Risk Factors for Poststroke Shoulder Pain: A Systematic Review and Meta-Analysis.

OBJECTIVE

To identify the risk factors identified within 1-month poststroke that predict the onset of poststroke shoulder pain (PSSP) within the first year after stroke.

METHODS

Five databases (AMED, CINAHL, EMBASE, Medline, and PubMed) were searched from inception to April 2019. Prospective cohort studies that measured a potential risk factor for PSSP within the first month after stroke were included. Two authors independently reviewed and selected articles for inclusion. Risk of bias was assessed using the Quality in Prognosis Studies tool. Data extracted included raw data for odds ratio (OR) calculations, definition and measurement of pain, study limitations, and baseline characteristics of participants. The review was conducted following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines.

RESULTS

Nine articles were retrieved that met the inclusion criteria, and 6 presented data to use in meta-analysis. Fifty-four different factors were identified as potential risk factors. Meta-analysis was possible for 4 factors; sex (OR .93, confidence interval [CI] .75-1.15), laterality (OR .78, CI .59-1.05), diabetes (OR 2.09, CI 1.16-3.78), and history of shoulder pain (OR 2.78, CI 1.29-5.97). Reduced motor function in the upper limb was also identified as a significant risk factor through qualitative synthesis.

CONCLUSIONS

Reduced motor function in the upper limb, diabetes, and a history of shoulder pain were identified as significant risk factors for the development of PSSP within the first year after stroke. Recommendations to standardize future studies in this area have been made, and it is suggested that defining subtypes of PSSP may aid future interventional studies.