THIEME

228

Patients’ Serum and Urine as Easily Accessible

Samples for the Measurement of Non–Vitamin K
Antagonist Oral Anticoagulants
Job Harenberg, MD1 Shanshan Du, PhD1 Sandra Krämer, MD1 Christel Weiss, PhD2
Roland Krämer, PhD3 Martin Wehling, MD1

1 Clinical Pharmacology, Medical Faculty Mannheim, Ruprecht-Karls Address for correspondence Job Harenberg, MD, Clinical
University Heidelberg, Mannheim, Germany Pharmacology, Medical Faculty Mannheim, Ruprecht-Karls University
2 Biometry and Statistics, Medical Faculty Mannheim, Ruprecht-Karls Heidelberg, Maybachstrasse 14, 68169 Mannheim, Germany
University Heidelberg, Mannheim, Germany (e-mail: [email protected]).
3 Institute Inorganic Chemistry, University of Heidelberg,
Heidelberg, Germany

Semin Thromb Hemost 2015;41:228–236.

Abstract Measurement of the anticoagulant effect of non–vitamin K antagonist oral antico-

agulants (NOAC) may be desirable, in particular in patients with acute medical
conditions. Useful methods should give results rapidly within minutes, should be
easy to perform, specific, and sensitive. Using plasma samples, chromogenic assays
can be made to be specific for the two types of NOAC (factor Xa and thrombin
inhibitors), and also hemoclot and ecarin clotting time specific for dabigatran. If plasma
Keywords samples anticoagulated with sodium citrate are not available, blood samples anti-
► direct factor Xa coagulated with ethylene diamine tetraacetic acid or serum samples may be regarded as
inhibitors alternatives for the determination of NOAC. At present, dabigatran cannot be deter-
► direct thrombin mined from serum samples because it may be consumed during the clotting process to
inhibitors obtain serum. NOAC can be determined in urine samples due to their renal elimination.
► new oral direct Quantitative methods are preferable to qualitative methods, although the latter may be
anticoagulant advantageous in some situations, being developed as point-of-care tests for oral factor
► serum Xa and thrombin inhibitors. In these tests, the presence and absence of NOAC in urine
► urine can be identified with the naked eye after a few minutes and these tests are highly
► plasma specific and sensitive. New assays such as a semiquantitative determination in urine
► chromogenic samples and measurement using other sample matrices are currently under
substrate methods development.

Besides the conventional vitamin K antagonists, the non– One of the common characteristics of these anticoagu-
vitamin K antagonist oral anticoagulants (NOAC) such as lants is the considerable, though differing, impact of renal
dabigatran, rivaroxaban, apixaban, and edoxaban are approved function on their elimination. Dabigatran is excreted by the
in many countries for the prevention of venous and arterial kidneys to approximately 80%, rivaroxaban to
60%, of
thromboembolism.1 All compounds were developed with the which one half represents the active drug, and apixaban
dual aims of avoiding laboratory monitoring and to be used at and edoxaban to
25%. Dabigatran is therefore
fixed daily dose. As a consequence, plasma samples for post contraindicated in patients with a creatinine clearance
hoc determination of the anticoagulant effect were not taken below 30 mL/min, and the factor Xa inhibitors at below
in the pivotal studies or only assessed in subgroups of patients. 15 mL/min.2 For intermediate decreases of renal function,

Issue Theme Anticoagulant DOI http://dx.doi.org/ Copyright © 2015 by Thieme Medical

Therapy: Present and Future; 10.1055/s-0035-1544158. Publishers, Inc., 333 Seventh Avenue,
Guest Editor: Job Harenberg, MD. ISSN 0094-6176. New York, NY 10001, USA.
published online Tel: +1(212) 584-4662.
February 15, 2015
 Serum and Urine Samples for Measurement of NOAC Harenberg et al. 229

individual reduction of doses of rivaroxaban, apixaban, and methods is limited to a small number of specialized labora-
edoxaban are required. In the clinical studies, patients with tories.3,4 Recently, these methods were used to compare the
impaired renal function with a creatinine clearance below sensitivity and specificity of chromogenic substrate assays
these limits were not included. and of coagulation assays for NOACs after spiking to human
In real-world situations after market approval, NOACs are plasma or after administration to man.5 The most relevant
administered to patient types not included into the clinical conclusion of those investigations is that the LC-MS/MS
studies, and patients may develop diseases over time or take methods show a strong correlation to the specific chromo-
additional drugs (e.g., nonsteroidal anti-inflammatory genic substrate assays after spiking human plasma samples
drugs) that result in reduction of renal function. Patients with dabigatran, rivaroxaban, apixaban, and edoxaban, as
may also have to undergo acute surgical interventions or may well as to a specific clotting test for dabigatran (Pearson
receive drugs interacting with NOAC metabolism at the level correlation r > 0.95). Following administration of these
of the cytochrome P450 3A4 system or with its transmem- NOAC to volunteers or patients, correlations were also very
brane transport (e.g., in the gut) at the level of P-glycoprotein. strong with values of r > 0.90. Consequently, not all studies
These and other clinical situations lead to the inevitable fact used LC-MS/MS method as reference to measure NOAC from
that laboratory testing of NOAC will be required in specific human plasma samples but used the specific chromogenic
clinical situations in the future. Importantly, these clinical substrate assays for the oral direct factor Xa and thrombin
situations most often require urgent medical decisions inhibitors.
necessitating rapid monitoring. Such rapid monitoring is LC-MS/MS results also demonstrated the limited utility of
feasible specifically with point-of-care (POC) methods. global coagulation assays such as prothrombin time (PT) and
Conventional methods, POC methods, and alternative activated partial thromboplastin time (aPTT) assays including
matrices for determination of NOACs are reviewed in this different reagents for both type of assays. For other coagulation
article (see ►Fig. 1). assays, such comparisons were not performed. Because of the
strong correlation of the LC-MS/MS methods with chromogenic
substrate assays, comparisons of the anticoagulant effects of
Liquid Chromatography Methods
NOAC with other clotting assays were reported in comparison
Liquid chromatography with high-resolution mass spectrom- with the specific chromogenic substrate assays.3,4,6–8
etry (LC-MS/MS) methods were used initially for the devel- A limitation of the LC-MS/MS methods3 is that they differ
opment of the NOAC. They were adopted to identify the purity in the experimental setting (e.g., turbulent chromatography6
of the compounds during the production process as well as for and ultra-performance liquid chromatography–tandem mass
the identification of metabolites following administration to spectrometry [UPLC-MS/MS])7 and that these specific meth-
animals and humans. However, the availability of these ods were not compared.

Fig. 1 Outline of the matrices and methods to measure NOAC. Identification of colors: red and yellow, color of the matrix (blood and urine);
green, anticoagulation ex vivo during blood sampling; no color, conventional matrix (citrated blood samples); blue, new matrices for
measurement of NOAC. POC tests for DTI (direct thrombin inhibitor: dabigatran and others) and DFXaI (direct factor Xa inhibitors: rivaroxaban,
apixaban, edoxaban and others): a negative (right side) and a positive (left side) tests for the absence or presence of DTI and DFXaI in urine samples
are shown. The colors can be identified by the naked eye within 10 to 15 minutes. Test details are described in Du et al.18

Seminars in Thrombosis & Hemostasis Vol. 41 No. 2/2015

230 Serum and Urine Samples for Measurement of NOAC Harenberg et al.

Influence of Coagulation Platforms on Germany). This assay was initially developed and certified for
Coagulation Assays hirudins. New studies showed the applicability of the assay for
samples containing dabigatran following spiking to human
Chromogenic substrate assays and coagulation assays are plasma and after administration to patients treated with
performed on a variety of coagulation platforms. Every assay dabigatran.13 The concentrations of samples were calculated
needs to be adapted to a specific coagulation platform. An from control samples containing known concentrations of
example is given here: human plasma samples spiked with dabigatran.5,9,14
different concentrations of apixaban were investigated in an
international collaborative study on factor Xa–specific chro-
S2238 Chromogenic Method
mogenic substrate methods, such as STA Liquid Anti-Xa (Diag-
nostica Stago, Asnières sur Seine, France), STA Rotachrom The chromogenic substrate S2238 is another substrate spe-
(Diagnostica Stago), Coamatic Heparin (Haemochrom, Essen, cific for the determination of any thrombin inhibitor. It has
Germany), HemosIL Anti-Xa liquid (Instrumentation Labora- also been used for the measurement of dabigatran in plasma
tory, Kirchheim, Germany), and Technochrome Anti-Xa (Tech- samples spiked with dabigatran and form patients on treat-
noclone, Vienna, Austria).9 Each laboratory used its own ment with dabigatran. As outlined above, some differences in
coagulation platform. Subanalysis and comparison of data the concentrations in plasma samples may be detected due to
were done for centers using the ACL coagulation platform the development of each chromogenic substrate assay for a
(Instrumentation Laboratory) versus centers not using the ACL specific coagulation platform. Preliminary data seem to sup-
coagulation platform, and also for centers using the Stago port this finding, which has been reported in more detail for
coagulation platform versus centers not using the Stago coag- apixaban and various chromogenic methods on different
ulation. As a result, the coefficient of variation (CV) of the data coagulation platforms.13
was lower for all concentrations of apixaban if reagents and
platforms were from the same producer compared with the
Ecarin Chromogenic Assay
data of centers using the chromogenic assays on platforms of
another producer. In detail, the chromogenic assays HemosIL Ecarin is a snake venom isolated from Echis carinatus; it is
and Coamatic (Instrumentation laboratory) resulted in a lower used to convert prothrombin to meizothrombin, a prothrom-
CV on the ACL platform compared with the other platforms. bin intermediate that is sensitive to inhibition by DTIs.15,16
The STA-Liquid Anti-Xa and Rotachrom chromogenic assays Ecarin actives factor V in the plasma sample resulting in a
(Diagnostica Stago) resulted in a lower CV when analyzed on conversion of prothrombin to thrombin. Thereby, thrombin is
the Stago coagulation platform compared with the other not added to this type of chromogenic assay. A thrombin-
coagulation platform.10 Similar data can be anticipated for specific chromogenic substrate is released dose dependently
coagulation assays when they are performed on coagulation by thrombin and inhibited by dabigatran inversely to the
platforms on which the coagulation assays were developed. concentration of the inhibitor. This assay is specific for
These results show the importance of analyzing and describing dabigatran but is not presently used as frequently as other
each assay for each coagulation platform. In addition, the data chromogenic substrate assays for the determination of
show that assays perform better if they are used on the dabigatran.
coagulation platform on which the analysis has been opti-
mized by specific adaptations of the experimental setting.
Clotting Time Assays for Oral Direct
These results also show that the variability of measurements
Thrombin Inhibitors
of NOAC concentrations or activities may result from such
differences in analytical techniques used in the publications. Hemoclot Thrombin Inhibitor Assay
The procedure of hemoclot thrombin inhibitor assay (HTI)
(HYPHEN BioMed, Neuville-sur-Oise, France) is a diluted
Chromogenic Assays for Oral Direct
thrombin clotting time method. Plasma is diluted 1 to 8
Thrombin Inhibitors
with human pool plasma and thrombin is added to induce
The direct thrombin inhibitor (DTI) dabigatran is a small clotting. The clotting times are prolonged by low (
10 to 20
synthetic oral anticoagulant with a nonpeptidic, potent, ng/mL) to high (
500 ng/mL) concentrations of dabigatran in
specific, competitive, and reversible binding to free and a linear fashion.17 This assay was introduced as a standard
fibrin-bound thrombin.11 Dabigatran etexilate is rapidly method by the US Food and Drug Administration (FDA) to
hydrolyzed to its active form dabigatran after oral adminis- determine dabigatran. In fact, many publications reported on
tration and intestinal absorption by nonspecific, ubiquitous the high correlation to the LC-MS/MS method and to chromo-
esterases in gut, blood, and liver.12 To improve oral absorp- genic methods and on the usefulness in patients. However, in
tion, tartaric acid is added to the oral preparation. patient samples, a high variation of the results was observed,
a finding that was attributed mainly to the variation of the
pharmacokinetics of dabigatran, and patients’ adherence to
Direct Thrombin Inhibitors Assay
therapy rather than to the performance of the assay. Using
One of the thrombin specific chromogenic substrate methods specific calibrators, the clotting times are converted to ng/mL
is the DTI assay (Siemens Healthcare Diagnostics Inc., Marburg, dabigatran per mL plasma.

Seminars in Thrombosis & Hemostasis Vol. 41 No. 2/2015

 Serum and Urine Samples for Measurement of NOAC Harenberg et al. 231

Ecarin Clotting Time reagents. All aPTT reagents are insensitive to low concen-
Ecarin is snake venom isolated from Echis carinatus as trations of dabigatran. High concentration of dabigatran in
described above. The ecarin reagent (Pentapharm, Basle, plasma prolong coagulation times of aPTT, but the coefficient
Switzerland) is added to initiate clotting. The clotting times of variation remains above 50% even for plasma samples
are linearly prolonged by dabigatran, and through calibrators spiked with dabigatran and analyzed within 1 day (intraday
the concentrations of dabigatran are calculated. As for the variation).23–26 For clinical use, large prolongations of aPTT
ecarin chromogenic assay, this test is presently also not used indicate high concentrations of dabigatran in plasma.
frequently for experimental analyses of the effect of However, the variation of the results is too wide to sensitively
dabigatran. detect overdosing of dabigatran. This underpins the limited
value of aPTT for the determination of dabigatran in plasma
Heptest Coagulation Assay samples of patients treated with this anticoagulant.
Recently, the effect of dabigatran was investigated for the
Heptest-STAT coagulation assay demonstrating a prolonga-
Chromogenic Assays for Oral Direct Factor Xa
tion over a concentration range from approximately 50 to 500
Inhibitors
ng/mL of the anticoagulant.18 The assay is less sensitive at
concentrations of dabigatran below 100 or 300 ng/mL and can Several chromogenic substrates have been proven to specifi-
be used if overdosing of dabigatran is suspected. These cally determine the inhibition of rivaroxaban, apixaban,
limitations in this assay are less pronounced for oral direct edoxaban, and other oral direct factor Xa inhibitors in human
factor Xa inhibitors, which more specifically prolong this plasma samples. The dependence of each chromogenic meth-
assay (see later). od on the coagulation platform used with regard to the
measured concentration of apixaban was described above.
Prothrombinase-Induced Clotting Time These findings should be applicable to other oral factor Xa
The prothrombinase-induced clotting time (PiCT) assay uses inhibitors as well. The chromogenic substrates differ in their
prothrombinase to modify factor V and consequently factor X affinity, association, and dissociation constants toward factor
and prothrombin to activated clotting factors. Therefore, PiCT Xa. Consequently, the kinetics of interaction are also influ-
is sensitive to unfractionated heparins, low-molecular- enced differently by the factor Xa inhibitors. This was inves-
weight heparins, danaparoid, fondaparinux, hirudins, direct tigated in detail for rivaroxaban using five chromogenic
thrombin, and direct factor Xa inhibitors. For specific indirect substrates.27 Human plasma samples were spiked with vari-
and direct factor Xa inhibitors, the test has to be performed as ous concentrations of rivaroxaban and incubated with the
a two-step method (reagents from Pentapharm, Basle, chromogenic substrate, and the release of para-nitroaniline
Switzerland) (normal mean value 29.9  4.6 seconds19). For by the residual activity of factor Xa was determined. The
other coagulation inhibitors, preincubation of plasma with optical density (OD at 405 nm) was plotted versus the
prothrombinase is not required.20 concentration of rivaroxaban in the sample. The results
demonstrated significant differences between the OD and
Prothrombin Time and Dilute Prothrombin Time concentration curves of the chromogenic substrates. By sta-
Assays tistical methods taking these differences into account, over-
PT assays can be performed with many thromboplastin lapping OD versus concentration curves for rivaroxaban with
reagents. It has been demonstrated in several studies that the various chromogenic substrates were obtained.27 By this
every thromboplastin possesses an individual sensitivity procedure, the variability between assay results was dramat-
toward dabigatran. A standardization of the reagents in ically reduced. The application of this method to clinical
relation to dabigatran has not been investigated. Importantly, samples remains to be validated. These data demonstrate
the international normalized ratio (INR) cannot be used to that differences between results of chromogenic substrate
standardize the influence of dabigatran on the PT as it was assays can be eliminated by mathematical approaches. Care-
done for patients on treatment with coumarins.21,22 This is ful adaptation of chromogenic methods to hemostasis instru-
due to the fact that standardization of INR requires plasma ment platforms may reduce this variability further.
samples of patients treated with coumarin, which is depletion
of the vitamin K–dependent clotting factors. In contrast,
Clotting Assays for Measurement of Oral
plasma of patients treated with dabigatran contains normal
Direct Factor Xa Inhibitors
concentration of these clotting factors rendering a calculation
of the INR as inappropriate. Heptest Coagulation Assay
Heptest STAT is a coagulation assay that is activated by a
Activated Partial Thromboplastin Time reagent containing calcium, phospholipids, and factor V
As for PT reagents, aPTT can be performed by using several (HemaXa Ltd, Tioga, TX). The previously available Heptest
commercially available activators for the assay. Transforma- coagulation assay sensitively determines the activities of
tion of the clotting times to ratios may be helpful to compare unfractionated heparins,28,29 low-molecular-weight hepa-
the effects of dabigatran, but relationships between concen- rins,30 fondaparinux,31 and heparinoids.32 Heptest was
trations and ratios still vary between reagents. Further stan- used in phase I,33 phase II, or phase III clinical trials to
dardization activities have not been published for aPTT determine the pharmacodynamics effects of rivaroxaban34

Seminars in Thrombosis & Hemostasis Vol. 41 No. 2/2015

232 Serum and Urine Samples for Measurement of NOAC Harenberg et al.

and apixaban.35 The utility of Heptest STAT under real-life has to be considered that results of a POC test are influenced
conditions has so far only been examined in patients treated by the sensitivity of the reagents toward the NOAC and by the
with low-molecular-weight heparins.36 A recent investiga- POC platform. This has been described in detail earlier for
tion demonstrated that Heptest STAT determined the con- apixaban and a chromogenic test. At present, one publication
centrations of the NOACs rivaroxaban, apixaban, and is available with results for dabigatran and rivaroxaban
dabigatran as accurately as specific chromogenic assays.18 determined on two POC devices.42 The results show a high
variability of the results with both anticoagulants after
Prothrombinase-Induced Clotting Time adopting an aPTT reagent for measuring dabigatran and
The PiCT assay is sensitive toward rivaroxaban and other two PT thromboplastin reagents for rivaroxaban. The authors
direct factor Xa inhibitors if plasma samples are preincubated concluded that both methods may be useful for detecting an
with prothrombinase.37 Newer publications on the influence overdose of dabigatran and rivaroxaban in acute clinical
of NOAC on PiCT are not available. conditions.42 Additional publications are currently not
available.
Prothrombin Time and Dilute Prothrombin Time
PT is prolonged dose dependently by all thromboplastin
Determination of NOAC from Serum Samples
reagents. The sensitivity of the thromboplastin reagents varies
substantially for the different direct factor Xa inhibitors. In Unlike plasma samples, serum samples are taken from all
addition, the factor Xa inhibitors differ substantially in their patients in acute and nonacute clinical situations. Blood for
inhibitory potency as determined by PT with rivaroxaban the determination of coagulation parameters needs to be
being the most potent and apixaban the least potent agent collected separately into plastic or siliconized glass tubes
in this assay.38,39 The results from clinical trials vary substan- containing sodium citrate to inhibit mechanical coagulation
tially. A dilution of the thromboplastin reagent did not improve activation in vitro. Potential pitfalls of blood drawing for
the sensitivity of the reagent toward the factor Xa inhibitors. coagulation parameters are incomplete filling of the tube
From a clinical point of view, the variability of the response of resulting in wrong ratios of anticoagulant to blood leading to
thromboplastin reagents is too high to accurately identify low incorrect coagulation results and the activation of blood
or high concentrations of rivaroxaban.40 clotting during and after blood sampling due to incorrect
handling. In addition, correct handling also includes centri-
fugation at given temperatures, analysis within a given time
Determination of NOAC from Plasma
frame, and specific procedures for freezing and thawing of
Samples Anticoagulated with EDTA
samples.43 Published data indicate that rivaroxaban and
Sodium citrate is routinely added to blood samples to prevent apixaban can be accurately determined from serum samples
blood coagulation in vitro for coagulation assays, and ethylene of patients on treatment and that values correlate well with
diamine tetraacetic acid (EDTA) is added for determination of those determined from plasma samples.44 This would seem
hematology and some clinical chemistry parameters. In con- advantageous given the wider availability of serum samples
trast to plasma obtained with citrate, EDTA plasma specimens compared with plasma samples, and the easier handling/
cannot be accurately tested by clotting-based assays due to the preanalytical stability of serum samples.
irreversible chelation of ionic calcium. While clotting assays In contrast, however, other data indicate that dabigatran
require ionized calcium for the enzymatic reaction, this may cannot be accurately determined from serum samples. The
not be the case for the interaction of factor Xa and a chromo- variability of the lower values of dabigatran in serum com-
genic substrate.39 Recent results demonstrated that rivarox- pared with plasma was twice as high using chromogenic
aban determinations in plasma samples anticoagulated with assays. An explanation for this finding includes a consump-
sodium citrate or EDTA and spiked with serial dilutions of tion of dabigatran during thrombin generation of blood
rivaroxaban were not statistically significantly different for samples to obtain serum. The determination from serum still
two chromogenic assays. For patients on treatment with requires improvement before a general applicability in
rivaroxaban, the original values for mean concentrations and patients may be achieved.13
variances of rivaroxaban were also not different in samples
anticoagulated by either sodium citrate or EDTA during sam-
Measurement of NOAC from Urine Samples
pling.41 Additional data need to be generated with other direct
factor Xa inhibitors and by increasing the number of patients As stated earlier, the elimination of NOAC variably decreases
on treatment. However, this approach is promising and of with decreasing renal function. Therefore, patients on treatment
clinical relevance, if only EDTA, but no citrate plasma, samples with NOAC and with impaired renal function may be exposed to
are available from patients. an increased risk of bleeding due to drug accumulation.2 For
example, the mean clearance of total dabigatran decreased from
62.9 mL/min in the healthy control group to 10.3 mL/min in
Point-of-Care Tests Using Plasma Samples
subjects with severe renal impairment. The mean elimination
To determine the presence of NOAC in blood samples in half-life of dabigatran increased from 13.8 hours in healthy
emergency situations, POC tests from blood may offer advan- subjects to 16.6, 18.7, and 27.5 hours in subjects with mild,
tages over conventional clotting assays such as PT or aPTT. It moderate, and severe renal impairment, respectively (►Table 1).

Seminars in Thrombosis & Hemostasis Vol. 41 No. 2/2015

 Serum and Urine Samples for Measurement of NOAC Harenberg et al. 233

Table 1 Ratios of Cmax and AUC (ng x h/mL) for samples from and rivaroxaban was analyzed by a set of prepared samples
patients with renal impairment and those from healthy subjects obtained from patients under therapy. The results of the
following a single oral administration of dabigatran45 sensitivity, specificity, accuracy, and positive and negative
predictive indices are given in ►Table 2.46
Parameter Renal impairment Ratio (90% CI) In specific clinical situations, the rapid determination of
Cmax (ng/mL) Mild 1.11 (0.61–2.03) NOAC may be required for clinical decision making. Typical
Moderate 1.70 (0.93–3.10) examples are where it is necessary to know if the drug is on
board or not, or there are indications for fibrinolytic therapy
Severe 2.12 (1.25–3.59)
in cerebral embolism, for placement of a lumbar catheter,
AUC (ng x h/mL) Mild 1.50 (0.78–2.90) and/or severe bleeding events. Renal function reduces with
Moderate 3.15 (1.63–6.08) increasing age. Advantages of using urine as test sample
Severe 6.31 (3.54–11.25) include its noninvasive availability and the possibility to
develop POC test systems to be performed by trained personal
Abbreviations: AUC, area under the concentration time curve; CI, or by the patients themselves. These tests can be repeated
confidence interval; Cmax, maximal concentration.
easily. In contrast, POC tests from blood require blood sam-
pling from the fingertip and accurate administration to the
test strip of the coagulation device. In addition, standardiza-
The determination of dabigatran in urine showed high tion of the POC test is not required for urine samples, as
concentrations in the range of 2 to 20 µg/mL.13 The value of sensitivity and specificity of the NOAC tests in urine have
measuring dabigatran in urine is that this method is not homogeneously been found to be above 95%.46 A scheme for
invasive, may be repeated easily, and detect patients with a the clinical disease management of specific patient groups is
lack of compliance even if considered as a semiquantitative given in ►Table 3 in relation to the results of the qualitative
assessment. The measurement of dabigatran in urine is POC test for NOAC.
feasible and may be used to check compliance as has been
already demonstrated with a qualitative test.9 The validity of
Discussion
the assay results remains to be determined in specific clinical
settings under real-life conditions of therapy. Similar reports Several chromogenic assays and clotting tests were evaluated
are not yet available for rivaroxaban and apixaban. in the last couple of years to measure the anticoagulant effects
of NOAC in specific patient populations.47 Specific problems
were identified in these investigations. Standardization of
Point-of-Care Test for NOAC from Urine
assays for each platform is required to improve comparability
Samples
of the measurements and related publications. It has still to be
Development of POC methods for DOACs in urine is based on stressed that INR values as being specifically designed and
their immediate glomerular filtration and excretion into the standardized with plasma samples of patients treated with
urine if present in blood. The test principle is based on the vitamin K antagonist are unsuitable for analyzing the effects
development of different colors in the presence and absence of NOAC. Specific liquid chromatography methods are not
of oral direct factor Xa or thrombin inhibitors, and which can suitable for routine monitoring due to the sophisticated
be identified by the naked eye. Two colors were chosen for equipment required and the specific knowledge of specialists
each class of inhibitors: for factor Xa (absence: yellow color; to run these assays. Of note, comparisons between these test
presence: no color; ►Fig. 1) and for thrombin inhibitors systems remain to be performed. On the other hand, high
(absence: green color; presence: blue color; ►Fig. 1).46 correlations of the LC-MS/MS methods with chromogenic
The ability of patients to correctly identify the colors with assays and Hemoclot limit the necessity to perform these
the naked eye as being positive and negative for dabigatran tests for experiments in the future.

Table 2 Sensitivity, specificity, NPV, and PPV of the POC method using urine samples of patients on treatment as determined by
color reading by patients46

Dabigatran (n ¼ 484) Rivaroxaban (n ¼ 457)

Mean 95% CI Mean 95% CI
Sensitivity 100 99–100 96 94–98
Specificity 99 98–100 98 96–99
Accuracy 99 99–100 97 96–98
PPV 99 98–100 98 97–99
NPV 100 99–100 96 94–98

Abbreviations: CI, confidence interval; NPV, negative predictive value; POC, point of care; PPV, positive predictive value.

Seminars in Thrombosis & Hemostasis Vol. 41 No. 2/2015

234 Serum and Urine Samples for Measurement of NOAC Harenberg et al.

Table 3 Position of the urinary POC test results in the context of obtained coagulation values. Therefore, efforts are under-
clinical disease management taken to tackle this unmet medical need of accurately and
feasibly measuring the anticoagulant effects of NOAC. So
Bleeding complication far, only the determination of NOAC in urine samples offers
Test is negative: a possibility to partly address this problem, which is related
• Patient did not take medication within the past 1–24 h to the high rate of renal excretion of these anticoagulants
reaching urinary concentrations of above 100 ng/mL as
• Stop bleeding, re-check POC after 1 h, determine
soon as a NOAC is present in blood. Thus, an overlap with
chromogenic test
controls was not observed with such testing.13 These assays
Test is positive (patient took NOAC within the past 24 h):
were also developed as qualitative POC methods using
• Perform a chromogenic test specific for NOAC or urine samples, which can be read by the naked eye within
hemoclot/ECT for dabigatran a couple of minutes by different colors for a negative and
• Wait for result, treat according to published guidance positive result (absence and presence of the NOAC).46 One
Thrombotic event limitation of these methods is the time difference of
approximately 1 to 2 hours between actual concentration
Test is negative
of NOAC in blood and appearance in urine. Strengths of the
• Patient did not take medication within the last 1–24 h
methods are that they are not invasive, can be repeated
• Start anticoagulation regimen easily, they are highly sensitive and specific, and that the
Test is positive POC test can be performed by the patients themselves.
• Perform specific test as above Further development of the test system is currently on
the way at specialized centers.
• Wait for result and treat according to published
In conclusion, current methods to measure NOAC in
suggestions
several matrices are chromogenic substrate assays and
None of above and no acute medical intervention required
some clotting methods. POC methods are developed to speed
Test is negative up the decision making in critical care medicine. There is a
• Patient education high possibility of solving the current methodological prob-
• Start alternative anticoagulation if indicated lems and limitations of the assays due to an intense collabo-
ration of producers of the drugs, of the assay reagents and
Test is positive
coagulation platforms, laboratory and clinical scientists, and
• Interrupt NOAC therapy, if medical intervention physicians.
requires this
• Redo POC within 24 h, take sample for specific
anticoagulant test Conflict of Interest
• Restart therapy after medical intervention J.H.: Lecturing and consulting fees from Novartis, Bayer
• Continue therapy with NOAC if medically indicated HealthCare, Boehringer Ingelheim, Roche Diagnostics,
Pfizer, Bristol-Myers Squibb, Daichii-Sankyo, LEO Pharma;
Adherence to therapy
support of research from Novartis, Bayer HealthCare,
Test is negative Boehringer Ingelheim, Roche Diagnostics, Pfizer, Bristol-
• Education program for patient to improve adherence Myers Squibb, HepLabs.
to therapy M.W.: Employed by AstraZeneca R&D, Mölndal, as director
Test is positive of discovery medicine (¼ translational medicine) from
• No action required, continue therapy 2004 to 2006, while on sabbatical leave from his profes-
sorship at the University of Heidelberg. After return to this
• Check creatinine if elevated values are suspected
position in January 2007, he received lecturing and con-
Abbreviations: ECT, ecarin clotting time; NOAC, non–vitamin K antag- sulting fees from Novartis, Pro Bono Bio, Bayer HealthCare,
onist oral anticoagulant; POC, point of care. Boehringer Ingelheim, Roche, Pfizer, Bristol-Myers Squibb,
Daichii-Sankyo, LEO, Lilly, Shire, Mundipharm, and Novo-
Nordisk.
Several recommendations or guidelines have been published R.K.: Support of research from Medinvent, Siemens
on the use of coagulation assays to monitor NOAC.10,16,40,48 Healthcare. The other authors do not have to declare a
However, careful reading of the recommendations may also conflict of interest.
convey the impression that a consensus on this topic is not yet
obtained.38
A major limitation of all available methods is the fact that Acknowledgments
a wide range of values is obtained from patients under The authors thank the technical assistance of Mrs. Chris-
treatment and consequently essentially overlap in values tina Giese, Antje Hagedorn, and Inge Träger. The research
exists between controls and treated patients. Clinical was supported by a grant from the Dietmar-Hopp Foun-
decision making may thus be difficult in relation to the dation to J.H.

Seminars in Thrombosis & Hemostasis Vol. 41 No. 2/2015

 Serum and Urine Samples for Measurement of NOAC Harenberg et al. 235

References 19 Harenberg J, Giese C, Hagedorn A, Traeger I, Fenyvesi T. Determi-

1 Hylek EM, Ko D, Cove CL. Gaps in translation from trials to practice: nation of antithrombin-dependent factor Xa inhibitors by pro-
non-vitamin K antagonist oral anticoagulants (NOACs) for stroke thrombin-induced clotting time. Semin Thromb Hemost 2007;
prevention in atrial fibrillation. Thromb Haemost 2014;111(5): 33(5):503–507
783–788 20 Calatzis A, Peetz D, Haas S, Spannagl M, Rudin K, Wilmer M.
2 Harenberg J, Wehling M. Future of anticoagulant therapy. Cardi- Prothrombinase-induced clotting time assay for determination of
ovasc Ther 2011;29(5):291–300 the anticoagulant effects of unfractionated and low-molecular-
3 Antovic JP, Skeppholm M, Eintrei J, et al. Evaluation of coagulation weight heparins, fondaparinux, and thrombin inhibitors. Am J Clin
assays versus LC-MS/MS for determinations of dabigatran concen- Pathol 2008;130(3):446–454
trations in plasma. Eur J Clin Pharmacol 2013;69(11):1875–1881 21 Samama MM, Martinoli JL, LeFlem L, et al. Assessment of labora-
4 Douxfils J, Dogné JM, Mullier F, et al. Comparison of calibrated tory assays to measure rivaroxaban—an oral, direct factor Xa
dilute thrombin time and aPTT tests with LC-MS/MS for the inhibitor. Thromb Haemost 2010;103(4):815–825
therapeutic monitoring of patients treated with dabigatran etex- 22 Tripodi A, Chantarangkul V, Guinet C, Samama MM. The Interna-
ilate. Thromb Haemost 2013;110(3):543–549 tional Normalized Ratio calibrated for rivaroxaban has the poten-
5 Harenberg J, Kraemer S, Du S, et al. Determination of direct oral tial to normalize prothrombin time results for rivaroxaban-treated
anticoagulants from human serum samples. Semin Thromb patients: results of an in vitro study. J Thromb Haemost 2011;9(1):
Hemost 2014;40(1):129–134 226–228
6 Gous T, Couchman L, Patel JP, Paradzai C, Arya R, Flanagan RJ. 23 Van Blerk M, Bailleul E, Chatelain B, et al. Influence of dabigatran
Measurement of the direct oral anticoagulants apixaban, dabiga- and rivaroxaban on routine coagulation assays. A nationwide
tran, edoxaban, and rivaroxaban in human plasma using turbulent Belgian survey. Thromb Haemost 2014;112(6):
flow liquid chromatography with high-resolution mass spectrom- 24 Hapgood G, Butler J, Malan E, Chunilal S, Tran H. The effect of
etry. Ther Drug Monit 2014;36(5):597–605 dabigatran on the activated partial thromboplastin time and
7 Schmitz EM, Boonen K, van den Heuvel DJ, et al. Determination of thrombin time as determined by the Hemoclot thrombin inhibitor
dabigatran, rivaroxaban and apixaban by ultra-performance liquid assay in patient plasma samples. Thromb Haemost 2013;110(2):
chromatography - tandem mass spectrometry (UPLC-MS/MS) and 308–315
coagulation assays for therapy monitoring of novel direct oral 25 Samama MM, Contant G, Spiro TE, et al. Laboratory assessment of
anticoagulants. J Thromb Haemost 2014;12(10):1636–1646 rivaroxaban: a review. Thromb J 2013;11(1):11
8 Delavenne X, Moracchini J, Laporte S, Mismetti P, Basset T. UPLC 26 Suzuki S, Otsuka T, Sagara K, et al. Dabigatran in clinical practice for
MS/MS assay for routine quantification of dabigatran - a direct atrial fibrillation with special reference to activated partial throm-
thrombin inhibitor - in human plasma. J Pharm Biomed Anal 2012; boplastin time. Circ J 2012;76(3):755–757
58:152–156 27 Harenberg J, Krämer R, Giese C, Marx S, Weiss C, Wehling M.
9 Harenberg J, Du S, Krämer S, et al. Novel methods for assessing oral Determination of rivaroxaban by different factor Xa specific
direct factor Xa and thrombin inhibitors: use of point-of-care testing chromogenic substrate assays: reduction of interassay variability.
and urine samples. Semin Thromb Hemost 2013;39(1):66–71 J Thromb Thrombolysis 2011;32(3):267–271
10 Harenberg J, Du S, Weiss C, Krämer R, Hoppensteadt D, Walenga J; 28 Emanuele RM, Racanelli A, Fareed J. Pharmacokinetics of heparins
working party: methods to determine apixaban of the Subcom- differing in mean molecular weight using a Xa amidolytic and
mittee on Control of Anticoagulation of the International Society Heptest clotting method. Ther Drug Monit 1988;10(2):153–159
of Thrombosis and Haemostasis. Report of the Subcommittee on 29 Yin ET, Wessler S, Butler JV. Plasma heparin: a unique, practical,
Control of Anticoagulation on the determination of the anticoag- submicrogram-sensitive assay. J Lab Clin Med 1973;81(2):
ulant effects of apixaban: communication from the SSC of the ISTH. 298–310
J Thromb Haemost 2014;12(5):801–804 30 Hoffmann U, Harenberg J, Bauer K, et al. Bioequivalence of
11 Hauel NH, Nar H, Priepke H, Ries U, Stassen JM, Wienen W. subcutaneous and intravenous body-weight-independent high-
Structure-based design of novel potent nonpeptide thrombin dose low-molecular-weight heparin Certoparin on anti-Xa, Hept-
inhibitors. J Med Chem 2002;45(9):1757–1766 est, and tissue factor pathway inhibitor activity in volunteers.
12 Stangier J, Stähle H, Rathgen K, Roth W, Reseski K, Körnicke T. Blood Coagul Fibrinolysis 2002;13(4):289–296
Pharmacokinetics and pharmacodynamics of dabigatran etexilate, 31 Jeske WP, Walenga JM, Samama MM, Hoppensteadt D, Mayuga M,
an oral direct thrombin inhibitor, with coadministration of digox- Fareed J. Functionality of fondaparinux (pentasaccharide) depends
in. J Clin Pharmacol 2012;52(2):243–250 on clinical antithrombin levels. Blood Coagul Fibrinolysis 2011;
13 Du S, Weiss C, Giese C, et al. Determination of dabigatran in plasma, 22(3):206–210
serum and urine samples: comparison of six methods. Clin Chem 32 Harenberg J, Jeschek M, Acker M, Malsch R, Huhle G, Heene DL.
Lab Med. In press Effects of low-molecular-weight dermatan sulfate on coagulation,
14 Harenberg J, Giese C, Marx S, Krämer R. Determination of dabiga- fibrinolysis and tissue factor pathway inhibitor in healthy volun-
tran in human plasma samples. Semin Thromb Hemost 2012; teers. Blood Coagul Fibrinolysis 1996;7(1):49–56
38(1):16–22 33 Kubitza D, Becka M, Mück W, Krätzschmar J. Pharmacodynamics
15 Nowak G. The ecarin clotting time, a universal method to quantify and pharmacokinetics during the transition from warfarin to
direct thrombin inhibitors. Pathophysiol Haemost Thromb 2003; rivaroxaban: a randomized study in healthy subjects. Br J Clin
33(4):173–183 Pharmacol 2014;78(2):353–363
16 Fenyvesi T, Jörg I, Weiss C, Harenberg J. Effects of lepirudin, 34 Mueck W, Lensing AW, Agnelli G, Decousus H, Prandoni P, Mis-
argatroban and melagatran and additional influence of phenpro- selwitz F. Rivaroxaban: population pharmacokinetic analyses in
coumon on ecarin clotting time. Thromb Res 2003;111(1-2):89–94 patients treated for acute deep-vein thrombosis and exposure
17 Stangier J, Feuring M. Using the HEMOCLOT direct thrombin simulations in patients with atrial fibrillation treated for stroke
inhibitor assay to determine plasma concentrations of dabigatran. prevention. Clin Pharmacokinet 2011;50(10):675–686
Blood Coagul Fibrinolysis 2012;23(2):138–143 35 Miyares MA, Davis K. Newer oral anticoagulants: a review of
18 Harenberg J, Du S, Krämer S, Krämer R, Wehling M, Weiss C. laboratory monitoring options and reversal agents in the hemor-
Measurement of non-vitamin K antagonist oral anticoagulants in rhagic patient. Am J Health Syst Pharm 2012;69(17):1473–1484
patient plasma using Heptest-STAT coagulation method (epub 36 Dempfle CE, Zharkowa U, Elmas E, Ahmad-Nejad P, Neumaier M,
ahead of print). Ther Drug Monit 2014 Borggrefe M. Heptest-STAT, a new assay for monitoring of low-

Seminars in Thrombosis & Hemostasis Vol. 41 No. 2/2015

236 Serum and Urine Samples for Measurement of NOAC Harenberg et al.

molecular-weight heparins, is not influenced by pregnancy-related 43 Lippi G, Salvagno GL, Montagnana M, Lima-Oliveira G, Guidi GC,
changes of blood plasma. Thromb Haemost 2009;102(5):1001–1006 Favaloro EJ. Quality standards for sample collection in coagulation
37 Kluft C, Meijer P, Kret R, Burggraaf J. Preincubation in the Pro- testing. Semin Thromb Hemost 2012;38(6):565–575
thrombinase-induced Clotting Time test (PiCT) is necessary for in 44 Harenberg J, Krämer S, Du S, et al. Measurement of rivaroxaban
vitro evaluation of fondaparinux and to be avoided for the and apixaban in serum samples of patients. Eur J Clin Invest 2014;
reversible, direct factor Xa inhibitor, rivaroxaban. Int J Lab Hem- 44(8):743–752
atol 2013;35(4):379–384 45 Stangier J, Rathgen K, Stähle H, Mazur D. Influence of renal
38 Lippi G, Favaloro EJ. Recent guidelines and recommendations for impairment on the pharmacokinetics and pharmacodynamics
laboratory assessment of the direct oral anticoagulants (DOACs): is of oral dabigatran etexilate: an open-label, parallel-group,
there consensus? Clin Chem Lab Med 2014 single-centre study. Clin Pharmacokinet 2010;49(4):259–268
39 Rosén S. Chromogenic methods in coagulation diagnostics. 46 Harenberg J, Krämer S, Du S, Weiss C, Krämer R. Concept of a point
Hamostaseologie 2005;25(3):259–266 of care test to detect new oral anticoagulants in urine samples.
40 Douxfils J, Mullier F, Loosen C, Chatelain C, Chatelain B, Dogné JM. Thromb J 2013;11(1):15
Assessment of the impact of rivaroxaban on coagulation assays: 47 Cuker A, Siegal DM, Crowther MA, Garcia DA. Laboratory
laboratory recommendations for the monitoring of rivaroxaban measurement of the anticoagulant activity of the non-vitamin
and review of the literature. Thromb Res 2012;130(6):956–966 K oral anticoagulants. J Am Coll Cardiol 2014;64(11):
41 Du S, Krämer S, Giese C, Weiss C, Wehling M, Krämer R, Harenberg 1128–1139
J. Chromogenic assays for measurement of rivaroxaban from EDTA 48 Baglin T, Hillarp A, Tripodi A, Elalamy I, Buller H, Ageno W.
anticoagulated plasma samples (epub ahead of print) Thromb Measuring Oral Direct Inhibitors (ODIs) of thrombin and factor
Haemost 2015;113(5) Xa: A recommendation from the Subcommittee on Control of
42 Mani V, Wang S, Inci F, De Libero G, Singhal A, Demirci U. Emerging Anticoagulation of the Scientific and Standardisation Committee
technologies for monitoring drug-resistant tuberculosis at the of the International Society on Thrombosis and Haemostasis.
point-of-care. Adv Drug Deliv Rev 2014;78C:105–117 J Thromb Haemost 2013;11(4):756–760

Seminars in Thrombosis & Hemostasis Vol. 41 No. 2/2015