Four‐factor prothrombin complex concentrates for warfarin reversal: do real‐world outcomes differ from controlled clinical trials?

In this issue of TRANSFUSION, Joseph and colleagues1 report on their retrospective analysis of outcomes of use of four-factor prothrombin complex concentrates (4-PCCs) for reversal of warfarin anticoagulation effect. Their report presents an example of “real world” outcomes with respect to thromboembolic events (TEEs) and death occurring within 60 days of receiving 4-PCCs. In their cohort (n = 113), the overall incidences of TEE and death were 6.2% (7/113) and 15% (17/113), respectively. Their criteria for attribution of events to 4-PCCs was occurrence within 3 days of PCC administration; the incidence of TEEs and death was approximately 1.7% each (two events each). Although the direct-acting oral anticoagulants are gaining market share, warfarin is still the most commonly prescribed oral anticoagulant in the United States; in 2011, 33.9 × 107 prescriptions were dispensed.2 Given its narrow therapeutic index, hemorrhage, the most common warfarin toxicity, accounts for a large proportion of drug toxicity–related emergency room visits.3 The incidence of warfarin-associated major hemorrhage ranges from 0.5% to 6.5%, fatal hemorrhage ranges from 0.1% to 1.0%,4-8 and warfarin is ranked among the top 10 drugs with the most reports of adverse events.9-11 Pending the future transition to the safer direct-acting anticoagulants, a large proportion of the population will continue to receive new prescriptions or will remain on warfarin. Thus, warfarin-related hemorrhage will continue to occur. Temporary cessation of warfarin, either in the setting of major hemorrhage or for an invasive procedure, puts patients at risk for thrombotic complications. In a recent report of a cohort of patients undergoing elective invasive procedures that required interruption of warfarin, the incidence of thrombotic complications was approximately 1.8%. Active cancer was a major predictor of both thrombosis and hemorrhage.12 Options for reversal of warfarin anticoagulant effect, for management of hemorrhage or urgent or emergent surgery, consist of administration of fresh-frozen plasma (FFP), vitamin K, recombinant activated Factor VII, and PCCs.13 In the United States, FFP has been the standard of care; however, the demonstrated efficacy and safety of three-factor PCCs (3-PCCs) and 4-PCCs has led to their increasing utilization. Studies reporting on efficacy and safety of PCCs include prospective or retrospective cohort studies with varying management strategies and follow-up. In a meta-analysis of all reported studies, the incidence of thromboembolic complications occurring within 4 days of administration of the PCCs was 1.4% (95% confidence interval [CI], 0.8%-2.1%). The incidences in patients receiving 4-PCCs and 3-PCCs were 1.8% (95% CI, 1.0%-3.0) and 0.7% (95% CI, 0.0%-2.4%), respectively, with an overall total mortality rate of 10.6% (95% CI, 5.9%-16.6%).14 Until recently, there were limited data comparing efficacy and safety of PCC to FFP.15, 16 In their pivotal Phase 3 randomized controlled clinical trial, comparing FFP to 4-PCC, Sarode and coworkers17 demonstrated that 4-PCC was noninferior with respect to efficacy and superior with respect to safety, when administered for reversal of warfarin anticoagulation effect. In this trial,17 the incidences of TEE and deaths attributable to 4-PCCs were 3.9% (4/103) and 1% (1/103), respectively. In another recent report comparing FFP to 4-PCC Goldstein and colleagues18 reported an incidence of TEEs of 7% (6/88) and there was one death (1%, 1/88) attributed to 4-PCC. Given the strict enrollment criteria in controlled clinical trials, when the same treatment is generally applied in clinical practice to uncontrolled patient populations (real-world experiences), outcomes may be worse than reported in controlled trials. With this in mind, the report by Joseph and colleagues1 is an important contribution to the real-world experiences with 4-PCCs. They included all patients receiving 4-PCCs except for patients less than 18 years of age and those not taking warfarin. Approximately 11% of their cohort had active or metastatic malignancy, a known risk factor for thrombosis. The most common indication for warfarin reversal was management of hemorrhage (approx. 60%) followed by urgent surgery (approx. 40%). Multivariate analysis suggested that active malignancy was a predictor of death (odds ratio [OR], 10.5; 95% CI, 2.97-37.43) and existence of a thrombophilia was predictive of TEEs (OR, 10.23; 95% CI, 1.50-69.58). The seemingly similar rates of TEEs when compared with the data from meta-analysis14 and from the Phase 3 trial17 suggests that these rates represent a “real” incidence. Reasons for the high incidence of TEE in the trial by Goldstein and coworkers18 are not apparent but the criteria for attribution were not clearly stated. Withholding the anticoagulant, the underlying indication for the anticoagulant and the administration of the PCC puts patients at risk of TEEs and existence of comorbidities, for example, malignancy and thrombophilia combine to further increase this risk. So how does one practically apply current knowledge for optimal patient care and what are the unanswered questions? Given its prothrombotic risk, patient selection for PCC use is important; PCCs should not be used for warfarin reversal for elective invasive procedures.19 Temporary cessation of warfarin, allowing the international normalized ratio (INR) to drift into a range deemed safe for the procedure and bridging heparin anticoagulation for those at high risk of thrombosis, is the recommended approach. In addition, PCCs should not be used in asymptomatic patients with a supratherapeutic INR at low risk for bleeding; temporary cessation of warfarin with or without oral vitamin K is typically recommended.13 Use of PCC should be restricted to management of major hemorrhage and for urgent or emergent invasive procedures, keeping in mind the overall risk of TEE and the comorbidities that increase this risk. Of the unanswered questions, the optimal dose that controls hemorrhage but does not significantly increase the risk of thrombosis remains to be defined; pending availability of such data, the current suggested dosing guidelines should be followed. The role of a fixed dose of PCC deserves further study.20 Other unanswered questions yet to be addressed include PCC efficacy and safety in patients less than 18 years of age and in those who are not on vitamin K antagonists, for example, the trauma patient, where the incidence of thrombosis is likely to be high. The author has disclosed no conflicts of interest. R.K. Pruthi, MBBS e-mail: [email protected]