Unraveling Rescue Thrombectomy for Mild Large Vessel Occlusion Stroke Following Medical Management: Insights From a Multicenter Retrospective Study.

Abstract

Background: Although rescue thrombectomy is performed in mild (National Institutes of Health Stroke Scale ≤ 5) large vessel occlusion (LVO) stroke patients who experience early neurological deterioration (END) following best medical management (BMM), clinical outcomes remain highly variable. This study aimed to identify key determinants influencing outcomes in this population.

Methods: We retrospectively analyzed consecutive mild LVO patients who initially received BMM and later underwent rescue thrombectomy for END, across four centers between January 2019 and June 2024. END was defined as an NIHSS increase of ≥ 4 points or a total score of ≥ 6 within the first 24 h, without hemorrhage. Multivariable logistic regression was performed to identify factors associated with outcomes. Receiver operating characteristic curve analysis was performed to assess the predictive performance using the area under the curve (AUC).

Results: Among 347 patients with mild LVO who underwent BMM, 66 patients who developed END and underwent rescue thrombectomy were included in this study. Of these, 31 (47.0%) achieved poor outcome (90-day modified Rankin Scale score of 3-6). Multivariable analysis identified prolonged deterioration-to-groin puncture time (OR: 1.79 per 10-minute increase, 95% CI: 1.48-2.54) and basilar artery occlusion (OR: 1.42, 95% CI: 1.16-2.08) were independently associated with poor outcomes. The AUC for predicting poor outcomes was 0.828 for deterioration-to-groin puncture time, 0.690 for basilar artery occlusion, and 0.906 for their combination.

Conclusion: Delayed initiation of thrombectomy and basilar artery occlusion were predictors for poor outcomes in patients who underwent rescue thrombectomy after BMM.