Thromboembolic events after vitamin K antagonist reversal with 4-factor prothrombin complex concentrate: exploratory analyses of two randomized, plasma-controlled studies
Vitamin K antagonists such as warfarin are commonly prescribed, with an estimated 3.4 million patients (~1% of the population) using them in the United States. They are also invariably associated with adverse events – and are one of the top 10 drugs with serious adverse side effects according to the FDA. Their action inhibits the synthesis of vitamin K-dependent clotting factors and proteins; patients who are actively bleeding, or who have an urgent need for an invasive procedure need rapid replacement of those clotting factors. According to current literature, combination therapy with exogenous vitamin K and either fresh frozen plasma (FFP) or prothrombin complex concentrate (PCC) is recommended. FFP is cheaper, but has drawbacks including: possibility of cross-infection, requirement for ABO typing, delay for thawing and prep, longer infusion times, and potential for transfusion-related acute lung injury. PCC has a more rapid infusion time, no requirement for cross-matching or thawing and less risk of virus transmission, although there has been a concern that PCC may carry higher risk of thromboembolic events due to the concentration of clotting factors. It is difficult to differentiate reversal-related events from the underlying risks that indicated anticoagulation in the first place.
Two recent randomised trials of 4 factor PCC (4F-PCC) were conducted, and this study aimed to pool the results of their safety analyses to allow a fuller evaluation of the safety profile of 4F-PCC. A secondary post-hoc analysis was conducted on safety data. The included population were patients on a vitamin K antagonist with acute major bleeding, or who required an urgent surgical or invasive medical procedure. Adverse events were recorded up to day 10 post presentation, and serious adverse events were recorded up to day 45. The primary analysis used data directly from the trial databases.
The pooled data comprised 388 patients, of whom 191 received 4F-PCC and 197 received FFP. For patients who experienced thromboembolic events, the demographics, disease characteristics, and patient disposition were similar between the two studies. Of the 388 total patients, 28 experienced one or more thromboembolic event, while three patients experienced a second thromboembolic event. The analysis for those experiencing one event showed an even split; 14 of 191 patients on 4F-PCC (7.3%) and 14 of 197 patients on FFP (7.1%) – a non-significant risk difference of 0.2%. When total number of thromboembolic events was explored, there were 16 events in the 4F-PCC group and 15 events in the FFP group. After review by a panel, 17 serious thromboembolic events were included, seven in the 4F-PCC administration group and nine in the FFP group. In terms of overall timing, 11 of the 16 events in the 4F-PCC group occurred more than 7 days after infusion, while 11 of the 15 events in the FFP group were within a week of administration. The most common indication for anticoagulation was arrhythmia, with gastrointestinal haemorrhage the most common reason for reversal. Despite having a high baseline risk, few patients had resumed anticoagulation at the time of their events; only three of 14 patients in the 4F-PCC group and four of 14 in the FFP group were anticoagulated. The risk of thromboembolic event was not proportional to the dose of reversal agent.
The authors acknowledge the limitations of their study: since it was a post hoc analysis, the study was not powered to detect differences in safety outcomes; staff and patients were not blinded; and events deemed ‘non-serious’ were not reviewed. There was also a lack of doppler screening, meaning subclinical disease may have been missed.
In their discussion, the authors mention that anticoagulation reversal may be associated with thromboembolic events, but that this effect may be a result of re-exposure to the underlying risk. This remains one of the largest controlled assessments of PCC against FFP, and provides a detailed analysis of the effect of reversal strategies on thromboembolic events. The percentage of events and patients affected was very similar, suggesting a similar overall risk. The increase in vitamin K-dependent clotting factors following cessation/reversal of vitamin K inhibition could potentially increase the probability of thromboembolic events, but the lack of a consistent relationship between timing and the events means it is difficult to verify this hypothesis.
This study is a good indication of the relative safety of 4F-PCC and FFP as reversal methods, and additionally highlights the need to reinitiate anticoagulation in a timely manner following reversal for a bleeding event.