Abstract Background and aims Outcome prediction after endovascular thrombectomy (EVT) in the extended 6–24 hour window remains challenging, particularly in Asian populations with high rates of intracranial atherosclerotic stenosis (ICAS). We aimed to develop a prognostic model integrating early post-procedural clinical data and perfusion imaging to predict 90-day functional independence.. Methods This prospective cohort study (2022–2024) at a comprehensive stroke center in Vietnam enrolled 200 patients with anterior circulation large vessel occlusion treated with EVT in the 6–24 hour window. The primary endpoint was functional independence (90-day modified Rankin Scale 0–2). Hemodynamic parameters were derived using RAPID software. Predictors were identified via multivariable logistic regression using stepwise Akaike Information Criterion (AIC) selection, validated with bootstrap resampling. Model performance was assessed for discrimination (AUC) and calibration. Results The cohort (mean age 66.9 years; 70.5% male) was notable for a high prevalence of ICAS (48%). Functional independence was achieved by 46% of patients. The final "RAPID PREDICT" model incorporated 11 variables: age, sex, unwitnessed onset, hypertension, dyslipidemia, 24-hour NIHSS, onset-to-door time, onset-to-puncture time, Tmax 4s volume, modified TICI score, and occlusion site. The model demonstrated excellent discrimination (AUC 0.89) and good calibration. An online calculator was developed (https://rapidpredict.com). Conclusions Integrating 24-hour clinical data with perfusion imaging, RAPID PREDICT enables accurate prognostication for late-window EVT patients, proving effective in a cohort with significant ICAS burden. This tool supports expectation management and care planning, though external validation is required to confirm generalizability. Conflict of interest BINH NGUYEN PHAM: nothing to disclose, An Thai Thanh Nguyen: nothing to disclose, Thang Ba Nguyen: nothing to disclose, Huy Thang Nguyen: nothing to disclose Figure 1 - belongs to Conclusions
Building similarity graph...
Analyzing shared references across papers
Loading...
Binh Nguyen Pham
An Thai Thanh Nguyen
T Nguyen
European Stroke Journal
Ho Chi Minh City Medicine and Pharmacy University
Children's Hospital 1
University Medical Center HCMC
Building similarity graph...
Analyzing shared references across papers
Loading...
Pham et al. (Fri,) studied this question.
www.synapsesocial.com/papers/69fd7e42bfa21ec5bbf0667b — DOI: https://doi.org/10.1093/esj/aakag023.1411