Objective: To assess the putative impact of perioperative blood transfusions on overall and disease-free survival in patients undergoing curative resection of stage I–III rectal cancer by applying propensity-scoring methods.
Background: Whether perioperative blood transfusions negatively impact survival remains a matter of great debate.
Methods: In a single-center study, 401 patients undergoing open curative resection of stage I–III rectal cancer between 1996 and 2008 were assessed. The median follow-up was 34.2 months. Patients who did and did not receive perioperative blood transfusions were compared using Cox regression and propensity score analyses.
Results: Overall, 217 patients (54.1%) received blood transfusions. Patients' characteristics were highly biased concerning transfusions (propensity score 0.77 ± 0.23 vs. 0.28 ± 0.25; P < 0.001). In unadjusted analysis, blood transfusions were associated with a 119% increased risk of mortality [hazard ratio (HR): 2.19, 95% confidence interval (CI): 1.34–3.57, P = 0.001]. In propensity score-adjusted Cox regression (HR: 1.02, 95% CI: 0.65–1.58, P = 0.970), blood transfusions did not increase the risk of overall survival. Similarly, in propensity score-adjusted Cox regression (HR: 0.86, 95% CI: 0.60–1.23, P = 0.672), blood transfusions were not associated with an increased risk of recurrence.
Conclusions: This is the first propensity score-based analysis providing compelling evidence that worse oncological outcomes after curative rectal cancer resection in patients receiving perioperative blood transfusions are caused by the clinical circumstances requiring transfusions, not due to the blood transfusions themselves. Therefore, concerns about overall and disease-free survival should be no issue in the decision-making regarding perioperative blood transfusions in patients undergoing curative rectal cancer resection.
Blood transfusions are associated with a deteriorated survival after curative resection of stage I–III rectal cancer patients because of clinical circumstances requiring transfusions, not due to the blood transfusions themselves. Hence, concerns about survival should not be a factor in the decision-making regarding perioperative blood transfusions.
*Department of Surgery, Kantonsspital St. Gallen, St. Gallen, Switzerland
†Institute of Medical Biometry and Informatics, University of Heidelberg, Heidelberg, Germany
‡Department of Oncology, Kantonsspital St. Gallen, St. Gallen, Switzerland; and
§University Clinic for Visceral Surgery and Medicine, University of Bern, Inselspital, Bern, Switzerland
Reprints: Ignazio Tarantino, MD, MSc, Department of Surgery, Kantonsspital St. Gallen, 9007 St. Gallen, Switzerland. E-mail: email@example.com.
Rene Warschkow and Ulrich Güller contributed equally to this work.
Disclosure: The authors declare no conflicts of interest.