Are nephrometry scores accurate for the prediction of outcomes in patients with renal angiomyolipoma treated with robot-assisted partial nephrectomy? A multi-institutional analysis.

Background: Prediction of complications and surgical outcomes is of outmost importance even in patients with benign renal masses. The aim of our study is to test the PADUA, SPARE and R.E.N.A.L. scores to predict nephron sparing surgery (NSS) outcomes in patients presenting with renal angiomyolipoma (RAML).

Methods: We retrospectively analyzed the clinical and pathological data of 93 patients with AML treated with robot-assisted partial nephrectomy (RAPN) at three tertiary care referral centers. Renal masses were classified according to the PADUA, SPARE and R.E.N.A.L. nephrometry scores. Surgical success was defined according to the novel Trifecta Score. Logistic regression models (LRM) were fitted to predict the achievement of novel Trifecta and the risk of high-grade Clavien-Dindo (CD) complication. The receiver operating characteristics (ROC) curve analysis was used to estimate the accuracy of LRMs.

Results: Of 93 patients, 66 (69.9%) were females; median tumor size was 42 (36-48) mm. Novel Trifecta was achieved in 79 patients (84.9%) and postoperative complications classified as CD>2 occurred in 7 (7.5%) patients. At univariate and multivariate LRMs all three nephrometry scores were significantly associated with novel Trifecta achievement. Similar findings were observed for the prediction of CD>2 complications. The AUCs to predict optimal surgical outcomes and CD>2 complications were 0.791 and 0.912 for PADUA, 0.767 and 0.836 for SPARE and 0.756 and 0.842 for RENAL Score, respectively.

Conclusions: RAPN appears to be a feasible and safe surgical technique for the treatment of RAML. PADUA, SPARE and RENAL scores can be safely adopted to predict surgical outcomes, with the first one showing a higher accuracy.