Efficacy and safety of prothrombin complex concentrate in patients treated with rivaroxaban or apixaban compared to warfarin presenting with major bleeding

Mortality is high in these patients with reported rates of up to 65% within 30-90 days (Hart et al., 2012; Hankey et al., 2014;). Direct oral factor Xa inhibitors such as rivaroxaban, apixaban and edoxaban, and the oral thrombin inhibitor dabigatran have become the preferred first-line treatment choice over VKAs, but there is evidence that these drugs may increase gastrointestinal bleeding, and for patients with arterial thrombosis and those who need anticoagulation due to mechanical valves or higher intensity anticoagulation VKAs are still the anticoagulant of choice. Although ICH is less frequent in patients on DOACs, as with any major bleeding event (MBE), providing circulatory support with blood products is an essential part of treatment management until the anticoagulant wears off.

In patients with ICH and other MBE, the cornerstone of VKA reversal are intravenous (IV) vitamin K and prothrombin complex concentrate (PCC) (Keeling et al., 2011). Newer treatment options are beginning to become available in the US, such as idarucizumab, (a Fab fragment humanised antibody against dabigatran), and andexanet alpha, (a recombinant variant of human factor Xa and a reversal agent for rivaroxaban- and apixaban-related MBE). Idarucizumab has also been approved in November 2015 by the European Medicine Agency. However, for this reason, several guidelines still recommend the administration of PCC or activated PCC, and this remains the approved strategy in the UK for the management of MBE in patients on rivaroxaban or apixaban (Keeling et al., 2011; Makris et al., 2013; NICE 2016). There are numerous studies which support this treatment approach, however none have previously included control groups to assess the thrombotic risk of PCC in reversing rivaroxaban or apixaban compared to VKAs.

Now, in a retrospective single centre observational UK study between January 2016 and April 2018, Arachchillage et al. have investigated the efficacy and safety of PCC in the management of MBE in 344 patients - this is the largest cohort to date of patients who had MBE and received PCC (median range dose 2000 units of factor IX [1000-4500]) to reverse the anticoagulant effect of warfarin, rivaroxaban or apixaban. All patients had a full blood count, coagulation screen (including fibrinogen level), and renal and liver function tests on presentation with MBE. Patients who received PCC prior to emergency surgery or procedure were excluded. Key data were collected from electronic patient records including haemoglobin levels, platelet count, prothrombin time (PT), activated partial prothrombin time (APTT), occurrence of thromboembolic events, recurrence of bleeding at the same site within 30 days after PCC administration, and 30-day mortality. In patients receiving warfarin the International Normalised Ratio (INR) was also measured prior to and within 24 hours of PCC or IV vitamin K administration.

Bleeding events, management of MBEs and assessment of outcomes

MBE were classified as ICH, gastrointestinal, visceral (including bleeding into pulmonary or abdominal organs), genitourinary or musculoskeletal. Points of interest for non-ICH bleeding were haemoglobin level, transfusion of blood products, intervention or surgery to stop bleeding and the administration of other haemostatic drugs prior- or post-PCC administration. There was no difference in the total dose of PCC or the units/kg administered in the three anticoagulant groups, however, a significantly higher proportion of patients receiving apixaban (17/40, 42.5%) and rivaroxaban (14/40, 35%) received tranexamic acid versus warfarin (51/264, 19.3%) (p=0.001). This trend was also observed when analysis was restricted to patients with ICH. Despite concerns amongst some clinicians, no difference in thrombosis was observed following PCC with or without tranexamic acid. Blood products use was the same between groups. In total, 86% of study patients on warfarin received vitamin K.

Safety endpoints were the occurrence of an objectively verified arterial (stroke, myocardial infarction, or arterial thromboembolism) or venous thromboembolism (deep vein thrombosis or pulmonary embolism) after treatment with a PCC.

Site of major bleeding

In all three anticoagulant groups, ICH was the most common indication for PCC (39.8%, 137/344), followed by gastrointestinal bleeding (27%, 93/344). For patients receiving rivaroxaban, ICH was more frequently the indication for PCC (62.5%, 25/40), compared to apixaban (52.5%, 21/40) or warfarin (34.5%, 91/264) (p=0.002). Meanwhile, warfarin was more frequently associated with visceral bleeding compared to rivaroxaban or apixaban (24.2% vs. 5% vs 10% respectively, p=0.003). There were no differences in the other bleeding sites based on anticoagulant.

Results showed that there was no difference in the efficacy (30-day mortality, re- bleeding) or safety (thrombosis) in the use of PCC for MBE in patients on rivaroxaban, apixaban or warfarin.

These findings are in keeping with results from a number of previous studies, including two prospective observational studies, one with 84 patients, where PCC was effective in the management of MBE in 58/84 (69.1%) of patients (Majeed et al., 2017), and another with 66 patients who were administered PCCs for the reversal of MBE due to rivaroxaban or apixaban where PCC was effective in 65% of patients (Schulman et al., 2018).

In conclusion, PCC has been shown to be an effective anticoagulant reversal for patients presenting with rivaroxaban-, apixaban-, or warfarin-associated MBE, yet it will be important to reassess treatment options once a specific reversal agent for rivaroxaban and apixaban is approved in the UK and Europe.