Roy G Elbers, John Verhoef, Erwin EH van Wegen, Henk W Berendse, Gert Kwakkel

Factors contributing to subjective fatigue in people with idiopathic Parkinson's disease (PD) are not well known. This makes it difficult to manage fatigue effectively in PD. To evaluate the effects of pharmacological and non-pharmacological interventions, compared to an inactive control intervention, on subjective fatigue in people with PD. We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library); MEDLINE (via PubMed); Ovid EMBASE; EBSCO CINAHL; Ovid PsycINFO; PEDro; and the WHO International Clinical Trials Registry Platform Search Portal up to April 2015. References of included studies and identified review articles were screened for additional studies. There were no restrictions based on language, date of publication or study setting. Randomised controlled trials (RCTs) that report on subjective fatigue in people with PD. Two review authors independently performed study selection, data collection and risk of bias assessments. Eleven studies were eligible for this systematic review, with a total of 1817 people. Three studies included only people who experienced clinically relevant fatigue (Fatigue Severity Scale score ≥ 4 out of 7 or Multidimensional Fatigue Inventory total score > 48 out of 100), whereas all other studies did not select participants on the basis of experienced fatigue. Nine studies investigated the effects of medication (i.e. levodopa-carbidopa, memantine, rasagiline, caffeine, methylphenidate, modafinil or doxepin) on subjective fatigue. All studies were placebo controlled. There was insufficient evidence to determine the effect of doxepin on the impact of fatigue on activities in daily life (ADL) or fatigue severity (one study, N = 12, standardised mean difference (SMD) = -1.50, 95% confidence interval (CI) -2.84 to -0.15; low quality evidence). We found high quality evidence that rasagiline reduced or slowed down the progression of physical aspects of fatigue (one study, N = 1176, SMD = -0.27, 95% CI -0.39 to -0.16, I(2) = 0%). None of the other pharmacological interventions affected subjective fatigue in PD. With regard to adverse effects, only levodopa-carbidopa showed an increase for the risk of nausea (one study, N = 361, risk ratio (RR) = 1.85, 95% CI 1.05 to 3.27; high quality evidence). Two studies investigated the effect of exercise on fatigue compared with usual care. We found low quality evidence for the effect of exercise on reducing the impact of fatigue on ADL or fatigue severity (two studies, N = 57, SMD = -0.45, 95% CI -1.21 to 0.32, I(2) = 44%). Based on the current evidence, no clear recommendations for the treatment of subjective fatigue in PD can be provided. Doxepin may reduce the impact of fatigue on ADL and fatigue severity; however, this finding has to be confirmed in high quality studies. Rasagiline may be effective in reducing levels of physical fatigue in PD. No evidence was found for the effectiveness of levodopa-carbidopa, memantine, caffeine, methylphenidate, modafinil or exercise. Studies are needed to investigate the effect of exercise intensity on exercise capacity and subjective fatigue. Future studies should focus on interventions that address the maladaptive behavioural or cognitive aspects of fatigue in people with PD. Characteristics, such as severity and nature of perceived fatigue and underlying mood disorders should be considered to identify responders and non-responders when studying interventions for fatigue. The development of a core-set of self-report fatigue questionnaires with established responsiveness and known minimal important difference values will facilitate the interpretation of change in fatigue scores.