Abstract
Serum hemoglobin plays an important role in bone metabolism. However, the association between serum hemoglobin levels and bone mineral density (BMD) remains unclear. Therefore, this study aimed to explore the relationship between serum hemoglobin levels and lumbar spine BMD in adults. We conducted a cross-sectional study by utilizing data from the National Health and Nutrition Examination Survey from 2011 to 2018. The serum hemoglobin level was examined as an independent variable, while the lumbar spine BMD was utilized as the dependent variable. Weighted multivariate linear regression models and stratified analysis by age, sex, and race/ethnicity were applied after controlling for confounding factors to assess the relationship between serum hemoglobin levels and the lumbar spine BMD. Additionally, smooth curve fitting and threshold effect analyses were utilized to depict the nonlinear relationship between the 2 variables. A total of 11,658 participants (6004 men and 5654 women) aged ≥ 18 years were included in this study. When the serum hemoglobin level was represented as a continuous variable and fully adjusted in the regression model, the hemoglobin level was significantly negatively correlated with the lumbar spine BMD (β = -0.0035, 95% confidence interval: -0.0065 to -0.0004, P = .024555); this significant negative correlation persisted when the serum hemoglobin level was transformed into a categorical variable, except in the Q2 group (β = -0.0046, 95% confidence interval: -0.0142 to -0.0050, P = .348413). When different confounding factors were used including sex, age, and race/ethnicity, the stratified subgroups exhibited a negative correlation between the serum hemoglobin level and the lumbar spine BMD. Additionally, smooth curve fitting and threshold effect analyses showed a negative correlation between the serum hemoglobin level and the lumbar spine BMD, with a saturation effect at 15 g/dL. Our findings demonstrated an association between hemoglobin levels and the lumbar spine BMD in adults, characterized by a nonlinear relationship. Thus, monitoring the serum hemoglobin level could aid in the early detection of risks associated with bone metabolic disorders such as osteoporosis.