The underlying mechanisms involved in the pathogenesis of amenorrhoea in hyperprolactinaemic states still remain unclear. Conflicting information exists on the role of endogenous opiates on gonadotrophin disturbances in this pathological condition. In this study we have undertaken a detailed investigation of LH and PRL secretion before and during administration of naloxone, an opioid receptor blocker, in hyperprolactinaemic women with or without ovarian function in order to assess the role of ovarian steroids upon naloxone induced LH and PRL release. Five anovulatory and six ovulatory subjects with hyperprolactinaemia were studied before and during naloxone infusion. Five normo-prolactinaemic ovulatory subjects were included as controls. All ovulatory subjects were studied during the luteal phase of a menstrual cycle. Blood was sampled every 10-20 minutes over a 16-hour period on two alternate days. On study day 1 (control day), subjects received two sets of saline infusion every 6 hours and one saline bolus at the beginning of the seventh hour; on study day 3 (naloxone day), they received a saline infusion during the first 6 hours, an intravenous bolus of naloxone (20 mg) at the beginning of the seventh hour and then a continuous naloxone infusion (1.6 mg/hour) during the ensuing 6 hours. Pituitary LH responsiveness and reserve were assessed on both study days by the subsequent administration of 5 and 95 micrograms of GnRH 4 hours before the completion of each sampling period. Serum concentrations of LH, PRL, oestradiol and progesterone were determined by radioimmunoassay. LH and PRL pulse detection and characteristics were analysed by the Cluster program. Serum PRL levels in hyperprolactinaemic anovulatory and ovulatory subjects were significantly elevated above the normal range. Oestradiol and progesterone serum levels during the luteal phase in women with hyperprolactinaemia and regular menses were similar to those in control ovulatory subjects. Mean LH concentrations increased during naloxone infusion (P < 0.05) in ovulatory hyperprolactinaemia and controls, whereas PRL increased (P < 0.05) only in the group of control subjects. LH pulse amplitude and pulse interval were increased by naloxone (P < 0.05) in all the ovulatory subjects, with no significant changes in anovulatory hyperprolactinaemic women. PRL pulse characteristics were modified significantly by naloxone only in the control group. On day 1, GnRH administration increased LH in all groups, whereas a consistently lower pituitary LH response was observed after naloxone (day 3). Serum PRL levels significantly increased after GnRH administration on day 1 only in normal women, whilst on day 3 this GnRH-dependent PRL releasing effect was significantly attenuated. The absence of stimulatory effects of naloxone on LH in anovulatory hyperprolactinaemia implies that endogenous opiates do not play a significant role in the mechanisms governing hypothalamic amenorrhoea in this syndrome. The results in subjects with ovulatory hyperprolactinaemia suggest the existence of an active role of ovarian steroids on naloxone induced LH release. These data, along with those previously reported in normal women throughout the menstrual cycle, are consistent with the concept that sex steroid hormones contribute to the underlying mechanisms involved in the opioidergic control of LH and PRL release. Whether PRL by itself or through other non-opioid neuroendocrine pathways alters the hypothalamic-gonadotroph unit still requires further investigation.