537897

research-article2014

AOPXXX10.1177/1060028014537897Annals of PharmacotherapyTilton et al

Research Report

Outcomes Associated With Prothrombin Complex Concentrate for International Normalized Ratio Reversal in Patients on Oral Anticoagulants With Acute Bleeding

Annals of Pharmacotherapy 2014, Vol. 48(9) 1106­–1119 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1060028014537897 aop.sagepub.com

Ryan Tilton, PharmD1, Elizabeth Landrum Michalets, PharmD1,2, Bethany Delk, PharmD3, Susan E. Sutherland, PhD4, and Scott A. Ramming, MD5

Abstract Background: Management of bleeding in patients on oral anticoagulants (OACs) is crucial in optimizing outcomes. No large studies examine 3-factor prothrombin complex concentrate (PCC) for OAC reversal. Objective: To assess outcomes after administration of 3-factor PCC for reversal of international normalized ratio (INR). Method: We conducted an institutional review board–approved retrospective cohort study in all patients admitted to our level II trauma center over a 5-year period from 2007 to 2012 who received PCC for INR reversal and bleeding management. The primary outcome was assessment of efficacy as measured by achievement of INR < 1.5. Secondary objectives were to evaluate: factors associated with achievement of target INR, cessation of bleeding, mortality, outcome differences with or without fresh frozen plasma (FFP) or protocol utilization, safety, and cost. Result: A total of 403 patients were evaluated. Target INR was achieved in 88.8% of patients and was influenced by baseline INR. Associated factors were younger age (P = 0.02), utilization of the institution’s protocol (P < 0.01), and concomitant administration of vitamin K (P < 0.01). Concomitant FFP did not affect achievement. Bleeding cessation occurred in 333 (82.6%) patients, and 68 (16.9%) patients died. Patients who achieved target INR were more likely to have bleeding cessation (P < 0.01). The odds of survival for those who reached target INR was 3.8 times greater (P < 0.01). The incidence of thromboembolism was 3.7%. Conclusion: Three-factor PCC administration with IV vitamin K was effective for INR reversal and bleeding cessation and should continue to be a mainstay of therapy pending head-to-head outcome and cost comparisons with 4-factor products. Keywords anticoagulation, clinical pharmacology, emergency medicine, trauma, warfarin

Introduction Background In 2011, there were nearly 34 million prescriptions filled in the United States for warfarin.1 Although the yearly incidence of major bleeding with chronic oral anticoagulants (OACs) is only 2% to 4%, depending on the agent used, time-critical management of these episodes is increasingly important.2-4 The management of patients on OACs who experience a traumatic injury is particularly challenging because bleeding is often spontaneous and potentially magnified because of the diminished presence of clotting factors in these patients. Data from the National Trauma Databank indicate that warfarin use is present in 12.8% of trauma patients older than 65 years, a cohort that may be more challenging to manage.5 In patients who present with intracranial bleeding, an international normalized ratio (INR) of greater than

3 has been associated with larger hematoma volumes, risk for expansion, and worse neurological outcomes.6,7 When managing acute bleeding episodes, achievement of hemostasis along with rapid and complete reversal is essential. In those patients on warfarin, practitioners generally target an INR 6 hours postdose All patients vitamin K 10 mg IV •• No information on concomitant FFP

Imberti et al

Holland et al21

Outcome Measures

a

•

•

•

•

(continued)

Downloaded from aop.sagepub.com at EMORY UNIV on April 19, 2015

1109

Tilton et al Table 1.  (continued)

Reference

Number of Patients

Lankiewicz et al22

58

Imberti et al23

46

Design and Patient Characteristics

PCC Dosing and Concomitant Interventions

Outcome Measures

Results

• Supplemental FFP following the PCC doses further ↓ the mean INR to 2.1 for lowdose PCC (P = 0.01) and 2.0 for high-dose PCC (P < 0.01) • Mean time from baseline INR to follow-up: 20.5 hours control versus 5 minutes per PCC protocol • No significant difference in INR reduction between lowdose and high-dose groups • Neither thromboembolism nor bleeding cessation reported •• Final mean PCC doses not •• Proplex-Tc 25-50 •• Evaluation of rapid •• Retrospective case given reversal protocol series units/kg + vitamin K •• INR correction to 6 (0%) underwent urgent •• Median INR remained ≤1.5 craniotomy in 96% of assessments up to 96 hours post dose •• Two thromboembolic events (4.3%) 37-47 days after hospital DC, not thought to be attributable to PCC because of extended time frame •• One nonfatal recurrence of ICH •• Mortality: 20% but none attributed to PCC (continued)

Downloaded from aop.sagepub.com at EMORY UNIV on April 19, 2015

1110

Annals of Pharmacotherapy 48(9)

Table 1.  (continued)

Number of Patients

Reference 24

Baggs et al

Chapman et al13

Cabral et al25

41 warfarin 9 hepatic failure

13 PCC 18 non-PCC

30

Design and Patient Characteristics

PCC Dosing and Concomitant Interventions b

•• •• Retrospective cohort •• Profilnine SD study •• 14 patients •• Only 82% on (28%) received warfarin and 18% concomitant vitamin received PCC for K 10 mg IV hepatic failure •• 22 patients (44%) received concomitant FFP (range 2-18 units) before final INR assessment

•• Retrospective analysis with comparison between PCC and non-PCC trauma patients •• Mean ISS: 17.8 PCC, 9.1 non-PCC (P < 0.001) •• Mean age (years): 69.3 PCC, 70.2 nonPCC •• Mean baseline INR 3.03 PCC,2.77 nonPCC •• Retrospective case series •• Mean age (years): 72.8, 37% > 80 •• Bleeds: 100% ICH •• Median INR: 2.3

•• Mean total Profilnine •• SDb mean ± SD dose: 2281 ± 1083 units or 25.6 units/ kg •• Concomitant vitamin K: 36% PCC, 61% non-PCC •• Concomitant FFP: 85% PCC (4 units), 89% non-PCC (5.1 units)

Outcome Measures Evaluation of difference in initial INR between patients whose INR was adequately reversed (final INR ≤ 1.5) versus not adequately reversed (INR > 1.5)

Comparison of achievement and time to INR ≤1.5 with PCC

•• Profilnine SDb 25-50 •• Evaluation of degree and time frame for units/kg based on INR ↓ INR + vitamin K 10 •• Evaluation of mg IV + 2-4 units thromboembolic FFP events, hematoma expansion

Results •• Median PCC doses similar between adequate reversal and inadequate reversal patients: 25.2 units/kg versus 24.5 units/kg, P = 0.2 •• 5 patients (10%) received a second PCC dose (median 24.4 units/kg) •• 29 (58%) patients achieved adequate INR reversal to ≤1.5 at a median of 1.5 hours post–PCC dose •• Baseline INRs and adequate reversal: 2-2.9 (72%), 3-3.9 (58%), ≥4 (11%) •• After adjusting for concurrent vitamin K and FFP, initial INR was a significant predictor of adequate INR reversal (odds ratio = 0.38, P = 0.02) •• Neither PCC dose nor concurrent vitamin K or FFP were predictors of achieving adequate INR •• Proportion of patients who received 1 or more units of PRBCs higher in inadequate reversal group (57% vs 28%, P = 0.04) •• Achieved INR ≤1.5: 92% PCC, 89% FFP (P = 1.0) •• Time to INR ≤1.5: 16.6 hours PCC, 30hours nonPCC (P = 0.048) •• Percentage change in INR: 47% PCC, 44% non-PCC •• Thromboembolic events: 3 (23.1%) PCC, 1 (5.6%) non-PCC (5-10 days after admission)

•• Mean PCC dose 2530 units or 28.3 units/kg •• Mean INR ↓ from 2.3 to 1.4 (39%) after dose (P < 0.001) at a median of 95 minutes post–PCC dose •• 100% of patients had INR ≤1.7 within 240 minutes of dose •• 10% hematoma expansion •• 10% thrombotic events within 60 days of DC •• Mean PCC cost/patient: $1688 •• 30% of patients went to neurosurgery with no procedure-related bleeding complications •• Survival to discharge: 80% (continued

Downloaded from aop.sagepub.com at EMORY UNIV on April 19, 2015

1111

Tilton et al Table 1.  (continued)

Reference

Number of Patients

Safaoui et al26

28

Siddiq et al11

10 PCC 9 FFP

Fredriksson et al10

17

Design and Patient Characteristics

PCC Dosing and Concomitant Interventions

Outcome Measures

•• Retrospective case •• PCC product •• Evaluation of dose series and time of PCC changed during •• Mean age: 78.2 •• Evaluation of INR study from years outcomes for Konyned to Profilninea •• All traumatic PCC dosing ≤60 •• Patients also ICH (42% minutes and >60 received vitamin grade 3); 85.7% minutes K 10 mg IV or SQ •• Evaluation of early underwent surgical (82% of patients) intervention adverse events and + FFP (96% of •• Mean baseline INR outcome patients) 5.1

•• Comparison of •• Retrospective chart •• Profilnine SDb 25 achievement and review comparison units/kg for INR 90 kg •• Mean INR ↓ from 5.1 to 1.9 (62.7%; P = 0.008) •• Mean time to INR correction 116 minutes for all patients •• Mean correction time 13.5 minutes in the 11 patients with repeat INR within 30 minutes •• No thrombotic events reported •• 41.6% mortality in surgical intervention patients •• Final mean PCC doses not provided •• INR ↓ for non-PCC group mean 1.84 to 1.34 (27.2% reduction, P < 0.05) and for PCC group mean 2.44 to 1.34 (45.1% reduction, P < 0.005) •• 33% of non-PCC patients vs 80% of PCC patients achieved target INR in 3-4 hours (P = 0.012) •• Mean time to target INR for non-PCC 8.5 hours vs 4.25 hours for PCC (P < 0.05) •• Rate of INR correction for non-PCC patients 0.06/hour compared with 0.27/hour for PCC patients (P < 0.005) •• No thromboembolic complications •• Mean PCC dose 25.8 units/kg •• PT reversed 4 times more rapidly after treatment with PCC: INR 2.83 ↓ to 1.22 (56.5%) within 4.8 hours compared with 2.97 to 1.74 (41.4%) in 7.3 hours with FFP (P < 0.001) •• Reaction Level Scale progressed a mean of 0.2 grades in PCC patients compared with 1.9 grades in FFP patients (P < 0.05) •• Residual neurological deficits at discharge less severe in PCC patients but not statistically significant •• 2 patients died in each group: PPC (20%), FFP (28.6%) •• Thromboembolism not reported (continued)

Downloaded from aop.sagepub.com at EMORY UNIV on April 19, 2015

1112

Annals of Pharmacotherapy 48(9)

Table 1.  (continued)

Reference Boulis et al12

Cartmill et al27

Chong et al28

Number of Patients

Design and Patient Characteristics

PCC Dosing and Concomitant Interventions

Outcome Measures

13

•• Prospective, randomized, nonblinded •• All ICH patients •• Mean age and baseline INR not available but stated to be no different between groups

•• PCC Group: 8 patients received Konyned + FFP + vitamin K •• Non-PCC group: 5 patients received FFP + vitamin K SQ

6 PCC 6 non-PCC

•• •Retrospective observational study •• •Mean age (years): 69 for PCC group and 71 for non-PCC group •• •Median baseline INR: 3.9 for PCC group and 5.0 for non-PCC group •• • All ICH patients

•• 6 PCC patients •• Achievement and received factor IXatime to target INR BPLf 50 units/kg + ≤1.5 vitamin K 10 mg IV •• 6 non-PCC patients received FFP 4 units + vitamin K 10 mg IV

7

•• Comparison of time and correction rate to reach target INR ≤1.3 between PCC and non-PCC patients •• Comparison of changes in Glasgow Coma Scale

• Profilnine SDb + FFP • Time to INR • Retrospective case normalization 6 (n=62)

Baseline INR

Figure 2.  Percentage achievement of target INR 1.4 or less (n = 403). Abbreviation: INR, international normalized ratio.

100

Concomitant Fresh Frozen Plasma

90 80

INR Reducon (%)

70 60 50 40 30 20 10 0 -10 0

2

4

6

8

10

12

14

16

18

20

-20 -30

Pre-Dose INR

Figure 3.  Relationship between percentage INR reduction and baseline INR. Abbreviation: INR, international normalized ratio.

achieved in approximately 90% of patients with a baseline INR of 6 or less. Overall, there was a 47.6% reduction in INR from baseline, but the degree of INR reduction was dependent on baseline INR (Figure 3). For patients with a baseline INR of 6. For the 45 (11.2%) patients who did not achieve target INR, the median (IQR) postdose INR was 1.8 (1.6-2.0). Table 3 summarizes comparisons between patients who achieved target INR and those who did not. Statistically significant factors for achieving target INR included younger age (P = 0.02), concomitant administration of vitamin K (P < 0.01), and use of the institution’s protocol (P < 0.01). Other factors, including baseline INR and the administration of concomitant FFP, were not statistically different.

In total, 116 patients (28.8%) received concomitant FFP (mean = 2.4 units per person). Dosing (mean units ± SD) of PCC (1790 ± 612 vs 1775 ± 748, P = 0.83) and administration of vitamin K (85.3% vs 80.5%, P = 0.25) did not differ between patients who received FFP when compared with those who did not. The patients who received FFP were younger (73.4 vs 77.0 years, P = 0.02), had a higher (mean ± SD) baseline INR (5.2 ± 4.6 vs 3.9 ± 3.3, P < 0.01), and were more likely to have been taken to surgery within 12 hours of admission (20.7% vs 9.4%, P < 0.01). The patients who received FFP reached target INR a mean of 2.8 hours sooner, although this was not statistically significant (P = 0.08). As shown in Table 3, FFP administration did not appear to offer any advantages in achievement of the target INR, and it was not associated with a more profound INR reduction or incidence of bleeding cessation (78.5% vs 84.3%, P = 0.34). In fact, the mortality rate was higher in patients who received FFP (23.3% vs 14.3%, P = 0.03).

Baseline INR To parallel our institution’s dosing protocol, we conducted an analysis between patients with a baseline INR 4 or greater and those with a baseline INR less than 4. Patients with a baseline INR 4 or greater were younger (71.2 vs 78 years), more likely to have a history of congestive heart failure (37.8% vs 22.3%), and more likely to have renal insufficiency, defined as creatinine clearance