Identification of the Best Anthropometric Predictors of Serum High- and Low-Density Lipoproteins Using Machine Learning

Serum high-density lipoprotein (HDL) and low-density lipoprotein (LDL) cholesterol levels are associated with risk factors for various diseases and are related to anthropometric measures. However, controversy remains regarding the best anthropometric indicators of the HDL and LDL cholesterol levels. The objectives of this study were to identify the best predictors of HDL and LDL cholesterol using statistical analyses and two machine learning algorithms and to compare the predictive power of combined anthropometric measures in Korean adults. A total of 13 014 subjects participated in this study. The anthropometric measures were assessed with binary logistic regression (LR) to evaluate statistically significant differences between the subjects with normal and high LDL cholesterol levels and between the subjects with normal and low HDL cholesterol levels. LR and the naive Bayes algorithm (NB), which provides more reasonable and reliable results, were used in the analyses of the predictive power of individual and combined measures. The best predictor of HDL was the rib to hip ratio (p = <; 0.0001; odds ratio (OR) = 1.895; area under curve (AUC) = 0.681) in women and the waist to hip ratio (WHR) (p = <;0.0001; (OR) = 1.624; AUC = 0.633) in men. In women, the strongest indicator of LDL was age (p = <;0.0001; OR = 1.662; AUC by NB = 0.653); AUC byLR = 0.636. Among the anthropometric measures, the body mass index (BMI), WHR, forehead to waist ratio, forehead to rib ratio, and forehead to chest ratio were the strongest predictors of LDL; these measures had similar predictive powers. The strongest predictor in men was BMI (p = <;0.0001; OR = 1.369; AUC by NB = 0.594; AUC by LR = 0.595). The predictive power of almost all individual anthropometric measures was higher for HDL than for LDL, and the predictive power for both HDL and LDL in women was higher than for men. A combination of anthropometric measures slightly improved the predictive power for both HDL and LDL cholesterol. The best indicator for HDL and LDL might differ according to the type of cholesterol and the gender. In women, but not men, age was the variable that strongly predicted HDL and LDL cholesterol levels. Our findings provide new information for the development of better initial screening tools for HDL and LDL cholesterol.

[1]  C. Molinari,et al.  Comparison of different anthropometric indices for identifying dyslipidemia in school children. , 2011, Clinical biochemistry.

[2]  G. Jensen,et al.  Elevated HDL cholesterol is a risk factor for ischemic heart disease in white women when caused by a common mutation in the cholesteryl ester transfer protein gene. , 2000, Circulation.

[3]  L. Mosca,et al.  Waist circumference, body mass index, and their association with cardiometabolic and global risk. , 2009, Journal of the cardiometabolic syndrome.

[4]  Duong Duc Pham,et al.  Thoracic-to-hip circumference ratio as a novel marker of type 2 diabetes, independent of body mass index and waist-to-hip ratio, in Korean adults. , 2014, Diabetes research and clinical practice.

[5]  M. Visser,et al.  What aspects of body fat are particularly hazardous and how do we measure them? , 2006, International journal of epidemiology.

[6]  Boncho Ku,et al.  Prediction of Fasting Plasma Glucose Status Using Anthropometric Measures for Diagnosing Type 2 Diabetes , 2014, IEEE Journal of Biomedical and Health Informatics.

[7]  Lora E. Burke,et al.  Prevalence of the Metabolic Syndrome Among Overweight and Obese College Students in Korea , 2010, The Journal of cardiovascular nursing.

[8]  Sergey A Dzugan,et al.  Hypercholesterolemia treatment: a new hypothesis or just an accident? , 2002, Medical hypotheses.

[9]  Nader Rifai,et al.  Non-HDL cholesterol and apolipoprotein B predict cardiovascular disease events among men with type 2 diabetes. , 2004, Diabetes care.

[10]  Tae-Yoon Hwang,et al.  Incidence of Metabolic Syndrome and Relative Importance of Five Components as a Predictor of Metabolic Syndrome: 5-Year Follow-up Study in Korea , 2013, Journal of Korean medical science.

[11]  Michael E Griswold,et al.  Age is positively associated with high-density lipoprotein cholesterol among African Americans in cross-sectional analysis: the Jackson Heart Study. , 2011, Journal of clinical lipidology.

[12]  H U Wanner,et al.  Anthropometric and lifestyle correlates of serum lipoprotein and apolipoprotein levels among normal non-smoking men and women. , 1989, Atherosclerosis.

[13]  Makoto Nishida,et al.  A new simple method for the measurement of visceral fat accumulation by bioelectrical impedance. , 2005, Diabetes care.

[14]  N. Milman,et al.  Relationship between serum ferritin and risk factors for ischaemic heart disease in 2235 Danes aged 30–60 years , 1999, Journal of internal medicine.

[15]  J. Després,et al.  Treatment of obesity: need to focus on high risk abdominally obese patients , 2001, BMJ : British Medical Journal.

[16]  Matthew Larkin,et al.  National Heart Lung and Blood Institute, National Institute of Health , 2012 .

[17]  W. Shau,et al.  Optimal cut-off values for obesity: using simple anthropometric indices to predict cardiovascular risk factors in Taiwan , 2002, International Journal of Obesity.

[18]  Yanqing Zhang,et al.  SVMs Modeling for Highly Imbalanced Classification , 2009, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[19]  M. Woodward,et al.  The discrimination of dyslipidaemia using anthropometric measures in ethnically diverse populations of the Asia–Pacific Region: The Obesity in Asia Collaboration , 2010, Obesity reviews : an official journal of the International Association for the Study of Obesity.

[20]  W. Aekplakorn,et al.  Obesity indices and cardiovascular risk factors in Thai adults , 2006, International Journal of Obesity.

[21]  F. Maxfield,et al.  Sterol and lipid trafficking in mammalian cells. , 2006, Biochemical Society transactions.

[22]  V. Pallottini,et al.  Age-Related Hypercholesterolemia and HMG-CoA Reductase Dysregulation: Sex Does Matter (A Gender Perspective) , 2010, Current gerontology and geriatrics research.

[23]  Andrew Smart,et al.  A multi-dimensional model of clinical utility. , 2006, International journal for quality in health care : journal of the International Society for Quality in Health Care.

[24]  G. Moneta,et al.  Major Lipids, Apolipoproteins, and Risk of Vascular Disease , 2010 .

[25]  P. Mirmiran,et al.  Waist-to-hip ratio is a better screening measure for cardiovascular risk factors than other anthropometric indicators in Tehranian adult men , 2004, International Journal of Obesity.

[26]  S-K Lee.  Should waist circumference be used to identify metabolic disorders than BMI in South Korea? , 2010, European Journal of Clinical Nutrition.

[27]  A Trichopoulou,et al.  Associations of anthropometric characteristics with blood cholesterol fractions among adults. The Greek EPIC study , 2006, European Journal of Clinical Nutrition.

[28]  A. von Eckardstein,et al.  High-density lipoprotein cholesterol as a predictor of coronary heart disease risk. The PROCAM experience and pathophysiological implications for reverse cholesterol transport. , 1996, Atherosclerosis.

[29]  Lee Sk,et al.  Should waist circumference be used to identify metabolic disorders than BMI in South Korea , 2010 .

[30]  Christian Jung,et al.  Anthropometric indices as predictors of the metabolic syndrome and its components in adolescents , 2009, Pediatrics international : official journal of the Japan Pediatric Society.

[31]  Rachel Hess,et al.  Obesity in Older Adults: A Systematic Review of the Evidence for Diagnosis and Treatment , 2006, Obesity.

[32]  James J. Chen,et al.  Class-imbalanced classifiers for high-dimensional data , 2013, Briefings Bioinform..

[33]  Gen-Min Lin,et al.  Low high-density lipoprotein cholesterol and low/normal body mass index are associated with increased mortality in coronary artery disease patients in Taiwan. , 2013, Circulation journal : official journal of the Japanese Circulation Society.

[34]  Kuninori Shiwaku,et al.  Predictive values of anthropometric measurements for multiple metabolic disorders in Asian populations. , 2005, Diabetes research and clinical practice.

[35]  T. Lam,et al.  Waist to stature ratio is more strongly associated with cardiovascular risk factors than other simple anthropometric indices. , 2003, Annals of epidemiology.

[36]  U. Pagotto,et al.  Prominent role of low HDL-cholesterol in explaining the high prevalence of the metabolic syndrome in polycystic ovary syndrome. , 2009, Nutrition, metabolism, and cardiovascular diseases : NMCD.

[37]  Roya Kelishadi,et al.  Is Family History of Premature Cardiovascular Diseases Appropriate for Detection of Dyslipidemic Children in Population-Based Preventive Medicine Programs? CASPIAN Study , 2006, Pediatric Cardiology.

[38]  Margaret Ashwell,et al.  Waist to Height Ratio Is a Simple and Effective Obesity Screening Tool for Cardiovascular Risk Factors: Analysis of Data from the British National Diet and Nutrition Survey of Adults Aged 19–64 Years , 2009, Obesity Facts.

[39]  U. Igbavboa,et al.  Amyloid β-protein stimulates trafficking of cholesterol and caveolin-1 from the plasma membrane to the Golgi complex in mouse primary astrocytes , 2009, Neuroscience.

[40]  Darin J Erickson,et al.  Sex‐Specific HDL Cholesterol Changes With Weight Loss and Their Association With Anthropometric Variables: The LIFE Study , 2011, Obesity.

[41]  M. Fernández,et al.  The LDL to HDL Cholesterol Ratio as a Valuable Tool to Evaluate Coronary Heart Disease Risk , 2008, Journal of the American College of Nutrition.

[42]  E. Poehlman,et al.  Obesity, body fat distribution, and coronary artery disease. , 2000, Journal of cardiopulmonary rehabilitation.

[43]  Kwang Hyun Cho,et al.  Induction of tissue inhibitor of matrix metalloproteinase-2 by cholesterol depletion leads to the conversion of proMMP-2 into active MMP-2 in human dermal fibroblasts , 2010, Experimental & Molecular Medicine.

[44]  P. Barter,et al.  High-density lipoproteins and coronary heart disease. , 1991, Journal of cardiovascular risk.

[45]  K. Jablonski,et al.  Non-HDL cholesterol as a predictor of cardiovascular disease in type 2 diabetes: the strong heart study. , 2003, Diabetes care.

[46]  R. Gillum,et al.  Distribution of waist-to-hip ratio, other indices of body fat distribution and obesity and associations with HDL cholesterol in children and young adults aged 4–19 years: The Third National Health and Nutrition Examination Survey , 1999, International Journal of Obesity.

[47]  Clement Adebamowo,et al.  Low HDL-Cholesterol with Normal Triglyceride Levels is the Most Common Lipid Pattern in West Africans and African Americans with Metabolic Syndrome: Implications for Cardiovascular Disease Prevention. , 2010, CVD prevention and control.

[48]  S. Grundy,et al.  Joint distribution of non-HDL and LDL cholesterol and coronary heart disease risk prediction among individuals with and without diabetes. , 2005, Diabetes care.

[49]  Haejung Lee,et al.  A Study on the Reliability of Sasang Constitutional Body Trunk Measurement , 2011, Evidence-based complementary and alternative medicine : eCAM.

[50]  Gary K Grunwald,et al.  Association of anthropometric measures with risk of diabetes and cardiovascular disease in Hispanic and Caucasian adolescents. , 2003, Preventive medicine.

[51]  N. Cook,et al.  Non-HDL cholesterol, apolipoproteins A-I and B100, standard lipid measures, lipid ratios, and CRP as risk factors for cardiovascular disease in women. , 2005, JAMA.

[52]  Pedro Larrañaga,et al.  A review of feature selection techniques in bioinformatics , 2007, Bioinform..

[53]  J. Y. Kim,et al.  A Comparison of the Predictive Power of Anthropometric Indices for Hypertension and Hypotension Risk , 2014, PloS one.

[54]  Demosthenes B. Panagiotakos,et al.  Importance of LDL/HDL cholesterol ratio as a predictor for coronary heart disease events in patients with heterozygous familial hypercholesterolaemia: a 15-year follow-up (1987-2002) , 2003, Current medical research and opinion.

[55]  Chu-Chih Chen,et al.  Heterogeneity of body mass index, waist circumference, and waist-to-hip ratio in predicting obesity-related metabolic disorders for Taiwanese aged 35-64 y. , 2009, Clinical nutrition.

[56]  Xue-wen Chen,et al.  Combating the Small Sample Class Imbalance Problem Using Feature Selection , 2010, IEEE Transactions on Knowledge and Data Engineering.

[57]  LarrañagaPedro,et al.  A review of feature selection techniques in bioinformatics , 2007 .