Association of basal serum testosterone levels with ovarian response and in vitro fertilization outcome

BACKGROUND To evaluate basal testosterone (T) levels during follicular phase of the menstrual cycle as a predictor for ovarian response and in vitro fertilization (IVF) outcome. METHOD We analyzed data retrospectively from hospital-based IVF center including one thousand two hundred and sixty Chinese Han women under their first IVF cycle reached the ovum pick-up stage, without polycystic ovary syndrome (PCOS) or endometriosis undergoing long IVF protocol. Patients were divided into 2 groups. Group 1: patients with diminished ovarian reserve (basal FSH >10 IU/L) (n = 187); Group 2: patients with normal ovarian reserve (basal FSH < = 10 IU/L) (n = 1073). We studied the association of basal T levels with ovarian response and IVF outcome in the two groups. Long luteal down-regulation protocol was used in all patients, that is, the gonadotropin releasing hormone agonist was administered in the midluteal phase of the previous cycle and use of recombinant FSH was started when satisfactory pituitary desensitization was achieved. RESULTS Basal T levels were markly different between pregnant and non-pregnant women in Group 1; whereas not in Group 2. A testosterone level of 47.85 ng/dl was shown to predict pregnancy outcome with a sensitivity of 52.8% and specificity of 65.3%; and the basal T was correlated with the numbers of large follicles (> 14 mm) on HCG day in Group 1. Significantly negative correlations were observed between basal T, days of stimulation and total dose of gonadotropins after adjusting for confounding factors in both groups. CONCLUSION In women with diminished ovarian reserve, basal T level was a predictor for the number of large follicles on HCG day and pregnancy outcome; but could not in those with normal serum FSH. Basal T levels were associated with both days of stimulation and total dose of gonadotropins, indicating that lower level of T might relate with potential ovarian poor response.