The objective was to estimate the effect of
antithrombin therapy on mortality in
disseminated intravascular coagulation (
DIC) of
severe sepsis and
septic shock. Randomized clinical trials (RCT) on patients with
DIC and
severe sepsis or
septic shock assigned to intravenous
antithrombin or placebo were searched. Eligible studies reported death as the outcome measure. Of 35 RCT, 32 trials were excluded because patients were not randomized to
antithrombin versus placebo, or no separate data on patients with
DIC were given. In three RCT, 364 patients with
severe sepsis or
septic shock and
DIC were randomized. The disease severity, definition of
DIC, dose and
duration of treatment varied among the trials. In two of the three RCT, data were from subgroup analyses (patients not stratified by
DIC). The combined odds ratio for short-term all-cause mortality in those who received
antithrombin was 0.649 (95% confidence interval, 0.422-0.998). Data on
bleeding complications in patients treated with
antithrombin were reported only in one of the RCT and were not considered suitable for systematic safety estimation. In
sepsis patients with
DIC, administration of
antithrombin concentrate may increase overall survival. Current available evidence, however, is not suited to sufficiently inform clinical practice.