Prognostic nomogram for renal insufficiency after radical or partial nephrectomy

Abstract

Purpose: We analyzed prognostic factors to predict renal insufficiency after partial or radical nephrectomy. We developed and performed internal validations of a postoperative nomogram for this purpose. We used a prospectively updated renal tumor database of more than 1,500 patients.

Materials and methods: From July 1989 to October 2003, 161 partial nephrectomies and 857 radical nephrectomies performed at Memorial Sloan-Kettering Cancer Center for renal cortical tumors were analyzed. Computerized tomography images were reviewed by a single radiologist. Kidney volume was calculated using the ellipsoid formula, V = L1 x L2 x L3 x pi/6, where V represents volume and L represents length. Renal insufficiency was defined by 2 serum creatinine values greater than 2.0 mg/dl at least 1 month postoperatively. Tumor histology was not an exclusion criterion and yet we excluded cases of bilateral synchronous disease. Prognostic variables were preoperative serum creatinine, American Society of Anesthesiologists score, percent change in kidney volume after surgery, and patient age and sex.

Results: Renal insufficiency was noted in 105 of the 857 patients with radical nephrectomy (12.3%) and in 6 of the 161 with partial nephrectomy (3.7%) studied. Patients had a median followup of 21.2 months (maximum 157.9). The 7-year probability of freedom from renal insufficiency in the cohort was 79.1% (95% CI 74.6 to 83.6). The nomogram was designed based on a Cox proportional hazards regression model. Following internal statistical validation nomogram predictions appeared accurate and discriminating with a concordance index of 0.835.

Conclusions: A nomogram was developed that can predict the 7-year probability of renal insufficiency in patients undergoing radical or partial nephrectomy.