To evaluate whether surgeon factors, such as training and experience, have a strong impact on selection of surgical approach for treating renal cancers. Nephron-sparing surgery (NSS) has become the reference standard for tumors that are amenable to such an approach. Tumor size and configuration are important predictors of usage of NSS. The RENAL nephrometry score (RNS) has been developed to standardize reporting of tumor complexity, but the performance of this method within individual surgeons' practices, particularly in the community-based setting, has not been evaluated previously.

Clinical data and RNS were collected retrospectively for 300 cases performed by 5 different surgeons with varying NSS usage rates (31-74%).

Mean RNS for patients undergoing NSS (6.0) and radical nephrectomy (RN) (9.3) differed significantly (P <.0001), as did tumor size (2.8 vs 6.3 cm, P <.0001). RNS was a better predictor of surgery type (R² = .55) than was tumor size (R² = .43) for all 5 surgeons. In univariable analysis, individual surgeon, surgery year, glomerular filtration rate, tumor size, RNS, and each RNS component predicted NSS vs RN (each P <.05). In multivariable analysis, surgeon, tumor size, exophytic amount, and nearness to collecting system were independent predictors of NSS usage (P <.001).

Despite significant variation in NSS usage by individual surgeons at a community-based health system, RNS appears to be valid for both low and high usage. With increasing usage of NSS worldwide, RNS appears to reflect current patterns and may inform future practice for surgeons of all backgrounds.