Chest ACCP Member Benefits
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Guest Access | Sign In via User Name/Password
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Levine, R. L.
Right arrow Articles by McCollum, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Levine, R. L.
Right arrow Articles by McCollum, D.
(Chest. 2006;129:1167-1175.)
© 2006 American College of Chest Physicians

Argatroban Therapy in Heparin-Induced Thrombocytopenia With Hepatic Dysfunction*

Robert L. Levine, MD; Marcie J. Hursting, PhD and David McCollum, MS

* From the University of Texas Health Science Center (Dr. Levine), Houston, TX; Clinical Science Consulting (Dr. Hursting), Austin, TX; and CTI Clinical Trial and Consulting Services (Mr. McCollum), Blue Ash, OH.

Correspondence to: Robert L. Levine, MD, University of Texas School of Medicine at Houston, 6431 Fannin, MSB 7.142, Houston, TX 77030; e-mail: rlevine{at}uth.tmc.edu

Abstract

Study objectives: We evaluated the dosing requirements in argatroban-treated patients with heparin-induced thrombocytopenia (HIT) and hepatic dysfunction, and compared efficacy and safety outcomes with historical control patients.

Design: Retrospective analysis.

Setting: Inpatient setting.

Patients: Patients with hepatic dysfunction, defined as total bilirubin > 25.5 µmol/L (1.5 mg/dL), aspartate aminotransferase >100 IU/L, and/or alanine aminotransferase >100 IU/L, were identified from previous multicenter, historical-controlled studies of argatroban therapy in HIT.

Interventions: Argatroban, adjusted to maintain activated partial thromboplastin times (aPTTs) 1.5 to 3 times baseline in the experimental group, vs no direct thrombin inhibition in the historical control patients.

Measurements and results: The analysis population included 82 argatroban patients and 34 historical control patients with hepatic impairment, of whom approximately 50% in each group had renal dysfunction (defined as a serum creatinine level > 1.3 mg/dL). The argatroban dosage was 1.6 ± 1.1 µg/kg/min (mean ± SD) over a mean 5-day course of therapy. Significantly lower doses were used in patients with elevated vs normal total bilirubin levels (0.8 ± 0.6 µg/kg/min vs 1.7 ± 0.8 µg/kg/min, p = 0.0063) and in patients with hepatic/renal dysfunction vs hepatic dysfunction alone (1.2 ± 1.1 µg/kg/min vs 2.0 ± 1.1 µg/kg/min, p < 0.001). The aPTT 24 h after argatroban initiation was 69 ± 22 s, with 80% of patients having a therapeutic level of anticoagulation. Thirty-four argatroban-treated patients (41.5%) and 17 control patients (50.0%) experienced the 37-day composite end point of death, amputation, or new thrombosis (p = 0.32). Argatroban significantly reduced new thrombosis (8.5% vs 26.5%, p = 0.012). Major bleeding was similar between treatment groups (4.9% vs 2.9%, p = 0.684).

Conclusions: Hepatic dysfunction affects argatroban dosing, with reduced doses required particularly in patients with serum total bilirubin levels > 25.5 µmol/L (1.5 mg/dL) or combined hepatic/renal dysfunction. Individual mean aPTT-adjusted doses typically remain ≥ 0.5 µg/kg/min, supporting the recommendation of 0.5 µg/kg/min as a conservative initial dose for most patients with hepatic impairment. Argatroban, with proper initial dosing and monitoring, can provide safe and effective antithrombotic therapy in patients with HIT and hepatic impairment.

Key Words: anticoagulants • heparin • hepatic dysfunction • renal dysfunction • thrombocytopenia • thrombosis

Heparin is one of the most frequently prescribed medications, with > 12 million patients receiving heparin annually in the United States for thromboprophylaxis or treatment in a variety of clinical settings.1 Heparin-induced thrombocytopenia (HIT) is a serious, immune-mediated complication of heparin therapy affecting 1 to 5% of treated patients.2 In HIT, antibodies develop to a complex of heparin and platelet factor 4, and these antibodies cause intense platelet activation, leading to thrombocytopenia, a hypercoagulable state, and often thrombosis. Thromboembolic complications, sometimes fatal, occur in 38 to 76% of patients with HIT unless alternative anticoagulation is initiated.3 According to consensus guidelines, when HIT is diagnosed or strongly suspected, whether or not complicated with thrombosis, all heparins should be immediately discontinued and a rapidly acting, nonheparin anticoagulant such as argatroban should be initiated.4 In multicenter, historical-controlled studies,56 argatroban improved outcomes in patients with HIT, particularly reducing new thrombosis, without increasing bleeding.

Argatroban is a direct thrombin inhibitor derived from L-arginine that selectively and reversibly inhibits thrombin, both clot-bound and free, at the catalytic site. Argatroban is primarily metabolized in the liver and eliminated in the feces through biliary excretion,7 and its elimination half-life is 39 to 51 min in healthy subjects.8 In a special population study of 5 patients with moderate hepatic impairment (Child-Pugh score > 6), compared with 12 healthy volunteers receiving argatroban at 2.5 µg/kg/min, hepatic impairment was associated with approximately fourfold decreased systemic clearance and threefold increased in elimination half-life of argatroban.8 Furthermore, anticoagulant responses returned to baseline within 2 to 4 h of argatroban cessation in the healthy subjects, compared with at least 6 h (and up to 20 h) in patients with hepatic impairment. Also, on the basis of altered pharmacokinetics, patients with hepatic impairment would be expected to more slowly reach steady-state levels of drug and anticoagulant effects. Renal elimination of argatroban is minimal, and the pharmacodynamic and pharmacokinetic parameters of argatroban are comparable between healthy subjects and individuals with renal insufficiency.8 The recommended initial dose of argatroban for the prophylaxis or treatment of thrombosis in HIT is 2 µg/kg/min, adjusted as needed to achieve activated partial thromboplastin times (aPTTs) of 1.5 to 3 times the patient’s baseline aPTT. To account for the estimated fourfold reduction in clearance in hepatic impairment, the recommended initial dose for patients with hepatic impairment is 0.5 µg/kg/min. Despite availability of this recommendation, the dosing patterns, efficacy, and safety of argatroban therapy in patients with HIT and hepatic disease have not yet been well described.

The primary objective of this retrospective analysis was to evaluate argatroban dosing, including anticoagulation responses, in a large cohort of argatroban-treated HIT patients with hepatic disease, either with or without impaired renal function. A secondary objective was to assess the effect of argatroban therapy, compared with no antithrombin therapy, in these patients.

Materials and Methods

Study Population
We retrospectively analyzed argatroban dosing patterns, anticoagulant response, and safety and efficacy of argatroban therapy in HIT patients with hepatic dysfunction, either with or without renal dysfunction. Our analysis population included patients who were enrolled in the previous multicenter, historical-controlled argatroban-911 and argatroban-915 studies56 of argatroban in HIT. These trials, in brief, enrolled adults who had HIT, defined as an unexplained 50% decrease in the platelet count from baseline or an absolute decrease in the platelet count to < 100 x 109/L while receiving heparin therapy, either with or without thrombosis. Historical control patients were patients who met the same inclusion and exclusion criteria on chart review and who were seen up to 4 years before initiation of the argatroban-911 study. Institutional Review Board approval was obtained at each center, and patients gave informed consent.

Patients and historical control patients from the prospective studies were eligible for this analysis if they had hepatic dysfunction, which we prospectively defined as a total serum bilirubin level > 25.5 µmol/L (1.5 mg/dL), aspartate aminotransferase (AST) level > 100 IU/L, or alanine aminotransferase (ALT) level > 100 IU/L, at baseline (ie, the date that argatroban was initiated for the prospectively treated patients or for historical control patients, that heparin was discontinued after the platelet count met inclusion criteria, or that the count met inclusion criteria after heparin initiation). Patients were also stratified according to the presence or absence of comorbid renal impairment, which we prospectively defined as a serum creatinine level ≥ 115 µmol/L (1.3 mg/dL) immediately before HIT diagnosis.

Anticoagulant Dosing and Monitoring
When HIT was diagnosed in the prospectively treated patients, heparin was discontinued and argatroban (GlaxoSmithKline; Philadelphia, PA) was initiated as a continuous infusion at 2 µg/kg/min. A lower dose was allowed if medically necessary (such as in hepatic dysfunction) at the investigator’s discretion. The aPTT was measured 2 h later, and dose adjustments up to 10 µg/kg/min were made to maintain aPTTs at 1.5 to 3 times the baseline value. The aPTT was assessed daily and 2 h after any dose adjustment. Argatroban was continued until the underlying condition resolved, appropriate anticoagulation was provided with other agents, or for 14 days. For most patients, each dose, its respective duration, and associated aPTT values were recorded. However, the argatroban-915 study had an extension phase that allowed continued availability of argatroban to patients needing the drug during its regulatory review; during the extension phase, argatroban dosing information and aPTTs were not routinely recorded.

Historical control patients received care according to the local standard of practice at the time of HIT diagnosis. They did not receive direct thrombin inhibition therapy but instead received rather typical treatments of heparin discontinuation with or without the initiation of warfarin anticoagulation.

Clinical Outcomes
During the prospective studies, argatroban-treated patients were followed up during therapy and for 30 days after argatroban cessation, and control patients were followed up for 37 days after baseline. Death, including death due to thrombosis; amputation, including amputation secondary to ischemic complications of HIT; new thrombosis; and major bleeding were recorded. Thrombotic events were identified clinically and/or by objective testing (duplex Doppler examination, ventilation/perfusion scan, or other imaging techniques). The cause of death or amputation was clinically determined by the physician investigator. Major bleeding was defined as overt and associated with a hemoglobin decrease of at least 1.3 mmol/L (2 g/dL) that led to a transfusion of at least 2 U or that was intracranial, retroperitoneal, or into a prosthetic joint. Efficacy outcomes included an all-cause composite end point of death, amputation, or new thrombosis within 37 days of baseline; a thrombosis-related composite end point of death due to thrombosis, amputation secondary to HIT, or new thrombosis within 37 days of baseline; and the individual components of the composite end points.

Data Analysis and Statistics
Statistical analysis was performed using statistical software (Version 8.2; SAS Institute; Cary, NC). Statistical tests were two sided and conducted at the 0.05 level of significance. For patients (overall and stratified by renal function) in the argatroban and control groups, categorical data (gender, race, and outcomes) were summarized using counts and percentages, and continuous data (age, serum chemistry data, dosing parameters, and aPTT responses) were summarized using descriptive statistics. Comparisons were made between the argatroban and control groups using analysis of variance or Cochran-Mantel-Haenszel test, controlling each for renal function category. Among argatroban-treated patients, comparisons were made between hepatic dysfunction vs hepatorenal dysfunction using Fisher Exact Test or the Wilcoxon rank-sum test. Argatroban dosing requirements were also evaluated for patients stratified by their total serum bilirubin level at baseline (≤ 1.5 mg/dL or > 1.5 mg/dL). Regression analysis was performed to analyze the effect of baseline serum total bilirubin (up to 5 mg/dL), AST, and ALT on the mean argatroban dose. The mean aPTT over the course of argatroban therapy was calculated for each patient by weighting each aPTT value by the time between consecutive assessments. The number and percentage of patients with aPTTs < 45 s, 45 to 90 s, or > 90 s at 24 h after initiation of argatroban was determined. A "therapeutic" aPTT was defined as 45 to 90 s, assuming a typical baseline aPTT of 30 s in the absence of anticoagulation.

Results

Patients
Our analysis population included 116 patients with HIT and hepatic dysfunction who received either argatroban (n = 82) or historical control (n = 34) therapy. Combined hepatic/renal dysfunction was present in 43 argatroban-treated patients (52%) and 16 control patients (47%). Argatroban-treated patients with hepatic dysfunction vs hepatic/renal dysfunction had similar demographic and baseline laboratory characteristics, except for serum creatinine levels, which were greater in patients with renal dysfunction (Table 1 ). The baseline characteristics of the argatroban and control groups were also similar, when controlling for renal function, with the exception of race distribution. There were no statistically significant differences between the argatroban and control groups overall, or between the renal function categories within each group, in AST or ALT levels. The indications for heparin were also comparable between groups. Overall and within each group, the most frequent indications for heparin included coronary artery bypass graft surgery (16 argatroban-treated patients, 11 control patients), unspecified surgery (12 argatroban-treated patients, 7 control patients), and line or filter patency (17 argatroban-treated patients, 5 control patients). There were no statistically significant differences between groups in ongoing medical conditions at study entry. By exception, the argatroban-treated patients, vs control patients, had significantly more COPD (17% vs 0%) and pulmonary embolism (15% vs 0%). Coronary artery disease (29 argatroban-treated patients, 7 control patients), atrial fibrillation (14 argatroban-treated patients, 5 control patients), and peripheral vascular disease (14 argatroban-treated patients, 3 control patients) occurred most commonly overall. Cancer was reported in four argatroban-treated patients (5%) and two control patients (6%).


View this table:
[in this window]
[in a new window]

  		Table 1. Baseline and Demographic Characteristics*

 
Before HIT was diagnosed, the patients had received at least 1 day and typically 4 to 8 days of heparin therapy. Unfractionated heparin therapy was the leading cause of HIT, and only one patient (in the argatroban group) had received low-molecular-weight heparin. Baseline platelet counts in the argatroban and control groups (mean values of 72 x 109/L and 87 x 109/L, respectively) were consistent with the clinical presentation of HIT. Serologic confirmation of HIT, which was not required for initiating treatment and often not recorded, occurred in 11 of 22 argatroban-treated patients (50%) with available data and 27 of 34 control patients (79%).

Argatroban Dosing and Anticoagulant Response
Patients with hepatic dysfunction, irrespective of renal function, were administered argatroban therapy at a mean (and median) dose of 1.6 µg/kg/min for a mean duration of 4.9 days (Table 2 ). Individuals with hepatic/renal dysfunction compared with hepatic dysfunction alone received significantly lower doses of argatroban initially (mean doses of 1.4 µg/kg/min vs 1.9 µg/kg/min, p = 0.009), over the course of therapy (1.2 µg/kg/min vs 2.0 µg/kg/min, p < 0.001), and at the end of therapy (1.2 µg/kg/min vs 1.9 µg/kg/min, p = 0.001). Overall, individual mean doses ranged from 0.1 to 8.0 µg/kg/min.


View this table:
[in this window]
[in a new window]

  		Table 2. Summary of Argatroban Therapy and aPTT Response*

 
Within a median 3.8 h of initiating therapy, a therapeutic aPTT of 45 to 90 s was achieved in each (100%) of 70 argatroban-treated patients with data available. The aPTT was also in the therapeutic range 24 h after therapy was initiated in 48 of 60 patients (80%) with data available, with no significant difference between renal function categories (Table 2). In argatroban-treated patients with international normalized ratio (INR) data, which were not routinely recorded, the median INR was 1.5 (interquartile range, 1.2 to 2.3) before argatroban initiation (n = 52), increasing to 3.4 (interquartile range, 2.7 to 6.5) by 24 ± 12 h after argatroban initiation (n = 40). Over the course of argatroban therapy and irrespective of renal function, patients with hepatic impairment achieved a mean aPTT of 65 s in association with a mean 4.8 dose adjustment (translating to approximately 1 dose adjustment per day of therapy), and platelet counts recovered substantially to a mean of 176 ± 167 x 109/L. The median time to the first aPTT assessment after argatroban discontinuation was 4.8 h, at which point the median aPTT was 47.1 (interquartile range, 36.5 to 58.8 s).

Thirty-four patients had baseline total serum bilirubin values as well as dosing information available. Among seven patients whose mean dose was < 0.5 µg/kg/min, six patients had a baseline serum total bilirubin level > 1.5 mg/dL, and all had hepatic/renal dysfunction (Fig 1 ). Patients with bilirubin levels > 1.5 mg/dL (n = 16, vs n = 18 with ≤ 1.5 mg/dL) received significantly lower argatroban doses during therapy (mean doses, 0.8 ± 0.6 µg/kg/min vs 1.7 ± 0.8 µg/kg/min, p = 0.006) and at the end of therapy (0.7 ± 0.7 µg/kg/min vs 1.5 ± 0.9 µg/kg/min, p = 0.016), regardless of renal function, and underwent more dose adjustments (6.6 vs 3.7, p = 0.095). Mean aPTTs over the course of therapy were 70 ± 22 s for patients with elevated bilirubin and 61 ± 14 s for patients without elevated bilirubin (p = 0.23). Regression analysis of individual patient results showed that dose requirements significantly decreased with increasing bilirubin (up to 5 mg/dL) in patients with or without renal dysfunction (Fig 1). By contrast, there was no significant relationship between a patient’s mean argatroban dose and baseline ALT (n = 61) or AST (n = 70) levels (data not shown).


Figure 1
View larger version (13K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Relationship between baseline serum total bilirubin and mean argatroban dose for patients with hepatic dysfunction (squares) or hepatic/renal dysfunction (diamonds). Linear regression analyses were performed on data from patients with bilirubin levels < 5 mg/dL (results shown in insert), and the best fit equations were y = – 0.50 x + 2.37 for hepatic dysfunction (n = 14, r2 = 0.29), and y = – 0.38 x + 1.70 for hepatic/renal dysfunction (n = 17; r2 = 0.32).

 
Outcomes
Thirty-four argatroban-treated patients (41.5%) and 17 control patients (50.0%) experienced the composite end point of death, amputation, or new thrombosis within 37 days of baseline (p = 0.32) [Table 3 ]. The thrombotic composite end point of death due to thrombosis, amputation secondary to HIT, or new thrombosis within 37 days occurred in 12 argatroban-treated patients (14.6%), compared with 10 control patients (29.4%) [p = 0.06]. Argatroban therapy significantly reduced the incidence of new thrombosis (8.5% vs 26.5%, p = 0.012). A total of 16 patients, including 7 patients receiving argatroban and 9 control patients, acquired new thrombosis, typically within 5 to 7 days after the diagnosis of HIT. Known arterial events included stroke in argatroban-treated patients 17 days after HIT was diagnosed (and 13 days after argatroban was discontinued), and an acute myocardial infarction in a control patient 3 days after the diagnosis of HIT; the site of venous events were not routinely specified. There were no significant between-group differences in death from any cause, death due to thrombosis, amputations of any cause, or amputation secondary to HIT. Mortality was significantly greater in patients with hepatic/renal dysfunction than patients with hepatic dysfunction alone in the argatroban group (41.9% vs 15.4%, p = 0.014) and control patients (50.0% vs 11.1%, p = 0.017).


View this table:
[in this window]
[in a new window]

  		Table 3. Incidence of Outcomes (37-Day Study Period)*

 
Four argatroban-treated patients (4.9%), each with hepatic/renal dysfunction, had major bleeding, most often at a surgical procedure wound site (n = 3). The bleeding occurred in three patients while receiving argatroban and in one patient on the day after argatroban cessation. In each of three patients with dosing information available, argatroban was initiated at 2 µg/kg/min and subsequently adjusted downward to 0.08 µg/kg/min, 0.2 µg/kg/min, and 0.6 µg/kg/min, respectively. One control patient (2.9%) had major bleeding (p = 0.684).

Discussion

Patients of any age receiving any dose or type of heparin by any route can acquire HIT. The ubiquitous use of heparins taken together with the high risk of thromboembolic complications in patients with HIT makes HIT one of the most important adverse drug reactions physicians face today.1

To treat HIT, all heparins should be discontinued immediately and an alternative anticoagulant such as argatroban should be initiated.4 Because argatroban is primarily hepatically metabolized,7 the use of argatroban may theoretically be a concern when treating HIT patients with hepatic impairment, unless adequate levels of anticoagulation can be achieved and maintained safely. We investigated the effects of hepatic and hepatic/renal dysfunction on argatroban dosing and anticoagulation response in 82 patients with HIT, and compared their clinical outcomes with those of 34 historical control patients who did not receive direct thrombin inhibition therapy. The study patients and control patients were identified from previous multicenter, historical-controlled studies56 of argatroban anticoagulation for prophylaxis or treatment of thrombosis in HIT. The two groups studied in this report were generally well matched in terms of demographics, total serum bilirubin, indications for heparin therapy, and ongoing medical conditions. Although mean liver enzyme levels appeared greater in the control patients than argatroban-treated patients, differences were not statistically significant. The severity of the thrombocytopenia and the frequency of comorbid cancer, which are independent predictors of thrombosis or poor outcomes in HIT,910 were also comparable as well between groups. Overall, therefore, the argatroban-treated patients and historical control patients were similar in baseline characteristics and acuity of illness.

In our analysis, hepatic dysfunction led to reduced argatroban dosing requirements for the treatment of HIT. Patients in the prospective studies were administered mean argatroban doses of 1.7 to 2.0 µg/kg/min for 5 to 7 days56 and underwent a median of 3.0 dosage adjustments during the course of therapy.11 We found that argatroban-treated patients with hepatic dysfunction, irrespective of renal function, received a mean dose of 1.6 µg/kg/min over a mean of 5 days and underwent a median of 4.0 dose adjustments. This finding seems to contradict the need of lower doses of argatroban in patients with hepatic malfunction. However, when bilirubin is used as the indicator for hepatic malfunction, it becomes clear that patients with total serum bilirubin of at least 25.5 µmol/L (1.5 mg/dL) at baseline required significantly lower argatroban doses (mean, 0.8 µg/kg/min), compared with those with normal total bilirubin, regardless of renal function. Regression analysis of by-patient data confirmed that required doses decreased with increasing serum bilirubin, regardless of renal function. There was not, however, a significant relationship between individual mean argatroban doses and either AST or ALT. Therefore elevated bilirubin may be a better indicator of the need for argatroban dosing adjustment than either ALT or AST.

Consistent with these findings is the observation that of the five patients with Child-Pugh scores > 6 and reduced argatroban clearance from a previous study,8 four patients had elevated bilirubin > 25.5 µmol/L (1.5 mg/dL) and one patient had renal impairment with serum creatinine of 220 µmol/L (2.5 mg/dL). Furthermore, case reports1213 have described HIT patients with acute hepatic dysfunction and comorbid renal insufficiency who required reduced doses to maintain therapeutic aPTTs. Clinical data suggest that renal elimination of argatroban is minimal, and renal impairment does not affect argatroban dosing.814 The need for reduced argatroban doses in patients with hepatic/renal dysfunction vs hepatic dysfunction alone may reflect the acuity of patients’ health when other factors like sepsis, multiple organ failure, and cardiovascular instability affect the dosage of argatroban. Baghdasarian et al15 expressed similar interpretation in a study of argatroban dosage in critically ill patients.

In the current study, despite the dose reductions needed in hepatic impairment, particularly in association with elevated bilirubin or hepatic/renal dysfunction, individual mean doses typically remained ≥ 0.5 µg/kg/min (median doses of 1.6 µg/kg/min overall and 0.9 µg/kg/min in the presence of hepatic/renal impairment). At these dosages, argatroban was effective in maintaining adequate anticoagulation, with the majority of patients (80%) maintaining therapeutic aPTTs at 24 h after initiation of therapy. The median time to a therapeutic aPTT was approximately 4 h. By indirect comparison, steady-state levels of anticoagulant effect are typically attained within 1 to 3 h of initiating argatroban in healthy subjects,8 and therapeutic aPTTs occurred a median 2.9 h after initiating argatroban, 2 µg/kg/min, in patients with HIT, irrespective of their hepatic or renal function status, in a prospective study.5 We also found that the reversal of anticoagulant effects following argatroban discontinuation was delayed, by indirect comparison, relative to the 2 to 4 h required for reversal in healthy subjects.8 Specifically, at a median of approximately 5 h after argatroban cessation, more than half of the patients with hepatic dysfunction still had an aPTT > 47 s; that is, a subtherapeutic level of anticoagulation was not yet present.

The clinical outcomes in this cohort of patients with hepatic impairment were generally comparable with those in patients from the previously published studies56 from which this cohort was identified. In a meta-analysis3 of the prospective studies, the all-cause composite end point of death, amputation, or new thrombosis occurred in approximately 35% of patients with HIT (with or without associated thrombosis) treated with argatroban and 43% of historical control patients, regardless of hepatic function status. By indirect comparison, in our cohort of patients, the all-cause composite end point occurred in 42% of argatroban-treated patients and 50% of historical control patients, and thrombosis-related outcomes occurred in 14% of argatroban-treated patients and 29% of control patients. In each treatment group, death occurred approximately three times more frequently in patients with hepatic-renal dysfunction than hepatic dysfunction alone, again reflective of the acuity of these patients’ health status. Argatroban therapy, compared with control patients, significantly reduced new thrombosis in patients with hepatic dysfunction, without significantly increasing bleeding.

With the availability of direct thrombin inhibitors such as argatroban, warfarin should not be used as a sole therapy for acute HIT. Rather, in patients with HIT requiring long-term anticoagulation, consensus guidelines recommend that warfarin should be initiated at the expected maintenance dose only after adequate alternative parenteral anticoagulation has been provided and platelet counts have substantially recovered, and then overlapped with the alternative anticoagulant for at least 5 days.4 Careful monitoring may be needed because direct thrombin inhibitors prolong the INR, with this effect being particularly pronounced for argatroban.16 Previously established relationships regarding bleeding risk and INRs during warfarin therapy are not fully applicable during direct thrombin inhibition. INRs > 5 commonly occur without bleeding complications during argatroban therapy or argatroban-warfarin co-therapy in HIT.1718 Guidelines are published21 for monitoring the transition from argatroban to coumarins, including warfarin, using the INR1920 or chromogenic factor Xa assay.

The retrospective nature of our analysis is a limitation. However, all study data, including outcomes and patient characteristics, were from prospective, multicenter clinical studies56 of argatroban therapy in HIT conducted between 1995 and 1998. In those studies,56 laboratory confirmation of HIT was not required for initiating treatment, which simulated the "real world" of managing HIT. In one study,5 57% of the argatroban-treated patients and 77% of the control patients had a positive HIT serologic result, and the remaining patients were not tested or had negative results. Of the patients in our analysis with available data, 50% in the argatroban group and 79% in the control group had a positive test result. Patients were to be excluded from the prospective studies if they had a documented coagulation disorder or bleeding diathesis unrelated to HIT, or their prothrombin time was > 16 s (no criterion specified for the INR) in the absence of warfarin therapy. However, we found that INRs at study entry of ≥ 1.5 occurred in 50% of our patients with available data. Our analyses are limited by small sample sizes; overall, however, this is the largest cohort of HIT patients with hepatic impairment (n = 116) yet reported.

Although the Child-Pugh score and creatinine clearance have been used to define hepatic impairment and renal impairment in a previous study8 of argatroban pharmacodynamics and pharmacokinetics, we prospectively chose to use conservative, laboratory-based definitions of hepatic and kidney dysfunction for this study. These definitions allow for evaluation of argatroban dosing requirements using routine and accepted clinical laboratory measures of hepatic and kidney dysfunction. Because the upper normal values of ALT and AST are 45 IU/L and 34 IU/L, respectively,22 values that are > 100 IU/L are at least 2 times to 2.5 times the upper limit, respectively. Values near 100 IU/L are associated with a variety of parenchymal and biliary liver diseases.22 Although elevated liver function test results, including the transaminases and bilirubin, are not specific and usually represent a wide spectrum of hepatic dysfunction, they are commonly used as a practical, sensitive, and noninvasive method to assess hepatic dysfunction.2223 The upper limit of the reference interval for serum creatinine has been reported to be 1.1 mg/dL and 0.92 mg/dL for men and women, respectively, at 19 to 67 years of age.24 Among European-Americans in the Modification of Diet in Renal Disease study,2 serum creatinine levels of 1.23 to 1.47 mg/dL in men and 0.95 to 1.13 mg/dL in women corresponded with glomerular filtration rates ≤ 60 mL/min/1.73 m that are indicative of high risk for chronic renal failure.25 Although there are more precise and accurate methods to assess renal function like creatinine clearance or glomerular filtration rate, in the setting of our study, we believe that a serum creatinine level > 1.3 mg/dL was a useful clinical tool for identifying patients with kidney dysfunction. Acknowledging that elevated liver function tests are not very specific and that a serum creatinine level of 1.3 mg/dL is at the lower end of abnormal creatinine values, we may have overestimated the number of patients with hepatic or hepatic/renal impairment. Baseline bilirubin values and dosing information were available in 34 of 82 argatroban-treated patients (41%) with hepatic impairment; liver enzymes were more commonly reported.

Historical control patients, which were used as a comparator group in the prospective studies, were also used for comparisons of outcomes in our analysis, and outcomes were not centrally adjudicated. The limitations associated with historical control comparator groups and approaches used to minimize potential sources of bias in the prospective studies have been previously described.56

Conclusions

These data support the current recommendation of 0.5 µg/kg/min in patients with hepatic impairment as a reasonable, conservative initial dosage of argatroban, and indicate that argatroban, vs control, provides effective antithrombotic therapy in patients with hepatic dysfunction and HIT, without significantly increasing bleeding. We also offer the following refinements for argatroban dosing and monitoring in this setting: (1) physicians should take into consideration the patient’s total serum bilirubin level, which is a better indicator than ALT or AST of argatroban dosing requirements, before starting argatroban therapy; (2) argatroban should be initiated at a dose of 0.5 µg/kg/min if a patient’s serum total bilirubin level is > 25.5 µmol/L (1.5 mg/dL), if combined hepatic/renal dysfunction is present, or if conservative initial dosing is desired; (3) because achievement of steady-state anticoagulation will be delayed in many patients with hepatic dysfunction, it would be prudent to check the aPTT at least 4 to 5 h after drug initiation or dose change to ensure the desired level of anticoagulation is present; and (4) because coagulation parameters will take longer to normalize on infusion cessation in patients with hepatic dysfunction (> 5 h in many patients in this study and up to 20 h in a previous study8), argatroban should be stopped for a more extended period and anticoagulant effects should be monitored carefully before an invasive procedure to ensure adequate hemostatic control.

Acknowledgements

The authors thank Karissa Kim, PharmD, for editorial assistance, and Eliezer Katz, MD, for critical comments on this article.

Footnotes

Abbreviations: ALT = alanine aminotransferase; aPTT = activated partial thromboplastin time; AST = aspartate aminotransferase; HIT = heparin-induced thrombocytopenia; INR = international normalized ratio

Dr. Levine has received grant support and honorarium from GlaxoSmithKline. Dr. Hursting has received consultancy fees from GlaxoSmithKline.

This study was supported by GlaxoSmithKline.

Received for publication July 22, 2005. Accepted for publication October 25, 2005.

References

  1. Jang, IK, Hursting, MJ (2005) When heparins promote thrombosis: review of heparin-induced thrombocytopenia. Circulation 111,2671-2683[Free Full Text]
  2. Warkentin, TE, Sheppard, JI, Horsewood, P, et al Impact of the patient population on the risk for heparin-induced thrombocytopenia. Blood 2000;96,1703-1708[Abstract/Free Full Text]
  3. Hirsh, J, Heddle, N, Kelton, JG Treatment of heparin induced thrombocytopenia: a critical review. Arch Intern Med 2004;164,361-369[Abstract/Free Full Text]
  4. Warkentin, TE, Greinacher, A Heparin-induced thrombocytopenia: recognition, treatment, and prevention. Chest 2004;126(suppl),311S-337S[Abstract/Free Full Text]
  5. Lewis, BE, Wallis, DE, Berkowitz, SD, et al Argatroban anticoagulant therapy in patients with heparin-induced thrombocytopenia. Circulation 2001;103,1838-1843[Abstract/Free Full Text]
  6. Lewis, BE, Wallis, DE, Leya, F, et al Argatroban anticoagulation in patients with heparin-induced thrombocytopenia. Arch Intern Med 2003;163,1849-1856[Abstract/Free Full Text]
  7. Izawa, O, Katsuki, M, Komatsu, T, et al Pharmacokinetic studies of argatroban (MD-805) in human: concentrations of argatroban and its metabolites in plasma, urine and feces during and after drip intravenous infusion. Jpn Pharmacol Ther 1986;14,251-263
  8. Swan, SK, Hursting, MJ The pharmacokinetics and pharmacodynamics of argatroban: effects of age, gender and hepatic or renal dysfunction. Pharmacotherapy 2000;20,318-329[CrossRef][ISI][Medline]
  9. Kelton, JG, Hursting, MJ, Lewis, BE The predictors of clinical outcome in patients with heparin-induced thrombocytopenia [abstract].J Thromb Haemost 2003;1(supp1),OC020
  10. Opatrny, L, Warner, MN Risk of thrombosis in patients with malignancy and heparin-induced thrombocytopenia. Am J Hematol 2004;76,240-244[CrossRef][ISI][Medline]
  11. Verme-Gibboney, CN, Hursting, MJ Argatroban dosing in patients with heparin-induced thrombocytopenia. Ann Pharmacother 2003;37,970-975[Abstract/Free Full Text]
  12. Dager, WE, White, RH Argatroban for heparin-induced thrombocytopenia in hepato-renal failure and CVVHD. Ann Pharmacother 2003;37,1232-1236[Abstract/Free Full Text]
  13. Williamson, DR, Boulanger, I, Tardif, M, et al Argatroban dosing in intensive care patients with acute renal failure and liver dysfunction. Pharmacotherapy 2004;24,409-414[CrossRef][ISI][Medline]
  14. Murray, PM, Reddy, BV, Grossman, EJ, et al A prospective comparison of three argatroban treatment regimens during in end-stage renal disease. Kidney Int 2004;66,2446-2453[CrossRef][ISI][Medline]
  15. Baghdasarian, SB, Singh, I, Militello, M, et al Argatroban dosage in critically ill patients with HIT [abstract].Blood 2004;104,1779
  16. Gosselin, RC, Dager, WE, King, JH, et al Effect of direct thrombin inhibitors, bivalirudin, lepirudin, and argatroban, on prothrombin time and INR values. Am J Clin Pathol 2004;121,593-599[CrossRef][ISI][Medline]
  17. Hursting, MJ, Lewis, BE, Macfarlane, DE Transitioning from argatroban to warfarin therapy in patients with heparin-induced thrombocytopenia. Clin Appl Thromb Haemost 2005;11,279-287
  18. Bartholomew, JR, Hursting, MJ Transitioning from argatroban to warfarin in heparin-induced thrombocytopenia: an analysis of outcomes in patients with elevated international normalized ratio (INR). J Thromb Thrombolysis 2005;19,183-188[Medline]
  19. Sheth, SB, DiCicco, RA, Hursting, MJ, et al Interpreting the international normalized ratio (INR) in individuals receiving argatroban and warfarin. Thromb Haemost 2001;85,435-440[ISI][Medline]
  20. Harder, S, Graff, J, Klinkhardt, U, et al Transition from argatroban to oral anticoagulation with phenprocoumon or acenocoumarol: effects on prothrombin time, activated partial thromboplastin time, and ecarin clotting time. Thromb Haemost 2004;91,1137-1145[ISI][Medline]
  21. Arpino, PA, Demirjian, Z, Van Cott, EM Use of the chromogenic factor X assay to predict the international normalized ratio in patients transitioning from argatroban to warfarin. Pharmacotherapy 2005;25,157-164[CrossRef][ISI][Medline]
  22. Pratt, DS, Kaplan, MM Evaluation of the liver: A. Laboratory tests. Schiff, ER Sorrell, MF Maddrey, WC eds. Schiff’s diseases of the liver 1999,218-221 Lippincott, Williams and Wilkins. Philadelphia, PA:
  23. Rothschild, HM, Khorasani, R, Silverman, S, et al Abdominal cross-sectional imaging for inpatients with abnormal liver function test results. Arch Intern Med 2001;161,583-588[Abstract/Free Full Text]
  24. Finney, H, Newman, DJ, Price, P Adult reference ranges for serum cystatin C, creatinine and predicted creatinine clearance. Ann Clin Biochem 2000;37,49-59
  25. Levey, AS, Coresh, J, Balk, E, et al National Kidney Foundation practice guidelines for chronic kidney disease: evaluation and stratification. Ann Intern Med 2003;139,137-147[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 The Annals of PharmacotherapyHome page
M. Beiderlinden, T. A Treschan, K. Gorlinger, and J. Peters
Argatroban Anticoagulation in Critically Ill Patients
Ann. Pharmacother., May 1, 2007; 41(5): 749 - 754.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Levine, R. L.
Right arrow Articles by McCollum, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Levine, R. L.
Right arrow Articles by McCollum, D.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS