High Plasma Concentration of Remnant Lipoprotein Cholesterol in Obese Children and Adolescents

A total of 135 healthy children and adolescents between the ages of 7 and 16 years volunteered to participate. Study participants were recruited through newspaper advertisements in the Gyoung-gi Province, the nearest province to the Seoul metropolis, and posters were placed at our hospital. This study was approved by the institutional ethics committee, and written informed consent was obtained from the guardians. Pubertal development was assessed by physical examination according to Tanner criteria by an independent pediatrician in our hospital and was confirmed by measurement of plasma testosterone in male subjects and estradiol in female subjects.

In the morning, we measured the weight and height of the subjects, under fasting conditions and wearing light clothes. BMI was calculated as weight in kilograms divided by the square of height in meters, and the percentile of each subject’s BMI according to age and sex was obtained based on the growth chart issued in 1998 by the Korean Pediatrics Society. Waist and hip circumferences were measured by a single investigator using a nonstretchable standard tape measure: the waist measurement was obtained over the unclothed abdomen at the narrowest point between the costal margin and iliac crest, and hip circumference was obtained over light clothing at the level of the widest diameter around the buttocks. Blood pressure was measured twice at 5-min intervals in the sitting position using a standard sphygmomanometer, and the mean values of two tests were used in analyses.

We measured total percent body fat using dual-energy X-ray absortiometry (EXPERT-XL; Lunar, Madison, WI). To assess the distribution of body fat in the abdomen, we performed computed tomography (High Speed Advantage; General Electric, Fairfield, CT) at the umbilical level and measured a total abdominal fat area (centimeters squared) equivalent to a Hounsfield unit range of −50 to −250. We determined the visceral and subcutaneous fat area by applying the peritoneum as a boundary and then calculated the visceral-to-subcutaneous fat area ratio. The clinical status of the participants was concealed from the investigator who collected and analyzed these data.

After >8 h of fasting, all the subjects visited the laboratory from 0800 to 1000, and their blood was drawn from the antecubital vein. The plasma was separated from the collected blood and stored at −70°C until analyzed. Glucose, insulin, total cholesterol, triglycerides, HDL cholesterol, RLP cholesterol, apolipoprotein (apo) A-I, and apoB were measured by standard methods. LDL cholesterol was calculated by the Friedewald equation. As a marker of insulin resistance, a homeostasis model assessment (HOMA) index was calculated from fasting insulin and glucose levels: [insulin (μIU/ml) × glucose (mmol/l)]/22.5.

RLP cholesterol was measured by an immunoseparation technique using an immunoaffinity gel containing monoclonal antibodies to human apoB-100 and apoA-I (Jimro II, Otsuka, Japan) (3). The intra- and interassay coefficients of variation for the RLP cholesterol measurements were 7.6 and 7.8%, respectively.

The subjects were divided into two groups based on the BMI percentile: the normal group: <85th percentile and the at-risk-for-overweight and overweight group: ≥85th percentile. Data are presented as means ± SD and n (%). Insulin, triglycerides, HOMA of insulin resistance (HOMA-IR), and RLP cholesterol were log transformed for the analyses.

The clinical characteristics and laboratory findings of the three groups were compared by Student’s t test. We used correlation analysis to examine the relationship between RLP cholesterol and each factor and multiple regression analysis to identify factors affecting RLP cholesterol. HOMA-IR was also categorized into high and normal groups, in which the high HOMA-IR group had values ≥1.97, arbitrarily defined as the 75th percentile of the distribution for the levels in the study patients. We further categorized the subjects into four groups: normal HOMA-IR with normal BMI, high HOMA-IR with normal BMI, high HOMA-IR with normal BMI, and high HOMA-IR with overweight. We compared the mean RLP cholesterol concentration among the four groups by one-way ANOVA, followed by post hoc testing with Tukey’s test. Statistical analyses were performed using SPSS for Windows (v. 11.0; SPSS, Chicago, IL), and P < 0.05 was considered statistically significant.

The mean age of the 135 subjects (67 boys and 68 girls) was 11.6 ± 2.0 years, with no significant difference between the overweight and normal groups (Table 1). The proportion of male subjects was higher in the overweight group than in the normal group (P = 0.003). Tanner stage did not differ significantly between the two groups. Systolic (P = 0.005) and diastolic (P = 0.019) blood pressure were significantly higher in the overweight group than in the normal group. Fasting blood glucose did not differ between the groups, whereas the fasting insulin concentration and HOMA-IR were significantly higher in the overweight compared with the normal group (P < 0.001 for each).

Waist circumference, total body fat percentage, total abdominal fat area, visceral fat area, and subcutaneous fat area differed significantly between the two groups (P < 0.001 for each). However, visceral-to-subcutaneous fat area ratio was lower in the overweight group than in the normal group. The concentrations of total cholesterol, HDL cholesterol, LDL cholesterol, triglyceride, and apoA-I did not differ significantly between the two groups. The apoB concentration was higher in the overweight than in the normal group (P = 0.004). The concentration of RLP cholesterol was significantly higher in the overweight group than in the normal group (P = 0.019), and the proportion of subjects with a RLP cholesterol concentration >7.5 mg/dl was significantly higher in the overweight (46.0%) than in the normal (23.5%) group (P = 0.008; Fig. 1).

The comparison of clinical and biochemical parameters and body fat distribution in the boys and girls is summarized in Table 1. Fasting blood glucose, HOMA-IR, triglycerides, and LDL cholesterol differed significantly between the two groups in the boys (P < 0.05 for each) but not in the girls. In the boys, the proportion of subjects with a RLP cholesterol concentration >7.5 mg/dl was significantly higher in the overweight (57.6%) than in the normal (20.6%) group (P = 0.003; Fig. 1), but this was not the case in the girls (Fig. 1). Moreover, the concentration of RLP cholesterol was significantly higher in the overweight group than in the normal group in the boys (P = 0.016) but not in the girls (Table 1).

The correlation analysis revealed significant correlations between RLP cholesterol concentration and age, sex (boys), BMI, waist circumference, systolic blood pressure, diastolic blood pressure, visceral and subcutaneous fat area, visceral-to-subcutaneous fat area ratio, total cholesterol, triglycerides, HDL cholesterol, apoB, and HOMA-IR (Table 2).

Multiple regression analysis was performed to assess which factors were independently associated with RLP cholesterol among age, sex, BMI, blood pressure, visceral fat, insulin resistance, and lipid profiles. The triglyceride concentration was shown to be significantly associated with RLP cholesterol (β = 0.928, P < 0.001; R² = 0.862, P < 0.001). The multiple regression excluding lipid profiles showed that RLP cholesterol was significantly associated with HOMA-IR (β = 0.231, P = 0.007) and systolic blood pressure (β = 0.169, P = 0.046; R² = 0.089, P = 0.002; Table 2). To assess the relationships of RLP cholesterol, HOMA-IR, and obesity, we categorized the subjects into four groups: normal HOMA-IR with normal BMI, high HOMA-IR with normal BMI, high HOMA-IR with normal BMI, and high HOMA-IR with overweight. We compared the mean RLP cholesterol for the four groups. The concentration of RLP cholesterol was significantly higher in the high HOMA-IR with overweight group than in the normal HOMA-IR with normal BMI group (P = 0.045; Fig. 2).

Childhood obesity has become an important issue worldwide. In the U.S., the prevalence of overweight children tripled between 1980 and 2000 (13) and was 17.1% for boys and 14.7% for girls in a 1998 nationwide survey in Korea (14).

We used the percentile of each subject’s BMI according to age and sex based on the growth chart issued in 1998 by the Korean Pediatrics Society as an indicator of childhood obesity. The use of BMI as an indicator of adiposity in youth has some limitations because of differences in individual growth rates and maturity levels (15). Moreover, in adults, the waist-to-hip ratio has been the most extensively used indirect measure of visceral fat, and the INTERHEART study (16) revealed that the waist-to-hip ratio is a more important predictor of myocardial infarction than BMI. However, in children, no correlation exists between the waist-to-hip ratio and visceral fat, and adiposity in both boys and girls is predominantly subcutaneous (17). For this reason, most studies on childhood obesity have used BMI as an obesity indicator.

Childhood obesity has been reported to cause insulin resistance, type 2 diabetes, and hypertension (8,18). Our study showed that blood pressure, HOMA-IR, and fasting insulin levels were significantly higher in the overweight group with BMI values at and above the 85th percentile compared with the normal group. Thus, our study confirmed previous findings (8,18) that overweight and obese children and adolescents have adverse blood pressure and insulin levels.

Although the triglyceride level has been recognized as an independent risk factor for CAD (19,20), the RLP cholesterol level has recently received attention as an independent risk factor for atherosclerosis and CAD (4,5,21). Many in vitro studies have revealed the role of RLP in atherosclerosis. RLP is easily phagocytized by macrophages and is involved in the formation of foam cells (22). Furthermore, RLP has been shown to promote the aggregation of platelets (23) and to stimulate the adhesion of monocytes to vascular endothelial cells (24).

Masuoka et al. (25) reported that patients with CAD but with a normal cholesterol level had a higher RLP cholesterol level than control subjects, and Fukushima et al. (26) and McNamara et al. (4) stated that RLP cholesterol was associated with CAD, particularly in postmenopausal women. Recently, Satoh et al. (27) demonstrated that RLP cholesterol levels were elevated in metabolic syndrome.

Similarly, our study showed that the RLP cholesterol concentration was significantly higher in those with high BMI values. According to a study on Japanese subjects that defined RLP cholesterol levels >7.5 mg/dl as hyper-RLP cholesterolemia (28), the proportion of those with an RLP cholesterol level >7.5 mg/dl in the high-BMI group was about twice that of the other groups.

These findings were more prominent in male children, but RLP cholesterol did not differ significantly between overweight and normal weight female children. One possible explanation for this difference could be related to the fact that the female study subjects showed a significantly larger subcutaneous fat area than the male subjects (data not shown). In female subjects, adipose tissue accumulates subcutaneously. In male subjects, fat typically accumulates in the upper segment of the body, both subcutaneously and viscerally (29). Sex-related patterns of body fat deposition become established during puberty (30). We could not evaluate the relationship between sex and RLP cholesterol before and after puberty. Further investigation is required for this purpose. Moreover, this study did not show prominent differences in lipid profiles, except RLP cholesterol, in contrast to those seen in adults. This might be partly related to the larger subcutaneous fat deposits in children compared with those in adults. These findings support an increased RLP cholesterol level in early childhood and adolescence in the presence of obesity. In this study, the triglyceride level did not differ significantly between the overweight and normal groups, and the mean triglyceride level in the overweight subjects was much lower than the criterion for hypertriglyceridemia in adults. However, RLP cholesterol was significantly higher in the overweight subjects, and the mean level of RLP cholesterol in the overweight group was higher than the criterion for hyper-RLP cholesterolemia in adults. These findings support an increased RLP cholesterol level in early childhood and adolescence in the presence of obesity and may suggest a clinical usefulness of RLP cholesterol in childhood obesity.

Controversy exists regarding the relationship between RLP cholesterol and CAD. In a large, 17-year follow-up prospective study, RLP cholesterol levels were significantly related to the incidence of CAD, but in models that simultaneously included RLP and triglycerides, neither variable was significant when adjusted for the other (31). However, several cross-sectional studies (4,5,25,26) have reported that RLP cholesterol was an independent risk factor for CAD. In several prospective studies, RLP cholesterol was shown to be a risk factor, independently predicting the further development of CAD in patients with existing CAD (5,26). More prospective studies on the usefulness of RLP cholesterol in predicting CAD independently of triglycerides are needed.

Several studies on adults have shown the association between insulin resistance and RLP cholesterol (32,33), and our study confirmed the same association in children and adolescents. Furthermore, we revealed that subjects with insulin resistance and obesity have significantly higher RLP cholesterol than those with no insulin resistance and normal weight. Insulin resistance has been shown to increase VLDL synthesis in the liver, decrease lipoprotein lipase activity, and increase RLP (34).

This study was performed on a small number of volunteers recruited from one of the provinces of South Korea and thus does not represent all children and adolescents. In addition, the association of RLP cholesterol with obesity before and after puberty could not be evaluated. However, our study is the first to show that RLP cholesterol was significantly higher in obese children and adolescents and that insulin resistance was strongly related to RLP cholesterol in obese children and adolescents. Long-term prospective studies are required to examine whether weight reduction in obese children could decrease the concentration of RLP cholesterol and whether children and adolescents with high RLP cholesterol are at greater risk of developing metabolic syndrome.

This study was supported by a grant of the Korean Health 21 Research and Development Project, Ministry of Health and Welfare, Republic of Korea (A050463).

We thank Korea Otsuka Pharmaceutical and Professor Jung Ho Kim in the Department of Laboratory Medicine at Yonsei University College of Medicine for helping us to conduct this study.