Background: The RENAL nephrometry score (RNS) allows description of the anatomy and the complexity of renal masses. This study aimed to investigate the interobserver reproducibility of the RNS between a radiologist and a urologist.
Methods: The computed tomography (CT) scans of patients undergoing partial nephrectomy in the authors' department between June 2010 and June 2013 were analyzed for determination of the RNS by a urologist and a radiologist blinded to the medical records. Cohen's kappa coefficient was used for interobserver reproducibility assessment. Correlations with per- and postoperative complication rates and renal function were assessed.
Results: The study included 52 consecutive patients with a mean age of 55 years. The average score was 7.4 ± 1.7 for the urologist and 7.3 ± 1.5 for the radiologist. The Cohen's kappa was 0.81 for R, 0.47 for E, 0.63 for N, 0.28 for A, and 0.21 for L. The Pearson's coefficient for the total RNS was 0.70. The operative time and the occurrence of major complications were significantly correlated with the complexity assessed by the score of both observers. In the univariate analysis, the RNS, the American Society of Anesthesiologists score, and the patient's age were significantly associated with major complication rates. In the multivariate analysis, the RNS remained significantly associated with major complications. No significant difference in postoperative renal function according to complexity group was found by either the urologist or the radiologist.
Conclusions: The reproducibility of the RNS between the radiologist and the urologist was not very good, especially for some items referring to the location of the tumor, although the major complication rates were significantly associated with the RNS for both observers.