Introduction of universal prestorage leukodepletion of blood components, and outcomes in transfused cardiac surgery patients

OBJECTIVE: To assess whether introduction of universal leukodepletion (ULD) of red blood cells (RBCs) for transfusion was associated with improvements in cardiac surgery patient outcomes. METHODS: Retrospective study (2005-2010) conducted at 6 institutions. Associations between leukodepletion and outcomes of mortality, infection, and acute kidney injury (AKI) were modeled by logistic regression, and intensive care unit length of stay (LOS) in survivors was explored using linear regression. To examine trends over time, odds ratios (ORs) for outcomes of transfused were compared with nontransfused patients, including a comparison with nontransfused patients who were selected based on propensity score for RBC transfusion. RESULTS: We studied 14,980 patients, of whom 8857 (59 ) had surgery pre-ULD. Transfusions of RBCs were made in 3799 (43 ) pre-ULD, and 2525 (41 ) post-ULD. Administration of exclusively leukodepleted, versus exclusively nonleukodepleted, RBCs was associated with lower incidence of AKI (adjusted OR 0.80, 95 confidence interval [CI] 0.65-0.98, P = .035), but no difference in mortality or infection. For post-ULD patients, no difference was found in mortality (OR 0.96, 95 CI 0.76-1.22, P = .76) or infection (OR 0.91, 95 CI 0.79-1.03, P = .161); however, AKI was reduced (OR 0.79 95 CI 0.68-0.92, P = .003). However, ORs for post-ULD outcomes were not significantly different in nontransfused, versus transfused, patients. Furthermore, those who received exclusively nonleukodepleted RBCs were more likely to have surgery post-ULD. CONCLUSIONS: Universal leukodepletion was not associated with reduced mortality or infection in transfused cardiac surgery patients. An association was found between ULD and reduced AKI; however, this reduction was not significantly different from that seen in nontransfused patients, and other changes in care most likely explain such changes in renal outcomes.