Stroke Etiology and Outcomes after Endovascular Thrombectomy: Results from the SITS Registry and a Meta-Analysis

Background and purpose: The influence of stroke etiology on outcomes after endovascular thrombectomy (EVT) is not well understood. We aimed to investigate whether stroke etiology subgrouped as large artery atherosclerosis (LAA) and cardiac embolism (CE) influences outcomes in large artery occlusion (LAO) treated by EVT.

Methods: We included EVT treated LAO stroke patients registered in the Safe Implementation of Treatment in Stroke (SITS) thrombectomy register between January 1, 2014 and September 3, 2019. Primary outcome was successful reperfusion (modified Treatment in Cerebral Infarction 2b-3). Secondary outcomes were symptomatic intracranial hemorrhage (SICH), 3-month functional independence (modified Ranking Scale 0-2) and death. Multivariable logistic regression models were used for comparisons. In addition, a meta-analysis of aggregate data from the current literature was conducted (PROSPERO, ID 167447).

Results: Of 7,543 patients, 1,903 (25.2%) had LAA, 3,214 (42.6%) CE, and 2,426 (32.2%) unknown, other, or multiple etiologies. LAA patients were younger (66 vs. 74, P<0.001) and had lower National Institutes of Health Stroke Scale score at baseline (15 vs. 16, P<0.001) than CE patients. Multivariable analyses showed that LAA patients had lower odds of successful reperfusion (odds ratio [OR], 0.70; 95% confidence interval [CI], 0.57 to 0.86) and functional independence (OR, 0.74; 95% CI, 0.63 to 0.85), higher risk of death (OR, 1.44; 95% CI, 1.21 to 1.71), but no difference in SICH (OR, 1.09; 95% CI, 0.71 to 1.66) compared to CE patients. The systematic review found 25 studies matching the criteria. The meta-analysis did not find any difference between etiologies.

Conclusions: From the SITS thrombectomy register, we observed a lower chance of reperfusion and worse outcomes after thrombectomy in patients with LAA compared to CE etiology, despite more favorable baseline characteristics. In contrast, the meta-analysis did not find any difference between etiologies with aggregate data.