Association of Postprandial Triglyceride and Retinyl Palmitate Responses With Newly Diagnosed Exercise-Induced Myocardial Ischemia in Middle-Aged Men and Women

Abstract

Although strong evidence exists linking fasting plasma levels of LDL cholesterol (LDL-C) and HDL cholesterol (HDL-C) to risk for development of coronary artery disease (CAD), the data in support of an independent role for fasting triglyceride (TG) concentrations are weak. Humans are in the postprandial state most of the day, however, and results from both basic and clinical studies suggest that postprandial TG levels may be atherogenic. Previous studies have not, however, attempted to determine if postprandial TG levels are associated with CAD independent of other traditional risk factors or plasma lipid levels, particularly fasting plasma concentrations of TG and HDL-C. Ninety-two men and 113 women (mean age, 51.6 and 53.6 years, respectively) were recruited from populations undergoing diagnostic exercise electrocardiographic or thallium stress tests at our medical centers. Twenty-six men and 24 women had positive tests. We chose exercise-induced myocardial ischemia (EIM) as the criterion for defining case and control subjects because we wanted participants who did not have a prior diagnosis of CAD. Blood samples were obtained for measurement of plasma TG, TG-rich lipoprotein TG, and retinyl palmitate (RP) levels 2, 3.5, 5, and 8 hours after the subjects had consumed a fatty test meal. Logistic regression models were developed to test for associations between each variable and case-control status. Among men but not women postprandial TG and RP responses were associated with EIMI independent of age, race, and smoking status. In the male group, the odds ratio (OR) for an increase in postprandial TG response of approximately 1 SD was 1.69 (P = .007); the OR for an increase in RP response of 1 SD was 2.47 (P = .011). However, when fasting TG was added to the model, the OR for postprandial TG area in the men was reduced to 1.44 (P = .17); the OR postprandial RP area in the men was reduced to 1.88 (P = .12). There was no effect of adding other risk factors, including LDL-C and HDL-C, to the model. Significant effect modification by body mass index (BMI) on the relationship between postprandial responses and case-control status was observed. In men with BMI < 30, the OR was 1.83 for postprandial TG (P = .041) and 2.77 for postprandial RP (P = .032) in models that included fasting TG, LDL-C, and hypertension.