Positive surgical margins during robotic radical prostatectomy: a contemporary analysis of risk factors.

OBJECTIVE: To determine the risk factors (clinical, pathological and technical) for positive surgical margins (PSMs) after robotically assisted radical prostatectomy (RARP), as a PSM is associated with an increased risk of biochemical recurrence and often responsible for significant patient anxiety. PATIENTS AND METHODS: Between November 2003 and March 2007, 216 consecutive patients had an RARP by one fellowship-trained urological oncologist. The surgical pathological specimens were fixed and processed using standard techniques, and assessed by a pathologist at the same institution. A PSM was defined as the presence of cancer adjacent to the inked margin. The clinical charts were reviewed retrospectively under an approved institutional review board protocol. Univariable and multivariable methods, including logistic regression models, were used to analyse the clinical, pathological and technical risk factors for PSM.
RESULTS: The overall prevalence of PSM was 14.8% (32/216), and 5.4% (8/149) for pT2 cancers. The only preoperative factor that was associated with a greater risk of a PSM was the serum prostate-specific antigen (PSA) level (P = 0.012) and PSA density (P = 0.005). Age, clinical stage and clinical Gleason grade were not predictors of a PSM. The overall and pT2 PSM rate remained constant throughout the series of 216 patients (P = 0.371), indicating that the initial experience for RARP was not associated with a significantly greater risk of a PSM. However, there was a small independent 'learning curve' effect, with a lower rate of PSM associated with each increment of 25 patients (odds ratio 0.8, 95% confidence interval 0.6-1.0), supported by the significantly decreasing trend in PSM for pT3 cancers over time (P = 0.031) Although there was no significant increase over time in PSM with the use of an endostapler to control the dorsal venous complex (DVC), there was a significant learning effect, with a decrease in the PSM rate specifically in pT3 cancers using the suture technique (P = 0.005). A nerve-sparing procedure increased the risk of PSM in multivariable analysis (P = 0.03). As expected, pathological stage and pathological Gleason grade were the strongest predictors of PSM (P < 0.001).
CONCLUSION: The most important risk factors for a PSM after RARP are the preoperative PSA level, PSA density, pathological stage and Gleason grade. PSM rates for a surgeon in their initial experience can be comparable to that of a surgeon experienced in RARP. Using a stapling device to control the DVC does not appear to increase the risk of a PSM, although nerve-sparing increases the rates of PSM in extraprostatic prostate cancer.