Publications

This study examined whether hormonal and body composition responses to 10-weeks of U.S. Army Basic Combat Training (BCT) differ by baseline testosterone and sex. Methods: Male (n=97) and female (n=105) trainees were stratified at baseline as having LOW (men <10.4 nmol/L; women <0.82 nmol/L) or NORMAL baseline testosterone. We measured serum total and free testosterone, sex hormone binding globulin (SHBG), luteinizing hormone (LH) and estradiol (E2) at weeks 0, 2, 4, 6, 8, and 9 and cortisol weekly; dual-energy X-ray absorptiometry (DXA) assessed body composition at week 0 and 9. Linear mixed models tested time and group effects; generalized linear models related baseline testosterone status and testosterone area under the curve with changes in body composition. Results: In LOW women, testosterone increased at each timepoint (all p<0.01), reaching concentrations 53% higher at week 9 compared to week 0 (30.2± 7.8 vs 19.7± 3.2 ng/dl). In men, testosterone increased (LOW: +76%; NORMAL: +23%) across BCT (all p<0.05), except at week 2. Among LOW men, concentrations stabilized by week 4 (344.8± 81.4 ng/dl) and remained normal. Cortisol was highest at baseline and declined in all men and in NORMAL women. LH and E2 remained stable in both sexes, while SHBG declined only in women, independent of baseline testosterone. Neither baseline testosterone nor cumulative exposure were associated with changes in lean or fat mass. Conclusions: SHBG and cortisol trajectories differed by sex and baseline testosterone status. Testosterone changes were sex specific, reflecting an anabolic training environment in men and modest increased androgen availability in LOW women.

Purpose: Anorexia nervosa (AN) is associated with significant changes in bone density and microstructure. However, the extent and nature of skeletal recovery following weight restoration is unclear. The aim of this study was to investigate these changes in bone during simulated AN in rats across short- and long-term recovery. Methods: Female Sprague Dawley rats were exposed to 50-60% food restriction for 30 days to simulate AN, followed by recovery periods of 5, 15, and 30 days. Bone parameters were assessed longitudinally using peripheral quantitative computed tomography (pQCT) and micro-computed tomography (µCT). Bodyweight and muscle strength were correlated with bone parameters. Results: AN resulted in significant reductions in total and trabecular bone density, as well as impairments in cortical and trabecular microarchitecture. Even with weight restoration, these deficits persisted and the compensatory responses were not sufficient for bone recovery. Bodyweight showed an association with total and trabecular bone density. Measures of muscle strength also correlated with total and trabecular density and were additive in multiple regression analysis prediction of total and trabecular bone density when combined with bodyweight. Conclusion: Recovery from AN-induced bone disorders is incomplete even after weight restoration and likely requires targeted interventions beyond nutritional rehabilitation. Furthermore, muscle function is an important predictor of bone health when combined with bodyweight.

CONTEXT: Cardiometabolic disease (CMD) may increase vertebral fracture risk through deterioration in bone structure and/or tissue properties, compromising bone strength.

OBJECTIVE: Evaluate the relationship between CMD and vertebral bone measures.

DESIGN: Cross-sectional.

PARTICIPANTS: Coronary Artery Risk Development in Young Adults (CARDIA) study participants with year 25 (2010-2011) abdominal computed tomography (CT) scans (n = 2728).

EXPOSURE: CMD (abdominal obesity, diabetes mellitus, dyslipidemia, or hypertension) status (≥1 vs 0) and burden (0, 1, 2, ≥3).

OUTCOME MEASURES: Vertebral strength, volumetric bone mineral density (vBMD), and bone mass, obtained via biomechanical CT analysis.

METHODS: Multivariable linear regression evaluated the association between CMD and vertebral bone measures. Sex-specific models were adjusted for age, race, height, weight, and additional social determinants of health, lifestyle factors, and comorbidities. Effect modification by race, age, and menopausal status was evaluated.

RESULTS: Overall, at least 1 CMD was identified in 72.3% of men and 70.5% of women. In the fully adjusted model, CMD status was significantly associated with lower total (β = -6.32 mg/cm3; P = .018), trabecular (β = -4.53 mg/cm3; P = .048), and cortical (β = -6.63 mg/cm3; P = .034) vBMD in men. Greater CMD burden was associated with lower total and trabecular vBMD. The negative association between CMD and vBMD was stronger in men ≤50 years. There was no association in women.

CONCLUSION: In this biracial cohort of Black and White adults from the United States, CMD was associated with significantly lower total and subcompartment vertebral vBMD in men.