ABSTRACT
Background: Multiparametric prostate MRI (mpMRI) can provide important information for surgical planning, yet its interpretation is not immediate and imaging consultation at the time of surgery can result in interruptions and delay. The use of three-dimensional (3D) models based on mpMRI might obviate these issues. We aimed to evaluate the role of the prospective integration of 3D models from mpMRI in the robotic console in reducing the rate of positive surgical margins (PSMs). Materials and Methods: PSMs at our center are evaluated intraoperatively using the Neurovascular Structure Adjacent Frozen Section Examination method. Based on the rate of PSMs on frozen section during the year before the implementation of 3D models during surgery (22.5%), we estimated that 151 subjects were needed to detect a statistically significant difference of at least 40%. Patients with biopsy-proven prostate cancer (PCa) who received a 3T mpMRI at our institution and had a PIRADS ≥3 on mpMRI were included. Results: One hundred fifty-one patients were included. Overall, 17 (11.3%) patients had a PSM, 6 (35%) of them had PSM in an area where the mpMRI did not demonstrate any lesions. The rates of PSMs on both frozen (22.5% vs 11.3%) and permanent section (13.1% vs 6.6%) were significantly lower (p ≤ 0.03) compared with the cohort of patients operated during 2018 (n = 358). No significant differences among clinical characteristics were found between the study cohort and the 2018 cohort (all p > 0.05). Conclusions: The use of 3D models at the time of surgery was shown to reduce the PSM rate on both frozen and permanent section. Integrating 3D models in the robotic console could lead to improved PCa outcomes.
