Association of Galectin-3 With Diabetes Mellitus in the Dallas Heart Study

Abstract

Context: Galectin-3 is a biomarker associated with inflammation and fibrosis in cardiac, liver, and renal disease. Galectin-3 is higher in overweight and obese individuals; whether an association with diabetes exists independent of weight is unknown.

Objective: To evaluate if galectin-3 is associated with diabetes mellitus.

Design: We performed measurements of galectin-3 among participants in the Dallas Heart Study (DHS) Phases 1 and 2 (DHS-1 and DHS-2; n = 3392, and n = 3194, respectively). Of these, 1989 participants were evaluated longitudinally in both studies. Associations of galectin-3 with prevalent and incident type 2 diabetes were determined using logistic regression models. Associations of galectin-3 with relevant biomarkers and fat compartments were evaluated using Spearman correlation coefficients and multivariable linear regression models, respectively.

Setting and participants: DHS is a population-based, single-site, multiethnic study conducted in Dallas County, Texas, with oversampling to comprise 50% blacks.

Results: Galectin-3 levels were associated with diabetes prevalence in DHS-1 [OR 1.56 per SD change in log-galectin (95% CI 1.41 to 1.73)] and DHS-2 [OR 1.86 (95% CI 1.67 to 2.06)]. Galectin-3 levels in DHS-1 also associated with incident diabetes mellitus over the 7.1 (interquartile range 6.6 to 7.6)-year follow-up period [OR 1.34 (95% CI 1.14 to 1.58)]. These associations maintained significance in models adjusted for traditional metabolic risk factors (age, sex, race, body mass index, and hypertension) and renal function. Galectin-3 levels correlated with levels of biomarkers implicated in inflammation (high-sensitivity C-reactive peptide, IL-18, monocyte chemoattractant protein 1, soluble TNF receptor 1A, myeloperoxidase), insulin secretion (C-peptide and C-peptide/homeostatic model assessment for insulin resistance), and subcutaneous adiposity.

Conclusions: Galectin-3 is associated with diabetes prevalence and incidence, possibly through the inflammatory pathway contributing to β-cell fibrosis and impaired insulin secretion.