Prediction of over-response to ovarian stimulation in an intrauterine insemination programme.

Abstract

Prediction of poor-response is of equal importance to prediction of over-response in intrauterine insemination programmes. The gonadotrophin-releasing hormone agonist (GnRHa) stimulation test (GAST) was assessed as a predictor of over-response to ovarian stimulation in 81 patients. Blood samples were taken on cycle day 2 (before and 24 h after starting the GnRHa). Day 2 and 3 samples were assayed for oestradiol, follicle stimulating hormone (FSH) and luteinizing hormone (LH). Linear and logistic regression analyses were used to assess age, day 2 FSH, day 2 FSH/LH, oestradiol ratio (oestradiol on day 3/oestradiol on day 2) and FSH ratio (FSH on day 3/FSH on day 2) as predictors of the number of follicles (total and > or = 14 mm), oestradiol on HCG day, and clinical pregnancy rate as appropriate. Several parameters were also compared between the patients who produced < or = 3 (> or = 14 mm) follicles (group A) and those who produced >3 (> or = 14 mm) follicles (group B). The mean +/- SEM age of the patients in the study was 32 +/- 0.4 years. The mean total dose of recombinant FSH was 800 +/- 20 IU and the mean duration of stimulation was 7.6 +/- 0.2 days. Nine (11%) and 12 (15%) patients were cancelled for poor and over-response respectively. The oestradiol ratio was significantly positively correlated with oestradiol on HCG day (P < 0.001), and with the number of mature follicles (> or = 14 mm) (P = 0.01). Age, day 2 FSH and FSH ratio were not significantly correlated with oestradiol on HCG day, total follicles and follicles > or = 14 mm. None of the above-mentioned variables was correlated with clinical pregnancy rate. Group A had significantly lower oestradiol ratio (P = 0.007), longer duration of stimulation (P = 0.002), higher total FSH dose (P = 0.001), and lower oestradiol on HCG day (P = 0.001). GAST is therefore useful in predicting the high responders to gonadotrophin stimulation.