Nephrometry scoring systems are used to predict the surgical complexity of partial nephrectomy (PN) but are insufficient to predict renal function after robotic PN (RPN). The study aimed to calculate a new location factor for predicting postoperative renal function and develop a classification including the location factor. We calculated a location index (L-index) and verified an optimal cutoff value to predict renal function after RPN in 163 patients (development cohort). Then, we developed a new classification with the L-index and validated it in 127 patients (external validation cohort). The primary endpoint was an estimated glomerular filtration rate (eGFR) reduction of ≥ 20% from baseline to 6 months after RPN. This outcome occurred in 24 patients (14.7%) in the development cohort and 28 patients (22.0%) in the external validation cohort. The accuracy for predicting the endpoint was evaluated using area under the receiver operating characteristic curve (AUC). The L-index cutoff values were ≤ 15 and ≤ 30 mm. Using the L-index and tumor volume, we developed the LIVED (L-index and volume for prediction eGFR decline) classification dividing patients into three groups. The classification showed a high AUC compared to other nephrometry scoring systems (AUC = 0.858 vs. 0.674–0.744) in a validation cohort. The LIVED classification integrating the L-index, quantified as a location factor, and tumor volume predicted renal function after RPN with high accuracy.
Ohsugi et al. (Tue,) studied this question.