Association of Non-Insulin-Based Markers of Insulin Resistance With Hypertension in Type 2 Diabetes: An Age- and Gender-Matched Cross-Sectional Study.

Abstract

Background: This study investigated the association and discriminative ability of 6 surrogate insulin resistance (IR) indices-TyG, TyG-BMI, TyG-WC, TyG-WHtR, METS-IR, and the TG/HDL-C ratio-for hypertension (HTN) in patients with type 2 diabetes (T2D).

Methods: This cross-sectional, age- and gender-matched case-control study included 4236 patients with T2D (2167 with HTN and 2069 without) who attended a diabetes clinic between January 2014 and December 2024. Associations between surrogate IR indices and HTN were assessed using RCS and multivariable logistic regression. Discriminative ability and calibration were evaluated using ROC and calibration plots. The net reclassification improvement and integrated discrimination improvement analyses quantified the incremental value beyond a basic model.

Results: Each surrogate IR index showed a significant non-linear association with HTN. After adjusting for confounding factors, the ORs for HTN increased with higher values of each index. All indices demonstrated significant discriminative ability (AUCs ≥ 0.690), with TyG-BMI showing the highest AUC (0.698; 95% CI, 0.664-0.733). A TyG-BMI cutoff of 144.5 identified HTN with 61% sensitivity and 71% specificity (P < .001). Calibration was relatively good, and the addition of surrogate IR indices to the basic model (age, gender, diabetes duration, LDL-C, HbA1c, and eGFR) significantly improved HTN identification.

Conclusions: These findings support the use of surrogate IR indices as practical tools for identifying and assessing the risk of HTN in patients with T2D and highlight the potential role of IR in the development of HTN.