Replicate systematic review and meta-analyses on robotic surgery: a quality appraisal and overlap investigation

Abstract

Background: The number of publications of systematic reviews and meta-analyses (MAs) on robotic surgery have been increasing, including many investigating the same topic. Their quality and extent of overlap remains unclear. We assessed the quality of the MAs in this area and investigated the extent of their overlap. Methods: Relevant studies were identified by searching the MEDLINE, EMBASE, and Cochrane Library databases up to August 1, 2017. Reporting and methodological quality levels were assessed using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and Assessment of Multiple Systematic Reviews (AMSTAR) checklists. A thorough investigation of the extent of overlap was performed. Results: In total, 90 MAs in 5 surgical subspecialties were included after full-text review. The mean reporting and methodological quality scores were 22.5 (83.2%) and 7.6 (69.2%), respectively. Authors from university-affiliated institutions and the presence of statistician or epidemiologist coauthors were associated with better-reporting quality scores. The topics with the most overlapping MAs (all ≥ 6) were robot-assisted thyroidectomy, prostatectomy, gastrectomy, colectomy, and fundoplication. 36 (40%) of the included MAs cited previous MAs on the same topic. Among the 7 MAs comparing robot-assisted radical prostatectomy to the open procedure, most (6/7) drew the same conclusion. 50 to 86% of MAs on this topic included the same trials as primary studies. Conclusion: Conducting multiple overlapping MAs with identical conclusions on the same topic that are of suboptimal quality may be a waste of resource and effort. Authors from university-affiliated institutes and experts in epidemiology and statistics are more likely to conduct MAs that have better quality. More guidelines and registries are needed to avoid overlapping MAs.