Effect of baseline glomerular filtration rate on survival in patients undergoing partial or radical nephrectomy for renal cortical tumors

OBJECTIVE: To evaluate the effect of baseline renal function and comorbidity index on survival in patients with renal tumors. PATIENTS AND METHODS: We retrieved demographic, clinicopathologic, body mass index (BMI), and comorbidity data (assessed by Charlson-Romano index and hypertension) for 1479 patients who underwent partial or radical nephrectomy between January 4, 1995, and June 9, 2005, for localized renal tumors. We used the abbreviated Modified Diet and Renal Disease equation to estimate glomerular filtration rate (eGFR) using the last preoperative serum creatinine measurement. Body mass index and eGFR time trends were analyzed with linear regression. The effect of BMI, comorbidity, and baseline eGFR on disease-free and overall survival was studied using Cox regression controlling for pathologic stage, nodal status, and metastasis. RESULTS: Over a 10-year interval, median BMI increased from 27 (interquartile range [IQR], 24-31) to 28 (IQR, 25-31; P=.004), and median baseline eGFR decreased from 70 (IQR, 58-80) to 63 mL/min per 1.73 m(2) (IQR, 57-78; P <.001). Multivarlate regression demonstrated an association between year of surgery and baseline eGFR (P < 001) even after adjusting for age, sex, comorbidity, BMI, and tumor size. We repeated the analysis for patients aged 18 to 70 years, and this association persisted (P <.001). Baseline eGFR, BMI, and comorbidity were not associated with disease-free survival after controlling for stage. However, moderately reduced baseline eGFR (45-60 mL/min per 1.73 m(2)) and severely reduced eGFR (< 45 ml./min per 1.73 m(2)) were significantly associated with overall survival (hazard ratio, 1.5; P <.003; and hazard ratio, 2.8; P <.001; respectively). CONCLUSION: Baseline eGFR has declined over the past decade. Nephron-sparing techniques should be considered for patients with severely diminished baseline eGFR.