a b s t r a c t

Objective: Patients experiencing an Intracranial Hemorrhage on oral anticoagulants often require rapid reversal. This study evaluated patients taking factor Xa inhibitors or warfarin that received reversal with 4-factor prothrombin complex concentrate for an ICH. The objective of the study was to determine if the efficacy of 4F-PCC for the reversal of factor Xa inhibitors is noninferior to its use in warfarin reversal in patients with ICH.

Methods: This was a retrospective, single center, noninferiority trial. Patients presenting to the emergency department with ICH were divided into two cohorts: those taking factor Xa inhibitors versus those taking warfarin. In each cohort, patients received anticoagulation reversal with weight-based 4F-PCC. The primary end- point was hemostatic efficacy defined as <=20% expansion in Hematoma volume on repeat Computed tomography imaging. A pre-specified noninferiority margin of -10% was selected to evaluate the difference between groups for the primary endpoint.

Results: A total of 221 patients were included in the study (factor Xa inhibitors, n = 87; warfarin, n = 134). Effective hemostasis was achieved in 70 patients (81%) on factor Xa inhibitors compared to 111 patients (83%) on warfarin, (-2.4% difference, [95% confidence interval, -12.87 to 8.12]; p = 0.654). There was no statistically significant difference between groups with regards to the primary outcome; however, the use of 4F-PCC in factor Xa inhibitor reversal was not noninferior when compared to 4F-PCC use for warfarin reversal. Hospital length of stay and discharge disposition were similar between cohorts.

Conclusions: The efficacy of 4F-PCC in reversing factor Xa inhibitor-related ICH compared to warfarin-related ICH was not significantly different between groups; however, these results did not prove noninferiority. Further study is warranted to delineate 4F-PCC’s role in reversing factor Xa inhibitors in patients with ICH.

(C) 2022

1. Introduction

Oral anticoagulants are commonly prescribed for the prevention and treatment of thrombosis and the prevention of stroke in atrial fibrilla- tion. Warfarin, a vitamin K antagonist, is the oldest option available in the United States (US) and requires frequent monitoring due to drug- drug, drug-food, and drug-disease state interactions. Factor Xa inhibi- tors, such as apixaban and rivaroxaban, are newer oral anticoagulants

* Corresponding author.

E-mail address: megan.heath@dchsystem.com (M. Heath).

¹ Present address: DCH Regional Medical Center, 809 University Blvd E, Tuscaloosa, AL 35401 U.S.

² Present address: University of South Florida Morsani College of Medicine, 12,901 Bruce B Downs Blvd, Tampa, FL 33612 U.S.

that require less monitoring. They have demonstrated superiority to warfarin in preventing stroke or systemic embolism in atrial fibrillation, and noninferiority to warfarin bridged with parenteral anticoagulation for acute venous thromboembolism treatment [1-4]. As a result, these newer anticoagulants now supersede warfarin in the number of annual prescriptions filled in the US. Currently, over 6 million Americans are taking oral anticoagulants [5].

Regardless of the agent chosen, bleeding is the main adverse event of concern with anticoagulant use. The annual rate of intracranial hemor- rhage (ICH) with warfarin is 0.3-0.6% while factor Xa inhibitors have a lower annual rate at <=0.2%. Additionally, warfarin has been associated with an increased risk of ICH expansion, leading to worse outcomes in- cluding higher mortality [7,8]. Patients with ICH have an increased risk for death secondary to anticoagulant use [9,10]. Despite the lower

https://doi.org/10.1016/j.ajem.2022.04.044

0735-6757/(C) 2022

overall risk of death from ICH with factor Xa inhibitors, hemorrhagic events remain a frequent emergency necessitating effective treatment. Both the American College of Cardiology (ACC) and Neurocritical care Society (NCS) guidelines recommend andexanet alfa for reversal of factor Xa inhibitor-associated bleeding [6,7]. Andexanet alfa was ap- proved by the Food and Drug Administration in 2018 – many years after the first factor Xa inhibitor was approved for use in the US. Four-factor prothrombin complex concentrate (4F-PCC) was intro- duced in 2013, and has been used to reverse the anticoagulant effects of factor Xa inhibitors. Due to the cost and lack of data comparing its use to supportive care or other reversal modalities (including 4F-PCC), integration of andexanet alfa into many hospital formularies continues to be limited. The use of 4F-PCC for factor Xa inhibitors is supported by the ACC and the NCS guidelines, and both continue to include this therapy as a reversal option [6,7]. It is important to note that 4F-PCC has not been tested in large, prospective studies for factor Xa inhibitor reversal, and has not received FDA approval for this indication [7,8,11-13]. Additionally, there are limited studies evaluating its effec- tiveness against a comparator group due to ethical considerations with placebo use and the limited availability of the direct comparator, andexanet alfa [14-18]. Due to these gaps in the literature and the grow- ing numbers of patients on factor Xa inhibitors, off-label agents, like 4F-PCC, require more studies to help determine optimal use. The purpose of this study is to determine the noninferiority of 4F-PCC for the reversal of factor Xa inhibitors compared to warfarin in patients

with ICH.

1. Methods

Study Design and Setting: This was a retrospective, active-control, noninferiority study conducted at a single-site, tertiary referral center in the US and approved by the site’s Institutional Review Board. The fa- cility is an 864-bed community hospital with a level II trauma center. Patients were screened for inclusion in reverse chronological order be- tween January 1, 2015 and September 30, 2020. The primary investiga- tor screened all patients and was unblinded to the patient cohorts. Data collection was completed utilizing the electronic medical record. Pa- tients were divided into cohorts based on outpatient anticoagulant use (factor Xa inhibitor or warfarin).

Selection of Participants: Participants were eligible for inclusion if they presented to the emergency department, and were at least 18 years of age, prescribed Oral anticoagulation prior to admission with ei- ther a factor Xa inhibitor or warfarin, and diagnosed with an acute ICH (as demonstrated on initial head computed tomography [CT] imaging) requiring reversal with 4F-PCC. Patients were excluded if they had a hemorrhage at an anatomical site other than intracranial, were taking a direct thrombin inhibitor, did not receive a repeat head CT scan within 48 h of admission, received blood products prior to the initial CT scan, or received massive transfusion of blood products.

Interventions: Administration of 4F-PCC was based on weight fol- lowing the dosing protocol listed in Table 1. Doses could be rounded to the nearest 250 units by the pharmacist upon order verification per hospital policy. Patients in the warfarin cohort received 10 mg IV vita- min K in addition to 4F-PCC per hospital protocol.

Outcomes: The primary outcome was effective hemostasis achieved (defined as <=20% increase in hemorrhage volume on repeat head CT) as determined by a panel of three radiologists blinded to the cohort

Table 1

Dosing protocol of 4F-PCC.

Warfarin Factor Xa inhibitors

assignments. This definition of hemostasis has been utilized in other studies examining Anticoagulant reversal, including the study that ini- tially gained FDA approval for 4F-PCC [16,19]. The radiologists reviewed pre- and post-treatment CT scans for changes in hemorrhAge SIze and volume to determine if the primary outcome was achieved. Patients were divided equally among the three radiologists. Secondary outcomes included hospital length of stay, discharge disposition (home, home health, skilled nursing facility, inpatient rehabilitation, hospice, or morgue), intubation within 48 h, and incidence of blood products administered within 48 h. The Safety outcomes for this study were any thromboembolic event defined as superficial or deep vein thrombosis, pulmonary embolism, ischemic stroke, or myocardial infarction prior to hospital discharge.

Analysis: The primary outcome was analyzed using a noninferiority margin of -10% from the lower limit of the 95% confidence interval [16]. Secondary objectives were analyzed using the student’s t-test for parametric data, Mann Whitney-U test for non-parametric data, and Chi square test for categorical data. A p-value of <0.05 was used to denote statistical significance.

1. Results

Characteristics of Study Subjects: A total of 636 patients were screened for inclusion. Of those, 221 patients met criteria and were in- cluded in the final analysis. There were 352 patients (85%) who were excluded due to bleeding at another site and 48 patients (12%) who did not have a follow up head CT scan within 48 h. The remaining 15 pa- tients met one of the following exclusion criteria: no documentation of oral anticoagulant use (2%), no initial head CT (1%), no 4F-PCC adminis- tered (<1%), or received blood products prior to scan (<1%). A total of 87 patients were included in the factor Xa inhibitor group and 134 pa- tients in the warfarin group. Baseline characteristics are listed in Table 2. The median Glasgow Coma Scale score at admission was 15 with no difference between groups (p = 0.097). Demographics were similar with regards to sex, age, and antiplatelet use. The majority of patients presented with an intraparenchymal or subdural hematoma in both groups. Patients in the factor Xa inhibitor group received higher doses of 4F-PCC compared to patients on warfarin (median, 47 units/kg vs. 26 units/kg; p < 0.0001). The indications for anticoagulation also

Table 2

Baseline characteristics and clinical indicators.

Characteristic	Factor Xa inhibitor group	Warfarin group	p-value
	(n = 87)	(n = 134)
Male sex – no. (%)	44 (51)	70 (52)	0.726
Age (yr) – mean (SD)	78 (9.07)	75 (11.4)	0.061
GCS at admission – median (IQR)	15 (14-15)	15 (13-15)	0.097
Hemorrhage location – no. (%) Epidural	0 (0)	1 (1)	0.757
Subdural	22 (25)	47 (35)	0.125
Subarachnoid	9 (10)	15 (11)	0.843
Intraparenchymal	30 (34)	33 (25)	0.113
Intraventricular	6 (7)	6 (4)	0.438
Multiple	20 (24)	32 (24)	0.879

Anticoagulant indication – no. (%)

Atrial fibrillation	70 (80)	86 (64)	0.009
Thrombus	17 (20)	28 (21)	0.807
Mechanical heart valve	0 (0)	17 (13)	0.002
Other	4 (5)	13 (10)	0.164
Antiplatelet use – no. (%)	18 (21)	24 (18)	0.709
Dose of 4F-PCC in u/kg – median (IQR)	47 (33-50)	26 (24-33)	<0.00001
Time from CT to 4F-PCC (min) –	67 (50-99)	59 (44-77)	0.039
median (IQR)

INR: international normalized ratio.

;

factor prothrombin complex concentrate; CT: computed tomography.

INR 2 to <4 = 25 unit(s) per kilogram INR 4 to 6 = 35 unit(s) per kilogram	50 unit(s) per kilogram		Time from 4F-PCC to repeat CT (hr) – median (IQR)	14 (9-22)	16 (10-23)	0.105
INR > 6 = 50 unit(s) per kilogram			SD: standard deviation; GCS: Glasg	ow Coma Scale; IQR:	interquartile range	4F-PCC: four

Table 3

Primary outcome of hemostasis achieved on repeat CT.

Factor Xa inhibitor group (n = 87)

Warfarin group

(n = 134)

p-value

Table 4

Secondary outcomes.

Factor Xa inhibitor

group (n = 87)

Warfarin group

(n = 134)

p-value

Hemostasis achieved – no. (%) 70 (81) 111 (83) 0.654

Stable hematoma volume 60 (86) 92 (83) 0.961

Thromboembolic event – no. (%) 9 (10) 7 (6) 0.151

Intubation within 48 h – no. (%) 11 (13) 24 (18) 0.295

Reduced hematoma volume 8 (11) 11 (10) 0.798

Increased hematoma volume <= 20% 2 (3) 8 (7) 0.140

Blood products given within 48 h – no. (%)

13 (15) 34 (25) 0.064

differed between cohorts with more patients receiving anticoagulation for atrial fibrillation in the factor Xa inhibitor group while the warfarin group had more patients with a mechanical heart valve.

Main Results: Effective hemostasis, defined as <=20% expansion on re- peat head CT, was achieved by 70 patients (81%) in the factor Xa inhib- itor group and 111 patients (83%) in the warfarin group (95% CI -12.87 to 8.12; p = 0.654) (Table 3); however, 4F-PCC use for the reversal of factor Xa inhibitors was not found to be noninferior compared to warfa- rin as the lower limit of the confidence interval crossed the pre-defined noninferiority margin of -10% (Fig. 1).

Fig. 1 shows the primary outcome by noninferiority analysis. Treat- ment difference of -2.4% refers to the absolute between-group differ- ence in the rate of hemostasis achieved. The dashed line indicates the noninferiority margin set at -10%.

Secondary outcomes are listed in Table 4. There were no statistically significant differences in any of the secondary outcomes. Eleven pa- tients (13%) required intubation in the factor Xa inhibitor group com- pared to 24 patients (18%) in the warfarin group (p = 0.295). The in- hospital mortality rate was 11% for the factor Xa inhibitor group and 12% for the warfarin group (p = 0.920). A thromboembolic event was observed in nine patients (10%) in the factor Xa inhibitor group and seven patients (6%) in the warfarin group (p = 0.151).

1. Discussion

Rapid Reversal strategies are imperative when anticoagulated pa- tients present with life-threatening ICH. Given the dynamics of treat- ment and limited availability of andexanet alfa, lower cost alternatives are often used preferentially for the reversal of factor Xa inhibitors. Off-label use of 4F-PCC was determined to be therapeutically effective to reverse factor Xa inhibitor-induced coagulopathy in a number of small studies [11-16]. The present study is the largest known retrospec- tive analysis to date of 4F-PCC use in ICH reversal and uses patients tak- ing warfarin as an active control group to assess hemostatic efficacy.

This retrospective chart review included 221 patients who pre-

sented with ICH complicated by anticoagulant use. Baseline characteris- tics between groups were similar with some notable exceptions. First, the indications for anticoagulation differed between groups. Warfarin is approved for use in mechanical heart valves while factor Xa inhibitors

-2.4% (-12.87 to 8.12)

-20

-10

0

10

20

Fig. 1. Noninferiority margin.

Length of stay (days) – median (IQR) 5.2 (3-9) 6 (3-11) 0.124

Discharge Disposition – no. (%)

Home	18 (21)	23 (17)	0.510
Home with home health	21 (24)	28 (21)	0.712
Skilled nursing facility	20 (23)	42 (31)	0.177
Inpatient rehabilitation	9 (10)	13 (10)	0.876
Hospice	9 (10)	12 (9)	0.731
Deceased	10 (11)	16 (12)	0.920

IQR: interquartile range.

are not. Secondly, there was a statistically significant difference be- tween the time from CT read to 4F-PCC administration with the factor Xa inhibitor group having a median Time to administration of 67 min compared to 59 min in the warfarin group (p = 0.039). This may be sec- ondary to the additional time required to reconstitute the larger doses of 4F-PCC necessary to reverse factor Xa inhibitors, and is not consid- ered clinically significant.

Determining efficacy in anticoagulation reversal studies is challeng-

ing leading many researchers to resort to the use of surrogate outcomes instead of clinical outcomes [19]. These studies use definitions of clinical hemostasis to determine efficacy based on parameters that vary by bleeding location [15]. A common definition for effective hemostasis proposed by the International Society of Thrombosis and Haemostasis is

<35% increase in hematoma volume on repeat imaging [20]. Sarode et al. developed a similar hemostatic efficacy scale during a phase III trial to gain FDA approval for the use of 4F-PCC in warfarin reversal. This scale defined “excellent” hemostasis as <=20% hemorrhage expan- sion, and "good" hemostasis as >20% but <=35% hemorrhage expansion in ICH. This scale has since been used in multiple studies [16,19-21].

There was no statistically significant difference in the primary out- come of “excellent” hemostasis between the factor Xa inhibitor or war- farin groups (p = 0.654); however, these results were not noninferior due to the confidence interval crossing the noninferiority margin of

-10%. This study’s definition of effective hemostasis was more strin-

gent than previous studies examining hemostatic control and may ac- count for the inconclusive results. The factor Xa inhibitor group demonstrated “excellent” hemostasis in 70 patients (81%). These find- ings were similar to a study performed by Korobey et al. that noted a 78% rate of “excellent” hemostatic efficacy for factor Xa inhibitor- related ICH with 4F-PCC. The Korobey study had a similar methodology to this study; however, the researchers did not utilize a comparator group [16]. The present study attempted to strengthen its results by cre- ating an active comparator group consisting of patients that were being treated for warfarin-associated ICH. Due to 4F-PCC holding FDA ap- proval for the treatment of warfarin-related bleeding (including ICH), this study elected to use this group as a benchmark for efficacy compar- ison. A study by Connolly et al. evaluated andexanet alfa for the treat- ment of factor Xa inhibitor-associated bleeding, and demonstrated an 80% rate of “good” or “excellent” hemostasis in patients with ICH. This is similar to the current study’s finding that 4F-PCC provides “excellent” hemostasis in 81% of patients. This suggests that 4F-PCC may be used to reverse factor Xa inhibitor-related ICH bleeds with a similar rate of effi- cacy as the FDA-approved product.

Additionally, 4F-PCC at the aforementioned doses was shown to be safe in these groups with a thrombotic rate of 10% in the factor Xa inhib- itor group and 6% in the warfarin group (p = 0.151). Although the dif- ference between Thromboembolic events was not statistically significant, this value did trend towards an increased risk in the factor

Xa inhibitor group. Two of the nine patients with thrombotic events in the factor Xa inhibitor group had a past medical history of thrombosis in the same limb area. This may indicate that the clot identified on ultra- sound was a re-demonstration of an old occlusion; however, this cannot be determined through a retrospective chart review. In published stud- ies of andexanet alfa, thrombotic events occurred in 10% of patients – a rate that was considered acceptable by the authors [21].

There are several important limitations to this study. First, it was a retrospective analysis which relied heavily on the validity of the docu- mentation in the electronic medical record for the patients’ baseline characteristics and secondary outcomes. Second, the study was con- ducted at a single-site ED at a community hospital which limits general- izability to other institutions with dissimilar processes, procedures, and patient populations. Third, part of the patients examined in this study were admitted during the COVID-19 pandemic. This could have con- founded some of the secondary results like Intubation rates. However, patients from both cohorts were included during this timeframe to min- imize this limitation. Lastly, while it was possible to objectively measure the volume of both subdural and epidural hemorrhages, the size of bleeds at other cerebral Anatomical sites could only be estimated allowing for possible interpretation variability. To mitigate any subjec- tivity, the radiologist panel members were blinded to the patient group assignments.

In conclusion, the efficacy of 4F-PCC to reverse factor Xa inhibitor- related ICH is similar, but not noninferior to its use in warfarin reversal. This study supports the continued off-label administration of 4F-PCC for factor Xa inhibitor reversal but should prompt further exploration into this subject area. Additional investigation is warranted to directly com- pare 4F-PCC to andexanet alfa in this population.

Sources of support

None to disclose.

Previous presentations

A summary of this research was presented at the virtual Florida Residency Conference on May 13th, 2021. This included a virtual PowerPoint presentation detailing the study design, objectives, and key data findings. The results were also presented internally at Lakeland Regional Health to the pharmacy department and ED physicians.

Credit authorship contribution statement

Megan Heath: Writing – review & editing, Writing – original draft, Validation, Supervision, Project administration, Methodology, Investi- gation, Formal analysis, Data curation, Conceptualization. Brad Hall: Writing – review & editing, Writing – original draft, Visualization, Supervision, Project administration, Methodology, Formal analysis, Conceptualization. Jason De Leon: Writing – review & editing, Writing – original draft. Rita Gillespie: Writing – review & editing, Writing – original draft. Shannon Hasara: Writing – review & editing, Writing –

original draft, Methodology. Bret Henricks: Investigation. Magge Lakshmi: Investigation. Davin Watson: Investigation. Kayla Wilson: Writing – review & editing, Writing – original draft, Resources, Methodology.

References

1. Agnelli G, Buller HRB, Cohen A, et al. Oral apixaban for the treatment of acute venous thromboembolism. N Engl J Med. 2013;369:799-808.
2. Granger CB, Alexander JH, McMurray JJV, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2011;365:981-92.
3. EINSTEIN Investigators. Oral rivaroxaban for symptomatic venous thromboembo- lism. N Engl J Med. 2010;363:2499-510.
4. Patel MR, Mahaffey KW, Garg J, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011;365(10):883-91.
5. Barnes GD, Lucas E, Alexander GC, Goldberger ZD. National trends in ambulatory oral anticoagulant use. Am J Med. 2015;128(12):1300-5.
6. Frontera JA, Lewin III JJ, Rabinstein AA, et al. Guideline for reversal of Antithrombotics in intracranial hemorrhage. Neurocrit Care. 2016;24:6-46.
7. Tomaselli GF, Mahaffey KW, Cuker A, et al. 2020 ACC expert consensus decision pathway on Management of Bleeding in patients on Oral Anticogaulants. JACC. 2020.;76(5).
8. Salem AM, Roh D, Kitagawa RS, Choi HA, Chang TR. Assessment and management of coagulopathies in neurocritical care. J Neurocrit Care. 2019;12(1):9-19.
9. Rosand J, Eckman MH, Knudsen KA, et al. The effect of warfarin and intensity of anticoagulation on outcome of intracerebral hemorrhage. Arch Intern Med. 2004; 164:880884.
10. Steiner T, Weitz JI, Veltkamp R. Anticoagulant-associated intracranial hemorrhage in the era of reversal agents. Stroke. 2017;48:1432-7.
11. Perzborn E, Heitmeier S, Laux V, Buchmuller A. Reversal of rivaroxaban-induced anticoagulation with prothrombin complex concentrate, activated prothrombin complex concentrate and recombinant activated Factor VIII in vitro. Thromb Res. 2014;133(4):671-81.
12. Eerenber ES, Kamphuisen PW, Sijpkens MK, Meijers JC, Buller HR, Levi M. Reversal of rivaroxaban and dabigatran by prothrombin complex concentrate: a randomized, placebo-controlled, crossover study in healthy subjects. Circulation. 2011;124(14): 1573-9.
13. Levi M, Moore KT, Castillejos CF, et al. Comparison of three-factor and four-factor prothrombin complex concentrates regarding reversal of the anticoagulant effects of rivaroxaban in healthy volunteers. J Thromb Haemost. 2014;12(9):1428-36.
14. Zahir H, Brown KS, Vandell AG, et al. Edoxaban effects on bleeding following punch biopsy and reversal by a 4-factor prothrombin complex concentrate. Circulation. 2015;131(1):82-90.
15. Allison TA, Gass JA, et al. Evaluation of the use of low-dose 4-factor prothrombin complex concentrate in the reversal of direct oral anticoagulants in bleeding patients. J Intensive Care Med. 2020;35(9):903-8.
16. Korobey MJ, Sadaka F, Javed M, Moynihan M, Alsaei A. Efficacy of 4-factor prothrom- bin complex concentrates in factor Xa inhibitor-associated Intracranial bleeding. Neurocrit Care. 2021;34(1):112-20.
17. Stratman R, Owen E, Gibson G, Human T. Prothrombin complex concentrate (Kcentra) for reversal of oral factor Xa inhibitors. Crit Care Med. 2015;42(12):164-5.
18. Smith MN, Deloney L, Carter C, Weant KA, Eriksson EA. Safety, efficacy, and cost of four-factor prothrombin complex concentrate in patients with factor Xa inhibitor-related bleeding: a retrospective study. J Thromb Thrombolysis. 2019;48 (2):250-5.
19. Sarode R, Molling TJ, Refaai MA, et al. Efficacy and safety of a 4-factor prothrombin complex concentrate in patients on vitamin K antagonists presenting with major bleeding: a randomized, plasma-controlled, phase IIIb study. Circulation. 2013; 128:1234-43.
20. Khorsand N, Majeed A, Beyer-Westendorf J, Meijer SK. Assessment of the effective- ness of major bleeding management: proposed definitions for effected hemostasis: communication from the SCC of the ISTH. J Thromb Haemost. 2015;14:211-4.
21. Connolly SJ, Crowther M, Eikelboom JW, et al. Full study report of andexanet alfa for bleeding associated with factor Xa inhibitors. N Engl J Med. 2019;380(4):1326-35.