INCIDENCE, TYPES AND NATURE OF POST-STROKE PAIN: SYSTEMATIC REVIEW OF LITERATURE AND META-ANALYSIS

Introduction
Disability following stroke is mainly a result of the stroke’s neurological deficit, but other factors contribute, including pain. Literature suggests that around half of stroke survivors experience pain after their stroke. However, it is possible that pain after stroke is under-reported, because of lack of systematic identification of pain, and because some stroke survivors cannot communicate if they are in pain or not. Three reviews (published in 2015, 2017 and 2017 respectively) of post-stroke pain have been identified in literature so far. None of the three reviews were prospectively registered to PROSPERO (the last literature searches were performed in 14th October 2018) and there was no meta-analysis of post-stroke frequency or pain type. New studies have been published since the last review, and therefore it is timely to perform a new systematic review and meta-analysis of post-stroke pain.

Methods
We systematically reviewed the literature using a comprehensive search strategy developed by the Cochrane Stroke Information Specialist. Electronic search of the databases CENTRAL, MEDLINE, EMBASE, Epistemonikos, CINAHL EBSCO, PsycINFO and Web of Science was performed, from time of inception until November 2018, to identify observational studies (cross-sectional, cohort, longitudinal, case-control studies) of adult human participants (18 years or older), who suffered a first ever or recurrent stroke (ischemic and/or haemorrhagic) that examined the type, frequency and nature of post-stroke pain. Covidence Review Software was used for collecting and processing the data by two different reviewers - two authors screened abstracts and then two authors screened full text papers.

Findings
Database search yielded a total of 18, 445 references. After initial citation screening, 240 studies were eligible for full text review. Among 63 eligible studies extracted to date, frequency of post-stroke pain ranged between 7.4 – 71%, with pain frequency increasing between the acute phase and 6 months post-stroke. Hemiplegic-shoulder pain was most common (4 - 58%), followed by headache (0.8 - 40.7%) and Central post-stroke pain (5-21.8%).

Conclusions
Frequency of post-stroke pain was found to affect up to 71% of this patient population, more than anticipated prior to conducing this systematic review and meta-analysis. There were no qualitative studies identified, and therefore it was not possible to establish how post-stroke pain manifests, how it is managed, and opportunities for addressing the issue.