Physical activity trajectories after mild ischemic stroke and their association with 6-month functional outcomes

Whether post-stroke physical activity (PA) patterns are associated with stroke prognosis remains unclear. We aimed to examine the association between post-stroke PA trajectories and 6-month functional outcomes in patients with mild ischemic stroke.

This prospective cohort study included patients with mild ischemic stroke (National Institutes of Health Stroke Scale [NIHSS] score ≤ 5) admitted to a teaching hospital between July 2024 and March 2025. PA was assessed using the Saltin-Grimby Physical Activity Level Scale at pre-stroke, 1, and 3 months. PA trajectories from pre-stroke to 3 months post-stroke were identified using growth mixture modeling, yielding three PA trajectories: stable low, increased, and decreased. The primary outcome was functional independence (defined as a score of 0-2 on the modified Rankin Scale; range, 0-6, with higher scores indicating more severe disability) at six months. Multivariable logistic regression was used to examine the associations between PA trajectories and the primary outcome, adjusted for prespecified baseline covariates (age, sex, hypertension, diabetes, and atrial fibrillation).

Among 502 patients (median age, 66 years [IQR 57-72]; 354 males [70.52%]), functional independence at 6 months was achieved in 166 patients with stable low PA (80.58%), 177 with increased PA (88.06%), and 74 with decreased PA (77.89%). The overall three-level test for the trajectory variable was not statistically significant (P = 0.086). In an exploratory pairwise comparison, compared with decreased PA, the odds of functional independence were 2.03 (95% CI 1.05-3.90; P = 0.034) for increased PA and 1.29 (95% CI 0.70-2.36; P = 0.418) for stable low PA.

Although the overall trajectory test was not significant, an exploratory pairwise comparison suggested that an increased PA trajectory was associated with better functional independence compared with a decreased trajectory. These findings should be considered exploratory and require external validation; causality cannot be inferred.