The introduction of dual‐energy X‐ray absorptiometry (DXA) in the 1980s for the assessment of areal bone mineral density (BMD) greatly benefited the field of bone imaging and the ability to diagnose and monitor osteoporosis. The additional capability of DXA to differentiate between bone mineral, fat tissue, and lean tissue has contributed to its emergence as a popular tool to assess body composition. Throughout the past 2 decades, technological advancements such as the transition from the original pencil‐beam densitometers to the most recent narrow fan‐beam densitometers have allowed for faster scan times and better resolution. The majority of reports that have compared DXA‐derived body composition measurements to the gold standard method of body composition appraisal, the four‐compartment model, have observed significant differences with this criterion method; however, the extent to which the technological advancements of the DXA have impacted its ability to accurately assess body composition remains unclear. Thus, this paper reviews the evidence regarding the trueness and precision of DXA body composition measurements from the pencil‐beam to the narrow fan‐beam densitometers.

This study suggests that approximately half of exercising women experience subtle menstrual disturbances, i.e. LPD and anovulation, and that one third of exercising women may be amenorrheic. Estimates of the prevalence of subtle menstrual disturbances in exercising women determined by the presence or absence of short or long cycles does not identify these disturbances. In light of known clinical consequences of menstrual disturbances, these findings underscore the lack of reliability of normal menstrual intervals and self report to infer menstrual status.

A high drive-for-thinness (DT) score obtained from the Eating Disorder Inventory-2 is associated with surrogate markers of energy deficiency in exercising women. The purposes of this study were to confirm the association between DT and energy deficiency in a larger population of exercising women that was previously published and to compare the distribution of menstrual status in exercising women when categorized as high vs. normal DT. A high DT was defined as a score ≥7, corresponding to the 75th percentile for college-age women. Exercising women age 22.9 ± 4.3 yr with a BMI of 21.2 ± 2.2 kg/m2 were retrospectively grouped as high DT (n = 27) or normal DT (n = 90) to compare psychometric, energetic, and reproductive characteristics. Chi-square analyses were performed to compare the distribution of menstrual disturbances between groups. Measures of resting energy expenditure (REE) (4,949 ± 494 kJ/day vs. 5,406 ± 560 kJ/day, p < .001) and adjusted REE (123 ± 16 kJ/LBM vs. 130 ± 9 kJ/LBM, p = .027) were suppressed in exercising women with high DT vs. normal DT, respectively. Ratio of measured REE to predicted REE (pREE) in the high-DT group was 0.85 ± 0.10, meeting the authors' operational definition for an energy deficiency (REE:pREE <0.90). A greater prevalence of severe menstrual disturbances such as amenorrhea and oligomenorrhea was observed in the high-DT group (χ2 = 9.3, p = .003) than in the normal-DT group. The current study confirms the association between a high DT score and energy deficiency in exercising women and demonstrates a greater prevalence of severe menstrual disturbances in exercising women with high DT.

No abstract

Percentage body fat (%BF) (21.0 ± 1.0 versus 26.8 ± 0.7%; P < 0.001) and leptin concentration (4.8 ± 0.8 versus 9.6 ± 0.9 ng/ml; P < 0.001) were lower in the exercising women with amenorrhea (ExAmen; n = 24) compared with the exercising ovulatory women (ExOvul; n = 26). However, the ranges in leptin were similar for each group (ExAmen: 0.30-16.98 ng/ml; ExOvul: 2.57-18.28 ng/ml), and after adjusting for adiposity the difference in leptin concentration was no longer significant. Significant predictors of log leptin in ExAmen included %BF (β = 0.826, P < 0.001), log insulin (β = 0.308, P = 0.012) and log glycerol (β = 0.258, P = 0.030), but in ExOvul only %BF predicted leptin. CONCLUSIONS These data suggest that leptin concentrations per se are not associated with FHA in exercising women, but the modulation of leptin concentrations may differ depending on reproductive status.