Intrathecal administration of opioids is a very efficient tool in the long-term control of intractable nonmalignant pain. However, despite the well known role of opioids in endocrine regulation, few data are available about possible effects on hypothalamic-pituitary function during this treatment. Seventy-three patients (29 men and 44 women; mean age, 49.2 +/- 11.7 yr) receiving opioids intrathecally for nonmalignant pain were enrolled for extensive endocrine investigation. At the time of hormonal determination, the mean duration of opioid treatment was 26.6 +/- 16.3 months; the mean daily dose of morphine was 4.8 +/- 3.2 mg. The control group consisted of 20 patients (11 men and 9 women; mean age, 54.2 +/- 14.0 yr) with a comparable pain syndrome but not treated with opioids. Decreased libido or impotency was present in 23 of 24 men receiving opioids. The serum testosterone level was below 9 nmol/L in 25 of 29 men and was significantly lower than that in the control group (P < 0.001). The free androgen index was below normal in 18 of 29 men and was significantly lower than that in the control group (P < 0.001). The serum LH level was less than 2 U/L in 20 of 29 men and was significantly lower than that in the control group (P < 0.001). Serum FSH was comparable in both groups. Decreased libido was present in 22 of 32 women receiving opioids. All 21 premenopausal females developed either amenorrhea or an irregular menstrual cycle, with ovulation in only 1. Serum LH, estradiol, and progesterone levels were lower in the opioid group. In all 18 postmenopausal females significantly decreased serum LH (P < 0.001) and FSH (P = 0.012) levels were found. The 24-h urinary free cortisol excretion was below 20 microg/day in 14 of 71 opioid patients and was significantly lower than that in the control group (P = 0.003). The peak cortisol response to insulin-induced hypoglycemia was below 180 microg/L in 9 of 61 opioid patients and was significantly lower than that in the nonopioid group (P = 0.002). The insulin-like growth factor I SD score was below -2 SD in 12 of 73 opioid patients and was significantly lower than that in the control group (P = 0.002). The peak GH response to hypoglycemia was below 3 microg/L in 9 of 62 subjects and was significantly lower than that in the control group (P = 0.010). Thyroid function tests and PRL levels were considered normal. No metabolic disturbances were recorded, apart from significantly decreased high density lipoprotein cholesterol levels (P = 0.041) and elevated total/high density lipoprotein cholesterol ratio (P = 0.008) in the opioid group compared to the control group. Supplementation with gonadal steroids improved sexual function in most patients. In conclusion, of all patients receiving intrathecal opioids, the large majority of men and all women developed hypogonadotropic hypogonadism, about 15% developed central hypocorticism, and about 15% developed GH deficiency. These findings suggest that further investigations are required to determine the need for systematic endocrine work-up in these patients and the necessity for substitutive therapy.