The Potential Role of GnRH Agonists and Antagonists in Inducing Thyroid Physiopathological Changes During IVF

We conducted an observational cohort study to evaluate whether drugs used for hypothalamic inhibition may impact thyroid function of infertile women scheduled for fresh nondonor in vitro fertilization/intracytoplasmic sperm injection treatment. We considered eligible for inclusion in the study only women with normal thyroid function (serum thyroid-stimulating hormone [TSH] range: 0.2-4.0 mIU/L, serum thyroxin values: 9-22 pmol/L) and negative personal history for previous thyroid disorders. According to which protocols were implemented to gain hypothalamic inhibition, patients were assigned to group A (70 women treated by long gonadotropin-releasing hormone [GnRH] agonist protocol) or to group B (86 women treated by flexible GnRH antagonist protocol). Before initiating controlled ovarian stimulation (COS), both groups were further stratified into 4 subgroups: A1 (46 of the 70 women) and B1 (61 of the 86 women) in women with a baseline TSH value <2.5 mIU/L, whereas those with a baseline value ≥2.5 mIU/L were assigned to groups A2 (24 of the 70 women) and B2 (25 of the 86 women). Prior to initiating stimulation (T-0), 17-β-estradiol (E(2)) and TSH serum values were dosed in all women and repeated on T-5 (day 5 of COS) and subsequently every 2 days until T-ov-ind (ovulation induction day) and T-pick-up (oocytes retrieval day). In case of detection of TSH levels above the cutoff, patients were screened for thyroxin and thyroid autoantibody serum values. In group A, E(2) at T-ov-ind was significantly increased compared to group B (P < .01), whereas TSH values showed an opposite trend (not significantly modified in group A, whereas significantly increased in group B; P < .001). A total of 64 women were found to have TSH values above the cutoff during COS: 7 in group A (11%) and 57 in group B (89%). Among them, 5 (71.4%) of the 7 in group A displayed hypothyroidism (and 4 of the 5 autoantibody positivity), whereas in group B, 6 (10.5%) of the 57 displayed hypothyroidism (and 2 of the 6 autoantibody positivity; P < .001). No pregnancies were observed in women with hypothyroidism, whereas in the 53 women with "isolated" increased TSH (normal T4, negative antibodies), we reported a 20.7% clinical pregnancy rate and a 54.5% ongoing pregnancy rate. Our preliminary data, despite requiring further confirmation, seem to suggest that the various drugs used for gaining hypothalamic control during COS could interfere through different mechanisms with physiological function of thyroid axis, potentially affecting its regulation.