In order to elucidate whether exercise and calcium intake can offset bone loss due to hypoestrogenemia in teenage girls, the lumbar and femoral bone mineral density (BMD) in elite amenorrheic runners (group 1 (n = 8)) was compared with those in eumenorrheic athletes (group 2 (n = 15)) and two groups of eumenorrheic non-athletic subjects, group 3 (n = 14), and group 4 (n = 15). BMD was measured by dual energy x-ray absorptiometry. The subjects in group 1 lived in the dormitory, and their calcium intake was controlled to > 1400 mg/day. The subjects in groups 1, 2 and 3 were age matched (16-17 years old), but the body weight of those in group 2 was significantly greater than that of others. The subjects in groups 1, 3 and 4 were weight matched, but group 4 subjects were older by 5 years (21.0 +/- 0.2 years old). The lumbar BMD of group 1 was 1.0460 +/- 0.0259, which was not statistically different from those in the other groups, while the femoral BMD of group 1 was significantly higher than those of groups 3 and 4 at trochanter. The lumbar and femoral BMD of group 2 was significantly higher than in groups 3 and 4, but was not statistically different from group 1. There were no significant differences in serum parathyroid hormone levels, but osteocalcin and calcitonin levels of group 1 were significantly lower than those of groups 2 and 3, and groups 2, 3 and 4, respectively, suggesting that the rate of bone formation is suppressed in amenorrheic runners.(ABSTRACT TRUNCATED AT 250 WORDS)