Markers of endothelial dysfunction in the prediction of coronary artery disease in Type 1 diabetes. The Pittsburgh Epidemiology of Diabetes Complications Study

• Published in: Journal of diabetes and its complications Vol. 19; no. 4; pp. 183 - 193
• Main Authors: Costacou, Tina, Lopes-Virella, Maria F., Zgibor, Janice C., Virella, Gabriel, Otvos, Jim, Walsh, Michael, Orchard, Trevor J.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01-07-2005; Elsevier Limited
• Subjects: Adult; Atherosclerosis; Biomarkers - blood; Cardiovascular disease; Case-Control Studies; Cell adhesion & migration; Cellular adhesion molecules; Coronary artery disease; Coronary Artery Disease - blood; Coronary Artery Disease - etiology; Diabetes; Diabetes Mellitus, Type 1 - blood; Diabetic Angiopathies - blood; E-Selectin - blood; Endothelium; Endothelium, Vascular - physiopathology; Epidemiology; Family medical history; Fatty acids; Female; Follow-Up Studies; Heart attacks; Humans; Leukocytes; Lipoproteins; Logistic Models; Male; Nitric oxide; Predictive Value of Tests; Proportional Hazards Models; Prospective Studies; Proteins; Risk Factors; Standard deviation; Type 1 diabetes; Cellular adhesion molecules; Type 1 diabetes; Coronary artery disease
• ISSN: 1056-8727; 1873-460X
• DOI: 10.1016/j.jdiacomp.2005.01.003

Low-density lipoprotein (LDL) oxidation, the immune response it provokes, and lipoprotein subclasses measured by nuclear magnetic resonance (NMR) spectroscopy have explained some of the enhanced coronary artery disease (CAD) risks in Type 1 diabetes. We examined whether cellular adhesion molecules further improve CAD prediction.
Participants were identified from the Epidemiology of Diabetes Complications (EDC) cohort, a 10-year prospective study of childhood-onset Type 1 diabetes. Mean age at baseline was 28 years, and diabetes duration was 19 years. CAD incidence was determined by EDC physician-diagnosed angina, confirmed myocardial infarction (MI), stenosis ≥50%, ischemic ECG, or revascularization. Cases were gender, age, and diabetes duration (±3 years) matched with the controls. The samples and risk factors used in the analyses were identified from the earliest exam prior to incidence in the cases. Sixty cases and 72 controls (including 43 pairs) had complete information on all covariates. Cox proportional hazard models with backward elimination and conditional logistic regression (for paired analyses) were conducted. Separate analyses were conducted to examine whether E-selectin related differently to soft (ischemic ECG and angina;
n=68) or hard (revascularization, MI, and fatal events;
n=37) CAD endpoints.
Mean E-selectin concentration was elevated among cases (P=.0009) compared to controls. Adjusting for previously established CAD risk factors, E-selectin remained an independent predictor of CAD (HR=1.07, 95% Cl=1.01-1.15). Multivariable models confirmed the importance of E-selectin as a risk factor of soft (HR=1.13, 95% Cl=1.03−1.24; HRs are per standard deviation increase) but not hard CAD.
Study results suggest that E-selectin may enhance CAD prediction beyond traditional risk factors or markers of oxidative stress in Type 1 diabetes.