If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, SwedenAcademic Specialist Center, Center of Neurology, Stockholm Health Services, SE-113 65 Stockholm, Sweden
Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, SwedenAcademic Specialist Center, Center of Neurology, Stockholm Health Services, SE-113 65 Stockholm, Sweden
Academic Specialist Center, Center of Neurology, Stockholm Health Services, SE-113 65 Stockholm, SwedenDepartment of Neurobiology, Care Sciences and Society, Karolinska Institutet, Stockholm, Sweden
Exercise studies including only fatigued persons with multiple sclerosis (PwMS) with fatigue as primary endpoint are lacking.
Objective
To evaluate the effects of high-intensity resistance training (HIRT) on self-reported fatigue in fatigued PwMS in a single center randomised controlled trial.
Methods
We recruited 71 PwMS scoring ≥ 53 on the Fatigue Scale for Motor and Cognitive Functions (FSMC), who were randomised 1:1 to either twice (group A) or once (group B) weekly supervised HIRT for twelve weeks. A non-randomised FSMC score-matched group (n=69) served as non-intervention control.
Results
Between HIRT-group differences were non-significant for primary and most secondary endpoints. Mean difference in FSMC score (95% confidence intervals) was -10.9 (-14.8; -6.9) in group A and -9.8 (-13.2; -6.3) in group B. Corresponding values for combined HIRT groups vs non-intervention control were -10.3 (-12.9; -7.7) and 1.5 (-0.6;3.6), respectively, p<0.001. Secondary endpoints also improved in both HIRT groups, though only Hospital Anxiety and Depression Scale anxiety and MS Impact Scale-29 psychological subscales significantly favoured the twice a week HIRT (group A). As an exploratory endpoint, changes in plasma inflammatory protein markers were associated with reduced FSMC scores in the pooled material.
Conclusion
The finding that HIRT in fatigued PwMS leads to clinically relevant reductions in self-reported fatigue, associated with changes in plasma inflammatory protein levels, provide evidence for recommending HIRT for fatigued PwMS.
Multiple sclerosis (MS) is one of the most frequent causes of neurological disability and loss of work capacity among young and middle-aged adults. Although the symptom spectrum experienced by persons with MS (PwMS) is very wide, both in terms of qualitative aspects and severity, fatigue is often cited as one of the most frequently reported and disabling impairments (
Fatigue can be described as a subjective feeling of exhausted physical and/or mental energy reserves to an extent that it restricts normal activities of daily living. It is present in approximately two-thirds of PwMS (
). Soluble inflammatory mediators such as pro-inflammatory cytokines like tumour necrosis factor (TNF) and Interleukin-6 (IL-6) have also been proposed to be involved in the pathophysiology of MS fatigue (
The evidence base for physical exercise as a therapeutic intervention for inflammatory diseases is growing both regarding its beneficial clinical effects and mechanism of action (
). This is in line with research identifying skeletal muscles as a secretary organ producing and releasing contraction dependent anti-inflammatory mediators (
). Adults with other chronic conditions are, in general, recommended muscle-strengthening activities twice a week (Physical Activity Guidelines Advisory
). However, frequent supervised exercise sessions over extended periods of time are resource demanding, which may constitute a problem when translating clinical study protocols to real world contexts.
We previously conducted a pilot-study of a 12-week supervised high-intensity resistance training (HIRT) programme in patients with relapsing remitting MS (RRMS), indicating clinically relevant improvements in fatigue, a decrease in anxiety and depression scores, improved HRQL and reduced levels of TNF in blood (
High-intensity resistance training in multiple sclerosis - an exploratory study of effects on immune markers in blood and cerebrospinal fluid, and on mood, fatigue, health-related quality of life, muscle strength, walking and cognition.
). However, to serve as an evidence base these preliminary findings need to be corroborated by a larger high-quality randomised controlled trial (RCT) in fatigued PwMS. Thus, the objectives here were to evaluate the effects of HIRT on functioning and HRQL in fatigued PwMS, and to explore associations between changes in fatigue with serum inflammatory protein markers, and to compare HIRT twice a week with once a week and using a matched non-randomised cohort as non-intervention control group.
2. Materials and methods
2.1 Study design and participants
This was a single-blinded RCT of HIRT in fatigued PwMS comparing HIRT twice a week with once a week, with an additional comparison of change in fatigue with a contemporary non-intervention control group.
Study participants were recruited during two screening epochs, in August 2020 and January 2021 respectively, at the largest MS clinic in Sweden; Center of Neurology (CfN), Stockholm. Inclusion criteria were age ≥ 18 years, a diagnosis of MS according to the revised 2017 McDonald Criteria, scoring at least moderate fatigue (score ≥ 53) on the Fatigue Scale for Motor and Cognitive Functions (FSMC) (
), ability to understand and communicate in Swedish, and not practicing high-intensity training within 6 months prior to the trial. Key exclusion criteria were comorbidity interfering with the possibility to engage in HIRT or to evaluate the endpoints, and pregnancy or breastfeeding. Participants were provided with both oral and written study information, and all provided a signed informed consent before study start. Approval was obtained from the Swedish Ethical Review Authority (2019-05105, 2020-05614) and procedures were conducted in accordance with the Helsinki Declaration. The study was registered with ClinicalTrials.gov (NCT04562376).
The contemporary non-intervention control group consisted of PwMS included in the prospective observational Combat-MS study (ClinicalTrials.gov NCT03193866) with a FSMC score ≥ 53 at assessments in September-October 2020 and who were re-assessed with the FSMC within 2-4 months.
2.2 Intervention
The HIRT programme followed the current guidelines concerning resistance training. The training consisted of a 5–10-minute warm-up on a stationary bicycle, four upper body exercises (pull down, push up, chest press, latissimus pull) and three lower body exercises (leg extension, leg curl, leg press) performed in training machines (HUR machines, https://www.hur.fi/en), and one whole body exercise (plank position), followed by a 5-10-minute cool down of stretching exercises. A progressive resistance training model was applied (Table 1). At the start of the intervention period, 1 repetition maximum (RM) for each machine-exercise was calculated after performance of 10RM tests, i.e., the load a person can manage for 10, but not 11 consecutive repetitions. This load is considered to represent 70% of 1RM. The maximum time for holding the plank position was considered 1RM for this exercise.
Table 1Description and progression of the intervention.
Participants had a 60 minutes HIRT session twice a week (group A) or once a week (group B) for 12 weeks at the Karolinska University Hospital, Stockholm, Sweden, under the supervision of a physiotherapist. Participants were provided with different options for HIRT sessions to increase adherence and they trained alone or in groups of maximum five persons/session.
2.3 Measures
Participant characteristics and disease-related data were collected for descriptive purposes. The primary endpoint was change in fatigue measured with the FSMC (
Development and preliminary testing of the new five-level version of EQ-5D (EQ-5D-5L). Quality of life research : an international journal of quality of life aspects of treatment.
As an exploratory endpoint, change of inflammatory protein markers in serum was measured using a highly sensitive multiplex technique (Olink Target 96 Inflammation, ref no. 95302, Uppsala, Sweden). A detailed description of the assay methodology and processing has previously been described (
). In short, the panel includes 92 immune-related proteins (Supplementary Table S1), and the assay utilizes epitope-specific binding and hybridization of a set of paired antibodies linked to oligonucleotide probes, which subsequently can be amplified using a quantitative polymerase chain reaction to quantify relative protein concentrations in terms of log base-two normalized protein expression (NPX) values.
2.4 Procedures
Data collections were performed at the CfN by a registered nurse, blinded to allocation, before (baseline) and after the 12-week intervention. Standard EDTA tubes were used for blood samples, which were spun down at 1500G for 15 min within 15-120 min directly after sampling. Serum samples were aliquoted and stored at -80◦C until analyses.
Randomisation was performed after baseline assessments by an independent nurse using a computer-generated random scheme (Sealed Envelope Ltd). A 1:1 allocation ratio of blocks of ten without stratification was used and implementation was done by means of sequentially numbered sealed opaque envelopes.
2.5 Data analysis
The sample size calculation was based on reported clinically relevant change in the primary endpoint fatigue and a standard deviation (SD) in a similar sample of fatigued PwMS (
). A total of 45 participants in each group were needed to detect a difference of 10 points in FSMC assuming a common SD of 11, with a power of 90%, at α=0.01 and a drop-out rate of 20%.
The cut-off for session adherence was set to attendance ≥75% of the sessions. Mean percentage increase in muscle strength for the seven body exercises performed in training machines were calculated for each participant and used as a parameter for adherence to the exercise protocol, i.e., content adherence. The cut-off for content adherence was set to a mean percentage increase ≥10%.
Descriptive statistics were used to present data. An intention to treat (ITT) approach was used for data analysis. The primary endpoint was after the 12-week intervention period. Linear repeated-measurement analyses were used for evaluating within-group (time) and between-group (time x group) effects for FSMC, FSS, HADS, OGQ, MSIS-29 and EQ-VAS. Drop-out analyses were performed with non-parametric methods. Based on FSMC scores participants were categorised as having no (20-42), mild (43-52), moderate (53-62) or severe (63-100) fatigue (
). Data from the 30-item OGQ identifying occupational gaps in the areas of instrumental activities of daily living, leisure, socialising, and work were summarized by calculation of the total number of identified gaps. Data from the EQ-5D-5L questionnaire were only presented with descriptive statistics.
Changes in inflammatory protein markers were analysed with paired samples t-tests. Analyses of associations between change in FSMC and immune-related protein levels were analysed using multivariable linear regression models adjusting for age, sex, and days between last HIRT session and follow-up sampling. In all analyses, only proteins with a call rate ≥75% were included, with additional corrections for potential effects of pre-analytical variability relating to handling procedures (see supplementary information). Volcano-plots were used to illustrate mean difference in NPX values and significance (P-values) from paired samples t-tests and multivariable linear regression analyses.
The significance level was set at P<0.05 for primary and secondary endpoints. For statistical analyses of inflammatory protein markers, a direct false discovery rate (FDR) correction of P-values with a significance cut-off of PFDR <0.05 were performed and an exploratory significance cut-off of P=0.05 were used. Analyses were performed using IBM SPSS Statistics 28 and SAS version 9.4 (SAS Institute).
3. Results
In the two screening epochs in August 2020 and January 2021, 81 and 63 potentially eligible PwMS, respectively, were screened, out of which 71 were enrolled in the study, 35 were allocated to HIRT twice a week (group A) and 36 to once HIRT a week (group B) (Fig. 1). Forty-four PwMS started their training period in September 2020 and 27 in February 2021. The contemporary non-intervention control group consisted of 69 PwMS. Descriptive baseline characteristics are presented in Table 2. Nineteen participants had been prescribed symptomatic pharmacological treatment for fatigue within one-year prior to baseline assessments and 16 had prescriptions of antidepressants.
Fig. 1Consort diagram giving an overview of the participant flow in the study.
Table 2Baseline characteristics of participants receiving high-intensity resistance training by group allocation, for all participants and for contemporary non-intervention control group.
Defined as minutes per week of aerobic physical activity according to the recommendation of physical activity for health benefits by the World Health Organization.
a Defined as minutes per week of aerobic physical activity according to the recommendation of physical activity for health benefits by the World Health Organization.
Eleven participants (15.5%) dropped-out of the study (Fig. 1), although only three (4%) were completely lost to follow-up (another five could not attend the follow-up meeting in person but sent in questionnaires by post). These 11 participants scored significantly higher fatigue (FSMC, p=0.016), anxiety (HADS, p=0.030) and HRQL (MSIS-29 physical, p=0.009) at baseline compared to participants who remained in the study.
Session adherence ranged between 25-100% in group A and between 33-100% in group B with 26 (74%) and 24 (67%) participants in group A and B, respectively, who completed at least 75% of the stipulated HIRT sessions. Thus, 50 (70%) participants fulfilled the criteria for session adherence. As for content adherence, 52 (73%) participants showed adherence to the exercise protocol. There were two reported adverse events: one patient dropped-out due to unpleasant sensory sensations, another patient reported chest pain which was diagnosed as Tietze's syndrome and who remained in the study.
The results of HIRT are shown in Table 3. There was no significant between-group difference in the primary endpoint with mean FSMC change score being reduced with approximately 10 points in both groups, i.e., representing a clinically meaningful change. A total of 28 participants (15 in group A and 13 in group B) decreased their FSMC score ≥ 10 points. The reduction in FSMC scores in the merged HIRT groups was larger than the non-intervention control, who retained similar scores between baseline and follow-up (Fig. 2, Table 3). At baseline, 66 HIRT participants were categorized as having severe fatigue and five as having moderate. At follow-up, 43 participants had severe fatigue, 16 moderate, five mild and four were categorized as having no fatigue. As for the secondary endpoints, there was a significant time effect for all except OGQ, and only HADS anxiety and MSIS-29 psychological showed a significant between-group difference in favour of group A (twice a week). Results from EQ-5D-5L questionnaire ratings of perceived problems are presented in Fig. 3.
Table 3Results of primary and secondary outcome measures, P-values are from intention to treat linear repeated-measurement analyses.
Fig. 2Mean Fatigue Scale for Motor and Cognitive Function (FSMC) scores and 95% confidence interval (CI) for combined high-intensity resistance training (HIRT) groups (n=68) and for the natural history cohort (n=69) at baseline and follow-up assessments.
Fig. 3Baseline (T0) and follow-up (T1) results of ratings of perceived problems (no; slight; moderate; severe; extreme) on the 5-level EQ-5D version (EQ-5D-5L) in combined high-intensity resistance training (HIRT) groups.
The median (interquartile range) time between last HIRT session and follow-up blood-sampling was 3 (2-10.5) days for group A, 6 (3-10) days for group B, and 5 (2-10) days for combined groups, n=63. As the change in FSMC was similar in both groups, results from the exploratory endpoint are reported from combined groups. Using a false discovery cut-off of 5% (PFDR <0.05) a significant up-regulation was found for 18 of 51 proteins, e.g., matrix metalloproteinase-1 (MMP-1), oncostatin-M (OSM) and vascular endothelial growth factor A (VEGF-A), in paired samples t-tests (Fig. 4). The multivariable linear regression model analyses showed that up-regulations in protein levels of TNF and Interleukin-17A (IL-17A) were significantly (PFDR <0.05) associated with improvement in FSMC (Fig. 4). Information on results in group A and B and exploratory analyses with significance cut-off of P=0.05 are presented in the supplementary information.
Fig. 4[A] Mean differences and significance (P) from paired samples t tests comparing baseline and follow-up levels of inflammatory protein markers in all participants (n=63) after supervised high-intensity resistance training (HIRT) for 12 weeks. [B] Associations between change in plasma profile of immune markers and decrease in Fatigue Scale for Motor and Cognitive Functions (FSMC) after HIRT for 12 weeks in all participants (n=63). Figure [B] summarize the mean difference in protein markers and significance (P) for every ten-point decrease in FSMC from multivariable linear regression model analyses, which were adjusted for age, sex, and days between last HIRT session and follow-up sampling. The horizontal dashed line indicates an exploratory cut-off of P=0.05 and proteins highlighted with filled circles are direct false corrected (PFDR<0.05).
This is, as far as we know, the first single-blind RCT specifically designed to assess effects of progressive resistance training on MS-related fatigue in fatigued PwMS. We found a significant and clinically relevant reduction in the primary endpoint fatigue in both our training groups as well as in comparison to the non-intervention control group. In addition, secondary endpoints on anxiety, depression and HRQL were significantly changed to the better over time in both training groups. We had hypothesised that HIRT twice a week would be superior to once a week. This was, however, only found in measures of anxiety (HADS) and HRQL (MSIS-29 psychological subscale), which favoured a higher exercise frequency.
Exercise RCTs having fatigue both as an inclusion criteria and primary endpoint in PwMS are rare. What has been studied previously are the effects of vestibular rehabilitation (
Efficacy of an internet-based program to promote physical activity and exercise after inpatient rehabilitation in persons with multiple sclerosis: a randomized, single-blind, controlled study.
Calling out MS fatigue: feasibility and preliminary effects of a pilot randomized telephone-delivered exercise intervention for multiple sclerosis fatigue.
) reported statistically significant reductions in self-reported fatigue compared to control. In contrast to our results, the change in fatigue after aerobic training was below the predefined cut-off for a clinically relevant change and for the self-management programme the effect size was similarly small. It cannot be deduced from current data if HIRT might be more effective in reducing fatigue than these other interventions and future studies are therefore needed to explore how fatigued PwMS respond to different types of interventions.
In addition to only include fatigued PwMS in our study, the limitation raised by Heijne et al (
). Thus, the mean reduction of approximately 1 point in FSS in both HIRT groups was not only statistically significant but also clinically relevant. Although the changes over time in the MSIS-29 subscales were significant and exceeded proposed change scores for moderate to high sensitivity and specificity for improvement (8 on the physical subscale and 6 on the psychological) (
Adherence and drop-out in randomized controlled trials of exercise interventions in people with multiple sclerosis: a systematic review and meta-analyses.
) we report important information for determining feasibility for HIRT from a service delivery perspective. The overall session and content adherence was 70% and 73%, respectively, marginally lower than reported pooled adherence of 77% (range 31-100%) for resistance interventions in MS. The drop-out rate of 15.5% was less than we had expected, but in line with other exercise interventions (
Adherence and drop-out in randomized controlled trials of exercise interventions in people with multiple sclerosis: a systematic review and meta-analyses.
). That the sessions were supervised by a physiotherapist might have contributed to the few adverse events. The fact that once weekly HIRT displayed the same effect as twice weekly facilitates clinical implementation. Based on our findings, we propose that healthcare providers should supply supervised HIRT for fatigued PwMS.
We found an up-regulation in several proteins after HIRT, e.g., MMP-1 and VEGF-A. These proteins are involved in muscle remodelling and angiogenesis, respectively, and have previously been found to increase in concentration in serum after resistance training in healthy men (
). Interestingly, TNF and IL-17A did not significantly change over time but were found to be significantly up-regulated and associated to a decrease in FSMC. This was surprising and contradicted findings from our pilot study (
High-intensity resistance training in multiple sclerosis - an exploratory study of effects on immune markers in blood and cerebrospinal fluid, and on mood, fatigue, health-related quality of life, muscle strength, walking and cognition.
). These findings can be explained by differences in time since last training session and degree of rest before sampling as these factors represents important confounders when interpreting inflammatory protein markers (
). Nevertheless, the findings reported here represent a biological correlate of training intensity, in turn correlating with impact on the primary endpoint.
This study had certain limitations. The study was conducted during the COVID-pandemic which to some degree affected the recruitment of study participants and their possibility to attend training sessions. Thus, the planned number of 45 participants in each HIRT group was not achieved, which potentially might have masked HIRT group differences. The fact that the non-intervention control group was not randomised restricts the validity of detected differences as compared with intervention groups. However, this choice was motivated by ethical aspects, also considering that randomisation to a control arm might lead persons to engage in voluntary physical activity. Sampling of plasma was not conducted at a set time period after last exercise session, which likely introduced a technical bias as changes in protein levels with physical activity displays dynamic temporal changes that are difficult to control for. Finally, most of the study participants had RRMS, an EDSS score ≤ 2.5 and were receiving highly potent monoclonal DMTs, which may impact on the generalizability of our findings to other MS populations.
In conclusion, we demonstrate that participation in once or twice weekly HIRT is associated with a clinically relevant reduction in self-reported fatigue scores among fatigued PwMS, with relevant improvements also in other patient-reported outcomes. In addition, improved fatigue scores were associated with changes in inflammatory protein levels in plasma. These findings provide evidence for recommending HIRT for fatigued PwMS.
5. Funding
Research reported in this article was partly funded through NEURO Sweden with additional framework support from the Swedish MRC (grant no. 2020-0270) and the Region of Stockholm. The funding sources had no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication.
CRediT authorship contribution statement
S. Englund: Software, Formal analysis, Investigation, Writing – original draft, Visualization. F. Piehl: Conceptualization, Methodology, Resources, Writing – original draft, Visualization, Funding acquisition. M. Kierkegaard: Conceptualization, Methodology, Software, Formal analysis, Resources, Writing – original draft, Visualization, Project administration, Funding acquisition.
Declaration of competing interest
Simon Englund declare that there is no conflict of interest; Fredrik Piehl received research grants from Merck KGaA and UCB, and fees for serving on DMC in clinical trials with Chugai, Lundbeck and Roche. Marie Kierkegaard has received honoraria for lectures from Novartis, Sanofi, Genzyme and Roche. There is no commercial entity relevant for potential conflicts.
Adherence and drop-out in randomized controlled trials of exercise interventions in people with multiple sclerosis: a systematic review and meta-analyses.
Efficacy of an internet-based program to promote physical activity and exercise after inpatient rehabilitation in persons with multiple sclerosis: a randomized, single-blind, controlled study.
Development and preliminary testing of the new five-level version of EQ-5D (EQ-5D-5L). Quality of life research : an international journal of quality of life aspects of treatment.
High-intensity resistance training in multiple sclerosis - an exploratory study of effects on immune markers in blood and cerebrospinal fluid, and on mood, fatigue, health-related quality of life, muscle strength, walking and cognition.
Calling out MS fatigue: feasibility and preliminary effects of a pilot randomized telephone-delivered exercise intervention for multiple sclerosis fatigue.
00613-7&id=gr1.jpg){kind=link}
00613-7&id=gr2.jpg){kind=link}
00613-7&id=gr3.jpg){kind=link}
00613-7&id=gr4.jpg){kind=link}