Antithrombin concentrate use in disseminated intravascular coagulation of sepsis: meta-analyses revisited.

Recently, two meta-analyses examined the effect of antithrombin on mortality in patients with sepsis and disseminated intravascular coagulation (DIC) with diverging results. In the meta-analysis observing no significant survival effect of antithrombin, the results of a large trial were included under the assumption that all 2314 participants of the trial had sepsis and DIC, which according to post hoc analyses was not the case. To the other meta-analysis reporting beneficial effect, the same trial contributed only those 229 patients that had confirmed sepsis-induced DIC. Replacing the mixed sepsis patient group with and without DIC with the group of sepsis patients with confirmed DIC provides evidence for a beneficial intervention effect of antithrombin also in the meta-analysis that reported no beneficial effect of antithrombin on mortality. This revised result does not change the overall conclusion that the quantity and quality of evidence supporting the use of antithrombin in sepsis patients with DIC is low.