A Nomogram for the Prediction of Intermediate Significant Renal Function Loss After Robot-assisted Partial...

Robot-assisted partial nephrectomy (RAPN) is increasingly adopted for the treatment of localized renal tumors; however, rates and predictors of significant renal function (RF) loss after RAPN are still poorly investigated, especially at a long-term evaluation.

To analyze the predictive factors and develop a clinical nomogram for predicting the likelihood of ultimate RF loss after RAPN.

We prospectively evaluated all patients treated with RAPN in a multicenter series (RECORd2 project).

Significant RF loss was defined as >25% reduction in estimated glomerular filtration rate (eGFR) from preoperative assessment at 48th month follow-up after surgery. Uni- and multivariable logistic regression analyses for RF loss were performed. The area under the receiving operator characteristic curve (AUC) was used to quantify predictive discrimination. A nomogram was created from the multivariable model.

A total of 981 patients were included. The median age at surgery was 64.2 (interquartile range [IQR] 54.3-71.4) yr, and 62.4% of patients were male. The median Charlson Comorbidity Index (CCI) was 1 (IQR 0-2), 12.9% of patients suffered from diabetes mellitus, and 18.6% of patients showed peripheral vascular disease (PVD). The median Preoperative Aspects and Dimensions Used for an Anatomical (PADUA) score was 7 (IQR 7-9). Imperative indications to partial nephrectomy were present in 3.6% of patients. Significant RF loss at 48th month postoperative evaluation was registered in 108 (11%) patients. At multivariable analysis, age (p = 0.04), female gender (p < 0.0001), CCI (p < 0.0001), CCI (p < 0.0001), diabetes (p < 0.0001), PVD (p < 0.0001), eGFR (p = 0.02), imperative (p = 0.001) surgical indication, and PADUA score (p < 0.0001) were found to be predictors of RF loss. The developed nomogram including these variables showed an AUC of 0.816.

We developed a clinical nomogram for the prediction of late RF loss after RAPN using preoperative and surgical variables from a large multicenter dataset.

We developed a nomogram that could represent a clinical tool for early detection of patients at the highest risk of significant renal function impairment after robotic conservative surgery for renal tumors.