Background: Intracranial hemorrhage (ICH) is a serious complication of warfarin therapy. Plasma infusion is the current standard of care in the US, even though most guidelines recommend prothrombin complex concentrate (PCC) as the preferred therapy; PCC has several advantages over plasma including no ABO blood type requirement, no lengthy thawing and infusion process, and a very small volume of virally inactivated pure vitamin K-dependent factors (VKDF) content that causes neither volume overload nor TRALI. Ideal PCC should contain adequate amounts of all VKDF (4-factor). As our group reported at the 2006 ASH meeting, PCC currently available in the USA have a very low FVII content (hence called a 3-factor PCC), often resulting in an incomplete correction of supratherapeutic INR. PCC is, therefore, often supplemented with 1–2 units of plasma to provide FVII; however, many elderly patients on warfarin do not tolerate an additional fluid volume and there is an urgency to correct coagulopathy in ICH. Recombinant FVIIa (rFVIIa) has been used to reverse warfarin effect with variable success. An experimental model has shown that rFVIIa corrects PT/INR in vitro but fails to correct bleeding because it does not address FII and X deficiencies critical for thrombin generation in vivo when compared with PCC. Therefore, we developed a trauma coumadin protocol (TCP) for urgent warfarin reversal for ICH, which includes both PCC and a low dose rFVIIa.

Methods: TCP was approved by the Blood Utilization Review Committee of the Parkland Memorial Hospital and consists of administration of 4000 IU PCC (Profilnine, Grifols) and 1.2 mg rFVIIa; 5 mg intravenous Vit K1 is also administered daily for 3 days. PCC contains FII no more than 150 units, FX no more than100 units, and FVII no more than 35 units for each 100 units of FIX. Pre- and post-TCP VKDF were measured when possible (Table 2). We compared 19 patients treated with TCP with a 12 patient-non-TCP control group (warfarin-ICH) treated either with plasma alone (n=3) or plasma plus PCC (n=9). We will also compare (software delay) the 2 groups for extension of hematoma volume; the radiologist will be blinded to the therapy.

Results: The TCP group included 12 females and 7 males aged 46–87 years (median 64). The control group included 4 females and 8 males aged 53–89 years (median 73). TCP patients had significantly faster as well as complete correction of PT/INR as compared to the controls. (Table-1). Post-TCP CT scans showed stable hematomas. 1/19 TCP patients with pituitary adenoma had pre-TCP INR of 7.2 and post-TCP INR of 0.9. He developed thrombotic stroke and acute myocardial infarction 2 days later; he was discharged home. The patient received 2.4 mg of rFVIIa because initially the dose of rVIIa was stratified based on INR (1.2 mg for INR 5.0 or less and 2.4 mg for INR greater than 5.0). . Examination of post-treatment FVII levels (Table-2) led us to modify TCP rFVIIa dose to a 1.2 mg, irrespective of INR.

Conclusion: TCP rapidly and effectively corrects warfarin-associated coagulopathy in patients with ICH. Because thrombotic complications remain a concern, 4-factor PCC should further be evaluated.

Table 1.

Test resultsTime relative to TCP treatmentTCP group n=19Control group n=12p value
INR Mean ± SD (range) Pre 3.1±2.0 (1.4–5.1) 3.2 ±1.6 (1.4–7.2) 0.991 
 Post 1.1±0.3 (0.9–2.0) 1.4 ±0.3 (0.9–2.0) 0.003 
PT Mean ± SD (range) Pre 28.1±17.3 (13.5–65.1) 28.9±14.0 (13.7–62.6) 0.887 
 Post 10.7±2.2 (9.3–18.1) 13.8±2.8 (9.3–18.3) 0.002 
PTT Mean ± SD (range) Pre 36.0±10.3 (21.8–59.8) 38.2 ±11.5 (25.4–60.6) 0.591 
 Post 24.7±4.2 (19.5–32.2) 27.8±4.5 (22.8–30.9) 0.058 
Time from TCP to INR correction (min) Mean ± SD (range) 145.4±107 (22–329) 267±76 (83–697) 0.023 
 Median 111 240  
Test resultsTime relative to TCP treatmentTCP group n=19Control group n=12p value
INR Mean ± SD (range) Pre 3.1±2.0 (1.4–5.1) 3.2 ±1.6 (1.4–7.2) 0.991 
 Post 1.1±0.3 (0.9–2.0) 1.4 ±0.3 (0.9–2.0) 0.003 
PT Mean ± SD (range) Pre 28.1±17.3 (13.5–65.1) 28.9±14.0 (13.7–62.6) 0.887 
 Post 10.7±2.2 (9.3–18.1) 13.8±2.8 (9.3–18.3) 0.002 
PTT Mean ± SD (range) Pre 36.0±10.3 (21.8–59.8) 38.2 ±11.5 (25.4–60.6) 0.591 
 Post 24.7±4.2 (19.5–32.2) 27.8±4.5 (22.8–30.9) 0.058 
Time from TCP to INR correction (min) Mean ± SD (range) 145.4±107 (22–329) 267±76 (83–697) 0.023 
 Median 111 240  

Table 2.

Test resultsINRPTPTTFIIFVIIFIXFX
Pretreatment Mean ± SD (range) 3.3±2.0 (1.5–7.4) 29.6±17.3 (14.4.–65.1) 35.7±9.9 (30.1–54.7) 54±41 (6–117) 23 ±12 (6–37) 57 ±28 (17–96) 44 ±30 (5–74) 
Posttreatment Mean ± SD (range) 1.1±0.4 (0.9–2.0) 11.0±3.0 (9.4–18.1) 25.9±4.1 (19.5–32.2) 148±84 (27–323) 394±141 (114–600) 97±22 (58–130) 144±54 (89–268) 
Test resultsINRPTPTTFIIFVIIFIXFX
Pretreatment Mean ± SD (range) 3.3±2.0 (1.5–7.4) 29.6±17.3 (14.4.–65.1) 35.7±9.9 (30.1–54.7) 54±41 (6–117) 23 ±12 (6–37) 57 ±28 (17–96) 44 ±30 (5–74) 
Posttreatment Mean ± SD (range) 1.1±0.4 (0.9–2.0) 11.0±3.0 (9.4–18.1) 25.9±4.1 (19.5–32.2) 148±84 (27–323) 394±141 (114–600) 97±22 (58–130) 144±54 (89–268) 

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author

Sign in via your Institution