Clinical safety and feasibility of a newly developed, simple algorithm for decision-making on neurovascular bundle preservation in radical prostatectomy

Abstract

Objective: We investigated the clinical safety and feasibility of an algorithm we developed for the decision-making on neurovascular bundle preservation in radical prostatectomy to decrease the incidence of positive surgical margins. Methods: We prospectively applied our algorithm to 82 patients (164 prostate sides) with clinically localized prostate cancer who underwent radical prostatectomy at our institution between October 2004 and September 2006. The algorithm was developed using the apical core characteristics, clinical T stage, preoperative prostate-specific antigen level and Gleason sum. All prostate sides were divided into two groups by the algorithm: 115 sides (70.1%) were qualified for neurovascular bundle preservation (favorable algorithm side group) and 49 sides (29.9%) for non-neurovascular bundle preservation (unfavorable algorithm side group). Results: Median patient age was 66 years (range: 52-77) and median prostate-specific antigen was 7.1 ng/ml (range: 1.4-29.6). Overall, a positive surgical margin was observed in 23 sides (14.0%). The incidence of positive surgical margins at the apex was significantly correlated with the maximal diameter of the tumor in the apex (P < 0.001). The incidence of positive surgical margins was 8.7% in the favorable algorithm group, whereas it was 26.5% in the unfavorable algorithm group (P = 0.003). When this algorithm was combined with surgeons' intraoperative assessments, the incidence of positive surgical margins was 2.1% in neurovascular bundle preservation sides, compared with 25.0% in non-neurovascular bundle preservation sides (P = 0.002). Conclusions: This simple algorithm is safe and feasible for the decision-making on neurovascular bundle preservation from the aspect of cancer control in radical prostatectomy patients. © The Author (2009). Published by Oxford University Press. All rights reserved.