Abstract
Background:
PCC (Kcentra®) is an Food and Drug Administration (FDA)–approved 4-factor PCC used for the treatment of warfarin-related coagulopathy (WRC), but it has also been used off-label to treat non-WRC. Three-factor PCC in the form of coagulation factor IX human (Bebulin®) has also been used for WRC and off-label to treat non-WRC. It is unclear whether the use of 3- or 4-factor PCCs is effective for the treatment of non-WRC,.
Objective:
Our aim is to characterize the use of 3- and 4-factor PCCs for patients identified with a non-WRC.
Methods:
A retrospective analysis of patients who received PCCs for both WRC and non-WRC between January 2012 and July 2015 was conducted.
Results:
A total of 187 patients with elevated international normalized ratio (INR) who received PCCs were analyzed; 53.9% of patients in the WRC group and 27.7% in the non-WRC group corrected to an INR of 1.3 or less after 3- or 4-factor PCC administration. In those patients with non-WRC and who had underlying liver disease, 3- and 4-factor PCCs reduced mean INR by 0.98 and 1.43, respectively.
Conclusion:
Three and 4-factor PCCs can reduce INR in patients with WRC and in those with non-WRC secondary to liver disease.
Background
Beginning in the early 1990s, prothrombin complex concentrates (PCCs) were originally used for treating major bleeding events. 1 Recent recommendations made by the American College of Chest Physicians however support the use of 4-factor prothrombin complex for patients presenting with vitamin K antagonist (VKA)–associated major bleeding. 2 Given the increasing prevalence of patients using VKA therapy, 4-factor PCCs are being utilized more frequently to treat the toxicity from such agents. 3
Data on PCC administration for the treatment of non-warfarin-related coagulopathy (non-WRC) are limited. Several small studies indicate that coagulopathy of trauma can be reversed with PCCs. 4 –6 PCC use in patients with coagulopathy secondary to liver failure has been evaluated in a few small case studies, yielding positive results regarding hemostasis. 7 Moreover, clinical outcome data regarding reversal agents for novel oral anticoagulants (direct thrombin inhibitors and direct factor Xa inhibitors) are currently lacking and are limited to case studies and animal models demonstrating hemostatic success. 8 –10
While 4-factor PCCs, such as Kcentra®, are Food and Drug Administration (FDA) approved for the treatment of WRC, there is limited evidence showing the utility of PCCs (3- or 4-factor) for the treatment of non-WRC. Examining patients treated with PCCs for non-WRC and their clinical outcomes might help to identify potential off-label indications. The purpose of this study is to characterize the use of PCCs in patients identified with a non-WRC. A secondary objective is to compare the effect between 3- and 4-factor PCCs on international normalized ratio (INR) correction to 1.3 or less in patients with non-WRC.
Methods
After approval by the institutional review board at 2 participating hospitals at a single academic medical center in the northeastern United States, a retrospective analysis of patients who received PCCs at 1 public 850-bed facility between January 1, 2012, and July 1, 2015, and 1 private 725-bed facility between January 1, 2013, and July 1, 2015, was conducted. Given the regulated nature of PCC distribution, relevant patients were identified by a prospectively maintained database by the blood bank at the public hospital and the pharmacy records at the private hospital—each of which is responsible for the distribution of PCCs at their respective institutions.
Inclusion criteria were as follows: adult subjects 18 years and older were included if they were identified as having received PCCs for elevated INR. In order to be included in the data analysis, identified cases needed to have both pre- and post-PCC administration coagulation panels, dosing of PCC, and clear documentation of when the PCCs were administered. Those who did not meet the inclusion criteria were excluded.
Charts were reviewed for patient demographic data (including age, sex, height, weight, body mass index, medical comorbidities, and tobacco and alcohol use), antiplatelet therapy, VKA therapy, novel oral anticoagulant therapy, heparin use, admitting diagnoses, and PCC indication and dosing. It should also be noted that “admitting diagnosis” and “PCC indication” were not necessarily the same. In cases when they were discordant, patients likely developed a new condition as an inpatient and were subsequently treated with PCCs. Routine laboratory panels prior to PCC administration were obtained and included complete blood counts, basic metabolic panels, liver function tests, and blood gases if available. Coagulation panels (prothrombin time, partial thromboplastin time, and INR) immediately preceding and following PCC administration were examined. Additional blood products (including vitamin K, fresh frozen plasma [FFP], packed red blood cells [pRBCs], and platelets) that were administered during 6 hours before and 6 hours after PCC administration were also recorded. Time to follow-up coagulation panel after PCC administration was noted. Thrombotic complications such as myocardial infarction, pulmonary embolism, cerebrovascular accident, and deep venous thrombosis were assessed after PCC administration during the course of the hospital admission. Patient outcomes were coded as “survival to discharge,” “transferred to hospice,” or “patient expiration.”
Subjects were divided into 2 groups: those with WRC and those with non-WRC. Patients in the latter group were subdivided according to their presumed cause of coagulopathy—liver disease, novel oral anticoagulant use, sepsis, or trauma. The primary end point was the change in INR before and after PCC administration. The secondary end point assessed patient outcome (eg, survival to discharge, patient expiration). Given that this study was designed to characterize the use of PCCs for non-WRC, no attempt was made to perform a detailed analysis of the data beyond descriptive statistics (including means, medians, ranges, and standard deviations).
Results
Baseline Data
A total of 249 patients with elevated INR were identified as having received PCCs at both sites during the analyzed time period. Six patients were younger than 18 years of age, 30 did not have documented pre-PCC administration coagulation panels, and 24 had unclear PCC dosing and were thus excluded. The demographic data for the remaining 189 patients are shown in Table 1. Approximately two-thirds of patients were classified as having a WRC.
Demographics.
Abbreviation: PCC, prothrombin complex concentrate.
aAbdominal aortic aneurysm leak, acute pancreatitis, acute respiratory failure, adrenal hemorrhage, altered mental status (x2), aortic aneurysm, aortic injury, atrial fibrillation with rapid ventricular response, cardiac arrest, cellulitis, cellulitis, chest pain, cholecystitis (x3), congestive heart failure exacerbation (x4), dyspnea, dyspnea due to bilateral pleural effusions, epistaxis, generalized weakness, hematuria (x3), hemorrhagic shock, hemorrhagic spleen, hyponatremia, incarcerated inguinal hernia, intra-abdominal abscess, intraperitoneal hemorrhage, lower extremity pain, left ventricular assist device dysfunction, menorrhagia, pericardial effusion, pneumonia, rectus abdominis hematoma (x3), retroperitoneal bleed (x3), seizure, sepsis (x3), small bowel obstruction, transcatheter aortic valve replacement, unwitnessed fall, and urosepsis (x5).
bAbdominal pain, altered mental status (x2), ascites, bacterial peritonitis (x5), cardiac tamponade, cervical spine compression, cholangitis (x2), cirrhosis (x2), flank hematoma, gallstone pancreatitis, generalized weakness, hemarthrosis, hepatic encephalopathy, hepatic encephalopathy (x3), incarcerated hernia, multi-organ failure, pancreatic mass, psoas hematoma, renal failure, sepsis (x2), shortness of breath, transjugular intrahepatic portosystemic shunt procedure, and volume overload.
cPer past medical records.
Note. x refers to the number of occurrences.
Mean baseline laboratory values are displayed in Table 2. On average, patients with non-WRC tended to have higher levels of serum creatinine, blood urea nitrogen (BUN), aspartate aminotransferase, and alanine aminotransferase as compared to those in the WRC group. Moreover, the patients in the non-WRC group also exhibited lower baseline levels of platelets, hemoglobin, and hematocrit.
Baseline Laboratory Values.
Abbreviation: BUN, blood urea nitrogen.
Medication Administration
Indications for PCC are shown in Table 3. The most common indications were intracranial hemorrhage in the WRC group (35%) and gastrointestinal bleed in the non-WRC group (28%). Periprocedural PCC was given to 16% and 21% of patients in the WRC- and non-WRC-related groups, respectively. Other specific indications are outlined in Appendix A.
Prothrombin Complex Concentrate Indications.
Abbreviation: INR, international normalized ratio.
aAbdominal wall hematoma (x4), altered mental status in the setting of elevated INR, cardiac arrest, chest wall hematoma, elevated INR and hemodynamic instability after trauma, elevated INR in the setting of adrenal hemorrhage, elevated INR in the setting of hemoperitoneum, elevated INR in the setting of altered mental status and thigh hematoma, elevated INR in the setting of multiple long bone fractures, elevated INR refractory to other blood products, epistaxis and hemoglobin drop, hematuria (x3), hemorrhagic spleen, INR > 10, pelvic hematoma, pericardial effusion, retroperitoneal bleed (x4), septic shock, unclear source of bleed in the setting of hemoglobin drop and elevated INR (x2), and vaginal bleeding.
bDisseminated intravascular coagulation (x2), elevated INR in the setting of ecchymosis, elevated INR in the setting of hemoglobin drop, elevated INR in the setting of liver disease, elevated INR refractory to other blood products (x2), hemoperitoneum after liver biopsy, hemoglobin drop without a known cause, lower extremity hematoma, multi-organ failure, pericardial tamponade, postoperative epidural hematoma/bleeding, psoas hematoma, and retroperitoneal bleed.
Note. x refers to the number of occurrences.
PCC dosing is listed in Table 4. The dosing of PCC was at the discretion of the individual clinician since PCC administration for non-WRC represents off-label use. Two different types of PCC were utilized during the study period. Three-factor PCC was given in the form of Bebulin® and 4-factor PCC in the form of Kcentra. During the study period, the private hospital site only administered PCCs in the form of Kcentra.
PCC Dosing.a
Abbreviation: PCC, prothrombin complex concentrate.
a3 Factor refers to Bebulin; 4 factor refers to Kcentra.
All patients in the WRC group received 10 mg of intravenous vitamin K versus only 39% of patients in the non-WRC group. FFP administration rates were similar for both groups (42% in the WRC group vs 38% in the non-WRC group). Patients in the non-WRC group tended to receive greater units of both pRBCs (2.50 vs 2.10 U) and platelets (1.92 vs 1.74 U). The specific quantities of other blood product administration are summarized in Table 5.
Blood Product Administration.a
Abbreviation: pRBCs, packed red blood cells.
aOnly products that were given within 6 hours of prothrombin complex concentrate administration were accounted for.
PCC Effect
Overall, 53.9% of patients in the WRC group and 27.7% in the non-WRC group corrected to an INR of 1.3 or less. For the 128 patients with WRC, mean pre-PCC administration INR was 4.57 and mean post-PCC INR was 1.42. This resulted in a mean INR drop of 3.14. Median time to follow-up INR was 3 hours. For the 61 non-WRC-related patients, mean pre- and post-PCC INR was 2.67 and 1.53, respectively. This resulted in a mean INR drop of 1.14. Median time to follow-up INR was 3 hours and 5 minutes. These effects are further subdivided by 3- versus 4-factor PCC in Table 6.
Three Versus 4 Factor PCC Effect.a
Abbreviations: INR, international normalized ratio; PCC, prothrombin complex concentrate.
a3-Factor refers to Bebulin; 4-factor refers to Kcentra.
Thirty-six out of the 61 patients in the non-WRC group exhibited liver-related pathology (ie, active viral hepatitis, cirrhosis, and nonalcoholic steatohepatitis). Table 7 delineates the effect of PCCs in this subgroup; these patients exhibited a mean INR decrease of 0.98 and 1.43 associated with 3- and 4-factor PCC administration, respectively. Of this group, 16 survived to discharge, 6 were transferred to hospice, and 14 died. The effect of PCC administration on INR for the other 25 patients in the non-WRC group is listed in Appendix A. Ten patients in the non-WRC cohort were taking a direct oral anticoagulant (eg, apixaban, dabigatran, or rivaroxaban). Five of these patients received PCC for intracranial hemorrhage, 2 for gastrointestinal bleed, and 1 each for hemarthrosis, cardiac tamponade, and periprocedural safety; all survived to discharge. The remaining patients were believed to be coagulopathic due to sepsis, trauma, or other unique clinical circumstances (see Appendix A).
PCC Effect in Non-Warfarin Coagulopathy Secondary to Liver Disease.a
Abbreviations: INR, international normalized ratio; PCC, prothrombin complex concentrate.
a3-Factor refers to Bebulin; 4-factor refers to Kcentra.
Of the 189 total patients with elevated INRs and who received PCCs, the only complication identified was a single case of deep venous thrombosis in the non-WRC cohort.
Patient Outcome
Figure 1 demonstrates patient outcomes. While the majority of patients in both groups survived to discharge (84% in the WRC group and 59% in the non-WRC group), there was a greater proportion of patients in the non-WRC group who were either transferred to hospice (13% vs 7%) or died (28% vs 9%) as compared to the WRC group.
Patient outcome in WRC and non-WRC. WRC indicates warfarin-related coagulopathy.
Discussion
PCCs are derived from a highly purified concentrate that is composed of variable amounts of the clotting factors II, VII, IX, X, protein C, and protein S. 11 When comparing PCCs to FFP in reversing WRC, PCCs have a diminished risk of infectious disease transmission, can be delivered in large doses without a concomitant large amount of fluid, and can be infused more rapidly. 12 This study was undertaken to characterize the use of 3- and 4-factor PCCs in patients presenting with a non-WRC at an academic medical center comprised of a public and a private hospital. Data regarding the WRC arm are included to permit comparisons on the effect of PCCs on decreasing the INR, whether the etiology is secondary to a WRC or not. Given the descriptive nature of this study, no a priori hypothesis was made to determine whether there was a statistical difference in the effectiveness of 3- versus 4-factor PCC for non-WRC.
Before comparing the 2 groups, baseline health and underlying disease status must be considered. Patients in the non-WRC group appeared to have more severe illness based on higher levels of serum creatinine, BUN, AST, and ALT and lower levels of platelets, hemoglobin, and hematocrit and received more units of pRBCs and platelets. Moreover, a greater proportion of these patients were transferred to hospice or ultimately died. These findings suggest that patients presenting with a non-WRC in this study were “sicker” compared to the WRC group and thus had a greater probability for a poor clinical outcome.
In the trial submitted to the US FDA to obtain approval, Kcentra was shown to reduce INR to ≤ 1.3 thirty minutes after the end of infusion in 62.2% of patients on VKA therapy versus 9.6% of patients who received FFP. 13 In our study, 53.9% of patients in the WRC group and 27.7% in the non-WRC group corrected to an INR of 1.3 or less after 3- or 4-factor PCC administration. While there are no randomized controlled trials comparing 3- versus 4-factor PCCs, there is some literature suggesting an inferior reduction in INR with 3-factor products. 14 Our study supports this notion as 4-factor PCC reduced the mean INR by a greater absolute value than did 3-factor PCC in the WRC group. This is not meant to imply, however, that a change in INR serves as the perfect proxy for clinical hemostasis—moreover, INR is not a reliable marker with regard to monitoring novel oral anticoagulant effectiveness. 15 Taking this into account, when comparing 3- versus 4-factor PCC administration in the non-WRC cohort, 3- and 4-factor PCCs reduced the mean INR by 1.13 and 1.15, respectively.
Of the 61 patients in the non-WRC group, 36 (59%) had underlying liver disease—32 of whom received 4-factor PCCs and 4 of whom received 3-factor PCCs. Liver disease was the only subgroup that demonstrated an INR decrease of close to or greater than 1.00 after 3- or 4-factor PCC administration. Given that the liver is the site of coagulation factor production and that synthetic function is impaired in liver disease, it is not entirely surprising that by restoring clotting factors, PCCs have the potential to improve the INR. These findings suggest that PCCs could have utility in treating non-WRC. However, as mentioned above, caution is warranted in using laboratory value changes as surrogates for clinical improvement.
With regard to PCC dosing for non-WRC, the amount of PCC that was administered was ultimately based on clinician discretion, as there is currently no standardized dosing regimen for such clinical scenarios; the same can be said for the administration of vitamin K in these patients. This heterogeneity in dosing has the capacity to yield inconsistent effects on INR and thus makes it difficult to isolate the true effect of PCCs in this setting.
There are several important limitations to this study. Given the retrospective nature of this project, it would be incorrect to assume that PCC administration alone was responsible for the improvements in INR in the non-WRC group. These patients were all significantly sicker at the baseline, and, in addition to being treated for the underlying disease processes, all received multiple blood products. Another critical limitation is that data were pooled from 2 dissimilar institutions that differ both in patient population and practice habits—such differences may influence study outcomes. The relatively small sample size of our population may not offer a true representation of the risks and benefits of PCCs. This is reflected by a complication rate of 0.5% in our cohort, whereas other studies and clinical trials demonstrate a thromboembolic risk anywhere between 6.8% and 7.3%. 16,17 Given the methodology of this study, our ability to assess the complication rate of PCC administration is limited, as these rates are completely dependent on health-care providers explicitly recording adverse events in the medical record. Finally, a control for the non-WRC group was not identified: patients with elevated INRs who could have been defined as non-WRC but did not receive PCCs.
While caution is always warranted in interpreting a descriptive retrospective study, our findings suggest that 3- and 4-factor PCCs may have some utility in impacting the INR in non-WRC, particularly in patients with liver disease. Future direction should include prospective studies that elucidate the true efficacy of PCCs in such clinical scenarios in order to draw more definitive conclusions.
Footnotes
Appendix A
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.