Outcomes of Robotic versus Laparoscopic Partial Nephrectomy: an Updated Meta-Analysis of 4,919 Patients.

ABSTRACT

PURPOSE: We evaluated the current literature comparing outcomes of robotic vs laparoscopic partial nephrectomy. MATERIALS AND METHODS: We performed a literature search according to Cochrane guidelines up to December 2015 including studies comparing robotic and laparoscopic partial nephrectomy, and we compared baseline patient and tumor characteristics. We performed a meta-analysis to evaluate safety, effectiveness and functional outcomes of robotic vs laparoscopic partial nephrectomy using weighted mean difference and inverse variance pooled risk ratios, respectively. RESULTS: A total of 4,919 patients were included from 25 studies (robotic partial nephrectomy 2,681, laparoscopic partial nephrectomy 2,238). There were no significant differences between the 2 groups in terms of age, gender, laterality and final malignant pathology. Patients treated with robotic partial nephrectomy had larger tumors (WMD 0.17 cm, p=0.001) and higher mean R.E.N.A.L. nephrometry scores (WMD 0.59, p=0.002), and were associated with a decreased likelihood of conversion to laparoscopic/open surgery compared to laparoscopic partial nephrectomy (RR 0.36, p <0.001), any (Clavien 1 or greater) (RR 0.84, p=0.007) and major (Clavien 3 or greater) (RR 0.71, p=0.023) complications, positive margins (RR 0.53, p <0.001) and shorter warm ischemia time by 4.3 minutes (p <0.001). Both approaches had similar operative times (WMD -12.2 minutes, p=0.34), estimated blood loss (WMD -24.6 ml, p=0.15) and postoperative change in estimated glomerular filtration rate. CONCLUSIONS: This updated meta-analysis of retrospective cohort studies demonstrated that robotic partial nephrectomy confers a superior morbidity profile compared to laparoscopic partial nephrectomy in most of the examined perioperative outcomes. Despite being the strongest available evidence (Level 2b) for outcomes of robotic vs laparoscopic partial nephrectomy thus far, there have been no completed or ongoing randomized trials to lend Level 1 support for either approach.