PULSATILE GNRH STIMULATES NORMAL CYCLIC OVARIAN-FUNCTION IN AMENORRHEIC LACTATING POSTPARTUM WOMEN

The postpartum period is characterized hormonally by elevated levels o f PRL and low levels of gonadotropins and sex steroids. In breast feed ing, this state of postpartum amenorrhea can persist for an extended p eriod, even though PRL levels decrease slowly. Although the action of PRL on multiple target sites has frequently been suggested as the caus e of this ovarian quiescence, a suckling-induced alteration in hypotha lamic gonadotropin-releasing hormone (GnRH) production has also been h ypothesized. To test this latter hypothesis, we provided a uniform pul satile GnRH stimulus to eight exclusively breast-feeding women for an 8-week duration beginning at 4 weeks postpartum. Five women with funct ional hypothalamic amenorrhea served as a comparison group. All women received GnRH administered at a dose of 200 ng/kg every 90 min se via a portable infusion pump. Serial blood sampling For LH, FSH, and PRL w as performed weekly for 5 h at 10-min inten;als beginning immediately before initiation of GnRH, during the period of GnRH, and 1 week after the cessation of GnRH. The women collected daily urine aliquots for e strone-3-glucuronide, pregnanediol-3-glucuronide, and LH determination s. Serial transvaginal sonography was used to monitor follicular devel opment. Before GnRH treatment, the urinary steroid and serum gonadotro pin levels of the two groups were low and similar. As expected, PRL le vels were higher in the postpartum women(87 mu g/mL us. 4.25 mu g/L P < 0.05). After initiation of pulsatile GnRH, LH values increased and F SH values decreased in both groups. The LH increase with GnRH was sign ificantly greater in the breast-feeding group than in the hypothalamic amenorrhea group (19.75 mIU/mL vs. 12.34 mIU/mL, P < 0.05). Analysis of pulse frequency and amplitude revealed a nearly complete 1:1 induct ion of LH pulses by the exogenous GnRH in both groups, with the breast -feeding group showing a greater amplitude (12.26 mIU/mL cs. 5.34 mlU/ mL, P < 0.05). The cycle lengths, urinary steroids. and vaginal ultras onography demonstrated a more rapid initial ovarian responsiveness in the breastfeeding group, as determined by the length of the first foll icular phase. The breast-feeding group also showed a brisker ovarian r esponse, as evidenced by a greater number of follicles that were 12 mm or greater (2.3 vs. 1.2, P < 0.05), and a greater luteal phase peak a nd integrated pregnanediol excretion, respectively (3.02 mu g/L creati nine and 39.87 mu g/L creatinine/cycle vs. 1.89 mu g/L creatinine and 7.69 mu g/L creatinine/cycle, P < 0.05). These results demonstrate tha t the pituitary gonadotropes and the ovary remain fully responsive to GnRH stimulation in amenorrheic breast-feeding women and that chronic treat ment with pulsatile GnRH can restore cyclic ovarian activity, ov ulation, and normal luteal function, with responses at least as brisk as in patients with hypothalamic amenorrhea. Thus, postpartum amenorrh ea seems to result entirely from reduced GnRH secretion, without signi ficant inhibition of intrinsic gonadotrope or ovarian function.