Prevalence of serious bleeding events and intracranial hemorrhage in patients receiving activated protein C: a systematic review and meta-analysis

Abstract

Background: Activated protein C reduces 28-day mortality in patients with severe sepsis, but its anticoagulant properties entail a risk of bleeding.

Objective: The aim of this systematic review was to evaluate the prevalence of serious bleeding events in patients receiving activated protein C.

Methods: We searched the MEDLINE and EMBASE databases for studies that described the prevalence of serious bleeding events and intracranial hemorrhage in patients receiving activated protein C. We calculated the bleeding rates by calculating proportions and 95% CIs for each study, and then pooled the data to derive a pooled proportion and 95% CI.

Results: Our search yielded 17 studies, which included 10,679 patients. The occurrence of serious bleeding events in patients receiving activated protein C ranged from 0.5% to 9.6%, and the pooled prevalence was 3.3% (95% CI 2.4-4.4%) by the random effects model. The occurrence of intracranial hemorrhage ranged from 0% to 1.4%, and the pooled prevalence was 0.44% (95% CI 0.31-0.6%). Sensitivity analysis showed a higher prevalence of bleeding in the observational studies than in the randomized controlled trials. There was substantial clinical and statistical heterogeneity, but no evidence of publication bias.

Conclusions: Activated protein C is associated with significant risk of bleeding, so strict inclusion and exclusion criteria should be set prior to administering activated protein C.