Simplified PADUA Renal (SPARE) Nephrometry Scoring System: External Validation, Interobserver Variability, and Comparison with RENAL and PADUA in a Single-center Robotic Partial Nephrectomy Series

Abstract

BackgroundThe RENAL (radius [R], exophytic/endophytic [E], nearness to collecting system/sinus [N], anterior/posterior [A], and location relative to polar lines [L]) and the PADUA (preoperative aspects and dimensions used for an anatomical classification) scores help in quantifying tumor complexity. However, nephrometry scoring systems have low interobserver variability. To simplify and improve score reproducibility, a new Simplified PADUA Renal (SPARE) scoring system was introduced. ObjectiveTo externally validate the SPARE nephrometry scoring system and to determine its interobserver variability. Design, setting, and participantsA total of 202 patients were included in the analysis. We performed a retrospective analysis of robot-assisted partial nephrectomy (RAPN) cases for a single renal mass performed at a single academic institution during the period 2008–2018. For each renal mass, PADUA, RENAL, and SPARE nephrometry scores were calculated. Outcome measurements and statistical analysisThree urology residents (URs), two urology attendings (UAs), two radiology residents (RRs), and one radiology attending (RA) retrospectively reviewed computed tomography scans blinded to clinical outcomes. The accuracy of the SPARE nephrometry score in the prediction of any complication (Clavien grade ≥1) was compared with other scoring systems in a univariable and a multivariate fashion. The area under the curve (AUC) and kappa statistics were used to assess interobserver variability of the SPARE score. Results and limitationsThe SPARE score was not inferior to the PADUA and RENAL scores (AUC 0.61, 0.59, and 0.57, respectively, p = 0.43). Patients with intermediate to high SPARE scores had longer operative time (158 vs 135 min, p = 0.10) and a higher rate of complications (28% vs 14%, p = 0.012). Univariable analysis predicting overall complications showed that RRs performed slightly better than URs and UAs using the SPARE score. Interobserver agreement was 84% between an RA and an RR (kappa 0.42), 85% between an RA and a UA (kappa 0.39), and 85% between an RA and a UR (kappa 0.45). ConclusionsThese findings confirm that the SPARE nephrometry scoring system is a reproducible and easy tool offering overall fair interobserver agreement regardless of years of training or type of practice, while maintaining the predictive capabilities of more established nephrometry scores. Patient summaryIn this study, a novel and simple classification system was assessed using a sample of cases from our institution to define surgical complexity renal masses detected on radiological imaging. Our findings suggest that this tool can be useful in clinical practice to facilitate the characterization of renal masses and predict the complications of surgical treatment.