Endovascular thrombectomy with or without intravenous thrombolysis in large-vessel ischemic stroke: A non-inferiority meta-analysis of 6 randomised controlled trials
Lisa Christina Horvath, Felix Bergmann, Arthur Hosmann, Stefan Greisenegger, Kerstin Kammerer, Bernd Jilma, Jolanta M. Siller‐Matula, Markus Zeitlinger, Georg Gelbenegger, Anselm Jorda
Abstract
BACKGROUND: It is unclear whether thrombectomy alone is non-inferior to thrombectomy with intravenous thrombolysis in patients with acute ischemic stroke due to large-vessel occlusion. PURPOSE: To perform a comprehensive, trial-level data, non-inferiority meta-analysis of randomised controlled trials comparing endovascular thrombectomy with and without intravenous thrombolysis in patients with ischemic stroke due to large-vessel occlusion of anterior circulation. METHODS: The prespecified primary efficacy outcome was functional independence, defined as a modified Rankin scale (mRS)score of 0 to 2 at 90 days. The two prespecified non-inferiority margins were risk differences of -10% and - 5%. The study was registered in PROSPERO (CRD42022361110) and conducted according to PRISMA guidelines. RESULTS: Six trials were included in this analysis (DIRECT-MT, DEVT, SKIP, MR CLEAN-NO IV, DIRECT-SAFE and SWIFT DIRECT) comprising a total of 2334 patients. Functional independence at 90 days was achieved by 570 (49·0%) of 1164 patients in the thrombectomy alone group and 595 (50·9%) of 1170 patients in the thrombectomy with thrombolysis group (pooled risk difference - 0·02, [95% CI -0·06-0·02]). Combined thrombectomy and thrombolysis were associated with significantly higher rates of successful reperfusion (pooled risk ratio 0·96 [95% CI, 0·93-0·99], p = 0·006) but at the expense of a significantly increased risk of overall - but not symptomatic - intracranial haemorrhage (pooled risk ratio 0·87 [95% CI, 0·77-0·98], p = 0·02). CONCLUSIONS: Compared with a combined treatment approach, thrombectomy alone was non-inferior at -10% non-inferiority margin, but not at a - 5% inferiority margin for functional independence. Current evidence cannot exclude clinically important differences between the two treatment approaches.