Discharge Neurological Deficit as a Predictor of Early Stroke Recurrence: A Nationwide Registry-Based Propensity-Matched Study.

BACKGROUND: Early stroke recurrence is a significant risk for survivors of ischemic stroke. Although admission severity is commonly used in prognosis, the predictive value of residual neurological deficits at discharge remains underexplored. We aimed to determine whether discharge National Institutes of Health Stroke Scale (NIHSS) score predicts 1-year ischemic stroke recurrence, using a large multicenter registry and propensity score matching. METHODS: We analyzed 39 947 patients with first-ever ischemic stroke from the KRS (Korean Registry of Regional Cardiocerebrovascular Center for Stroke) (2014-2022). NIHSS scores were recorded at admission (baseline) and at discharge. Patients were dichotomized by discharge NIHSS score (<5 versus ≥5); supplementary cutoffs of 16 and 21 were explored. Propensity score matching (1:1 nearest neighbor) was used to adjust baseline covariates. The primary outcome was stroke recurrence within 1 year, defined as a new neurological deficit confirmed by imaging. Kaplan-Meier survival and Cox regression were used to evaluate predictors. RESULTS: After propensity score matching, discharge NIHSS score was a significant predictor of recurrence (hazard ratio per point, 1.03 [95% CI, 1.00-1.05]; P=0.028). Higher discharge mRS score and hypertension were also significantly associated with recurrence. Patients with NIHSS score ≥5 had 20% to 30% higher recurrence rates across analyses. Stroke subtype analysis showed that small artery occlusion conferred lower recurrence risk. CONCLUSIONS: Discharge NIHSS score significantly reflects risk of early recurrence compared with initial stroke severity. These findings support incorporating discharge neurological status into poststroke risk stratification and follow-up planning.