Abstract
Background
The prevention of osteoarthritis through controlling body measurements has received increasing attention in recent years, but the relationship between lean body mass (LBM) and osteoarthritis remains unclear. Hence, we used large-scale data to explore this association.
Methods
To explore the correlation between LBM and osteoarthritis, we utilized logistic regression equations to explore associations between covariates, exposures, and outcomes. We used multivariate regression models to further eliminate confounding factors. Restricted cubic splines were employed to investigate non-linear relationships. And the inflection point was identified by recursive algorithm. Furthermore, stratified analyses of gender and age were conducted.
Results
Osteoarthritis was negatively correlated with LBM [0.74 (0.67, 0.80)] in the model adjusting for all confounders. A further linear trend test was performed, and the OR (95% CI) were 0.72 (0.61, 0.85), 0.54 (0.44, 0.65), and 0.45 (0.36, 0.56) for Q2, Q3, and Q4, respectively, compared with Q1. Based on the restricted cubic spline curve, an inflection point of 52.26 kg was found to confirm the non-linear relationship. LBM was negatively correlated with osteoarthritis at 0.66 (0.50, 0.86) before the inflection point, and the association was not statistically significant afterward.
Conclusion
This large-scale study revealed that LBM was non-linearly correlated with osteoarthritis in the general American population. Differences in age and gender were further identified.