The triglyceride-glycated hemoglobin index as a predictor of cardiometabolic multimorbidity: evidence from a Chinese nationwide cohort

Technical Summary: The Triglyceride-Glycated Hemoglobin Index as a Predictor of Cardiometabolic Multimorbidity

Problem Statement
Cardiometabolic multimorbidity (CMM), defined as the concurrent presence of two or more conditions among diabetes, heart disease, and stroke, represents a significant global health burden associated with increased mortality and reduced life expectancy. While insulin resistance (IR) is a central pathophysiological driver of CMM, current risk stratification tools have limitations. The triglyceride-glucose (TyG) index, a widely used surrogate for IR calculated from fasting blood glucose (FBG) and triglycerides (TG), is susceptible to short-term fluctuations in FBG, which may not accurately reflect long-term glycemic control. Although glycated hemoglobin (HbA1c) offers a more stable measure of average glucose over 2–3 months, the utility of a novel index combining TG and HbA1c—the triglyceride-glycated hemoglobin (TyGH) index—for predicting CMM risk remains unexplored. This study aimed to evaluate the association between the TyGH index and incident CMM and compare its clinical utility against the established TyG index.

Methodology

• Data Source: The study utilized data from the China Health and Retirement Longitudinal Study (CHARLS), a nationally representative prospective cohort.
• Study Population: From an initial pool, 6,822 participants aged 45 years or older were included after excluding those with baseline CMM, those under 45, individuals without overnight fasting, and those missing key data (TG, FBG, or HbA1c) or follow-up data.
• Exposure Variables: Baseline measurements were used to calculate the TyG index [Ln(TG × FBG/2)] and the TyGH index [Ln(HbA1c × TG/2)].
• Outcome Definition: The primary outcome was incident CMM, defined as the new onset of a second distinct cardiometabolic condition (diabetes, heart disease, or stroke) during the follow-up period (2011–2020).
• Statistical Analysis:
  • Cox Regression: Multivariable Cox proportional hazards models were employed to estimate hazard ratios (HRs) for CMM risk, adjusting for demographics, lifestyle factors, and baseline comorbidities.
  • Non-linearity: Restricted cubic spline (RCS) models were used to characterize the dose-response relationships.
  • Predictive Performance: Receiver operating characteristic (ROC) curves and Area Under the Curve (AUC) values were calculated at 2, 4, 7, and 9 years.
  • Clinical Utility: Decision curve analysis (DCA) was performed to assess the net benefit of using these indices for clinical decision-making.
  • Subgroup Analysis: Interactions were tested across strata including age, sex, BMI, hypertension, and dyslipidemia.
  • Missing Data: Multiple imputation via chained equations was used to handle missing covariate data.

Key Results

• Incidence: Over a median follow-up of 8.43 years, 1,081 participants (15.85%) developed CMM.
• Risk Association: Both indices were independent predictors of CMM risk after full adjustment.
  • TyG Index: Each one-unit increment was associated with a 40% increased risk (HR 1.40, 95% CI 1.29–1.52, p < 0.001).
  • TyGH Index: Each one-unit increment was associated with a 43% increased risk (HR 1.43, 95% CI 1.30–1.58, p < 0.001).
• Dose-Response Patterns:
  • The TyG-CMM relationship exhibited a non-linear pattern in unadjusted and partially adjusted models, though it appeared linear in the fully adjusted model.
  • The TyGH-CMM association demonstrated a linear relationship across all adjusted models.
• Predictive Accuracy:
  • ROC analysis showed comparable AUC values for both indices at various time points (e.g., at 9 years: TyG AUC 0.621 vs. TyGH AUC 0.615). While some differences reached statistical significance (e.g., at 4 years, p=0.049), the absolute differences were marginal.
  • Decision Curve Analysis: The DCA curves for TyG and TyGH were nearly superimposed, indicating that both indices offer broadly similar clinical net benefits across a range of threshold probabilities.
• Subgroup Consistency: No significant interactions were observed in subgroup analyses (all p for interaction > 0.05), suggesting the predictive value of the TyGH index is consistent across different demographic and clinical strata.

Significance and Claims
The authors claim that the TyGH index is a practical and effective alternative to the TyG index for CMM risk stratification. The study establishes that:

1. Equivalence in Utility: Despite the TyGH index incorporating HbA1c, its predictive performance and clinical net benefit are comparable to the widely used TyG index.
2. Operational Advantage: The TyGH index offers a distinct operational advantage due to the stability of HbA1c. Unlike FBG, which fluctuates with recent diet and exercise, HbA1c provides a robust measure of chronic glycemic exposure, potentially offering more consistent dose-response relationships and reducing measurement variability caused by acute stress or illness.
3. Clinical Application: The TyGH index serves as a simple, low-cost tool derived from routine clinical measurements. It can be utilized to refine risk stratification and identify high-risk populations for early preventive interventions, particularly in settings where HbA1c stability is preferred over fasting glucose.

The paper concludes that while the TyGH index does not significantly outperform the TyG index in terms of statistical discrimination (AUC), its stability and linear association with risk make it a valuable, equivalent tool for assessing cardiometabolic multimorbidity risk in the Chinese population.