Does intrafascial dissection during nerve-sparing laparoscopic radical prostatectomy compromise cancer control?

To assess whether oncological outcomes are compromised by adopting the curtain dissection (CD) technique (high incision of the peri-prostatic fascia) during nerve-preserving radical prostatectomy (RP).

In all, 973 laparoscopic RPs (LRPs) were performed or supervised by one surgeon between March 2000 and October 2007 for cT1-3 N0M0 prostate cancer, of which 510 included bilateral neurovascular bundle preservation. A CD technique was used in 240 men and a standard dissection (StD) technique was used in 270, considered the control group. The technique was extraperitoneal, used five ports and included preservation of the seminal vesicle tips. Thermal energy was not used posterior or lateral to the prostate in either group. Patient, operative and oncological outcome variables were compared using an independent-sample t-test if continuous or with Fisher's exact test for rates.

Patient and cancer characteristics before LRP were similar for the CD and StD groups, and there were no significant perioperative differences either. Positive margins occurred in 11.7% of the CD group and 11.1% of the StD group (P = 0.95). At a mean (range) follow-up of 11.7 (3-24) months for the CD group and 13.1 (3-24) months for the StD group, biochemical recurrence rates were 0% and 1.1%, respectively (P = 0.30). Potency (CD, 62%; StD, 61%; P = 0.89) and continence rates (StD, 97%; CD, 98%; P = 0.83) were comparable between the groups, but there was a statistically significant earlier return to continence in the CD group (P < 0.001 at 3 months).

For carefully selected men there appears to be no compromise in cancer control with intrafascial dissection in the short term. However, equally there appears to be no significant improvement in potency after LRP. The earlier return to continence after intrafascial nerve-sparing suggests reduced dissection of periurethral supports rather than preservation of additional autonomic nerve fibres.