Learning Curve for Endovascular Treatment of Anterior Circulation Large Vessel Occlusion at a Single Center

Background and purpose: Data concerning the learning curve for endovascular treatment (EVT) of anterior circulation large vessel occlusion are scarce. This study aimed to investigate the relationship between operator experience and the outcome of EVT and to further identify the number of cases needed to acquire the ability to perform successful reperfusion. Materials and methods: Four hundred and thirty-four patients who underwent EVT by seven operators at a single center from January 2016 to September 2019 were enrolled. Procedural experience was defined by the number of cases performed by each operator. Multivariable backward regression analyses were used to investigate the association between procedural experience and functional independence (defined as a modified Rankin Scale score of 0–2), 90-days mortality, successful reperfusion (defined as a modified Thrombolysis in Cerebral Infarction score of 2b-3), and puncture-to-reperfusion time after adjusting for covariates. A risk-adjusted cumulative sum (RA-CUSUM) chart was utilized to identify the number of caseloads needed to overcome the learning curve effect. Results: Procedural experience was independently associated with functional independence, 90-days mortality, successful reperfusion, and puncture-to-reperfusion time reduction (per 10-case increment: OR 1.219, 95% CI: 1.079–1.383, P < 0.001; OR 0.847, 95% CI: 0.738–0.968, P = 0.016; OR 1.553, 95% CI: 1.332–1.830, P < 0.001 and β 8.087 min, 95% CI: 6.184–9.991, P < 0.001, respectively). The RA-CUSUM chart indicated that at least 29 cases were required to overcome the learning curve effect. Conclusions: There was a dose-response relationship between operator case volume and clinical outcome, procedure time, and successful reperfusion. The experience needed for successful EVT was at least 29 cases.