A Nomogram to Predict Significant Estimated Glomerular Filtration Rate Reduction After Robotic Partial Nephrectomy

Abstract

Background: Decreased functional outcome after partial nephrectomy is associated with overall mortality.

Objective: To create a model that predicts ≥25% reduction from baseline estimated glomerular filtration rate (eGFR) in patients undergoing robot-assisted partial nephrectomy (RAPN) and to investigate the role of acute kidney injury (AKI) in this patient population.

Design, setting, and participants: A total of 999 patients were identified from a multi-institutional database. Renal function was defined according to the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines for chronic kidney disease (CKD). AKI was defined as >25% reduction in eGFR from pre-RAPN period to discharge.

Outcome measurements and statistical analysis: A nomogram to predict significant eGFR reduction (≥25% from baseline) in the time-frame between 3 and 15mo after RAPN was built based on the coefficients of Cox survival function that ultimately included age, sex, Charlson comorbidity index, baseline eGFR, RENAL nephrometry score, AKI in patients with normal baseline renal function, and AKI on CKD. Such landmark analysis was chosen in order to account for eGFR fluctuations occurring within the first 3mo of RAPN. The proportional hazard assumption was evaluated through the Schönfeld test. Internal validation was performed using the leave-one-out cross validation. Calibration was graphically investigated. The decision curve analysis (DCA) was used to evaluate the net clinical benefit.

Results and limitations: Median (interquartile range [IQR]) age at surgery was 61yr (51, 68). Overall, 146 patients experienced significant eGFR reduction; median follow-up for survivors was 12.4mo. The 15-mo probability of significant eGFR reduction was 19%. All variables fitted into the model, including AKI in patients with normal renal function (hazard ratio [HR]: 4.51; 95% confidence interval [CI]: 3.12, 6.60; p<0.001) and AKI on CKD (HR: 4.90; 95% CI: 2.17, 11.1; p<0.001), emerged as predictors of significant eGFR reduction (all p≤0.048) and were considered to build a nomogram. The internally validated c index was 73%. The model demonstrated excellent calibration and a net benefit at the DCA with probabilities ≥4%.

Conclusions: We developed a nomogram that accurately predicts significant eGFR reduction after RAPN. This model may serve as a tool for early identification of patients at high risk for significant renal function decline after surgery.

Patient summary: We have developed a model for the prediction of renal function loss after partial nephrectomy for renal cancer.

Keywords: Acute kidney injury; Acute on chronic renal failure; Chronic kidney disease; Functional outcome; Kidney cancer; Partial nephrectomy.