- Email: [email protected]
Relationship of serum 25-Hydroxyvitamin D (25[OH]D) levels and components of metabolic syndrome in prepubertal children
Nutrition 31 (2015) 1324–1327
Contents lists available at ScienceDirect
Nutrition journal homepage: www.nutritionjrnl.com
Applied nutritional investigation
Relationship of serum 25-Hydroxyvitamin D (25[OH]D) levels and components of metabolic syndrome in prepubertal children Jung Hyun Kwon M.D. a, Seung Eun Lee M.D. a, Hye Ah Lee M.D. b, Young Ju Kim M.D. c, Hwa Young Lee M.D. d, Hye Sun Gwak Pharm.D., Ph.D. e, Eun Ae Park M.D., Ph.D. a, Su Jin Cho M.D., Ph.D. a, Se Young Oh Ph.D. f, Eun Hee Ha M.D., Ph.D. b, Hyesook Park M.D., Ph.D. b, *, Hae Soon Kim M.D., Ph.D. a, ** a
Department of Pediatrics, School of Medicine, Ewha Womans University, Seoul, Korea Department of Preventive Medicine, School of Medicine, Ewha Womans University, Seoul, Korea c Department of Obstetrics and Gynecology, School of Medicine, Ewha Womans University, Seoul, Korea d Department of Anatomy, School of Medicine, Ewha Womans University, Seoul, Korea e College of Pharmacy, Ewha Womans University, Seoul, Korea f Department of Food and Nutrition, Kyung Hee University, Seoul, Korea b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 February 2015 Accepted 30 April 2015
Objective: The relationship between 25-hydroxyvitamin D (25[OH]D) levels and the lipid and metabolic levels of the prepubertal normal population is unclear. Our goals were to investigate the association of serum 25(OH)D concentrations with lipid and metabolic levels in Korean prepubertal children ages 7–9 y. Methods: We followed 205 children, ages 7–9 y in the Ewha Birth and Growth Cohort study, a prospective cohort sample established in 2001–2006, from July to August 2011. We studied the association of serum 25-hydroxyvitamin D (25[OH]D) levels with components of metabolic syndrome and insulin resistance indices using multivariate regression analysis adjusted for body mass index (BMI) z-scores. Results: The mean age of the 205 subjects was 7.89 0.85 y, and the sample included 109 boys (53.2%). The average 25(OH)D levels of all participants was 25.0 5.4 ng/mL. After adjustment for age and sex, triacylglycerol levels were significantly associated with 25(OH)D (b ¼ –0.02, P ¼ 0.02) concentrations even after adjustment for BMI z-scores (b ¼ 0.02, P ¼ 0.04). However, other metabolic components were not correlated with 25(OH)D status. Those with the lowest quartile of 25(OH)D levels had the highest serum triacylglycerol levels (P ¼ 0.04, Ptrend ¼ 0.01). Conclusions: We found that serum 25(OH)D levels were negatively associated with serum triacylglycerol levels, even independently of adiposity, in prepubertal children. This study suggests that 25(OH)D insufficiency is related to metabolic syndrome via the derangement of triacylglycerol metabolism. Ó 2015 Elsevier Inc. All rights reserved.
Keywords: Vitamin D Dyslipidemia Children
Introduction Recent literature has suggested an association between vitamin D deficiency and obesity and metabolic syndrome in addition to its well-known effects on calcium homeostasis and
This work was supported by National Research Foundation of Korea Grant funded by the Korean Government (2010-0026225). * Corresponding author. Tel.: þ82.2-2650-5275; fax: þ82.2-2650-5569. ** Co-corresponding author. Tel.: þ82.2-2650-5275; fax: þ82.2-2650-5569. E-mail addresses: [email protected] (H. Park), [email protected] (H. S. Kim). http://dx.doi.org/10.1016/j.nut.2015.04.023 0899-9007/Ó 2015 Elsevier Inc. All rights reserved.
bone metabolism [1,2]. Vitamin D deficiency is becoming a widespread global problem, and it has been estimated that up to 1 billion people may have insufficient 25-hydroxyvitamin (OH)D levels (25[OH]D) [3,4]. Indeed, vitamin D deficiency-related obesity and metabolic syndrome are important public health problems in Korea. A study based on the Fourth Korea National Health and Nutrition Examination Survey (KNHANES IV) determined that the serum 25 (OH)D levels of 93% of males and 96% of females ages 10–19 y were insufficient [5]. Vitamin D deficiency is thought to be associated with lipid metabolism. However, studies regarding the association
J. H. Kwon et al. / Nutrition 31 (2015) 1324–1327
1325
between vitamin D and serum lipid levels have reported a variety of results. A cross-sectional study of children and adolescents conducted by Delvin et al. reported that serum 25(OH)D levels were positively correlated with total cholesterol and triacylglycerol levels in females [6], and Shin et al. reported an inverse relationship between serum 25(OH)D levels and triacylglycerol and low-density lipoprotein cholesterol (LDL-c) levels in Korean adolescents [7]. Additionally, it has been proposed that, independent of obesity, low serum 25(OH)D concentration is a risk factor for the development of glucose intolerance and type 2 diabetes because it reduces insulin secretion and sensitivity in adults [8,9]. A recent study in adolescents identified a negative association between 25(OH)D levels and serum glucose and triacylglycerol levels and a positive association between 25(OH)D levels and serum high-density lipoprotein cholesterol (HDL-c) levels, independent of weight [10]. Furthermore, as there is a complex interaction between metabolic syndrome and the reproductive axis [11], prepubertal lipid and glucose metabolism differs from that of pubertal individuals and adults, but studies demonstrating relationships among vitamin D deficiency and lipid and metabolic levels in Asian prepubertal children are rare. We already reported that 25(OH)D concentrations were inversely related to adiposity indices in the prepubertal children ages 7–9 y who participated in the Ewha Birth and Growth Cohort study [12]. In this follow-up study among prepubertal children in Korea, we aimed to investigate whether serum 25(OH)D concentrations were associated with lipid and metabolic levels as well as with insulin resistance indices, independent of adiposity.
total cholesterol/5.0). Serum insulin was measured to enable use of an immunoradiometric assay kit (Biosource Europe, Nivelles, Belgium). Insulin resistance indices were analyzed to enable use of the homeostasis model assessment of insulin resistance (HOMA-IR); (plasma glucose [mmol/L] insulin [mIU/mL])/22.5 and the glucose-to-insulin ration (GIR).
Methods
Results
Study population In this study, our goal was to examine the relationship between serum 25(OH)D level and components of metabolic syndrome in Korean preadolescent children, by extending the range of the prior study [12]. This study was performed as part of the Ewha Birth and Growth Cohort study, which involves an ongoing long-term research cohort established in 2001–2006. When participants reached 7–9 y of age, we collected data from July to August 2011. During this period, both general physical measurements and blood samples were taken. A total of 205 children with complete data were included in the present study (109 males and 96 females). The parents or guardians of the children provided consent for their child’s participation in this study. This study was approved by the Institutional Review Board of Ewha Woman’s University Hospital. Data collection We used body mass index (BMI), calculated by dividing weight by the square of height, as an adiposity indicator. A detailed description of the methods used to measure height and weight were presented in a previous publication [12]. We calculated BMI z-scores based on the 2007 Korean Children and Adolescents Growth Standard. Blood pressure (BP) was measured using an automatic device (Critikon, Tampa, FL, USA) using the correct cuff size after participants rested in a stable position with their arm properly supported. Two measurements, taken within 5 min of each other, were averaged. High BP was defined as systolic blood pressure (SBP) and diastolic blood pressure (DBP) higher than the 90th percentile of the age- and sex-specific norms in the 2007 Korean Children and Adolescents Growth Standard [13]. All blood samples were collected after at least 8 h of fasting and immediately stored in 70 C freezers until analysis. The 25(OH)D blood concentration (ng/ mL), an indicator of vitamin D levels, was measured using a radioimmunoassay (DiaSorin/Stillwater, MN, USA) with a Gamma counter (Wallac 1470 Wizard, Perkin Elmer, Waltham, MA, USA). The mean and standard deviation of serum 25(OH)D levels was 25.0 5.4 ng/mL (males, 23.2 4.4 ng/mL; females, 27.0 5.9 ng/mL). The total cholesterol, triacylglycerol, HDL cholesterol, and glucose levels of each child were measured using an automatic analyzer (Model: 7180, Hitachi, Tokyo, Japan), and LDL cholesterol levels (mg/dL) were calculated using the Friedewald formula equations (LDL cholesterol ¼ cholesterol–HDL cholesterol–
Statistical analysis All statistical analyses were conducted using SAS version 9.3 (SAS Institutes, Cary, NC, USA). Descriptive analyses presented the distributions of numeric variables as means and standard deviations and those of categorical variables as numbers and percentages. t tests were used to assess mean differences in 25(OH) D concentrations according to sex, overweight status, dietary supplement intake, and metabolic components. We assessed the magnitude of the association between 25(OH)D levels and metabolic components using analysis of covariance (ANCOVA) and multiple regression analysis. Because the distributions of triacylglycerol and insulin levels did not satisfy the statistical assumption of normality, these values were log transformed. To investigate the mean differences in metabolic components and insulin resistance indices across serum 25(OH)D levels, we used an ANCOVA controlling for age, sex, and BMI z-scores. Only 15.6% of study participants reached an optimal level of 25(OH)D, as recommended by the Institute of Medicine (IOM) (30 ng/mL). Thus, we categorized the data into quartiles according to 25(OH)D concentration (Q1 group: 25[OH]D < 21.4 ng/mL, Q2 group: 21.4–23.9 ng/mL, Q3 group: 23.9–27.9 ng/mL, Q4 group: >27.9 ng/mL) to increase the statistical power. According to the multiple regression analysis, sex, age, birth order (first, second, and third or more), maternal education level (graduated from high school, college or university), and fruit/fruit juice intake (always, generally, and seldom; treated as dummy variables) were considered as confounding factors. These confounding factors, which had a significance of <0.3 according to the descriptive analysis of previous publications, were selected for inclusion in our analyses [12]. Additionally, BMI z-scores were considered because of the strong association of metabolic components with obesity. All analyses were two-tailed, and a P value of <0.05 was considered to indicate statistical significance.
The clinical characteristics, insulin resistance indices, and lipid profiles of the prepubertal children included in the present study are presented in Table 1. A total of 205 children ages 7.89 0.85 y, including 109 males (53.2%), were enrolled. A total of 27 (13.2%) were obese, and 27 (13.3%) had high BP. No children reached a glucose intolerance state, a fasting glucose level higher than 100 mg/dL. The mean differences in metabolic components and insulin resistance indices by 25(OH)D levels are presented in Table 2. The highest serum 25(OH)D quartile showed significantly lower serum triacylglycerol levels and a strong inverse relationship between 25(OH)D and triacylglycerol levels (P ¼ 0.04, Ptrend ¼ 0.01). Those in the lowest 25(OH)D quartile showed higher mean total cholesterol and LDL cholesterol levels, LDL/ HDL ratios, SBP, DBP, log-transformed insulin levels, glucose levels, and HOMA values and lower HDL cholesterol levels, but these relationships did not reach significance. Table 3 presents the results of the multiple regression analysis of the relationship between 25(OH)D concentrations and insulin resistance indices controlling for confounding factors. When controlling for sex and age (Model 1), only the relationship involving triacylglycerol levels reached statistical significance (b ¼ 0.02, P ¼ 0.02). In addition, triacylglycerol levels remained significantly associated with 25(OH)D concentrations even after adjusting for BMI z-scores, age, sex, birth order, fruit/fruit juice intake, and maternal educational level (b ¼ 0.02, P ¼ 0.04 in Model 2). However, no associations among other lipid markers, except triacylglycerol levels (total cholesterol, LDL cholesterol, and HDL cholesterol) and insulin resistance indices, and 25(OH)D concentrations were detected in the total sample.
1326
J. H. Kwon et al. / Nutrition 31 (2015) 1324–1327
Table 1 Characteristics, geometric mean hormone levels, and 95% confidence intervals of 205 prepubertal children ages 7–9 y in the Ewha Birth and Growth Cohort in 2001–2006 Characteristic
Mean (SD) or Geometric mean (95% confidence interval)
Age (years) Sex, % Boy (number) Girl (number) Anthropometrics Mean BMI z-score Overweight/obese, (%)
7.89 (0.85) 109 (53.2%) 96 (46.8%) 0.25 (1.22) 27 (13.2) Boy 17 (8.3); Girl 10 (4.9)
Blood pressure Systolic blood pressure, mmHg Diastolic blood pressure, mmHg High blood pressure, (%) Glycometabolism-related measure Fasting glucose, mg/dL Insulin, mIU/mL HOMA GIR Lipids-related measure HDL cholesterol, mg/dL LDL cholesterol, mg/dL TG, mg/dL
101.71 (9.26) 59.22 (6.84) 27 (13.3) 77.75 7.86 1.61 10.28
(6.65) (1.02); 7.55 (6.37–9.31) (0.71) (2.85)
59.98 (11.19) 87.24 (19.16) 64.65 (1.04); 64.00 (44.00–94.00)
GIR, glucose-to-insulin ratio; HDL-c, high-density lipoprotein cholesterol; HOMA, homeostasis model assessment of insulin resistance; LDL-c, low-density lipoprotein cholesterol; TG, triacylglycerol High blood pressure was defined as systolic or diastolic blood pressure 90th percentile for age and sex based on the 2007 Korean Children and Adolescents Growth Standard Overweight/obese was defined as a body mass index 85th percentile for age and sex based on the 2007 Korean Children and Adolescents Growth Standard LDL-c (mg/dL) was calculated using the Friedewald formula equations (LDL-c ¼ (TC–HDL-c–TG)/5.0)
Discussion We found a strong inverse relationship between 25(OH)D and serum levels in prepubertal children after adjusting for sex and age. Furthermore, the association remained significant after adjustment for adiposity and potential confounders. In particular, those in the lowest 25(OH)D quartile tended to have the highest levels of triacylglycerol. These results are consistent with other findings of a negative association with elevated triacylglycerol levels [14–16]. Lee et al. found that vitamin D deficiency in 9-year-old Korean children was associated with obesity,
hypertension, hyperlipidemia, and hyperglycemia, but that study did not address prepubertal status [14]. Several studies involving adolescents and adults have supported the hypothesis that poor 25(OH)D status is associated with obesity, metabolic syndrome, and insulin resistance [10,17]. Alemzadeh et al. reported a negative association between 25(OH)D and body fat mass in obese children and adolescents, and vitamin D-deficient subjects showed reduced insulin sensitivity compared with vitamin D-sufficient subjects [17]. Therefore we expected that insulin resistance would show the inverse correlation with 25(OH)D levels in our prepubertal children. Those in the lowest 25(OH)D quartile tended to show higher log-transformed insulin, glucose, and HOMA values, although these differences were not significant. Because of the small sample, our results are subject to several limitations, and additional larger longitudinal studies regarding the insulin resistance of prepubertal children are needed. We also did not assess the all elements of metabolic syndrome, because there is no definition of metabolic syndrome in children and none of our subjects reached fasting glucose levels higher than 100 mg/dL. Although only one of the metabolic components changed meaningfully in our study, we might presume that changes associated with 25(OH)D are precipitated from prepubertal period. A study of 237 black American and Caucasian children with an average age of 12.7 2.2 y found that 25(OH)D levels were inversely correlated with adiposity and positively correlated with HDL-c levels [18]. A study among American adolescents found inverse relationships between 25(OH)D and serum glucose and triacylglycerol levels and a positive relationship between this variable and HDL-c levels [10]. However, our results showed only a weak correlation between serum 25(OH)D and HDL-c levels. The association between serum 25(OH)D and lipid profiles has been discussed in many other studies, but the results remain controversial. One study among elderly individuals reported a strong positive relationship between serum 25(OH)D and HDL-c levels [19] and another reported no notable relationship between serum 25(OH)D and triacylglycerol levels [20]. Chiu et al. [8] found an inverse relationship between serum 25(OH)D levels and both total cholesterol and LDL-c levels, whereas no relationship between serum 25(OH)D and triacylglycerol levels was observed. Another study found that lower 25(OH)D concentrations were correlated with higher insulin resistance,
Table 2 Metabolic components and insulin resistance indices in relation to serum 25(OH)D quartiles Metabolic components
Quartile of serum 25(OH)D Q2
Q3
Q4
(<21.4 ng/mL)
(21.4–23.9 ng/mL)
(23.9–27.9 ng/mL)
(>27.9 ng/mL)
(n ¼ 51) TC (mg/dL) Log triacylglycerol HDL-c (mg/dL) LDL-c (mg/dL) LDL/HDL ratio SBP (mmHg) DBP (mmHg) Log insulin Glucose (mg/dL) HOMA GIR
P
Q1
163.1 4.3 57.8 88.0 1.57 103.4 60.1 2.1 79.1 1.7 10.1
3.39 0.07 1.55 2.81 0.06 1.26 0.98 0.04 0.95 0.09 0.37
(n ¼ 51)
(n ¼ 51)
(n ¼ 52)
161.37 4.16 59.89 86.75 1.49 100.83 58.80 2.03 76.09 1.50 10.16
162.83 4.11 61.71 87.16 1.44 101.03 58.43 2.04 77.45 1.57 10.49
160.67 4.07 60.43 87.02 1.47 101.64 59.54 2.07 78.34 1.65 10.38
(3.32) (0.07) (1.52) (2.76) (0.05) (1.23) (0.96) (0.04) (0.93) (0.09) (0.37)
(3.31) (0.07) (1.52) (2.75) (0.05) (1.21) (0.95) (0.04) (0.93) (0.09) (0.37)
(3.37) (0.07) (1.54) (2.80) (0.06) (1.24) (0.97) (0.04) (0.94) (0.09) (0.37)
0.95 0.04* 0.38 0.99 0.43 0.46 0.60 0.57 0.12 0.30 0.87
DBP, diastolic blood pressure; GIR, glucose-to-insulin ratio; HDL-c, high-density lipoprotein cholesterol; HOMA, homeostasis model assessment of insulin resistance; LDL-c; low-density lipoprotein cholesterol; SBP, systolic blood pressure; TC, total cholesterol Results are presented as least-squares means with standard errors. P values were obtained from ANCOVAs adjusted for age, sex, and BMI z-scores. *P for trend was 0.01
J. H. Kwon et al. / Nutrition 31 (2015) 1324–1327 Table 3 Results of multiple regression of serum 25(OH)D levels on metabolic components and insulin resistance indices Metabolic components
Total-Cholesterol (mg/dL) Log triacylglycerol HDL-Cholesterol (mg/dL) LDL-Cholesterol (mg/dL) LDL/HDL ratio SBP (mmHg) DBP (mmHg) Glucose (mg/dL) Log insulin HOMA GIR
Model 1
Model 2
b (P value)
b (P value)
0.06 0.02 0.07 0.23 0.002 0.16 0.06 0.03 0.004 0.01 0.06
0.03 0.02 0.06 0.19 0.002 0.08 0.01 0.03 0.003 0.002 0.05
(0.85) (0.02) (0.66) (0.41) (0.71) (0.20) (0.53) (0.73) (0.31) (0.52) (0.14)
(0.92) (0.04) (0.71) (0.51) (0.78) (0.54) (0.95) (0.73) (0.44) (0.87) (0.19)
DBP, diastolic blood pressure; GIR, glucose-to-insulin ratio; HOMA, homeostasis model assessment of insulin resistance; SBP, systolic blood pressure P value was obtained from multiple linear regression Model 1: Adjusted for age and sex Model 2: Model 1 þ body mass index z-score, age, sex, birth order, fruit/fruit juice intake, and maternal educational level
triacylglycerol concentrations, and increments in blood pressure among elderly Koreans and related the risk for hypertriglyceridemia to vitamin D deficiency [21]. It has been suggested that the following mechanisms underlie the effect of 25(OH)D on lipid profiles. First, it has been proposed that vitamin D inhibits the secretion of parathyroid hormones, which have been known to reduce lipolysis [1]. It has also been proposed that a vitamin D-induced increment in calcium absorption, that is, 25(OH)D, increases calcium levels, which tends to reduce the formation of calcium-fatty soaps in the gut and therefore to increase fat absorption and reduce hepatic triacylglycerol formation and secretion [18,22]. Serum 25(OH)D concentrations are largely determined by environmental factors, primarily vitamin D intake (cholecalciferol and ergocalciferol) and the ultraviolet radiation of 7-hydro cholesterol in the skin [1]. Although we could not investigate the daily outside activities of children, we measured 25(OH)D levels, which reflect the vitamin D nutrient state within the previous 3 wk, and controlled for the intake of dietary supplements. Most studies of vitamin D deficiency in prepubertal children or adults have focused on populations who were obese. Very few previous studies have found an association of vitamin D deficiency with hyperlipidemia and metabolic risk in a general sample of prepubertal children. One strength of this study is that the sample included normal children from the Ewha Birth and Growth Cohort. We used physical examinations to select prepubertal children ages 7–9 y to exclude the effects of puberty on insulin resistance. The measurements of 25(OH)D levels in all subjects were performed during summer to preclude overestimation of vitamin D deficiency. Additional studies with a longitudinal design are warranted to identify the temporal sequence among these variables and to evaluate the mechanisms underlying vitamin D insufficiency or deficiency, such as parathyroid hormones and lipid profiles. Conclusion Our study found a significant inverse relationship between serum 25(OH)D concentrations and triacylglycerol levels in
1327
prepubertal children that was independent of obesity. Although only one of metabolic component had changed, this finding suggests a key mechanism underpinning the effects of vitamin D on lipid profiles and metabolic syndrome in prepubertal children. These results provide evidence in support of recommendations that prepubertal children consume sufficient amounts of vitamin D. References [1] Muscogiuri G, Mitri J, Mathieu C, Badenhoop K, Tamer G, Orio F, et al. Mechanisms in endocrinology: vitamin D as a potential contributor in endocrine health and disease. Eur J Endocrinol 2014;171:R101–10. [2] Shin YH, Shin HJ, Lee YJ. Vitamin D status and childhood health. Korean J Pediatr 2013;56:417–23. [3] Andiran N, Celik N, Akca H, Dogan G. Vitamin D deficiency in children and adolescents. J Clin Res Pediatr Endocrinol 2012;4:25–9. [4] Huh SY, Gordon CM. Vitamin D deficiency in children and adolescents: epidemiology, impact and treatment. Rev Endocr Metab Disord 2008;9:161–70. [5] Choi HS, Oh HJ, Choi H, Choi WH, Kim JG, Kim KM, et al. Vitamin D insufficiency in Koreada greater threat to the younger generation: The Korea National Health and Nutrition Examination Survey (KNHANES) 2008. J Clin Endocrinol Metab 2011;96:643–51. [6] Delvin EE. Vitamin D status is modestly associated with glycemia and indicators of lipid metabolism in French–Canadian children and adolescents. J Nutr 2010;140:987–91. [7] Shin YH, Kim KE, Lee C, Shin HJ, Kang MS, Lee HR, et al. High prevalence of vitamin D insufficiency or deficiency in young adolescents in Korea. Eur J Pediatr 2012;171:1475–80. [8] Chiu KC, Chu A, Go VL, Saad MF. Hypovitaminosis D is associated with insulin resistance and beta cell dysfunction. Am J Clin Nutr 2004;79:820–5. [9] Isaia G, Giorgino R, Adami S. High prevalence of hypovitaminosis D in female type 2 diabetic population. Diabetes Care 2001;24:1496. [10] Reis JP, von Mühlen D, Miller ER 3rd, Michos ED, Appel LJ. Vitamin D status and cardiometabolic risk factors in the United States adolescent population. Pediatrics 2009;124:e371–9. [11] Michalakis K, Mintziori G, Kaprara A, Tarlatzis BC, Goulis DG. The complex interaction between obesity, metabolic syndrome and reproductive axis: a narrative review. Metabolism 2013;62:457–78. [12] Lee HA, Kim YJ, Lee H, Gwak HS, Park EA, Cho SJ, et al. Association of vitamin D concentrations with adiposity indices among preadolescent children in Korea. J Pediatr Endocrinol Metab 2013;26:849–54. [13] Moon JS, Lee SY, Nam CM, Choi J, Choe B, Seo J, et al. 2007 Korean national growth charts: review of developmental process and an outlook. Korean J Pediatr 2008;51:1–25. [14] Lee SH, Kim SM, Park HS, Choi KM, Cho GJ, Ko BJ, et al. Serum 25hydroxyvitamin D levels, obesity and the metabolic syndrome among Korean children. Nutr Metab Cardiovasc Dis 2013;23:785–91. [15] Pacifico L, Anania C, Osborn JF, Ferraro F, Bonci E, Olivero E, et al. Low 25(OH)D3 levels are associated with total adiposity, metabolic syndrome, and hypertension in Caucasian children and adolescents. Eur J Endocrinol 2011;165:603–11. [16] Pacifico L, Anania C, Martino F, Poggiogalle E, Chiarelli F, Arca M, et al. Management of metabolic syndrome in children and adolescents. Nutr Metab Cardiovasc Dis 2011;21:455–66. [17] Alemzadeh R, Kichler J, Babar G, Calhoun M. Hypovitaminosis D in obese children and adolescents: relationship with adiposity, insulin sensitivity, ethnicity, and season. Metabolism 2008;57:183–91. [18] Rajakumar K, de las Heras J, Chen TC, Lee S, Holick MF, Arslanian SA. Vitamin D status, adiposity, and lipids in black American and Caucasian children. J Clin Endocrinol Metab 2011;96:1560–7. [19] Jorde R, Figenschau Y, Hutchinson M, Emaus N, Grimnes G. High serum 25hydroxyvitamin D concentrations are associated with a favorable serum lipid profile. Eur J Clin Nutr 2010;64:1457–64. [20] Gannage-Yared MH, Chedid R, Khalife S, Azzi E, Zoghbi F, Halaby G. Vitamin D in relation to metabolic risk factors, insulin sensitivity and adiponectin in a young Middle-Eastern population. Eur J Endocrinol 2009;160:965–71. [21] Park HY, Lim YH, Kim JH, Bae S, Oh SY, Hong YC. Association of serum 25-hydroxyvitamin D levels with markers for metabolic syndrome in the elderly: a repeated measure analysis. J Korean Med Sci 2012;27:653–60. [22] Zemel MB, Shi H, Greer B, Dirienzo D, Zemel PC. Regulation of adiposity by dietary calcium. FASEB J 2000;14:1132–8.
Contents lists available at ScienceDirect
Nutrition journal homepage: www.nutritionjrnl.com
Applied nutritional investigation
Relationship of serum 25-Hydroxyvitamin D (25[OH]D) levels and components of metabolic syndrome in prepubertal children Jung Hyun Kwon M.D. a, Seung Eun Lee M.D. a, Hye Ah Lee M.D. b, Young Ju Kim M.D. c, Hwa Young Lee M.D. d, Hye Sun Gwak Pharm.D., Ph.D. e, Eun Ae Park M.D., Ph.D. a, Su Jin Cho M.D., Ph.D. a, Se Young Oh Ph.D. f, Eun Hee Ha M.D., Ph.D. b, Hyesook Park M.D., Ph.D. b, *, Hae Soon Kim M.D., Ph.D. a, ** a
Department of Pediatrics, School of Medicine, Ewha Womans University, Seoul, Korea Department of Preventive Medicine, School of Medicine, Ewha Womans University, Seoul, Korea c Department of Obstetrics and Gynecology, School of Medicine, Ewha Womans University, Seoul, Korea d Department of Anatomy, School of Medicine, Ewha Womans University, Seoul, Korea e College of Pharmacy, Ewha Womans University, Seoul, Korea f Department of Food and Nutrition, Kyung Hee University, Seoul, Korea b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 February 2015 Accepted 30 April 2015
Objective: The relationship between 25-hydroxyvitamin D (25[OH]D) levels and the lipid and metabolic levels of the prepubertal normal population is unclear. Our goals were to investigate the association of serum 25(OH)D concentrations with lipid and metabolic levels in Korean prepubertal children ages 7–9 y. Methods: We followed 205 children, ages 7–9 y in the Ewha Birth and Growth Cohort study, a prospective cohort sample established in 2001–2006, from July to August 2011. We studied the association of serum 25-hydroxyvitamin D (25[OH]D) levels with components of metabolic syndrome and insulin resistance indices using multivariate regression analysis adjusted for body mass index (BMI) z-scores. Results: The mean age of the 205 subjects was 7.89 0.85 y, and the sample included 109 boys (53.2%). The average 25(OH)D levels of all participants was 25.0 5.4 ng/mL. After adjustment for age and sex, triacylglycerol levels were significantly associated with 25(OH)D (b ¼ –0.02, P ¼ 0.02) concentrations even after adjustment for BMI z-scores (b ¼ 0.02, P ¼ 0.04). However, other metabolic components were not correlated with 25(OH)D status. Those with the lowest quartile of 25(OH)D levels had the highest serum triacylglycerol levels (P ¼ 0.04, Ptrend ¼ 0.01). Conclusions: We found that serum 25(OH)D levels were negatively associated with serum triacylglycerol levels, even independently of adiposity, in prepubertal children. This study suggests that 25(OH)D insufficiency is related to metabolic syndrome via the derangement of triacylglycerol metabolism. Ó 2015 Elsevier Inc. All rights reserved.
Keywords: Vitamin D Dyslipidemia Children
Introduction Recent literature has suggested an association between vitamin D deficiency and obesity and metabolic syndrome in addition to its well-known effects on calcium homeostasis and
This work was supported by National Research Foundation of Korea Grant funded by the Korean Government (2010-0026225). * Corresponding author. Tel.: þ82.2-2650-5275; fax: þ82.2-2650-5569. ** Co-corresponding author. Tel.: þ82.2-2650-5275; fax: þ82.2-2650-5569. E-mail addresses: [email protected] (H. Park), [email protected] (H. S. Kim). http://dx.doi.org/10.1016/j.nut.2015.04.023 0899-9007/Ó 2015 Elsevier Inc. All rights reserved.
bone metabolism [1,2]. Vitamin D deficiency is becoming a widespread global problem, and it has been estimated that up to 1 billion people may have insufficient 25-hydroxyvitamin (OH)D levels (25[OH]D) [3,4]. Indeed, vitamin D deficiency-related obesity and metabolic syndrome are important public health problems in Korea. A study based on the Fourth Korea National Health and Nutrition Examination Survey (KNHANES IV) determined that the serum 25 (OH)D levels of 93% of males and 96% of females ages 10–19 y were insufficient [5]. Vitamin D deficiency is thought to be associated with lipid metabolism. However, studies regarding the association
J. H. Kwon et al. / Nutrition 31 (2015) 1324–1327
1325
between vitamin D and serum lipid levels have reported a variety of results. A cross-sectional study of children and adolescents conducted by Delvin et al. reported that serum 25(OH)D levels were positively correlated with total cholesterol and triacylglycerol levels in females [6], and Shin et al. reported an inverse relationship between serum 25(OH)D levels and triacylglycerol and low-density lipoprotein cholesterol (LDL-c) levels in Korean adolescents [7]. Additionally, it has been proposed that, independent of obesity, low serum 25(OH)D concentration is a risk factor for the development of glucose intolerance and type 2 diabetes because it reduces insulin secretion and sensitivity in adults [8,9]. A recent study in adolescents identified a negative association between 25(OH)D levels and serum glucose and triacylglycerol levels and a positive association between 25(OH)D levels and serum high-density lipoprotein cholesterol (HDL-c) levels, independent of weight [10]. Furthermore, as there is a complex interaction between metabolic syndrome and the reproductive axis [11], prepubertal lipid and glucose metabolism differs from that of pubertal individuals and adults, but studies demonstrating relationships among vitamin D deficiency and lipid and metabolic levels in Asian prepubertal children are rare. We already reported that 25(OH)D concentrations were inversely related to adiposity indices in the prepubertal children ages 7–9 y who participated in the Ewha Birth and Growth Cohort study [12]. In this follow-up study among prepubertal children in Korea, we aimed to investigate whether serum 25(OH)D concentrations were associated with lipid and metabolic levels as well as with insulin resistance indices, independent of adiposity.
total cholesterol/5.0). Serum insulin was measured to enable use of an immunoradiometric assay kit (Biosource Europe, Nivelles, Belgium). Insulin resistance indices were analyzed to enable use of the homeostasis model assessment of insulin resistance (HOMA-IR); (plasma glucose [mmol/L] insulin [mIU/mL])/22.5 and the glucose-to-insulin ration (GIR).
Methods
Results
Study population In this study, our goal was to examine the relationship between serum 25(OH)D level and components of metabolic syndrome in Korean preadolescent children, by extending the range of the prior study [12]. This study was performed as part of the Ewha Birth and Growth Cohort study, which involves an ongoing long-term research cohort established in 2001–2006. When participants reached 7–9 y of age, we collected data from July to August 2011. During this period, both general physical measurements and blood samples were taken. A total of 205 children with complete data were included in the present study (109 males and 96 females). The parents or guardians of the children provided consent for their child’s participation in this study. This study was approved by the Institutional Review Board of Ewha Woman’s University Hospital. Data collection We used body mass index (BMI), calculated by dividing weight by the square of height, as an adiposity indicator. A detailed description of the methods used to measure height and weight were presented in a previous publication [12]. We calculated BMI z-scores based on the 2007 Korean Children and Adolescents Growth Standard. Blood pressure (BP) was measured using an automatic device (Critikon, Tampa, FL, USA) using the correct cuff size after participants rested in a stable position with their arm properly supported. Two measurements, taken within 5 min of each other, were averaged. High BP was defined as systolic blood pressure (SBP) and diastolic blood pressure (DBP) higher than the 90th percentile of the age- and sex-specific norms in the 2007 Korean Children and Adolescents Growth Standard [13]. All blood samples were collected after at least 8 h of fasting and immediately stored in 70 C freezers until analysis. The 25(OH)D blood concentration (ng/ mL), an indicator of vitamin D levels, was measured using a radioimmunoassay (DiaSorin/Stillwater, MN, USA) with a Gamma counter (Wallac 1470 Wizard, Perkin Elmer, Waltham, MA, USA). The mean and standard deviation of serum 25(OH)D levels was 25.0 5.4 ng/mL (males, 23.2 4.4 ng/mL; females, 27.0 5.9 ng/mL). The total cholesterol, triacylglycerol, HDL cholesterol, and glucose levels of each child were measured using an automatic analyzer (Model: 7180, Hitachi, Tokyo, Japan), and LDL cholesterol levels (mg/dL) were calculated using the Friedewald formula equations (LDL cholesterol ¼ cholesterol–HDL cholesterol–
Statistical analysis All statistical analyses were conducted using SAS version 9.3 (SAS Institutes, Cary, NC, USA). Descriptive analyses presented the distributions of numeric variables as means and standard deviations and those of categorical variables as numbers and percentages. t tests were used to assess mean differences in 25(OH) D concentrations according to sex, overweight status, dietary supplement intake, and metabolic components. We assessed the magnitude of the association between 25(OH)D levels and metabolic components using analysis of covariance (ANCOVA) and multiple regression analysis. Because the distributions of triacylglycerol and insulin levels did not satisfy the statistical assumption of normality, these values were log transformed. To investigate the mean differences in metabolic components and insulin resistance indices across serum 25(OH)D levels, we used an ANCOVA controlling for age, sex, and BMI z-scores. Only 15.6% of study participants reached an optimal level of 25(OH)D, as recommended by the Institute of Medicine (IOM) (30 ng/mL). Thus, we categorized the data into quartiles according to 25(OH)D concentration (Q1 group: 25[OH]D < 21.4 ng/mL, Q2 group: 21.4–23.9 ng/mL, Q3 group: 23.9–27.9 ng/mL, Q4 group: >27.9 ng/mL) to increase the statistical power. According to the multiple regression analysis, sex, age, birth order (first, second, and third or more), maternal education level (graduated from high school, college or university), and fruit/fruit juice intake (always, generally, and seldom; treated as dummy variables) were considered as confounding factors. These confounding factors, which had a significance of <0.3 according to the descriptive analysis of previous publications, were selected for inclusion in our analyses [12]. Additionally, BMI z-scores were considered because of the strong association of metabolic components with obesity. All analyses were two-tailed, and a P value of <0.05 was considered to indicate statistical significance.
The clinical characteristics, insulin resistance indices, and lipid profiles of the prepubertal children included in the present study are presented in Table 1. A total of 205 children ages 7.89 0.85 y, including 109 males (53.2%), were enrolled. A total of 27 (13.2%) were obese, and 27 (13.3%) had high BP. No children reached a glucose intolerance state, a fasting glucose level higher than 100 mg/dL. The mean differences in metabolic components and insulin resistance indices by 25(OH)D levels are presented in Table 2. The highest serum 25(OH)D quartile showed significantly lower serum triacylglycerol levels and a strong inverse relationship between 25(OH)D and triacylglycerol levels (P ¼ 0.04, Ptrend ¼ 0.01). Those in the lowest 25(OH)D quartile showed higher mean total cholesterol and LDL cholesterol levels, LDL/ HDL ratios, SBP, DBP, log-transformed insulin levels, glucose levels, and HOMA values and lower HDL cholesterol levels, but these relationships did not reach significance. Table 3 presents the results of the multiple regression analysis of the relationship between 25(OH)D concentrations and insulin resistance indices controlling for confounding factors. When controlling for sex and age (Model 1), only the relationship involving triacylglycerol levels reached statistical significance (b ¼ 0.02, P ¼ 0.02). In addition, triacylglycerol levels remained significantly associated with 25(OH)D concentrations even after adjusting for BMI z-scores, age, sex, birth order, fruit/fruit juice intake, and maternal educational level (b ¼ 0.02, P ¼ 0.04 in Model 2). However, no associations among other lipid markers, except triacylglycerol levels (total cholesterol, LDL cholesterol, and HDL cholesterol) and insulin resistance indices, and 25(OH)D concentrations were detected in the total sample.
1326
J. H. Kwon et al. / Nutrition 31 (2015) 1324–1327
Table 1 Characteristics, geometric mean hormone levels, and 95% confidence intervals of 205 prepubertal children ages 7–9 y in the Ewha Birth and Growth Cohort in 2001–2006 Characteristic
Mean (SD) or Geometric mean (95% confidence interval)
Age (years) Sex, % Boy (number) Girl (number) Anthropometrics Mean BMI z-score Overweight/obese, (%)
7.89 (0.85) 109 (53.2%) 96 (46.8%) 0.25 (1.22) 27 (13.2) Boy 17 (8.3); Girl 10 (4.9)
Blood pressure Systolic blood pressure, mmHg Diastolic blood pressure, mmHg High blood pressure, (%) Glycometabolism-related measure Fasting glucose, mg/dL Insulin, mIU/mL HOMA GIR Lipids-related measure HDL cholesterol, mg/dL LDL cholesterol, mg/dL TG, mg/dL
101.71 (9.26) 59.22 (6.84) 27 (13.3) 77.75 7.86 1.61 10.28
(6.65) (1.02); 7.55 (6.37–9.31) (0.71) (2.85)
59.98 (11.19) 87.24 (19.16) 64.65 (1.04); 64.00 (44.00–94.00)
GIR, glucose-to-insulin ratio; HDL-c, high-density lipoprotein cholesterol; HOMA, homeostasis model assessment of insulin resistance; LDL-c, low-density lipoprotein cholesterol; TG, triacylglycerol High blood pressure was defined as systolic or diastolic blood pressure 90th percentile for age and sex based on the 2007 Korean Children and Adolescents Growth Standard Overweight/obese was defined as a body mass index 85th percentile for age and sex based on the 2007 Korean Children and Adolescents Growth Standard LDL-c (mg/dL) was calculated using the Friedewald formula equations (LDL-c ¼ (TC–HDL-c–TG)/5.0)
Discussion We found a strong inverse relationship between 25(OH)D and serum levels in prepubertal children after adjusting for sex and age. Furthermore, the association remained significant after adjustment for adiposity and potential confounders. In particular, those in the lowest 25(OH)D quartile tended to have the highest levels of triacylglycerol. These results are consistent with other findings of a negative association with elevated triacylglycerol levels [14–16]. Lee et al. found that vitamin D deficiency in 9-year-old Korean children was associated with obesity,
hypertension, hyperlipidemia, and hyperglycemia, but that study did not address prepubertal status [14]. Several studies involving adolescents and adults have supported the hypothesis that poor 25(OH)D status is associated with obesity, metabolic syndrome, and insulin resistance [10,17]. Alemzadeh et al. reported a negative association between 25(OH)D and body fat mass in obese children and adolescents, and vitamin D-deficient subjects showed reduced insulin sensitivity compared with vitamin D-sufficient subjects [17]. Therefore we expected that insulin resistance would show the inverse correlation with 25(OH)D levels in our prepubertal children. Those in the lowest 25(OH)D quartile tended to show higher log-transformed insulin, glucose, and HOMA values, although these differences were not significant. Because of the small sample, our results are subject to several limitations, and additional larger longitudinal studies regarding the insulin resistance of prepubertal children are needed. We also did not assess the all elements of metabolic syndrome, because there is no definition of metabolic syndrome in children and none of our subjects reached fasting glucose levels higher than 100 mg/dL. Although only one of the metabolic components changed meaningfully in our study, we might presume that changes associated with 25(OH)D are precipitated from prepubertal period. A study of 237 black American and Caucasian children with an average age of 12.7 2.2 y found that 25(OH)D levels were inversely correlated with adiposity and positively correlated with HDL-c levels [18]. A study among American adolescents found inverse relationships between 25(OH)D and serum glucose and triacylglycerol levels and a positive relationship between this variable and HDL-c levels [10]. However, our results showed only a weak correlation between serum 25(OH)D and HDL-c levels. The association between serum 25(OH)D and lipid profiles has been discussed in many other studies, but the results remain controversial. One study among elderly individuals reported a strong positive relationship between serum 25(OH)D and HDL-c levels [19] and another reported no notable relationship between serum 25(OH)D and triacylglycerol levels [20]. Chiu et al. [8] found an inverse relationship between serum 25(OH)D levels and both total cholesterol and LDL-c levels, whereas no relationship between serum 25(OH)D and triacylglycerol levels was observed. Another study found that lower 25(OH)D concentrations were correlated with higher insulin resistance,
Table 2 Metabolic components and insulin resistance indices in relation to serum 25(OH)D quartiles Metabolic components
Quartile of serum 25(OH)D Q2
Q3
Q4
(<21.4 ng/mL)
(21.4–23.9 ng/mL)
(23.9–27.9 ng/mL)
(>27.9 ng/mL)
(n ¼ 51) TC (mg/dL) Log triacylglycerol HDL-c (mg/dL) LDL-c (mg/dL) LDL/HDL ratio SBP (mmHg) DBP (mmHg) Log insulin Glucose (mg/dL) HOMA GIR
P
Q1
163.1 4.3 57.8 88.0 1.57 103.4 60.1 2.1 79.1 1.7 10.1
3.39 0.07 1.55 2.81 0.06 1.26 0.98 0.04 0.95 0.09 0.37
(n ¼ 51)
(n ¼ 51)
(n ¼ 52)
161.37 4.16 59.89 86.75 1.49 100.83 58.80 2.03 76.09 1.50 10.16
162.83 4.11 61.71 87.16 1.44 101.03 58.43 2.04 77.45 1.57 10.49
160.67 4.07 60.43 87.02 1.47 101.64 59.54 2.07 78.34 1.65 10.38
(3.32) (0.07) (1.52) (2.76) (0.05) (1.23) (0.96) (0.04) (0.93) (0.09) (0.37)
(3.31) (0.07) (1.52) (2.75) (0.05) (1.21) (0.95) (0.04) (0.93) (0.09) (0.37)
(3.37) (0.07) (1.54) (2.80) (0.06) (1.24) (0.97) (0.04) (0.94) (0.09) (0.37)
0.95 0.04* 0.38 0.99 0.43 0.46 0.60 0.57 0.12 0.30 0.87
DBP, diastolic blood pressure; GIR, glucose-to-insulin ratio; HDL-c, high-density lipoprotein cholesterol; HOMA, homeostasis model assessment of insulin resistance; LDL-c; low-density lipoprotein cholesterol; SBP, systolic blood pressure; TC, total cholesterol Results are presented as least-squares means with standard errors. P values were obtained from ANCOVAs adjusted for age, sex, and BMI z-scores. *P for trend was 0.01
J. H. Kwon et al. / Nutrition 31 (2015) 1324–1327 Table 3 Results of multiple regression of serum 25(OH)D levels on metabolic components and insulin resistance indices Metabolic components
Total-Cholesterol (mg/dL) Log triacylglycerol HDL-Cholesterol (mg/dL) LDL-Cholesterol (mg/dL) LDL/HDL ratio SBP (mmHg) DBP (mmHg) Glucose (mg/dL) Log insulin HOMA GIR
Model 1
Model 2
b (P value)
b (P value)
0.06 0.02 0.07 0.23 0.002 0.16 0.06 0.03 0.004 0.01 0.06
0.03 0.02 0.06 0.19 0.002 0.08 0.01 0.03 0.003 0.002 0.05
(0.85) (0.02) (0.66) (0.41) (0.71) (0.20) (0.53) (0.73) (0.31) (0.52) (0.14)
(0.92) (0.04) (0.71) (0.51) (0.78) (0.54) (0.95) (0.73) (0.44) (0.87) (0.19)
DBP, diastolic blood pressure; GIR, glucose-to-insulin ratio; HOMA, homeostasis model assessment of insulin resistance; SBP, systolic blood pressure P value was obtained from multiple linear regression Model 1: Adjusted for age and sex Model 2: Model 1 þ body mass index z-score, age, sex, birth order, fruit/fruit juice intake, and maternal educational level
triacylglycerol concentrations, and increments in blood pressure among elderly Koreans and related the risk for hypertriglyceridemia to vitamin D deficiency [21]. It has been suggested that the following mechanisms underlie the effect of 25(OH)D on lipid profiles. First, it has been proposed that vitamin D inhibits the secretion of parathyroid hormones, which have been known to reduce lipolysis [1]. It has also been proposed that a vitamin D-induced increment in calcium absorption, that is, 25(OH)D, increases calcium levels, which tends to reduce the formation of calcium-fatty soaps in the gut and therefore to increase fat absorption and reduce hepatic triacylglycerol formation and secretion [18,22]. Serum 25(OH)D concentrations are largely determined by environmental factors, primarily vitamin D intake (cholecalciferol and ergocalciferol) and the ultraviolet radiation of 7-hydro cholesterol in the skin [1]. Although we could not investigate the daily outside activities of children, we measured 25(OH)D levels, which reflect the vitamin D nutrient state within the previous 3 wk, and controlled for the intake of dietary supplements. Most studies of vitamin D deficiency in prepubertal children or adults have focused on populations who were obese. Very few previous studies have found an association of vitamin D deficiency with hyperlipidemia and metabolic risk in a general sample of prepubertal children. One strength of this study is that the sample included normal children from the Ewha Birth and Growth Cohort. We used physical examinations to select prepubertal children ages 7–9 y to exclude the effects of puberty on insulin resistance. The measurements of 25(OH)D levels in all subjects were performed during summer to preclude overestimation of vitamin D deficiency. Additional studies with a longitudinal design are warranted to identify the temporal sequence among these variables and to evaluate the mechanisms underlying vitamin D insufficiency or deficiency, such as parathyroid hormones and lipid profiles. Conclusion Our study found a significant inverse relationship between serum 25(OH)D concentrations and triacylglycerol levels in
1327
prepubertal children that was independent of obesity. Although only one of metabolic component had changed, this finding suggests a key mechanism underpinning the effects of vitamin D on lipid profiles and metabolic syndrome in prepubertal children. These results provide evidence in support of recommendations that prepubertal children consume sufficient amounts of vitamin D. References [1] Muscogiuri G, Mitri J, Mathieu C, Badenhoop K, Tamer G, Orio F, et al. Mechanisms in endocrinology: vitamin D as a potential contributor in endocrine health and disease. Eur J Endocrinol 2014;171:R101–10. [2] Shin YH, Shin HJ, Lee YJ. Vitamin D status and childhood health. Korean J Pediatr 2013;56:417–23. [3] Andiran N, Celik N, Akca H, Dogan G. Vitamin D deficiency in children and adolescents. J Clin Res Pediatr Endocrinol 2012;4:25–9. [4] Huh SY, Gordon CM. Vitamin D deficiency in children and adolescents: epidemiology, impact and treatment. Rev Endocr Metab Disord 2008;9:161–70. [5] Choi HS, Oh HJ, Choi H, Choi WH, Kim JG, Kim KM, et al. Vitamin D insufficiency in Koreada greater threat to the younger generation: The Korea National Health and Nutrition Examination Survey (KNHANES) 2008. J Clin Endocrinol Metab 2011;96:643–51. [6] Delvin EE. Vitamin D status is modestly associated with glycemia and indicators of lipid metabolism in French–Canadian children and adolescents. J Nutr 2010;140:987–91. [7] Shin YH, Kim KE, Lee C, Shin HJ, Kang MS, Lee HR, et al. High prevalence of vitamin D insufficiency or deficiency in young adolescents in Korea. Eur J Pediatr 2012;171:1475–80. [8] Chiu KC, Chu A, Go VL, Saad MF. Hypovitaminosis D is associated with insulin resistance and beta cell dysfunction. Am J Clin Nutr 2004;79:820–5. [9] Isaia G, Giorgino R, Adami S. High prevalence of hypovitaminosis D in female type 2 diabetic population. Diabetes Care 2001;24:1496. [10] Reis JP, von Mühlen D, Miller ER 3rd, Michos ED, Appel LJ. Vitamin D status and cardiometabolic risk factors in the United States adolescent population. Pediatrics 2009;124:e371–9. [11] Michalakis K, Mintziori G, Kaprara A, Tarlatzis BC, Goulis DG. The complex interaction between obesity, metabolic syndrome and reproductive axis: a narrative review. Metabolism 2013;62:457–78. [12] Lee HA, Kim YJ, Lee H, Gwak HS, Park EA, Cho SJ, et al. Association of vitamin D concentrations with adiposity indices among preadolescent children in Korea. J Pediatr Endocrinol Metab 2013;26:849–54. [13] Moon JS, Lee SY, Nam CM, Choi J, Choe B, Seo J, et al. 2007 Korean national growth charts: review of developmental process and an outlook. Korean J Pediatr 2008;51:1–25. [14] Lee SH, Kim SM, Park HS, Choi KM, Cho GJ, Ko BJ, et al. Serum 25hydroxyvitamin D levels, obesity and the metabolic syndrome among Korean children. Nutr Metab Cardiovasc Dis 2013;23:785–91. [15] Pacifico L, Anania C, Osborn JF, Ferraro F, Bonci E, Olivero E, et al. Low 25(OH)D3 levels are associated with total adiposity, metabolic syndrome, and hypertension in Caucasian children and adolescents. Eur J Endocrinol 2011;165:603–11. [16] Pacifico L, Anania C, Martino F, Poggiogalle E, Chiarelli F, Arca M, et al. Management of metabolic syndrome in children and adolescents. Nutr Metab Cardiovasc Dis 2011;21:455–66. [17] Alemzadeh R, Kichler J, Babar G, Calhoun M. Hypovitaminosis D in obese children and adolescents: relationship with adiposity, insulin sensitivity, ethnicity, and season. Metabolism 2008;57:183–91. [18] Rajakumar K, de las Heras J, Chen TC, Lee S, Holick MF, Arslanian SA. Vitamin D status, adiposity, and lipids in black American and Caucasian children. J Clin Endocrinol Metab 2011;96:1560–7. [19] Jorde R, Figenschau Y, Hutchinson M, Emaus N, Grimnes G. High serum 25hydroxyvitamin D concentrations are associated with a favorable serum lipid profile. Eur J Clin Nutr 2010;64:1457–64. [20] Gannage-Yared MH, Chedid R, Khalife S, Azzi E, Zoghbi F, Halaby G. Vitamin D in relation to metabolic risk factors, insulin sensitivity and adiponectin in a young Middle-Eastern population. Eur J Endocrinol 2009;160:965–71. [21] Park HY, Lim YH, Kim JH, Bae S, Oh SY, Hong YC. Association of serum 25-hydroxyvitamin D levels with markers for metabolic syndrome in the elderly: a repeated measure analysis. J Korean Med Sci 2012;27:653–60. [22] Zemel MB, Shi H, Greer B, Dirienzo D, Zemel PC. Regulation of adiposity by dietary calcium. FASEB J 2000;14:1132–8.