Diabetes complications

The cholesterol, high-density lipoprotein, and glucose index and its derived indices predict diabetic kidney disease development and progression: evidence from two nationwide cohorts and a biopsy-proven clinical cohort

Cardiovasc DiabetolPublished Prognostic cohort study

Performance · Incident DKD (CHARLS)Result — go to the resultHazard ratio 2.19

Authors' conclusion

The CHG index and CHG-BMI independently predict DKD development and progression, outperforming traditional insulin resistance markers.

Study context

Population
Diabetic patients without baseline renal injury (China Health and Retirement Longitudinal Study (CHARLS) 2011–2020); validation in NHANES 1999–2018; patients with biopsy-proven diabetic kidney disease (DKD)CHARLS
Participants
984 patients
Comparison
CHG index and derived indices (e.g. CHG-BMI) vs traditional insulin resistance surrogate markers for predicting DKD
Outcome
Incident DKD
Methodological details10-year longitudinal cohort (CHARLS), external validation in cross-sectional NHANES, independent biopsy-proven clinical cohort
  • 10-year longitudinal cohort (CHARLS)
  • External validation in cross-sectional NHANES
  • Independent biopsy-proven clinical cohort

Performance

Incident DKD (CHARLS)

Hazard ratio

2.19

P = 0.020

As reported in the abstract

Significant integrated discrimination improvement of CHG-BMI over traditional markers
Integrated discrimination improvement (IDI), P = 0.030

Further findings

Incident DKD by subgroup (CHARLS)

Higher risk estimates for CHG-BMI in non-overweight and normolipidemic individuals
P for interaction < 0.05

NHANES validation

Associations replicated; significant J-shaped trajectories for CHG and CHG-BMI

Reduced eGFR in biopsy-proven DKD cohort

CHG and CHG-BMI independently associated with reduced eGFR, unlike traditional markers
P < 0.05
Adjusted AUC for predicting DKD progression, CHG
0.733
Adjusted AUC for predicting DKD progression, CHG-BMI
0.730
Significant interactions between CHG-BMI and Renal Pathology Society (RPS) grades, with significant associations in classes IIa and IIb

E-value for unmeasured confounding, CHG-BMI (Q3)

E-value
4.81

Original publication

The cholesterol, high-density lipoprotein, and glucose index and its derived indices predict diabetic kidney disease development and progression: evidence from two nationwide cohorts and a biopsy-proven clinical cohort. Cardiovasc Diabetol. Published 13 April 2026. doi:10.1186/s12933-026-03167-3

Composed by AI from the published abstract and checked against it. How Briefs are made →

Get them every week

New studies on this topic in My Week and in your Monday e-mail.

We e-mail you a sign-in link and a code. No password, no spam.