Persistent shoulder pain in the first 6 months after stroke: results of a prospective cohort study.

OBJECTIVE: To identify factors associated with persistent poststroke shoulder pain (pPSSP) in the first 6 months after stroke. DESIGN: Prospective inception cohort study. SETTING: Stroke units of 2 teaching hospitals. PARTICIPANTS: Patients (N=31) with a clinical diagnosis of stroke. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: The development of pPSSP within the first 6 months after stroke. Clinical assessment of motor, somatosensory, cognitive, emotional, and autonomic functions, undertaken within 2 weeks (t0), at 3 months (t1), and at 6 months (t2) after stroke. RESULTS: Patients with pPSSP (n=9) were compared with patients without pPSSP (n=22). Bivariate logistic regression analyses showed that pPSSP was significantly associated with impaired voluntary motor control (t0, t1, t2), diminished proprioception (t0, t1), tactile extinction (t0), abnormal sensation (t1, t2), spasticity of the elbow flexor muscles (t1, t2), restricted range of motion (ROM) for both shoulder abduction (t2) and shoulder external rotation (t1, t2), trophic changes (t1), and type 2 diabetes mellitus (t0). CONCLUSIONS: These findings suggest a multifactorial etiology of pPSSP. The association of pPSSP with restricted, passive, pain-free ROM and signs indicative of somatosensory sensitization may implicate a vicious cycle of repetitive (micro)trauma that can establish itself rapidly after stroke. Intervention should therefore be focused on maintaining and restoring joint ROM as well as preventing injury and somatosensory sensitization. In this perspective, strategies that aim to intervene simultaneously at various levels of function can be expected to be more effective than treatment directed at merely 1 level.