Authors: Sun M, Trinh QD, Bianchi M, Hansen J, Hanna N, Abdollah F, Shariat SF, Briganti A, Montorsi F, Perrotte P, Karakiewicz PI
Title: A non-cancer-related survival benefit is associated with partial nephrectomy.
Journal: Eur Urol 61(4):725-31
Date: 2012 Apr
Abstract: BACKGROUND: Partial nephrectomy (PN) may better protect against other-cause mortality (OCM) when compared with radical nephrectomy (RN) in patients with localized renal cell carcinoma (RCC). OBJECTIVE: Test the effect of treatment type on OCM. DESIGN, SETTING, AND PARTICIPANTS: Using the Surveillance Epidemiology and End Results-Medicare-linked database, 4956 RN patients (82%) and 1068 PN patients (18%) with T1a RCC were identified (1988-2005). MEASUREMENTS: To adjust for inherent differences between treatment types, we relied on propensity-matched analyses. One-to-one matching was performed according to age, sex, race, baseline Charlson comorbidity index (CCI), baseline diagnosis of hypercalcemia and hyperlipidemia, socioeconomic status (SES), population density, tumor size, and year of surgery. The 2- and 5-yr OCM rates were computed using cumulative incidence. Univariable and multivariable competing-risks regression analyses for prediction of OCM were performed according to treatment type. Adjustment was made for cancer-specific mortality (CSM), patient age, CCI, sex, race, SES, tumor grade, and year of surgery. RESULTS AND LIMITATIONS: Following propensity-based matching, 1068 RN patients were matched with 1068 PN patients. The 2- and 5-yr OCM rates after nephrectomy were 5.0% and 16.0% for PN versus 6.9% and 18.1% for RN, respectively. In the postpropensity multivariable analyses, patients who underwent PN were significantly less likely to die of OCM compared with their RN-treated counterparts (hazard ratio [HR]: 0.83; 95% confidence interval, 0.69-0.98; p=0.04). Increasing age (HR: 1.08, p<0.001), higher CCI (HR: 1.14, p<0.001), female gender (HR: 0.79, p=0.02), baseline hypercalcemia (HR: 2.05, p=0.03), baseline hyperlipidemia (HR: 0.73, p=0.003), and year of surgery (HR: 0.95, p=0.003) were independent predictors of OCM. CONCLUSIONS: Compared with PN-treated patients, RN-treated patients are more likely to die of OCM after surgery, even after adjusting for CSM, as well as baseline CCI. Consequently, PN should be offered whenever technically feasible.