Impact of positive surgical margins on overall survival after partial nephrectomy—A matched comparison based on the National Cancer Database

Abstract

Introduction The impact of positive surgical margins (PSM) in partial nephrectomy (PN) has been a controversy. Previous studies on the relationship between PSM and overall survival (OS) were either underpowered or had highly dissimilar groups. We used the National Cancer Database with propensity score matching to determine the association between PSM and OS after PN. Materials and methods We identified patients with T1/T2 N0M0 renal cancer treated with PN between 2004 and 2009, and divided them into 2 groups based on their margin status. We used propensity score matching to ensure similarities in age, comorbidity score (CCI), tumor size, histology, and grade between groups. Covariates were compared by χ2 test. Cox multiple regression was used to estimate the hazard ratios (HR) for all-cause mortality. OS between matched groups were compared by log-rank, Breslow and Tarone-Ware tests. Results After excluding those with missing data on margin or survival status, 20,762 patients were eligible for matching. Each matched group had 1,265 patients, similar in age, sex, race, CCI, tumor size, histology, and grade. There were 386 recorded all-cause mortalities over a median follow-up duration of 72.6 months. Cox multiple regression showed a higher risk of all-cause mortality among cases with PSM (HR: 1.393, P = 0.001). Old age, high CCI, and large tumors had higher risks, while papillary and chromophore histologic subtypes had lower risks. PSM was associated with significantly worse OS by log-rank, Breslow, and Tarone-Ware tests. Conclusion PSM is associated with significantly worse OS after PN.