Sarcopenia is an important complication in patients with rheumatoid arthritis (RA) that severely impairs quality of life and prognosis, and there is an urgent need for simple and effective screening tools. This study aims to evaluate the association between triglyceride-glucose body mass index (TyG-BMI) and sarcopenia in RA patients.

A case-control study design was adopted. A total of 478 RA patients admitted to Department of Rheumatology and Immunology of the Affiliated Hospital of Shandong University of Traditional Chinese Medicine from May 2025 to January 2026 were enrolled, and based on the presence or absence of sarcopenia, they were assigned into a sarcopenia group (n=77) and a non-sarcopenia group (n=401). To balance baseline confounding factors, propensity score matching (PSM) was performed at a ratio of 1∶1, with matching variables including age, sex, C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), Disease Activity Score-28 for Rheumatoid Arthritis with ESR (DAS28-ESR), and creatinine. After matching, conditional logistic regression models were used to analyze the association between TyG-BMI and sarcopenia. Least absolute shrinkage and selection operator (LASSO) regression was employed to screen baseline candidate factors significantly associated with sarcopenia, followed by multivariate logistic regression analysis. The discriminative ability of the model was evaluated using receiver operating characteristic (ROC) curve and area under the curve (AUC).

After PSM, 76 pairs of patients were successfully matched, with balanced baseline characteristics. Multivariate logistic regression analysis further confirmed that, after adjusting for age, sex, CRP, renal function, and DAS28-ESR, TyG-BMI (as a continuous variable) remained independently and negatively associated with sarcopenia risk (OR=0.225, 95%CI: 0.153 to 0.332, P<0.001). When TyG-BMI was categorized by quartiles, sarcopenia risk showed a dose-dependent decrease with increasing quartiles. Multivariate association model analysis indicated that adding TyG-BMI to baseline variables significantly improved the efficacy of identifying sarcopenia (AUC=0.820), with good model calibration. Subgroup analyses demonstrated that the negative association between TyG-BMI and sarcopenia remained consistent across different age groups and in individuals with BMI <24 kg/m² but was attenuated in those with BMI ≥24 kg/m².

TyG-BMI is an independent protective factor for sarcopenia in RA patients, and its negative association with sarcopenia risk is robust. TyG-BMI enhances the discriminative performance of multivariate association models for sarcopenia and may serve as a promising early biomarker for identifying sarcopenia in RA.