Prothrombin Complex Concentrates for Perioperative Vitamin K Antagonist and Non–vitamin K Anticoagulant Reversal

DESPITE the increasing use of non–vitamin K antagonist oral anticoagulants, vitamin K antagonists, such as warfarin, are still widely used in patients with atrial fibrillation, venous thromboembolism, and mechanical heart valves. In 2015, approximately 3 million patients were prescribed warfarin in the United States alone.¹  As with all anticoagulants, the main risk associated with vitamin K antagonist therapy is an increased risk for bleeding. Thus, annual rates of major hemorrhagic events ranged from 1.0 to 7.4% in a systematic review of patients with atrial fibrillation receiving vitamin K antagonist therapy for stroke prevention, while rates of intracranial hemorrhage in the same population ranged from 0.1 to 2.5%.² 

Patients receiving vitamin K antagonist therapy who require surgery or an invasive procedure present a specific challenge to clinicians, with an estimated 250,000 to 400,000 patients affected per year in North America alone.³  Data from the Randomized Evaluation of Long-Term Anticoagulation Therapy trial demonstrated that major bleeding (defined as 2 g/dl or more reduction in hemoglobin, transfusion of two or more units of red blood cells, or a critical area or organ bleed) occurred in 3.3% of warfarin-treated patients undergoing elective surgery, increasing to 21.6% in patients who required emergency surgery.⁴  Consequently, effective perioperative management is a key consideration in this population. In patients undergoing elective surgery, current guidelines recommend discontinuing vitamin K antagonist therapy 5 days before the procedure to restore patients’ international normalized ratio to a normal range and to minimize the risk of perioperative bleeding.³  However, in patients who require an emergency surgical procedure, rapid vitamin K antagonist reversal is recommended by replacing the vitamin K–dependent coagulation factors II, VII, IX, and X.⁵ 

Intravenous vitamin K monotherapy is recommended only for vitamin K antagonist reversal in patients in whom surgery can be delayed⁶  because it can take more than 48 h to normalize functional factor levels and restore them to the normal range.⁵  Therefore, in situations requiring rapid vitamin K antagonist reversal, treatment with prothrombin complex concentrates, concomitantly with vitamin K, is more commonly administered. Although fresh frozen plasma (plasma frozen within 8 h of collection) or plasma (frozen within 24 h of collection) was traditionally used for rapid reversal of anticoagulation with vitamin K antagonists, there are multiple limitations to its use, including the need for blood type matching before administration; time required to thaw the product; and risks of fluid overload, pathogen transmission, and transfusion-related acute lung injury.⁵  Furthermore, only minimal benefits have been shown from plasma when reducing the international normalized ratio to less than 1.7 in adults, as well as minimal efficacy for anticoagulation reversal.^7,8 

Prothrombin complex concentrates, which are classed as either four-factor prothrombin complex concentrates (containing coagulation factors II, VII, IX, and X) or three-factor prothrombin complex concentrate (containing factors II, IX and X, but only minimal levels of factor VII; table 1), are stored at room temperature, administered in a smaller volume and shorter infusion time than plasma, and are virally inactivated to minimize the risk of pathogen transmission. Current treatment guidelines recommend prothrombin complex concentrates, specifically four-factor prothrombin complex concentrates, with concomitant intravenous vitamin K, as the preferred therapy for urgent vitamin K antagonist reversal (table 2).^5,6,9,10 

The perioperative management of hemostasis in patients receiving vitamin K antagonists was previously reviewed in this journal in 2008.¹¹  Since then, multiple new studies have investigated vitamin K antagonist reversal in perioperative and periprocedural settings, and prothrombin complex concentrates have become more widely available in the United States and are recommended in guidance documents. Despite the fact that prothrombin complex concentrate is recommended in all guidelines, plasma is still frequently administered for vitamin K antagonist reversal.¹²  This article provides an update on the latest evidence for the use of prothrombin complex concentrates in patients requiring urgent vitamin K antagonist reversal for emergency surgery, but it also reviews current use for non–vitamin K antagonist oral anticoagulant reversal.

A Cochrane Library and PubMed search for publications between January 2008 and December 2017 was conducted with the following search terms: prothrombin complex concentrate* AND (warfarin OR [vitamin K antagonist*]). The search retrieved 423 English-language papers, which were then screened for relevance to the perioperative setting (fig. 1). We excluded preclinical studies and reviews but included all other studies, including case studies.

In total, 35 papers investigating the use of prothrombin complex concentrate for vitamin K antagonist reversal in perioperative settings were identified and included in this review. A further paper investigating prothrombin complex concentrate use in cardiac surgery was identified through a recent meta-analysis of warfarin reversal with prothrombin complex concentrate or fresh frozen plasma,¹³  bringing the total number of papers included to 36. Of these papers, six studies in cardiac surgery and three in neurosurgical settings were identified.

The majority of studies identified in the search were of a retrospective observational design, with limited numbers of patients and lacking a comparator treatment arm. In general, perioperative bleeding episodes were well controlled with prothrombin complex concentrate therapy. The percentage of patients achieving effective hemostasis (no reports of excessive bleeding or bleeding controlled with no requirement for additional products) ranged from 90 to 100%,^14–19  while in a study of 20 patients treated with a four-factor prothrombin complex concentrate, blood loss decreased significantly, from an average of 829 ml in the 6 h preceding four-factor prothrombin complex concentrate administration to 283 ml 6 h after administration.²⁰ 

Reversing vitamin K antagonist anticoagulation, as reflected by a normalized international normalized ratio, is often required for most surgeries and procedures. In the studies identified, prothrombin complex concentrate therapy consistently reduced patients’ international normalized ratio to 1.1 to 1.9 from baseline values of 1.6 to 4.2.^{15,16,20–35}  These reduced international normalized ratios are in line with the target international normalized ratio for patients undergoing surgery of less than 1.5.³ 

As well as reducing the risk for bleeding during surgery, rapid vitamin K antagonist reversal with prothrombin complex concentrates may also reduce time to surgery. For minor procedures such as a lumbar puncture, the time between administration of prothrombin complex concentrate and the start of the procedure was as short as 15 to 30 min.^16,24  In patients requiring more extensive surgery (e.g., heart transplantation, neurosurgery), this period ranged from 2.5 to 5.2 h.^22,23,36 

All-cause mortality rates were generally between 10% and 25%,^{14,22,23,30,37}  although one study in patients requiring neurosurgery because of a life-threatening intracranial hemorrhage reported a mortality rate of 43.5%.²¹  It should be noted that this study included high-risk patients with serious head trauma; the authors also highlighted that delays in therapy administration, subtherapeutic doses of prothrombin complex concentrate, and incorrect vitamin K use may also have been factors contributing to this high mortality rate.²¹ 

Given that fresh frozen plasma or plasma is still often used by clinicians for the urgent reversal of vitamin K antagonist anticoagulation, it is pertinent to look at studies that specifically compared this treatment option with prothrombin complex concentrate in patients undergoing emergency surgical procedures. Overall, three randomized trials^38–40  and one retrospective study⁴¹  comparing these treatment options in this setting were identified. Outside of the literature search, a further study was identified that investigated the administration of prothrombin complex concentrate versus fresh frozen plasma in patients who experienced coagulopathy while undergoing elective pulmonary endarterectomy (table 3).⁴²  All studies used either fresh frozen plasma or plasma frozen within 24 h of collection (frozen plasma). Compared with fresh frozen plasma, only levels of factors V and VIII are slightly reduced in frozen plasma; therefore, for the purposes of this review, the terms fresh frozen plasma and plasma can be used interchangeably.

In the randomized trials, prothrombin complex concentrate was consistently shown to reduce the international normalized ratio more rapidly than plasma. One study, by Goldstein et al., in various surgical indications demonstrated superiority of four-factor prothrombin complex concentrate over plasma for rapid international normalized ratio reduction, with 55% of patients treated with a four-factor prothrombin complex concentrate achieving a target international normalized ratio of 1.3 or less versus 10% of patients in the plasma group at 30 min after the end of infusion (treatment difference, 45.3%; 95% CI, 31.9 to 56.4%; P < 0.0001).⁴⁰  Patients undergoing cardiac surgery with cardiopulmonary bypass demonstrated a significant treatment difference 15 min after infusion, with 17.5% of patients receiving four-factor prothrombin complex concentrate achieving a target international normalized ratio of 1.5 or less compared to no patients who received fresh frozen plasma (P = 0.0068).³⁸  These quicker international normalized ratio reduction times seen with prothrombin complex concentrate compared with plasma should also be considered in the context of the smaller volume that needs to be administered (40 to 100 ml with prothrombin complex concentrate compared with 520 to 1,200 ml with plasma), which leads to a shorter infusion time.^38–40  Thus, in the Goldstein et al. study, mean infusion times were 21 min for four-factor prothrombin complex concentrate and 141 min for plasma. Therefore, despite plasma having almost two additional hours to start exerting a treatment effect, international normalized ratio reduction 30 min after end of infusion was still superior with four-factor prothrombin complex concentrate.⁴⁰ 

An important advantage of a more rapid and predictable international normalized ratio reduction is the ability to proceed to surgery quickly in emergency situations. The length of time from start of infusion to start of surgery was reported in one study: patients who received four-factor prothrombin complex concentrate had a significantly shorter median time to surgery than did patients who received plasma (3.6 h vs. 8.5 h, respectively; P = 0.0098).⁴⁰  In a post hoc analysis of patients with gastrointestinal bleeding requiring procedures in the Goldstein et al. study and in another study investigating four-factor prothrombin complex concentrate for warfarin reversal in patients with acute bleeding,⁴³  the mean time between the start of treatment and the first procedure was significantly shorter in patients given four-factor prothrombin complex concentrate than in those given plasma (P = 0.037).⁴⁴ 

Unlike the randomized trials, the retrospective analysis comparing prothrombin complex concentrate versus fresh frozen plasma in patients undergoing emergency neurosurgery did not investigate the time taken to achieve international normalized ratio reversal. However, both prothrombin complex concentrate and fresh frozen plasma were shown to significantly decrease the international normalized ratio from baseline (P < 0.001), with no significant difference between either group for posttreatment values.⁴¹ 

As well as demonstrating more rapid international normalized ratio reduction, four-factor prothrombin complex concentrates have also been associated with greater clinical efficacy than plasma. In a study by Goldstein et al. in patients undergoing various surgical or invasive procedures, effective hemostasis (defined as intraoperative blood loss not exceeding predicted loss by 50 ml or 30%, normal hemostasis, and no requirement for additional coagulation products) was achieved in 90% of patients who received four-factor prothrombin complex concentrate compared with 75% of patients who received plasma. This treatment difference was significant (P = 0.0142) and demonstrated the superiority of four-factor prothrombin complex concentrate over plasma.⁴⁰  In another study involving patients who received four-factor prothrombin complex concentrate or plasma for vitamin K antagonist reversal while undergoing elective pulmonary endarterectomy, cumulative blood loss was significantly lower up to 12 h postoperatively in the four-factor prothrombin complex concentrate group than in the plasma group (277 ml and 650 ml, respectively; P = 0.0078).⁴² 

In general, similar numbers of patients receiving prothrombin complex concentrate and plasma required transfusions of additional blood products (i.e., platelets, erythrocytes, cryoprecipitate).^40,42  However, one study in patients undergoing cardiopulmonary bypass reported a significantly greater proportion of patients who received plasma requiring additional doses of plasma or four-factor prothrombin complex concentrate versus patients who originally received four-factor prothrombin complex concentrate (100% vs. 30% of patients receiving plasma or four-factor prothrombin complex concentrate, respectively; P < 0.001).³⁸ 

Mortality rates were reported in two studies. Although fewer deaths occurred among patients who received four-factor prothrombin complex concentrate than among patients who received plasma (3.4% vs. 9.1%⁴⁰  and 6.7% vs. 7.3%⁴² ), this difference did not reach statistical significance.^40,42  A recent systematic review and meta-analysis of 13 studies comparing prothrombin complex concentrate versus plasma in patients with warfarin-related bleeding also demonstrated a nonsignificant reduction in mortality outcomes in a subgroup analysis of studies evaluating patients who underwent urgent surgical procedures.¹³  By contrast, when all warfarin-related bleeding events were included, and not just those in the perioperative setting, this meta-analysis demonstrated that prothrombin complex concentrate therapy was associated with a significant reduction in all-cause mortality compared with plasma (P = 0.006).¹³ 

Both three- and four-factor prothrombin complex concentrates were used in the studies identified in our search, although no studies directly compared these different formulations in a surgical setting. However, current guidelines recommend the use of four-factor prothrombin complex concentrates for patients who require rapid vitamin K antagonist reversal.^6,9  These recommendations are aligned with the findings of retrospective studies conducted in patients experiencing major bleeding, which have demonstrated that a greater proportion of patients achieved vitamin K antagonist reversal (as measured by achievement of target international normalized ratios ranging from 1.3 or less to 1.5) with four-factor prothrombin complex concentrate than three-factor prothrombin complex concentrate,^45–48  reaching statistical significance in two studies.^45,47  One study also reported a significantly higher mortality rate (P = 0.001) in patients who received three-factor than in patients who received four-factor prothrombin complex concentrate.⁴⁸ 

Despite being off label, use of recombinant FVIIa has been reported for vitamin K antagonist reversal. Although recombinant FVIIa completely normalizes the international normalized ratio, it does not correct the coagulation defect based on peak thrombin levels and endogenous thrombin potential.^49,50  Two retrospective studies investigated the use of recombinant FVIIa in comparison with a three-factor prothrombin complex concentrate. In one analysis, recombinant FVIIa was shown to reduce the international normalized ratio more rapidly than prothrombin complex concentrates, although this difference did not result in clinical benefit, with a greater proportion of patients receiving recombinant FVIIa experiencing hematoma expansion.⁵¹  The second study also reported more rapid international normalized ratio reduction with recombinant FVIIa versus prothrombin complex concentrate; however, there were no significant differences in thromboembolic events or mortality rates.⁵²  In another retrospective review, a significantly greater proportion of patients achieved a target international normalized ratio of less than 1.3 when receiving a combination of three-factor prothrombin complex concentrate and recombinant FVIIa (79.4%) compared with patients who received either recombinant FVIIa (45.7%) or four-factor prothrombin complex concentrate (50%) alone; however, this combination therapy was associated with a significantly higher proportion of deep vein thromboses (18.7%) than was associated with either recombinant FVIIa (4.2%) or four-factor prothrombin complex concentrate (6.1%).⁵³  The high number of patients achieving international normalized ratio less than 1.3 and experiencing deep vein thromboses with the combination therapy might be indicative of “double dosing” of coagulation factors and the fact that three-factor prothrombin complex concentrates lack small amounts of anticoagulant factors (protein C and S) present in four-factor prothrombin complex concentrates.⁵³ 

Current guidelines do not recommend recombinant FVIIa for urgent vitamin K antagonist anticoagulation reversal.^5,6  Further investigative studies would be beneficial to compare the efficacy and safety of prothrombin complex concentrates and rFVIIa to help inform future practice.

The majority of surgeries carry an inherent risk of bleeding; however, certain surgical indications are associated with an increased bleeding risk in patients receiving vitamin K antagonists.³  Furthermore, uncontrolled bleeding in patients undergoing cardiac, intracranial, or spinal surgery can result in serious clinical consequences.³ 

Intracranial hemorrhage is a particular concern in patients treated with vitamin K antagonists. A report from a large U.S. cohort of more than 13,500 patients with atrial fibrillation demonstrated that almost 88% of deaths resulting from warfarin-associated bleeding were intracranial hemorrhage events, and more than 40% in patients who developed an intracranial hemorrhage died.⁵⁴  Although surgical intervention in cases of intracranial hemorrhage remains controversial, it can be considered in patients who are deteriorating neurologically, have brainstem compression or hydrocephalus as a result of ventricular obstruction,⁵⁵  or those with supratentorial intracranial hemorrhage and a Glasgow coma score of 9 to 12.⁵⁶ 

Few studies have investigated vitamin K antagonist reversal in patients with intracranial hemorrhage in the perioperative setting, and a comprehensive examination of prothrombin complex concentrate use in patients presenting with intracranial hemorrhage, not just those requiring surgical intervention, is outside the scope of this review. A retrospective analysis by Agarwal et al. investigated prothrombin complex concentrate use versus plasma in warfarin-treated patients undergoing emergency surgery for treatment of intracranial hemorrhage.⁴¹  As highlighted earlier, both prothrombin complex concentrate and plasma significantly reduced the international normalized ratio from baseline (P < 0.001); however, no difference between the posttreatment international normalized ratio values were seen between prothrombin complex concentrate and plasma. In-hospital mortality rates were similar between the two treatments with a rate of 17.9% and 14.3% in the prothrombin complex concentrate and plasma groups, respectively.⁴¹ 

In studies investigating plasma and prothrombin complex concentrate treatment in warfarin-treated patients presenting with intracranial hemorrhage and not just patients undergoing neurosurgery, prothrombin complex concentrate use versus plasma resulted in more rapid international normalized ratio reversal,^57–60  and a greater proportion of patients achieved the target international normalized ratio.^58,59,61,62  Mortality rates were not significantly different between treatments,^{41,57,59,63,64}  and fewer patients experienced neurologic deterioration⁶³  or required neurosurgical intervention after prothrombin complex concentrate treatment.⁵⁷  In a study comparing plasma, three-factor prothrombin complex concentrate, and recombinant FVIIa in patients with intracranial hemorrhage, time to anticoagulation reversal was almost twice as long with plasma than with three-factor prothrombin complex concentrate or recombinant FVIIa; international normalized ratio rebound was seen more frequently in patients who received recombinant FVIIa than in those who received either plasma or prothrombin complex concentrate, and the mortality rate was lowest in patients who received three-factor prothrombin complex concentrate (although it should be noted that the population size for these groups was small).⁶⁵  In another retrospective study conducted in the intracranial hemorrhage setting, recombinant FVIIa was shown to reduce the international normalized ratio to 1.3 or less in 83% of patients, compared with only 20% of patients treated with three-factor prothrombin complex concentrate. However, this improved international normalized ratio reversal did not translate into clinical efficacy, with hematoma expansion occurring in a greater proportion of patients receiving recombinant FVIIa.⁵¹ 

Major bleeding events in patients undergoing cardiac surgery have been shown to significantly increase the risk of operative mortality and are also a precursor to reoperation and increased erythrocyte transfusions, both of which are associated with increased morbidity and mortality.⁶⁶  As such, rapid vitamin K antagonist reversal is essential for patients requiring emergency cardiac surgery. In a study in patients undergoing cardiopulmonary bypass, international normalized ratio reversal to 1.5 or less within 15 min was achieved in 35% of patients who were administered four-factor prothrombin complex concentrate compared with 0% of plasma recipients,³⁸  whereas another study of prothrombin complex concentrate in patients undergoing heart transplantation showed that 12% and 75% of patients achieved an international normalized ratio less than 1.5 and less than 1.7, respectively, before transplantation.⁶⁷  International normalized ratio reduction to less than 1.5 was achieved in all four patients in a small case series of patients undergoing heart transplantation; however, the average time to achieve this was 2.45 h (still within the recommended 2- to 3-h window between dosing and incision).³⁶ 

In comparison with plasma, prothrombin complex concentrate treatment was associated with a more rapid international normalized ratio decrease,³⁸  a greater proportion of patients achieving the target international normalized ratio before cardiac surgery,³⁹  and less cumulative postoperative blood loss.^38,42  Nonsignificant decreases in blood product use (e.g., red blood cells, plasma, platelets, and cryoprecipitate), patients requiring reoperation for bleeding, and in-hospital mortality were also seen in patients undergoing heart transplantation treated with four-factor prothrombin complex concentrate compared with a nonfactor concentrate historical control group.⁶⁷ 

In a retrospective analysis of patients undergoing orthotopic heart transplantation, significantly fewer units of cryoprecipitate and packed erythrocytes were transfused in patients who received four-factor prothrombin complex concentrate than in those who did not (P < 0.001).⁶⁸  Furthermore, the median time to chest closure was significantly shorter in patients receiving four-factor prothrombin complex concentrate (547.9 min) versus those who did not (618.8 min; P = 0.008).⁶⁸  No significant difference in in-hospital mortality was observed.⁶⁸ 

Patients who receive vitamin K antagonist therapy and present with trauma represent a challenging medical emergency. In the United States, the proportion of trauma patients who take warfarin has been shown to be approximately 4%, which increases to almost 13% when considering patients older than 65 yr.⁶⁹  Furthermore, anticoagulant use before trauma has been associated with an increased risk of mortality, even when adjusting for confounders such as age and preexisting medical conditions.^69,70 

While rapid reversal of the anticoagulant effect is essential in any vitamin K antagonist–treated patient suffering a traumatic injury, simply replenishing the vitamin K–dependent coagulation factors does not provide volume replacement, which is often required in patients who are in hypovolemic shock after major blood loss. For patients with a suspected massive bleed, current European guidelines recommend transfusion of fresh frozen plasma (or pathogen-inactivated plasma) in conjunction with packed erythrocytes in a plasma–erythrocyte ratio of at least 1:2.⁷¹  The recent Pragmatic, Randomized Optimal Platelet and Plasma Ratios (PROPPR) trial investigated the effectiveness and safety of a 1:1:1 plasma, platelet, erythrocyte transfusion ratio compared with a 1:1:2 ratio.⁷²  No significant differences were seen in overall mortality after 24 h or 30 days; however, the 1:1:1 transfusion ratio was associated with a significantly greater proportion of patients achieving hemostasis (86.1% vs. 78.1%; P = 0.006) and significantly fewer patients dying from exsanguination (9.2% vs. 14.6%; P = 0.03).⁷² 

However, the administration of large amounts of fresh frozen plasma and erythrocytes to restore blood volume can lead to a dilution of the coagulation factors, which delays coagulopathy reversal.⁷¹  As such, the administration of prothrombin complex concentrate in conjunction with fresh frozen plasma has been proposed as an alternative therapeutic option for the rapid correction of traumatic coagulopathy while also restoring volume and should be considered in a vitamin K antagonist–treated patient. While studies have also investigated three-factor prothrombin complex concentrate for the treatment of traumatic coagulopathy,^73,74  four-factor prothrombin complex concentrate is recommended by current guidelines for the reversal of vitamin K antagonist–related anticoagulation,^5,6,9,10  and a number of studies have investigated the use of four-factor prothrombin complex concentrate to reverse the anticoagulant effect in vitamin K antagonist–treated trauma patients.^{47,60,75–77}  In a retrospective study of four-factor prothrombin complex concentrate administered to 26 trauma patients on warfarin, the mean international normalized ratio was shown to significantly decrease from 5.7 to 1.5 (P < 0.001), and this decrease was sustained for more than 2 days.⁷⁵  No patients developed venous thromboembolic events, and no in-hospital mortality was reported.⁷⁵  A prospective study investigated four-factor prothrombin complex concentrate treatment for warfarin-associated coagulopathy after traumatic brain injury.⁶⁰  In five patients treated with four-factor prothrombin complex concentrate, the international normalized ratio was corrected to 1.2 or less from a baseline of more than 2.0 in all patients; for patients requiring surgery, the time to anesthesia induction was 159 min, which compared favorably to patients who received fresh frozen plasma (307 min).⁶⁰  Administration of four-factor prothrombin complex concentrate has also been shown to result in significantly lower transfusion requirements of erythrocyte and platelet concentrate units (P < 0.001), as well as fewer trauma patients requiring transfusion compared with patients receiving fresh frozen plasma.⁷⁶ 

When comparing with prothrombin complex concentrates, four-factor prothrombin complex concentrate has been shown to result in a significantly lower international normalized ratio (1.3 vs. 1.6; P < 0.001) and a significantly greater proportion of trauma patients achieving successful reversal of anticoagulation (83% vs. 50%; P = 0.022).⁴⁷  Three-factor prothrombin complex concentrate was also associated with a greater number of venous thromboembolic events in patients, compared with four-factor prothrombin complex concentrate (15% vs. 0%), although this difference did not reach statistical significance.⁴⁷  In a retrospective analysis of warfarin-treated trauma patients comparing four-factor prothrombin complex concentrate with three-factor prothrombin complex concentrate plus recombinant FVIIa, the combination therapy of a three-factor prothrombin complex concentrate and recombinant FVIIa achieved a significantly lower international normalized ratio than did four-factor prothrombin complex concentrate (0.75 vs. 1.28; P < 0.001); however, no difference was seen between treatments for patients achieving a target international normalized ratio less than 1.5.⁷⁷  Furthermore, the combination therapy was associated with a significantly increased risk of deep vein thrombosis development compared with the four-factor prothrombin complex concentrate group (22.6% vs. 2.9%; P = 0.01).

These studies demonstrate that prothrombin complex concentrate results in a rapid reversal of the coagulopathy, as measured by the international normalized ratio. However, as stated earlier, it is important to remember that because of its concentrated nature, prothrombin complex concentrate does not provide the volume support that can be required to correct hypoperfusion associated with major blood loss, and, therefore, administration of plasma is still recommended in this patient population.⁷¹ 

Historically, the use of prothrombin complex concentrates has been associated with a potential increase in venous thromboembolic events, possibly because activated coagulation factors were included in the previous formulations of prothrombin complex concentrates⁷⁸  but also because patients on anticoagulants are treated for hypercoagulable disorders. Current formulations use nonactivated clotting factors and include antithrombotic components (protein C and S), which may mitigate the risk of developing venous thromboembolic events.⁷⁹  In a meta-analysis of 27 studies investigating prothrombin complex concentrate therapy for vitamin K antagonist–treated patients in various settings, the overall risk of venous or arterial venous thromboembolic events was only 1.4%, which decreased to 0.8% in the subset of patients undergoing a surgical procedure.⁸⁰ 

In the studies identified in our search, rates of venous thromboembolic events in patients receiving prothrombin complex concentrates varied considerably, from 0 to 26.3%.^{19,22,23,25,31,37,41,64,81–83}  We also identified two case studies that each reported a patient undergoing surgery who developed a venous thromboembolic event within 1 h after three-factor prothrombin complex concentrate administration.^35,84  It should be noted that many of these studies included few patients, and the patient populations investigated often had a number of comorbidities; moreover, once vitamin K antagonist therapy is reversed, the underlying risk that first necessitated anticoagulation is restored, and as a result, caution should be taken when interpreting these findings.

In a comparative study of patients requiring vitamin K antagonist reversal before heart transplantation, venous thromboembolic events were reported more frequently in patients receiving three-factor prothrombin complex concentrate than with a historical cohort who received vitamin K and plasma (18.7% vs. 10%, respectively), although this difference was not significant.⁶⁷  In another comparative study of four-factor prothrombin complex concentrate and plasma in patients undergoing emergency surgery, no significant difference was noted in the percentage of patients with venous thromboembolic events, with 7% and 8% of patients who received prothrombin complex concentrate and plasma, respectively, experiencing a venous thromboembolic event,⁴⁰  which is in line with a recent meta-analysis demonstrating no increase in risk of venous thromboembolic events with prothrombin complex concentrates compared with plasma.¹³  However, none of the studies were designed to compare the incidence of thromboembolic events between prothrombin complex concentrates and plasma.

Across four studies, no significant differences in overall adverse event rates were seen in patients who received prothrombin complex concentrate compared with patients who received plasma.^38–40,42  One study reported similar rates of late bleeding events in four-factor prothrombin complex concentrate–treated patients and plasma-treated patients,⁴⁰  whereas another study reported abnormal bleeding in two patients who received plasma but none in those treated with four-factor prothrombin complex concentrate.³⁸  These abnormal bleeding events were likely linked to plasma’s lower effectiveness at reducing patients’ international normalized ratio.³⁸ 

Because rapid infusion of prothrombin complex concentrates have potential safety concerns, a multinational trial evaluated 43 patients given prothrombin complex concentrates for emergency warfarin reversal to evaluate the effect of the infusion rate on international normalized ratio correction and thrombogenicity.⁸⁵  The infusion speed ranged from 2.0 to 40.0 ml/min (median of 7.5 ml/min). The investigators noted that the speed of infusion did not affect the international normalized ratio measured at 30 min after prothrombin complex concentrate completion, and measured thrombogenicity parameters were not affected by infusion speed.⁸⁵  Currently, recommendations for four-component prothrombin complex concentrate administration is reconstitution in 20 ml, and the solution should be administered intravenously (not more than 3 U · kg⁻¹ · min⁻¹, maximum 210 U/min, approximately 8 ml/min).

Owing to the increased volumes administered with plasma compared with prothrombin complex concentrate, there is a greater risk of fluid overload in patients treated with plasma.⁸⁶  In the study by Goldstein et al., fluid overload or similar cardiac events were reported in 3% of patients who received four-factor prothrombin complex concentrate, compared with 13% of patients who received plasma.⁴⁰  In another study, one patient who received plasma experienced a significant increase in pulmonary and/or atrial pressure after plasma administration, which is indicative of fluid overload; no patients treated with four-factor prothrombin complex concentrate demonstrated fluid overload events.³⁸  Taken together, these safety findings are in line with those reported in a recent meta-analysis of warfarin-treated patients who required urgent reversal owing to major bleeding or urgent surgical intervention: no significant difference between prothrombin complex concentrate and plasma was seen in relation to thromboembolic risk, and fluid overload was less likely in patients treated with prothrombin complex concentrate than in patients treated with plasma.¹³  In summary, large volumes of plasma are required to reverse vitamin K antagonists; however, they ineffectively increase the concentration of coagulation factors, expose patients to allogeneic blood products with all the inherent risks, and should not be recommended or used for vitamin K antagonist reversal as also recommended in guidelines.

In contrast with vitamin K antagonists, the non–vitamin K oral anticoagulants specifically inhibit either coagulation factors IIa or Xa, and unlike vitamin K antagonists, have few drug–drug interactions.⁸⁷  However, as with vitamin K antagonists, increased bleeding risk remains a concern with non–vitamin K oral anticoagulants.⁸⁸  In cases of emergency surgical intervention, there are currently no approved specific reversal agents for factor Xa inhibitors, although andexanet alfa, a recombinant factor Xa decoy receptor protein, was approved in May 2018 for patients treated with rivaroxaban and apixaban, when reversal of anticoagulation is needed because of life-threatening or uncontrolled bleeding. The use of specific reversal strategies for non–vitamin K oral anticoagulants, also called antidotes, is an evolving strategy for treating bleeding with these agents.⁸⁹  However, andexanet has not been studied in surgical patients and will be available initially in a limited number of medical centers; its role for perioperative use remains to be determined.

Based on preclinical evidence and recent reports, current guidelines suggest that prothrombin complex concentrates could be used as part of a multimodal approach in patients requiring urgent surgery or experiencing life-threatening bleeding.^10,90–93  Infusion of four-factor prothrombin complex concentrate has been shown to reduce prothrombin time and/or increase endogenous thrombin potential in studies of healthy volunteers or patients who received apixaban, edoxaban, or rivaroxaban.^94–98  Furthermore, infusion of four-factor prothrombin complex concentrate after edoxaban administration demonstrated a dose-dependent effect on reducing bleeding duration and volume within 30 min, with a dose of 50 U/kg decreasing bleeding duration and volume below baseline levels in patients receiving therapeutic doses.⁹⁸ 

An increasing amount of clinical data on prothrombin complex concentrate use for treatment of acute major bleeding associated with factor Xa anticoagulation is emerging from large patient registries and observational studies. Data from a large prospective registry of patients receiving non–vitamin K oral anticoagulants, the Dresden registry,⁹⁹  demonstrated the rates, management, and outcome of rivaroxaban-related bleeding. Of 1,776 patients, 66 patients experienced a major bleeding event and six patients received prothrombin complex concentrates (dose range, 18 to 47 U/kg). Only one patient had a significant improvement in coagulation parameters (international normalized ratio, prothrombin time ratio, and activated partial thromboplastin time); however, five of the six patients demonstrated hemorrhage stabilization.⁹⁹  In a retrospective review of patients developing hemorrhage secondary to dabigatran or rivaroxaban therapy, a median dose of 40 U/kg prothrombin complex concentrate was administered in 3 of 25 patients.¹⁰⁰  All three patients had rivaroxaban-associated bleeds (one major, two life threatening), and administration of prothrombin complex concentrate successfully resolved the bleeding in all cases.¹⁰⁰  With regards to the perioperative setting, a retrospective, multicenter study investigated patients who received four-factor prothrombin complex concentrate for treatment of the anticoagulation effects of factor Xa inhibitors when developing a pericardial effusion during or after atrial fibrillation ablation.¹⁰¹  In total, 11 patients were administered four-factor prothrombin complex concentrate. Two patients required further surgery for treatment of the pericardial effusion, and the other nine patients were hemodynamically stable and there was no recurrence of the pericardial effusion, demonstrating that four-factor prothrombin complex concentrate is an effective management option in this patient population.¹⁰¹ 

There have also been a few case reports of patients on factor Xa inhibitors (either apixaban or rivaroxaban) being treated with prothrombin complex concentrate before undergoing a surgical procedure.^102–104  Overall, administration of prothrombin complex concentrate was associated with successful completion of surgery and no bleeding complications were reported.^102–104 

A recent prospective evaluation reported 84 patients receiving rivaroxaban or apixaban who were treated with prothrombin complex concentrates for major bleeding and evaluated for thromboembolic events and all-cause mortality within 30 days.¹⁰⁵  Prothrombin complex concentrates were administered at a median dose of 2,000 U (1,500 to 2,000 U) for patients with an intracranial hemorrhage (n = 59; 70.2%) or gastrointestinal bleeding (n = 13; 15.5%). Treatment to stop bleeding was considered effective in 58 (69.1%) and ineffective in 26 (30.9%) treated patients. The majority of the patients with ineffective hemostasis had intracranial hemorrhage (n = 16; 61.5%), and two patients developed an ischemic stroke 5 and 10 days after prothrombin complex concentrate administration. A total of 27 (32%) patients died within 30 days; however, there was no control group in the report.¹⁰⁵ 

An additional report from Canada evaluated major bleeding in 66 apixaban- or rivaroxaban-treated patients treated with 2,000 units of prothrombin complex concentrates and evaluated thromboembolism or mortality 30 days later.¹⁰⁶  Using a specific evaluation scale, the investigators reported cessation of bleeding was good in 65%, moderate in 20%, and poor or none in 15% of patients and included patients with intracranial hemorrhage or gastrointestinal bleeding. Overall reversal was considered to be effective in 68% of patients and ineffective in 32%, and mortality was 14% in 30 days, with an 8% risk of thromboembolic events.¹⁰⁶ 

Overall, the studies identified in this review support current guideline recommendations that four-factor prothrombin complex concentrate is a preferred treatment option for urgent reversal of vitamin K antagonist anticoagulation in patients requiring urgent surgical or invasive procedures. Prothrombin complex concentrates consistently and rapidly reduced patients’ international normalized ratio. Comparative studies with plasma demonstrated greater clinical efficacy with prothrombin complex concentrates in patients requiring emergency surgery. Furthermore, prothrombin complex concentrate treatment was associated with lower rates of fluid overload owing to its lower infusion volume compared to plasma and no instances of viral transmission. Prothrombin complex concentrates are recommended in guidelines for rapid reversal of anticoagulation in vitamin K antagonist–treated patients and represent an important therapeutic option for emergency surgical interventions.

The authors had full editorial control and no additional financial support was received. Writing assistance was also provided with a grant from CSL Behring (Marburg, Germany).

Dr. Levy serves on research and advisory committees for Boehringer-Ingelheim (Ingelheim am Rhein, Germany), CSL Behring (Marburg, Germany), Grifols (Barcelona, Spain), Instrumentation Laboratories (Bedford, Massachusetts), Janssen (Beerse, Belgium), Merck (Kenilworth, New Jersey), Octapharma (Lachen, Switzerland), and Portola (South San Francisco, California). Dr. Steiner has received a research grant form Octapharma, consultancy fees from Bayer (Leverkusen, Germany), BMS-Pfizer (New York, New York), Boehringer-Ingelheim, and Daiichi Sankyo (Tokyo, Japan), and speakers’ honoraria from Bayer, BMS-Pfizer, Boehringer-Ingelheim, and Daiichi Sankyo, and holds shares from Novo Nordisk (Bagsværd, Denmark). Dr. Goldstein has received research funding from Boehringer-Ingelheim, Pfizer (Groton, Connecticut), and Portola. Dr. Milling serves on research steering committees for Population Health Research Institute (Hamilton, Canada), Portola, Boehringer-Ingelheim, and CSL Behring, and a speakers’ bureau for Janssen. He has also received funding from the Seton Dell Medical School Stroke Institute (Austin, Texas) as a principal investigator; grant No. NHLBI K23 1K23HL127227-01A1. Dr. Douketis declares no competing interests.