Which exercise or behavioral interventions are most effective for treating fatigue in multiple sclerosis?

Abstract

Background: Fatigue is one of the most common and disabling symptoms of Multiple Sclerosis (MS). There are no recognised biomedical treatments for MS fatigue. The aim of this study was to determine the most promising exercise and behavioral treatments for MS fatigue by (a) providing a narrative synthesis of all the interventions where fatigue was a primary outcome including a breakdown of the key contextual and treatment components (b) a network meta-analysis of all controlled exercise and behavioural trials conducted to date where fatigue was a primary or secondary outcome. Methods: 17 databases were searched, papers hand-searched, and relevant authors contacted. Studies were screened and assessed for eligibility by two independent reviewers. Information was extracted according to Cochrane and TIDieR guidance and statistical data were double extracted. Treatment manuals were obtained and coded by content of intervention, treatment context and relevant BCTs and interventions grouped accordingly. Effect sizes were expressed as standardised mean differences (SMD). The network meta-analysis was estimated using restricted maximum likelihood by the mvmeta command and network packages. Results: 1880 abstracts and 227 full text articles were screened. 82 studies with 179 treatment arms were included in the meta-analysis. SMD ranged from -0.07 (-0.33, 0.47) for resistive exercise to -0.61 (-0.85,-0.37) for exercise with more than one overload component and -.91 (-1.37, -0.45) for balance exercise. Cognitive behavioural therapy (CBT) was the most effective behavioral treatment (SMD=-.61; 0.85, -0,37). CBT for fatigue had higher estimates SMD=-.77 (-1.21, -.33) than CBT for distress SMD=-.56 (-.88, -.23). Combining behavioural and exercise programs showed no added benefit. Discussion: Balance and exercise programmes with more than one overload/progression component and CBT for fatigue appear the most promising interventions for MS fatigue. BCTs did not discriminate across interventions. More detailed knowledge of the content was important for developing relevant groups of interventions.