The relationship of small dense LDL cholesterol and various parameters of metabolic syndrome in Korean population

Abstract

The purpose of this cross-sectional study was to describe the relationship between small dense low density lipoprotein cholesterol (sdLDL-C) and various metabolic parameters of 890 men and women aged 20 - 70 years from a Korean populations sampled by the Korean Metabolic Syndrome Research Initiatives (KMSRI)-Seoul cohort study. The study population consists of 318 MetS only patients (MetS+DM-), 83 DM only patients (MetS-DM+), 129 patients with both disease (MetS+DM+), and 360 gender matched healthy controls (MetS-DM-).Fasting plasma concentrations of sdLDL-C were higher in men than in women (42.67±0.85 mg/dL vs 36.66±0.86 mg/dL, p < 0.0001), and the tendency was continuous over all age group except for those who are > 60. The 25th-percentile and 75th-percentile concentrations in healthy control were 20.54 mg/dL and 40.92 mg/dL, respectively.Among four disease groups, MetS+DM+ and MetS+DM- group showed significantly higher level of sdLDL-C over MetS-DS+ and MetS-DS- group (p < 0.0001). Association of sdLDL-C concentration with each metabolic disease also differed. sdLDL-C level was higher in MetS group comparing with that in non-MetS group (52.88±9.29 mg/dL vs 47.48±9.93 mg/dL, p < 0.0001), and within MetS group, a stepwise increase of sdLDL-C in parallel to the number of MetS components was observed (p for trend < 0.0001). Furthermore, its correlation with HOMA-IR as a surrogate marker of insulin resistance, which may play a central role in metabolic disease progression, was significant (R = 0.352, p < 0.0001). In the contrary, sdLDL-C level was not different in comparison between diabetic and non-diabetic group (40.97±19.3 mg/dL vs. 39.29±18.08 mg/dL, p = 0.2468), yet sdLDL-C/LDL-C ratio differed (34.88±14.54 mg/dL vs. 32.15±12.75 mg/dL, p = 0.015), which implies possible usage of the ratio in clinic.We performed multiple linear regression and found that most strongly associated factors with sdLDL-C level in both gender were TG, LDL-C and fasting insulin (p < 0.05). Age and fasting glucose played additive effect on sdLDL-C only in women (p < 0.05).The receiver operating characteristic area under the curve value of these markers for MetS was 0.733 and 0.755, respectively. Logistic regression analyses showed that the odds ratio (OR) of the MetS as predicted by sdLDL-C in the presence of potential confounding variables was 1.0511 (95% CI, 1.0414 ~ 1.0608) and OR of MetS as predicted by sdLDL-C/LDL-C ratio was 1.0768 (95% CI, 1.0623 ~ 1.0915). Although sdLDL-C and sdLDL-C/LDL-C ratio do not give superiority over constitutional parameters of MetS in regard to diagnosis of MetS, they have significant association with MetS status and its severity.The association between metabolic parameters and sdLDL-C concentrations observed in this study was consistent with previous reports, mostly grounded on phenotypic studies of LDL subfraction, and we could calculate degree of association more precisely in relatively larger population by virtue of using fully automated quantitative analysis. In summary, sdLDL-C concentration is closely related to various parameters of metabolic disorders, especially with insulin resistance and dyslipidemia. However, the level of its association with each metabolic disorder differed in that, whereas MetS showed significant correlation with it, even well with its severity assessed by the number of metabolic components, sole presence of DM did not alter the concentration.Thus measurement of sdLDL-C might improve clinical assessment of MetS severity and explain some of the additional risk for CHD in diabetic patient with MetS.