The time-course of post-stroke fatigue: A systematic review of longitudinal cohort studies

To systematically review longitudinal studies to determine the prevalence and time-course of fatigue after stroke (post-stroke fatigue, PSF). A study protocol was registered on PROSPERO. Five databases (PUBMED, MEDLINE, EMBASE, PSYCHINFO and CINAHL) were searched (10th to 13th June 2022). Citations were imported into Covidence software, abstracts screened by one author, full texts of potentially eligible studies retrieved, and one author applied inclusion criteria (longitudinal cohort studies of patients with acute stroke). Quality assessment of included studies was performed using the Joanna Briggs institute tool for observational studies. A meta-analysis was performed for the prevalence of PSF at different time-points after stroke onset, and changes over time. Subgroup analyses were performed by type of stroke and study location. A total of 13,991 records were returned from the searches. Nine studies were eligible and were included. Five studies were of strong and four of moderate quality. Of the studies suitable for meta-analysis, the prevalence of PSF was 42% (95% CI – 39–44%) at six months after ischaemic stroke; and 34% (95% CI – 28–40%) at one year in stroke survivors excluding subarachnoid haemorrhage. Subgroups analyses found no differences in PSF prevalence between Asian countries and others. Of those with PSF at first assessment, 66% (95% CI – 61–71%) remained fatigued at follow-up; of those without PSF initially, 15% (95% CI – 11–20%) developed PSF at follow-up. PSF is common and around two-thirds with fatigue remain fatigued. This justifies the development of new interventions for PSF treatment.