Which behavioural and exercise interventions targeting fatigue show the most promise in multiple sclerosis? A systematic review with narrative synthesis and meta-analysis

Highlights

• Exercise and behavioural interventions had moderate to large effects on MS fatigue.

• The quality of evidence was moderate for behavioural but poor for exercise studies.

• There was good quality evidence of non-significant effects of energy conservation.

• Web-based CBT and balance interventions showed promise but require large trials.

• Based on limited evidence, combined interventions did not show added benefits.

Abstract

Fatigue is a common and highly debilitating symptom of multiple sclerosis (MS). This meta-analytic systematic review with detailed narrative synthesis examined randomised-controlled (RCTs) and controlled trials of behavioural and exercise interventions targeting fatigue in adults with MS to assess which treatments offer the most promise in reducing fatigue severity/impact. Medline, EMBASE and PsycInfo electronic databases, amongst others, were searched through to August 2018. Thirty-four trials (12 exercise, 16 behavioural and 6 combined; n = 2,434 participants) met inclusion criteria. Data from 31 studies (n = 1,991 participants) contributed to the meta-analysis. Risk of bias (using the Cochrane tool) and study quality (GRADE) were assessed. The pooled (SMD) end-of-treatment effects on self-reported fatigue were: exercise interventions (n = 13) -.84 (95% CI -1.20 to -.47); behavioural interventions (n = 16) -.37 (95% CI -.53 to -.22); combined interventions (n = 5) -.16 (95% CI: -.36 to .04). Heterogeneity was high overall. Study quality was very low for exercise interventions and moderate for behavioural and combined interventions. Considering health care professional time, subgroup results suggest web-based cognitive behavioural therapy for fatigue, balance and/or multicomponent exercise interventions may be the cost-efficient therapies. These need testing in large RCTs with long-term follow-up to help define an implementable fatigue management pathway in MS.

Introduction

Multiple sclerosis (MS) is a chronic, incurable, demyelinating disease of the central nervous system, usually diagnosed during young adulthood (Compston & Coles, 2008). An estimated 2.3 million people worldwide have MS with a 2:1 ratio of women to men (Atlas of MS, 2013). Around 85% of people with MS (pwMS) are diagnosed with relapsing remitting MS (RRMS), which includes periods of remission, interspersed with symptom relapses. After 10–20 years, many patients with RRMS go on to develop secondary-progressive MS (SPMS), where impairment accumulates over time. Around 15% of pwMS are diagnosed with primary-progressive MS (PPMS) characterised by chronic-progressive worsening of symptoms and disability from onset (Compston & Coles, 2008; Reich, Lucchinetti, & Calabresi, 2018).

The disease causes a range of symptoms and associated disabilities, including loss of mobility, spasticity, sensory disturbances, impaired balance, slowed cognitive processing, incontinence, pain, and fatigue depending on the site of neuronal damage (Compston & Coles, 2008). Fatigue is one of the most common, reported by around 80% of pwMS. Two-thirds consider fatigue their most troubling symptom (Giovannoni, 2006). It is one of the most commonly cited reasons for stopping work and a predictor of poor quality of life even when controlling for disease severity (Branas, Jordan, Fry-Smith, Burls, & Hyde, 2000; Krupp, Serafin, & Christodoulou, 2010).

In the context of chronic medical illnesses, such as MS, fatigue is defined as a lack of energy, feeling of exhaustion or overwhelming sense of tiredness that can be physical or mental or both. This fatigue is not relieved by rest and may be unrelated to physiological exertion (Bleijenberg, 2003; Krupp, 2003; NICE, 2015). Fatigue is one of the least understood symptoms in MS. Evidence to date suggests that primary disease factors, such as demyelination, axonal loss or damage, and inflammatory disease activity only play a small part in MS fatigue (Krupp et al., 2010; van Kessel & Moss-Morris, 2006). According to older evidence significantly higher levels of fatigue were observed among people with progressive forms of MS compared to those with RRMS (Bergamaschi, Romani, Versino, Poli, & Cosi, 1997; Colosimo et al., 1995). However, more recent evidence suggests that there are no significant differences in fatigue by disease course after controlling for disability and duration of symptoms (Kroencke, Lynch, & Denney, 2000; Lerdal, Gulowsen Celius, Krupp, & Dahl, 2007; Patrick, Christodoulou, Krupp, & Consortium, 2009). Fatigue directly related to the disease mechanisms of MS can be referred to as primary fatigue (Langeskov-Christensen, Bisson, Finlayson, & Dalgas, 2017).

Recent systematic reviews suggest medications often used to treat MS fatigue, such as amantadine and aspirin, have low efficacy and that non-pharmacological interventions (both exercise and psychological/educational) may have more beneficial effects on reducing the severity of fatigue (Asano & Finlayson, 2014; Khan, Amatya, & Galea, 2014). This may be because a wide range of psychosocial and secondary factors contribute to fatigue in MS, including poor sleep, low mood, deconditioning, and unhelpful cognitive behavioural responses to fatigue (Krupp et al., 2010; van Kessel & Moss-Morris, 2006). Fatigue associated with these non-disease-specific factors is defined by some researchers as secondary fatigue which may be treatable through behavioural methods (Langeskov-Christensen et al., 2017). Clinical guidelines suggest behavioural methods and exercise be incorporated in treatments for MS fatigue, but the nature of these treatments is poorly specified (National Institute for Health and Care Excellence, 2014).

Meta-analytic systematic reviews of exercise and/or behavioural interventions for the management of fatigue in MS already exist. Two focused on the effects of exercise therapy only (Heine, van de Port, Rietberg, van Wegen, & Kwakkel, 2015; Pilutti, Greenlee, Motl, Nickrent, & Petruzzello, 2013), one on yoga (Cramer, Lauche, Azizi, Dobos, & Langhorst, 2014), three on behavioural interventions, including energy conservation (EC; Blikman et al., 2013), cognitive behavioural therapy (CBT; van den Akker et al., 2016), and patient education (Wendebourg et al., 2017). Only Asano and Finlayson (2014)'s systematic review included pharmacological, exercise and behavioural interventions, but no consideration was given to the heterogeneity of interventions within the exercise and behavioural categories. Except for three systematic reviews (Asano & Finlayson, 2014; van den Akker et al., 2016; Wendebourg et al., 2017), none of the other reviews specified fatigue as a primary outcome. Furthermore, van den Akker et al. (2016)'s review found variability in intervention content, suggesting that even CBT is not one entity.

The combined evidence from these reviews has shown small to moderate effects of exercise and behavioural interventions on fatigue in MS; however, given the heterogeneity and complexity of such interventions, this evidence fails to unravel differences in efficacy by subtypes of exercise or behavioural interventions, limiting the clinical utility of the evidence syntheses. Firstly, previous reviews have not focused exclusively on interventions aimed at fatigue, and instead pooled outcomes across fatigue-specific and non-specific interventions. Although, improvements in mood or self-management are likely to lead to secondary benefits in fatigue, essential differences in content of therapeutically-similar interventions based on intervention target are overlooked. For instance, CBT for depression focuses on thoughts and behaviours relevant to low mood, while fatigue-specific thoughts and behaviours need to be addressed in CBT aimed at fatigue. In fact, according to a systematic review in cancer, psychosocial interventions were only superior to exercise interventions when aimed specifically at fatigue (Kangas, Bovbjerg, & Montgomery, 2008). Inevitably this also limits the consideration of treatment mechanisms pertinent to fatigue. Another important caveat of this is that trials evaluating interventions not aimed at fatigue specifically are less likely to screen patients for fatigue, which is likely to introduce a ceiling effect, again doing little to discern what therapeutic approaches need to be considered in the management of fatigue.

It is also still unclear which exercise and behavioural interventions are likely to have the greatest clinical utility, as an in-depth analysis of intervention components is not presented in the current reviews and interventions are often pooled in meta-analyses without any clear indication of how similar or different these interventions may be, making the relative effectiveness hard to determine. For example, whereas CBT for fatigue involves establishing balance in activity and rest by often gradually increasing activity (Chalder, 2014), energy conservation involves a systematic assessment of all daily activities and identifying ways of reducing energy expenditure (Packer, Brink, & Sauriol, 1995). This clearly highlights that pooling behavioural interventions without consideration of the potentially conflicting mechanisms of action specific to each approach may not be appropriate. Issues such as intervention delivery, dose, and homework practice are often not elucidated. Therefore, although a number of meta-analytic reviews are already available, key research questions remain unanswered. Hoffman and colleagues have recently argued for methods of reviewing complex intervention studies which enhance the clinical utility of the reviews (Hoffmann et al., 2017; Hoffmann & Walker, 2015). These include using the Template of Intervention Description and Replication (TIDieR) to extract and summarise the contextual factors relevant to an intervention (Hoffmann et al., 2014) and establishing the key components of complex interventions. The current systematic review incorporates these elaborated methods alongside meta-analysis of treatment efficacy. The overall aim was to provide a detailed description of all behavioural and exercise interventions for MS fatigue trialled to date against their relative potential efficacy and future treatment utility. The specific objectives are to:

• Provide a narrative synthesis of all the exercise and behavioural interventions explicitly designed to treat fatigue in MS, including a breakdown of the treatment components of each intervention, the delivery methods, and acceptability of the interventions (uptake and adherence).

• Calculate the effect size for outcomes of self-reported fatigue for each intervention based on the primary post-randomisation end-point (defined as being within two months following the stated duration of the intervention) and summarise risk of bias for each study.

• Create subgroups within the exercise and behavioural interventions based on key intervention components and conduct meta-analyses of post-treatment effect sizes of self-reported severity or impact of fatigue across each of these intervention subtypes.

• Where possible conduct meta-analyses of effect sizes of longer-term follow-up of self-reported fatigue outcomes across each of these intervention subtypes.¹

• Compare the overall standardised intervention effect sizes of the exercise and behavioural interventions.

• If possible, conduct exploratory moderator and sensitivity analyses to explore how treatment effects vary according to whether interventions were guided by theory or not, different levels of health care professional contact, types of MS, comparators used, and study quality.