The relationship between perioperative blood transfusion and overall mortality in patients undergoing radical cystectomy for bladder cancer

Objectives: The relationship between perioperative blood transfusion (PBT) and oncologic outcomes is controversial. In patients undergoing surgery for colon cancer and several other solid malignancies, PBT has been associated with an increased risk of mortality. Yet, the urologic literature has a paucity of data addressing this topic. We sought to evaluate whether PBT affects overall survival following radical cystectomy (RC) for patients with bladder cancer. Methods: The medical records of 777 consecutive patients undergoing RC for urothelial carcinoma of the bladder were reviewed. PBT was defined as transfusion of red blood cells during RC or within the postoperative hospitalization. The primary outcome was overall survival. Clinical and pathologic variables were compared using chi(2) tests, and Cox multivariate survival analyses were performed. Results: A total of 323 patients (41.6%) underwent PBT. In the univariate analysis, PBT was associated with increased overall mortality (HR 1.40, 95% CI 1.11-1.78). Additionally, an independent association was demonstrated in a non-transformed Cox regression model (HR, 95% CI 1.01-1.36) but not in a model utilizing restricted cubic splines (HR 1.03, 95% CI 0.77-1.38). The c-index was 0.78 for the first model and 0.79 for the second. Conclusions: In a traditional multivariate model, mirroring those that have been applied to this question in the general surgery literature, we demonstrated an association between PBT and overall mortality after RC. However, this relationship is not observed in a second statistical model. Given the complex nature of adequately controlling for confounding factors in studies of PBT, a prospective study will be necessary to fully elucidate the independent risks associated with PBT. (C) 2013 Elsevier Inc. All rights reserved.