Effects of controlled ovarian stimulation on thyroid stimulating hormone in infertile women

To study the effects of long-acting gonadotropin-releasing hormone agonist (GnRH-a) on thyroid function in euthyroid patients of in vitro fertilization (IVF)/ intracytoplasmic sperm injection of embryo transfer (ICSI-ET) and to investigate the timing and alteration of thyroid stimulating hormone (TSH) during controlled ovarian stimulation(COS).

Euthyroid patients scheduled for IVF/ICSI were enrolled. Euthyroidism was defined as having no history of hypothyroidism with normal TSH before IVF. Long GnRH-a protocol was chosen as COS protocol. 207 patients were divided into two groups based on basal serum TSH level: group A with 0.35mIU/L＜TSH＜2.5mIU/L (n = 137) and group B with 2.5mIU/L ≤ TSH＜4.5mIU/L (n = 70). Serum TSH was tested on 6 time points: before COS (2-5days in menstrual cycle, before GnRH-a injection), Gn injection day 1, Gn injection day 5, human chorionic gonadotropin (HCG) day, 14 and 28 days after transplantation. The serum TSH, clinical pregnancy and abortion rate were investigated.

The serum TSH value was significantly (P < 0.05) increased after injection of long-acting GnRH-a in all patients. Both groups had significant (P < 0.05) increases in serum TSH level after long-acting GnRH-a injection. The TSH level was increased in 131(63.3%) patients after GnRH-a injection, of which twenty (9.7%) had subclinical hypothyroidism with TSH level over 4.5 mIU/L. The other 76 (36.7%) patients had decreased TSH. In group A, 79 (57.7%) patients showed an increase of TSH, including three patients (2.2%) with simultaneous rise of TPOAb and four (2.9%) diagnosed of subclinical hypothyroidism with TSH level over 4.5 mIU/L, and the rest fifty-eight (42.3%) patients had decreased TSH with one patient with elevated TPOAb who was diagnosed with subclinical hyperthyroidism. In group B, fifty-two (74.3%) patients showed an increase of TSH, including thirteen (18.6%) patients with elevated TPOAb and sixteen (22.9%) patients diagnosed of subclinical hypothyroidism with TSH level over 4.5 mIU/L, and the rest eighteen (25.7%) patients had decreased TSH with one patient diagnosed with subclinical hyperthyroidism. Group B had a significant higher proportion of patients with elevated serum TSH than group A (P ＜ 0.05). Compared to the baseline level, serum TSH ascended distinctly and reached peak level on HCG day in all patients. Group A and B had similar trends of alteration. Patients in group A had significantly (P＜0.05) higher clinical pregnancy rate than in group B. No significant (P＞0.05) difference in abortion rate were observed between the two groups.

GnRH-a can significantly increase serum TSH levels with possible development of subclinical thyroid dysfunction. Infertile patients with serum TSH > 2.5 mIU/L are more susceptible to GnRH-a while patients with basal TSH less than 2.5 mIU/L may get a higher clinical pregnancy rate when receiving IVF/ICSI.