OBJECTIVE Hyperprolactinaemic amenorrhoea is associated with disturbances of pulsatile gonadotrophin secretion. The underlying mechanism remains unclear and the aim of this study was to investigate the 24-hour secretory pattern of gonadotrophins in women with hyperprolactinaemic amenorrhoea. The effect of opioid blockade using naloxone infusion on LH secretory pattern was also studied. METHODS The secretory patterns of LH, FSH, PRL and their responses to naloxone infusion were studied by serial blood samples collected at 10-minute intervals for 24 hours. On the following day, naloxone was infused at a dose of 1.6 mg per hour for 4 hours. METHODS Eight women with hyperprolactinaemic amenorrhoea, two women hyperprolactinaemic but with normal ovarian cycles, and nine control subjects in the early follicular phase of menstrual cycle. METHODS Concentrations of LH, FSH and PRL were measured in plasma samples obtained at 10-minute intervals for 24 hours. In one woman, concentrations of urinary oestrone glucuronide were measured daily during treatment with pulsatile GnRH. RESULTS The number of LH pulses per 24 hours was significantly fewer in women with hyperprolactinaemic amenorrhoea than in those with hyperprolactinaemia with normal cycles or control subjects (mean +/- SEM 4.5 +/- 2.4 vs 13.5 +/- 2.5 vs 17.3 +/- 0.8, P < 0.001). The magnitude of each episode of secretion was significantly higher in the hyperprolactinaemic amenorrhoeic women (P < 0.05) so the overall mean concentrations of LH throughout the 24-hour period was similar in the three groups (5.2 +/- 1.1, 4.8 +/- 0.8 and 5.2 +/- 0.4 U/l respectively). In women with hyperprolactinaemic amenorrhoea there was no significant change in the pattern of LH secretion during sleep in contrast to the control women in whom there was a slowing in the LH pulse frequency during the night. There was no significant change in the mean concentrations of LH, FSH and PRL during the naloxone infusion. There were also no significant changes in the LH pulse frequency in response to naloxone infusion when compared with an equivalent period of time in the previous 24 hours. In one hyperprolactinaemic amenorrhoeic woman, follicular development, ovulation and pregnancy were induced when gonadotrophin releasing hormone (GnRH) was infused in a pulsatile manner at a dose of 5 micrograms every 90 minutes. CONCLUSIONS The suppression of normal ovarian cycles in women with hyperprolactinaemic amenorrhoea is due to a significant reduction in frequency of LH (GnRH) secretion which is not due to an increase in hypothalamic opioid activity. As normal ovarian cycles can occur or be induced by exogenous GnRH in hyperprolactinaemia, it is unlikely that a high level of prolactin by itself inhibits follicular development and ovulation.