Safety and effectiveness of a four-factor prothrombin complex concentrate for vitamin K antagonist reversal following a fixed-dose strategy

Abstract

Objectives

Early reversal of anticoagulation improves outcomes in major bleeding and emergency surgery. To reverse vitamin K antagonists (VKA), vitamin K in addition to prothrombin complex concentrate (PCC) is recommended. Dosing recommendations for VKA reversal provided by the manufacturer are 25–50 IU/kg depending on the baseline international normalised ratio (INR). Nevertheless, we recommend an initial fixed dose of 1000 IU, and additional 500 IU doses evaluated on a case-by-case basis. As there is a paucity of clinical data demonstrating the efficacy and safety of this strategy, we designed this study to assess the effectiveness and safety of a four-factor (4F)-PCC for VKA reversal following a fixed-dose strategy.

Methods

This was a retrospective study of adult patients who received 4F-PCC for VKA reversal. The primary outcome was INR correction. INR correction was achieved if the first INR draw after 4F-PCC was ≤1.5. Safety outcome was any confirmed thromboembolic event within 3 months after 4F-PCC. Secondary outcomes included activated partial thromboplastin time (aPTT) correction, as well as haemostatic effectiveness for bleeding patients.

Results

A total of 145 patients were included: 106 (73.1%) in the bleeding group and 39 (26.9%) in the emergency surgery group. The INR target was reached in 102 (70.3%) patients (p<0.0001). In one case, a thromboembolic complication was possibly related to 4F-PCC. The aPTT ratio target was reached in 113 (77.9%) patients (p<0.0001), and 79 of the 106 (74.5%) patients reversed for bleeding achieved haemostatic effectiveness.

Conclusions

After 4F-PCC, the majority of patients achieved the target INR, meaning 4F-PCC is a useful modality for rapid INR reduction. The safety profile may be considered acceptable. Fixed-dose 4F-PCC was able to restore haemostasis rapidly while minimising the risk of adverse events and optimising available resources.

Introduction

Major bleeding in emergency medical services can be a significant problem frequently related to oral anticoagulation treatment (OAT). Indeed, major bleeding during OAT is associated with higher morbidity and mortality.¹ Therefore, early reversal of anticoagulation is essential to improve outcomes in major bleeding, while any emergency surgery or invasive procedure under OAT may also require anticoagulation reversal.²

Oral anticoagulants available include vitamin K antagonists (VKA). To reverse VKA activity, intravenous vitamin K in addition to prothrombin complex concentrate (PCC), or fresh frozen plasma (FFP) if PCC is not available, is recommended.³ International guidelines recommend the use of PCC over FFP because PCC is available for rapid intravenous administration in small volumes and can be promptly dispensed as there is no need for thawing or blood-type matching. Furthermore, clinical studies support PCC as a better option than FFP to provide fast reversal of the international normalised ratio (INR).⁴

PCC is referred to as three-factor (3F)-PCC (containing factors II, IX and X) or four-factor (4F)-PCC (also containing factor VII). Our study focused on the use of 4F-PCC for anticoagulation reversal. To assess the effectiveness of 4F-PCC for anticoagulation reversal, clinical outcomes may be the most relevant measures. However, patients in need of acute reversal therapy have diverse disease profiles, which make clinical outcomes difficult to standardise. Thus, surrogate measures, such as INR, are often used to evaluate treatment results. Regarding the 4F-PCC safety profile, the literature considers it may be tolerable, with thromboembolic complications as the most serious adverse events reported. Some cohort studies have reported an incidence of 0–2%, even though most published case series were too small to estimate the risk for thromboembolism.^{5 6}

4F-PCC is approved in Spain under the proprietary names Prothromplex, Octaplex, and Beriplex. At the time of the study, only Octaplex was dispensed in the hospital. Its main indication is VKA reversal in patients with acute major bleeds and in those requiring urgent surgeries or invasive procedures. 4F-PCC has also been studied for the reversal of the direct-acting oral anticoagulants^{7 8} and to treat patients with coagulopathy bleeding,⁹ both off-label uses.

Dosing recommendations for VKA reversal provided by the manufacturer are 25–50 IU/kg, depending on the baseline INR.¹⁰ Our institutional protocol recommends 4F-PCC for VKA reversal in adult patients with major bleeding or who require emergency surgery. Nevertheless, for the reversal of VKA, we recommend an initial fixed dose of 1000 IU, and additional 500 IU doses evaluated on a case-by-case basis depending on INR reversal. If administering 4F-PCC for off-label indications, we use doses similar to those of VKA reversal.

In December 2017, the American College of Cardiology (ACC) expert consensus decision pathway on management of bleeding in patients on oral anticoagulants included a fixed-dose 4F-PCC strategy for urgent reversal of VKA.¹¹ Dosing strategies have previously demonstrated positive outcomes, such as low rates of thromboembolic events,¹² and a fixed-dose strategy adds other benefits. The first benefit is an improved time to drug administration in a patient population where time is critical to outcomes. A recent study noted a decrease in medication administration times, from 51 to 38 min, following a fixed-dose protocol.¹³ The second benefit is the decreased cost of pharmaceutical goods. In 2018, Astrup et al reported US$36.365 as potential savings using a fixed-dose strategy over an 8 month study period with 37 patients included in the analysis.¹³ However, a paucity of clinical data exist demonstrating the efficacy and safety of this strategy. The aim of this study was to assess the effectiveness and safety of 4F-PCC for VKA reversal following a fixed-dose strategy.

Methods

Study design and population

This retrospective study was carried out at La Paz University Hospital (1300 beds), which is a tertiary-care institution.

Data were collected for 17 months (November 2015 to March 2017). At the time of the study, one 4F-PCC (Octaplex) was dispensed in the hospital, which was available 24 hours a day. This 4F-PCC was a four-factor concentrate (II, VII, IX, and X) with 25 IU/mL factor IX. Other constituents of Octaplex, according to product labelling,¹⁰ were factor II 11–38 IU/mL, factor VII 9–24 IU/mL, factor X 18–30 IU/mL, and antithrombotic content such as protein C 7–31 IU/mL and protein S 7–32 IU/mL. An initial fixed-dose of 1000 IU was used, and additional 500 IU doses were administered based on INR reversal (figure 1).

Figure 1.

VKA reversal algorithm. 4F-PCC, four-factor prothrombin complex concentrate; INR, international normalised ratio; VKA, vitamin K antagonist.

All adult patients who received 4F-PCC in the emergency department (ED) for emergency reversal of VKA anticoagulation, either due to bleeding, emergency surgery or invasive procedures, were included. No exclusion criteria were applied. All patients included had haematology and coagulation control within 24 hours.

The ethics committee of the medical centre approved the study, and, given its retrospective nature, informed consent was waived.

Data extraction

Data were extracted from each patient’s electronic medical record. Data collected included demographic information, medical history, laboratory values, medication administration records and daily physician progress notes. A retrospective analysis of all patients was carried out for 3 months after administration of 4F-PCC.

Endpoints

The primary effectiveness outcome was INR correction. It was described as INR ≤1.5 in the first INR draw after 4F-PCC administration.

Secondary outcomes included activated partial thromboplastin time (aPTT) correction and haemostatic effectiveness for bleeding patients:

• aPTT correction was achieved if the first aPTT ratio after 4F-PCC administration was ≤1.2

• Haemostatic effectiveness was achieved if a decrease in haemoglobin levels <20% was found within the first 24 hours.

Safety evaluations were performed for all patients included in the study. The primary safety outcome was any confirmed thromboembolic event during the 3 month follow-up.

Information on gender, patient baseline and dose range were collected to determine potential correlation with the effectiveness of 4F-PCC.

Statistical analysis

Data are expressed as mean±SD or median (IQR) for continuous variables. For categorical data, the frequencies and percentages are presented. The continuous variables were analysed using the Student t-test for parametric data and the Mann-Whitney U test for non-parametric data. Categorical data were analysed using either the χ² test or Fisher’s exact test. The statistical analysis was carried out using SPSS 22.0 statistical software (SPSS Inc, Chicago, IL, USA). A p value of 0.05 was considered statistically significant.

Results

Patients and treatment

One hundred and forty-five patients received 4F-PCC during the 17 month period. The patients’ demographics are shown in table 1.

Table 1.

Patients’ demographics

	All patients (n=145)	Bleeding (n=106)	Surgery (n=39)
Age, years	84 (76 to 88)	84 (78 to 89)	84 (76 to 88)
Sex, male	74 (51%)	50 (47.2%)	24 (61.5%)
Weight, kg	73 (64 to 80)	71 (60 to 80)	76 (69 to 80)
Dose of 4F-PCC, IU/kg	16.3 (13 to 20)	16.9 (13 to 21)	15.4 (13 to 19)
Dose of 4F-PCC
500 IU	13 (9%)	10 (9.4%)	3 (7.8%)
1000 IU	74 (51%)	53 (50%)	21 (53.8%)
1500 IU	48 (33%)	35 (33%)	13 (33.3%)
2000 IU	6 (4%)	4 (3.8%)	2 (5.1%)
>2000 IU	4 (3%)	4 (3.8%)	–
Baseline INR	2.8 (2.1 to 3.6)	2.9 (2.2 to 3.9)	2.6 (2 to 3.3)
Baseline aPTT ratio	1.4 (1.2 to 1.6)	1.4 (1.2 to 1.7)	1.4 (1.2 to 1.6)
Bleeding patients
Intracranial haemorrhage		42 (39.6%)
Gastrointestinal haemorrhage		48 (45.3%)
Other haemorrhage		16 (15.1%)
Vitamin K administration
Yes	129 (89%)	95 (89.6%)	33 (84.6%)
No	16 (11%)	11 (10.4%)	6 (15.4%)

All numbers expressed as n (%) or median (IQR).

aPTT, activated partial thromboplastin time; 4F-PCC, four factor-prothrombin complex concentrate; INR, international normalised ratio.

One hundred and six patients (73.1%) were reversed for acute bleeding and 39 (26.9%) for emergency surgery or invasive procedure. The most common types of bleed were gastrointestinal haemorrhage (GIH) (45.3%) and intracranial haemorrhage (ICH) (39.6%).

The majority of the patients (60%) received a single 4F-PCC dose; 74 patients (51%) received 1000 IU and 13 (9%) 500 IU. Forty-eight patients (33%) needed an additional 500 IU dose and only 10 patients (7%) needed two or more 500 IU additional doses.

The median (IQR) dose of 4F-PCC, adjusted for body weight, was 16.3 IU/kg (13–20): 16.9 IU/kg (13–21) in the bleeding group and 15.4 IU/kg (13–19) in the emergency surgery group. The majority of the patients (89%) received concomitant vitamin K administration.

Effectiveness

VKA reversal

INR correction

All 145 patients had pre- and post-administration INR values. The median (IQR) baseline INR was 2.8 (2.1–3.6). One hundred and two patients (70.3%) achieved an INR ≤1.5 (p<0.0001) within 24 hours after 4F-PCC administration (table 2).

Table 2.

Correction of INR and aPTT ratio after 4F-PCC administration

	All patients (n=145)	Bleeding (n=106)	Surgery (n=39)
Achieved INR ≤1.5	102 (70.3%)	78 (73.6%)	24 (61.5%)
Achieved aPTT ratio ≤1.2	113 (77.9%)	91 (85.8%)	22 (56.4%)

All data collected within 24 hours post 4F-PCC. All numbers expressed as n (%).

aPTT, activated partial thromboplastin time; 4F-PCC, four factor-prothrombin complex concentrate; INR, international normalised ratio.

Data regarding gender, patient baseline and dose range as independent variables in all 145 patients are presented in table 3.

Table 3.

Subgroup analysis

	Achieved INR ≤1.5	Did not achieve INR ≤1.5
Gender
Male (n=74)	49 (66.2%)	25 (33.8%)
Female (n=71)	53 (74.6%)	18 (25.4%)
Patient baseline
Bleeding (n=106)	78 (73.6%)	28 (264%)
Intracranial haemorrhage (n=42)	36 (85.7%)	6 (14.3%)
Gastrointestinal haemorrhage (n=48)	32 (66.7%)	16 (33.3%)
Other haemorrhage (n=16)	10 (62.5%)	6 (37.5%)
Surgery (n=39)	24 (61.5%)	15 (38.5%)
Dose of 4F-PCC, IU/ kg
<15 (n=63)	42 (66.7%)	21 (33.3%)
15–22.5 (n=59)	44 (74.6%)	15 (25.4%)
22.5–32.5 (n=15)	10 (66.7%)	5 (33.3%)
>32.5 (n=8)	6 (75%)	2 (25%)

All numbers expressed as n (%).

4F-PCC, four factor-prothrombin complex concentrate; INR, international normalised ratio.

Gender

Fifty-three of the 71 women (74.6%) achieved an INR ≤1.5 in contrast to 49 of the 74 men (66.2%) with a median (IQR) 4F-PCC dose of 15.9 IU/kg (13.3–19.4) and 16.4 IU/kg (12.7–20.3), respectively.

Patient baseline

Seventy-eight of the 106 (73.6%) bleeding patients achieved an INR ≤1.5 in contrast to 24 of the 39 (61.5%) surgery patients. This could be due to the higher doses administered in the bleeding group, as more than one administration was allowed if the INR was not corrected. Figure 2A shows patients who achieved an INR ≤1.5 and the degree of INR correction after 4F-PCC depending on the patient baseline.

Figure 2.

Patient baseline and INR values (A) and 4F-PCC dose and INR values (B). 4F-PCC, four-factor prothrombin complex concentrate; GIH, gastrointestinal haemorrhage; ICH, intracranial haemorrhage; INR, international normalised ratio; INRc, control INR; INRi, initial INR; OH, other haemorrhages; S, surgery.

Dose range

Different dose ranges showed similar rates of INR correction (table 3). Figure 2B shows patients who achieved an INR ≤1.5 and the degree of INR correction after 4F-PCC depending on the dose range. Even though rates of INR correction were similar between dose ranges, it should be noted that more dose was used in those patients with higher levels of initial INR. Doses under 22.5 IU/kg were administered in the majority of the cases. These patients had an initial median (IQR) INR of 2.7 (2–3.6). Patients who received >32.5 IU/kg of 4F-PCC had an initial median (IQR) INR of 4.9 (2.4–6.1).

A univariate comparison of independent variables was performed between the 102 patients who achieved INR ≤1.5 and the 43 who did not, but this comparison did not reach statistical significance (p>0.05).

aPTT correction

All 145 patients had pre- and post-administration aPTT ratio values. The median (IQR) baseline aPTT ratio was 1.4 (1.2–1.6). One hundred and thirteen patients (77.9%) achieved an aPTT ratio ≤1.2 (p<0.0001) within 24 hours after 4F-PCC administration (table 2).

Haemostatic effectiveness

One hundred and six bleeding patients were evaluated, from whom 79 (74.5%) achieved haemostatic effectiveness. Patients who achieved this outcome were: 23 of the 42 ICH (54.8%), 44 of the 48 GIH (91.7%) and 12 of the 16 other haemorrhages (75%).

Adverse events

Safety evaluations were performed for all patients included in the study within 3 months after the administration of 4F-PCC. In one case, the occurrence of thromboembolic complications was possibly related to 4F-PCC (table 4). The patient had received 4F-PCC 5 days before the thromboembolic event to reverse VKA anticoagulation in the context of ICH. The dose of 4F-PCC administered was 14 IU/kg and a correction of 1.2 units in INR, with control INR 1.3, was achieved. On the fifth day after receiving 4F-PCC the patient attended the ED with chest pain, where a pulmonary embolism was diagnosed and heparin anticoagulation was initiated.

Table 4.

Safety outcomes

	All patients (n=145)	Bleeding (n=106)	Surgery (n=39)
Thromboembolism within 3 months	1 (0.7%)	1 (0.9%)
Pulmonary embolism	1 (100%)	1 (100%)
Emergency department mortality	10 (6.9%)	7 (6.6%)	3 (7.7%)
Traumatic brain injury	4 (40%)	4 (57.1%)
Sepsis	4 (40%)	2 (28.6%)	2 (66.7%)
Gastrointestinal haemorrhage	1 (10%)	1 (14.3%)
Cardiorespiratory arrest	1 (10%)		1 (33.3%)

All numbers expressed as n (%).

Ten patients (6.9%) died during the ED episode after treatment with 4F-PCC, seven (6.6%) in the bleeding group and three (7.7%) in the emergency surgery group. Death occurred due to a direct effect of bleeding in five of the seven patients (table 4) and none was related to 4F-PCC administration.

Discussion

We report results from a large cohort of patients treated with 4F-PCC for major bleeding or emergency surgery following a fixed-dose strategy. Our main finding is the high effectiveness of 4F-PCC for VKA reversal in this setting and the low risk of thromboembolic events.

In patients with life-threatening bleeding, guidelines recommend the use of 4F-PCC over other alternatives.¹⁴ The literature shows that two-thirds of cases of 4F-PCC use in the ED are related to major haemorrhage, mainly ICH or GIH. Indeed, the most frequent major complications in patients receiving OAT are gastrointestinal (30–60%) or intracranial (17–30%).^{15 16} In our population, 73.1% received 4F-PCC for major bleeding, mainly for GIH (33.1%) and ICH (29%).

Following a fixed-dose regimen, 84% of patients received 1000–1500 IU of 4F-PCC. Doses administered were lower than suggested by the package insert,¹⁰ with a median (IQR) dose, adjusted for body weight, of 16.3 IU/kg (13–20). The fixed-dose regimen used in our study is supported by the Canadian Advisory Committee on Blood and Blood Products, which recommends a dose of 1000 IU of 4F-PCC, with vitamin K administration, for VKA reversal.¹⁷ The majority of our patients received concomitant vitamin K. Indeed, concomitant vitamin K should be given if a sustained INR reduction is desired when administering 4F-PCC for VKA reversal. This recommendation is included in the American College of Chest Physicians guidelines to prevent the incidence of INR rebound.¹⁸

The majority of patients achieved INR correction. A definition of INR ≤1.5 at the first administration draw was used because different clinical guidelines support INR ≤1.5 as a surrogate measure of anticoagulation reversal.^{19 20} The achievement of the INR target was reached by 70.3% of patients, confirming published results that show that anticoagulation reversal is effective in 60–80% of patients.^{21 22} Indeed, in their review, Tornkvist et al confirmed that elevated INR values due to VKA treatment could be reversed (INR ≤1.5) in 63.1% of study subjects after treatment with 4F-PCC.²

Different dose ranges, adjusted for body weight, showed similar rates of INR correction. Even though it should be noted that more dose per kg was used in those patients with higher initial INR, these doses were always lower than those recommended by the manufacturer. Indeed, the majority of the patients had an initial INR between 2–3 and received doses under 22.5 IU/kg, instead of the 22.5–40 IU/kg recommended.¹⁰ The literature also differs from manufacturer recommendations²³ and the ACC recommends 25 IU/kg of 4F-PCC for VKA reversal in bleeding patients with baseline INR between 2–4.¹¹ Our data support the use of fixed doses as a strategy to optimise efficacy. Our results are supported by recent retrospective studies that have evaluated doses ranging from 1000 to 1500 IU that appear to successfully reverse patients’ INR to <1.5.^{13 24}

In regards to the secondary outcomes, the majority (77.9%) achieved aPTT ratio correction. We used a definition of aPTT ≤1.2 at the first administration draw, as the literature supports an aPTT ratio ≤1.2 as the normal value.²⁵ The aPTT ratio has been assessed in studies evaluating VKA reversal together with INR because both parameters in conjunction help to quantify the severity of over-anticoagulation.^{26 27} However, to our knowledge, there is no published data concerning the aPTT ratio as a surrogate measure of the efficacy of 4F-PCC for VKA reversal.

Vitamin K is essential for the formation of factors II, VII, IX and X in the coagulation cascade. Factor VII participates in the extrinsic pathway, factor IX in the intrinsic pathway, and factors II and X in the common pathway. INR, the most commonly used assay to monitor VKA treatment and reversal of anticoagulation by 4F-PCC, measures the extrinsic pathway and it is very sensitive to factor VII variations, while the aPTT ratio measures the intrinsic pathway and is sensitive to factor IX variations. As factor IX is a component of 4F-PCC, we believe aPTT ratio inclusion as a secondary objective strengthens our results.

The rate of haemostatic effectiveness in our bleeding patients was 74.5%, which is supported by the 72.4% rate reported by Sarode et al in their prospective trial.²⁸ Our definition of haemostatic effectiveness was different from the one used by Sarode et al due to the retrospective nature of our study. Nevertheless, in a previously published retrospective study, Sin et al reported 73.6% as the rate of bleeding control, which supports our results.²⁹ Surgery patients were not evaluable for haemostatic effectiveness as a prospective follow-up would have been required.

This study reported a 0.7% incidence of thromboembolism after 4F-PCC administration, which is numerically lower than the reported rates of two large randomised controlled trials (RCTs) with values of 7.8% and 6.8%.^{28 30} We chose a 90 day follow-up period to look for thromboembolism, and the thromboembolic event was detected 5 days after the 4F-PCC administration. Thus, the low incidence of thromboembolism may not be attributed to the follow-up period. Moreover, as OAT was discontinued after the ICH episode, thromboembolism might have occurred because of underlying risks irrespective of the means used to reverse VKA.

Tornkvist et al, in their review, reported a 1.6% incidence of thromboembolism in VKA patients treated with 4F-PCC,² which supports our data. The low incidence of thromboembolic events in our population may be attributed to the 4F-PCC dose used. Goldstein et al and Sarode et al in their RCTs used 25 IU/kg of 4F-PCC with baseline INR between 2–4, in contrast to our study where a median (IQR) dose of 16.3 IU/kg (13–20) was administered.^{28 30} Likely, there may have been non-reported thromboembolic events, since not all patients may have attended the ED at La Paz Hospital when suffering a post-4F-PCC thromboembolic episode. Also, there may have been an underestimation of 4F-PCC association with thromboembolic risk, since the majority of patients probably restarted OAT within the 90 day period.

The high number of patients, the inclusion of patients with active bleeding as well as non-bleeding patients in need of emergency surgery, and the 3 month follow-up, render clinical relevance and strength to this study. However, the retrospective nature of the study results in limitations, most notably the lack of a control group and the use of surrogate measures to evaluate 4F-PCC effectiveness. Other limitations were the patients’ variability, as all patients who received 4F-PCC in the ED for VKA reversal were included, and the unicentric study design, even though a large cohort of patients was assessed.

In conclusion, this study reports results from a large cohort of patients treated with 4F-PCC for VKA reversal for major bleeding or emergency surgery following a fixed-dose strategy. After 4F-PCC, the majority of patients achieved the target INR, meaning 4F-PCC is a useful modality for rapid INR reduction. The safety profile of 4F-PCC may be considered acceptable. Fixed-dose 4F-PCC was able to rapidly restore haemostasis while minimising the risk of adverse events and optimising available resources. Future comparative prospective studies are required to better evaluate this strategy.

What this paper adds.

What is already known on this subject

• Early reversal of anticoagulation is essential to improve outcomes in the event of major bleeding or emergency surgery.

• Dosing strategies have demonstrated positive outcomes.

What this study adds

• This study reports results from a large cohort of patients treated with four factor-prothrombin complex concentrate (4F-PCC) for major bleeding or emergency surgery following a fixed-dose strategy.

• This study supports fixed-dose 4F-PCC as a strategy to rapidly restore haemostasis while minimising the risk of adverse events and optimising available resources.

Data availability statement

Data are available upon reasonable request.

Ethics statements

Patient consent for publication

Not required.