The Association between Bone Quality and Atherosclerosis: Results from Two Large Population-Based Studies

Abstract

Objective: It is highly debated whether associations between osteoporosis and atherosclerosis are independent of cardiovascular risk factors. We aimed to explore the associations between quantitative ultrasound (QUS) parameters at the heel with the carotid artery intima-media thickness (IMT), the presence of carotid artery plaques, and the ankle-brachial index (ABI).

Methods: The study population comprised 5680 men and women aged 20-93 years from two population-based cohort studies: Study of Health in Pomerania (SHIP) and SHIP-Trend. QUS measurements were performed at the heel. The extracranial carotid arteries were examined with B-mode ultrasonography. ABI was measured in a subgroup of 3853 participants. Analyses of variance and linear and logistic regression models were calculated and adjusted for major cardiovascular risk factors.

Results: Men but not women had significantly increased odds for carotid artery plaques with decreasing QUS parameters independent of diabetes mellitus, dyslipidemia, and hypertension. Beyond this, the QUS parameters were not significantly associated with IMT or ABI in fully adjusted models.

Conclusions: Our data argue against an independent role of bone metabolism in atherosclerotic changes in women. Yet, in men, associations with advanced atherosclerosis, exist. Thus, men presenting with clinical signs of osteoporosis may be at increased risk for atherosclerotic disease.

Figure 1

Adjusted mean number of plaque sites according to QUS-based osteoporotic fracture risk by sex. ANOVA was adjusted for age, body mass index, smoking status, physical inactivity, risky alcohol consumption, diabetes mellitus, dyslipidemia, hypertension, and, in women, additionally intake of estrogens (oral contraceptives or hormone replacement therapy) and years since menopause. The number of arterial segments affected by plaque was transformed (log (number of plaque sites + 1)) before being entered in the model and back-transformed for display in the figure.