Author(s): Pinheiro MM, Castro CM, Szejnfeld VL
Abstract Share this page
Abstract BACKGROUND: Prospective and cross-sectional studies have confirmed a significant association between bone mineral density (BMD) measurements and fracture risk. However, the relationship among incident fracture risk, mortality, BMD, and quantitative ultrasound is controversial and less studied. METHODS: At baseline, 275 postmenopausal elderly women were evaluated by clinical questionnaire regarding fracture risk factors and had radiological analysis of the spine, spine and femur dual energy x-ray absorptiometry, and calcaneous quantitative ultrasound measurements. Five years later, 42 (15.3\%) women had died, 25 (9.1\%) were lost to follow-up, and 208 (75.6\%) continued the study. Specific questionnaire items regarding fracture risk were reevaluated, and thoracic and lumbar spine x-rays were taken to identify new fractures. Causes of mortality in this population were also assessed. All reported deaths were confirmed by review of death certificates or hospital records and were classified according to International Classification of Diseases, 10th Revision (ICD-10) code. RESULTS: After adjustments for age, weight, body mass index, smoking status, previous fracture, physical activity, drug use, and presence of chronic diseases, each 1 standard deviation (SD) reduction in stiffness index (SI) at baseline was significantly associated with future fracture (hazard ratio [HR] = 2.23; 95\% confidence interval [CI], 1.30-3.83) and total mortality 5 years later (HR = 1.57; 95\% CI, 1.10-2.47). Femoral neck and trochanter BMD values at baseline were also related to new fracture (HR = 2.01; 95\% CI, 1.27-3.18 and HR = 1.62; 95\% CI, 1.08-2.42, respectively) and total mortality (HR = 1.44; 95\% CI, 1.06-2.22 and HR = 1.59; 95\% CI, 1.07-2.36, respectively). Cardiovascular mortality was associated with decreased baseline femur BMD (HR = 1.28; 95\% CI, 1.08-2.26) and lower SI values (HR = 1.54; 95\% CI, 1.08-2.79). CONCLUSIONS: Our results demonstrate that low femoral BMD and low SI are able to predict fracture risk and are related to non-cause-specific and cardiovascular mortality, independently of other factors associated with osteoporosis, death, or aging.
This article was published in J Gerontol A Biol Sci Med Sci
and referenced in Internal Medicine: Open Access