Risk for postoperative infection after transfusion of white blood cell-filtered allogeneic or autologous blood components in orthopedic patients undergoing primary arthroplasty

BACKGROUND: This study was designed to obtain data on the incidence of postoperative infection in patients undergoing elective orthopedic surgery and receiving white blood cell (WBC)-filtered blood components prepared according to current standards. STUDY DESIGN AND METHODS: A total of 308 consecutive orthopedic patients who opted for preoperative autologous blood donation (PAD) for primary unilateral hip and knee replacement surgery were enrolled in a prospective observational study of the incidence of postoperative infection. Patients with contraindications for PAD or with any infectious disease were not included in the study. To identify probably confounding factors, differences between patient groups were analyzed first. Identified factors, which differed between groups, and variables describing blood supply were further tested in uni- and multivariate logistic regression analysis for their independent influence on development of postoperative infection. Infection rates were compared on the basis of actual transfusion groups. RESULTS: Of the 308 study patients, 101 were not transfused, 85 received their PAD, 100 received allogeneic WBC-filtered red blood cells (RBCs), and 22 were given autologous RBCs and additionally allogeneic WBC-filtered RBCs. Overall the infection rate was 6.82 percent (21/308). Infection rates varied significantly between transfusion groups (no transfusion, 6.9%; autologous RBCs, 1.2%; allogeneic WBC-filtered RBCs, 12.0%; both transfusion types, 4.6%; p = 0.03). Allogeneic recipients showed significantly more infections compared to autologous recipients (p = 0.0053). Multivariate regression analysis confirmed transfusion of allogeneic WBC-filtered RBCs as an independent variable predicting postoperative infection (odds ratio, 23.65; confidence interval, 1.3-422.1; p = 0.01). CONCLUSION: Differences in postoperative infection rates between allogeneic and autologous recipients are still observable, although universal WBC filtration has been introduced into clinical practice.