A Cochrane review [Abstract] 1 included eight studies with a total of 310 subjects. The studies differed widely in transfusion policies, criteria for neutropenia (range < 0.1 to < 1.0 × 109 /l), and definition of infection requiring treatment.
For mortality, which was extracted from six trials, the summary RR = 0.64 in favour of transfusion (95% CI 0.33, 1.26), but with evidence of significant statistical heterogeneity. The data for the combined RR for mortality for the four studies transfusing higher granulocyte doses greater than 1×1010 indicated a significant summary RR= 0.37 (95% CI 0.17, 0.82); Chi-square 3.9, I2 23%. Data on rates of reversal of infection could be extracted from four studies, and the combined RR was 0.94 (95% CI 0.71, 1.26), again with evidence of heterogeneity.
Authors' comment:All published studies had limitations and were undertaken over 25 years ago. In addition to the observed clinical diversity between all studies, uncertainty about the quantitative and qualitative analyses for these studies is compounded by methodological deficiencies.
Primary/Secondary Keywords