Abstract
BACKGROUND: Stroke recurrence after symptomatic non-acute middle cerebral artery occlusion (SNMCAO) remains difficult to predict, with existing models showing limited accuracy. OBJECTIVE: To investigate the factors associated with recurrent stroke in SNMCAO and validate a prediction model. METHODS: This retrospective study included 150 patients with SNMCAO admitted to our hospital between August 2023 and October 2024. Data were obtained from the electronic medical record system. Patients were divided into recurrent (n = 35) and non-recurrent groups (n = 115) based on stroke recurrence. Risk factors were screened using univariate analysis, least absolute shrinkage and selection operator (LASSO) regression, and multifactorial logistic analysis. A nomogram prediction model was established and validated using R software. RESULTS: Hypertension (yes/no, 25/10 vs. 46/69; 71.4% vs. 40.0%), smoking (yes/no, 9/26 vs. 9/106; 25.7% vs. 7.8%), mRS score (≤ 2/>2, 20/15 vs. 93/22; 57.1% vs. 80.9%), deep tiny flow voids (DTFV) (yes/no, 18/17 vs. 89/26; 51.4% vs. 77.4%), intraplaque hemorrhage (yes/no, 21/14 vs. 36/79; 60.0% vs. 31.3%), and incomplete anterior circulation (yes/no, 22/13 vs. 38/77; 62.9% vs. 33.0%) showed significant differences. LASSO regression identified hypertension, smoking, DTFV, and incomplete anterior circulation without multicollinearity. Logistic regression showed these were risk factors for recurrent stroke (OR = 3.750, 4.077, 3.233, 3.653, P < 0.05). The nomogram showed an AUC of 0.684 (95% CI: 0.586-0.782), with good calibration (Brier Score: 0.163, p = 1) and positive clinical decision curve analysis. CONCLUSION: Hypertension, smoking, DTFV, and incomplete anterior circulation are risk factors for recurrent stroke in SNMCAO. The nomogram provides moderate predictive performance for risk stratification.