Statins and longitudinal changes in vertebral bone mineral density in an emulated target trial: the multi-ethnic study of atherosclerosis.
prospective_cohort · Level II
Where this comes from
- Record sourced from PubMed, PMID 40992679.
- Also identified by DOI 10.1016/j.bone.2025.117659.
- No licence information is recorded for this record.
- Because redistribution is not established, this page shows the abstract only. Follow the links below for the full text.
Abstract
Existing reports on the protective effects of statins on bone health have been conflicting, which may have been related to the diverse selectivity of studied populations. We aimed to evaluate the association between statin initiation and longitudinal changes in vertebral bone mineral density (BMD) over six years. This emulated target trial used data from the Multi-Ethnic Study of Atherosclerosis (MESA). Participants with a baseline 10-year atherosclerotic cardiovascular disease (ASCVD) risk ≥7.5 % at MESA Exam 4 or 5 and no history of prevalent statin use were included. Statin initiators were propensity score matched to non-initiators. The primary outcome was longitudinal change in CT-derived multilevel vertebral (T1-T10) BMD, assessed using linear mixed-effects models. Prespecified interaction terms and subsequent subgroup analyses were performed by age, sex, race/ethnicity, body mass index, high-sensitivity C-reactive protein (hs-CRP), physical activity, diabetes status, smoking, and alcohol use. We used an array approach to evaluate the magnitude and prevalence of residual confounding necessary to fully explain our main observation. A total of 384 participants (173 statin initiators; 211 non-initiators) were included in the per-protocol analysis. Statin initiation was associated with a slower decline in vertebral BMD (estimated difference, β = 1.00 g/cm<sup>3</sup>/year; 95 % CI, 0.65-1.35; p < 0.001), corresponding to a 0.47 % slower annual decline. An unmeasured confounder associated with a 1 % annual relative difference in BMD change would need to be at least three times as prevalent among statin initiators compared with non-initiators to nullify the observed effect size. Subgroup analyses suggested stronger protective associations among participants with impaired fasting glucose or diabetes, lower hs-CRP, and higher physical activity. In this emulated target trial, statin initiation was associated with beneficial bone health effects.
Medical subject headings
- Atherosclerosis
- Bone Density
- Ethnicity
- Hydroxymethylglutaryl-CoA Reductase Inhibitors
- Spine