Although the impact of malnutrition in patients with acute stroke has been reported, its significance after rehabilitation is not well understood. The geriatric nutritional risk index (GNRI) is a simple and well-established nutritional screening tool that predicts poor prognosis in elderly patients and in those with a high risk of cardiovascular events. We investigated the associations between GNRI and all-cause mortality, cardiovascular events, and infectious diseases in patients with stroke after rehabilitation. This study included 138 patients aged 80 years or below who were discharged between 2010 and 2013 in a single center, and followed up for more than 1 year. Malnutrition was defined as a GNRI of 96 or lower. The mean age was 63.9 ± 11.0 years, the mean GNRI at discharge was 98.8 ± 6.5, and the mean total functional independence measure (FIM) score at discharge was 91.8 ± 25.8. Among the patients, 37 (27%) had malnutrition. During the follow-up period, all-cause mortality, cardiovascular events, and infectious diseases were recorded in 11 (8%), 21 (15%), and 20 (15%) patients, respectively. Kaplan-Meier curves showed a significantly higher incidence of each outcome in patients with a GNRI of 96 or lower. In the Cox proportional analysis, GNRI was an independent determinant of all-cause mortality (hazard ratio [HR], .71; 95% confidence interval [CI], .61-.83), cardiovascular events (HR, .87; 95% CI, .80-.95), and infectious diseases (HR, .80; 95% CI, .74-.87) after adjusting for age, gender, and total FIM score. Malnutrition has a negative impact on prognosis in patients with stroke even after rehabilitation.