Numerous studies have shown that low estrogen states, such as those observed in hyperprolactinemia and post-menopause, are associated with low levels of skeletal bone mass. Physical activity has been demonstrated to inhibit and even reverse bone loss in postmenstrual women. Since the exertions of athletes far exceed those of post-menopausal women, such activity might be expected to have a protective effect against bone loss in the amenorrheic athlete. The present study was designed to test that assumption.Twenty-eight women athletes, 14 of them amenorrheic and 14 eumenorrheic, participated in the study. The two groups were matched for the following variables: sport, age, weight, height, and the frequency and duration of daily training sessions. Single-photon and dual-photon absorptiometry were used to measure regional bone mass at the distal radius and lumbar vertebrae, respectively. Single-photon measurements were made on the nondominant arm at two sites, S1and S2. The bone mineral density of the axial skeleton was determined by dual-photon ab-sorptiometry.Neither bone mineral content nor bone mineral density at the two sites along the radius differed between the two groups, but the mineral density of the lumbar vertebrae was significantly lower in the amenorrheic athletes. Although the mineral content and density at S1and S2were significantly related (r = 0.72), there was no significant relationship between the mineral density of the vertebrae and that at either radial site (S1, r = 0.31; S2, r = 0.34).Estradiol levels, expressed either as an average of four samples or as the peak value, were significantly lower in the amenorrheic group. Eight of these women had estradiol values under 45 pg/ml in each of their four samples. Three of the other women had at least one sample each in which the value was between 45 and 100 pg/ml, and the remaining three had one or more samples each in which the value exceeded 100 pg/ml. Twelve of the eumenorrheic women had estradiol profiles that were characteristic of women with regular cycles, and at least one estradiol value was above 130 pg/ml. The other two women reported menstruating within 12 days of entering the study and having had regular periods during the previous 12 months.All four progesterone values for 13 of the amenorrheic women were less than 0.65 ng/ml, which suggests that ovulation had not occurred. One woman had a peak concentration of 13.1 ng/ml that was coincident with an estradiol level of 93.6 pg/ml, but she did not experience menstrual flow. Progesterone concentrations for the eumenorrheic group indicated the presence of ovulatory cycles in 11 of the subjects. The peak progesterone level of two of the other women was under 2.0 ng/ml, which suggests the presence of an anovulatory cycle. Both women had peak estradiol levels in excess of 250 pg/ml.Prolactin levels differed between the two groups also. Two women had at least one prolactin value each above 70 ng/ml and a four-sample average that was above 60 ng/ml. Both were in the eumenorrheic group and reported the occurrence of regular cycles during the previous 12 months. The relationship between prolactin and estradiol was significant for both mean values (r = 0.54; P 0.01) and peak concentrations (r = 0.44; P 0.02). Testosterone levels did not differ between groups and were within the normal range for women.The dietary intake of the two groups was similar, but the lower total intake of calories (P 0.06) and of fat (P 0.06) of the amenorrheic group closely approached the level of significance (a = 0.05) established for this study. There was no difference between the amenorrheic and eumenorrheic women in calcium intake, either through diet alone or through diet plus supplements. For both groups, intake exceeded the current recommended dietary allowance of 800 mg per day.Although both mean and peak estradiol values were significantly different between groups, neither value was significantly correlated with bone mineral density at any site. With the exception of a significant relationship between age and mineral density of the radius (S1, r = 0.52; S2, r = 0.48), none of the physical characteristics, training factors, hormone levels, or dietary variables was related to the mineral density of the radius or of the lumbar vertebrae.
展开▼
