Efficacy and safety of four-factor prothrombin complex concentrate fixed, weight-based dosing for reversal of warfarin anticoagulation

ABSTRACT

Background

Four-factor prothrombin complex concentrate (4F-PCC) is widely used for urgent reversal of anticoagulation with warfarin, but the optimal 4F-PCC dosing approach is unknown. Herein, we sought to determine the efficacy of a novel fixed, weight-based dosing nomogram.

Methods

We retrospectively studied consecutive adult patients receiving fixed, weight-based 4F-PCC dosing for warfarin reversal between 30 April 2009 and 31 December 2010. The primary outcome was reversal of warfarin anticoagulation, defined as INR ≤1.5 within 6 h. Secondary outcome was the occurrence of thromboembolic events.

Results

A total of 227 patients (56% male), with a median age of 74 years and a median weight of 76kg were evaluated. The most common indications for 4F-PCC were active bleeding (37.4%: 12.7% intracranial, 12.3% gastrointestinal, 4.0% trauma, 8.4% other), reversal for a procedure (22.0%), reversal for surgery (29.5%) or other (11.1%). 66.1% of patients achieved an INR ≤1.5 within 6 h of 4F-PCC administration. 95.0% (57/60) of patients completed a planned procedure and 95.7% (67/70) of patients completed a planned surgery. The median baseline INR was 2.9 (1.5–10) and decreased significantly to a median of 1.3 (1.0–3.7) (p < .001) post-4F-PCC administration. There was no statistically significant difference in response to a fixed, weight-based dose of 4F-PCC based on pre-PCC INR, as long as the pre-treatment INR was ≤ 4.5. Although the majority of patients in our study (99%) received doses over 1000IU, rates of thrombosis were low (1.8%).

Conclusion

Fixed, weight-based dosing of 4F-PCC is effective for reversing warfarin anticoagulation in patients with a pre-dosing INR ≤ 4.5.

KEYWORDS:

• INR
• prothrombin complex concentrate
• 4F-PCC
• weight-based dosing
• warfarin reversal
• ‌anticoagulant reversal
• ‌warfarin antidote
• PCC dosing

Introduction

Four-factor prothrombin complex concentrate (4F-PCC) is widely used for the urgent reversal of patients anticoagulated with warfarin [1] and it has been shown that they are superior to Fresh Frozen Plasma (FFP) in patients who require urgent procedures or experience major or life-threatening bleeding [2]. 4F-PCC has the advantage of being infused in lower volumes than FFP and has a better safety profile due to viral inactivation [3]. However, optimal dosing guidelines have yet to be defined [4,5] and currently, there are varying doses (25 or 50 IU/kg) that have been used both in clinical trials and in practice [6–9]. The generally accepted strategy is the use of a variable dose based on pre-treatment INR and patient body weight; however, recent evidence suggests that a fixed dosing strategy may provide superior hemostatic control [10]. While there is some evidence in the literature supporting a fixed, weight-based dosing approach for 4F-PCC, studies are small and cannot necessarily be generalized to all types of bleeding [11]. Due to the current lack of substantiated dosing regimens for 4F-PCC, coupled with concern for potential thrombogenicity, there is a clear need for studies supporting specific dosing strategies of 4F-PCC focusing specifically on their efficacy and safety profile.

Our institution has been using a fixed, weight-based dosing protocol for 4F-PCC since 2009, which dispenses to a maximum of 3000 IU for adults weighing over 100 kg. Thus, in the present study, we aimed to determine the efficacy and safety of this fixed, weight-based dosing nomogram in a large cohort of patients with various types of bleeding or necessitating urgent reversal of anticoagulation prior to an invasive procedure or surgery.

Methods

Study design, subjects and outcomes

We conducted a retrospective cohort study at the London Health Sciences Centre including all consecutive adult patients (age ≥18 years) who received a fixed, weight-based 4F-PCC (Octaplex® and Beriplex®) dosing regimen for warfarin reversal between 30 April 2009 and 31 December 2010 (n = 227 patients). Patients who received 4F-PCC for any reason other than warfarin reversal were excluded. The fixed, weight-based regimen consisted of 1000 IU for patients weighting less than 40 kg, 1500 IU for 41–60 kg, 2000 IU for 61–80 kg, 2500 IU for 81–100 kg, and 3000 IU for > 100 kg [10]. The primary outcome was the reversal of warfarin anticoagulation, defined as an INR ≤ 1.5 within 6 h of 4F-PCC administration. Secondary outcomes were evaluating the incidence of adverse events, including thrombotic events (stroke or deep vein thrombosis [DVT]). The study was approved by the Institutional Research Ethics Board (Review Number 176E1).

Data extraction and analysis

Data were obtained from clinical and blood bank records. Demographic data included age, sex and weight. Admission diagnosis was coded into one of the following categories: intracerebral hemorrhage or subdural hematoma; lower or upper gastrointestinal bleed, urgent/emergent surgery (requiring surgery within 6 h); elective/scheduled surgery (requiring surgery in greater than 6 h after admission); other procedures (i.e. central lines, lumbar punctures, thoracentesis); trauma; infection/sepsis or other. Data was collected on co-morbidities (including atrial fibrillation, congestive heart failure, coronary artery disease, DVT or pulmonary embolism, bleeding, cerebrovascular accident and stroke). Medications used within 7 days prior to 4F-PCC administration including aspirin, clopidogrel, warfarin, unfractionated heparin, or low-molecular weight heparin were also documented as were laboratory parameters at baseline, 6 and 24 h. Each patient’s indication for receiving 4F-PCC which was coded into one of the following categories: active bleeding; elevated INR but no active bleeding; reversal for procedure; reversal for surgery; active bleed and procedure/surgery; or other. Date and time 4F-PCC was given and the dose category in IU (1000, 1500, 2000, 2500, 3000, >3000) was recorded. For patients who received vitamin K as an adjunct therapy to 4F-PCC, the dose and route of administration were also recorded. We additionally assessed the use of blood products within 6 h prior or after 4F-PCC administration. New bleeding events within 24 h of 4F-PCC administration without an alternative explanation were noted, as were any thrombotic events (objectively confirmed by imaging) within 30 days following 4F-PCC, or until time of discharge (whichever came first). Death and cause of death were also recorded, where applicable. Statistical analysis was performed, including descriptive statistics and chi-squared tests for group comparisons using IBM SPSS Statistics Version 27.0 (IBM Corp., Armonk, NY, USA).

Results

Patient characteristics

Between 30 April 2009 and 31 December 2010, 227 patients met our inclusion criteria. The median age was 74 years (range 40–95), 56% of patients were male and the median weight of all patients was 76 kg (range 44-200) (Table 1). Medical indications for warfarin anticoagulation included atrial fibrillation (68%), DVT (22%) and ischemic or cardioembolic stroke (21%) with some patients falling into more than one of these categories. The most common indications for 4F-PCC administration were active bleeding (37.4%), reversal for a procedure (22.0%), reversal for surgery (29.5%) and other not already listed (11.1%). Active bleeding patients were further classified as: 12.7% intracranial, 12.3% gastrointestinal, 4.0% trauma, and 8.4% were considered other. Prior to receiving 4F-PCC, 29% of patients were taking Aspirin and 4% were taking clopidogrel.

Table 1. Patient characteristics.

Patient demographics	N = 227
Median age in years (range)	74 (40–95)
Median weight in kg (range)	76 (44–200)
Males	127 (56%)
Medical history	N = 227
Atrial fibrillation	154 (68%)
DVT	49 (22%)
Stroke	47 (21%)
Indication for PCC	N = 227
Active bleeding	85 (37.4%)
Intracranial	29 (12.7%)
Gastrointestinal	28 (12.3%)
Trauma	9 (4.0%)
Other	19 (8.4%)
Reversal for procedure	50 (22.0%)
Reversal for surgery	67 (29.5%)
Other not already listed	25 (11.1%)

4F-PCC Doses, adjunctive therapy and responses

The percentages of patients in each dose category are shown in Table 2. Within 6 h of the first dose of 4F-PCC, a second dose of 4F-PCC was given to one patient (0.4%). Fresh frozen plasma was used as an adjuvant therapy within 6 h prior to 4F-PCC administration in 7.4% of patients and in 14.5% of patients within 6 h post-PCC administration. Additionally, 64.7% of patients received vitamin K as an adjuvant therapy within 6 h post-PCC. These adjuvant therapies were prescribed at the discretion of each patient’s attending physician. The median baseline INR was 2.9 (range 1.5–10.0) and decreased significantly (p < 0.001) to a median of 1.3 (range 1.0–3.7) post-PCC administration.

Table 2. PCC doses, adjunctive therapy and responses.

PCC dose	N = 227
1000 IU	3 (1.3%)
1500 IU	40 (17.6%)
2000 IU	93 (41.0%)
2500 IU	55 (24.2%)
3000 IU	36 (15.9%)
Adjunctive therapy	N = 227
Additional dose of PCC within 6 h	1 (0.4%)
FFP within 6 h prior to PCC	17 (7.4%)
FFP within 6 h post-PCC	33 (14.5%)
Vitamin K within 6 h of PCC	147 (64.7%)
INRs	N = 227
Median Pre-PCC INR (range)	2.9 (1.5–10.0)
Median Post-PCC INR (range)	1.3 (1.0–3.7)
Completion of Procedures/Surgeries
% of patients with pre-planned procedure that was completed	95.0% (57 of N = 60)
% of patients with pre-planned surgery that was completed	95.7% (67 of N = 70)
Thrombosis^a
Total thrombotic events within 30 days of PCC	4 (1.8%)
DVT^b	3 (1.3%)
Ischemic Foot^b	1 (0.4%)
Stroke^c	1 (0.4%)
Death due to thrombosis^c	1 (0.4%)

Outcomes

Within 6 h of 4F-PCC administration, 66.1% of patients achieved a target INR of ≤ 1.5, as seen in Table 3. There was a statistically significant difference in response based on pre-PCC INR when this value was >4.5. As seen in Table 3, there was no statistically significant difference in the proportion of patients who achieved a target INR of ≤ 1.5 post-PCC administration between patients with a pre-PCC INR of 1.5–3.0 versus those with a pre-PCC INR of 3.1–4.5. However, there was a statistically significant difference in the proportion of patients who achieved a target INR of ≤ 1.5 post-PCC administration between patients with a pre-PCC INR of 1.5–3.0 versus those with higher pre-PCC INR of >4.5. Pre-planned procedures were completed in 95.0% of 60 scheduled patients, and pre-planned surgeries were completed in 95.7% of 70 scheduled patients within 24 h of receiving 4F-PCC. Four patients (1.8%) suffered a thrombotic event within 30 days of receiving 4F-PCC. Three patients (1.3%) suffered a DVT (one of which also additionally developed an ischemic foot), while one patient (0.4%) suffered a stroke. Of the three patients who suffered a DVT, one was anticoagulated due to a previous DVT, while the other two, including the patient who subsequently developed an ischemic foot, were anticoagulated due to underlying atrial fibrillation. Of note, the patient who suffered a stroke was also anticoagulated due to underlying atrial fibrillation.

Table 3. Proportion of patients within various INR ranges achieving post-PCC INR ≤ 1.5.

INR range pre-PCC	Proportion with post-PCC INR ≤ 1.5 (N = 227)	p-value (2 tailed, mid p value exact)	95% Confidence interval	Total patients reaching target INR ≤ 1.5
1.5–3.0	94/121 (77.7%)	Reference	Reference	150 (66.1%)
3.1–4.5	29/44 (65.9%)	0.0843	−1.22–25.92%
>4.5	27/62 (43.5%)	<0.0001*	24.34–53.01%

Discussion

In the present study, we found that the use of our fixed, weight-based 4F-PCC dosing algorithm resulted in approximately two-thirds of patients achieving an INR ≤ 1.5 within 6 h of 4F-PCC administration, with 95% of patients completing scheduled procedures or surgeries and less than 2% suffering thrombotic complications within 30 days. However, the proportion of patients achieving the target INR was lower for those with a baseline INR > 4.5. These findings suggest that this algorithm is effective and safe for patients with a baseline INR ≤ 4.5, but for those patients with an INR > 4.5 modifications may be required.

Other studies have also demonstrated the efficacy of fixed, weight-based dosing of 4F-PCC for INR reversal [11–13]. Mohrien et al. used a fixed, weight-based (30 IU/kg) dosing protocol of 3-factor PCC (3F-PCC) which demonstrated 85.7% reversal of INR to target without significant adverse events. However, this study only looked at patients with intracranial bleeding requiring rapid INR reversal and excluded any patients where FFP was used as an adjuvant agent to stop bleeding. Further, like many other studies which investigated a fixed, weight-based dosing schedule, this was a small study, assessing the outcome of only 35 patients [11]. Our study achieved a target INR of ≤ 1.5 in 76% of patients following administration of Octaplex and Beriplex, consistent with the findings of similar studies [14].

Adverse outcomes noted included DVTs in three patients (1.3%) (one of which also additionally developed an ischemic foot) and stroke (arterial thrombus) in one patient (0.4%). These thrombotic events were deemed not directly related to 4F-PCC administration. Of the 52 patients (22.9%) who died in our study, only one patient’s death (0.4%) was related to a thrombotic event, the arterial thrombus leading to stroke. This patient was noted to have decreased level of consciousness prior to 4F-PCC administration which was given for a suspected retroperitoneal bleed. No other patient’s cause of death was related to thrombosis (arterial or venous) or to hemorrhage and no deaths were deemed to be directly related to 4F-PCC administration. The rates of thrombosis in our study (1.8%) are comparable to other studies despite the fact that the majority of patients in our study (98.7%) received doses of 4F-PCC over 1000 IU [14,15].

Prothrombin complex concentrates dosing algorithms have undergone many modifications. In 2008, the Canadian National Advisory Committee (NAC) on blood product utilization recommended a single dose of 1000 International Units (IU) for all adults over 18 years of age, irrespective of INR levels a – ‘one dose fits all’ approach. A revision in 2011, suggested dosing based on the INR elevation to be the ideal option and in 2014, the recommendation was updated to incorporate an algorithm including both INR and weight. Although this approach may offer a more tailored approach to dosing, it is more complex and currently, there is insufficient evidence to suggest that one approach is more efficacious than another [5]. In fact, our retrospective cohort study suggests that for most patients with INR < 4.5, a fixed, weight-based dosing shows a similar efficacy. The lack of strong randomized control trial evidence on clinical effectiveness, and high morbidity and mortality in patients requiring urgent warfarin reversal highlights the increased need for ongoing data collection on clinical outcomes in order to ensure best practice guidelines [1]. Fixed, weight-based dosing of 4F-PCC has many advantages over a combination approach of weight-and INR-based dosing algorithms as the former allows clinicians to avoid waiting for blood work results prior to initiating treatment. Fixed, weight-based dosing, however, may be limited by the challenge of obtaining an urgent weight value. However, our local dosing algorithm is flexible enough that an estimated weight within a 10 kg range is sufficient for dosing.

Data on the use of 4F-PCC amongst other agents for direct oral anticoagulants (DOAC) anticoagulation reversal is limited in the literature. Therefore, validating the most efficacious 4F-PCC dosing algorithm for INR reversal will be highly valuable for the future management of patients, practices, policies and guidelines [16]. Our study is subject to the inherent limitations of a retrospective design including difficulty in establishing true cause–effect relationships and absence of a control group for comparison [17]. Further, our retrospective study was missing some post-treatment INR values (approximately 10%). It is also important to recognize that our study utilized INR as a surrogate marker for clinical hemostasis, but INR is not the optimal marker of clinical outcomes. Additionally, only 64.7% of our patient population received vitamin K post-PCC administration despite vitamin K being recommended for all patients after receiving PCC. While vitamin K was administered in accordance with institutional guidelines at the time of study completion, this may have affected patients’ post-PCC INR values. Lastly, our study did not collect data on or measured clinical hemostatic efficacy; therefore, it is possible that patients may not have undergone full reversal of INR but may still have achieved clinical hemostasis. Our study is likely underpowered to detect clinical events, therefore INR was used as a surrogate marker.

In conclusion, our study suggests that a fixed, weight-based dosing algorithm is effective in the reversal of anticoagulation for patients on warfarin with a pre-treatment INR ≤ 4.5. Fixed, weight-based dosing of PCC achieved comparable rates of successful INR reversal to previous studies which utilized alternative dosing strategies. Further studies comparing the clinical and laboratory response using different dosing strategies are warranted.

Acknowledgements

The authors would like to thank Kathy Eckert, Transfusion Safety Officer at LHSC, and Dr Amanda Devraj for their help with initial data collection. Miss Kaitlin Endres: literature review; final analysis and interpretation of data; revision and finalization of statistical analysis; writing of the full manuscript. Dr Rosanne St. Bernard: literature review; REB approval; initial acquisition, analysis and interpretation of data; initial statistical analysis; revisions to the final manuscript. Dr Ian Chin-Yee: REB approval; initial data collection; initial statistical analysis; revisions to the final manuscript. Dr Cyrus Hsia: study conception and design, revisions to the final manuscript. Dr Alejandro Lazo-Langner: study conception and design, revisions to the final manuscript.

Disclosure statement

Dr Lazo-Langner has the following disclosures: Pfizer: honoraria; Bayer: honoraria; Daiichi Sankyo Company: research funding; Alexion: research funding. None of these entities was associated with any aspect of this study. The remaining authors have no conflicts of interest to disclose.

Notes

Note: DVT deep vein thrombosis.

Note: PCC prothrombin complex concentrate; IU international units; FFP fresh frozen plasma; INR international normalized ratio; DVT deep vein thrombosis.

^aThere is no evidence that the thrombotic events are directly related to PCC.

^bIschemic foot and DVT occurred in same patient.

^cStroke and death occurred in the same patient.

Note: INR, international normalized ratio; PCC, prothrombin complex concentrate.

*p-value is considered significant at p < .05.