Thrombectomy Use in the United States for Basilar Artery Occlusion in the Era of Neutral Clinical Trials: 2018 to 2020 Analysis of the National Inpatient Sample

Abstract

In 2019 and 2020, 2 randomized trials of basilar artery occlusion (BAO) thrombectomy reported no improvement in functional outcomes with thrombectomy, despite observational evidence and weak guidelines recommending thrombectomy. The objective of this study was to assess whether neutral clinical trials may have influenced BAO thrombectomy practice. The National Inpatient Sample (2018–2020) was queried for US patients with BAO, and comparisons were made between patients admitted in 2020 versus 2018 to 2019 for the primary outcome of thrombectomy. Unadjusted and adjusted multivariable regression was used, accounting for demographic and clinical covariates, with propensity‐score matching to balance clinical and hospital‐level characteristics between years. Of the 14 945 patients with BAO, 2345 (15.6%) underwent thrombectomy, with no differences in the unadjusted rate of thrombectomy between 2020 and 2018 to 2019 (14.9% versus 16.1%; P =0.41). Following multivariable adjustment, BAO thrombectomy was independently associated with a private insurance beneficiary (odds ratio [OR] 1.46, 95% CI 1.16–1.85) as compared with Medicare beneficiary; having a National Institutes of Health Stroke Scale 10–19 (OR 1.89, 95% CI 1.39–2.50) or >19 (OR 1.67, 95% CI 1.25–2.26) versus <10; but not with year of admission (OR 0.78, 95% CI 0.60–1.01; P =0.06). These relationships were preserved in the propensity‐score matching cohort, and admission year lacked association with thrombectomy for BAO (OR 0.98, 95% CI 0.73–1.33; P =0.92). Following publication of neutral BAO randomized clinical trials in late 2019 and 2020, there was no significant change in thrombectomy rate among US patients with BAO. The latest trials support thrombectomy for select patients with moderate‐to‐severe BAO symptoms, and this appears consistent with the practice of US clinicians before these trials.