Abstract
This large-scale study utilizing the US National Health and Nutrition Examination Survey (NHANES, 1999-2018; n = 14,175) compared the diagnostic and prognostic utility of two novel obesity indices-A Body Shape Index (ABSI), and Triglyceride Glucose-A Body Shape Index (TyG-ABSI), evaluating their stage-specific utility for CKD risk identification and mortality prediction within the same cohort. TyG-ABSI demonstrated superior performance for identifying prevalent CKD, with the highest quartile (Q4) showing a significantly increased risk after full adjustment (OR = 1.58, 95% CI:1.23-2.03). Conversely, ABSI emerged as a stronger, nonlinear predictor of mortality in established CKD patients. Restricted cubic splines identified significant ABSI z-score thresholds: all-cause mortality risk increased steeply above z = 0.624 (HR = 1.44, 95% CI:1.17-1.77), and cardiovascular disease (CVD) mortality above z = 0.416 (HR = 1.35, 95% CI:1.09-1.83). After full adjustment, ABSI Q4 remained significantly associated with both all-cause (HR = 1.50) and CVD mortality (HR = 2.40), while TyG-ABSI associations attenuated. ABSI demonstrated superior discriminative accuracy for both all-cause mortality (area under the curve [AUC] = 0.68 vs. TyG-ABSI AUC = 0.66) and CVD mortality (AUC = 0.64 vs. TyG-ABSI AUC = 0.59). Subgroup analyses confirmed the robustness of ABSI for mortality prediction. These findings suggest a shift in the primary driver of adverse outcomes as CKD progresses: TyG-ABSI, capturing metabolic dysregulation, is optimal for identifying CKD risk, whereas ABSI, reflecting anatomical fat distribution, becomes the superior predictor of mortality in established CKD. This differential utility supports stage-specific risk stratification strategies and highlights the broader clinical relevance of obesity phenotyping for cardiovascular risk assessment, metabolic disease screening, and geriatric prognosis.