B99-06 Computed Tomography (CT)-Based Muscle-Fat Ratio as Predictor of Lung Function Decline and Exacerbations After Pulmonary Tuberculosis Treatment

Rationale Changes in body composition, particularly skeletal muscle loss and visceral fat accumulation, are increasingly recognized as important contributors to chronic lung disease. However, their prognostic relevance among individuals recovering from tuberculosis (TB) remains unclear. This study examined whether chest computed tomography (CT)–derived indices of muscle and fat distribution are associated with subsequent lung function decline and acute exacerbations (AEs) following TB treatment. Methods We retrospectively screened 3,744 patients treated for pulmonary TB between 2015 and 2023. 1,961 patients completed treatment, and after exclusion of individuals lacking both pre- and post-treatment spirometry and CT imaging, 202 patients were included in the final analysis. The pectoralis muscle area (PMA) and visceral fat area (VFA) were quantified from standardized axial CT slices at the levels of the aortic arch and T12, respectively. PMA values were translated into fat-free mass (FFM) using a validated, body size–adjusted equation. Longitudinal change was expressed as the ratio of post- to baseline FFM divided by the corresponding ratio for VFA, yielding the FFM-to-VFA ratio as an index of relative muscle preservation versus visceral fat accumulation. Associations between this ratio and longitudinal changes in post-bronchodilator FEV1 and FEV1/FVC, as well as the frequency of moderate and severe AEs, were evaluated. Results In the overall cohort, higher FFM-to-VFA ratios were associated with a slower decline in FEV1/FVC (ρ = 0.19, p = 0.027). Among non-current smokers, the ratio showed stronger correlations with annual changes in both FEV1/FVC (ρ = 0.52, p < 0.001) and FEV1 (ρ = 0.35, p = 0.025). In multivariable regression, the FFM-to-VFA ratio independently predicted FEV1/FVC decline (β = 2.10, p = 0.031), and linear mixed-effects modeling demonstrated more favorable FEV1/FVC trajectories in those with higher ratios (β = 5.13±2.37, p = 0.03). In 68 patients with COPD, after propensity score matching for age, sex, smoking quantity, and percent predicted FEV1, the high-ratio group (≥1.11) experienced fewer moderate AEs (incidence rate ratio [IRR] 0.26, 95% CI 0.07–0.94, p = 0.041). Kaplan–Meier analysis also demonstrated superior AE-free survival in the high-ratio group (p = 0.031). Conclusion Greater preservation of muscle relative to visceral fat after TB treatment is associated with slower lung function decline, particularly in non-current smokers, with reduced AE risk in patients with COPD. Simple, routinely obtainable CT-derived metrics may improve post-TB lung disease risk stratification and monitoring while highlighting modifiable targets for nutritional and rehabilitative interventions. This abstract is funded by: None