Urology Video Journal (Jun 2023)
Vascular perfusion 3D models guidance during robot-assisted partial nephrectomy: The end of empirical selective clamping era
Abstract
Objective: In the last years selective clamping strategies have been tested during partial nephrectomy (PN) to decrease the functional impairment linked to the ischemic damage. Recently, 3D virtual models (3DVM) have been reported to increase the accuracy of surgical strategy, leading to a higher rate of successful selective clamping. However, sometimes this empirical definition of vascular regions, based on arterial branches’ direction, fails intraoperatively and forcing towards a global clamping.Our aim is to present our new generation of 3DVM, built taking in consideration the different perfusion areas of the kidney instead of vascular regions empirically estimated, and to evaluate their accuracy intraoperatively and to assess if there are factors influencing the tumor location relative to these perfusion areas. Patients and surgical procedure: By using a mathematical tool, the Voronoi diagram, we calculate vascular perfusion regions of the organs and built the 3DVMs for all consecutive patients scheduled in our center for robot-assisted PN (RAPN) from 2019 to 2021. Based on the info given by the 3DVMs, a selective clamping strategy was planned end performed intraoperatively if feasible. To assess selective clamping success, near-infrared-fluorescence imaging (NIRF) was used. The concordance between virtual and real perfusion regions was automatically assessed by superimposition of 3DVMs over the real endoscopic view. The comparison between their extension was then assessed via k-Cohen test.At last, each lesion location was evaluated in relation to the number of its crossing perfused areas and a multivariate regression model (MLR) was fitted to assess potentially related preoperative factors. Results: 48 patients were enrolled. The renal pedicle management strategy was conducted as preoperatively planned in 100% of them. 68.7% of cases underwent selective clamping. The concordance between the NIRF enhanced areas and their corresponding 3DVM areas was confirmed (k = 0.94). Distribution of perfused areas crossing the tumor was as follows: 16.6%, 29.1%, 43.7%, 8.3%, 2.4% for 1, 2, 3, 4 and 5 crossing areas respectively. At MLR the clear cell histotype was the only factor related to a lower number (3) of perfusion areas. Conclusions: The implementation of mathematical algorithms to 3DVMs allows a precise estimation of the perfusion area of each arterial branch feeding the organ, leading to perform an effective pedicle management. Moreover, with this new technology it is possible to open new perspectives on the knowledge of renal tumors features, considering their growth in relation to their surrounding environment.
