Thromboembolic Events after Vitamin K Antagonist Reversal with 4-Factor Prothrombin Complex Concentrate: Exploratory Analyses of Two Randomized, Plasma-Controlled Studies

Study objective We evaluated thromboembolic events after vitamin K antagonist reversal in post hoc analyses of pooled data from 2 randomized trials comparing 4-factor prothrombin complex concentrate (4F-PCC) (Beriplex/Kcentra) with plasma. Methods Unblinded investigators identified thromboembolic events, using standardized terms (such as "myocardial infarction," "deep vein thrombosis," "pulmonary embolism," and "ischemic stroke"). A blinded safety adjudication board reviewed serious thromboembolic events, as well as those referred by an independent unblinded data and safety monitoring board. We descriptively compared thromboembolic event and patient characteristics between treatment groups and included detailed patient-level outcome descriptions. We did not power the trials to assess safety. Results We enrolled 388 patients (4F-PCC: n=191; plasma: n=197) in the trials. Thromboembolic events occurred in 14 of 191 patients (7.3%) in the 4F-PCC group and 14 of 197 (7.1%) in the plasma group (risk difference 0.2%; 95% confidence interval -5.5% to 6.0%). Investigators reported serious thromboembolic events in 16 patients (4F-PCC: n=8; plasma: n=8); the data and safety monitoring board referred 2 additional myocardial ischemia events (plasma group) to the safety adjudication board for review. The safety adjudication board judged serious thromboembolic events in 10 patients (4F-PCC: n=4; plasma: n=6) as possibly treatment related. There were 8 vascular thromboembolic events in the 4F-PCC group versus 4 in the plasma group, and 1 versus 6 cardiac events, respectively. Among patients with thromboembolic events, 3 deaths occurred in each treatment group. All-cause mortality for the pooled population was 13 per group. We observed no relationship between thromboembolic event occurrence and factor levels transiently above the upper limit of normal; there were no notable differences in median factor or proteins C and S levels up to 24 hours postinfusion start in patients with and without thromboembolic events. Conclusion The incidence of thromboembolic events after vitamin K antagonist reversal with 4F-PCC or plasma was similar and independent of coagulation factor levels; small differences in the number of thromboembolic event subtypes were observed between treatment groups.