Cystatin C is associated with osteoporosis and fractures: An observational study based on Mendelian randomization analysis

Abstract

Objectives: Osteoporosis is characterized by decreased bone mass and damaged bone microstructure, often leading to fragility fractures. Low bone mineral density is a key risk factor for fractures. Serum cystatin C (CysC), an endogenous marker of glomerular filtration rate, is negatively correlated with bone mineral density and may be a potential risk factor for osteoporosis. This study aims to investigate the association and potential pathogenic mechanisms between CysC and osteoporosis and fractures in the general population by combining cohort analysis and Mendelian randomization (MR) analysis.

Methods: Large-scale prospective cohort data from the UK Biobank and summary statistics from genome-wide association study (GWAS) in European populations were utilized, with strict exclusion criteria applied (excluding non-white individuals, those with thyroid diseases, gastrointestinal dysfunction, kidney diseases, rheumatoid diseases, malignant tumors, chronic infections or inflammatory diseases, diabetes, hypertension, and individuals taking medications that affect bone metabolism). Multivariable linear regression, logistic regression, and Cox proportional hazards models were used to analyze the relationship between CysC and bone mineral density, osteoporosis, and fracture risk. All analyses were performed using three sequential models to adjust for confounding factors: Model 1 adjusted for demographic characteristics and lifestyle factors; Model 2 further adjusted for renal function based on Model 1; and Model 3 further adjusted for physical activity based on Model 2. Restricted cubic spline models were used to explore non-linear relationships, and MR analysis was conducted to assess the causal associations between CysC and osteoporosis and fractures.

Results: Multivariate analysis showed that after adjusting for basic variables (Model 1), there was no correlation between CysC and estimated bone mineral density (eBMD) in the overall study population; however, when stratified by gender, both males and females exhibited a significant negative correlation (P<0.001). After further adjustment for renal function (Model 2) and physical activity level (Model 3), CysC became negatively correlated with eBMD in the overall population (P<0.001). Moreover, multivariable logistic regression consistently demonstrated that CysC concentration was significantly positively associated with osteoporosis risk (P<0.01), and this association remained stable across all models. In all populations and models, multivariate Cox regression analysis indicated that subjects in the highest quartile (Q4) of CysC had a significantly increased risk of developing osteoporosis (P<0.001). In the overall population, the positive association between Q4 CysC levels and fractures was observed only in Models 2 and 3, with a hazard ratio of 1.118 (both P<0.001); however, after gender stratification, this association disappeared in males (P>0.05). Additionally, restricted cubic spline regression analyses revealed a significant non-linear relationship between CysC and the incidence of osteoporosis and fractures (P<0.05). MR analysis, using 167 single nucleotide polymorphisms (SNPs) as instrumental variables, showed no direct causal relationship between CysC and osteoporosis or fractures (P≥0.05), a finding that differs from previous studies in special populations.

Conclusions: Elevated levels of CysC are significantly associated with an increased risk of osteoporosis and fractures, and this association is more pronounced in females. Renal function and physical activity levels may be important factors influencing this relationship. The link between CysC and osteoporosis and fractures may be mediated by several mechanisms: Eelevated CysC may lead to abnormalities in vitamin D and mineral metabolism, thereby inhibiting bone formation; renal dysfunction may exacerbate inflammation, affecting bone resorption; or in the osteoporosis state, increased osteoclast differentiation may result in elevated CysC levels. These findings support the potential use of CysC as a biomarker for predicting the risk of osteoporosis.

Keywords: cystatin C; mendelian randomization analysis; osteoporosis, fractures; prospective cohort study.

图1

研究设计流程图 Figure 1 Study design flowchart UK Biobank: United Kingdom Biobank; SNPs: Single nucleotide polymorphisms; GWAS: Genome-wide association study; IEU-openGWAS: MRC integrative epidemiology unit open GWAS database; FinnGen: Finnish genomics biobank; MR: Mendelian randomization.

图2

CysC对骨质疏松症和骨折的发生风险具有非线性影响 Figure 2 Non-linear relationship between cystatin C and the incidence of fracture and osteoporosis A: Association between CysC and risk of osteoporosis in the total population (P _non-linear=0.007); B: Association between CysC and risk of osteoporosis in males (P _non-linear<0.001); C: Association between CysC and risk of osteoporosis in females (P _non-linear=0.542); D: Association between CysC and risk of fracture in the total population (P _non-linear=0.039); E: Association between CysC and risk of fracture in males (P _non-linear=0.634); F: Association between CysC and risk of fracture in females (P _non-linear=0.929). The red solid line represents the adjusted HR, and the light red area represents the 95% confidence interval. The horizontal dashed line indicates the reference value (HR=1). CysC: Cystatin C; HR: Hazard ratio; CI: Confidence interval.