Thrombin is a key protein for the regulation of the hemostatic process, since it has procoagulant, anticoagulant, fibrinolytic and antifibrinolitic properties.1 Thrombin formation can be assessed by clot-based assays, prothrombin time and activated partial thromboplastin time, which are limited by an endpoint that occurs when only 5% of thrombin is produced. The thrombin generation test (TGT) was introduced by Macfarlane & Biggs in 1953 and the development of a commercial semi-automated Calibrated Automated Thrombogram (CAT) method (Thrombinoscope BV®, Maastricht, Netherlands), allowed for the measurement of thrombin generation (TG) potential in the plasma, over 60min, after addition of the tissue factor (TF), phospholipids and a fluorescent substrate containing calcium with consequent activation of coagulation.2–7 Using Thrombinoscope® software it is possible to obtain the TG curve (time versus thrombin concentration) and calculate TG parameters, such as the endogenous thrombin potential (ETP), peak, and lag time.

The TG curve is characterized by a lag time phase, followed by the formation of a large amount of thrombin (propagation), which culminates in a peak of thrombin concentration and finally, in the inhibition of thrombin by natural anticoagulants.8 The area under the curve, termed ETP, represents the total amount of thrombin formed over 60min.4 It is known that clot formation occurs at the end of the lag time, therefore the duration of this parameter corresponds to the coagulation time.9 From the analysis of these parameters in a TG curve, it can be suggested that a prolonged lag time and decrease of both the ETP and peak indicate a state of hypocoagulability, characterized by a lower thrombin formation. On the other hand, the increase in TG is characterized by a decrease in the lag time and an increase in ETP and peak values.10 Therefore, this method is sensitive to hypo- and hypercoagulability states and has numerous research and potential clinical applications, such as screening for genetic or acquired thrombotic disorders or estimation of bleeding risk for hemophiliac patients.10 Using platelet-poor plasma (PPP), the CAT method is sensitive to detect all coagulation factor deficiencies (except FXIII) and the effect of vitamin K antagonists, heparinoids, and direct oral anticoagulants. In platelet-rich plasma, the method is sensitive to the von Willebrand disease and can detect the effect of platelet inhibitors. In addition, in the presence of activated protein C or thrombomodulin, it is possible to study congenital or acquired disorders of the anticoagulant system.7,11,12

However, substantial differences in pre-analytical and analytical variables for the TG technique among laboratories currently still exist, which make the comparison of results from different studies difficult.1,13,14 Therefore, standardized pre-analytical conditions and a validated method are still needed to introduce this test in clinical practice. Since studies on the TGT standardization are still scarce, this study proposed to analyze some important analytical variables, as well as the sensitivity of this method in hypo- and hypercoagulability states in order to contribute to the standardization of the TGT by the CAT method.

All data were analyzed for normal distribution by the Shapiro Wilks test. The results were presented as mean and standard deviation (SD) when they presented normal distribution, and as median and interquartile range (IQR), when their distribution was not normal. The Student t and Mann–Whitney tests were used for comparison of the two groups, when the data were normal or non-normal, respectively. The Kruskal–Wallis test (non-normal data) was used to compare three or more groups. The PRISMA 6.0 software was used to make these analyses. A p value less than 0.05 was considered statistically significant.

The TGT is a method with potential clinical application as a global assay directed to screen genetic or acquired thrombotic and bleeding disorders. However, substantial differences in pre-analytical and analytical variables still exist, which make the comparison of results from different studies difficult.1,13,14 Since studies on TGT standardization are still scarce, this study aimed at evaluating the analytical variables, as well as the performance of the TGT in hypo- and hypercoagulability states. We found that OC use was associated with increased TG, which was more pronounced in reactions performed with low TF. In FVIII-deficient plasma, higher TG was associated with higher FVIII levels with a better discrimination with a low TF. In FVII-deficient plasma, the TG was lower when compared to normal plasma, and a more pronounced difference was observed when the reactions were performed with a high TF. The intra-assay CV was below 10% when using both a low and a high TF.

Under the conditions of hypocoagulability, in the test with FVIII-deficient lyophilized plasma, with a low TF, an increase of 121% in the ETP and 563% in the peak was observed in the plasma with 50% of FVIII:C compared to <1% of FVIII:C. Furthermore, there was an increase of 22% in the ETP and 50% in the peak in plasma with 100% of FVIII, compared to 50% of this factor.

Salvagno et al.18 studied the TG in a normal pool plasma diluted with FVIII-deficient plasma to obtain concentrations between 5% and 100% of FVIII. The reaction was performed with 0.179pM of TF and the researchers observed an increase in the TG, with an increase in the FVIII concentration. They concluded that the TG is an adequate technique to evaluate the global hemostatic capacity of patients with hemophilia A.

In reactions with a high TF, there was a reduced discrimination between different concentrations of FVIII, since an increase of only 1% in the ETP and 6% in the peak was observed in plasma with 50% of FVIII:C compared to <1%, and an increase of 38% in the ETP and 23% in the peak in plasma with 100% of FVIII:C compared to plasma with 50% FVIII:C. In the experiment with plasma samples from patients with hemophilia A at 1, 3 and 7 days after the infusion of FVIII, we observed that the TGT was able to detect differences in the hemostatic profile after the infusion of recombinant FVIII at different times. An analysis of the TG profile revealed decreasing values as a function of the time of infusion of FVIII, i.e., the longer the time elapsed after taking FVIII concentrate, the higher the hypocoagulability state became. Consistent with the results already observed for the FVIII-deficient lyophilized plasma, 1pM TF reactions were more sensitive to detect different factor concentrations.

As expected, the TG profile with a low TF better discriminated the different concentrations of FVIII, since the TF at low concentrations makes the reaction slower. Therefore, a low TF is more sensitive to study the deficiencies of factors related to the intrinsic pathway (FVIII, IX and XI). In conclusion, the concentration of TF influences the sensitivity of the test to different coagulation factors and when high TF was used, a good discrimination between the different concentrations of FVIII was not found.10,19

In the experiment with FVII-deficient plasma, the results with low and high TF were similar, showing a decrease of 23% in the ETP and 42% in the peak using high TF, and a decrease of 28% in the ETP and 36% in the peak using a low TF. A more expressive difference was in the lag time, which increased 83% with a high TF and 66% with a low TF in FVII-deficient plasma, when compared to normal plasma. This can be explained because the lag time corresponds to the initiation phase of the TG and, according to Dielis,11 this parameter is determined by the FVII. The coagulation cascade is initiated by the binding of TF to FVII. In addition, a longer lag time was observed in the reactions with a high TF because the use of high concentrations of TF makes the reaction faster and more sensitive to the changes in extrinsic pathway factors.

The TGT in hypercoagulable states showed that reactions with the TF in low concentration (1pM), OC users had higher TG when compared to non-users. In reactions with a high TF, OC users also had a higher potential for TG in relation to non-users, but the increase was less obvious. A comparison of the TG results using a low and a high TF reveals that OC users appeared to be more “hypercoagulable” compared to non-users, with a more pronounced difference when the reaction was performed with a low TF.

Combined oral contraceptives may induce significant changes in the coagulation system, such as increased coagulation factors (fibrinogen, VII, VIII, IX, X, XII and XIII), reduction of the anticoagulants protein S (PS) and antithrombin, and acquired resistance to the activated protein C, culminating in the increase in the TG.20 In this study, reactions with a low TF better discriminated the differences for ETP and peak values between OC users vs non-users. The increase in the TG in OC users may be related to an increase in FVIII and a decrease in the PS and (TFPI). The reaction with a low concentration of TF is more sensitive to changes in the intrinsic pathway and deficiencies in the TFPI, which requires a slower reaction to exert its effect. The decrease in PS levels may lead to a decrease in the action of the TFPI, since the PS is a cofactor of the TFPI.21–23 The results obtained in the present study suggest that the plasma of OC users show parameters in the TGT which are compatible with hypercoagulability, when compared to non-users. Such differences are more evident when the reactions occur with low TF concentrations.

As a limitation of this study, it should be noted that, unfortunately, it does not solve the main problems with the TGT, that is, the variation in the results of the inter-laboratory tests. However, this is beyond the scope of our study. As a manner of minimizing errors and decreasing intra- and inter-assay coefficients of variation, we decided to work only with Stago® kits and not with homemade reagents. According to Dargaud et al.,24 the use of reagents from the same source, as well as the use of the same procedure/technical protocol, significantly reduces the variability in the results.

Finally, in agreement with previous findings, we recommend the ideal conditions for the performance of the TGT as: (i) if the TG test is performed with Stago® reagents (PPP-Reagent Low® or PPP Reagent High®), 20μL of 1pM (lower concentration of TF) should be used or 5pM of TF (higher concentration of TF), according to the coagulation pathway to be evaluated, whether intrinsic or extrinsic; (ii) using low concentrations of TF (1pM – 20μL of PPP-Reagent Low®) the test is more sensitive to changes in the intrinsic pathway and (iii) using high concentrations of TF (5pM – 20μL of PPP-Reagent High®) the test is more sensitive to changes in the extrinsic pathway.

The TGT by the CAT method using a low and a high TF at concentrations of 1.0 and 5.0pM, respectively, can be reproducible with acceptable precision, with low inter- and intra-assay coefficients of variation. In addition, the TGT proved to be sensitive to detect hypo- and hypercoagulability states. The TGT may have useful applications in the hemostatic evaluation of bleeding and thrombotic disorders. However, a standardized protocol for the TGT is important to minimize the large inter-laboratory variability in results. We suggest that a multicenter study be undertaken to validate the use of the TGT in hyper- and hypocoagulability states in clinical practice.

The authors declare no conflicts of interest.