Clinical outcomes after 4F-PCC for warfarin-associated ICH and baseline GCS less than or equal to 8

a b s t r a c t

Purpose: There is limited evidence describing the Mortality benefit of utilizing 4-factor prothrombin complex concentrate (4F-PCC) in patients presenting with a warfarin-associated intracerebral hemorrhage (ICH) and a Glasgow Coma Scale (GCS) of <=8. The aim of this study is to determine the potential mortality benefit of 4F-PCC in this patient population.

Methods: This was a retrospective chart review, performed at a comprehensive stroke center from October 2013 through August 2020. Patients were included if they were >= 18 years of age, experienced a spontaneous ICH with baseline GCS <= 8, treated with warfarin prior to admission, had a baseline INR >= 1.7, and received 4F-PCC for INR normalization due to warfarin-associated ICH. The primary outcome was in-hospital mortality at 30 days.

Results: A total of 252 patients received 4F-PCC in the specified time period. Of those patients, 25 patients met inclusion criteria. Sixteen patients (64%) experienced in-hospital mortality. When compared to a historical esti- mated 80% mortality rate in the studied patient population, there was no statistically significant difference (p = 0.208) in mortality when 4F-PCC was utilized to reverse INR.

Conclusion: The administration of 4F-PCC in patients presenting with warfarin-related ICH and GCS <= 8 did not result in statistically significant mortality benefit. Our results are limited by study design and sample size. Thus, larger studies are needed to determine if a benefit exists for 4F-PCC in this patient population. Although the results are not statistically significant, our small study suggests that there may be a clinically significant mortality benefit when 4F-PCC is utilized.

(C) 2022

  1. Introduction

Spontaneous, non-traumatic intracerebral hemorrhage (ICH) repre- sents 10 to 15% of all strokes [1]. Up to 25% of ICHs are associated with oral anticoagulant use. At 30 days, these patients have an estimated mortality of 40%. Independent predictors that lead to increased 30-day mortality are greater ICH volume, infratentorial location, low score on Glasgow Coma Scale (GCS), older age, and intraventricular extension of the hemorrhage [1-3]. Specifically, a low GCS, 9 or less, score after ICH has shown to be predictive of a mortality rate between 40 and 90% [4-7].

Warfarin, a vitamin K antagonist (VKA), is an oral anticoagulant pre- scribed for multiple health conditions requiring systemic anticoagulation. Warfarin inhibits Vitamin K-dependent clotting factors II, VII, IX, and X [8]. The level of anticoagulation is measured by the International normalized ratio , a ratio of the patient’s prothrombin time to a control PT.

* Corresponding author at: Rush University Medical Center, 1653 West Congress Parkway, Atrium 0036, Chicago, IL 60612, United States of America.

E-mail address: [email protected] (G.W. Slocum).

[8] INR goals vary depending on the indication, but a range of 2 to 3 is uti- lized most. The risk of bleeding doubles when the INR is >3, while the risk of ICH doubles for each increase of 1 in the INR [9].

Warfarin use is a risk factor for early mortality after ICH [3]. Patients on warfarin are at a two- to five-fold increased risk of ICH compared to patients not on anticoagulation. Up to 76% of patients with an anticoagulant-associated ICH either expire or become dependent [2]. A reversal agent commonly utilized to reverse elevated INR with associ- ated bleeding due to warfarin is inactivated four factor prothrombin complex concentrates (4F-PCC), which quickly replete factors II, VII, IX, and X. Sarode et al. studied 4F-PCC versus fresh frozen plasma in pa- tients with warfarin-associated major bleeding [10]. Patients with ICH were included in this study but were excluded if they had a GCS less than or equal to 8. They found that there was no difference in hemo- static efficacy between the groups at 24 h, but 4F-PCC was superior to FFP in rapid INR reduction and correction of INR (INR <= 1.3) within 1 h. The researchers concluded that 4F-PCC was an effective alternative to FFP for the urgent reversal of warfarin in major bleeding events [10]. Currently, there is a paucity of evidence describing the mortality benefit of utilizing 4F-PCC in patients presenting with a warfarin-

0735-6757/(C) 2022

related ICH with a GCS less than or equal to 8. The aim of this study is to determine the in-hospital mortality benefit of administering 4F-PCC in patients with warfarin associated spontaneous ICH, and a GCS of less than or equal to 8.

  1. Methods

This was a retrospective chart review, performed at a 600-bed, aca- demic medical center and comprehensive stroke center that included adult patients with warfarin associated ICH and a baseline GCS of less than or equal to 8 between October 2013 to August 2020. The study was approved by the Institutional Review Board.

The study included adult patients (>18 years of age) with a sponta- neous ICH and a baseline GCS of less than or equal to 8. Patients were included if they were on warfarin prior to admission, had a baseline INR of >1.7, and received 4F-PCC for INR normalization. Patients were excluded if 4F-PCC was ordered but never administered, patient was on a direct oral anticoagulant prior to admission, patient was pregnant, or if the ICH was due to trauma, aneurysm rupture, hemorrhagic trans- formation of cerebral infarct, or tumor presence.

The primary outcome was to evaluate the potential hospital mortal- ity benefit of administering 4F-PCC to patients with a baseline GCS of less than or equal to 8, a patient population that was excluded from the Sarode et al. study. Secondary outcomes included disposition at hos- pital discharge and adverse events related to ICH or 4F-PCC. With lim- ited available data, we predicted a 30-day mortality of 80% for patients who experienced ICH while on warfarin. We predicted that patients who received 4F-PCC reversal for warfarin-related ICH would experi- ence a 10% mortality benefit. The electronic medical record system was utilized to extract data, including: patient age, sex, race, past med- ical history, location or unit in hospital during receipt of 4F-PCC, indica- tion for warfarin, concomitant use of antiplatelet agent with warfarin, 4F-PCC product received with dosing, timing of PCC administration, quantity of administered blood products (fresh frozen plasma, Packed red blood cells, platelets, or cryoprecipitate), administration of vitamin K with route and dose, pre-treatment lab values (INR, hemoglobin, and platelet count), post-treatment lab values (INR, hemoglobin, and platelet count up to 30 h post reversal agent administration), head CT (initial, repeat, location of hemorrhage, size of hemorrhage), baseline and discharge Glasgow Coma Scale (GCS) modified Rankin Score (mRS), Deep vein thrombosis prophylaxis (drug, dose, timing of initiation post-ICH), length of stay in an intensive care unit (ICU), length of hospital stay, mortality during hospital admission, incidence of thrombotic events (DVT, pulmonary embolism [PE], myocardial infarc- tion [MI] and ischemic Cerebrovascular events) during hospital admission, and discharge disposition (long-term acute care hospital, hospice, home). If the patient survives to discharge, their electronic medical record will be used to collect additional data, including: if the patient returned to our institution within 3 months of discharge, the date the patient returned, chief complaint upon return, disposition after return (emergency department, admit to observation, admit to general medicine floor, admit to intensive care unit).

A single investigator collected data and 10% of the data was verified and agreed upon by a second investigator. The data collection tool uti- lized was reviewed and agreed upon by all investigators. A sample size calculation was performed and determined 137 patients were required to meet 80% power. Statistical significance was defined as a p-value

<=0.05. Descriptive statistics were utilized to describe baseline character- istics of the study populations and analyze the data. A chi-square test was utilized to determine statistical significance for the primary outcome.

  1. Results

A total of 251 patient encounters were reviewed with 25 patient en- counters meeting inclusion criteria. The most common criteria requiring exclusion were 4F-PCC utilization for non-ICH indications, GCS >8, and

ICH due to trauma (Fig. 1). Of the included patients, the median age was 69.2 years (IQR 59.5, 79.5) with 10 patients (40%) being African American (Table 1). The majority of patients were male (64%). Median baseline INR was 2.33 (IQR 2.15, 3.17) and median baseline GCS was 5 (IQR 3, 7). On head Computerized tomography , 16 patients (64%) had a Midline shift present (Table 2). The majority of hemorrhages were intraparenchymal with 14 hemorrhages extending beyond the primary location (Table 2). Only 6 patients had surgical intervention which included 2 External ventricular drain placements, 2 craniotomies with evacuation, 1 craniotomy, and 1 craniectomy with partial evacuation.

A total of 16 patients (64%) experienced in-hospital mortality after receiving 4F-PCC for warfarin reversal (Table 2). When compared to the predicted mortality rate of 80%, there was no statistical difference (p = 0.21). Eight of 16 patients had plans for inpatient hospice or com- fort care, while the other eight had cardiac arrest and/or brain death. Of those patients who survived their initial hospital stay, the median dis- charge GCS was 14 (IQR 11, 15) with a minimum GCS of 6. Two patients returned to the institution after discharge. One patient was admitted for a complication that was deemed to be related to the prior ICH. The pa- tient was admitted to the ICU for seizures 76 days after initial discharge. Additionally, 1 patient entered hospice care upon discharge from the hospital (Table 2). There were no adverse events, such as thromboem- bolic events, deaths, or late bleeding, events related to the use of 4F-PCC.

  1. Discussion

Our results show that the utilization of 4F-PCC within our study pop- ulation did not result in a significant mortality benefit. Our study also provides insight in regard to the use of 4F-PCC in the patient population that was previously excluded from the Sarode et al. study.

Spontaneous ICH is known to have a high morbidity and mortality, particularly, when patients are on anticoagulants and have low baseline GCS. Given the previous study results and the combination of warfarin use and low GCS, we chose a predicted mortality of 80% in the study population. It is important to recognize that in the following studies, 4F-PCC was not utilized to reverse anticoagulation in the study partici- pants. A single-center, retrospective chart review aimed to identify fac- tors associated with early warfarin-related ICH mortality and identify variables responsible for poor functional outcome. They found that 39.7% of patients died within 7 days of presentation and these patients had a median GCS score of 9 [6]. In another single-center chart review, patients who were on warfarin, suffered from a spontaneous ICH, and presented with a GCS less than or equal to 7 had a mortality rate of 94.4%. The authors concluded that as GCS decreased, mortality rate in- creased. A fatal prognosis was predicted in patients who presented with a GCS <7 [7]. Another retrospective study observed 342 patients and attempted to identify predictors of 30-day mortality in patients with spontaneous ICH. Eighty-six patients died within 30 days. These patients had a mean GCS score of 6 +- 3. The investigators identified a “High risk group” of patients with spontaneous ICH as having an initial GCS < 11 and ICH volumes >32 mL supratentorially or 21 mL infratentorially. The 30-day mortality for this group of patients exceeded 50% in their study [11]. In a retrospective review in Finland, investigators observed 3218 patients with spontaneous ICH to compare ICU severity scores (APACHE II, SAPS II, and SOFA) to age and GCS to ob- serve usefulness of ICU severity scores in predicting six-month mortal- ity. There were 1999 patients with a baseline GCS score between 3 and 8. Within that group, 1345 patients (67.2%) expired within 6 months [12]. The Finnish investigators retrospectively reviewed 972 pa- tients, aiming to determine if adding radiologic long-term prognostic factors to clinical factors predicted mortality in ICU-treated ICH patients. In these patients, 508 patients had a baseline GCS score of 3 to 8, with 340 patients (70%) expiring at 12 months [13]. As previously mentioned, 16 patients (64%) in this study experienced in-hospital

Total Included n = 25

(n = 65, 28.8%)

(n = 56, 24.8%)

(n = 43, 19.0%)

(n = 40, 17.7%)

(n = 13, 5.7%)

(n = 4, 1.8%)

(n = 4, 1.8%)

(n = 1, 0.4%)

(n = 0, 0.0%)

Excluded Patients n = 226

4F-PCC for non-ICH indication GCS greater than 8

Traumatic ICH

On a DOAC prior to admission

ICH due to hemorrhage transformation of cerebral infarct

Did not receive 4F-PCC reversal Pre-treatment INR less than 1.7 ICH due to aneurysm rupture Pregnancy

Screened for Inclusion n = 251

Fig. 1. Inclusion and exclusion.

mortality. The mentioned studies align with the high mortality rate that we found within our similar patient population.

A recent retrospective review observed the relationship between ex- clusion criteria based upon the Sarode et al. study with mortality and disability. The investigators found that a baseline GCS of <7 in combina- tion with one other exclusion criterion from the Sarode et al. study increased the in-hospital mortality rate from 30% to 78.4%. Patients with a GCS of <7 with intraventricular, subdural, infratentorial, and subarachnoid bleeds were found to have a mortality rate of 85%, 78.6%, 80%, and 100%, respectively [14]. Twelve of the 14 hemorrhage extensions extended intraventricularly. As previously mentioned,

Table 1

Baseline characteristics.

Table 2

Clinical characteristics.

Midline Shift Noted on CT scan, n (%) 16 (64)

Hemorrhage Type Size, in cm3?

Intraparenchymal (n = 19??) 36.7 (15.0, 87.5)

Subdural hemorrhage (n = 6?) 1.2 (1.0, 1.2)

Subarachnoid hemorrhage (n = 1) –

Hemorrhage Extension (n = 14)

Intraventricular, n (%) 9 (64.3)

Subarachnoid, n (%) 1 (7.1)

Intraventricular + subarachnoid, n (%) 2 (14.3)

Subdural + subarachnoid, n (%) 1 (7.1) Intraventricular, subarachnoid + subdural, n (%) 1 (7.1)

Location 4F-PCC Administered

Baseline GCS 5 (3, 7)

Age, in years

69.4 (59.5, 79.5)

Emergency Department, n (%)

Intensive Care Unit, n (%)

2 (8)

22 (88)

Sex, male, n (%)

16 (64)

General Medicine Unit, n (%)

1 (4)

Weight, in kilograms

75.1 (62.2, 90.5)

Dose of 4F-PCC


in units

1698.0 (1638.5,

Caucasian, n (%)

7 (28)


African American, n (%)

10 (40)

in units per kilogram

25.0 (22.0, 27.4)

Hispanic, n (%)

2 (8)

IV Vitamin K dose, in milligrams (n = 22)

10 (10, 15)

Asian, n (%)

1 (4)

Blood Products, in units

Other, n (%)

5 (20)

Fresh Frozen Plasma (n = 10)

2 (1.75, 2.75)

Warfarin Indication

Platelets (n = 3)


Atrial fibrillation, n (%)

7 (28)

Post-4F-PCC INR (n = 23)

1.24 (1.09, 1.43)

History of DVT, n (%)

7 (28)

Post-4F-PCC Hemoglobin, in grams per deciliter (n =

10.1 (8.7, 12.4)

History of PE, n (%)

4 (16)


History of Aortic BVR, n (%)

2 (8)

Post-4F-PCC Platelet Count,x 109/L (n = 23)

205 (149, 226)

History of Mitral BVR, n (%)

1 (4)

ICU length of stay, in days

3 (2, 7)

Other/undocumented, n (%)

7 (28)

Hospital Length of Stay, in days

4 (2, 9.5)

Home Antiplatelet

Discharge Disposition

Aspirin, n (%)

7 (28)

Home, n (%)

2 (8)

Clopidogrel, n (%)

2 (8)

SAR, n (%)

5 (20)

Baseline INR

2.33 (2.15, 3.17)

LTACH, n (%)

1 (4)

Baseline Hemoglobin, in grams per deciliter

11.7 (9.8, 13.1)

Hospice, n (%)

1 (4)

Baseline Platelet Count,x 109/L

214 (155, 258)

Expired prior to discharge, n (%)

16 (64)

Discharge GCS 14 (11, 15)

All continuous variables represented by medians (interquartile range).

BVR: bioprosthetic valve replacement, DVT: deep vein thrombosis, GCS: Glasgow Coma Scale, PE: pulmonary embolism.

CT: computerized tomography.

* Size for Intraparenchymal hemorrhage is represented by ABC/2, size for subdural represented by maximum thickness (in centimeters).

?? One patient had two distinct intraparenchymal hemorrhages.

? One patient had two distinct subdural hemorrhages.

intraventricular extension is an independent risk factor for mortality [1-3]. This study provides support for the high mortality rate that was found within our patient population given the combination of low GCS and location of hemorrhages.

Within our study, we did not find any adverse event related to the use of 4F-PCC. 4F-PCC has a black box warning for arterial and venous thromboembolic complications as there is a risk of reversing warfarin in patients with underlying diseases that predispose them to thrombo- embolic events. Sarode et al. found similar rates of adverse events in re- spect to death or thromboembolic event between those patients who received 4F-PCC compared to those who received only plasma [10].

There were several limitations within our study. First, due to the sample size our study was underpowered and at risk for Type 2 error. Second, functional outcomes were not assessed. It is important to assess functional outcome of patients who suffer ICH, but due to limitations in documentation in the medical record, we were not able to collect this data. Additionally, there were patients transferred from outside facilities that may have received 4F-PCC prior to arriving to our hospital. These patients were not captured in our study population due to not having an order for 4F-PCC at our facility.

  1. Conclusions

The administration of 4F-PCC for patients presenting with warfarin- related ICH and a GCS less than or equal to 8 did not result in a signifi- cant mortality benefit. Our results are limited by study design and sam- ple size. Larger prospective studies are needed to determine if a benefit exists for 4F-PCC in this patient population.


This research did not receive any specific grant from funding agen- cies in the public, commercial, or not-for-profit sectors.



Conflicts of interest

None of the authors have any actual or potential conflicts of interest.

CRediT authorship contribution statement

Kristen E. Koehl: Writing – review & editing, Writing – original draft, Validation, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Nicholas G. Panos: Writing – review & editing, Supervision, Methodology, Conceptualization. Gary D. Peksa: Writing – review & editing, Supervision, Formal analysis, Data curation.

Giles W. Slocum: Writing – review & editing, Supervision, Methodol- ogy, Conceptualization.


Tara Kimbrough, MD, MPHa; Hannah Breit, MDa for analyzing head computerized tomography of included patients.


  1. Houben R, Schreuder FHBM, Bekelaar KJ, Claessens D, van Oostenbrugge RJ, Staals J. Predicting prognosis of intracerebral hemorrhage (ICH): performance of ICH Score is not improved by adding oral anticoagulant use. original research. Front Neurol. 2018;9(100). 2018-February-28.
  2. An SJ, Kim TJ, Yoon B-W. Epidemiology, risk factors, and clinical features of intrace- rebral hemorrhage: an update. J Stroke. 2017;19(1):3-10.
  3. Pinho J, Costa AS, Araujo JM, Amorim JM, Ferreira C. Intracerebral hemorrhage out- come: a comprehensive update. J Neurol Sci. Mar 15 2019;398:54-66. https://doi. org/10.1016/j.jns.2019.01.013.
  4. Parry-Jones AR, Abid KA, Napoli MD, et al. Accuracy and clinical usefulness of intra- cerebral hemorrhage grading scores. Stroke. 2013;44(7):1840-5. 1161/STROKEAHA.113.001009.
  5. Qureshi AI, Tuhrim S, Broderick JP, Batjer HH, Hondo H, Hanley DF. Spontaneous in- tracerebral hemorrhage. N Engl J Med. 2001;344(19):1450-60. 1056/nejm200105103441907.
  6. Zubkov AY, Mandrekar JN, Claassen DO, Manno EM, Wijdicks EFM, Rabinstein AA. Predictors of outcome in warfarin-related intracerebral hemorrhage. Arch Neurol. 2008;65(10):1320-5.
  7. Secil Y, Ciftci Y, Tokucoglu F, Beckmann Y. intracranial hemorrhages related with warfarin use and comparison of warfarin and acetylsalicylic acid. J Stroke Cerebrovasc Dis. 2014;23(2):321-6. 2013.03.008.
  8. Coumadin [package insert]. Princeton, NJ: Bristol-Myers Squibb Company; Revised 2011.
  9. Schulman S, Beyth RJ, Kearon C, Levine MN. hemorrhagic complications of anticoag- ulant and Thrombolytic treatment: american college of chest physicians evidence- based clinical practice guidelines (8th edition). Chest. 2008;133(6,Supplement): 257S-98S.
  10. Sarode R, Milling TJ, Refaai MA, et al. Efficacy and safety of a 4-factor prothrombin com- plex concentrate in patients on vitamin K antagonists presenting with major bleeding. Circulation. 2013;128(11):1234-43.


  1. Safatli DA, Gunther A, Schlattmann P, Schwarz F, Kalff R, Ewald C. Predictors of 30- day mortality in patients with spontaneous primary intracerebral hemorrhage. Surg Neurol Int. 2016;7(Suppl. 18):S510-7. 187493.
  2. Fallenius M, Skrifvars MB, Reinikainen M, Bendel S, Raj R. Common intensive care scoring systems do not outperform age and glasgow coma scale score in predicting mid-term mortality in patients with spontaneous intracerebral hemorrhage treated in the intensive care unit. Scand J Trauma Resusc Emerg Med. Oct 25 2017;25(1):


  1. Fallenius M, Skrifvars MB, Reinikainen M, et al. Spontaneous intracerebral hemor- rhage. Stroke. 2019;50(9):2336-43. 024560.
  2. Zaeem M, Porter BA, Delibert S, Jones CMC, Acquisto NM. The association of ICH ex- clusion criteria with mortality and disability rates in ICH patients receiving 4F-PCC for anticoagulation reversal. Ann Pharmacother. 2022. 1060028020962115. 0(0):1060028020962115.