Metabolic phenotypes modify the association between composite risk indices and chronic kidney disease: findings from a national survey

Technical Summary: Metabolic Phenotypes Modify the Association Between Composite Risk Indices and Chronic Kidney Disease

Problem Statement
Chronic kidney disease (CKD) is a major contributor to global morbidity and mortality, often driven by the complex interplay of obesity, diabetes, and cardiovascular dysfunction, collectively termed the cardiovascular-kidney-metabolic (CKM) syndrome. While composite metabolic risk indices—such as the triglyceride-glucose index (TyG) and its derivatives (TyG-WHtR, CHG, CHG-WHtR, TyHGB)—have been proposed as predictors for CKD, their utility across distinct metabolic phenotypes remains unclear. Previous studies have largely evaluated these indices in heterogeneous populations, implicitly assuming consistent performance across the spectrum of metabolic health. This study addresses the gap in understanding whether these indices provide meaningful discrimination within clinically relevant metabolic phenotypes, ranging from normal metabolism to severe metabolic dysfunction, or if they primarily distinguish between broad risk groups.

Methodology
The study utilized cross-sectional data from the 2021 Iran STEPS survey, a nationally representative sample of Iranian adults. The final analysis included 17,814 participants with complete data on anthropometrics, blood pressure, fasting plasma glucose, lipid profiles, and renal function markers (eGFR and UACR).

• Metabolic Phenotyping: Participants were stratified into four metabolic phenotypes based on the presence of metabolic syndrome (MetS) components and the status of glucose and blood pressure regulation:
  1. Metabolically Healthy: <3 MetS components, no hypertension or diabetes.
  2. Partial Metabolic Impairment: 3 MetS components OR diabetes OR hypertension.
  3. Class I Phenotype: ≥3 MetS components AND either impaired glucose OR blood pressure regulation.
  4. Class II Phenotype: ≥3 MetS components AND both impaired glucose AND blood pressure regulation.
• Indices Evaluated: The study assessed TyG, TyG-WHtR, CHG, CHG-WHtR, and TyHGB.
• Statistical Analysis: Associations were examined using logistic regression (adjusted for age, sex, residence, smoking, physical activity, cardiovascular history, statin use, wealth, and sodium intake). Dose-response relationships were analyzed using restricted cubic splines (RCS). Discriminative ability was quantified via covariate-adjusted Area Under the Receiver Operating Characteristic Curve (AUC).

Key Results

• General Population: All evaluated indices showed strong, non-linear associations with CKD in the general population (all p-values < 0.001).
• Phenotype-Specific Associations:
  • High-Risk Phenotypes (Class I & II): The associations between the indices and CKD remained robust in individuals with established metabolic dysfunction (Class I and Class II).
  • Normal Metabolism: No consistent associations or independent discriminative ability were observed among individuals with a normal metabolic profile. Notably, the base TyG index showed a significant inverse association with CKD in the normal metabolism group.
• Discriminatory Power:
  • In the general population, indices showed modest discriminatory ability (AUCs ranging from 0.609 to 0.663).
  • Among high-risk individuals (Class II), indices maintained independent discriminatory power (AUCs 0.560–0.592).
  • Among individuals with normal metabolism, AUCs included 0.5, indicating no predictive value.
  • For the intermediate risk group (Class I), only TyG-WHtR and CHG-WHtR demonstrated minimal independent discriminatory power.
• Sex Interactions: Female participants with higher values for TyG-WHtR and CHG-WHtR were more likely to have CKD compared to their male counterparts.

Key Contributions

1. Phenotype-Dependent Utility: The study demonstrates that the utility of composite cardiometabolic risk indices is heavily dependent on the baseline metabolic phenotype. These indices effectively stratify risk within groups of individuals who already exhibit metabolic dysfunction but fail to identify risk in metabolically healthy individuals.
2. Superiority of Anthropometric Derivatives: Indices incorporating waist-to-height ratio (TyG-WHtR and CHG-WHtR) provided additional discriminative power for individuals with intermediate metabolic risk compared to their base counterparts, suggesting that quantifying visceral adiposity adds value in early-stage CKM dysfunction.
3. Refutation of Universal Applicability: The findings challenge the assumption that these indices serve as universal early markers for CKD, highlighting that their predictive value emerges primarily after the onset of clinical metabolic dysfunction.

Significance and Claims
The authors conclude that while composite risk indices are strongly associated with CKD in the general population and high-risk subgroups, they lack independent predictive capacity in metabolically healthy individuals. The study posits that these indices are better suited for detailed risk stratification among individuals with established cardiometabolic risk rather than as early screening tools for the general population.

The paper emphasizes that the lack of association in "normal" phenotypes may be due to the absence of a link between metabolic markers and CKD independent of the metabolic profile, or attenuated associations prior to clinical dysfunction. Consequently, the authors suggest that future research should focus on longitudinal risk stratification in individuals with early-stage metabolic dysfunction and explore indices capable of detecting residual metabolic risk in those with normal phenotypes, potentially focusing on underrepresented domains like blood pressure dysregulation. The study explicitly states that its cross-sectional design precludes causal inference and positions the findings as a preliminary examination to guide future clinical research rather than a definitive assessment of clinical utility.