Document resource
Background Observational studies have shown that endovascular thrombectomy (EVT) benefits patients with acute basilar artery occlusion (aBAO) in the ultra-late window, yet subgroups at high risk of unfavourable outcome remain unclear. We sought to identify predictors and develop and validate a nomogram for identifying these patients.Methods This observational, retrospective study enrolled patients with aBAO who underwent EVT beyond 24 hours between January 2019 and March 2025. Eight hospitals were assigned to the training cohort and seven hospitals to the validation cohort, with a 7:3 random allocation based on hospital geographical location. A nomogram was constructed to estimate the probability of 90-day unfavourable outcomes (modified Rankin Scale score, 4–6) and generate actionable risk categories. We assessed the performance of the nomogram and conducted external validation.Results Overall, 203 patients were included in the analysis. Independent predictors, including late neurological deterioration (OR 3.12, 95% CI 1.22 to 8.02), National Institutes of Health Stroke Scale score (OR 1.10, 95% CI 1.04 to 1.17), Glasgow Coma Scale score ≤8 (OR 3.98, 95% CI 1.06 to 14.96), male (OR 5.75, 95% CI 1.59 to 20.74) and coronary heart disease (OR 6.53, 95% CI 1.40 to 30.55), were identified in the training cohort and entered into the nomogram. The area under the receiver operating characteristic curve of the nomogram was 0.798 (95% CI 0.727 to 0.870) and 0.721 (95% CI 0.587 to 0.856) in the training and external validation cohorts, respectively.Conclusion We developed a nomogram that can identify patients with aBAO at high risk of unfavourable outcome following EVT in the ultra-late window.