Context.—Direct thrombin inhibitors (DTIs) and fondaparinux represent a new class of anticoagulants. The effects of DTIs on activated partial thromboplastin time and prothrombin time measurements have been reported previously, but there are limited data on the impact of these anticoagulants on other coagulation tests.

Objective.—To determine the effects of fondaparinux and 3 DTIs (argatroban, bivalirudin, and lepirudin) on miscellaneous coagulation tests.

Design.—Bivalirudin, lepirudin, argatroban, and fondaparinux were added to pooled normal plasma and tested for fibrinogen, antithrombin (thrombin and Xa substrate methods), plasminogen, protein C (clot and chromogenic methods), protein S, von Willebrand factor, D-dimer, lupus anticoagulant testing (dilute Russell viper venom test [DRVVT] with ratio), and factors II, IX, and X activities.

Results.—We found no drug interference on antithrombin, plasminogen, chromogenic protein C, von Willebrand factor, or D-dimer results. All DTIs falsely decreased fibrinogen values, while falsely increasing protein C and protein S levels. All DTIs prolonged the DRVVT, and only argatroban yielded DRVVT ratios less than 1.2. Lepirudin demonstrated no effect on factor II activity, and only argatroban demonstrated decreased factor X activity. All DTI samples demonstrated a linear, dose-dependent, false decrease of factor IX activity.

Conclusions.—Using in vitro methods, we demonstrated DTI effects on numerous clot-based assays, but we found no interference with latex agglutination, chromogenic, or platelet aggregation methods. Fondaparinux only affected measurement of protein S activity. Caution must be used when interpreting coagulation test results on patients receiving these drugs.

Direct thrombin inhibitors (DTIs) represent a new class of anticoagulants that inhibit thrombin independently of antithrombin. Patients with documented heparin-induced thrombocytopenia or high suspicion for heparin-induced thrombocytopenia, may also be acutely anticoagulated with DTIs until long-term anticoagulation with warfarin is achieved. Similar to heparin, DTI pharmacodynamics are monitored by the activated partial thromboplastin time (aPTT), with a target aPTT ratio from 1.5 to 2.5 or 3.0.1 

Fondaparinux is another new anticoagulant consisting of a pentasaccharide chain of glycosaminoglycans, which selectively inhibits activated factor X via the antithrombin protein.2 Present use of fondaparinux is primarily for thromboprophylaxis, but it may also be efficacious in the treatment of acute thromboembolism. The pharmacodynamic anticoagulation effects of fondaparinux are not routinely monitored and have a minimal impact on the aPTT at therapeutic doses.

The presence of an anticoagulant may interfere with clot-based assays. There have been several reports on the effects of argatroban on international normalized ratio (INR) determinations, but these data are limited to a few prothrombin time (PT) reagents.3–5 We recently described the effects of direct thrombin inhibitors on INR using multiple reagents.6 Data on the effects of these novel anticoagulants on other coagulation tests are also limited.7,8 Interpretation of clot-based assays on patients receiving either class of these drugs should be carried out with caution until any form of interference is determined. In this study, we tested the effects of pooled normal plasma treated with various concentrations of these novel anticoagulants and measured various coagulation tests that are commonly used for diagnosis and/or treatment.

Bivalirudin (Angiomax; Ben Venue Laboratories, Bedford, Ohio), lepirudin (Refludan; Berlex Laboratories, Wayne, NJ), and argatroban (Texas Biotechnology Laboratory, Houston, Tex) were prepared according to the manufacturers' instructions. Single-dose (5 mg/mL), prefilled syringes of fondaparinux (Arixtra; Sanofi-Synthlabo, Paris, France; NV Organon, OSS, the Netherlands) did not require additional preparation. Each drug was added to a pooled normal plasma preparation (from at least 20 normal donors) to achieve concentrations of 0.1, 0.2, 0.4, 0.6, 0.8, 1.0, and 1.2 μg/mL. These selected drug concentrations were based on previous published studies for DTIs6 and fondaparinux.9 An aliquot of the pooled normal plasma and each drug preparation was frozen at −70°C prior to testing.

Each sample was tested for fibrinogen; antithrombin activity (AT) using thrombin (AT-IIa) or factor Xa (AT-Xa) as the substrate; plasminogen activity; protein C activity, using both clotting (PC-Cl) and chromogenic (PC-Chr) methods; protein S activity, von Willebrand factor activity, D-dimer; lupus anticoagulant testing, using dilute Russell viper venom test (DRVVT) with confirmation method; and factors II, IX, and X activities. The methods are further characterized in Table 1. Testing of pooled plasma (baseline) and drug samples were performed using the BCS (Dade Behring, Inc, Deerfield, Ill).

Table 1. 

Methodology of Testing Performed on Samples Prepared With Various Concentrations of Direct Thrombin Inhibitors or Fondaparinux*

Methodology of Testing Performed on Samples Prepared With Various Concentrations of Direct Thrombin Inhibitors or Fondaparinux*
Methodology of Testing Performed on Samples Prepared With Various Concentrations of Direct Thrombin Inhibitors or Fondaparinux*

To determine whether the presence of these agents affected a coagulation test, we first determined the variation of a single normal sample that was run 8 times contemporaneously (replicate precision) and calculated the standard deviation. Since all drug samples were run concurrently for each analyte, variation exceeding 2 SD from the replicate precision for a given test would indicate an analytical bias that was due to drug effect. For those coagulation tests demonstrating a drug effect, a variation greater than 10% from baseline value was considered to be clinically significant. A DRVVT ratio of more than 1.2 indicates a false-positive result for the lupus anticoagulant.

The testing methods used on the BCS were clotting (fibrinogen, PC-Cl, protein S activity, DRVVT, factor II, factor X, and factor IX), chromogenic (PC-Chr, AT-IIa, AT-Xa, and plasminogen activity), platelet aggregation (von Willebrand factor activity), and immunoturbidimetric (D-dimer) (Table 1). The standard deviation for the replicate precision for each parameter is listed in Table 2. For all testing measuring the functional properties of the coagulation protein, the replicate precision was 0.05 U/mL or less. For DRVVT and D-dimer tests, the replicate precision was less than 1.0 second and less than 0.14 mg/L, respectively. Using the replicate precision cutoff of less than 2 SD as indicating no clinically significant drug interference on the test, neither the pentasaccharide nor DTI affected AT-IIa, AT-Xa, plasminogen activity, PC-Chr, or D-dimer testing (Table 2). Fondaparinux also did not affect fibrinogen, PC-Cl, DRVVT, factor II, factor X, or factor IX activities. Fondaparinux only affected protein S activity testing, with a greater than 10% increase of protein S activity from baseline occurring at 0.2 μg/mL.

Table 2. 

Comparison of the Standard Deviations (SDs) Between the Within-Run Precision and Drug Samples for Each Analyte*

Comparison of the Standard Deviations (SDs) Between the Within-Run Precision and Drug Samples for Each Analyte*
Comparison of the Standard Deviations (SDs) Between the Within-Run Precision and Drug Samples for Each Analyte*

For samples containing a DTI, there was some degree of effect on fibrinogen, protein C, protein S, DRVVT, factor II, factor X, and factor IX measurements. For fibrinogen, argatroban and bivalirudin demonstrated a greater than 10% variation, resulting in a decrease of reported values from baseline values at concentrations of 0.4 μg/mL and 0.8 μg/mL, respectively. For lepirudin samples, there was a greater than 2-SD bias noted for the clottable fibrinogen method, but no fibrinogen result exceeded the greater than 10% threshold difference from the baseline value at any drug concentration (Figure, A). Both argatroban and bivalirudin demonstrated a greater than 2-SD variation for fibrinogen, with falsely decreased fibrinogen values reached at drug concentrations of 0.4 μg/mL and 0.8 μg/ mL, respectively. Protein C and protein S were all markedly biased by all DTI samples, with as little as 0.1 μg/ mL of drug (Figure, B and C) resulting in falsely increased results. All DTI samples prolonged the DRVVT test (Figure, D), but only argatroban yielded DRVVT ratios less than 1.2 (Figure, E). For factor activity measurements, only lepirudin demonstrated no deviation from baseline for factor II activity (Figure, F), and only argatroban demonstrated a greater than 10% deviation, resulting in decreased factor X activity from baseline at a 0.8-μg/mL concentration (Figure, H). All DTI samples demonstrated a linear, dose-dependent effect, resulting in a falsely depressed factor IX activity (Figure, G).

Changes seen in a dose-dependent manner of argatroban (closed square), bivalirudin (open triangle), lepirudin (closed triangle), and fondaparinux (open square) on fibrinogen (A), clottable protein C (B), functional protein S (PS) (C), dilute Russell viper venom test (DRVVT) (D), DRVVT ratio (E), prothrombin time (PT)–based factor assays for prothrombin activity (F), and factor X activity (G), and activated partial thromboplastin time (aPTT)-based factor IX activity (H). The increasing concentration of each respective drug on fibrinogen, protein C–clotting (PC-Cl), prothrombin activity (Factor II), factor X activity (Factor X), and factor IX activity (Factor IX) tests are represented by the percent change from baseline (no drug) values obtained from pooled normal plasma

Changes seen in a dose-dependent manner of argatroban (closed square), bivalirudin (open triangle), lepirudin (closed triangle), and fondaparinux (open square) on fibrinogen (A), clottable protein C (B), functional protein S (PS) (C), dilute Russell viper venom test (DRVVT) (D), DRVVT ratio (E), prothrombin time (PT)–based factor assays for prothrombin activity (F), and factor X activity (G), and activated partial thromboplastin time (aPTT)-based factor IX activity (H). The increasing concentration of each respective drug on fibrinogen, protein C–clotting (PC-Cl), prothrombin activity (Factor II), factor X activity (Factor X), and factor IX activity (Factor IX) tests are represented by the percent change from baseline (no drug) values obtained from pooled normal plasma

Close modal

There was a slight increase in the standard deviation for von Willebrand factor activity with bivalirudin, but no linear response was observed with increasing drug concentration, suggesting random analytical error. While the DTI sample standard deviation for antithrombin activity was not higher than replicate precision, there was a trend of increasing antithrombin activity using the thrombin substrate at higher drug concentrations.

Direct thrombin inhibitors are gaining use as an alternative to heparin anticoagulation. Unlike heparin, which requires a mediator (antithrombin) to potentiate anticoagulation, DTIs can inhibit thrombin directly. There are 3 DTIs available for use in the United States. Lepirudin is a recombinant hirudin derived from yeast cells and forms an equimolar complex with thrombin to inhibit thrombin's activity. Argatroban is a synthetic, low-molecular-weight, reversible inhibitor of thrombin derived from l-arginine that binds to thrombin's catalytic site. Bivalirudin is a synthetic inhibitor of both free and bound thrombin, by binding to both the catalytic and anion-binding exosite of thrombin. Fondaparinux is a synthetic inhibitor that is selective for neutralizing factor Xa after forming a complex with antithrombin. Unique differences in each drug may produce different effects on various clot-based assays.

Our results demonstrate the variability of DTIs' effects on coagulation tests. Not surprisingly, modified aPTT coagulation studies, such as clotting protein C, protein S, and factor IX activities are affected by the presence of DTIs. Clottable protein C method is based on snake venom activation of protein C, resulting in the inhibition of thrombin-activated factors VIII and V. Clottable protein S method is based on the addition of activated protein C, which in the presence of protein S, will accelerate the inhibition of thrombin-activated factors VIII and V. For both methods, the prolongation of clotting time is proportional to the amount of factor activity. Therefore, the presence of DTI in plasma prolongs the clot times of these modified aPTT tests. For protein C and protein S, the reported factor activity results in a 2-fold false increase in activity at minimal drug concentrations. Fondaparinux produced biased protein S activity, seen at minimal concentrations (0.2 μg/ mL) of drug. Fondaparinux did not affect other aPTT-based assays (protein C or factor IX).

Prothrombin time–based coagulation studies, such as factor II (prothrombin) and factor X activities, were affected in a dose-dependent fashion for argatroban, consistent with previously reported studies.3,4,7 It is unclear why factor X activity, a PT-based measurement, was less affected by argatroban concentration than prothrombin activity. Prothrombin activity is measured by combining DTI plasma with factor II–depleted plasma and then recording the resultant PT. Such decreases in the prothrombin level, as a result of the mixing of DTI plasma with prothrombin-depleted plasma, may account for the magnifying prolongation due to the pharmacodynamic properties (binding to thrombin's catalytic site) of argatroban. We recently published a patient management report in which oral anticoagulation therapy monitoring required factor activity measurement due to a strong antiphospholipid antibody biasing the INR measurement.10 This patient also had concomitant heparin-induced thrombocytopenia, complicating both DTI and oral anticoagulation. In these rare circumstances in which factor activity measurements may be required for the transition of DTI to oral anticoagulation, it is necessary to appreciate the bias of factor activity measurements associated with DTI therapy.

The potential weakness of our study is that we used pooled normal plasma and added known amounts of either DTI or pentasaccharide to achieve a desired concentration. In anticoagulation studies, it is generally accepted that spiking normal plasma with an anticoagulant (eg, heparin) is not an effective means for determining the dose response of a drug or potential influences of metabolites. Variation in an individual's factor levels, as well as individual responses to the anticoagulation effect of the drug itself, can impact a given laboratory test, such as the PT or aPTT. This study was not intended to provide dose responses, but was intended to approximate the effect that DTIs have on coagulation tests other than the aPTT and PT. Since this was an in vitro study, the effects of DTI or pentasaccharide on coagulation studies in patients receiving these drugs may require further investigation.

The presence of a DTI, and to a lesser degree of fondaparinux, in plasma can dramatically affect clot-based coagulation assays. Argatroban, at minimal concentrations, will falsely increase protein C, protein S, and DRVVT assay values, while falsely decreasing fibrinogen, prothrombin activity, and factor IX activity measurements. Lepirudin can falsely increase protein C and protein S values, as well as falsely decrease factor IX activity levels. Lepirudin (at 0.2 μg/mL) also biased DRVVT clotting times with a concomitant false-positive lupus anticoagulant ratio. Bivalirudin falsely increased clottable protein C and protein S results at minimal drug concentrations. Falsely decreased values of fibrinogen, prothrombin, and factor IX activities were seen in plasma containing more than 0.8 μg/mL of bivalirudin. Bivalirudin also falsely elevated DRVVT clotting times and yielded false-positive ratios. Fondaparinux only affected the protein S assay, resulting in false elevations of protein S levels even at minimal drug concentrations (0.2 μg/mL). Chromogenic assays for plasminogen, protein C, and antithrombin were not affected by any drug up to concentrations of 1.2 μg/ mL, although there was an upward trending of falsely increased antithrombin activity using thrombin substrate for the DTI samples. There was no effect of DTI or fondaparinux on D-dimer or von Willebrand factor activity. This information can be useful to clinicians, who should consider the possibility of biased results when interpreting results for patients receiving concurrent DTI or fondaparinux therapy.

Dager
,
W. E.
and
R. H.
White
.
Treatment of heparin-induced thrombocytopenia.
Ann Pharmacother
2002
.
36
:
489
503
.
Hirsh
,
J.
and
J. I.
Weitz
.
New antithrombic agents.
Lancet
1999
.
353
:
1431
1436
.
Sheth
,
S. B.
,
R. A.
DiCicco
,
M. J.
Hursting
,
T.
Montague
, and
D. K.
Jorkasky
.
Interpreting the international normalized ratio (INR) in individuals receiving argatroban and warfarin [published correction appears in Thromb Haemost. 2001;86:727].
Thromb Haemost
2001
.
85
:
435
440
.
Hursting
,
M. J.
,
J. L.
Zehnder
,
J. L.
Joffrion
,
J. C.
Becker
,
G. D.
Knappenberger
, and
R. P. Jr
Schwarz
.
The international normalized ratio during concurrent warfarin and argatroban anticoagulation: differential contributions of each agent and effects of the choice of thromboplastin used.
Clin Chem
1999
.
45
:
409
412
.
Fenyvesi
,
T.
,
I.
Jorg
, and
J.
Harenberg
.
Effect of phenprocoumon on monitoring of lepirudin, argatroban, melagatran and unfractionated heparin with the PiCT method.
Pathophysiol Haemost Thromb
2002
.
32
:
174
179
.
Gosselin
,
R. C.
,
W. E.
Dager
, and
J. H.
King
.
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
.
Walenga
,
J. M.
,
A. R.
Fasanella
, and
O.
Iqbal
.
et al
.
Coagulation laboratory testing in patients treated with argatroban.
Semin Thromb Hemost
1999
.
25
:(
suppl 1
).
61
66
.
Tobu
,
M.
,
O.
Iqbal
,
D. A.
Hoppensteadt
,
C.
Shultz
,
W.
Jeske
, and
J.
Fareed
.
Effects of a synthetic factor Xa inhibitor (JTV-803) on various laboratory tests.
Clin Appl Thromb Hemost
2002
.
8
:
325
336
.
Boneu
,
B.
,
J.
Necciari
, and
R.
Cariou
.
et al
.
Pharmacokinetics and tolerance of the natural pentasaccharide (SR90107/Org31540) with high affinity to antithrombin III in man.
Thromb Haemost
1995
.
74
:
1468
1473
.
Trask
,
A. S.
,
R. C.
Gosselin
,
J. A.
Diaz
, and
W. E.
Dager
.
Initiating and monitoring warfarin using clotting factors II, VII and X.
Ann Pharmacother
2004
.
38
:
251
256
.
Epub December 30, 2003
.

The authors have no relevant financial interest in the products or companies described in this article.

Author notes

Reprints: John T. Owings, MD, Trauma Division, Department of Surgery, University of California, Davis Medical Center, 2315 Stockton Blvd, Sacramento, CA 95817 ([email protected])