Arterial Tortuosity Is a Potent Determinant of Safety in Endovascular Therapy for Acute Ischemic Stroke.

Background: Subarachnoid hemorrhage (SAH) associated with vessel injury during endovascular therapy for acute ischemic stroke is a known complication. Arterial anatomy may predispose to increased risk of SAH and technical safety, yet factors such as clot location, arterial size, and tortuosity have not been explored. We examined these anatomic factors with respect to SAH during thrombectomy. Methods: Arterial anatomy at the site of occlusion and mechanical thrombectomy during device deployment was detailed by the STRATIS (Systematic Evaluation of Patients Treated With Neurothrombectomy Devices for Acute Ischemic Stroke) core laboratory. Luminal diameters, arterial branching, and segmental tortuosity were measured. Arterial tortuosity was quantified using the distance factor metric. Statistical analyses included descriptive variables of arterial anatomy, with univariable and multivariable modeling to predict SAH. Results: Arterial tortuosity in each segment from the proximal cerebral arteries to the site of occlusion was quantified in 790 subjects treated with mechanical thrombectomy in STRATIS. Cumulative arterial tortuosity to the site of vessel occlusion was greater in distal lesions. SAH was clearly linked with more distal thrombectomy (P = 0.02), occurring in 19.0% of distal M2, 16.7% of M3, 7.3% of distal M1, 5.8% of proximal M2, 2.4% of distal internal carotid artery, and 2.1% of proximal M1. In multivariable analysis after adjusting for arterial diameter at the site of occlusion, arterial tortuosity was a significant predictor of SAH (upper tertile versus 1: odds ratio, 3.08 [95% CI, 1.04-9.09]; P = 0.04), while arterial diameter was unrelated to SAH (P = 0.30) when accounting for tortuosity. Conclusion: This novel analysis of arterial tortuosity and angiographic anatomy during mechanical thrombectomy establishes tortuosity as a determinant of SAH, providing insight for future techniques and innovative device designs.