Safety of selective nerve sparing in high risk prostate cancer during robot-assisted radical prostatectomy.

D'Amico high risk prostate cancer is associated with higher incidence of extra prostatic disease. It is recommended to avoid nerve sparing in high risk patients to avoid residual cancer. We report our intermediate term oncologic and functional outcomes in patients with preoperative D'Amico high risk prostate cancer, who underwent selective nerve sparing robot-assisted radical prostatectomy (RARP). Between Jan 2008 till June 2013, 557 patients underwent RARP for D'Amico high risk prostate cancer. The criteria for nerve sparing were as follows-complete: non palpable disease with <3 cores involvement on prostate biopsy; partial: non palpable disease with <4 cores involvement on prostate biopsy; none: clinically palpable disease with ≥4 cores involvement on prostate biopsy and intraoperative visual cues of locally advanced disease (loss of dissection planes, focal bulge of prostatic capsule). Degree of nerve sparing (NS) was graded intraoperatively by the surgeon independently at either side as side specific margins were assessed to predict subjectivity of the intraoperative judgment. Various data were collected and analyzed. Of 557 patients who underwent RARP 140 underwent complete (group 1), 358 patients underwent partial (group 2), and 59 patients underwent non-nerve-sparing procedure (group 3). There were no difference in preoperative characteristic between the groups (p = 0.678), but group 3 had higher Gleason score sum (p = 0.001), positive cores on biopsy (p = 0.001) and higher t stage (p = 0.001). Postoperatively Extra prostatic extension (p = 0.001), seminal vesicle invasion (p = 0.001), and tumor volume (p < 0.001) were higher in Group 3. Side specific positive surgical margins (PSMs) rates were higher for non-nerve-sparing compared to partial and complete nerve sparing RARP (p < 0.001; overall PSMs = 25.2 %). On univariate and multivariate analysis, nerve sparing did not affect PSMs (p > 0.05). The overall biochemical recurrence (BCR) rate at mean follow-up of 24.3 months was 19.21 %. The continence rate at 3 month was significantly higher in complete NS group in comparison to non-NS group (p = 0.020), however, this difference was not statistically significant at 1 year. Similarly, mean time to continence was significantly lower in complete NS group in comparison to non-NS group (p = 0.030). The potency rate was significantly higher and mean time to potency was significantly lower in complete NS group in comparison to non-NS group (p = 0.010 and 0.020, respectively). In high risk prostate cancer patients, selective nerve sparing during RARP, using the preoperative clinical variables (clinical stage and positive cores on biopsy) and surgeon's intraoperative perception, could provide reasonable intermediate term oncologic, functional outcomes (continence and potency) with acceptable perioperative morbidity and positive surgical margins rate. Use of these preoperative factors and surgeon's intraoperative judgment can appropriately evaluate high risk prostate cancer patients for nerve sparing RARP.