Meta-Analysis of the Impact of the Learning Curve in Robotic Rectal Cancer Surgery on Histopathologic Outcomes.

Abstract

INTRODUCTION: Although the process of learning robotic surgery for rectal cancer is associated with a prolonged operating time and higher complication rates, its impact on histopathologic outcomes is unknown. The aim of this meta-analysis was to evaluate the impact of the learning curve in robotic surgery for rectal cancer on histopathologic outcomes. METHODS: The PubMed, EMBASE, Cochrane Library, MEDLINE via Ovid, CINAHL, and Web of Science databases were systematically searched. The inclusion criterion was any clinical study comparing the outcomes of robotic surgery for rectal cancer between different phases of the learning curve (LC) including competence (C). The primary endpoint was the circumferential resection margin (CRM) involvement rate defined as CRM ≤1 mm. The Mantel-Haenszel method with odds ratios with 95% confidence intervals (OR (95%CI)) was used for dichotomous variables. RESULTS: Ten studies including a total of 907 patients (521 LC and 386 C) were selected. Nine studies were found to have a low risk of bias, and one had a moderate risk of bias. The CRM involvement rate was 2.9% (13/441) for learning curve vs. 4.6% (13/284) for competence. This difference was not significant (OR (95%CI) = 0.70 (0.30, 1.60); p=0.39; I2=0%). CONCLUSION: A surgeon's learning curve seems to have no impact on CRM involvement rates compared to surgeon competence in robotic surgery for rectal cancer.