THE MIDCYCLE GONADOTROPIN SURGE IN NORMAL WOMEN OCCURS IN THE FACE OFAN UNCHANGING GONADOTROPIN-RELEASING-HORMONE PULSE FREQUENCY

The midcycle gonadotropin surge is a critical event in normal reproduc tive cycles and requires functional integration of the hypothalamus, p ituitary, and ovary. To determine whether a change in GnRH frequency o ccurs coincident with the onset or termination of the surge in normal women, 20 studies were performed at a sampling interval of every 5 min for up to 36 h. The frequency of pulsatile GnRH secretion was assesse d by the use of two surrogate markers of its secretion, LH and free al pha-subunit (FAS). The timing of the studies was prospectively determi ned by serial ultrasound and previous cycle history, whereas measureme nts of LH, FSH, estradiol, and progesterone in daily blood samples wer e used retrospectively to locate the frequent sampling study in relati on to the day of ovulation in each individual. The frequent sampling s tudies were divided into late follicular phase (LFP; days -4 to -2) an d early, mid-, and late portions of the midcycle surge (days -1 to 1) in relation to the 95% confidence limits of the LH peak derived from d aily samples in 69 normal ovulatory women.The patterns of LH and FAS s ecretion were pulsatile at all times during the midcycle surge. The am plitude of LH pulsations increased from the LFP and early surge to the midportion of the midcycle surge (5.9 +/- and 15.1 +/- 5 vs. 39.0 +/- 3 IU/L; P < 0.0001) and decreased from the mid- to the late portion o f the surge (13.4 +/- 5 IU/L; P < 0.0001). Likewise, the amplitude of FAS pulses increased from the LFP and early surge to the midportion of the surge (82.4 +/- 59 and 153.1 +/- 50 vs. 421.4 +/- 35 ng/L; P < 0. 0001) and decreased from the mid- to the late portion of the surge (19 0.8 +/- 49 ng/L; P < 0.0002). Although there was excellent concordance of pulsatile secretion of LH and FAS, significantly more pulses of FA S were detected than of LH (P < 0.0001). There was no change in freque ncy (expressed as interpulse interval) between the LFP and the early a nd midportions of the surge for LH (70.0 +/- 8, 67.5 +/- 7, and 65 +/- 5 min, respectively) or FAS (55.1 +/- 7, 54.6 +/- 6, and 60.0 +/- 4 m in). However, there was an increase in LH interpulse interval (decreas e in pulse frequency) in the late portion of the surge (87.0 +/- 6 min ) compared to the early and midportions of the surge (P < 0.02 and P < 0.0005, respectively). FAS pulse frequency also decreased significant ly from the early to the late (71.5 +/- 6 min) portion of the surge (P < 0.05). These studies indicate that LH, FAS, and, by inference, GnRH continue to be secreted in a pulsatile manner throughout the duration of the midcycle surge in normal women. Despite the marked increase in pulse amplitude of both LH and FAS that accompanies the onset the mid cycle gonadotropin surge, there is no increase in GnRH pulse frequency , as assessed by two independent markers of its secretion. However, a decrease in GnRH pulse frequency may contribute to the termination of the surge.