I 2 = 78%) and quality of life (I 2 = 66%), to explain the heterogeneity between included studies and find modifiable factors of exercise, meta-regression analysis, subgroup analysis, and sensitivity analysis were further performed.

Quality of life
We conducted a subgroup analysis by type of intervention.Aerobic exercise (WMD, 11.68; 95% CI, 5.31 to 18.05, P = 0.0003, I 2 = 0%) and multicomponent training (WMD, 7.28; 95% CI, 2.77 to 11.79, P = 0.002, I 2 = 24%) were effective in improving quality of life in people with MS, with aerobic exercise being the most effective intervention type (Supplementary Figure 6).

Sensitivity analysis
Sensitivity analyses showed that there is no change in the direction or level of compatibility of the overall effect of exercise on fatigue (Supplementary Figure 7) and quality of life (Supplementary Figure 8) in people with MS when any of the included studies are omitted.

Risk of bias
The quality of included studies was assessed by the Cochrane Collaboration tool in terms of selection bias, performance bias, attrition bias, reporting bias, detection bias, and other bias (Supplementary Figure 9).The results of PEDro scale showed that of the 40 included studies, 39 were of good quality and one was of fair quality (Supplementary Table 2).

Publication bias
Possible publication bias was assessed by examining the funnel plot (Supplementary Figure 10).Visual inspection of the funnel plot suggested the absence of funnel plot asymmetry.The results of the egger's test indicated that the small sample size studies were not enough to affect the final results (TUG, P = 0.575; BBS, P = 0.705; 6MWT, P = 0.586; MSWS-12, P = 0.137; quality of life, P = 0.791; Supplementary Table 3), with the exception of fatigue (P = 0.002).Therefore, we performed the Dsuval and Tweedie's trim and fill procedure, and the results indicated that no evidence of publication bias was found for fatigue.

FIGURE
Meta-analysis results of the e ect of exercise on timed up and go test (TUG) in people with MS.Exercise had a significant e ect on improving TUG in people with MS (P < . ).

FIGURE
Meta-analysis results of the e ects of exercise on walking ability in people with MS.Exercise had a significant e ect on improving walking ability in people with MS (P = . ).

Discussion
In this systematic review and meta-analysis, exercise significantly improved balance, walking ability, walking endurance, fatigue, and quality of life in people with MS.Subgroup analyses showed that a younger age was associated with larger improvement in fatigue.In addition, resistance exercise and aerobic exercise were the most effective interventions for improving fatigue and quality of life, respectively.
Loss of balance and walking ability are two of the primary impairments of MS that leads to increased fatigue perception and disease severity, and loss of autonomy (13).Imbalance, gait dysfunction and falls are common in people with MS, with the overwhelming majority having abnormal postural control and gait even early in the course of the disease.It has been reported that 50-80% people with MS have balance and gait dysfunction and over 50% fall at least once each year (64).Exercise has been shown to improve physical function and psychological rehabilitation in people with MS, and to help reduce the risk of falls (65, 66).Our study showed that exercise significantly improved balance function (TUG and BBS) in people with MS, which was consistent with a previous study (13), showing that the combination of resistance and aerobic exercise training is effective in improving balance in people with MS and supports functional and psychological FIGURE Meta-analysis results of the e ects of exercise on walking endurance in people with MS.Exercise had a significant e ect on improving walking endurance in people with MS (P < . ).

FIGURE
Results of the e ects of exercise on fatigue in people with MS.Exercise had a significant e ect on improving fatigue in people with MS (P < . ).
Frontiers in Public Health frontiersin.org

FIGURE
Results of the e ects of exercise on quality of life in people with MS.Exercise had a significant e ect on improving quality of life in people with MS (P = . ).
therapeutic effects through exercise.In addition, a meta-analysis showed that yoga was the best intervention to improve static and dynamic balance, and aquatic training was the best intervention to improve walking ability in people with MS (67).The mechanisms by which exercise improves balance may be that exercise improves neurological control of muscles, increases unconscious deliberate muscle responses to dynamic joint stabilization signals, and enhances core area muscle strength to strengthen body stability.
Our results showed that exercise significantly improved walking ability (MSWS-12) and walking endurance (6MWT) in people with MS, which was in agreement with previous studies, showing that aerobic exercise, aquatic exercise, virtual reality training, and assisted gait training significantly improved walking ability (67-69), as well as that Pilates, aerobic exercise, resistance exercise, highintensity training, and intermittent walking training significantly improved walking endurance in people with MS (28,68,(70)(71)(72).Furthermore, fast-velocity concentric resistance training may have a greater effect on walking endurance with greater neural adaptations in a shorter period of time (28).A meta-analysis showed that walking training programs significantly improved functional ability (mobility, walking endurance, and gait speed), possibly due to improved walking economy (68).The mechanisms by which exercise improves walking ability and walking endurance in people with MS may be improvements in maximal oxygen uptake, muscular strength, and fitness.The increase in muscle strength is due to improved firing and synchronization of motor units and improved synergistic coordination of agonists and antagonists (73).Moreover, another mechanism may be increased bilateral symmetry, which reduces the amount of time the lower limbs are supported on the ground (74).
Early fatigue in people with MS presents with common symptoms such as decreased endurance and muscle strength (75).Statistically, fatigue affects approximately two-thirds of people with MS (76).Current evidence suggests that pharmacological interventions are largely ineffective and that exercise significantly reduces fatigue in people with MS (77, 78).Our results showed that exercise significantly improved fatigue in people with MS, which was consistent with the results of Taul-Madsen et al. (79), showing that aerobic exercise is effective in reducing perceived fatigue in people with MS.The mechanism by which exercise improves fatigue may be an improvement in cardiorespiratory fitness, which increases available energy reserve and reduces fatigue.In addition, exercise may induce upregulation of neuroendocrine growth factor secretion, which increases neuronal plasticity and thus may improve compensatory cortical activation (80, 81).Furthermore, exercise-induced upregulation of anti-inflammatory cytokines may have beneficial effects on fatigue (82)(83)(84).
Resistance exercise has been reported to be an effective intervention to ameliorate physical and generalized fatigue and result in significant changes in muscle strength and postural stability (85).Subgroup analysis showed that aerobic exercise, resistance exercise, and multicomponent training were effective in improving fatigue in people with MS, with resistance exercise being the most effective intervention type, which may be due to the fact that resistance exercise is well-tolerated in people with MS, restores the ability to respond quickly to stimuli, and improves autonomy when walking (13).Previous studies have shown that motor and cognitive function deteriorate with age in adult people with MS and that older people with MS exhibit worse cognitive performance (86)(87)(88)(89).Therefore, we conducted a subgroup analysis based on the participants' age and the results showed that a younger age was associated with larger improvement in fatigue.Horton et al. (90) showed that with age, people with MS develop a sedentary lifestyle, which increases the risk of secondary disease.Although exercise is an effective therapy, dyskinesia is common in older adult patients.Increased fatigue is a severe barrier when exercise energy expenditure is relatively high, and older patients can lose confidence in their ability to exercise and may feel at risk of injury, especially when exercise equipment is involved (91-94).
In addition, exercise significantly improved the quality of life in people with MS, which was consistent with a previous study, showing that exercise seems to be the most effective way to improve the quality of life in people with MS by increasing strength and balance, thereby reducing the risk of falls (94).Previous studies have shown that multicomponent training is well-tolerated and can effective in improving the quality of life in people with MS (13), and that group exercise is an effective intervention for people with MS to cope with fatigue, with the Baduanjin playing a more prominent role in improving the quality of life through respiration and psychology (46).Improvements in quality of life may be related to exercise-induced increases in fitness, mobility, balance, muscle strength, and sleep quality (53,95).Subgroup analysis showed that aerobic exercise and multicomponent training were effective in improving quality of life, with aerobic exercise being the most effective intervention type, which was in agreement with previous studies, showing that aerobic exercise increases aerobic capacity and improves physical and mental health, thereby enhancing functional independence and fatigue resistance in people with MS (96).In addition, aerobic exercise may stimulate the activity of the sympathetic nervous system and activate the activity of the parasympathetic nervous system, which leads to the release of acetylcholine, resulting in a sedative effect (97).

Strengths and limitations of this systematic review
In this systematic review and meta-analysis, we included studies on the effect of exercise on balance, walking ability, walking endurance, fatigue, and quality of life in people with MS, and excluded studies where participants in the control group received exercise interventions, which can better reflect the effect of exercise interventions.Our findings provide an alternative treatment strategy for people with MS, clinically recommending engagement in resistance exercise and aerobic exercise, respectively, to alleviate fatigue and enhance quality of life.
However, this study has some limitations that should be noted.First, the heterogeneity between each of the original studies is unavoidable (the proportion of male and female participants from different regions, the age of subjects, etc.), which may affect the scientific validity of the meta-analysis.Second, many of the included studies had small sample sizes, which may have had some impact on the results.Finally, it was not possible to exclude a placebo effect, as blinding could not be performed during the exercise intervention.Future reviews could reduce the heterogeneity between included studies by restricting the inclusion criteria more strictly.

Conclusion
This meta-analysis revealed that exercise had beneficial effects in improving balance, walking ability, walking endurance, fatigue, and quality of life in people with MS.The effect of exercise on improving fatigue was associated with the age of the participants, with the younger the age, the greater the improvement in fatigue.To improve fatigue and quality of life, this meta-analysis provides clinicians with evidence to recommended that people with MS participate in resistance exercise and aerobic exercise, respectively.

FIGURE
FIGUREPRISMA flow diagram of the study selection process.

FIGURE
FIGUREMeta-analysis results of the e ects of exercise on Berg balance scale (BBS) in people with MS.Exercise had a significant e ect on improving BBS in people with MS (P < .).