White-blood-cell-containing allogeneic blood transfusion, postoperative infection and mortality: a meta-analysis of observational 'before-and-after' studies

Abstract

Background and objectives: An association of white blood cell (WBC) reduction with decreased mortality was reported by one observational, before-and-after study. A meta-analysis was undertaken to examine whether this finding is supported by all the evidence currently available from before-and-after studies, and whether these studies support an association of WBC reduction with a decreased risk of postoperative infection.

Materials and methods: Observational, before-and-after studies were retrieved that reported on postoperative infection and/or mortality between January 1997 and June 2003. Six studies met the criteria for meta-analysis. Unadjusted summary odds ratios (ORs) of postoperative infection or mortality in patients transfused after (compared with before) WBC reduction were calculated across the studies if the hypothesis of homogeneity was not rejected. Adjusted summary ORs were calculated across three studies that had reported multivariate analyses.

Results: There was an unadjusted association of WBC reduction with a decreased risk of postoperative infection [summary OR = 0.93; 95% confidence interval (95% CI), 0.88-0.99; P < 0.01] that did not persist following adjustment for confounding factors (summary OR = 0.94; 95% CI, 0.85-1.04; P > 0.05). There was neither an unadjusted nor an adjusted association of WBC reduction with decreased mortality (summary OR = 0.94; 95% CI, 0.71-1.23; P > 0.05; and OR = 0.92; 95% CI, 0.80-1.05; P > 0.05, respectively).

Conclusions: An association of WBC reduction with decreased mortality was not detected across the results available from all before-and-after studies. An unadjusted association of WBC reduction with a decreased risk of postoperative infection exists, but this was not detected across the three studies that reported multivariate analyses.