The association between plasminogen activator inhibitor-1 (PAI-1) and insulin resistance is established in western countries. The major component of this association is obesity. Accordingly, we examined this association in Japan where the prevalence of obesity is low. Data for fasting PAI-1 levels of 404 subjects were obtained from a general population in a farming area. We measured body mass index (BMI), systolic and diastolic blood pressure, high-density lipoprotein (HDL)-cholesterol, triglycerides, fasting plasma glucose (FPG), insulin, creatinine, and uric acid. The use of alcohol was ascertained by a questionnaire. The formula for the homeostasis model assessment (HOMA) score was used as an index of insulin resistance. Uni- and multivariate analyses were applied for the determinants of plasma PAI-1. Age and sex did not affect plasma PAI-1. The average BMI was 23.0 +/- 3.2 kg/m(2). Thus, most of the subjects were not obese. Because, even in this population, BMI (P <.001) was the strongest determinant for PAI-1 after univariate analysis, we performed multiple linear regression analyses after adjustment for BMI. The significance of triglycerides, FPG, insulin, and the HOMA score still remained. PAI-1 levels were linearly related to the HOMA score. From the subanalysis of the non-obese subjects (BMI < 25; n = 298), waist-hip ratio, triglycerides, FPG, and HOMA scores were significant determinants of PAI-1. This is the first demonstration that increased PAI-1 levels were significantly related to insulin resistance in a Japanese general population. PAI-1 levels are associated with insulin resistance, irrespective of obesity.
Copyright 2003, Elsevier Science (USA). All rights reserved