Shoulder pain in people with a stroke: a population-based study

Abstract

Objective: To measure the occurrence of shoulder pain after stroke. To identify the factors that predict risk of shoulder pain after stroke. Design: Auckland Stroke Study, population-based case-cohort study. Setting and subjects: All cases of stroke, including those managed outside hospital, over a 12-month period ending February 1992 were considered in Auckland. Outcome measures: Self-reported shoulder pain at one week, one month and six months after the onset of stroke for each person. Results: A total of 1761 stroke events were identified. Self-reported shoulder pain among survivors increased from 256/1474 (17%) at one week, to 261/1336 (20%) at one month and 284/1201 (23%) at six months. Shoulder pain was positively associated with motor deficit, side of deficit and severity of deficit. In those surviving to six months after stroke, the risk of shoulder pain was higher in those with severe upper limb motor deficit (odds ratio (OR) 4.94; 95% confidence interval (CI) 3.06–7.98) and in diabetics (OR 1.57, 95% CI 1.15–2.14). Risk of shoulder pain increased with time and was lower for those in institutional care. Conclusion: Shoulder pain after stroke is common, especially in patients with severe sensorimotor deficits, diabetics and those living at home. Appropriate management may reduce the rate of occurrence.