A multi-institutional comparison of radical retropubic prostatectomy, radical perineal prostatectomy, and robot-assisted laparoscopic prostatectomy for treatment of localized prostate cancer.

To evaluate the pathological stage and margin status of patients undergoing radical retropubic prostatectomy (RRP), radical perineal prostatectomy (RPP) and robot-assisted laparoscopic prostatectomy (RALP). We performed a retrospective analysis of 196 patients who underwent RRP, RPP, and RALP as part of our multi-institution program. Fifty-seven patients underwent RRP, 41 RPP, and 98 RALP. Patient age, preoperative prostate specific antigen (PSA), preoperative Gleason score, preoperative clinical stage, pathological stage, postoperative Gleason score, and margin status were reviewed. The three groups had similar preoperative characteristics, except for PSA (8.4, 6.5, and 6.2 ng/ml) for the retropubic, robotic, and perineal approaches. Margins were positive in 12, 24, and 36% of the specimens from RALP, RRP, and RPP, respectively (P = 0.004). The positive margin rates in patients with pT2 tumors were 4, 14, and 19% in the RALP, RRP, and the RPP groups, respectively (P = 0.03). Controlling for age and pre-operative PSA and Gleason score, the rate of positive margins was statistically lower in the RALP versus both the RRP (P = 0.046) and the RPP groups (P = 0.02). In the patients with pT3 tumors, positive margins were observed in 36% of patients undergoing the RALP and 53 and 90% of those patients undergoing the RRP and RPP, respectively (P = 0.015). Controlling for the same factors, the rate of positive margins was statistically lower in the RALP versus the RPP (P = 0.01) but not compared with the RRP patients (P = 0.32). The percentage of positive margins was lower in RALP than in RPP for both pT2 and pT3 tumors. RRP had a higher percentage of positive margins than RALP in the pT2 tumors but not in the pT3 tumors.