Laboratory Assessment of Novel Oral Anticoagulants: Method Suitability and Variability Between Coagulation Laboratories

Transcription

1 Papers in Press. Published February 1, 2013 as doi: /clinchem The latest version is at Clinical Chemistry 59: (2013) Hemostasis and Thrombosis Laboratory Assessment of Novel Oral Anticoagulants: Method Suitability and Variability Between Coagulation Laboratories Tuukka A. Helin, 1 Anja Pakkanen, 2 Riitta Lassila, 1,3 and Lotta Joutsi-Korhonen 1* BACKGROUND: Laboratory tests to assess novel oral anticoagulants (NOACs) are under evaluation. Routine monitoring is unnecessary, but under special circumstances bioactivity assessment becomes crucial. We analyzed the effects of NOACs on coagulation tests and the availability of specific assays at different laboratories. METHODS: Plasma samples spiked with dabigatran (Dabi; 120 and 300 g/l) or rivaroxaban (Riva; 60, 146, and 305 g/l) were sent to 115 and 38 European laboratories, respectively. International normalized ratio (INR) and activated partial thromboplastin time (APTT) were analyzed for all samples; thrombin time (TT) was analyzed specifically for Dabi and calibrated anti activated factor X (anti-xa) activity for Riva. We compared the results with patient samples. RESULTS: Results of Dabi samples were reported by 73 laboratories (13 INR and 9 APTT reagents) and Riva samples by 22 laboratories (5 INR and 4 APTT reagents). Both NOACs increased INR values; the increase was modest, albeit larger, for Dabi, with higher CV, especially with Quick (vs Owren) methods. Both NOACs dose-dependently prolonged the APTT. Again, the prolongation and CVs were larger for Dabi. The INR and APTT results varied reagent-dependently (P 0.005), with less prolongation in patient samples. TT results (Dabi) and calibrated anti-xa results (Riva) were reported by only 11 and 8 laboratories, respectively. CONCLUSIONS: The screening tests INR and APTT are suboptimal in assessing NOACs, having high reagent dependence and low sensitivity and specificity. They may provide information, if laboratories recognize their limitations. The variation will likely increase and the sensitivity differ in clinical samples. Specific assays measure NOACs accurately; however, few laboratories applied them American Association for Clinical Chemistry Novel oral anticoagulants (NOACs), 4 such as thrombin inhibitor dabigatran (Dabi) and activated factor X (Xa) inhibitor rivaroxaban (Riva), offer effective alternatives for anticoagulation (1, 2). Whereas vitamin K antagonists (VKAs) have highly variable pharmacokinetics and pharmacodynamics and require monitoring with the international normalized ratio (INR) assay, those aspects of the NOACs are more predictable and routine monitoring is not necessary (3, 4). Occasionally, bioactivity assessment of the NOAC may be vital, e.g., in cases of severe bleeding or thrombotic complications, acute infection, emergency surgery, or suspected overdose. Typical plasma concentrations during treatment range from 50 to 400 g/l (ng/ ml); however, no therapeutic window per se exists (3, 4). Some fatal and major bleeds have been associated with high plasma concentrations of Dabi, especially with renal impairment (5, 6). The effect of NOACs has not been studied in many other coinciding clinical conditions with increased bleeding diathesis, e.g., anemia, thrombocytopenia, or liver failure (7, 8). Coagulation screening assays, including the prothrombin time (PT), INR, and activated partial thromboplastin time (APTT), are widely available. For measuring PT, both Owren and Quick methods are used, the former influenced only by coagulation factors II, VII, and X, instead of the complete extrinsic and common pathways. When spiked to plasma, Dabi dosedependently prolongs the APTT. The prolongation is nonlinear, however, and plateaus at high concentra- 1 Coagulation Disorders Unit, Clinical Chemistry, HUSLAB Laboratory Services, Helsinki University Central Hospital, Helsinki, Finland; 2 Labquality Ltd., Helsinki, Finland; 3 Aplagon Ltd., Helsinki, Finland. * Address correspondence to this author at: Lotta Joutsi-Korhonen, Clinical Chemistry and Hematology, HUSLAB Laboratory Services, POB 340, Helsinki, Finland. Fax ; [email protected]. The information in this article was previously presented as a poster in the ISTH SSC 2012 scientific meeting, June 29, 2012, Liverpool, UK. Received November 4, 2012; accepted January 9, Previously published online at DOI: /clinchem Nonstandard abbreviations: NOAC, novel oral anticoagulant; Dabi, dabigatran; Xa, activated factor X; Riva, rivaroxaban; VKA, vitamin K antagonist; INR, international normalized ratio; PT, prothrombin time; APTT, activated partial thromboplastin time; TT, thrombin time; EQA, external quality assessment; ANOVA, analysis of variance; ISI, international sensitivity index. 1 Copyright (C) 2013 by The American Association for Clinical Chemistry

2 tions (4, 9). The INR response is more linear, but not sensitive enough. Thrombin time (TT) is too sensitive at therapeutic concentrations but is a useful indicator of the presence of Dabi. A Dabi-calibrated TT performed on diluted samples is able to quantify Dabi linearly (9 11). Riva prolongs the APTT in a nonlinear manner but only modestly at trough concentrations (12). INR responses vary widely depending on the thromboplastin assay reagent, but a calibrated anti-xa assay appears to be a suitable method for determination of Riva concentration (13 16). We aimed to assess the effects of NOACs on coagulation screening tests, as well as on more specific assays between different laboratories. This survey was conducted in conjunction with an international coagulation external quality assessment (EQA) round to examine the variety of reagents and the availability of more specific methods in a large number of laboratories. In addition to these spiked samples, we controlled some of the findings in clinical patient samples. Materials and Methods DABI AND RIVA SAMPLES AND METHODS Lyophilized pooled normal human plasma samples spiked with Dabi (Aniara) at final concentrations of 120 and 300 g/l were delivered to 115 European laboratories (Poland 49, Finland 35, Lithuania 17, Latvia 9, Estonia 3, and Iceland 2) by the EQA organization Labquality in July Samples were kindly donated by Boehringer Ingelheim Finland. Laboratories were instructed to reconstitute the plasma with 1.0 ml distilled water, shake thoroughly, and incubate for 30 min at room temperature, mixing periodically, before analysis. Laboratories were then asked to use screening tests and more specific assays (such as TT) available on the two samples. Altogether 73 laboratories (63.5%) reported results; TT was reported by only 11 laboratories. Lyophilized pooled normal human plasma samples spiked with Riva (Technoclone) at final concentrations of 60, 146, and 305 g/l were delivered to 38 laboratories (Finland 18, Norway 9, Ireland 5, Denmark 2, Sweden 2, Estonia 1, and Lithuania 1) by Labquality in January Samples were kindly donated by Bayer Schering Pharma. Laboratories were instructed to reconstitute the plasma with 1.0 ml distilled water, mix by rotating carefully avoiding foaming, and incubate for at least 10 min at room temperature before analysis. Laboratories were then asked to use the screening tests and more specific assays (such as anti-xa) available on the 3 samples; 22 (57.9%) laboratories reported results. In addition, calibrators with Riva concentrations of 0, 15, 60, 100, and 150 g/l were sent to 18 Finnish laboratories for the local anti-xa assay of Riva. Riva-calibrated anti-xa results were reported by 8 laboratories. The different reagents used by the participating laboratories are summarized in Table 1. In total, 23 different coagulation analyzers from 8 different manufacturers were used. The effects of analyzers were not assessed in this survey. To compare the differences between patient samples and spiked samples, we used clinical patient samples from Helsinki University Central Hospital. Ten samples from patients using Dabi and 10 using Riva were assessed with PT, APTT, and a test to determine the drug concentrations (Dabi-calibrated TT or Rivacalibrated anti-xa). Nycotest PT (Axis-Shield, Owren method) and Actin FSL (Siemens Healthcare Diagnostics) reagents were used to measure PT and APTT, respectively. Patients used Dabi for either orthopedic surgery or atrial fibrillation indication; for Riva, all samples were from patients undergoing orthopedic surgery. Time intervals from drug ingestion to blood sampling were unknown. STATISTICAL ANALYSIS We analyzed the effects of the reagents with analysis of variance (ANOVA) and Student t-test. We used CVs to examine the variation between laboratories in different assays. Results are represented as box-plots, where bar lengths indicate the interquartile range (25th to 75th percentile). Outliers (circles in figures) were defined as values bar lengths from the box edge, and extreme outliers (asterisks) as 3.0 bar lengths. In the APTT assay, because the baseline values were not known, we used the upper limit of the local reference interval reported by the laboratory to determine the prolongation of APTT at increasing drug concentrations. In comparisons of different reagents, results obtained with reagents used only in a single laboratory were excluded from statistical analysis. Statistical analyses were performed with Microsoft Excel (2007) and IBM SPSS (18.03) programs. Results Coagulation screening test INR and APTT results from different laboratories for Dabi and Riva samples are summarized in Table 2. DABI AND INR/PT INR values for the plasma samples observed with Dabi (120 and 300 g/l) were reported by 71 laboratories. Thirteen different thromboplastin reagents were used (Table 1). INR values were slightly increased in both plasma samples but were only modestly higher at 300 g/l Dabi. The interlaboratory agreement was not optimal (Table 2). There were differences in sensitivity 2 Clinical Chemistry 59:5 (2013)

3 Laboratory Assessment of Novel Oral Anticoagulants Table 1. Reagents used by the participating laboratories from 10 European countries. Assay and reagent Reagent manufacturer Laboratories in dabigatran round, n Laboratories in rivaroxaban round, n PT/INR Owren method Nycotest PT Axis-Shield 8 2 Owren PT Medirox 10 7 SPA Stago Diagnostica Stago 19 9 Thrombotest Axis-Shield 7 2 Quick method Bioksel System PT Bio-Ksel 2 HemosIL PT Instrumentation Laboratory 1 Innovin Siemens Healthcare Diagnostics 2 2 RecombiPlastin 2G Instrumentation Laboratory 5 0 Roche Thromboplastin Roche Diagnostics 1 0 STA Neoplastin CIPlus Diagnostica Stago 2 Technoplastin HIS Technoclone 1 Thromborel S Siemens Healthcare Diagnostics 12 TriniCLOT PT HTF Tcoag 1 APTT Actin FSL Siemens Healthcare Diagnostics 16 8 Bioksel System APTT Bio-Ksel 2 Dapttin TC Technoclone 1 Grifols APTT Grifols 2 HemosIL APTT SP Instrumentation Laboratory 17 5 Pathromtin SL Siemens Healthcare Diagnostics 9 STA Cephascreen Diagnostica Stago 4 2 STA PTT Automate Diagnostica Stago 18 7 TriniCLOT aptt HS Tcoag 3 TT Comesa Thrombin time Comesa 1 Siemens Thrombin Reagent Siemens Healthcare Diagnostics 7 STA Thrombin Diagnostica Stago 3 Anti-Xa Berichrom Heparin Siemens Healthcare Diagnostics 2 Chromogenix Coamatic Heparin Instrumentation Laboratory 4 Rotachrom Liquid Heparin Diagnostica Stago 1 STA Liquid Anti-Xa Diagnostica Stago 1 between thromboplastin reagents (ANOVA P 0.001). The mean relative change in INR between the two Dabi samples was 1.21 (range ). The type of method used for INR analysis had an impact: the Quick method resulted in both higher INR levels and larger increase in INR with increasing Dabi concentration than the Owren method (P 0.001) (Fig. 1). The PT in seconds was somewhat longer with the higher Dabi concentration, but only 17 laboratories reported results. The prolongation of PT was small; 21.3 s (range s) and 27.5 s (range s) with a wide variation (CV 36.4% and 39.0%, respectively). DABI AND APTT APTT values for the 2 plasma samples with Dabi (120 and 300 g/l) were reported by 72 laboratories. Nine different APTT reagents were used (Table 1). APTT detected the presence of Dabi well, with 70 of 72 labo- Clinical Chemistry 59:5 (2013) 3

4 Table 2. Summary of results reported by different laboratories for Davi- and Riva-spiked plasma samples. INR APTT Anti-Xa n Mean (range) CV(%) n Mean (range), s CV, % n Mean (range), g/l CV, % Dabi, g/l ( ) ( ) ( ) ( ) 18.0 Riva, g/l ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) 33.8 ratories reporting values above their upper reference limits at 120 g/l Dabi. Dabi dose-dependently prolonged the APTT (Table 2). The mean prolongation of the APTT was 23.4 s (range s), exceeding the upper reference limit by 53.8 s at 300 g/l Dabi (range s). However, the interlaboratory variation was wide and reagent dependent (P 0.001). Relative prolongation of APTT was mean 1.36 (range ), and relative increase above the upper reference limit was mean 2.45 (range ) (Fig. 2) at 300 g/l Dabi. DABI AND TT Only 11 laboratories (15% of the 73 laboratories reporting results) reported TT results. All 11 laboratories reported a TT value above the measurement range for both Dabi concentrations (data not shown), thus demonstrating the presence of a thrombin inhibitor. Only 1 laboratory reported results as Dabi concentrations, obtained with the Hemoclot (Aniara) assay: 110 and 320 g/l. Fig. 1. INR results for the Dabi samples obtained with thromboplastin reagents. Boxplot shows median (line), lower and upper quartiles (box), total range (whiskers), outliers (E), and extreme outliers (*). Number of laboratories using the reagent is shown in parentheses. Owren-method reagents exhibited lower INR values and less variation (P 0.001) than Quick-method reagents. SPA, SPA Stago; Nycotest, Nycotest PT; ThromboT, Thrombotest; Trel S, Thromborel S; RecombiPl, RecombiPlastin 2G; BiokselPT, Bioksel System PT; Neoplastin, STA Neoplastin CIPlus. DABI IN CLINICAL PATIENT SAMPLES The mean concentration of Dabi as measured by Dabicalibrated TT in 10 clinical patient plasma samples was 119 g/l (range g/l). For the patient samples, INR was mean 1.08 (range ), whereas for the spiked sample of 120 g/l with the same reagent (Nycotest PT), INR was slightly higher, mean 1.16 (P 0.01) (Fig. 1). For both clinical and spiked samples, the INR was within reference range. Mean APTT for the patient samples was 39.8 s (range s), whereas in the spiked sample of 120 g/l with the same reagent (Actin FSL), APTT was higher, mean 54.2 s (P 0.001) (Fig. 2). This difference is clinically relevant, since in the patient sample group, APTT prolongation was clearly less pronounced, in 1 sample even being within local reference interval. RIVA AND INR/PT INR values were reported by 22 laboratories for the 3 plasma samples with Riva (60, 146, and 305 g/l). Five different thromboplastin reagents were used (Table 1). INR values were only slightly influenced by Riva. The INR value of 1.9 was not reached by any of the laboratories. The agreement between laboratories was rela- 4 Clinical Chemistry 59:5 (2013)

5 Laboratory Assessment of Novel Oral Anticoagulants Fig. 2. APTT results for the Dabi samples obtained with different reagents. Boxplot shows median (line), lower and upper quartiles (box), total range (whiskers), outliers (E), and extreme outliers (*). The upper limit of the local reference interval was used to simulate baseline. Almost all laboratories (70 of 72) reported values above the upper reference limit, even at the lowest Dabi concentration. Number of laboratories using the reagent is shown in parentheses. Ref. range, range of upper limits of local reference intervals; STA PTT, STA PTT Automate; HemosIL, HemosIL APTT SP; Pathromtin, Pathromtin SL; Cephascreen, STA Cephascreen; TriniCLOT, TriniCLOT aptt HS; BiokselAPTT, Bioksel System APTT; Grifols, Grifols APTT. tively good, with low CVs (Table 2). However, there were significant differences in INR responses between reagents (P 0.005). The relative change in INR between the lowest and highest Riva concentrations was mean 1.25 (range ). The method (Quick vs Owren) had little effect on the result (Fig. 3). Only 11 laboratories reported PT results in seconds. The effect of Riva was modest: 23.3 s (range s), 24.6 s ( s), and 31.2 s ( s), with widely varying CVs of 31.3%, 35.3%, and 35.7%, respectively. Increasing Riva concentrations clearly prolonged the PT. For the highest concentration of Riva, the PT exceeded the upper reference limit in all laboratories. Fig. 3. INR results for the Riva samples obtained with thromboplastin reagents. Boxplot shows median (line), lower and upper quartiles (box), total range (whiskers), and outliers (E). Only 2 laboratories used a Quick-method reagent. Number of laboratories using the reagent is shown in parentheses. SPA, SPA Stago; Nycotest, Nycotest PT; ThromboT, Thrombotest. RIVA AND APTT APTT values were reported by 22 laboratories. Four different reagents were used (Table 1). APTT prolongation was dose dependent, but modest for the 3 samples (60, 146, and 305 g/l Riva). However, the increase seemed nonlinear (Fig. 4). Again, there was large variation between laboratories (P 0.001) (Table 2). The reagent used had a distinct effect on the APTT: there was a small but significant difference in the magnitude of response to increasing concentrations of Riva between the different reagents (P 0.05) (Fig. 4). RIVA AND ANTI-Xa CALIBRATED FOR RIVA Eight laboratories reported anti-xa (36% of the 22 laboratories reporting results). Two laboratories reported accurate concentration results for all the samples, whereas 6 underestimated the highest concentration, probably because the highest calibration sample concentration (150 g/l) was clearly too low for these samples (Table 2). Two laboratories used anti-xa reagents with exogenous antithrombin (reagent Berichrom Heparin, in Table 1), whereas 6 did not, without a significant effect on the results. Mean results with added antithrombin were 62.4, 148.3, and g/l and without antithrombin addition 68.2, 129.5, and g/l (P values 0.210, 0.043, and 0.203), respectively. Clinical Chemistry 59:5 (2013) 5

6 Fig. 4. APTT results for the Riva samples obtained with different reagents. Boxplot shows median (line), lower and upper quartiles (box), and total range (whiskers). APTT results were within the reference interval in most reagent groups at the lowest rivaroxaban concentration. At the highest rivaroxaban concentration, almost all laboratories reported values above the upper reference limit. Number of laboratories using the reagent is shown in parentheses. Ref. range: range of upper limits of local reference intervals. STA PTT, STA PTT Automate; HemosIL, HemosIL APTT SP; Cephascreen, STA Cephascreen. RIVA IN CLINICAL PATIENT SAMPLES The mean concentration of Riva as measured by Rivacalibrated anti-xa in 10 clinical patient plasma samples was 63 g/l (range g/l). Mean INR was 1.00 (range ). Results did not differ significantly from the spiked sample of 60 g/l with the same reagent (Nycotest PT), and all INR results were within the reference range. The mean APTT value was 27.3 s (range s). Again, the results did not differ significantly from the spiked sample of 60 g/l with the same reagent (Actin FSL), and all APTT results for both clinical and spiked samples were within the local reference interval. Discussion This survey aimed to assess the effect of NOACs Dabi and Riva on both routine anticoagulant assays and drug-specific assays in relation to a large number of methods and reagents in 10 European countries. Laboratory assessment of the degree of anticoagulation may be needed in special situations, and readily available screening assays are required. Therefore, laboratories should be aware of the drug-specific sensitivity of their own PT, INR, and APTT assays (17). As expected, a wide variety of coagulation screening reagents were used: in total, 13 INR and 9 APTT reagents in numerous combinations with 23 different coagulometers. Here, only a few laboratories (15% and 36%) reported specific assay results on the Dabi and Riva rounds, respectively. The routine anticoagulation test PT/INR is of limited value in measuring the anticoagulant effects of the NOACs. Here, the INR values varied widely between different reagents, especially with Dabi samples, in concordance with previous studies (11, 18). Our new observation was that the Owren-method PT reagents were particularly insensitive to Dabi, and there was little variation (Fig. 1). In addition, in our small group of clinical patients using Dabi, all had INR values in the reference range measured with the Owren method. In contrast, with the Quick-method reagents, the range was 3-fold even at the lower concentration of Dabi. Our data contradict the suggested INR levels of 1.5 (Owren) and 2.0 (Quick) for Dabi overdose screening (18). Furthermore, point-of-care INR devices may in some cases provide falsely high readings when used with patients on Dabi (19). With Riva samples, the INR value of 1.9 was not reached by any of the laboratories even at the highest concentration of 305 g/l. Neoplastin CI Plus reagent (Diagnostica Stago), owing to its sensitivity, has been suggested to be suitable as a qualitative test reagent (13, 20). According to this survey, that is currently of limited practical value, since of the laboratories reporting results in the Riva round, none used that reagent. When assessing the effects of NOACs, the PT rather than the INR has been suggested, as the INR is calibrated for VKAs only. The differences between reagents are enlarged with thromboplastin reagents of high international sensitivity index (ISI) (16, 21). The ISI values of the reagents in our study were fairly similar and close to 1.0 (according to the manufacturers range, ). The differences in PT results were slightly larger with increasing drug concentrations for both NOACs. Recently, there have been efforts to standardize PT methods by creating an ISI for Riva, analogous to the ISI for VKAs. However, this calibration applies only to the Quick method and is available in only a limited number of laboratories (21). The INR/ PT, being widely available, is likely to be used to measure NOACs, despite not being optimal. Here, INR results were similarly inert between the clinical patient samples and in vitro samples with the low drug concentration. 6 Clinical Chemistry 59:5 (2013)

7 Laboratory Assessment of Novel Oral Anticoagulants All the APTT reagents detected the presence and dose-dependence of Dabi well. The APTT was prolonged even at the lower concentration almost uniformly, but with a 2-fold variance between reagents. Of the most frequently used reagents, Actin FSL (Siemens Healthcare Diagnostics) was one of the least responsive, whereas Stago PTT (Stago) responded more vigorously (Fig. 2). These reagent-related findings agree with those of Lindahl et al. (18) but are at odds with those of Douxfils et al. (10). The APTT is recommended as an indicator test of the presence of Dabi because of its high sensitivity, a statement supported by the results from our spiked samples (17). However, in patient samples, APTT was less sensitive than in spiked samples. Indeed, clinicians must exercise caution when using the APTT to assess the effects of Dabi, as APTT may be normal even at therapeutic concentrations (22). The dose response is curvilinear and clearly affected by the clinical situation. The effect of Riva on the APTT was modest and nonlinear, in accord with previous studies (13). Specific assays were available in relatively few centers. TT results were provided by 11 laboratories (15%), but only 1 of them had the calibrated diluted TT assay available to measure the concentration with accurate results. Anti-Xa was able to measure the Riva concentration, when specifically calibrated, which is compatible with previous studies (16). Generally, anti-xa assays are readily available and widely used and can be relatively easily calibrated to accurately measure factor Xa inhibitor concentrations. Nonetheless, hospital laboratories are also validating suitable tests for detection of anticoagulant effects for routine use (11, 13, 15, 16, 18, 23). Increased availability of calibrants and further EQA rounds are needed to help laboratories set up and maintain the specific assays. When examining the results of our survey, the limitations of the sample materials should be recognized. Not being optimal, lyophilized plasma may reduce coagulation factor activities and prolong PT and APTT values compared with fresh plasma (24, 25). However, the samples sent to different laboratories were produced in a similar standardized manner. Furthermore, currently the use of patient plasma rather than spiked normal plasma would be possible only in a limited number of laboratories, and with the NOACs, such material is still sparse. Thus, applying these data to clinical practice should be done with caution. The samples from patients using NOACs differ somewhat from the spiked samples. At similar drug concentrations, the APTT was less sensitive to Dabi in patient samples than in spiked samples, a fact bearing clinical significance as APTT does not necessarily alert for the presence of Dabi. As significant variation between PT and APTT reagents was observed with in vitro spiked normal plasma, the variation between patient samples is also likely to exceed this amount in vivo. The clinical situations with varying coagulation status, i.e., renal or liver impairment, infection, surgery, bleeding, or thrombosis, modify both the results of global coagulation assays and the efficacy of NOACs, further complicating the interpretation. Unfortunately, there are still limited data on laboratory monitoring of patients with bleeds (6, 26). As routine monitoring is not performed, experience of the plasma concentrations of the NOACs at desired or sub-/supratherapeutic doses will not accumulate. Most often, the knowledge on NOAC concentrations is needed suddenly during on-call hours. In the absence of specific antidotes, nonspecific bypassing agents (i.e., prothrombin complex concentrates and recombinant FVIIa) may provide clinical benefit (27, 28). However, these agents influence both PT and APTT, which may not reflect the clinical efficacy of the reversal, further complicating the interpretation. Instead, calibrated TT and anti-xa assays offer specific and accurate concentration monitoring under special situations. In conclusion, routine anticoagulation assays PT, INR, and APTT are highly reagent dependent and do not provide accurate measurement of the activity of NOACs. However, each laboratory should test the sensitivity of their assays to NOACs. We also believe that laboratories need to validate calibrated specific assays more widely. Therefore, EQA rounds for NOACs must be established to standardize and harmonize laboratory practices, as an increasing number of laboratories will apply these tests for clinical practice. Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article. Authors Disclosures or Potential Conflicts of Interest: Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest: Employment or Leadership: None declared. Consultant or Advisory Role: None declared. Stock Ownership: None declared. Honoraria: None declared. Research Funding: L. Joutsi-Korhonen, Boehringer Ingelheim Finland, Bayer Schering Pharma Finland. Expert Testimony: None declared. Other Remuneration: T. Helin, Leo Pharma Ltd. Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or preparation or approval of manuscript. Acknowledgments: Bayer Finland and Boehringer Ingelheim Finland are acknowledged for providing the Riva and Dabi plasma samples, respectively. Laboratories participating in these EQA surveys are acknowledged for their efforts. Clinical Chemistry 59:5 (2013) 7

8 1. Connolly SJ, Ezekowitz MD, Yusuf S, Eikelboom J, Oldgren J, Parekh A, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med 2009;361: Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med 2011;365: Mueck W, Lensing AW, Agnelli G, Decousus H, Prandoni P, Misselwitz F. Rivaroxaban: population pharmacokinetic analyses in patients treated for acute deep-vein thrombosis and exposure simulations in patients with atrial fibrillation treated for stroke prevention. Clin Pharmacokinet 2011;50: Stangier J, Rathgen K, Stähle H, Gansser D, Roth W. The pharmacokinetics, pharmacodynamics and tolerability of dabigatran etexilate, a new oral direct thrombin inhibitor, in healthy male subjects. Br J Clin Pharmacol 2007;64: Mack DR, Kim JJ. Pharmacokinetic and clinical implications of dabigatran use in severe renal impairment for stroke prevention in nonvalvular atrial fibrillation. Ann Pharmacother 2012;46: Lillo-Le Louët A, Wolf M, Soufir L, Galbois A, Dumenil AS, Offenstadt G, et al. Life-threatening bleeding in four patients with an unusual excessive response to dabigatran: implications for emergency surgery and resuscitation [Letter]. Thromb Haemost 2012;108: Favaloro EJ, Lippi G, Koutts J. Laboratory testing of anticoagulants: the present and the future. Pathology 2011;43: Salmela B, Joutsi-Korhonen L, Armstrong E, Lassila R. Active online assessment of patients using new oral anticoagulants: bleeding risk, compliance, and coagulation analysis. Semin Thromb Hemost 2012;38: van Ryn J, Stangier J, Haertter S, Liesenfeld KH, Wienen W, Feuring M, et al. Dabigatran etexilate: a novel, reversible, oral direct thrombin inhibitor: interpretation of coagulation assays and reversal of anticoagulant activity. Thromb Haemost 2010; 103: Douxfils J, Mullier F, Robert S, Chatelain C, Chatelain B, Dogné JM. Impact of dabigatran on a References large panel of routine or specific coagulation assays. Laboratory recommendations for monitoring of dabigatran etexilate. Thromb Haemost 2012;107: Harenberg J, Giese C, Marx S, Krämer R. Determination of dabigatran in human plasma samples. Semin Thromb Hemost 2012;38: Mani H, Hesse C, Stratmann G, Lindhoff-Last E. Rivaroxaban differentially influences ex vivo global coagulation assays based on the administration time. Thromb Haemost 2011;106: Hillarp A, Baghaei F, Fagerberg Blixter I, Gustafsson KM, Stigendal L, Sten-Linder M, et al. Effects of the oral, direct factor Xa inhibitor rivaroxaban on commonly used coagulation assays. J Thromb Haemost 2011;9: Samama MM, Contant G, Spiro TE, Perzborn E, Guinet C, Gourmelin Y, et al. Evaluation of the anti-factor Xa chromogenic assay for the measurement of rivaroxaban plasma concentrations using calibrators and controls. Thromb Haemost 2012;107: Samama MM, Martinoli JL, LeFlem L, Guinet C, Plu-Bureau G, Depasse F, et al. Assessment of laboratory assays to measure rivaroxaban: an oral, direct factor Xa inhibitor. Thromb Haemost 2010;103: Barrett YC, Wang Z, Frost C, Shenker A. Clinical laboratory measurement of direct factor Xa inhibitors: anti-xa assay is preferable to prothrombin time assay. Thromb Haemost 2010;104: Baglin T, Keeling D, Kitchen S. Effects on routine coagulation screens and assessment of anticoagulant intensity in patients taking oral dabigatran or rivaroxaban: guidance from the British Committee for Standards in Haematology. Br J Haematol 2012;159: Lindahl TL, Baghaei F, Blixter IF, Gustafsson KM, Stigendal L, Sten-Linder M, et al. Effects of the oral, direct thrombin inhibitor dabigatran on five common coagulation assays. Thromb Haemost 2011;105: van Ryn J, Baruch L, Clemens A. Interpretation of point-of-care INR results in patients treated with dabigatran. Am J Med 2012;125: Samama MM, Guinet C. Laboratory assessment of new anticoagulants. Clin Chem Lab Med 2011; 49: Tripodi A, Chantarangkul V, Guinet C, Samama MM. The International Normalized Ratio calibrated for rivaroxaban has the potential to normalize prothrombin time results for rivaroxabantreated patients: results of an in vitro study [Letter]. J Thromb Haemost 2011;9: Smith TW, Zypchen L, Carter CJ, Tran A, Colley P, Gin K et al. Effects of dabigatran and rivaroxaban on routine and specialized coagulation assays: a study using actual patient samples [Abstract]. Blood 2012;120:A Asmis LM, Alberio L, Angelillo-Scherrer A, Korte W, Mendez A, Reber G, et al. Rivaroxaban: quantification by anti-fxa assay and influence on coagulation tests: a study in 9 Swiss laboratories. Thromb Res 2012;129: Samama MM, Contant G, Spiro TE, Perzborn E, Flem LL, Guinet C et al. Evaluation of the prothrombin time for measuring rivaroxaban plasma concentrations using calibrators and controls: results of a multicenter field trial. Clin Appl Thromb Hemost 2012;18: Martinaud C, Civadier C, Ausset S, Verret C, Deshayes AV, Sailliol A. In vitro hemostatic properties of French lyophilized plasma. Anesthesiology 2012;117: Warkentin TE, Margetts P, Connolly SJ, Lamy A, Ricci C, Eikelboom JW. Recombinant factor VIIa (rfviia) and hemodialysis to manage massive dabigatran-associated postcardiac surgery bleeding [Letter]. Blood 2012;119: Eerenberg ES, Kamphuisen PW, Sijpkens MK, Meijers JC, Buller HR, Levi M. Reversal of rivaroxaban and dabigatran by prothrombin complex concentrate: a randomized, placebo-controlled, crossover study in healthy subjects. Circulation 2011;124: Marlu R, Hodaj E, Paris A, Albaladejo P, Crackowski JL, Pernod G. Effect of non-specific reversal agents on anticoagulant activity of dabigatran and rivaroxaban: a randomized crossover ex vivo study in healthy volunteers. Thromb Haemost 2012;108: Clinical Chemistry 59:5 (2013)