A Retrospective Comparison of 3-Factor Prothrombin Complex Concentrate Products for Warfarin Reversal

Abstract

Background and Purpose:

Current guidelines suggest that 3-factor prothrombin complex concentrate is a possible alternative to 4-factor products for the emergent reversal of bleeding secondary to warfarin. While multiple observational studies have evaluated various forms of 3-factor prothrombin complex concentrate individually, no study has compared the efficacy of the 2 products. The purpose of this study is to compare the efficacy and safety of Bebulin™ and Profilnine™ for the emergent reversal of warfarin-associated major bleeding.

Methods:

We conducted a retrospective cohort study of patients receiving both Bebulin™ and Profilnine™ at an urban, academic medical center with comprehensive stroke center designation and a neurosurgical center of excellence. All patients were treated at a single center that utilized a fixed, weight-based dosing protocol. The primary outcome was the percentage of patients in each group achieving a goal international normalization ratio of 1.4 or less.

Results:

There was a significant difference in goal international normalization ratio achieved favoring Bebulin™ (85.5% vs 27.3%; P < .001) over Profilnine™. Median dose per kilogram of actual body weight was the same between the groups. When we assessed results by baseline™ international normalization ratio subgroup, more patients in the Bebulin™ group achieved goal when baseline values were 6 or less. No thrombotic events were documented in either group.

Conclusions:

We found that patients treated with Bebulin™ experienced significantly higher rates of successful international normalization ratio reversal when compared to those who received Profilnine™. Further research is needed to determine the comparative efficacy between the 2 agents.

Introduction

An estimated, 3.4% of patients treated with warfarin will experience a major hemorrhage each year, accounting for nearly 60 000 annual emergency department visits across the United States.^1,2,3 Given the associated high morbidity and mortality of these hemorrhages, emergent reversal with prothrombin complex concentrate (PCC) has become the standard of care recommended by numerous guidelines.^4-8 While 4-factor PCC (4PCC) is the only product approved by the United States Food and Drug Administration for emergent warfarin reversal, 3-factor PCC (3PCC) products have been widely used off-label and may offer potential advantages, including lower medication costs and a possible reduction in the risk of thrombosis.⁹ In the United States, 2 primary 3PCC products have been evaluated in the literature: Bebulin™ (Baxter Healthcare, Deerfield, Illinois) and Profilnine™ (Grifols Biologicals, Los Angeles, California). These products primarily replete levels of factors II, IX, and X. Both products contain low levels of factor VII, with the package insert for Bebulin™ listing a lower content than what is listed for Profilnine™.^10,11 Numerous observational studies have validated the role of these products in emergent warfarin reversal, and several retrospective comparative studies have demonstrated similar efficacy to 4PCC.^9,12-17 However, no study, observational or randomized, has compared the efficacy and safety of both 3PCC products. Given the lack of direct comparisons, our study aimed to compare rates of international normalized ratio (INR) reversal and risk of thrombosis between Bebulin™ and Profilnine™ in patients requiring emergent reversal for warfarin-associated major hemorrhage.

Methods

This retrospective cohort study was conducted at Methodist University Hospital in Memphis, Tennessee, a comprehensive stroke center that utilized a fixed, weight-based dosing protocol for 3PCC administration (Figure 1).¹⁸ Patients received 1 of the two 3PCC products: Bebulin™ or Profilnine™. While we previously utilized Bebulin™ as our PCC agent of choice, we began utilizing Profilnine™ in 2018 after a change in preferred agent by our supplier of factor products. We used the same weight-based dosing protocol (30 U/kg) for both agents. The research was determined to be exempt by the University of Tennessee Institutional Review Board due to its retrospective nature. The requirement for informed consent was also waived.

Figure 1.

Hospital protocol for emergent warfarin reversal. CBC indicates complete blood count; CPOE, computerized physician order entry; INR, international normalization ratio; IVPB, intravenous piggyback; kg, kilogram; PCC, prothrombin complex concentrate.

We included patients >18 years of age who received 3PCC for reversal of warfarin due to intracranial hemorrhage (ICH), gastrointestinal hemorrhage, or emergent reversal for acute surgical intervention. From January 2012 to February 2018, Bebulin™ was the formulary agent used for warfarin reversal, with Profilnine™ used from March 2018 to December 2018. Indication for reversal was abstracted directly from physician documentation in the emergency department, and the decision to administer agents was at the discretion of the treating provider, with the electronic treatment protocol used for all patients. We excluded those receiving reversal for any other indication as well as those with documentation in the medical record of having liver disease–associated coagulopathy. Three investigators extracted all data from the electronic medical record. The primary outcome was the percentage of patients achieving INR of 1.4 or less on initial follow-up measurement after 3PCC administration. We also reported the number of thrombotic events after treatment with 3PCC as a secondary outcome measure. Data collected included age, body weight, indication for and 3PCC dosing, baseline and subsequent INR results, discharge status, and occurrence of any thrombotic events within 7 days of 3PCC administration. We also evaluated whether additional reversal agents were administered in the first 24 hours of admission including but not limited to phytonadione (vitamin K) and fresh frozen plasma (FFP). Our institution specific protocol includes INR monitoring to be drawn approximately 30 minutes after PCC infusion completed as well as every 12 hours for 1 day and then daily for 3 days. We planned a priori to complete a subgroup analysis of the primary outcome based upon 3 ranges of baseline INR. Because no previous literature had established a definite expected rate of goal INR achievement with each agent, we were unable to conduct a power analysis prior to completing the study. We included all patients within the defined period to arrive on the final sample presented.

We analyzed continuous data using the Student t test for parametric data and the Mann-Whitney U test for nonparametric data, with analysis of categorical data completed using either the χ² test or the Fisher exact test. Normally distributed data are presented using mean (±standard deviation) and median (25%-75% interquartile range) for non-normally distributed data. All tests for statistical significance were 2-tailed, and statistical significance was established at the P < .05 threshold during the initial analysis. We planned a priori to assess the primary outcome based upon 3 baseline INR categories, and a Bonferroni correction was used to adjust for multiple comparisons.¹⁹ When differences were compared in the preplanned subgroup analysis, statistical significance was determined if P was <.01667 (0.05/3). All analysis was performed using SPSS (IBM Corp Released 2017. IBM SPSS Statistics for Macintosh, Version 25; IBM Corp, Armonk, New York).

Results

We included 127 patients: 83 patients treated with Bebulin™ and 44 with Profilnine™. No data were missing for any included patient. A summary of baseline characteristics is provided in Table 1. Patients treated with Bebulin™ were slightly older (median 75 vs 66 years; P = .02), less likely to be African American (42.2% vs 61.4%; P = .04), and more likely to have an ICH (95.2% vs 59.1%; P <.001) as their indication for emergent warfarin reversal. Baseline INR measurements were also different, with patients in the Profilnine™ group having higher median baseline values (2.6 vs 3.9; P < .001; Table 2).

Table 1.

Patient Characteristics.^a

	Bebulin™, n = 83	Profilnine™, n = 44	P
Age, years	75 [64-83]	66 [57-78.5]	.02
Male, n (%)	45 (54.2)	20 (45.5)	.35
African American, n (%)	35 (42.2)	27 (61.4)	.04
Height, cm	172 [162.3-180.0]	168.3 [162.0-175.5]	.22
Actual body weight, kg	85.0 [70.2-100.8]	88.4 [64.1-115.3]	.61
Obese, n (%)	40 (48.2)	22 (50.0)	.85
Anticoagulation-related past medical history, n (%)b
Atrial fibrillation	45 (54.2)	27 (61.4)	.44
Deep vein thrombosis	19 (22.9)	6 (13.6)	.21
Pulmonary embolism	11 (13.3)	3 (6.8)	.27
Ischemic stroke	10 (12.0)	5 (11.4)	.91
Valve replacement	7 (8.4)	2 (4.5)	.42
Peripheral vascular disease	1 (1.2)	3 (6.8)	.12
Other thrombusc	4 (4.8)	1 (2.3)	.48
Reversal indication
Intracranial hemorrhage, n (%)	79 (95.2)	26 (59.1)	<.001
Gastrointestinal hemorrhage, n (%)	2 (2.4)	12 (27.3)	<.001
Emergent surgery, n (%)	2 (2.4)	6 (13.6)	.01
In-hospital mortality, n (%)	26 (31.3)	11 (25.0)	.46

^a Data are presented as median [25%-75% interquartile range] unless otherwise noted.

^b Patients may have had more than one documented indication for anticoagulation.

^c Included four patients with a cardiac thrombus and one with a superior vena cava thrombus.

Table 2.

Anticoagulation Reversal Information.^a

	Bebulin™, n = 83	Profilnine™, n = 44	P
INR measurements
Baseline	2.6 [2.2-3.5]	3.9 [2.6-5.9]	.001
Immediately after PCC	1.3 [1.2-1.4]	1.7 [1.4-2.0]	<.001
24 hours post PCC	1.2 [1.2-1.3]	1.4 [1.3-1.7]	<.001
Minutes between PCC dose and follow-up INR	48 [31-86]	63 [41-102]	.03
INR at goal (<1.4), n (%)	71 (85.5)	12 (27.3)	<.001
Initial PCC dose (units)	2472 [2228-3045]	2520 [2000-3500]	.53
Units per kilogram actual body weight	30.5 ± 2.8	30.9 ± 2.6	.33
Repeat dose given, n (%)	5 (6.0)	6 (13.6)	.19
Supplemental reversal agents receivedb
Vitamin K prior to follow-up INR, n (%)	61 (73.5)	34 (77.3)	.64
Vitamin K in first 24 hours, n (%)	75 (90.4)	38 (86.4)	.49
FFP prior to follow-up INR, n (%)	0 (0)	5 (11.4)	.004
FFP in first 24 hours, n (%)	5 (6.0)	9 (20.5)	.01

Abbreviations: FFP, fresh frozen plasma; INR, international normalization ratio; PCC, prothrombin complex concentrate.

^a Data are presented as median [25%-75% interquartile range] unless otherwise noted.

^b Patients received no other reversal agents in the first 24 hours of presentation.

A similar percentage of patients in both the groups also received supplemental vitamin K, with more patients in the Profilnine™ group also treated with repeat 3PCC doses and supplemental FFP (Table 2). No patients in the Bebulin™ group received FFP prior to the first follow-up INR, while 11.4% of patients in the Profilnine™ group did (P = .004). We observed no thrombotic events or documented infusion reactions in either group.

There was a significant increase in the percentage of patients meeting goal INR in the Bebulin™ group (85.5% vs 27.3%; P < .001). This difference persisted in the subgroup of patients with ICH (88.6% Bebulin™ vs 34.6% Profilnine™; P < .001). Median INR measurements were also lower in the Bebulin™ group (1.3 vs 1.7; P = .001) upon first follow-up and at 24 hours (1.2 vs 1.4; P < .001). When we assessed the primary outcome based upon baseline INR values, we also observed significant differences. More patients who received Bebulin™ met goal when baseline INR was 2 to 3.9 (92.5% vs 45.5%; P < .001) and 4 to 6 (54.5% vs 0%; P = .01). While the numerical percentage of patients achieving goal INR was higher in the Bebulin™ group when baseline INR was >6 (60% vs 18.2%), this difference did not meet the significance threshold (<0.0167) after the adjustment for multiple comparisons (Table 3).

Table 3.

International Normalization Ratio (INR) Outcomes by Baseline Measurements.^a

	Baseline INR 2-3.9, n = 89			Baseline INR 4-6, n = 22			Baseline INR > 6, n = 16
INR Values	Bebulin™, n = 67	Profilnine™, n = 22	P b	Bebulin™, n = 11	Profilnine™, n = 11	P b	Bebulin™, n = 5	Profilnine™, n = 11	P b
Baseline	2.5 [2.1-2.8]	2.6 [2.2-3.1]	.42	4.9 [4.7-5.1]	4.5 [4.3-4.7]	.010	7.3 [7.1-10.2]	9.8 [7.7-10.0]	.91
After PCC	1.3 [1.2-1.4]	1.5 [1.4-1.6]	<.001	1.4 [1.3-1.7]	2.0 [1.9-2.3]	<.001	1.4 [1.4-1.9]	2.0 [1.7-2.7]	.12
INR < 1.4, n (%)	62 (92.5)	10 (45.5)	<.001	6 (54.5)	0 (0)	.010	3 (60.0)	2 (18.2)	.25

Abbreviations: PCC, prothrombin complex concentrate

^a Data are presented as median [25%-75% interquartile range] unless otherwise noted.

^b After adjustment for multiple comparisons, P < .0167 was used to determine statistical significance.

Discussion

Current guidelines suggest that 3PCC may be an option for warfarin-associated major bleeding, including those with ICH. However, no previous study has compared the 2 commercially available agents, and our results suggest their effects are not equal.²⁰ Our study found that administration of 30 U/kg of Bebulin™ resulted in a higher percentage of patients achieving goal INR when compared to an equal dose of Profilinine, including the subgroup of patients with ICH. We also observed increased efficacy of Bebulin™ across multiple predefined baseline INR ranges. Despite the difference in INR reversal, we found no thrombotic events with either agent.

We identified a plethora of observational studies using both 3PCC agents for warfarin reversal, with Profilnine™ used in a larger number of studies than Bebulin™. Two studies demonstrated higher efficacy (80% and 89%) when targeting goal INR of <1.4 and using an INR-based dosing regimen for Profilinine™ (25 U/kg used if INR <4.0 and 50 U/kg if 4.0 or higher).^21,22 The same Profilnine™ dosing regimen achieved goal in 100% of patients when using an INR target of 1.7.²³ When studying patients with baseline INR >5, Holland and colleagues found that Profilnine™ at 25 and 50 U/kg attained goal INR <3.0 in only 55% and 43% of patients, respectively.²⁴ Multiple studies also demonstrated reduced efficacy at achieving goal INR when using Profilnine™. When using a median Profilnine™ dose of 25.2 U/kg, goal INR <1.5 was seen in only 58% of patients.²⁵ Similarly, Switzer and colleagues observed that 62.9% of patients achieved a goal INR <1.4 with a single 50-unit/kg dose.²⁶ A single-center, observational study observed even lower rates of successful reversal (INR < 1.5) when using 25 U/kg (12%) and 35 U/kg (69%).²⁷ Multiple nonrandomized studies have also utilized Profilnine™ and compared results to patients receiving 4PCC. Two comparative studies demonstrated low rates of success (42.9% [INR 1.3 or less] and 45.6% [INR 1.5 or less]) with Profilnine™ as the 3PCC agent, while those in the 4PCC groups achieved goal INR in a much higher percentage of cases.^12,14 While a 2015 study reported that 80% of those in the 3PCC group achieved goal INR <1.5 and was similar to that of 4PCC, the results are confounded due to the large proportion of 3PCC patients also receiving FFP.¹⁷ Several other observational studies that utilized Profilnine™ for warfarin reversal are also limited due to FFP coadministration as well as very small sample sizes of less than 10 patients per report.^28-31,15 Overall, the majority of studies that utilized Profilnine™ reported low rates of successful INR reversal regardless of the dosing regimen used, which is consistent with the low success rate observed in our study.

While fewer studies evaluating 3PCC use have utilized Bebulin™, available evidence has demonstrated improved results compared to the body of evidence available using Profilnine™. In 2014, a single-center study of warfarin-associated patients with ICH that received a fixed, weight-based dose of Bebulin™ without adjunctive FFP demonstrated 85.7% efficacy at achieving goal INR of 1.4 or less. Mean INR was reduced from 3.2 on admission to 1.3 post-3PCC.¹⁸ Another study using Bebulin™ at 50 U/kg achieved goal INR of <1.4 in 87.5% of patients.³² In addition, 3 studies comparing 3PCC to 4PCC used Bebulin™. A multicenter, retrospective, propensity-matched cohort analysis of 148 patients (72 in the propensity-matched arm) demonstrated 85.7% efficacy (84.2% in matched analysis) at achieving goal INR 1.4 or less, which was similar to that of 4PCC.¹⁶ A single-center study comparing 3PCC to 4PCC included 137 total patients (68 receiving 3PCC) and found that a similar number of patients in each group achieved goal INR of 1.5 or less (79% Bebulin™ vs 81% 4PCC).³³ Conversely, a multicenter study of 64 patients with trauma receiving Bebulin™ ranging from 25 to 50 U/kg found that only 51% of patients on warfarin in the 3PCC arm reached goal INR of 1.5 or less, compared to 81% in the 4PCC group.³⁴ While several other studies used Bebulin™ in their assessment, inclusion of less than 10 patients per study limits applicability of their results.^35,28

Generally, there is more consistency in published studies demonstrating higher rates of successful INR reversal with Bebulin™, which is similar to what we observed in our study. While patients treated with Profilnine™ in our study had a higher median baseline INR, we completed a prespecified subgroup analysis based upon 3 INR ranges derived from the recommendations set forth for 4PCC dosing.³⁶ Across all ranges, Bebulin™ demonstrated more successful INR reversal than Profilnine™, although this was not at the threshold needed to achieve statistical significance in the group with baseline INR >6 despite the 45% difference between groups. It is notable that previous research has hypothesized that 3PCC efficacy is possibly reduced when compared to that of 4PCC due to lower factor VII content.^37,16 We did observe that rates of successful INR reversal decreased with both products as baseline INR increased above 4, which highlights the theory that factor VII activity decreases above this INR threshold and adequate reversal may only be achieved by supplementing factor VII. This is supported by several other studies that demonstrated low rates of INR reversal when using Profilnine™ in patients with higher baseline INRs.^24,25,22,14

In our study, Bebulin™ still resulted in achieving goal INR in more patients when the exact same dosing regimen was used with both agents. Per package labeling, Bebulin™ (0.05-0.2 units of factor VII per unit of factor IX)¹⁰ appears have a lower amount of factor VII than Profilnine™ (no more than 0.35 units of factor VII per unit of factor IX).¹¹ However, prior studies have noted variability between PCCs in actual factor composition compared to product labeling, degree of thrombin generation, and level of impurities, all of which have the potential to effect efficacy.^38,39 Additionally, we identified 2 additional studies that provided insight into hemophilia resources that list Bebulin™ as having greater concentrations of factor VII and factor X as well as containing a more well-balanced mixture of coagulation factors.^40,41

Our study has several limitations not previously discussed. As this was a retrospective study, sample size and the observational nature of the data limit widespread applicability. Additionally, patients with higher baseline INR did not receive increased doses of 3PCC, which could have resulted in lower rates of success at higher baseline INR values. Nonetheless, our study does have several strengths. Although it is retrospective, it is the first comparison of two 3PCC products and used the same standardized dosing protocol for both agents. Since 3PCC is not labeled for warfarin reversal, there are no clear package labeling or guideline recommendations for dosing in this population, further emphasizing the importance of the standardized dosing utilized in our study. Our data also provide meaningful information on utilizing 3PCC in the real-world setting and the ability to achieve adequate INR reversal outside the controlled trial environment. It is evidence to support that certain patient populations, notably those with baseline INR values less than 4, may achieve adequate reversal with Bebulin™. Currently, neurocritical care guidelines for the management of warfarin-induced ICH recommend that both 3PCC and 4PCC are valid treatment options. Given this recommendation, we feel our results are even more notable as Bebulin™ was recently discontinued by the manufacturer. Notably, the significant cost difference between 3PCC and 4PCC agents may cause hospital systems to consider use of Profilnine™ as a cost-savings measure. Based upon purchasing information from our wholesaler, the average cost per unit of KCentra was approximately 160% higher than Profilnine™. Given that our study has a large population of patients with ICH and our results demonstrated a significantly higher percentage of patients achieving goal INR with Bebulin™, we feel current guideline recommendations stating that 3PCC and 4PCC are both viable options for warfarin-associated hemorrhage should be reevaluated.

Conclusions

Our study demonstrated higher rates of successful INR reversal with Bebulin™ when compared to Profilnine™. More research on the comparative efficacy of various 3PCC products is warranted, and further research is needed to determine the comparative efficacy between Bebulin™ and 4PCC before definite recommendations can be made in international guidelines.