Aspiration, stent retriever, or combined approach for basilar artery occlusion : a three-way comparative analysis

Abstract

Background: Basilar artery occlusion (BAO) is a rare but devastating form of ischaemic stroke, with high rates of disability and mortality. While randomized trials have demonstrated the benefit of mechanical thrombectomy (MT) in BAO, the optimal first-line technique – aspiration, stent retriever, or a combined approach – remains undefined. Objectives: This multicentre study aimed to provide a three-way comparison of MT techniques in terms of efficacy, safety and subgroup-specific outcomes. Design: A retrospective observational study. Methods: We prospectively included 517 consecutive patients with acute isolated BAO treated with MT across seven comprehensive stroke centres between January 2019 and December 2023. Patients were grouped by first-line technique: aspiration (n = 200), stent retriever (n = 260), or combined approach (n = 57). The primary outcome was favourable functional outcome at 90 days (mRS 0–3). Inverse probability weighting (IPW) adjusted for baseline imbalances. Secondary outcomes included successful recanalization, excellent outcome (mRS 0–1), functional independence (mRS 0–2), mortality, symptomatic intracranial haemorrhage (sICH) and haemorrhagic transformation (HT). Predefined subgroup analyses were performed. Results: After adjustment, 90-day outcomes were similar across groups. Stent retrievers achieved higher recanalization rates (RR 1.86 vs aspiration, p < 0.001), while the combined technique was associated with less HT (RR 0.39 vs aspiration, p = 0.008). In patients ⩾80 years, stent-retriever use led to better outcomes than aspiration (39.2% vs 18%; p = 0.021). No other significant subgroup interactions were found. Conclusion: While overall functional outcomes were comparable, stent retrievers yielded superior recanalization and the combined technique reduced haemorrhagic complications. Technique selection may benefit from individualized, anatomy-driven decision-making. Randomized studies are warranted.