Exercise induces two types of human luteal dysfunction: confirmation by urinary free progesterone

We have previously reported that during 2 months of strenuous exercise, untrained young women with documented ovulatory menstrual cycles developed secondary oligoamenorrhea and luteal phase defects. In this study we tested the hypothesis that such abnormalities arise by altered neuroendocrine regulation of menstrual hormone secretion and that weight loss potentiates such effects. We supply a detailed analysis of the 20 cycles, of the total of 53, in which luteal phase abnormalities occurred. During the control month and 2 exercise months, all subjects collected daily overnight urine samples for the determination of LH, FSH, estriol (E3), and free progesterone (P) excretion by RIAs and creatinine by chemical assay. The characteristics of the abnormal luteal phase cycles were determined by comparing the excreted hormone levels and patterns during the control cycles with those of exercise cycles. The area under the curve (AUC) for each hormone was calculated for the follicular and luteal phases of each cycle. Six of the exercise cycles exhibited an inadequate luteal phase. This was characterized by a mean integrated P area of 202.4 (SEM, -61.8) nmol/day.nmol creatinine, compared with 331.7 (SEM, 64.7) during the corresponding control cycles, over a period of 9 or more days after the urinary LH peak to the onset of menses. Fourteen of the exercise cycles exhibited a short luteal phase. This was characterized by a mean integrated P area of 75.9 (30.9) nmol/day.nmol creatinine, compared to 267 (61.7) during the corresponding control cycles, over a span of 8 days or less from the urinary LH peak to the onset of menses. Additional abnormalities occurred only in the short luteal phase cycles. These included an increase in the length and AUC for E3 of the follicular phase and a decrease in the AUC of LH during the luteal phase. We conclude that the initiation of strenuous endurance training in previously ovulating untrained women frequently leads to corpus luteum dysfunction associated with insufficient P secretion and, in the case of short luteal phase cycles, decreased luteal phase length. That exercise may alter the neuroendocrine system is suggested by a delay in the ovulatory LH peak in spite of increased E3 excretion; moreover, less LH is excreted during the luteal phase. The lack of positive feedback to estrogens and decreased LH secretion during the luteal phase could compromise corpus luteum function. In contrast, decreased free P excretion was the sole abnormality noted in menstrual cycles with an inadequate luteal phase.