Putting the record straight on aprotinin as safe and effective: results from a mixed treatment meta-analysis of trials of aprotinin.

Abstract

Meta-analysis of small, randomized, placebo-controlled trials demonstrated efficacy and safety of aprotinin. After highly publicized retrospective studies and the early stopping of the Blood Conservation Using Antifibrinolytics in a Randomized Trial (BART), aprotinin was withdrawn. We conducted a new meta-analysis (including BART) on safety and efficacy of aprotinin in cardiac surgery. We conducted a mixed treatment comparisons network meta-analysis estimating the effects of aprotinin and alternative agents in reducing blood loss during surgery. We implemented a combination of direct and indirect evidence in mixed treatment comparisons and estimated relative effects for different agents on all-cause mortality and return to the operating room for bleeding and conducted a supportive analysis of the effects of different agents with only directly randomized trials. Mixed treatment analysis of 88 trials randomizing 15,528 patients to 1 of 3 antifibrinolytic agents demonstrated no difference in mortality between placebo and antifibrinolytic agents. Analysis of aprotinin versus tranexamic acid and ε-aminocaproic acid in 17 and 6 trials, respectively and tranexamic acid versus ε-aminocaproic acid in 5 trials demonstrated no difference in mortality between treatment allocations. All agents were superior to placebo in reducing reexploration for bleeding, with aprotinin numerically superior: aprotinin odds ratio, 2.6 (95% confidence interval, 1.9-3.7); tranexamic acid odds ratio, 1.79 (1.2-2.9), and ε-aminocaproic acid odds ratio, 2.4 (1.3-6.6). This mixed treatment comparisons meta-analysis demonstrates no increased mortality risk with aprotinin versus other antifibrinolytic agents. All agents were superior to placebo in reducing reexploration for bleeding after adult cardiac surgery.