Thrombin Generation Decrease After LMWH Administration in an Antithrombin-Deficient Pregnant Woman With a Homozygous HBS II Mutation

Abstract

The cases of antithrombin (AT)-deficient pregnant women with a homozygous HBS II mutation are relatively rare and are accompanied by an increased thrombophilic risk, which is manifested by increased thrombin generation (TG). It is very difficult to ensure their prophylactic treatment during pregnancy. We aimed to determine the utility of the thrombin generation assay (TGA) and anti-factor Xa (anti-FXa) test to monitor the effects of a prophylactic dose of low-molecular-weight heparin (LMWH) in a 28-year-old woman with homozygous AT deficiency caused by mutation c.391C > T#, (p.Leu131Phe†) in the SERPINC1 gene and to compare the findings with those from a group of pregnant and non-pregnant women also treated with LMWH. TG monitoring was chosen due to severe AT deficiency that was manifested by low levels of anti-FXa activity when monitoring the efficacy of LMWH treatment. A significant decrease in TG was detected in all monitored groups (P < .05). There were no thrombotic complications during the whole pregnancy of the woman with AT deficiency. Consistent monitoring of TG with LMWH anticoagulant therapy administration during pregnancy together with AT administration before and after delivery may improve the overall condition of pregnant women and the quality of their care.

Introduction

Physiological pregnancy is accompanied by many changes in the hemostatic system due to thrombogenesis.^1,2 There is an increase in the levels of coagulation factors, especially fibrinogen, factor VIII, VII, V, and von Willebrand factor.^1,2 Protein S activity is reduced, and there is also a lower effect of activated protein C due to higher prothrombin activity. At the same time, there is an increase in the concentration of fibrinolysis inhibitors PAI-1 and PAI-2 (plasminogen activator inhibitors) and TAFI (thrombin-activated fibrinolysis inhibitor). These procoagulant changes may be manifested by increased thrombin generation (TG).^1,3

If there is a pregnancy with an increased thrombophilic risk, for example, due to a congenital thrombophilic condition, it may be accompanied by even higher values of factor activity or an inadequate response to prophylactic doses of anticoagulant treatment.^4,5

Congenital antithrombin (AT) coagulation inhibitor deficiency may be such a case. AT production disorder is an autosomal dominant disease that is less common than, for example, the Leiden mutation; however, it is much more serious. AT deficiency has several distinct types. ⁶ Type I is manifested by a lower amount of AT antigen in the plasma as well as reduced activity. Nevertheless, the AT molecule is functional. Type II is characterized by normal amount of AT, but its activity is greatly reduced. Type II is further divided into 3 subtypes, the most important of which is type IIB, in which damage to the heparin-binding site occurs, reducing the effectiveness of treatment with exogenous heparins. ⁶ It is important to monitor AT deficiency, especially during pregnancy, where even fatal consequences can occur.^4,7,8

The aim of this study was to draw attention to the high generation of thrombin in a patient with AT deficiency, in whom a homozygous mutation of AT in the heparin-binding site for heparin was proven and TG reduction after low-molecular-weight heparin (LMWH) administration was demonstrated. We wanted to compare her results with a control group of healthy non-pregnant women and pregnant women. Furthermore, we wanted to compare the reduction of TG after LMWH administration in control groups of healthy non-pregnant women and pregnant women and, in comparison with these groups, take into account the inconsistent effect of LMWH in a patient with AT deficiency which, however, was manifested by certain reduction in TG.

Material and Methods

Selection of Patients

A group P consisted of 30 pregnant Czech women (age 32 ± 6 years, the gestation week [GW] 37 ± 1, BMI 28.9 ± 2.2 kg/m²) monitored at a pregnancy center of the Obstetrics and Gynecology Clinic and at the Thrombotic Centre of General University Hospital and indicated for LMWH prophylaxis before the delivery (Enoxaparinum natricum [Clexane, Sanofi, Paris, France] 4000 IU/0.4 mL, SQ once a day, LOT ECI47A). A group N consisted of 30 non-pregnant Czech women (age 34 ± 7 years, BMI 24.1 ± 1.8 kg/m²) without a coagulation disorder or any thrombotic manifestation, with 1 or 2 previous uncomplicated births and without postpartum complications, who had LMWH administered voluntarily with the same dose as the pregnant women (Enoxaparinum natricum [Clexane, Sanofi, Paris, France] 4000 IU/0.4 mL, SQ once a day, LOT ECI47A). A patient (28-year-old Czech woman) with AT deficiency (D) was monitored from the beginning of the pregnancy and treated with LMWH in an adjusted dose (GW 9-10 Enoxaparinum natricum [Clexane, Sanofi, Paris, France] 6000 IU/0.6 mL, SQ once a day, LOT ECI11C; GW 11-12 Enoxaparinum natricum [Clexane, Sanofi, Paris, France] 6000 IU/0.6 mL, SQ twice a day, LOT ECI11C; GW 13-38 Nadroparinum calcicum [Fraxiparine, Mylan IRE Healthcare Limited, Dublin, Ireland] 7600 IU/0.8 mL, SQ twice a day, LOT 3264A) at the Thrombotic Centre of General University Hospital.

The first venous blood collection took place before the administration of LMWH at the time of standard pregnancy blood collection (time 0).

The second sampling was performed 4 h after LMWH SQ (time 4).

The collected blood samples were subsequently processed in the Central Hematological Laboratory of the Institute of Medical Biochemistry and Laboratory Diagnostics, General University Hospital. We determined basic coagulation parameters in plasma samples taken before LMWH administration (prothrombin time, activated partial thromboplastin time, fibrinogen, and D-dimer were determined in accordance with routine laboratory methods), AT activity with thrombin and activated factor X, AT antigen, and TG with thrombin generation assay (TGA). ⁹ We then measured anti-factor Xa (anti-FXa) activity and TG using the TG test in plasma samples collected after LMWH application.^10,11

The study was performed within the scope of the projects of the Ministry of Health, the Czech Republic (no.: RVO VFN 64 165), which was approved by the ethics committee of the First Faculty of Medicine of the Charles University and General University Hospital in Prague. Each participant gave written informed consent.

Laboratory Methods

Four and a half of milliliters of venous blood was collected in BD Vacutainer^® tubes (Becton Dickinson, Plymouth, UK) containing buffered 0.129 M sodium citrate. Platelet-poor plasma (PPP) was isolated from citrated blood by centrifugation at 2000 g for 15 min at room temperature, aliquoted, and a part of PPP was stored at −80 °C until TGA analysis.^12,13

We used Siemens diagnostic kits Berichrom^® Antithrombin III (IIa) and INNOVANCE^® Antithrombin (Xa) to determine AT activity using BCS XP analyzer (Siemens, Marburg, Germany). When determining the AT antigen, we used the immunochemical reaction (N Antiserum to Human Antithrombin III, Siemens, BN ProSpec analyzer, Siemens, Marburg, Germany). To monitor the effect of LMWH, we used the chromogenic method of anti-FXa test with BIOPHEN^® Heparin LRT (Hyphen Biomed, Neuville-sur-Oise, France). D-dimer concentration was assessed using MediRox D-Dimer^® (Alere AB, Lidingö, Sweden), BCS XP analyzer (Siemens, Marburg, Germany). Monitoring of TG is based on monitoring the thrombin formation with a fluorogenic substrate using the Ceveron^® Alpha analyzer (Technoclone GmbH, Vienna, Austria) upon activation of the coagulation cascade by a tissue factor. We have performed the TG comparison between the AT-deficient pregnant woman and the control groups of pregnant and non-pregnant women using the TGA. TG in PPP was initiated using a recombinant human tissue factor (rhTF) in this assay. The maximum concentration of thrombin (peak thrombin in nM) generated and the endogenous thrombin potential (ETP; calculated from the area under the concentration–time curve [AUC]) were used for the analysis. TGA was performed using the assay kit for thrombophilia tendency testing Technothrombin^® TGA RC Low (RCL) with a low concentration of phospholipid micelles containing approximately 5 pmol rhTF in Tris-Hepes-NaCl buffer) and the anticoagulant activity assay kit Technothrombin^® TGA RC High (RCH) with a high concentration of phospholipid micelles containing approximately 5 pmol rhTF in Tris-Hepes-NaCl buffer) on an automated Ceveron^® alpha analyzer with a TGA fluorometric module in accordance with the manufacturer's instructions. The TGA kits, additional normal control plasma Technothrombin^® TGA C1, and the analyzer were purchased from Technoclone GmbH (Vienna, Austria). Peak thrombin and AUC levels (mean ± SD) obtained from the TGAs in the control group of healthy subjects were used to define normal TG in PPP.

Normal values of our laboratory are summarized in Table 1.

Table 1.

Results of Control Groups and an Antithrombin-Deficient Pregnant Woman Before and After LMWH Administration.

Parameters (units)	Normal Laboratory Values	Non-Pregnant (N0)	Non-Pregnant (N4)	Pregnant (P0)	Pregnant (P4)	AT-Deficient Pregnant Woman (D0)	AT-Deficient Pregnant Woman (D4)
	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	36 GW	36 GW
AT Activity Xa (%)	99.6 ± 10.80	97.0 ± 7.5	NA	99.0 ± 8.7	NA	16.0*	NA
AT Activity IIa (%)	100.5 ± 11.90	97.0 ± 7.6	NA	98 .0 ± 7.9	NA	63.7*	NA
AT Antigen (g/L)	0.24 ± 0.02	0.24 ± 0.03	NA	0.24 ± 0.03	NA	0.17*	NA
anti-FXa (IU/mL)	0.020 ± 0.020	0.026 ± 0.025	0.394 ± 0.077*	0.032 ± 0.025	0.253 ± 0.093*	0.0	0.05
D-dimer (μg/L)	97.97 ± 164.60	76.7 ± 57.6	NA	507.6 ± 54.7*	NA	791.3*	NA
TGA RCL Peak (nM/L)	265.83 ± 109.38	217.4 ± 88.6	119.3 ± 50.1*	527.2 ± 217.4*	361.9 ± 126.6*	799.2*	792.1*
TGA RCL AUC (nM/L.min)	2095.78 ± 330.79	2060.0 ± 317.6	1539.2 ± 347.5*	3104.2 ± 149.9*	2821.9 ± 241.1*	5116.2*	5005.8*
TGA RCH Peak (nM/L)	427.86 ± 138.71	312.0 ± 133.4	180.0 ± 88.9*	690.9 ± 147.7*	564.8 ± 484.9*	954.6*	954.1*
TGA RCH AUC (nM/L.min)	2168.78 ± 318.47	2091.1 ± 370.3	1552.8 ± 414.9*	3113.7 ± 163.8*	2856.3 ± 266.5*	5111.2*	5105.8*

*Statistically significant difference versus laboratory values.

Abbreviations: AT, antithrombin; FXa, activated factor X; TGA, thrombin generation assay; RCL, low concentration of phospholipid; RCH, high concentration of phospholipid; AUC, area under the curve; N0/N4, non-pregnant before/4 h after LMWH administration; P0/P4, pregnant before/4 h after LMWH administration; D0/D4, AT-deficient pregnant woman before/4 h after LMWH administration; GW, gestational week.

Genomic DNA for AT mutation confirmation was extracted from the whole blood treated with citrate.^14,15 The coding regions of the SERPINC1 gene including intron–exon boundaries were amplified by PCR using corresponding intronic primers. Amplification was carried out by AmpliTaq Gold DNA polymerase with Buffer I (ThermoFisher Scientific, Waltham, Massachusetts, USA). The PCR products were purified with ExoSAP-IT (ThermoFisher Scientific) and sequenced with a Big Dye Terminator v3.1 Cycle Sequencing Kit (ThermoFisher Scientific) according to the manufacturer's protocol. Sequences of SERPINC1 were analyzed on the ABI3500 Genetic Analyzer (ThermoFisher Scientific, Waltham, Massachusetts, USA) using Vector NTI software (ThermoFisher Scientific).

The mutation p.Leu131Phe is the most prevalent mutation in the Czech Republic. It was present in one-fourth (26%) of all AT-deficient Czech patients. ¹⁵

Statistical Methods

The statistical software STATISTICA CZ (StatSoft, Inc., Version 12, Tulsa, USA) was used for statistical analysis. Two-sample t-test was used to compare differences between pregnant and control group data with normal distribution, whereas the Mann–Whitney U test was used for data with non-normal distribution. Matched pair analysis was used to compare the group before and after LMWH administration. Paired t-test was used to compare differences between dependent variables with normal distribution, and the Wilcoxon test was used for data with non-normal distribution. The level of statistical significance was determined at P˂.05 for each test. Mean and SD were used for descriptive characteristics.

Results

The normal values of our laboratory and the results of the 3 monitored groups are presented in Table 1. These values showed that there is no difference between AT levels in the 2 groups of pregnant and non-pregnant (P = .75; .87), but the activity and antigen are significantly decreased in a pregnant woman with proven AT deficiency (P < .05).

AT activity determined with a kit containing factor Xa in a patient with AT deficiency has significantly lower values (by 80%) than the 1 determined with a kit containing thrombin (by 36%) (Table 1).

Before the LMWH administration, the anti-FXa level in all 3 groups ranged from 0.00 to 0.03. The level of anti-FXa activity after LMWH administration differs significantly in the group of non-pregnant women, where the activities are close to the upper limit of the prophylactic range, and in the group of pregnant women, where the activities are more at the lower limit. A pregnant woman with AT deficiency did not respond to treatment according to the level of anti-FXa.

The level of D-dimers in non-pregnant women does not differ from normal values. In the group of pregnant women and in a pregnant woman with proven AT deficiency, the values are higher compared to the normal range, but they correspond with the D-dimer values in pregnancy (Figure 1).

Figure 1.

Anti-FXa activity, N0/N4—non-pregnant before/4 h after LMWH administration, P0/P4—pregnant before/4 h after LMWH administration.

Abbreviations: anti-FXa, anti-factor Xa; LMWH, low-molecular-weight heparin.

The TG values of peak thrombin and AUC RCL and RCH in the group of non-pregnant women do not differ from normal values. TG is increased in the pregnant group compared to normal values in the parameters peak thrombin and AUC RCL and RCH, but corresponds to TG values in pregnancy. TG values in a pregnant woman with AT deficiency are doubled compared to normal values in the parameters peak thrombin and AUC RCL and RCH and significantly increased also compared to the group of pregnant women.

The AUC RCL parameter, which is shown in Figure 2, shows the different values best.

Figure 2.

Thrombin generation via method using TGA RClow AUC, N0/N4—non-pregnant before/4 h after LMWH administration, P0/P4—pregnant before/4 h after LMWH administration, D0/D4-AT-deficient pregnant woman before/4 h after LMWH administration.

Abbreviations: TGA, thrombin generation assay; AUC, area under the curve; LMWH, low-molecular-weight heparin.

TG values (RCL Peak) are lower 4 h after LMWH administration in a group of non-pregnant women, pregnant women and a pregnant woman with AT deficiency (P < .05). For the RCL AUC parameter, they are also lower in all cases (P < .05) (Figure 2), but the reduction of both parameters occurs proportionally with respect to the baseline value before LMWH administration.

There is a significant reduction in TG in the non-pregnant and pregnant groups (P < .05), as for RCH peak and RCH AUC values, so the effect of LMWH is evident.

There is no significant reduction in TG (P = .19) in a pregnant woman with AT deficiency, so the result corresponds to an ineffective anti-FXa level (Table 1).

Discussion

The main conclusion of this study is that LMWH prophylaxis in pregnant women and in pregnant women with severe AT deficiency reduces TG.

We also demonstrated different generation of thrombin in the monitored groups of non-pregnant and pregnant women and a pregnant woman with AT deficiency.

Non-pregnant and pregnant women had AT levels in the normal range, ¹⁶ so the response to LMWH treatment was also adequate. The level of anti-FXa was almost zero before the LMWH administration, while 4 h after the administration of heparin, the upper limit of the target prophylactic range (0.2-0.4 IU/mL) was approached by the non-pregnant group and the lower limit by the pregnant group.^17,18

The significant reduction of both AT activity and antigen was demonstrated in a pregnant woman with proven AT deficiency, where the values of the AT determination method with the kit containing factor Xa were significantly lower than those of the determination with the thrombin-containing kit.^19,20

We supposed that the response to the administered LMWH would not be adequate because of the low AT level. When determining the activity of anti-FXa, values close to zero were indeed obtained. ¹⁸

TGA was chosen for its main role in the coagulation process as a marker for monitoring thrombophilic risk.^{21,22,23,24,25} We assumed that TG reduction would also be manifested in response to the effectiveness of the treatment, which was manifested in our observed groups, but unfortunately ambiguously.

It was shown in this study that pregnant women have increased TG for the parameters RCL peak thrombin, RCL AUC, RCH peak thrombin, and RCH AUC compared to the non-pregnant group, ²⁶ and a pregnant woman with AT deficiency had significantly increased values compared to both groups. TG values (RCL peak and RCL AUC) were lower in all cases 4 h after LMWH administration, but the reduction was proportional to the baseline value before LMWH administration. In 2017, Kraft et al ²⁷ published a study about homozygous AT deficiency type II HBS pregnant women treated LMWH with positive pregnancy outcome. LMWH treatment with positive pregnancy outcome is also recommended by Bates et al ²⁸ in American Society of Hematology 2018 guidelines. In humans, there is no evidence in Summary of Product Characteristics that enoxaparine crosses the placental barrier during the second and third trimester of pregnancy. Nadroparine should not be used in pregnant women unless the therapeutic benefits to the patients outweigh the possible risks.

Effective anticoagulant treatment with LMWH was manifested in the group of non-pregnant and pregnant women by decrease in RCH peak and RCH AUC values, but even there the response of anti-FXa activity was reflected. Anti-FXa values in pregnant women were also lower than in the non-pregnant group due to their initially higher values of TG RCH peak and RCH AUC. Pregnancy is accompanied by many changes in the hemostatic system due to thrombogenesis and may be manifested by increased TG. At the same time, the weight of pregnant women increases (BMI [N] 24.1 ± 1.8 and BMI [P] 28.9 ± 2.2), so the response to administered LMWH might be lower.

The monitored patient with AT deficiency had significantly increased TG already at the beginning of pregnancy (GW 9) compared to both control groups. TG increased gradually during the pregnancy, which corresponds to the increase in TG in individual GW 9 to 12, GW 13 to 24, and GW 25 to 38. ²⁶ But there was no further significant increase in TG during pregnancy. Therefore, the adjusted dose of heparin was continued and AT concentrate was administered before preparation for delivery.

However, there was also no decrease in TG in the RCH peak and RCH AUC parameters, which we assumed due to zero anti-FXa values and the fact that the RCH peak and RCH AUC parameters should reflect the effect of anticoagulant treatment.

There was no significant increase in the level of D-dimers during the entire pregnancy of a woman with AT deficiency compared to the initial values and the control group of pregnant women. ²⁹ We therefore assumed that the patient benefited from the administered treatment, even if in neither case did it reach the prophylactic dose.

A significant decrease in TG subsequently occurred with AT substitution (Kybernin) before delivery and 1 day and 3 days after delivery, after normal AT values were reached. ³⁰ There were no thrombotic complications in the observed woman during the entire pregnancy.

Limitation

Our study was of exploratory nature only. It was designed in such a way that we could compare the differences between the pregnant and non-pregnant groups and relate the results of pregnant woman with AT deficiency to them. Our results require further confirmation. It would be perfect to have the results of the same number of AT-deficient pregnant women to compare. However, this is not possible due to the rare occurrence of this deficit. Only 1 woman with this type of deficiency in childbearing age is registered at the Thrombotic Centre of General University Hospital. Therefore, we also included her results from the entire pregnancy, while the other 2 groups only had a 1-time follow-up. The statistical comparison is not consequently entirely correct.

Conclusion

Peak thrombin and AUC parameters used for measuring TG appear to be good indicators of procoagulant activity and effectiveness of anticoagulant treatment in pregnant women. TG RCL parameters are then more suitable for pregnant women with AT deficiency. These parameters should be regularly checked in pregnant patients with proven severe thrombophilic condition, whereas sudden increase in TG indicates worsened thrombophilic condition.

The monitoring of TG together with an AT level during AT substitution and simultaneous LMWH administration also appears to be a good prognostic indicator of treatment efficiency. It is more appropriate to monitor AT activity by a kit with factor Xa, in case of such a severe AT deficiency, because it better reflects very low levels of AT.

There is in this study only 1 case of a pregnant woman with this serious deficit, but it was her first pregnancy that ended successfully with delivery without serious thrombotic complications during pregnancy, delivery, and postpartum.

Additional criteria for the evaluation of thrombophilic status during pregnancy were set thanks to this study. It is planned to continue monitoring with a larger group of pregnant women in the future.