Introduction: To date, Martini nomogram represents the best attempt to stratify the risk of renal function decline in patients treated with robot-assisted partial nephrectomy (RAPN) for kidney tumor. However, this tool relied on short follow-up and adopted a questionable endpoint such as 25% eGFR reduction. To address this void, we developed a novel nomogram to accurately predict the risk of 5-years CKD upstaging in patients treated with RAPN. Moreover, we aimed at stratifying patients in different risk categories based nomogram-derived probability. Methods: Whitin a multi-institutional database, we identified patients treated with RAPN for localized renal tumor (cT1-2, cN0, cM0). Exclusion criteria was end-stage renal disease. Kaplan-Meier and Cox Regression addressed significant CKD upstaging (sCKD-upstaging), defined as any upstaging to CKD=3a. The predictive model was built as follow: 1) univariable Cox regression analysis excluded non-significant predictors (p=0.05); 2) predictors were checked for multicollinearity; 3) stepwise bidirectional Cox regression analysis identified the best model. Model accuracy was reported as Harrel C-index. Internal validation (bootstrapping) and calibration were performed. Nomogram predicting 5-years sCKD-upstaging risk was displayed. Regression tree analysis identified potential cut-offs in nomogram-derived probability. Based on this cut-offs, three risk class were derived and tested with Kaplan-Meier analysys (log-rank test). Results: Overall, 965 patients were identified. At Kaplan-Meier analysis 5-years sCKD-upstaging free-survival was 69.1% (95%CI 64.8-73.7%). The final model included baseline eGFR, postoperative AKI, WIT, hypertension, age at surgery, solitary kidney status and multiple lesions. Model accurately predicted 5-years sCKD-upstaging (C-index 83.3%). Two nomogram cut-off (16 and 37%) were identified. Based on this cut-off, the overall cohort was stratified in three risk categories (low: <16% vs intermediate:16-37% vs high:>37%). Kaplan-Meier analysis depicted a significant reduction in five-years sCKD-upstaging free survival rates between low vs intermediate vs high-risk (97.0% vs 71.9% vs 32.2%, respectevely, p <0.001). Conclusions: We developed a novel nomogram that accurately predicted five-years significant CKD-upstaging. Moreover, we were able to identify three subgroups with substantial difference in risk profiles. If externally validated this nomogram may represent a useful tool to improve patient counselling and management. SOURCE OF Funding: no
