key: cord-0742829-8b8whwq2
authors: Amiral, Jean; Amiral, Cédric; Dunois, Claire
title: Optimization of Heparin Monitoring with Anti-FXa Assays and the Impact of Dextran Sulfate for Measuring All Drug Activity
date: 2021-06-21
journal: Biomedicines
DOI: 10.3390/biomedicines9060700
sha: 45eb44dc917e8e6dd5fea6fb55108320566f2704
doc_id: 742829
cord_uid: 8b8whwq2

Heparins, unfractionated or low molecular weight, are permanently in the spotlight of both clinical indications and laboratory monitoring. An accurate drug dosage is necessary for an efficient and safe therapy. The one-stage kinetic anti-FXa assays are the most widely and universally used with full automation for large series, without needing exogenous antithrombin. The WHO International Standards are available for UFH and LMWH, but external quality assessment surveys still report a high inter-assay variability. This heterogeneity results from the following: assay formulation, designed without or with dextran sulfate to measure all heparin in blood circulation; calibrators for testing UFH or LMWH with the same curve; and automation parameters. In this study, various factors which impact heparin measurements are reviewed, and we share our experience to optimize assays for testing all heparin anticoagulant activities in plasma. Evidence is provided on the usefulness of low molecular weight dextran sulfate to completely mobilize all of the drug present in blood circulation. Other key factors concern the adjustment of assay conditions to obtain fully superimposable calibration curves for UFH and LMWH, calibrators’ formulations, and automation parameters. In this study, we illustrate the performances of different anti-FXa assays used for testing heparin on UFH or LMWH treated patients’ plasmas and obtained using citrate or CTAD anticoagulants. Comparable results are obtained only when the CTAD anticoagulant is used. Using citrate as an anticoagulant, UFH is underestimated in the absence of dextran sulfate. Heparin calibrators, adjustment of automation parameters, and data treatment contribute to other smaller differences.

Heparin and its derivatives, including unfractionated heparin (UFH), low molecular weight heparin (LMWH), and fondaparinux, comprise a major group of anticoagulants with multiple indications in various clinical situations associated with occurrence or prevention of thrombosis [1] [2] [3] . Since its discovery [4, 5] , heparin has been used for the treatment of thrombotic diseases, and this has reversed their prognosis [6] [7] [8] [9] . Therapy monitoring is required for drug dosage adjustment [10, 11] , especially when clearance is impaired [12, 13] , or in the presence of heparin resistance [14] [15] [16] [17] . Rare and severe side effects can develop such as heparin-induced thrombocytopenia (HIT) [18, 19] ; however, heparin remains to be the anticoagulant of choice in many critical circumstances [1, 8, [20] [21] [22] and has anticoagulant activity through additional mechanisms [23] [24] [25] . UFH and LMWH drug dosages need to be accurately adjusted for each treated patient [2, 6, 13, 26, 27] . Inappropriate drug adjustment can generate severe and life-threatening thrombotic or bleeding complications. In addition, measurement accuracy is important for therapy monitoring. Many assays have been developed over time for monitoring heparin [11, 28, 29] , such as the activated clotting time (ACT) used in cardiology [28] [29] [30] and the activated partial thromboplastin time (APTT) performed on citrated plasma [31, 32] . APTT is still the first line method in many countries, despite its

Normal citrated plasmas and plasma pools were supplied frozen by Precision Biologic Inc. (Halifax, Canada) , and stored at < −70 • C until use. Plasmas from hospitalized patients treated with heparin for post-surgery thrombosis prevention, with either UFH or LMWH, were obtained from Beaujon University Hospital (Clichy, France), as the left-over residual plasma from an ongoing clinical study and obtained in agreement with CLSI. According to the primary study protocol, blood was collected either on 0.109 M citrate or CTAD, and plasma was decanted following 20 min centrifugation (2000× g), at room temperature (RT), and then stored frozen at <−70 • C until use. Plasmas were thawed for 5 min in a water bath at 37 • C just before use.

Heparin anti-FXa kinetic chromogenic assays were obtained from various manufacturers: STA-Liquid Anti-Xa (reagent A), STA-Multi-Hep Calibrator, STA-Quality HNF/UFH, and STA-Quality HBPM/LMWH from Diagnostica Stago (Asnières, France); HemosIL Liquid Anti-Xa (reagent B), HemosIL Heparin Calibrators, HemosIL UF Heparin Controls, and HemosIL LMW Heparin Controls from IL-Werfen (Le Pré Saint Gervais, France); INNO-VANCE Heparin (reagent C), INNOVANCE Heparin Calibrator, INNOVANCE Heparin UF Controls, and INNOVANCE Heparin LMW Controls from Siemens (Aubervilliers, France); BIOPHEN Heparin LRT (reagent D), the 2-stage assays BIOPHEN Heparin Anti-Xa-2-stages and BIOPHEN Heparin Anti-IIa-2-stages, BIOPHEN UFH calibrator and controls, and BIOPHEN LMWH calibrator and controls, were from HYPHEN BioMed (Neuville sur Oise, France). For each system, calibrators were tested in duplicate. IL-Werfen and Diagnostica Stago propose plasma calibrators prepared by mixing or combining UFH and LMWH with traceability to International Standards and claim a hybrid calibration curve which can be used irrelevantly for UFH or LMWH. Siemens and Hyphen BioMed propose plasma calibrators were prepared with LMWH only, and assay conditions allowing to obtain full superimposition of UFH and LMWH calibration curves; different types of LMWH were used for Siemens and Hyphen BioMed heparin calibrators. Lastly, Siemens, IL-Werfen, and HYPHEN BioMed anti-Xa reagents (reagents B, C and D) contained dextran sulfate (DS), whilst that from Diagnostica Stago (reagent A) did not.

Reference materials used for UFH or LMWH were the WHO International Standards (IS), obtained from the National Institute for Biological Standards and Controls (NIBSC, Potters Bar, UK), IS 11/176 for LMWH (1068 anti-FXa and 342 anti-FIIa IU per ampoule) and IS 07/328 for UFH (2145 IU per ampoule). These ISs were restored as indicated on the product instructions for use, and a stock solution was prepared at exactly 100 International Units (IU)/mL in a 0.05 M Tris, 0.15 M NaCl, 1% BSA buffer, at pH 7.40 (TBSA). This stock solution was used for preparing UFH or LMWH concentration ranges in the Cryocheck plasma pool, from 0 to 1.8 IU/mL: first a twenty-fold concentrated range was prepared in TBSA (0 to 36 IU/mL); then, 50 µL of each stock solution was spiked in 950 µL of Cryocheck citrate plasma pool to obtain UFH or LMWH concentrations in plasma ranging from 0.00 to 1.80 IU/mL. All spiked plasmas had the same matrix, i.e., 95% Cryocheck plasma pool and 5% TBSA.

Each heparin anti-FXa assay was used according to the associated manufacturer's instrument: Diagnostica Stago reagent A with STA-R Max (Diagnostica Stago, Asnières, France); IL-Werfen reagent B with ACL-Top 550 (IL Werfen, Le Pré St Gervais, France); Siemens reagent C with CS-5100 (an automated instrument from Sysmex, Kobe, Japan, and distributed by Siemens Healthineers, Aubervilliers, France); HYPHEN BioMed reagent D with the Sysmex CS-5100 instrument (Sysmex, Kobe, Japan); the BIOPHEN anti-FIIa and anti-Xa 2-stage assays were used with the CS-2400 instrument (Sysmex, Kobe, Japan). Reagents were used respecting strictly the recommended manufacturers' protocols. HY-PHEN BioMed reagents, proposed for multiplatform applications, were used with the CE marked protocols developed and validated for CS-5100. All assayed plasmas were tested undiluted (reagent B) or diluted, as claimed in the instructions for use for each assay. Plasma diluent was Owren Veronal Buffer (reagents A and C) or 0.15 M sodium chloride (reagent D).

The citrate plasma pool supplemented with either UFH or LMWH ISs was assayed in duplicate for each reagent-instrument combination (reagents A, B, C, and D), comparatively to the manufacturers' calibrators. Each proposed manufacturer's heparin calibrator and the UFH or LMWH WHO ISs spiked in plasma, for concentrations from 0.00 to 1.80 IU/mL as described before, were tested with each anti-FXa reagent-instrument combination (reagents A, B, C, and D). For each combination, the 3 calibration curves obtained (heparin assay manufacturer's calibrator, UFH IS, and LMWH IS) were compared.

All plasmas, from UFH or LMWH treated patients, and anticoagulated with citrate or CTAD, were tested using the 4 anti-FXa assay combinations and correlation diagrams were established. Then, a subanalysis was performed for the different groups: plasmas from UFH or LMWH treated patients; use of citrate or CTAD anticoagulant and four groups were obtained, i.e., UFH-citrate, UFH-CTAD, LMWH-citrate, and LMWH-CTAD.

Heparin calibrators from the various manufacturers were assayed in duplicate with the 2-stage anti-FXa or anti-FIIa assays with the CS-2400 instrument and compared with the calibrations obtained with UFH or LMWH WHO ISs spiked in plasma. This measurement allowed analyzing the content of each plasma calibrator by establishing the anti-FXa/anti-FIIa ratios: UFH has a ratio of 1.00, whilst depending on the branded material LMWH has a ratio from 1.6 to 9.7 [4, 43] .

Statistics were performed using Analyse-it software version 5.11 (Analyse-it Software, Leeds, UK). Pearson's correlation diagrams and ordinary least square regression analyses were performed. The heparin concentrations measured in patients' plasmas were compared using the Friedman's test; to determine whether any of the differences between the medians were statistically significant, we compared the p-value at the 1% significance level to assess the null hypothesis. According to sample size and the non-normal distribution of values, Friedman's test was more appropriate than the ANOVA multivariate analysis for this study.

The calibration curves obtained with each anti-FXa combination for the manufacturer's calibrator and the UFH or LMWH WHO ISs are shown in Figure 1 . Superimposition between the manufacturer calibration curves and those obtained with the WHO International UFH or LMWH standards are globally good, although some slight deviations can be seen depending on the system used.

The calibration curves obtained with each anti-FXa combination for the manufacturer's calibrator and the UFH or LMWH WHO ISs are shown in Figure 1 . Superimposition between the manufacturer calibration curves and those obtained with the WHO International UFH or LMWH standards are globally good, although some slight deviations can be seen depending on the system used. In combination reagent A, UFH IS calibration lacks linearity, especially in the low range, and absorbances are above the manufacturer's calibration, which can result in underestimation of UFH concentrations, especially for low heparin concentrations. Superimposition is better in the high range. The lack of linearity (R 2 = 0.979) contributes to underestimating UFH concentration, i.e., the measured absorbance, when plotted against the manufacturer's calibrator gives significantly lower UFH measurements than when plotted against the UFH WHO IS. In combination reagent B, UFH and LMWH IS calibrations have an acceptable superimposition, and the manufacturer's calibration has absorbances slightly lower than ISs. This can also contribute to underestimating UFH or LMWH concentrations. In combination reagent C, superimposition is acceptable; the assay calibrator behaves similar to UFH IS and is slightly above that of LMWH IS; this low deviation tends In combination reagent A, UFH IS calibration lacks linearity, especially in the low range, and absorbances are above the manufacturer's calibration, which can result in underestimation of UFH concentrations, especially for low heparin concentrations. Superimposition is better in the high range. The lack of linearity (R 2 = 0.979) contributes to underestimating UFH concentration, i.e., the measured absorbance, when plotted against the manufacturer's calibrator gives significantly lower UFH measurements than when plotted against the UFH WHO IS. In combination reagent B, UFH and LMWH IS calibrations have an acceptable superimposition, and the manufacturer's calibration has absorbances slightly lower than ISs. This can also contribute to underestimating UFH or LMWH concentrations. In combination reagent C, superimposition is acceptable; the assay calibrator behaves similar to UFH IS and is slightly above that of LMWH IS; this low deviation tends to slightly underestimate LMWH; superimposition for all the curves is also obtained for the combination reagent D.

Differences in heparin concentrations measured among the assays are higher for UFH, especially for low concentrations, and when DS is not present in the assay system. A better accuracy and exactness are obtained when heparin plasma calibrator concentrations are regularly distributed over the dynamic range rather than concentrated in the lower part, as for combination reagent B.

Correlation studies are performed on the global group, using plasmas obtained from UFH or LMWH treated patients and anticoagulated with citrate or CTAD. Figure 2 shows the correlation diagrams, for the cross-comparison of anti-FXa manufacturers' devices: reagents D vs. B, reagents D vs. C, reagents B vs. C, reagents D vs. A, reagents C vs. A, and reagents B vs. A. The ordinary least square fits and correlation coefficients are shown on each graph. The reagents containing DS (reagents B, C, and D) present global acceptable correlations between them on the full assay range, whilst differences tend to be higher when these reagents are compared with reagent A, designed without DS, especially in the low range. This is incompletely reflected by the Pearson's correlation coefficients calculated on the full dynamic assay range (Figure 1 ), as r values are always >0.90. Then, a more detailed analysis is necessary.

to slightly underestimate LMWH; superimposition for all the curves is also obtained for the combination reagent D.

Differences in heparin concentrations measured among the assays are higher for UFH, especially for low concentrations, and when DS is not present in the assay system. A better accuracy and exactness are obtained when heparin plasma calibrator concentrations are regularly distributed over the dynamic range rather than concentrated in the lower part, as for combination reagent B.

Correlation studies are performed on the global group, using plasmas obtained from UFH or LMWH treated patients and anticoagulated with citrate or CTAD. Figure 2 shows the correlation diagrams, for the cross-comparison of anti-FXa manufacturers' devices: reagents D vs. B, reagents D vs. C, reagents B vs. C, reagents D vs. A, reagents C vs. A, and reagents B vs. A. The ordinary least square fits and correlation coefficients are shown on each graph. The reagents containing DS (reagents B, C, and D) present global acceptable correlations between them on the full assay range, whilst differences tend to be higher when these reagents are compared with reagent A, designed without DS, especially in the low range. This is incompletely reflected by the Pearson's correlation coefficients calculated on the full dynamic assay range (Figure 1 ), as r values are always >0.90. Then, a more detailed analysis is necessary. Differences between assays are higher for UFH samples than for LMWH. The correlation line tendency for reagents A and B shows that heparin concentrations tend to be underestimates, as compared with reagents C and D, and as expected from the calibration curves' analysis.

The mean values for the different subgroups of plasmas tested (UFH or LMWH), and anticoagulated with citrate or CTAD, are shown in Table 1 . The statistical analysis with the Friedman's test shows that at a significance level of 1%, measured concentrations of citrate anticoagulated samples differ: for A compared with C and D, but not B; for B compared with C and D, but not A; for C compared with A and B, but not D; for D compared with A and B, but not C. Table 2 presents the p-values for the citrate or CTAD subgroups. When the CTAD anticoagulated plasmas were tested, the significant differences remained only for: reagents A with C; reagents B with C and D; reagents C with A and B; reagents D with B. Mean heparin concentrations are lower when measured with reagents A and B than with reagents C and D. Differences result partly from the presence of dextran sulfate in the assay formulation, and partly from the manufacturers' heparin calibrator deviations from UFH or LMWH ISs. Interestingly, no significant differences are noted for the citrate or the CTAD groups between assays for reagents A and B, designed with a combination of UFH and LMWH for calibrators, and between reagents C and D, designed with the use of only LMWH. 

To understand and illustrate which major factors are responsible for heparin concentration differences between assays, the correlation diagrams were drawn by identifying each patient's plasma group. Figure 3 and Table 3 show, for each combination, the correlation diagrams with a separate identification of each subgroup: UFH-citrate, UFH-CTAD, LMWH-citrate, and LMWH-CTAD. These diagrams clearly show that the differences are mainly due to UFH-citrate, and to a lesser extent LMWH-citrate. When CTAD is used as an anticoagulant, a much better coherence of heparin concentrations measured is obtained for all assays.

To confirm the factors explaining the heparin concentration differences measured with the different reagents, especially when designed with or without DS, correlations were analyzed separately for each group of plasma samples. Figure 4 presents the correlation diagrams for UFH or LMWH plasmas anticoagulated either with citrate or CTAD, for the comparison of reagents A and D. Similar correlations are obtained for reagent A as compared with reagents B or C (data not shown).

tration differences between assays, the correlation diagrams were drawn by identifying each patient's plasma group. Figure 3 and Table 3 show, for each combination, the correlation diagrams with a separate identification of each subgroup: UFH-citrate, UFH-CTAD, LMWH-citrate, and LMWH-CTAD. These diagrams clearly show that the differences are mainly due to UFH-citrate, and to a lesser extent LMWH-citrate. When CTAD is used as an anticoagulant, a much better coherence of heparin concentrations measured is obtained for all assays. . When r > 0.95, correlation between assays looks acceptable, and measurements differ when r ≤ 0.95. The highest differences are observed for UFH-citrate for the comparisons between anti-FXa reagents containing dextran sulfate (reagents B, C, and D) with that without (reagent A). The highest dispersion of results between reagents A and D concerns UFH samples collected with the citrate anticoagulant. When the same samples are collected with the CTAD anticoagulant, a much better correlation is obtained which was also the case for reagent A as compared with reagents B or C, whilst correlations were better when reagents B, C, and D were compared (r > 0.95). The highest dispersion of results between reagents A and D concerns UFH samples collected with the citrate anticoagulant. When the same samples are collected with the CTAD anticoagulant, a much better correlation is obtained which was also the case for reagent A as compared with reagents B or C, whilst correlations were better when reagents B, C, and D were compared (r > 0.95).

These data suggest that UFH is partially inhibited ex vivo by heparin neutralizing proteins, and its concentration is underestimated when reagent A is used. The presence of DS prevents this inhibition.

The mean heparin concentrations measured with the four anti-FXa assays' combinations were analyzed for each of the subgroups treated with either UFH or LMWH, and anticoagulated with citrate or CTAD. Table 4 shows the values obtained for each subgroup, underlining the important impact of the anticoagulant used for blood collection, and of the assay design without DS, on the heparin concentrations measured, especially for the low concentration range. Other differences observed with the different assays and the different groups can be explained by the calibration curves biases, when compared with the UFH or LMWH ISs reference curves. An additional impact is noted for reagent B in the low UFH range, and in a lesser extent for reagent C, as the manufacturers' calibration curves deviate slightly from ISs for UFH or LMWH, as shown in Figure 1 . 

As heparin anti-FXa reagents are indicated for testing all heparin types, manufacturer proposed superimposed curves or hybrid curves can be used irrelevantly for testing UFH or LMWH with the same heparin calibrator. We evaluated the specific FXa or FIIa anticoagulant activity of each heparin plasma calibrator, with the two-stage assays. The specific anti-FXa to anti-FIIa ratios were calculated for each calibrator. The results are presented in Table 5 .

UFH has an anti-FXa/anti-FIIa ratio of 1.00 and the different LMWHs have ratios ranging from 1.6 to 9.7, partly dependent on the MW size distribution, and on the pentasaccharide density. From these data, it can be deduced that the Stago heparin calibrator set contains two calibrators (Calibrators 2 and 4) obtained with UFH and two calibrators obtained with LMWH (Calibrators 3 and 5), whilst all the IL HemosIL heparin calibrators contain a mixture of UFH and LMWH. Siemens and HYPHEN BioMed heparin calibrators are homogenous and prepared with only LMWH. The anti-FXa/anti-FIIa ratio shows that different LMWHs are used: this ratio (mean of 2.10) is lower for the Siemens calibrators, similar to that of certoparin or fragmin, and higher for HYPHEN BioMed (mean of 4.02), similar to that of enoxaparin [43] . The WHO International Standard for LMWH 11/176 has an anti-FXa/FIIa ratio of 3.12 (1068 IU for anti-FXa and 342 IU for anti-FIIa). The use of a single heparin calibration curve appropriateness for measuring UFH or LMWH depends, first, on the accuracy of the superimposition of curves obtained with the corresponding ISs. Both WHO standards were proposed for the two heparin types, UFH or LMWH, which present different inhibition kinetic characteristics.

Recent reports and ECAT surveys have pointed out the variability of heparin measurements using various commercially available anti-FXa assays [55] [56] [57] . This debate has been reactivated with the extended use of heparin therapy in COVID-19 patients, with accumulation when drug clearance is decreased, or drug resistance when strong inflammation, NETs, and histones are present [14, 15, 17, 66] . Then, heparin concentrations measured can deviate from expected values. Some studies suggest an overestimation of measured heparin concentrations, especially for UFH, when DS is used for anti-FXa assay formulation, whilst other reports support this technical choice as providing the most accurate estimation of anticoagulant activity [55, 59, 60] . This debate also questions the appropriate residual heparin concentration following neutralization with protamine sulfate in extra-corporeal circulation, when the rebound effect is observed [59, [67] [68] [69] . Studies using heparinase or heparanase have shown that the measured residual heparin does not always match with the anticoagulant activity measured [70, 71] , and the presence of DS can provide a more reliable measurement. As developers of heparin measurement methods, we have analyzed these different reports and evaluated the various anti-FXa assays' performances through our experience. In a recent study, Hollestelle et al. already reported the high variability of UFH and LMWH measurements depending on the anti-FXa assay used, and our study confirmed their results using a different and complementary investigation [57] . Hollestelle et al. analyzed the multicentric and longitudinal results from an external quality survey performed on a few samples spiked with UFH or LMWH and provided to all centers participating in an external quality assessment program. Our study is monocentric and performed in the real-life conditions by testing plasmas from UFH or LMWH treated patients and aims to investigate the causes for discrepancies. We show that the presence of DS is important for assay reactivity, in line with former studies, and the heparin calibrator formulation used for UFH and LMWH is also essential, a better linearity being obtained when only LMWH is used.

In this study, we analyzed the various factors impacting the measurement of UFH or LMWH in plasma using four major commercially available anti-FXa assays. We investigated the incidence of reagents' formulations and that of heparin calibrators, tested by comparison with the UFH or LMWH WHO International Standards. Three of the anti-FXa assays (reagents B (IL-Werfen), C (Siemens-Innovance), and D (HYPHEN BioMed-Biophen)) contained DS, to mobilize all plasma heparin and to avoid its inhibition [61, 62] , whilst the fourth one (reagent A (Stago-STA Liquid Anti-Xa)) did not. Tested plasma samples, from UFH or LMWH treated patients, were provided by Beaujon University' Hospital (Clichy, France) anticoagulated with citrate or CTAD. As these plasmas were the left-over tubes from a clinical study, blood was collected using citrate or CTAD, depending on the patient tested. Nowadays, CTAD tends to be used less and less, as laboratories wish to use only one anticoagulant tube type for standardizing hemostasis testing, although it was developed to increase blood and plasma stability for heparin testing [65, 72] . CTAD prevents platelet activation and release of heparin neutralization proteins. The assay-to-assay comparison was performed on the global group including 68 plasmas and analyzed for each subgroup of heparin type and anticoagulant used. Then, four groups were obtained: UFH-citrate (n = 17); UFH-CTAD (n = 11); LMWH-citrate (n = 25); LMWH-CTAD (n = 15). Globally, correlations were acceptable on the global group among reagents B, C, and D, and there was a higher dispersion of values and a poorer correlation between reagent A and the three others, especially in the low range. When subgroups were analyzed separately, it was obvious that major deviations were observed for UFH-citrate. A better homogeneity was obtained when samples were collected using CTAD, probably because PF4 was then present at a lower concentration. These results strongly suggest that UFH is partly neutralized ex vivo by platelet released proteins, and that this inhibition is prevented using CTAD anticoagulant. When reagents contain DS, this inhibition does not occur, and heparin concentrations measured match better with those present in blood circulation, as reported by various studies [60] [61] [62] 71] . For reagent D, we confirmed that adding PF4, up to a final concentration of 1.0 µg/mL, had no impact on measured values (data not shown).

When comparing the mean heparin concentrations and standard deviations, the lowest values were obtained with reagents A and B, especially for UFH. Although reagents B, C, and D all contained DS, differences were observed among the mean concentrations measured, especially for reagent B as compared with reagents C and D. The calibration curve used for measuring irrelevantly plasma samples containing UFH or LMWH contributes by explaining these differences. Clinical laboratories need a 24/7 available anti-FXa assay for measuring heparin and monitoring treated patients, an analysis that is often requested in emergencies. The heparin type is not always known, and therefore using a single calibration curve is necessary. This procedure needs to be carefully established and validated. Concerning traceability, the WHO proposes two separate International Standards for UFH or LMWH, and manufacturers need to establish anti-FXa assay conditions to obtain superimposable calibration curves for all heparin types. When the correct conditions are fulfilled, calibration curves obtained with UFH or LMWH in plasma can be used without any difference. The heparin calibrations proposed by the various manufacturers differ significantly. Reagents C and D use a calibration curve prepared with only LMWH spiked in a platelet-depleted plasma pool. However, LMWH, used by both manufacturers for reagents C and D, differs as demonstrated by the anti-FXa/anti-FIIa ratios. LMWH is similar to certoparin or fragmin for reagent C and similar to enoxaparin for reagent D. Conversely, reagents A and B propose a hybrid calibration curve obtained by mixing or combining UFH and LMWH in plasma. The anti-FXa/anti-FIIa ratios suggest that calibrators used for reagent B are obtained using a mixture of UFH and LMWH. In contrast, reagent A uses a combination of plasmas containing UFH (Calibrators 2 and 4) or LMWH (Calibrators 3 and 5), with an anti-FXa/anti-FIIa ratio similar to that of certoparin.

These choices impact calibration linearity, for UFH mainly, and hybrid calibration curves are intermediate between UFH and LMWH with the introduction of biases. Then, the UFH concentrations tend to be underestimated, as observed with reagents A and B, especially in the low range, and this underestimation is additive with that resulting from heparin inhibition in the absence of DS (reagent A). Different anti-FXa reagents were used along with each manufacturer's proposed instrument for reagents A, B, and C, by adhering strictly to the instructions for use. Reagent D is proposed for multiplatform applications, developed for all major instruments available. In this study, reagent D was combined with the Sysmex CS-5100 instrument.

The limitations of our study are that the same plasmas were not collected with both anticoagulants tested, i.e., citrate and CTAD, and the number of samples analyzed in each test. However, the results are significant enough to support our analysis and conclusions. Additional and prospective studies would be useful to better document the anti-FXa assays' differences.

Lastly, when using homogenous (same manufacturer) reagent-instrument systems, reagent weaknesses can be masked by the assay software adaptation, or by introducing algorithms for optimizing the assay apparent performances. This approach is used for adjusting the intrinsic low anti-FXa activity present in all plasma samples, and it is variable from plasma to plasma. In the absence of heparin, this intrinsic anti-FXa activity can account for 0-0.05 IU/mL in normal plasmas, and more rarely up to 0.10 IU/mL. This background activity is likely due to the anti-FXa activity of TFPI, S protein, Z protein (ZP), and ZP inhibitor (ZPI), or the β-AT form. The anti-FXa heparin assay is an inverse relationship between absorbance change measured at 405 nM, and heparin concentration. Therefore, normal plasmas exhibit anti-FXa activity background. Assay systems can manage this variability by masking that effect and "starting to measure" the change in absorbance only from a threshold value, corresponding to plasmas with the highest anti-FXa intrinsic activity. The apparent heparin concentrations in all plasmas are then 0 IU/mL in the absence of heparin, but low concentrations of heparin, in the range 0-0.10 IU/mL, or even up to 0.15 IU/mL, can be missed, which contributes to the underestimation in the low range. This approach is, of course, not possible when the reagent is a multiplatform application, and no adjustment assay software can be used. Then, heparin concentrations measured in plasma are obtained without any data treatment.

In this report, we provide evidence that supports the usefulness of dextran sulfate for anti-FXa assays used for measuring plasma concentrations of UFH or LMWH, as shown by the good correlations between all assays, designed with or without dextran sulfate, with CTAD anticoagulated plasmas (platelet activation and release of heparin neutralizing proteins is prevented), but not with citrate anticoagulant. In addition, assay variability can result from the heparin calibration type used, the exactness of the UFH and LMWH calibration curves' superimposition, and the assay software for treatment of assay raw data. The analyses of these factors help to better understanding the differences reported in many studies for the heparin concentrations measured with different anti-FXa reagents. 

Heparin and anticoagulation

Monitoring Unfractionated Heparin: A Review of Activated Partial Thromboplastin Time Versus Antifactor Xa

Heparin: An essential drug for modern medicine

History of Heparin. Organotypic Models Drug Dev

A century of heparin: Past, present and future

Efficacy of low doses of heparin in prevention of deep-vein thrombosis after major surgery. A double-blind, randomised trial

New Applications of Heparin and Other Glycosaminoglycans

The Anticoagulant and Nonanticoagulant Properties of Heparin

The Multiple Faces of Heparin: Opportunities in COVID-19 Infection and Beyond

Problems in laboratory monitoring of heparin dosage

Monitoring high-dose heparinization during cardiopulmonary by-pass-a comparison between prothrombinase-induced clotting time (PiCT) and two chromogenic anti-factor Xa activity assays

A systematic review on the accumulation of prophylactic dosages of low-molecular-weight heparins (LMWHs) in patients with renal insufficiency

Chronic kidney disease and anticoagulation: From vitamin K antagonists and heparins to direct oral anticoagulant agents

Neutralisation of the anti-coagulant effects of heparin by histones in blood plasma and purified systems

Neutrophil extracellular traps (NETs) as markers of disease severity in COVID-19

Monitoring of Unfractionated Heparin in Severe COVID-19: An Observational Study of Patients on CRRT and ECMO

Heparin resistance in COVID-19 patients in the intensive care unit

Heparin-induced thrombocytopenia

An update on evidence based diagnostic and confirmatory testing strategies for heparin induced thrombocytopenia using combined immunological and functional assays

Histidine-rich glycoprotein inhibits the antiproliferative effect of heparin on smooth muscle cells

Effects of heparin and related molecules upon neutrophil aggregation and elastase release in vitro

Non-Anticoagulant Low Molecular Weight Heparins for Pharmaceutical Applications

A supersulfated low-molecular-weight heparin (IK-SSH) increases plasma levels of free and total tissue factor pathway inhibitor after intravenous and subcutaneous administration in humans

Tissue factor pathway inhibitor and anti-FXa kinetic profiles of a new low-molecular-mass heparin, Bemiparin, at therapeutic subcutaneous doses

Molecular weight dependent tissue factor pathway inhibitor release by heparin and heparin oligosaccharides

British Committee for Standards in Haematology. Guidelines on the use and monitoring of heparin

Comparison of activated coagulation time and whole blood heparin measurements with laboratory plasma anti-Xa heparin concentration in patients having cardiac operations

The use of activated clotting times to monitor heparin therapy during and after interventional procedures

Anti-factor Xa (Anti-Xa) Assay

In Vitro Modeling of the Influence of FVIII Activity and Heparin Induced Prolongation of APTT

Activated Partial Thromboplastin Time Monitoring of Unfractionated Heparin Therapy: Issues and Recommendations

Effects of pre-analytical variables on the anti-activated factor X chromogenic assay when monitoring unfractionated heparin and low molecular weight heparin anticoagulation

Monitoring unfractionated heparin with the aPTT: Time for a fresh look

Establishing the heparin therapeutic range using aPTT and anti-Xa measurements for monitoring unfractionated heparin therapy

Performance of Anti-Factor Xa Versus Activated Partial Thromboplastin Time for Heparin Monitoring Using Multiple Nomograms

Should we abandon the APTT for monitoring unfractionated heparin?

Accuracy, reproducibility and costs of different laboratory assays for the monitoring of unfractionated heparin in clinical practice: A prospective evaluation study and survey among Swiss institutions

The Mode of Action of Heparin in Plasma

Revisiting antithrombin in health and disease, congenital deficiencies and genetic variants, and laboratory studies on α and β forms

Heparin chain-length dependence of factor Xa inhibition by antithrombin in plasma

Molecular mechanisms of antithrombin-heparin regulation of blood clotting proteinases. A paradigm for understanding proteinase regulation by serpin family protein proteinase inhibitors

Heparin and low-molecular-weight heparin

Interactions of Heparins in the Vascular Environment

Heparin-protein interactions: From affinity and kinetics to biological roles. Application to an interaction network regulating angiogenesis

Binding of platelet factor 4 to heparin oligosaccharides

Heparin binding properties of human histidine-rich glycoprotein. Mechanism and role in the neutralization of heparin in plasma

Neutralization and binding of heparin by S protein/vitronectin in the inhibition of factor Xa by antithrombin III. Involvement of an inducible heparin-binding domain of S protein/vitronectin

Assay of heparin in plasma using a chromogenic substrate for activated factor X

Assay of plasma heparin using thrombin and the chromogenic substrate H-D-Phe-Pip-Arg-pNA (S-2238)

Comparative studies on biological activity of generic and branded enoxaparin in vivo and vitro

Automated amidolytic method for determining heparin, a heparinoid, and a low-Mr heparin fragment, based on their anti-Xa activity

Collaborative study for the calibration of replacement batches for the heparin low-molecular-mass for assay biological reference preparation

Standardisation of Unfractionated and Low-Molecular-Weight Heparin. Organotypic Models Drug Dev

Studies on hemostasis in COVID-19 deserve careful reporting of the laboratory methods, their significance, and their limitations

Monitoring unfractionated heparin therapy: Lack of standardization of anti-Xa activity reagents

Quality performance for indirect Xa inhibitor monitoring in patients using in-ternational external quality data

APTT therapeutic range for monitoring unfractionated heparin therapy. Significant impact of the anti-Xa reagent used for correlation

Dextran sulfate included in factor Xa assay reagent overestimates heparin activity in patients after heparin reversal by protamine

Response to "Studies on hemostasis in COVID-19 deserve careful reporting of the laboratory methods, their significance and their limitation": Don't throw the baby out with the bathwater

Modification of an Amidolytic Heparin Assay to Express Protein-Bound Heparin and to Correct for the Effect of Antithrombin III Concentration

Development of a Rapid and Sensitive Chromogenic Heparin Assay for Clinical Use

Monitoring Unfractionated Heparin (UFH) therapy: Which Anti Factor Xa assay is appropriate?

Using a Single Calibration Curve with the Anti-Xa Chromogenic Assay for Monitoring Heparin Anticoagulation

Heparin inactivation during blood storage: Its prevention by blood collection in citric acid, theophylline, adenosine, dipyridamole-C.T.A.D. mixture

COVID-19: Review on latest available drugs and therapies against SARS-CoV-2. Coagulation and inflammation cross-talking

Is Heparin Reversal with Protamine after Carotid Endarterectomy Dangerous?

Protamine reversal of low molecular weight heparin: Clinically effective?

Pharmacokinetic model of unfractionated heparin during and after cardiopulmonary bypass in cardiac surgery

Heparanase neutralizes the anticoagulation properties of heparin and low-molecular-weight heparin

Monitoring Incomplete Heparin Reversal and Heparin Rebound After Cardiac Surgery

Monitoring treatments with unfractionated heparin: CTAD must be used instead of citrate as the anticoagulant solution when using partial-draw collection tubes. Results of a multicenter evaluation