Novel obesity and metabolic indices better predict non-alcoholic fatty liver disease in elderly T2DM patients: evidence from cross-sectional and longitudinal analysis

Objective The associations between obesity- and metabolism-related indices and the risk of non-alcoholic fatty liver disease (NAFLD) in elderly patients with type 2 diabetes mellitus (T2DM) remain unclear. This study aimed to investigate these associations and assess their predictive value for NAFLD in this high-risk population. Methods A total of 789 elderly T2DM patients recruited between 2020 and 2022 were included in the cross-sectional analysis, and 382 patients without NAFLD were followed in the longitudinal cohort for a median of 25.37 months. Binary logistic regression and Cox models were used to assess associations between obesity- and metabolism-related indices and NAFLD risk. Kaplan–Meier curves, restricted cubic spline (RCS) models, subgroup analysis, and receiver operating characteristic (ROC) curves were applied to explore these relationships further. Results In the cross-sectional analysis, all obesity- and metabolic-related indices were significantly and positively associated with NAFLD risk, with odds ratio (OR) ranging from 1.014 (95% CI: 1.010–1.018) for LAP to 3.288 (95% CI: 2.414–4.533) for WHtR. RCS analysis revealed significant nonlinear associations for LAP, MetS scores, VAI, CMI, METS-IR, and ABSI. In the cohort analysis, 67 participants developed NAFLD, with an incidence rate of 8.35 per 100 person-years. Baseline LAP (HR = 3.10, 95% CI: 1.48–6.51), and MetS scores (HR = 4.26, 95% CI: 1.99–9.11) were independently associated with increased risk of incident NAFLD. Subgroup analysis demonstrated consistent positive associations across most subgroups. Time-dependent ROC analysis showed that LAP had the highest AUCs at 24 months (AUC = 0.725). Conclusion The findings from cross-sectional and cohort studies collectively supported that MetS score and LAP may be the most effective predictive indicators for the risk of NAFLD among Chinese elderly T2DM Patients.