Objective: Musculoskeletal pain commonly occurs in the elderly, many of whom are also prone to suffer from strokes. We studied whether short-term use (≤ 4 weeks) of cyclooxygenase-2 (COX-2) inhibitors for musculoskeletal pain in stroke patients helped them to participate in their therapies and was safe and efficacious. Materials and Methods: Three hundred and three patients admitted consecutively with first ischemic stroke were studied. Two cohorts were defined, based on whether patients with acute stroke had sufficient musculoskeletal pain that warranted oral COX-2 inhibitors (COX-2 group) or not (case-matched controls). Primary efficacy measures were change in Fugl-Meyer (F-M) pain score and change in total functional independence measure (TFIM) scores on discharge from hospital. Safety was judged by the incidence of vascular episodes during the study period. Results: From the original 303 patients, 64 patients in the COX-2 group were matched with 64 patients in the non-COX-2 group. The groups were matched for age (±5 years), gender, and admission TFIM score (± 5 points). Baseline characteristics between the 2 groups were similar. The primary and secondary outcome measures were similar between the 2 groups, except for ambulation endurance, which favored the non-COX-2 group (P < 0.03). Greater change in the pain score (less pain) was found in the COX-2 group; this effect was strongest in patients who were independent prior to their stroke (on post hoc analysis). There were too few adverse events in either group of any significance. Conclusions: The short-term use of COX-2 inhibitors reduced musculoskeletal pain in acute stroke patients, improved functional motor outcome, and were found to be safe.