Abstract
Objective
To determine the prevalence of hypertriglyceridemia and the mean serum triglyceride concentrations in different degrees of glucose tolerance—non-insulin-dependent diabetes mellitus (NIDDM), impaired glucose tolerance (IGT), and normal glucose tolerance (NGT). In addition, we analyzed the correlates of serum triglyceride concentration to explain why it is more prevalent in diabetic subjects.
Resarch Design and Methods
This study was a cross-sectional survey of 4000 people aged 45-64 yr randomly drawn from the population register of the Finnish population of the provinces of North Karelia and Kuopio in eastern Finland and Turku/Loimaa area in southwestern Finland and stratified by four 10-yr age- and sex groups. The final material comprised 96 subjects with NIDDM, 102 subjects with IGT, and 323 subjects with normal glucose tolerance classified on the basis of two 2-h oral glucose tolerance tests. The prevalence of hypertriglyceridemia by the glucose tolerance status and the variation in serum triglycerides associated with selected life-style and biochemical factors were executed as the main outcome measures.
Results
The prevalence of hypertriglyceridemia (≥ 2.3 mM) was 47.6% (95% confidence interval [CI] 32.5–62.7%) in NIDDM men, 21.9% (95% CI 7.6–36.2%) in IGT men, and 15.4% (95% CI 9.3–21.5%) in NGT. In women, hypertriglyceridemia was found in 51.9% (95% CI 38.6–65.2%) among those with NIDDM, 25.7% (95% CI 15.5–35.9%) among those with IGT, and 10.7% (95% CI 6.3–15.1%) in women with NGT. After adjusting for body mass index (BMI) and age, the difference in the prevalence of hypertriglyceridemia between the glucose tolerance groups remained significant in both men (P = 0.008) and women (P = 0.0001). High serum total cholesterol, high BMI, high waist-hip ratio, and low high-density lipoprotein (HDL) cholesterol were significantly associated with high serum triglycerides in all glucose tolerance groups. No synergistic effect between these parameters and glucose tolerance status was found. In multiple linear regression analyses, fasting plasma insulin, diabetes status, and serum uric acid were significant predictors of serum triglyceride concentration after taking into account age, BMI, and HDL and total cholesterol. The association between BMI and serum triglycerides in the regression analysis was significant only when plasma insulin was not included in the model.
Conclsions
Hypertriglyceridemia is common in subjects with NIDDM and IGT and is often associated with low HDL cholesterol, high total cholesterol, hyperinsulinemia, and elevated serum uric acid concentration.