Variability of Argatroban Effects on the Multiple APTT Reagents in High and Low Coagulation Activity Samples

ABSTRACT

Introduction

Argatroban is routinely monitored using activated partial thromboplastin time (APTT), with a recommended target range of 1.5–3.0 times. Although this range was established based on clinical trial data, including several APTT reagents, the differences in reactivity among APTT reagents remain unclear. This study compared the reactivity of six commercial APTT reagents to argatroban in normal and abnormal plasma samples.

Materials and Methods

Normal samples were spiked with argatroban and five abnormal samples: low coagulation factor activity, high factor VIII, high factor VIIa, low fibrinogen, and high fibrinogen. Drug concentrations were adjusted to 0, 0.5, 1.0, 1.5, and 2.0 μg/mL in each plasma. Six APTT reagents, including silica or ellagic acid activator and phospholipids derived from synthetic or natural sources, were tested.

Results

The APTT ratio range among the six reagents was 2.4–3.2 in normal plasma at a concentration of 2.0 μg/mL. The sample showing the most expanded range was low coagulation activity (3.3–5.2), and the range in the sample with high factor VIII activity decreased (1.5–2.2).

Conclusions

A reactivity difference was observed in argatroban‐spiked samples, which increased in abnormal plasma samples. APTT may not reflect the anticoagulant activity of argatroban in patients with abnormal coagulation activity. The reactivity of APTT should be confirmed in each laboratory, and patient background coagulation status should be assessed before monitoring is conducted using the APTT ratio.

1. Introduction

Argatroban, a direct thrombin inhibitor, has been widely used for blood extracorporeal circulation, acute ischemic stroke, and chronic limb‐threatening ischemia. It has been adapted for percutaneous coronary intervention (PCI) and the prevention of thrombosis in heparin‐induced thrombocytopenia (HIT). Activated partial thromboplastin time (APTT) is the preferred method for monitoring the effectiveness of argatroban in the prophylaxis or treatment of thrombosis in patients with HIT. The recommended target APTT ratio is 1.5–3.0 times the baseline value and a time in seconds not exceeding 100 s based on clinical trial data [1, 2, 3, 4]; clinical trials were conducted at multiple sites, and several types of APTT reagents were employed for monitoring. However, there are some limitations in using APTT for argatroban monitoring. First, the dose–response curve in APTT is non‐linear. Although it was reported that the linearity was recognized at low concentration ranges, the reactivity decreased and flattened from around 1 μg/mL [5, 6, 7]. It implies that it is difficult to determine the exact drug concentration, especially at high levels. Second, APTT is affected by low levels of coagulation factors and the presence of lupus anticoagulants, which lead to an underestimation of drug concentration [8]. Especially in patients at the intensive therapy unit, elevated levels of coagulation factors, such as factor VIII, also lead to lower values than expected, which can lead to an overestimation of drug concentration [9, 10]. Third, APTT reagents are also composed of activators, such as silica or ellagic acid, along with various phospholipids; owing to this variability in raw materials, the sensitivity of APTT can be variable [11, 12].

Several studies have investigated whether variability in APTT reagents affects argatroban monitoring. Francis and Hursting compared 21 commercial APTT reagents by spiking normal plasma samples with argatroban, and they concluded that most reagents were similarly sensitive and that reagent choice was unlikely to significantly affect argatroban monitoring [6]. In contrast, Guy et al. reported that reagent variability leads to overestimation or underestimation in clinical plasma samples and suggested that drug concentration measurements should be considered if there is concern that the APTT might not reflect the concentration of argatroban [13]. Furthermore, a consensus meeting on the use of argatroban suggested that each laboratory generate its own dose–response calibration curve by spiking normal plasma ex vivo because the variation of tests depends on both the test kit and the laboratory device itself [14]. The current guidelines also suggested that other more specific assays such as dilute thrombin time or ecarin clotting time were the preferred monitoring methods, followed by APTT monitoring [2, 15, 16]. However, laboratories conducting these specific assays are limited, and APTT monitoring is still performed and described in these guidelines. Due to these issues, argatroban monitoring methods remain debatable. Patients taking argatroban may have coagulopathy rather than a normal coagulation status; in such cases, the effectiveness of the drug and APTT prolongation would also differ, and the reagent variability would increase. Keyl et al. demonstrated that argatroban half‐life is markedly increased in critically ill patients and that patient age and serum albumin concentration significantly contribute to variability in its anticoagulant activity [17].

In this study, we compared the reactivity to argatroban of six commercial APTT reagents with variable characteristics using both commercial normal plasma as well as those with low or high coagulation factor activities. The variability and validity of the 1.5–3.0 times baseline value as the therapeutic range were investigated.

2. Materials and Methods

2.1. Plasma Samples

To prepare argatroban‐spiked samples, Coagtrol N (Sysmex Corporation, Kobe, Japan) was used as normal plasma. In addition to the normal spiked samples, five types of abnormal spiked samples were prepared: low coagulation factor activity, high factor VIII activity, high factor VIIa, low fibrinogen, and high fibrinogen. For the sample with low coagulation factor activity, Coagtrol N was diluted 1.67‐fold with Imidazole buffer (Sysmex Corporation) to adjust the coagulation activity to approximately 0.6 U/mL. Samples with high factor VIII activity were prepared by adding Advate (Takeda Pharmaceutical Co. Ltd., Tokyo, Japan) to normal plasma, and factor VIII levels were adjusted to 1.5, 2.5, 3.2, and 4.7 IU/mL. Factor VIII levels were measured using the Revohem FVIII Chromogenic kit (Sysmex Corporation) with SHP calibrator (Sysmex Corporation). The calibration curve was created in duplicate, and the measurement was conducted once by the 4‐fold dilution assay. The data was confirmed within the calibration range and used to determine the FVIII activity in the Advate‐spiked samples. Blood coagulation factor VIIa (human, 2nd International Standard 07/228) was purchased from The National Institute for Biological Standards and Control (NIBSC) in Hertfordshire, United Kingdom, and was spiked into normal plasma to adjust its activity to 2.0 and 4.5 IU/mL, confirmed using Hemoclot Factor VIIa (HYPHEN BioMed, Neuville‐sur‐Oise, France) with an in‐house calibrator created from the 2nd International Standard 07/228. The measurements were conducted once in the dilution assay because the target values were above the calibration curve range. For the preparation of low‐fibrinogen concentration samples, fibrinogen deficient human plasma (Affinity Biologicals Inc., Ancaster, Canada) was used, in which the activities of coagulation factors II, V, and VIII were 0.86, 0.28, and 0.64 U/mL, respectively, according to the certificate of analysis. The deficient plasma was mixed with normal plasma, and low fibrinogen samples were prepared to adjust the concentration to 0.5 and 1.0 g/L. In high fibrinogen samples, normal plasma was spiked with fibrinogen (Merck Group Japan, Tokyo, Japan) and adjusted to concentrations of 5.0 and 7.0 g/L. Fibrinogen concentration was measured using Thrombocheck Fib(L) (Sysmex Corporation) by the Clauss method. The measurements of factor VIII, factor VIIa, and Fbg were conducted in each prepared blank sample (0 μg/mL) without the interference of argatroban [18, 19]. Argatroban (Mitsubishi Tanabe Pharma America Inc., Osaka, Japan) was spiked into these plasma samples at concentrations of 0, 0.5, 1.0, 1.5, and 2.0 μg/mL. The concentration range was decided based on the previous study in clinical samples [17].

2.2. APTT Reagents and Laboratory Measurements

Six APTT reagents, including Thrombocheck APTT‐SLA (TC SLA) (Sysmex Corporation), Revohem APTT SLA (Rev SLA) (Sysmex Corporation), Actin FSL (FSL) (Sysmex Corporation), Data‐fi APTT (Data‐fi) (Sysmex Corporation), Coagpia APTT‐N (Coagpia) (Sekisui Medical, Tokyo, Japan), and HemosIL Synthasil APTT (Synthasil) (Werfen, Barcelona, Spain), were tested. The characteristics of these reagents were summarized in Table 1. For APTT measurements, plasma (50 μL) was incubated for 1 min at 37°C in a cuvette, to which 50 μL of APTT reagent was added, and then incubated for 3 min at 37°C. CaCl₂ solution (50 μL) was then added, and clotting time was measured. All measurements were performed using a CN‐3000 (Sysmex Corporation). For TC SLA, Rev SLA, FSL, and Data‐fi, the assay parameters provided by the manufacturer were used. For Coagpia and Synthasil, the assay parameters were established for this study based on the information in the instructions for use. The clotting times of all samples were determined twice, and mean values were used for the data analysis.

TABLE 1.

Characteristics of APTT reagents.

Reagents	Abbreviation	CaCl2 concentration	Activator	Phospholipids
Thrombocheck APTT‐SLA	TC SLA	20 mM	Ellagic acid	Synthetic
Revohem APTT SLA	Rev SLA	25 mM	Ellagic acid	Synthetic
Actin FSL	FSL	25 mM	Ellagic acid	Soy and rabbit phospholipids
Data‐fi APTT	Data‐fi	20 mM	Ellagic acid	Rabbits
Coagpia APTT‐N	Coagpia	25 mM	Ellagic acid	Rabbits
HemosIL Synthasil APTT	Synthasil	20 mM	Silica	Synthetic

2.3. Data Analysis

APTT ratios were calculated by dividing each clotting time by the clotting time of Coagtrol N for each reagent. The calculation formula was defined as patient APTT/mean normal APTT in British guidelines, and the clotting time of Coagtrol N and the spiked plasma samples were used as the substitutions for mean normal APTT and patient APTT, respectively, in this study [2]. To investigate the validity of the 1.5–3.0 APTT ratio as a therapeutic range, the following analysis was conducted (Figure 1). First, the dose–response curve was described for each pair of reagents and samples as a graph of argatroban concentration and APTT ratio. Second, the highest and lowest reactive dose–response curves were selected for each reagent, and the polynomial approximation formulas (y = ax² + bx + c) were calculated, respectively. Third, the intersections between the polynomial approximation formula and APTT ratios of 1.5 and 3.0, respectively, were then calculated, and the difference between them was defined as the calculated range of drug concentration in each dose–response curve. Finally, the difference between minimum and maximum values of the calculated range in the highest and lowest reactive dose–response curves was calculated, and this whole difference was defined as the estimated range corresponding to the APTT ratio 1.5–3.0 for each reagent.

FIGURE 1.

Method to for the calculation of estimated range corresponding to APTT ratio 1.5–3.0. The highest and lowest reactive dose–response curves were selected in each reagent (red and blue curves, respectively). The polynomial approximation formula was established in each curve. The intersections between the polynomial approximation formula and APTT ratio of 1.5 and 3.0 were calculated in each curve, and the calculated range from the maximum or minimum dose–response curve was calculated. The estimated range corresponding to APTT ratio 1.5–3.0 was defined as the whole range from the intersection between 1.5 and the highest reactivity dose–response curve to the intersection between 3.0 and the lowest reactivity dose–response curve.

3. Results

3.1. Reactivity to Argatroban in Normal Plasma Sample

The clotting times of Coagtrol N were 27.5, 27.8, 26.6, 26.8, 27.8, and 31.0 s in TC SLA, Rev. SLA, Data‐fi, FSL, Coagpia, and Synthasil, respectively; these clotting times were used as the mean normal APTT. The reagent differences in the APTT ratio among the six reagents increased as argatroban concentration increased, and clear differences among reagents were observed at high drug concentrations (Figure 2). Synthasil reagent, having silica as the activator and synthetic phospholipids, exhibited the highest reactivity. In contrast, the FSL and Coagpia APTT reagents, including ellagic acid and phospholipids derived from natural sources, showed lower reactivity than the other reagents. Synthasil reagent gave 109.6 s in the sample with argatroban 2.0 μg/mL, while other reagents did not show more than 100 s in all samples.

FIGURE 2.

Dose–response curve to argatroban in normal plasma by six APTT reagents. The dose–response curves in normal plasma sample spiked argatroban at concentrations of 0, 0.5, 1.0, 1.5, and 2.0 μg/mL were described in six reagents. The reagent characteristics were shown in Table 1.

3.2. Reactivity to Argatroban in Abnormal Plasma Samples

The ranges of APTT ratios among the six reagents in the abnormal samples were compared with those in normal plasma (Table 2). The dose–response curves of APTT reagents are shown in the Supporting Information (Figure S1–S5). In the low coagulation factor activity sample, the APTT ratios at 0 μg/mL were more than 1.1, and the ranges elevated as the drug concentration increased. The APTT ratio range was 3.3–5.2 at 2.0 μg/mL. The difference in the range was 1.9, in which it was more than twice in normal plasma. FSL showed the lowest reactivity in both normal and diluted plasma. In contrast, Data‐fi had the highest reactivity in low coagulation factor plasma, higher than that of Synthasil, which had the highest reactivity in normal plasma. The clotting times exceeded 100 s in the samples with 1.0, 1.5, and 2.0 μg/mL argatroban in TC SLA, Rev. SLA, Data‐fi, and Synthasil, while FSL and Coagpia did not give clotting times of more than 100 s.

TABLE 2.

The ranges of APTT ratios among six APTT reagents in argatroban‐spiked samples.

Sample types		Argatroban concentration (μg/mL)
		0	0.5	1.0	1.5	2.0
Normal		1.0	1.6–1.9	1.9–2.4	2.2–2.8	2.4–3.2
Low coagulation factor activity		1.1–1.6	2.1–3.4	2.6–4.2	3.0–4.8	3.3–5.2
High factor VIII activity	1.5 IU/mL	0.9	1.4–1.7	1.6–2.1	1.8–2.4	1.9–2.7
	2.5 IU/mL	0.8–0.9	1.2–1.6	1.4–1.9	1.6–2.2	1.7–2.4
	3.2 IU/mL	0.8	1.1–1.5	1.4–1.8	1.5–2.1	1.7–2.3
	4.7 IU/mL	0.7–0.8	1.1–1.4	1.3–1.7	1.4–2.0	1.5–2.2
High factor VIIa activity	2.0 IU/mL	1.0	1.7–2.0	2.0–2.5	2.2–2.9	2.4–3.4
	4.5 IU/mL	1.0	1.7–1.9	2.0–2.4	2.2–2.8	2.4–3.3
Low fibrinogen	0.5 g/L	0.8–1.2	1.6–2.3	2.0–2.8	2.2–3.1	2.3–3.3
	1.0 g/L	0.8–1.1	1.4–2.1	1.8–2.5	2.0–2.9	2.2–3.1
High fibrinogen	5.0 g/L	1.0–1.1	1.6–2.0	1.9–2.4	2.1–2.9	2.3–3.3
	7.0 g/L	1.0–1.2	1.6–2.1	1.8–2.6	2.0–3.0	2.2–3.4

Note: The ranges of APTT ratios among six APTT reagents were summarized in each argatroban concentration and sample. The values of cells indicate the ranges in the six reagents at the sample.

The APTT ratios in samples with high factor VIII activity were lower than those in normal plasma at all drug concentrations. As factor VIII activity increased from 1.5 to 4.7 IU/mL, the APTT ratio ranges narrowed and decreased from 1.9–2.7 to 1.5–2.2. The lowest reactivity was observed for FSL for all factor VIII activities. Data‐fi and Synthasil showed similar APTT ratios; these two reagents and Rev. SLA showed higher reactivity than the other three reagents. All clotting time data was within 100 s in all reagents.

In samples with high factor VIIa activity, the APTT ratios and ranges at both 2.0 and 4.5 IU/mL were not significantly different from those of normal plasma. Although coagulation activity is accelerated by factor VIIa, no significant changes in APTT ratios were observed at any drug concentration. Only Synthasil reagent gave more than 100 s clotting time for both factor VIIa activity samples with 2.0 μg/mL argatroban.

In low fibrinogen samples, the APTT ratio ranges at the concentrations from 0.5 to 2.0 μg/mL were 1.6–2.3 to 2.3–3.3 for 0.5 g/L and 1.4–2.1 to 2.2–3.1 for 1.0 g/L, respectively. The ranges at the concentrations of 0.5, 1.0, and 1.5 μg/mL were higher than those of normal plasma, while significant differences were not observed between fibrinogen 0.5 and 1.0 g/L and normal plasma at 2.0 μg/mL. Data‐fi showed much higher reactivity than the other reagents, and Synthasil showed lower reactivity than Rev. SLA and Data‐fi; these tendencies were similar to those of the low coagulation factor activity sample. No samples showed more than 100 s clotting times.

The APTT ratios did not decrease in high fibrinogen 5.0 and 7.0 g/L samples; the APTT ratio ranges at concentrations from 0.5 to 2.0 μg/mL were 1.6–2.0 to 2.3–3.3 for 5.0 g/L and 1.6–2.1 to 2.2–3.4 for 7.0 g/L, respectively. The range at 7.0 g/L was slightly higher than that at 5.0 g/L, indicating that high fibrinogen concentrations increased APTT ratio variability. Synthasil and Coagpia showed the highest and lowest reactivity, respectively. Both 5.0 and 7.0 g/L samples exceeded 100 s clotting time at 2.0 μg/mL concentration in Synthasil reagent.

3.3. Estimated Ranges Corresponding to APTT Ratio 1.5–3.0

The dose–response curves showing the highest and lowest reactivity were low coagulation factor activity and high factor VIII activity of 4.7 IU/mL samples, respectively. The ranges between the minimum value of the highest reactivity and the maximum value of the lowest reactivity samples were established based on each dose–response curve as the estimated drug concentration ranges corresponding to the APTT ratio 1.5–3.0 (Table 3). The APTT ratio increased without argatroban in the low coagulation factor activity sample, and the drug concentration corresponding to APTT ratio 1.5 was 0–0.15 μg/mL among the six reagents, indicating that APTT ratios underestimated drug concentrations. In contrast, the drug concentration was overestimated at more than 2.0 μg/mL corresponding to APTT ratio 3.0 in the high factor VIII activity (4.7 IU/mL) sample for all reagents. The drug concentration was largely dependent on the reagent and sample background, which induced overestimation or underestimation.

TABLE 3.

The estimated drug concentration ranges corresponding to APTT ratio 1.5–3.0.

	Thrombocheck APTT‐SLA	Revohem APTT SLA	Data‐fi APTT	Actin FSL	Coagpia APTT	Synthasil APTT
The ranges calculated in normal sample	0.29–≥ 2.0	0.28–1.70	0.25–1.70	0.37–≥ 2.0	0.34–≥ 2.0	0.25–1.34
The ranges calculated in low coagulation activity sample	0.04–0.61	0.08–0.67	0–0.39	0.14–1.2	0.13–1.2	0.15–0.79
The ranges calculated in high factor VIII sample	0.92–≥ 2.0	0.77–≥ 2.0	0.71–≥ 2.0	1.62–≥ 2.0	1.20–≥ 2.0	0.81–≥ 2.0
The estimated ranges	0.04–≥ 2.0	0.08–≥ 2.0	0–≥ 2.0	0.14–≥ 2.0	0.13–≥ 2.0	0.15–≥ 2.0

Note: The highest and lowest reactivity samples were low coagulation activity and high factor VIII samples in all reagents. These polynomial approximation formulas (y = ax²+bx + c) were calculated from the dose–response curve with that of the normal sample in each pair of sample and reagent. The intersections corresponding to the APTT ratio 1.5–3.0 were shown in each sample and reagent pair. 0 and ≥ 2.0 indicate that there are no intersections between the polynomial approximation formulas and the 1.5 or 3.0 lines, respectively. The estimated ranges were established as the minimum and maximum values calculated from these polynomial approximation formulas.

4. Discussion

Argatroban is mainly used for PCI and prevention of thrombosis in HIT, and monitoring is routinely conducted using APTT. There are no recommended APTT reagents for monitoring because previous studies have reported no clinically significant differences between reagents [6, 20, 21]. We found that the drug concentration ranges corresponding to APTT ratio 1.5–3.0 were different among six APTT reagents, with differences increasing in abnormal plasma samples. Although in vitro spiked plasma samples were used, the influence of coagulation factors, including factor VIII, factor VIIa, and fibrinogen, was confirmed. The low coagulation factor activity sample having around 0.6 U/mL particularly affected the APTT ratio and increased differences between the six reagents, suggesting that knowing coagulation factor activity is important to prevent overestimation and underestimation of argatroban concentration; additionally, the risk of bleeding or thrombosis might exist in abnormal coagulability status.

Synthasil had the highest reactivity in most samples, whereas Data‐fi had the highest reactivity in low coagulation factor activity and low fibrinogen samples. In contrast, the reactivities of FSL and Coagpia were lower than those of the other reagents in all samples. It has been widely reported that there are several differences between APTT reagents in their reactivity to argatroban as well as factor deficiency, lupus anticoagulants, and heparin [11, 12, 22, 23]. In particular, the sensitivity to unfractionated heparin, which mainly inhibits thrombin through anti‐thrombin, has also been investigated; Gouin‐Thibaut et al. showed that APTT ratios corresponding to 0.3–0.7 anti‐Xa IU/mL as the therapeutic ranges were 1.9–4.0, 2.2–5.8, and 2.3–6.6 with Synthasil, FSL, and Data‐fi, respectively [24]. The difference in ranges between Synthasil and Data‐fi was large, indicating that reagent reactivity with unfractionated heparin was different from that of argatroban. The reactivity of APTT reagents to drugs differed among drugs; reagent variability also differed.

The difference in reactivity between reagents resulted from the different source materials employed in the reagents. Commercial APTT reagents contained activators and various phospholipids. Synthasil and Data‐fi showed the highest reactivity; they contain a silica activator with synthetic phospholipids and an ellagic acid activator with phospholipids derived from rabbits, respectively. FSL and Coagpia showed low reactivity, containing ellagic acid with soy and rabbit phospholipids and ellagic acid with phospholipids derived from rabbits, respectively. The clear relationship between reactivity to argatroban and reagent composition was not observed. Although all measurements with multiple reagents were conducted by only one instrument in this study, the reactivity was dependent not only on the reagent but also on the analyzer. This indicates that the establishment of a local therapeutic range for the APTT ratio in each clinical laboratory may be useful. The consensus meeting on the use of argatroban also suggested that each laboratory generate its own dose–response calibration curve for accurate monitoring [14]. The sensitivity to argatroban was clearly different between FSL and Synthasil in our study, while Guy et al. showed similar results between FSL and Synthasil in their studies conducted with spiked plasma and clinical samples [13, 25]. Especially, the APTT mean ratios were 1.58 and 1.69 in Actin FSL and Synthasil, respectively; the correlation between these APTT ratios and argatroban levels was R ² = 0.29 and 0.16, respectively [25]. Although several causes of differences between our study and theirs were considered, one of the possibilities is the analyzer. They conducted the measurements on a different analyzer between FSL and Synthasil in the spiked plasma samples [13]. Therefore, the establishment of the local therapeutic range in each clinical laboratory with the pair of reagent and analyzer may be helpful. In addition, it was reported that the reagent variability was different between spiked plasma and clinical samples in reagents [13, 25]. The APTT ratio ranges in spiked plasma samples (0–2.7 μg/mL) were 0.9–2.6, 0.9–2.2, and 0.9–2.1 in Actin FS, Actin FSL, and Synthasil, respectively, while the ranges and correlation coefficients were 0.9–3.7 and 0.26 for Actin FS and 0.9–3.2 and 0.21 for Synthasil in clinical samples (0–1.37 μg/mL), respectively [13]. The reactivity of Actin FSL and Synthasil in that study was lower than those of our study. In addition, the APTT ratios in clinical samples were higher than those of spiked plasma samples despite the considerably lower argatroban concentration. Furthermore, Keyl et al. showed that the correlation between APTT ratio and the drug concentration was poor in clinical samples (R ² = 0.28) using Pathromtin‐SL‐reagent on a BCS‐XP analyzer, and the clear linearity was not recognized [17]. Although our study was conducted only in spiked plasma samples and the levels of reagent variability might be different from clinical samples, it was confirmed that each reagent has its own sensitivity to argatroban and the variability was spread in abnormal coagulation status.

Kits to measure exact argatroban concentrations have been developed, and the usefulness of diluted thrombin time and ecarin chromogenic/clotting assays, anti‐IIa methods to detect thrombin inhibition, has been reported in some studies [5, 8, 13]. These studies suggest that the anti‐IIa assay is more suitable for argatroban monitoring than APTT, especially in patients with low or high coagulation factor activity and lupus anticoagulants. Intensive care unit patients often have different coagulation factor activities in some coagulation factors, including prothrombin or high factor VIII activity, and show prolonged APTT clotting times [10], making exact monitoring of the APTT ratio difficult. In addition, the argatroban plasma half‐life is increased in critically ill cardiac surgical patients and is further prolonged by hepatic dysfunction due to impaired metabolism [17]. Based on these studies, there is a limitation in monitoring drug effectiveness using only APTT. The consensus statement on the use of argatroban in patients with HIT recommends that it would be advisable to employ anti‐IIa assays for the measurements of the drug concentration levels [14]. It would be advisable to employ the thrombin time method for patients in the intensive care unit because these patients are often deficient in prothrombin and abnormal results are obtained in APTT and ecarin assays. Thrombin time is independent of prothrombin level, in which creating the calibration curve with the specific thrombin time in the laboratory would be useful [14]. The recent guidelines also suggested that the anti‐IIa assay was the preferred monitoring method [2, 15, 16], and the proposed target argatroban concentration is 0.4–1.5 μg/mL and 0.5–1.5 μg/mL in Swiss and French guidelines, respectively [15, 16]. Although the anti‐IIa assay was not employed in our study, APTT ratios were largely dependent on coagulation factor activities, and APTT reagent reactivity differed. From this, our data also support the need for an accurate monitoring method, with the anti‐IIa assay being one solution.

This study had some limitations. First, the differences in the APTT reagent were investigated using only drug‐spiked samples; clinical samples with argatroban should be collected to confirm whether there are any differences between the spiked and clinical samples. Second, samples with low or high coagulation factor activity were also artificially prepared in vitro. The coagulability of patients was variable; this variability was not fully reflected in this study, and reagent reactivity might vary depending on patient background. Especially, a low coagulation factor activity sample was prepared by diluting normal plasma with buffer, in which the sample matrix changed. Although the reagent variability was confirmed in this artificial sample to investigate the effect of low coagulation activity, the sample matrix may affect the results. Third, the estimated ranges corresponding to an APTT ratio of 1.5–3.0 were established based on the polynomial approximation formulas calculated from the dose–response curve between 0 and 2.0 μg/mL of the drug. The formulas may differ when data of > 2.0 μg/mL are included in the calculation, and the estimated ranges may also be different. Fourth, argatroban concentration was calculated based on the assigned value in the pharmaceutical company. Although the assigned value is accurate, the matrix effect in plasma may not be reflected in spiked plasma samples. When the assigned values in spiked plasma samples are compared with the values measured by the anti‐IIa assay, the difference might be observed.

In conclusion, a reactivity difference among the six APTT reagents was observed in argatroban‐spiked samples, with higher variability observed in samples with low or high coagulation factor activity and fibrinogen concentration. APTT may not reflect the anticoagulant activity of argatroban in patients with low or high coagulation factor activity. The reactivity of the selected APTT reagent to argatroban should be confirmed in each clinical laboratory, and the background coagulation status in patients should be assessed before monitoring is conducted using the APTT ratio.

Author Contributions

Osamu Kumano performed the sample measurements, analysis, interpreted the data, and wrote the manuscript. Masahiro Ieko, Teruto Hashiguchi, Takashi Ito, Satoshi Yamazaki, and Sumiyoshi Naito contributed to study conception, interpreted the data, and revised the manuscript. Masako Yamazaki contributed to the study conception, interpreted the data, revised the manuscript, and organized the study.

Ethics Statement

The authors have nothing to report.

Consent

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1. Dose–response curve in low coagulation factor activity sample.

Figure S2. Dose–response curve in high factor VIII activity samples.

Figure S3. Dose–response curve in high factor VIIa activity samples.

Figure S4. Dose–response curve in low fibrinogen samples.

Figure S5. Dose–response curve in high fibrinogen samples.

Kumano O., Ieko M., Hashiguchi T., et al., “Variability of Argatroban Effects on the Multiple APTT Reagents in High and Low Coagulation Activity Samples,” International Journal of Laboratory Hematology 47, no. 5 (2025): 923–930, 10.1111/ijlh.14488.

Data Availability Statement

Due to the nature of this research, participants of this study did not agree for their data to be shared publicly, so supporting data is not available.