Head-to-head comparison of nephrometry scores for partial nephrectomy: implications for clinical guidelines

Introduction: Several nephrometry scores aim to predict partial nephrectomy (PN) outcomes. Since consensus regarding the best score is lacking, we performed a prospective head-to-head comparison of the most widely used. Materials and methods: A dedicated uroradiologist prospectively reviewed preoperative CT scan to assign points to variables of interest included in RENAL, PADUA, SPARE, C-Index, DAP, and MAP score, in 202 renal cell carcinoma (RCC) surgical candidates. The primary outcome was surgical success, defined as PN completion, absence of grade > II Clavien-Dindo complications, ischemia time ≤20 min, and negative surgical margins. The secondary outcome was PN completion relative to radical nephrectomy (RN). Multivariable logistic regression (MLR) models predicted study outcomes after adjusting for age, gender, ECOG performance status, preoperative renal function, and surgical approach. Receiver operating characteristic (ROC) and area under the curve (AUC) compared each nephrometry accuracy. Results: Surgical success and PN completion rates were 31 % and 60 %, respectively. At MLR, RENAL, PADUA, SPARE, DAP, and MAP scores independently predicted both outcomes (p < 0.001). The highest predictive accuracy for surgical success and PN completion was recorded for SPARE (AUC: 0.79 and 0.89). Intermediate accuracy was recorded for RENAL (AUC: 0.78 and 0.85), PADUA (AUC: 0.78 and 0.86), DAP (AUC: 0.76 and 0.85) and C-Index (AUC: 0.73 and 0.77). The lowest accuracy was recorded for MAP (AUC: 0.65 and 0.68). Conclusion: Nephrometries are associated with surgical success and PN completion. In clinical practice, the use of SPARE should be privileged based on the highest predictive accuracy for both endpoints.