Mediation effect of body mass index on the association between serum uric acid to high-density lipoprotein cholesterol ratio and osteoporosis: A cross-sectional study

Emerging evidence suggests that serum uric acid to high-density lipoprotein cholesterol ratio (UHR) is associated with bone loss. This study aims to explore the mediation effect of body mass index (BMI) underlying the relationship between UHR and osteoporosis risk in older adults, and compares the diagnostic performance of UHR and BMI for osteoporosis. This cross-sectional study analyzed data from 7388 adults aged 60 years and older from National Health and Nutrition Examination Survey (2005–2020). Multivariable linear regression models, multivariable logistic models, restricted cubic splines, subgroup analysis, threshold effect analysis, mediation analysis, receiver operating characteristic curve, and net reclassification improvement analysis were used. Results revealed a positive association between UHR and bone mineral density at the femoral neck, total femur, and lumbar spine. We found that UHR was negatively associated with osteoporosis prevalence (odds ratios (OR) = 0.734, 95% confidence interval (CI): 0.599–0.899, P = .005). In overall population, an L-shaped relationship between UHR and osteoporosis was identified, with a threshold of 0.096. Among those aged 60 to 74 years, UHR was consistently and negatively associated with osteoporosis prevalence (UHR < 0.096: OR = 0.305, 95% CI: 0.136–0.686, P = .005; UHR ≥ 0.096: OR = 0.498, 95% CI: 0.329–0.754, P = .002). BMI mediated the association between the UHR and osteoporosis, accounting for 58.4% of the mediation ( P < .001). BMI had a higher area under the curve value (0.695 vs 0.669, P = .012) and a greater net reclassification index (95% CI: 0.06–0.27, P = .009) compared to UHR. In conclusion, a higher UHR was associated with a lower prevalence of osteoporosis, particularly in adults aged 60 to 74 years. BMI mediated this relationship and demonstrated superior diagnostic performance compared to UHR, suggesting that BMI monitoring may be more effective for assessing osteoporosis risk. Although causality cannot be inferred due to the cross-sectional design, these findings provide valuable insights for future prospective research.