Plasma ceramide Cer24:0 and insulin resistance: associations with TyG and TG/HDL-C in a multicenter study of coronary artery disease cohorts.

Plasma ceramide Cer24:0 showed a direct conditional association with the triglyceride-glucose index (TyG) independent of other variables in the network.

• Mixed graphical model (MGM) analysis was used to estimate conditional associations within a multivariable network.
• The partial correlation between Cer24:0 and TyG was r=0.23 (95% CI, 0.17–0.29).
• This association was independent of other ceramide species (Cer16:0, Cer18:0, Cer24:1) and other variables included in the network.
• The study enrolled n=987 adults with established coronary artery disease (CAD) across multiple centers.

Per 1-unit increase in ln(Cer24:0) was associated with a significantly higher TyG index in double machine learning analyses.

• Double machine learning (DML) with causal-forest estimators was used to provide covariate-adjusted association estimates.
• The estimated association was 0.459 (95% CI, 0.252–0.665; P = 0.001) per 1-unit increase in ln(Cer24:0).
• Exposures were standardized per standard deviation.
• These estimates demonstrated moderate robustness to unmeasured confounding, with a robustness value (RV) theta of 0.223.

A ceramide-augmented model classified clinical insulin resistance (IR) with an ROC-AUC of 0.770.

• Clinical IR was prespecified as TyG ≥ 9.
• The ceramide-augmented model achieved an area under the receiver operating characteristic curve (ROC-AUC) of 0.770 (95% CI, 0.741–0.799).
• The model incorporated plasma ceramide species alongside conventional clinical variables.
• This finding suggests ceramide profiling may add discriminative value beyond conventional IR indices.

Plasma ceramides were correlated with multiple surrogate insulin resistance indices in a cohort of CAD patients.

• Four ceramide species were quantified: Cer16:0, Cer18:0, Cer24:0, and Cer24:1.
• Three surrogate IR indices were assessed: TyG, metabolic score for insulin resistance (METS-IR), and triglyceride-to-high-density lipoprotein cholesterol ratio (TG/HDL-C).
• The study was a prospective, multicenter observational cohort with n=987 participants with established CAD.
• Ceramide associations were observed across lipid-centric IR metrics (TyG and TG/HDL-C).

The association between Cer24:0 and TyG was consistent across multiple complementary analytic frameworks.

• Both MGM network analysis and DML with causal-forest estimators were applied to assess the Cer24:0–TyG association.
• Findings were consistent across these distinct methodological approaches.
• The robustness value (RV theta = 0.223) indicated moderate resistance to potential unmeasured confounding.
• The authors noted the study was observational in design, limiting causal inference.