A predictive model for functional outcomes in patients with acute anterior circulation stroke based on DWI-ASPECTS and serum homocysteine

Background Precise forecasting of functional outcomes is crucial for managing acute anterior circulation stroke (AACS). We sought to develop and validate a predictive model for functional outcomes by integrating the DWI-ASPECTS score and serum homocysteine levels. Methods We retrospectively examined information from consecutive AACS patients admitted between January 2022 and January 2025. The individuals were classified into favorable (modified Rankin Scale [mRS] ≤ 2) and unfavorable (mRS >2) outcomes groups depending on their 90-day mRS score. Upon admission, clinical parameters, including the National Institutes of Health Stroke Scale score, were assessed. Diffusion-weighted imaging was performed within 24 h to calculate the DWI-ASPECTS. Blood samples taken from a vein after an overnight fast were gathered within a 24-h timeframe for laboratory testing, including serum homocysteine, glycated hemoglobin, and lipid profiles. Results We included 122 patients, among whom 68 cases belonged to the favorable and 54 cases belonged to the unfavorable outcome group. Unfavorable outcome patients were significantly older (P = 0.017), exhibited higher baseline NIHSS scores, lower DWI-ASPECTS scores, and demonstrated elevated serum homocysteine levels (all P < 0.001). Multivariate analysis showed that DWI-ASPECTS score (OR = 0.270, P < 0.001) and serum homocysteine level (OR = 1.539, P < 0.001) independently predicted functional outcome. DWI-ASPECTS (AUC = 0.852) and homocysteine (AUC = 0.844) demonstrated high predictive accuracy. The final nomogram model, which combined these two variables, showed excellent discrimination with an AUC of 0.939 and was well-calibrated. Conclusion The developed nomogram, incorporating DWI-ASPECTS and serum homocysteine, demonstrates potential as an effective and accurate method for forecasting functional outcomes in AACS patients, offering a novel approach that integrates neuroanatomical and systemic metabolic information beyond models relying on single parameters.