We read with interest the report of the recent meta-analysis by Turpie and colleagues1 of data from 4 randomized, double-blind clinical trials in which patients undergoing major orthopedic surgery were randomized to receive fondaparinux sodium or enoxaparin sodium to prevent venous thromboembolism (VTE). The authors report that fondaparinux reduced the risk of VTE by approximately 50% but that it also increased the risk of major bleeding by 53%. They point out that the difference in bleeding events was accounted for mainly by an excess risk among fondaparinux-treated patients of major bleeding with a "bleeding index of 2 or more" (number of units of packed red blood cells or whole blood transfused plus prebleeding minus postbleeding hemoglobin [grams per deciliter] values). They suggest that this measure is not "clinically relevant" but provide no data to substantiate this claim.
Oster G, Vera M. Clinical Relevance of Bleeding Index. Arch Intern Med. 2003;163(22):2794. doi:10.1001/archinte.163.22.2794-a