219. ASSOCIATION BETWEEN VIDEO-BASED SURGICAL PERFORMANCE AND PATIENT OUTCOMES IN MINIMALLY INVASIVE ESOPHAGECTOMY IN THE NETHERLANDS

Background Increasing evidence shows substantial variation in surgical performance measured by a (video-based) competency assessment tool (CAT). Moreover, suboptimal surgical performance has been associated with less favorable patient outcomes in complex minimal invasive procedures. It is likely this also applies for minimally invasive esophagectomy (MIE). In a previous study a CAT for MIE (MIE-CAT) was developed and validated. The present study investigated the association between surgical performance and postoperative outcomes of MIE in the Netherlands. Methods A nationwide observational video analysis study was performed. All fifteen Dutch hospitals performing MIE voluntarily submitted all patient outcomes from the 2020–2021 Dutch Upper-GI Clinical Audit registry, and two representative surgical videos from 2022. Surgical performance was assessed by 7 blinded and independent expert MIE surgeons with the MIE-CAT. Hospitals were divided into quartiles based on their performance score. Multilevel logistic regression, with clustering of patients within hospitals, was used to study associations between surgical performance and patient outcomes. Primary outcome was severe postoperative complications (Clavien-Dindo ≥3) within 30 days after surgery. Results Highest- (mean MIE-CAT 113.6, SD 5.5, n = 3) versus lowest-performance-quartile (mean 94.1, SD 5.9, n = 3) had more MIE experience (Pearson’s r = 0.74, 95%CI 0.32–0.92), larger volumes (Pearson’s r = 0.46, 95%CI -0.13-0.81) and favorable outcomes (Figure 1). Performance of highest- versus lowest-quartile was statically significantly associated with less complications (severe postoperative (RR, converted from OR, = 0.50, 95%CI 0.24–0.99), peroperative (RR = 0.21, 95%CI 0.04–0.94) and any postoperative (RR = 0.54, 95%CI 0.24–0.96)), less conversions (RR = 0.21, 95%CI 0.21–0.21) and a 11.5% absolute chance reduction of severe postoperative complications for an average patient. Increased anastomotic-phase performance was associated with less anastomotic leakage (RR = 0.14, 95%CI 0.06–0.31). Conclusion This Dutch nationwide study showed statistically significant and clinically relevant associations between surgical performance and outcomes of patients undergoing MIE. An average patient has a 11.5% absolute chance reduction of a severe postoperative complication when operated in a highest-performance quartile hospital, compared to a lowest-performance quartile hospital. These findings show that good surgical performance is an essential aspect for good clinical outcomes.