Hormone replacement therapy and muscle loss: repeated measures analysis from the Baltimore Longitudinal Study of Aging using DXA.

Abstract

To investigate the association between hormone replacement therapy (HRT) and change in body composition in post-menopausal women in the Baltimore Longitudinal Study of Aging (BLSA). We included postmenopausal women with repeated dual-energy X-ray absorptiometry measurements and documented HRT status. HRT users were propensity score-matched to non-users based on baseline characteristics including age, height, weight, race, smoking, alcohol use, physical activity, education, age at menopause, age at first pregnancy, oral contraceptive use, and oophorectomy. Linear mixed-effects models assessed associations between current HRT use and longitudinal changes in fat and lean mass. Age was modeled as a time-varying continuous variable using piecewise linear splines at 60, 70, and 80 years. Interaction terms between HRT use and age were included to evaluate differences in age-related trajectories. HRT users (n = 91) and matched non-users (n = 220) had similar baseline characteristics. No significant interactions were found between HRT use and age for lean mass (all p > 0.05), and the likelihood ratio test (LRT) showed no improvement in model fit with interaction terms (p = 0.20). In contrast, fat mass trajectories differed by HRT status. Before age 60, HRT users experienced significantly greater fat mass gain (β = 596 g/year; 95% CI 75, 1117). After age 60, this reversed, with HRT users gaining 869 g/year less fat than non-users (β = -869 g/year; 95% CI - 1583, - 155), indicating a protective association of HRT after age 60. No significant differences were observed at ages 70 or 80. The LRT supported including interaction terms (p = 0.04). HRT use is associated with less fat accumulation beyond the age of 60, although this association attenuates with advancing age. No differential association with lean mass was observed.