Abstract
Background: Hip or knee osteoarthritis (OA) frequently culminates in joint arthroplasty, yet reliable biomarkers to predict this progression are lacking. We aimed to investigate whether baseline levels of seven serum lipid metabolites are associated with the risk and timing of subsequent joint replacement among participants diagnosed with hip or knee OA. Methods: Leveraging prospective data from the UK Biobank, we identified 56 978 participants newly diagnosed with hip or knee OA after the baseline measurement of seven serum lipid markers. We applied Fine & Gray competing risk models to estimate the impact of lipid indices on time to arthroplasty, complemented by accelerated failure time (AFT) modeling, threshold-effect analysis, subgroup stratification, and polygenic risk score (PRS) assessment to comprehensively characterize associations between lipid metabolites and OA progression. Results: Both Fine & Gray competing risk and AFT analyses consistently demonstrated that elevated total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), apolipoprotein A1 (ApoA1), and apolipoprotein B (ApoB) levels were significantly associated with increased risk of joint replacement and accelerated disease course ( P < 0.001). Lipoprotein(a) and triglycerides (TG) showed no significant relationship. Threshold analyses revealed inflection points beyond which risk increases plateaued. Findings remained robust in multiple-imputation and alternative model sensitivity analyses. Subgroup analyses indicated that participants younger than 60 years, those with hip site osteoarthritis, and especially obese women with high genetic risk exhibited heightened sensitivity to lipid fluctuations, underscoring the need for tailored monitoring and intervention. Conclusion: Elevated TC, HDL-C, LDL-C, ApoA1, and ApoB levels independently predict progression to arthroplasty and shorten time to surgery, providing an evidence base for lipid-targeted strategies in OA management.