Up to date, clinical trials have yielded conflicting results regarding the efficacy and safety of mechanical thrombectomy (MT) for acute ischemic strokes (AIS) of the posterior circulation (PC). It is urgent to identify factor associated with better outcomes and to improve patients’ selection.

We collected data from patients with AIS of the PC who underwent MT in a comprehensive stroke center, from 2015 to 2022. We registered demographic, clinical, and imaging information. Main outcome of interest was functional recovery at ninety-days, measured with the modified Rankin Scale (mRS); we defined good functional recovery as mRS 0-2, and fair as mRS 0-3. We used Bayesian Model Averaging (BMA) to identify variables correlated with good and fair functional outcome, with a posterior probability cut-off of 60%. Analyses were conducted using R software.

Eighty-nine patients met our inclusion criteria. Median age was 65 (IQR 55-76) and 37.9% (n= 36) were female. Mean basal mRS was 0.4 (SD 0.9), median pc-ASPECTS was 9 (IQR 8-10), 38.2% (n=34) received IV-tPA, and median NIHSS at admission was 17 (IQR 10-28). Median SO-to-puncture time was 534.5 min (IQR 339-944). In 77.5% (n=60) reperfusion (mTICI 2b-3) was achieved, 29.8% (n=25/84) had good mRS at 90-days, and 47.6% (n=40/84) fair mRS. In BMA analysis for fair mRS, NIHSS at admission (prob-not-0=66.6, OR-post-mean=0.96) and hemorrhagic conversion (prob-not-0=61.1, OR-post-mean=0.4) were selected as predictors in the 5 best models. For good mRS, only coronary artery disease (prob-not-0=62.9, OR-post-mean=0.25) was selected.

Lower NIHSS at admission was positive correlated with better functional recovery at 90-days, while hemorrhagic conversion and CAD showed a negative correlation. Our results are limited for the small sample size and retrospective design, however, using BMA allows for strongly reliable results.