Bayesian network analysis of long-term oncologic outcomes of open, laparoscopic, and robot-assisted radical cystectomy for bladder cancer.

Background: To understand the long-term oncologic outcomes of open radical cystectomy (ORC) versus laparoscopic radical cystectomy (LRC) versus robot-assisted radical cystectomy (RARC) for bladder cancer (BCa). Therefore, we performed the conventional meta-analysis and network meta-analysis to evaluate the long-term oncologic outcomes of ORC, LRC, and RARC for BCa.

Methods: A systematic search of PubMed, Embase, Cochrane Library, Medline, and Web of science was performed up until July 1, 2021. Long-term oncologic outcomes include the 5-year overall survival (OS) rate, the 5-year recurrence-free survival (RFS) rate, and the 5-year cancer specific-survival (CSS) rate. The Bayesian network analysis has been registered in PROSPERO (CRD42020208396).

Results: We found that 10 articles (including 3228 patients) were included in our Bayesian network analysis. No significant differences were found between ORC, LRC, and RARC in long-term oncologic outcomes in either direct meta-analysis or network meta-analysis. Therefore, the clinical effects of 5-year OS, RFS, and CSS of RARC, LRC, and ORC are similar. But LRC may be ranked first in 5-year OS, RFS, and CSS compared to other surgical approaches by probabilistic analysis ranking via Bayesian network analysis.

Conclusions: We found that there were no statistical differences in the 3 surgical approaches of RAPC, LPC, and OPC for Bca in long-term oncologic outcomes by direct meta-analysis. However, Subtle differences between these surgical approaches can be concluded that LRC may be a better surgical approach than RARC or ORC in long-term oncologic outcomes by probabilistic analysis ranking via Bayesian network analysis. Moreover, we need a large sample size and more high-quality studies to improve and verify further.