Supplementary growth hormone treatment of women with poor ovarian response to exogenous gonadotrophins: changes in serum and follicular fluid insulin-like growth factor-1 (IGF-1) and IGF binding protein-3 (IGFBP-3)

The effects of supplementary growth hormone (GH) treatment uponinsulin-like growth factor-1 (IGF-1), IGF binding protein-3(IGFBP-3) concentrations in serum and ovarian follicular fluidwere investigated in women undergoing buserelin human menopausalgonadotrophin (HMG) ovulation induction for in-vitro fertilization.Women (n = 40), aged 24–39 (mean 35 years), who showedpoor ovarian responses to HMG, were recruited and randomly dividedinto two groups. Each patient received two cycles of ovulationinduction, one with GH (12 IU/day x 12 days/HMG/buserelin) andanother with placebo/HMG. Serum IGF-1 increased substantiallyduring the GH treatment and remained significantly higher thanthe control 2 days after the last GH injection. Serum IGFBP-3fell steadily during the placebo/HMG treatment and to a nadiron the day of oocyte retrieval (P <0.05 compared to serumbefore any treatment). In contrast, IGFBP-3 was increased (P<0.01) during the GH administration and returned to the controllevel 2 days after GH injection. Serum oestradiol concentrationson the eighth day of HMG and the day of human chorionic gonadotrophin(HCG) were not significantly different between the two groups.Serum IGF-1 was highly correlated with IGFBP-3 before any treatment(r = 0.433, P < 0.001). This correlation disappeared afterbuserelin, placebo/HMG treatment in the control group, but itwas maintained during GH/HMG treatment (r = 0.343, P = 0.04).Follicular fluid concentrations of GH and IGF-1, not IGFBP-3or oestradiol, were significantly elevated in the GH-treatedwomen. Serum IGF-1 on the day of oocyte retrieval was highlycorrelated to the follicular fluid IGF-1 in both groups. Therelationships between the follicular fluid GH and IGF-1 werecompletely opposite in the two groups, being positive in thecontrol group and negative in the GH-treated group. In the controlgroup, significant correlations were found between follicularfluid concentrations of IGF-1 and IGFBP-3, and GH and IGFBP-3which were not found in the GH-treated group. There were nocorrelations found between follicular fluid concentrations ofGH or IGF-1 or IGFBP-3 and oestradiol. The results clearly demonstratethat the normal GH, IGF-1, IGFBP-3 relationships can be alteredby treatments which influence the ovarian—pituitary axis;the significance of such changes to ovulation remains to bediscovered.