An updated approach to incremental nerve sparing for robot-assisted radical prostatectomy

Objectives: To update the algorithm for performing incremental nerve sparing (NS) using our multiparametric magnetic resonance imaging (mpMRI)-based nomogram. Patients and methods: We applied the coefficients of the nomogram to the observations extracted from our population of patients who underwent robot-assisted radical prostatectomy between February 2014 and October 2015 and who received preoperative mpMRI. The information considered were PSA level, highest side-specific biopsy Gleason grade group, highest ipsilateral percentage core involvement with the highest Gleason grade group, and extracapsular extension (ECE) on mpMRI. The nomogram-derived probability [P (%)], after internal validation, was used as the independent variable on a classification tree to identify the most significant thresholds for ECE prediction. Incremental NS was performed as follows: Grade 1 NS: intrafascial dissection between the peri-prostatic veins and the pseudocapsule of the prostate; Grade 2 NS: inter-fascial dissection along the peri-venous plane; Grade 3 NS: inter-fascial dissection through the outer compartment of the lateral prostatic fascia; Grade 4 NS: extrafascial dissection. Results: Data from 561 patients were considered, and 829 prostatic lobes with biopsy-documented tumour were analysed. Overall, 142 lobes presented ECE that was focal in 27 (19%) cases. The classification tree identified four risk categories. In the low- [P (%) ≤10], intermediate- [P (%) 10–21], high [P (%) 21–73] and very-high-risk [P(%) >73] groups, the ECE rates were 3.3%, 16%, 61.6% and 90%, respectively. Amongst those, ECE was focal in 41.7%, 31.7%, 7.9% and 0%, respectively. Conclusion: We suggest that Grade 1 NS (intrafascial) should be performed in the low-risk group. The inter-fascial approach, namely grades 2 and 3 NS, should be performed in the intermediate- and high-risk categories, respectively. Grade 4 NS (extrafascial) should be performed in the very-high-risk group. The current algorithm yields a better accuracy than the previous one; however, prospective validation is warranted.