Seung-Hwan Lee1,2, Hun-Sung Kim1,2, Yong-Moon Park3, Hyuk-Sang Kwon4, Kun-Ho Yoon1,2, Kyungdo Han5, Mee Kyoung Kim4. 1. Division of Endocrinology and Metabolism, Department of Internal Medicine, Seoul St. Mary's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea. 2. Department of Medical Informatics, College of Medicine, The Catholic University of Korea, Seoul, Korea. 3. Epidemiology Branch, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, North Carolina. 4. Division of Endocrinology and Metabolism, Department of Internal Medicine, Yeouido St. Mary's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea. 5. Department of Medical Statistics, College of Medicine, The Catholic University of Korea, Seoul, Korea.
Abstract
CONTEXT: The bidirectional relationship between low high-density lipoprotein cholesterol (HDL-C) and glucose intolerance is well established. Recent studies suggested an association of lipid variability with various health outcomes. OBJECTIVE: To investigate the combined effect of HDL-C levels and their variability on the risk of diabetes. DESIGN: A population-based cohort study. SETTING AND PARTICIPANTS: In all, 5,114,735 adults without known diabetes in the Korean National Health Insurance System cohort who underwent three or more health examinations from 2009 to 2013 were included. Visit-to-visit HDL-C variability was calculated using variability independent of the mean (VIM) and the coefficient of variation (CV). Low mean and high variability groups were defined as the lowest and highest quartiles of HDL-C mean and variability, respectively. MAIN OUTCOME MEASURES: Newly developed diabetes. RESULTS: There were 122,192 cases (2.4%) of incident diabetes during the median follow-up of 5.1 years. Lower mean or higher variability of HDL-C was associated with higher risk of diabetes in a stepwise manner, and an additive effect of the two measures was noted. In the multivariable-adjusted model, the hazard ratios and 95% CIs for incident diabetes were 1.20 (1.18 to 1.22) in the high mean/high VIM group, 1.35 (1.33 to 1.37) in the low mean/low VIM group, and 1.40 (1.38 to 1.42) in the low mean/high VIM group compared with the high mean/low VIM group. Similar results were observed when modeling the variability using CV and in various subgroup analyses. CONCLUSIONS: Low mean and high variability in HDL-C were independent predictors of diabetes with an additive effect. Both elevating and stabilizing HDL-C may be important goals for reducing diabetes risk.
CONTEXT: The bidirectional relationship between low high-density lipoprotein cholesterol (HDL-C) and glucose intolerance is well established. Recent studies suggested an association of lipid variability with various health outcomes. OBJECTIVE: To investigate the combined effect of HDL-C levels and their variability on the risk of diabetes. DESIGN: A population-based cohort study. SETTING AND PARTICIPANTS: In all, 5,114,735 adults without known diabetes in the Korean National Health Insurance System cohort who underwent three or more health examinations from 2009 to 2013 were included. Visit-to-visit HDL-C variability was calculated using variability independent of the mean (VIM) and the coefficient of variation (CV). Low mean and high variability groups were defined as the lowest and highest quartiles of HDL-C mean and variability, respectively. MAIN OUTCOME MEASURES: Newly developed diabetes. RESULTS: There were 122,192 cases (2.4%) of incident diabetes during the median follow-up of 5.1 years. Lower mean or higher variability of HDL-C was associated with higher risk of diabetes in a stepwise manner, and an additive effect of the two measures was noted. In the multivariable-adjusted model, the hazard ratios and 95% CIs for incident diabetes were 1.20 (1.18 to 1.22) in the high mean/high VIM group, 1.35 (1.33 to 1.37) in the low mean/low VIM group, and 1.40 (1.38 to 1.42) in the low mean/high VIM group compared with the high mean/low VIM group. Similar results were observed when modeling the variability using CV and in various subgroup analyses. CONCLUSIONS: Low mean and high variability in HDL-C were independent predictors of diabetes with an additive effect. Both elevating and stabilizing HDL-C may be important goals for reducing diabetes risk.