P-622 Prothrombotic biomarkers during controlled ovarian stimulation for assisted reproductive techniques Free

Does the evolution of prothrombotic biomarkers over time differ between antagonist and long agonist stimulation protocols for assisted reproductive techniques (ART) ?

The hypercoagulable state was higher and persistent in the agonist and antagonist with hCG triggering groups compared to the antagonist with GnRH agonist triggering group.

Controlled ovarian stimulation (COS) for ART is associated with supra-physiological serum estradiol levels, a hypercoagulable state and an increased risk of venous thrombosis. Most thromboembolic events associated with COS occur in the context of ovarian hyperstimulation syndrome (OHSS). The use of hCG for final follicular maturation increases the risk of OHSS. In antagonist protocols, GnRH agonist triggering is known to prevent or reduce OHSS and is therefore widely used in women at risk. The impact of the different IVF protocols on pro-thrombotic biomarkers is unknown.

In this prospective observational cohort study, infertile women undergoing COS for ART in 2017-2019 at the University Hospitals of Geneva and Lausanne (Switzerland) were included. We evaluated changes in key coagulation parameters (D-dimers, factor VIII, fibrinogen activity, protein S and protein C) and thrombin generation, our primary outcome, (using 5 pM of tissue factor) by calibrated automated thrombinography before stimulation (T1), on the day of ovulation triggering (T2) and seven days after triggering (T3).

COS was started without hormonal pre-treatment. Protocols were prescribed according to the standards used in each centre taking into account the risk of OHSS (agonist protocol with hCG trigger in women without OHSS risk (Group 1); antagonist protocol in women at risk of OHSS with hCG trigger (Group 2;) or GnRH agonist trigger (Group 3;); variation of endogenous thrombin potential (ETP) was measured and compared among groups using mixed effects linear regression model.

A total of 64 women were included: 24 were in group 1, 16 in group 2, and 24 in group 3. The mean age (SD) was 37.8 (2.8), 35.9(5.2) and 34(4.6) years in groups 1, 2 and 3 respectively. As expected, women in group 1 had a statistically lower level of anti-müllerian hormone (p = < 0.001), a lower antral follicular count (p = < 0.001) and lower number of MII oocytes and embryos obtained (p = < 0.001). Mean serum estradiol levels were 1836 (1160), 1628 (815) and 3754 (2165) ng/L at T2, and 945 (471), 1061 (495) and 413 (729) ng/L at T3, in group 1 to 3, respectively. In multivariable regression analysis, the levels in group 3 were statistically higher at T2 and lower at T3 (overall time*group interaction: p < 0.001).

The mean ETP was similar between all groups at T1, and increased in all groups at T2 (1442, 1426 and 1486 nM/min in groups 1, 2 and 3, respectively) (p = 0.013). Overall, ETP evolution over time was statistically different between groups, with the lowest increase of ETP between T1 and T3 in group 3. Protein C and protein S levels were stable, while D-dimers, fibrinogen and factor VIII increased at T2 and T3 in all groups.

Stimulation protocols were prescribed according to the clinical profile and OHSS risks; groups therefore differ substantially in regards to age and ovarian reserve. Thromboembolic events are rare events after COS, we therefore evaluated biological markers of hypercoagulability and not clinical events.

Women with GnRH agonist triggering protocol did not increase mean ETP in the week after ovulation, while women with hCG triggering did. This different prothrombotic profile was independent of the variation of the other coagulation parameters investigated. This effect of ovulation triggering should be confirmed by further studies.

NCT04188444