Bone density and hemoglobin levels in older persons: results from the InCHIANTI study

Abstract

Hypoxemia has been recognized as a risk factor for bone loss. The aim of the present study is to investigate the relationship of bone mass and density measures with anemia and hemoglobin levels in a large sample of older community-dwelling persons. The study is based on data from 950 participants enrolled in the "Invecchiare in Chianti" (Aging in the Chianti area, InCHIANTI) study. All the analyses were performed considering continuous hemoglobin levels as well as the dichotomous anemia variable (defined according to WHO criteria as hemoglobin < 12 g/dl in women and < 13 g/dl in men). A peripheral quantitative computerized tomography (pQCT) scan of the right calf was performed in all participants to evaluate total bone density, trabecular bone density, cortical bone density, and the ratio between cortical and total bone area. Linear regression analyses were used to assess the multivariate relationship of pQCT bone measures with anemia and hemoglobin levels after adjustment for demographics, chronic conditions, muscle strength and biological variables. Participants were 75.0 (SD 6.9) years old. In our sample, 101 participants (10.6%) were anemic. In women, coefficients from adjusted linear regression analyses evaluating the association between pQCT bone measures (per SD increase) and hemoglobin levels/anemia showed significant associations of anemia with total bone density (beta = -0.335, SE = 0.163; P = 0.04) and cortical bone density (beta = -0.428, SE = 0.160; P = 0.008). Relationships with borderline significance were found for the associations of anemia with trabecular bone density and the ratio between cortical and total bone area. Significant associations were found between hemoglobin levels and trabecular bone density (beta = 0.112, SE = 0.049; P = 0.02), total bone density (beta = 0.101, SE = 0.046; P = 0.03), cortical bone density (beta = 0.100, SE = 0.046; P = 0.03) and the ratio between cortical bone and total area (beta = 0.092, SE = 0.045; P = 0.04). In men, significant associations were found for hemoglobin levels with total bone density (beta = 0.076, SE = 0.036; P = 0.03) and cortical bone density (beta = 0.095, SE = 0.41; P = 0.02). A borderline significance was reported for the association between anemia and cortical bone density. We concluded that anemia and low hemoglobin levels are negatively and independently associated with bone mass and density. The bone loss associated with hemoglobin levels mainly occurs in the cortical bone. Women with lower hemoglobin levels demonstrate a higher bone loss than male counterparts.