Abstract

OBJECTIVES: To measure the long term effects of dance training and the contribution of the timing and duration of any menstrual disruption on bone mineral density (BMD). DESIGN: Measurement of BMD in 57 premenopausal, previously professionally dance trained women and the relationship to menstrual and training history. MAIN OUTCOME MEASURES: Bone density measurements at lumbar spine and femoral neck by dual energy x-ray absorptiometry. RESULTS: The average Z score for BMD at the lumbar spine in the amenorrhoeic dancers was significantly below that for the normal population. The average Z score for BMD at the femoral neck in the eumenorrhoeic dancers was significantly above that for the normal population. There was a significant difference between the average Z score for BMD at both the lumbar spine and femoral neck between the amenorrhoeic and eumenorrhoeic dancers. Significant negative relationships were found between BMD at the lumbar spine and (1) age at menarche, (2) duration of amenorrhoea, (3) BMD at the femoral neck, and (4) the variable of ideal minus lowest weight, which was independent of amenorrhoea. No significant relationships were found between duration of oral contraceptive pill usage and BMD at either the lumbar spine or the femoral neck in eumenorrhoeic or amenorrhoeic dancers. In order to quantify the effect of a combination of these significant factors, a model of BMD was constructed using multiple regression incorporating the variables duration of amenorrhoea, age at menarche, and ideal minus lowest body weight. In this model R2 was 33.6%, in other words 33.6% of the total variation in the Z score for BMD at the lumbar spine could be accounted for by these factors. CONCLUSION: Professional female dancers with a history of delayed menarche and amenorrhoea have been identified as another group of premenopausal women potentially at risk of developing osteoporosis because of a decrease in BMD at the lumbar spine. The femoral neck in dancers with a history of amenorrhoea was partially protected from loss of BMD by virtue of being the major weight bearing site in previous dance training, and in eumenorrhoeic dancers BMD was significantly increased at this site.