Impact of Surgical Factors on Robotic Partial Nephrectomy Outcomes: Comprehensive Systematic Review and Meta-Analysis

Purpose: Utilization of robotic partial nephrectomy has increased significantly. We report a literature wide systematic review and cumulative meta-analysis to critically evaluate the impact of surgical factors on the operative, perioperative, functional, oncologic and survival outcomes in patients undergoing robotic partial nephrectomy. Materials and Methods: All English language publications on robotic partial nephrectomy comparing various surgical approaches were evaluated. We followed the PRISMA (Preferred Reporting Items for Systematic Review and Meta-Analyses) statement and AHRQ (Agency for Healthcare Research and Quality) guidelines to evaluate PubMed (R), Scopus (R) and Web of Science (TM) databases (January 1, 2000 to October 31, 2016, updated June 2017). Weighted mean difference and odds ratio were used to compare continuous and dichotomous variables, respectively. Sensitivity analyses were performed as needed. To condense the sheer volume of analyses, for the first time data are presented using novel summary forest plots. The study was registered at PROSPERO (https://www.crd.york.ac.uk/prospero/, ID CRD42017062712). Results: Our meta-analysis included 20,282 patients. When open partial nephrectomy was compared to robotic partial nephrectomy, the latter was superior for blood loss (weighted mean difference 85.01, p < 0.00001), transfusions (OR 1.81, p < 0.001), complications (OR 1.87, p < 0.00001), hospital stay (weighted mean difference 2.26, p = 0.001), readmissions (OR 2.58, p = 0.005), percentage reduction of latest estimated glomerular filtration rate (weighted mean difference 0.37, p = 0.04), overall mortality (OR 4.45, p < 0.0001) and recurrence rate (OR 5.14, p < 0.00001). Sensitivity analyses adjusting for baseline disparities revealed similar findings. When robotic partial nephrectomy was compared to laparoscopic partial nephrectomy, the former was superior for ischemia time (weighted mean difference 4.21, p < 0.0001), conversion rate (OR 2.61, p = 0.002), intraoperative (OR 2.05, p > 0.0001) and postoperative complications (OR 1.27, p = 0.0003), positive margins (OR 2.01, p < 0.0001), percentage decrease of latest estimated glomerular filtration rate (weighted mean difference -1.97, p = 0.02) and overall mortality (OR 2.98, p = 0.04). Hilar control techniques, selective and unclamped, are effective alternatives to clamped robotic partial nephrectomy. An important limitation is the overall suboptimal level of evidence of publications in the field of robotic partial nephrectomy. No level I prospective randomized data are available. Oxford level of evidence was level II, III and IV in 5%, 74% and 21% of publications, respectively. No study has indexed functional outcomes against volume of parenchyma preserved. Conclusions: Based on the contemporary literature, our comprehensive meta-analysis indicates that robotic partial nephrectomy delivers mostly superior, and at a minimum equivalent, outcomes compared to open and laparoscopic partial nephrectomy. Robotics has now matured into an excellent approach for performing partial nephrectomy for renal masses.