OBJECTIVE—The metabolic syndrome is associated with a high incidence of cardiovascular disease even when the abnormalities present in the syndrome are mild. The underlying mechanism of the metabolic syndrome has not been elucidated. We investigated whether a strong atherogenic lipoprotein, remnant-like particle (RLP) lipoprotein cholesterol, is elevated in the metabolic syndrome.
RESEARCH DESIGN AND METHODS—We performed a health examination among the residents of a rural community in Japan. Complete datasets, including fasting RLP cholesterol levels, were obtained in 1,261 subjects (509 men and 752 women) without diabetes and who were not taking lipid-lowering drugs. The subjects’ medical history, use of alcohol, and smoking habits were ascertained by a questionnaire.
RESULTS—All of the components of the metabolic syndrome were significantly related to RLP cholesterol by univariate analysis. Total cholesterol and smoking habits were also positively associated with RLP cholesterol. The subjects with the metabolic syndrome showed only mild abnormalities of each component. When RLP cholesterol levels were stratified by the number of the components of the metabolic syndrome, there was a strong association between RLP cholesterol levels and the number of components (P < 0.001 and F = 72.7).
CONCLUSIONS—RLP cholesterol levels are elevated in the metabolic syndrome, and this elevation may underlie the high incidence of cardiovascular disease in the metabolic syndrome.
It is well known that the metabolic syndrome is associated with a very high incidence of cardiovascular disease even when the respective components of the syndrome are only mildly abnormal (1,2). The underlying mechanism has not been elucidated. Remnant-like lipoprotein particle (RLP) cholesterol derived from VLDL and/or chylomicron has been progressively recognized as a stronger atherogenic factor than cholesterol. Therefore, RLP cholesterol has been highlighted as a novel risk factor for coronary artery disease (3) and sudden cardiac death (4). Although it is known that RLP cholesterol levels increase in diabetic patients (5), the level of RLP cholesterol in the metabolic syndrome without diabetes has not been reported. We hypothesized that RLP cholesterol is elevated in the metabolic syndrome and that this elevation may be responsible for the association between the metabolic syndrome and a high incidence of cardiovascular disease (1,2). Accordingly, we studied a large group of rural residents and measured RLP cholesterol in subjects with the metabolic syndrome but without diabetes.
RESEARCH DESIGN AND METHODS
We performed a health examination in 1,999 individuals in a farming community of southwestern Japan (Tanushimaru town). Invitations to participate in the health examinations were mailed to all of the residents in the area. Of the total population 40 years of age, we examined 48.2% of the men and 62.0% of the women. Thus, a total of 1,492 people received a health examination (6).
We excluded 231 subjects from the subsequent analyses. Of these, 122 subjects had diabetes with a fasting plasma glucose concentration of ≥126 mg/dl and/or HbA1c (A1C) levels of ≥6.0%, 40 subjects were taking an oral hypoglycemic agent or receiving insulin injection, 71 subjects were taking lipid-lowering drugs, and 27 subjects were missing RLP cholesterol values. Finally, a complete dataset was available for 1,261 subjects (509 men and 752 women; mean age 62.3 years).
Data collection
The subjects’ medical history, use of alcohol, and smoking habits were ascertained by a questionnaire. Alcohol intake and smoking were classified as current habitual use or not. Height and weight were measured, and BMI was calculated as weight in kilograms divided by the square of height in meters as an index of obesity. Waist circumference was measured at the level of the umbilicus in the standing position. Blood pressure was measured in the supine position twice at 3-min intervals using an upright standard sphygmomanometer. Vigorous physical activity and smoking were avoided for at least 30 min before blood pressure measurement. The second blood pressure value with the fifth-phase diastolic pressure was used for analysis.
Blood was drawn from the antecubital vein in the morning after a 12-h fast for determinations of lipids (RLP cholesterol, total cholesterol, triglycerides, and HDL cholesterol), glucose, insulin, A1C, and uric acid levels. Fasting blood samples were centrifuged within 1 h after collection. A homeostasis model assessment (HOMA) index [glucose (mg/dl) × insulin (μU/ml)/405] was calculated from fasting and insulin levels as a marker of insulin resistance (7). Serum RLP cholesterol was measured by an immunoseparation technique (using an immunoaffinity gel containing monoclonal antibodies to human apolipoprotein [apo] B-100 and apo A-I) (8). Intra- and interassay coefficients of variation of RLP cholesterol in the commercially available laboratory (The Kyodo Igaku Laboratory, Fukuoka, Japan) were 7.6 and 7.8%, respectively. All other chemical examinations were measured at the same laboratory.
We defined the metabolic syndrome according to Adult Treatment Panel III (ATP III) guidelines (9). The ATP III identified five components of the metabolic syndrome: abdominal obesity (given as waist circumference [men >101.6 cm, women >88.9 cm]), triglycerides (≥150 mg/dl [≥1.69 mmol/l]), HDL cholesterol (men <40 mg/dl [<1.03 mmol/l], women <50 mg/dl [<1.29 mmol/l[), blood pressure (≥130/≥85 mmHg), and fasting glucose (≥110 mg/dl [≥6.11 mmol/l]). However, Japanese are much smaller than Caucasians; therefore, it is not appropriate to use the ATP III criteria for abdominal obesity (9). Accordingly, we adopted a waist circumference of >85 cm for men and >90 cm for women, as proposed by the Japanese Society for Obesity (10). The metabolic syndrome was defined as the presence of at least three of these components.
This study was approved by the Tanushimaru branch of the Japan Medical Association and by the mayor and the welfare section of the Tanushimaru town office as well as by the ethics committee of Kurume University School of Medicine. All participants gave informed consent.
Statistical methods
Because of skewed distributions, natural logarithmic (ln) transformations were performed for RLP cholesterol, triglycerides, glucose, insulin, and the HOMA index. Log-transformed values were used for the calculation and reconverted to antilogarithm forms. Results are presented as means ± SD. The difference between the two groups was determined by Student’s t test. The χ2 test was used for categorical parameters to test differences between groups. Antihypertensive medication was used as a dummy variable. The RLP cholesterol levels stratified by quintiles were compared using ANCOVA, adjusted for age, sex, and antihypertensive medication as covariates (Table 3). Mean RLP cholesterol levels stratified by the number of the components of the metabolic syndrome were compared using ANCOVA, adjusted for age, sex, and antihypertensive medication (Fig. 1). Statistical significance was defined as P < 0.05. All analyses were performed with the use of the SPSS system (SPSS, Chicago, IL).
RESULTS
Demographic data for 1,261 subjects are presented in Table 1. As is apparent from the table, the enrolled subjects had almost normal mean values of risk factors and other variables. This is the first report of RLP cholesterol levels in a large number of subjects that showed skewed distributions in both sexes.
Table 2 shows the results of univariate analysis performed for correlates of RLP cholesterol levels. Age, BMI, total cholesterol, uric acid, insulin, HOMA index, A1C, and current smoking status were significantly related to RLP cholesterol levels. Age had a negative impact on RLP cholesterol levels. Mean RLP cholesterol levels stratified by age and sex in men were 4.3 ± 2.1 mg/dl at 40–49 years, 4.0 ± 1.9 mg/dl at 50–59 years, 3.0 ± 1.8 mg/dl at 60–69 years, and 3.0 ± 1.7 mg/dl at >70 years (P < 0.05 for trend) and in women were 2.7 ± 1.7 mg/dl at 40–49 years, 3.4 ± 1.6 mg/dl at 50–59 years, 3.3 ± 1.7 mg/dl at 60–69 years, and 3.3 ± 1.6 mg/dl at >70 years (P < 0.05 for trend). At <60 years of age, mean RLP cholesterol levels were higher (P < 0.01) in men than in women, but the difference disappeared at >60 years of age owing to the decline in the number of men and the increase in the number of women. There was a decrease in BMI in old age (BMI 23.2 ± 2.9 kg/m2 at 40–49 years, 24.1 ± 3.0 kg/m2 at 50–59 years, 23.0 ± 2.8 kg/m2 at 60–69 years, and 22.4 ± 2.9 kg/m2 at >70 years; P < 0.001 for trend). The five components of the metabolic syndrome (waist circumference, systolic blood pressure, triglycerides, HDL cholesterol, and glucose) were significantly related to RLP cholesterol levels. Figure 1 shows RLP cholesterol levels stratified by the number of the components of the metabolic syndrome adjusted for age, sex, and antihypertensive medication. There was a strong association between RLP cholesterol levels and the number of components (P < 0.001 and F = 72.7). RLP cholesterol levels were progressively higher (P < 0.0001) as the number of components of the metabolic syndrome increased. In Table 3, the mean values of the components of the metabolic syndrome are shown in relation to the accumulation of components after adjustment for age, sex, and antihypertensive medication. All components showed progressive changes (P < 0.001 for trend) as the number of components increased. However, even in the presence of the metabolic syndrome (component ≥3), the mean values of each component were normal or only slightly abnormal.
CONCLUSIONS
Values of RLP cholesterol and its distribution
In the past it was difficult to measure remnant lipoproteins, because they have heterogeneous properties and because the measurement was time-consuming and complex (11). About 10 years ago, Nakajima et al. (8) developed a simple rapid assay method for RLP cholesterol. Since then some reports describing plasma levels of RLP cholesterol have been published, mostly from Japan, but the number of subjects reported was <250 (3,4,12). Moreover, the subjects enrolled in these studies were heterogeneous and included subjects with cardiovascular disease and diabetes (4,12). It is well known that RLP cholesterol is elevated in such diseases (5). We enrolled a large number (n = 1,261) of relatively healthy subjects without diabetes and not receiving lipid-lowering drugs from among the residents in a rural community. Thus, the reported values of RLP cholesterol in our study represent those of an average Japanese population >40 years old (13). In the present study, the mean levels of RLP cholesterol in men and women were 3.34 and 3.31 mg/dl, respectively. These values were similar to those reported in Japanese control subjects (14) but much lower than those (6.8–7.2 mg/dl for women and 8.0 mg/dl for men) of Caucasians reported in the Framingham Heart Study (15,16). There are several possible reasons for this difference. The first may be race, the second may be diet, and a third may be other demographic differences, such as lower body weight and lower levels of lipid profiles. Although physical activity may be a factor, we do not consider this likely because physical activity in farmers is not so high in Japan because of the widespread use of advanced automated farming machines. It has been reported that RLP cholesterol levels are higher in men than in women (13,16). However, our univariate analysis did not confirm this. RLP cholesterol levels were higher in men than in women at <60 years, but the difference disappeared at >60 years old. Thus, when the impacts of age and sex on RLP cholesterol levels are discussed, one must be careful to specify what age range is being considered.
RLP cholesterol and the metabolic syndrome
In this study, we not only showed a strong, positive association between RLP cholesterol levels and each of the components of the metabolic syndrome but also demonstrated that RLP cholesterol levels rose significantly as the number of components increased. Although high RLP cholesterol levels were reported in type 2 diabetes (5), we excluded diabetes from the present analysis. Elevated RLP cholesterol levels have been found more frequently in individuals with insulin resistance without diabetes (5,12). Moreover, it was reported that insulin resistance was closely related to RLP cholesterol levels in epidemiological studies (11,13). In our analysis, there was a strong association between RLP cholesterol levels and the HOMA index (Table 2), an index of insulin resistance (7). Thus, our study confirmed the previous finding (13). However, insulin resistance is not the sine qua non of the metabolic syndrome. The relationship shown in Fig. 1 between RLP cholesterol and the number of components was very strong even after adjustment for the HOMA index, indicating that insulin resistance was not the primary pathway for the association. Thus, our study is the first to demonstrate a strong association between RLP cholesterol and the metabolic syndrome. Although it was reported from Japan (17) that the risk of cardiovascular events increased in the high RLP cholesterol groups (>4.7–5.1 mg/dl), we cannot comment on this issue because our study was not prospective. Instead, we would like to emphasize that the RLP cholesterol level in the metabolic syndrome was very high at >7.9 mg/dl, compared with the average Japanese value of 3.3 mg/dl. A prospective study is necessary to investigate whether a population with the metabolic syndrome and high RLP cholesterol level (>7.9 mg/dl) will show a significantly increased incidence of cardiovascular events.
In Table 1, we showed that not only the five components of the metabolic syndrome defined by ATP III (9) but also other factors related to the metabolic abnormality (BMI, cholesterol, uric acid, insulin, HOMA index, smoking, and A1C) were significantly associated with RLP cholesterol levels. Thus, RLP cholesterol may be a useful marker of the total risk for cardiovascular disease. It is interesting to note that RLP cholesterol showed a significant association (P < 0.05) with current smoking. At present, the reason for this is not clear. A decrease in clearance of RLP cholesterol in chronic smokers may be responsible for their high levels of RLP cholesterol.
In summary, we reported the mean values and the distribution of RLP cholesterol in a group of relatively healthy people >40 years of age who were residents of a rural community in Japan. The subjects with the metabolic syndrome had very high levels of RLP cholesterol. Given that RLP cholesterol is even more atherogenic than cholesterol, the high level of RLP cholesterol in subjects with the metabolic syndrome may underlie the high incidence of cardiovascular disease/events observed in this syndrome, even when the degree of abnormality of each individual component is mild.
Age-, sex-, and antihypertensive medication–adjusted means of RLP cholesterol stratified by the number of the components of the metabolic syndrome.
Age-, sex-, and antihypertensive medication–adjusted means of RLP cholesterol stratified by the number of the components of the metabolic syndrome.
Demographical data
. | Men . | Women . | Total . |
---|---|---|---|
n | 509 | 752 | 1,261 |
Age (years) | 63.5 ± 10.5 | 61.5 ± 10.7 | 62.3 ± 10.6 |
BMI (kg/m2) | 23.0 ± 2.9 | 22.9 ± 3.2 | 22.9 ± 3.1 |
Waist circumference (cm) | 81.2 ± 8.2 | 73.3 ± 8.3 | 76.5 ± 9.1 |
Systolic blood pressure (mmHg) | 135.1 ± 19.5 | 130.4 ± 21.0 | 132.3 ± 20.5 |
Diastolic blood pressure (mmHg) | 81.1 ± 11.5 | 76.8 ± 11.4 | 78.5 ± 11.4 |
RLP cholesterol (mg/dl)* | 3.34 ± 1.90 | 3.31 ± 1.69 | 3.32 ± 1.78 |
Total cholesterol (mg/dl) | 187.7 ± 31.0 | 205.6 ± 33.9 | 198.4 ± 33.9 |
HDL cholesterol (mg/dl) | 53.3 ± 13.3 | 58.9 ± 13.9 | 56.7 ± 13.9 |
Triglycerides (mg/dl)* | 101.3 ± 62.9 | 88.3 ± 57.1 | 93.3 ± 59.3 |
Uric acid (mg/dl) | 5.8 ± 1.4 | 4.3 ± 1.0 | 4.9 ± 1.4 |
Glucose (mg/dl)* | 94.9 ± 23.6 | 92.3 ± 14.9 | 93.9 ± 19.6 |
Insulin (μU/ml)* | 4.0 ± 1.9 | 4.9 ± 1.8 | 4.5 ± 1.8 |
HOMA (mg/dl × μU/ml)* | 0.99 ± 0.64 | 1.12 ± 0.62 | 1.04 ± 0.63 |
A1C (%) | 5.0 ± 0.4 | 5.1 ± 0.4 | 5.0 ± 0.4 |
Frequency (%) | |||
Hypertension medication | 17.5 | 15.8 | 16.5 |
Alcohol intake | 48.2 | 2.7 | 21.0 |
Current smoking | 37.9 | 1.9 | 16.4 |
. | Men . | Women . | Total . |
---|---|---|---|
n | 509 | 752 | 1,261 |
Age (years) | 63.5 ± 10.5 | 61.5 ± 10.7 | 62.3 ± 10.6 |
BMI (kg/m2) | 23.0 ± 2.9 | 22.9 ± 3.2 | 22.9 ± 3.1 |
Waist circumference (cm) | 81.2 ± 8.2 | 73.3 ± 8.3 | 76.5 ± 9.1 |
Systolic blood pressure (mmHg) | 135.1 ± 19.5 | 130.4 ± 21.0 | 132.3 ± 20.5 |
Diastolic blood pressure (mmHg) | 81.1 ± 11.5 | 76.8 ± 11.4 | 78.5 ± 11.4 |
RLP cholesterol (mg/dl)* | 3.34 ± 1.90 | 3.31 ± 1.69 | 3.32 ± 1.78 |
Total cholesterol (mg/dl) | 187.7 ± 31.0 | 205.6 ± 33.9 | 198.4 ± 33.9 |
HDL cholesterol (mg/dl) | 53.3 ± 13.3 | 58.9 ± 13.9 | 56.7 ± 13.9 |
Triglycerides (mg/dl)* | 101.3 ± 62.9 | 88.3 ± 57.1 | 93.3 ± 59.3 |
Uric acid (mg/dl) | 5.8 ± 1.4 | 4.3 ± 1.0 | 4.9 ± 1.4 |
Glucose (mg/dl)* | 94.9 ± 23.6 | 92.3 ± 14.9 | 93.9 ± 19.6 |
Insulin (μU/ml)* | 4.0 ± 1.9 | 4.9 ± 1.8 | 4.5 ± 1.8 |
HOMA (mg/dl × μU/ml)* | 0.99 ± 0.64 | 1.12 ± 0.62 | 1.04 ± 0.63 |
A1C (%) | 5.0 ± 0.4 | 5.1 ± 0.4 | 5.0 ± 0.4 |
Frequency (%) | |||
Hypertension medication | 17.5 | 15.8 | 16.5 |
Alcohol intake | 48.2 | 2.7 | 21.0 |
Current smoking | 37.9 | 1.9 | 16.4 |
Data are means ± SD, unless indicated otherwise.
Log-transformed values were used.
Association between RLP cholesterol and other parameters by univariate analysis
. | β . | SE . | P value . |
---|---|---|---|
Age | −0.064 | 0.010 | <0.05 |
Sex | −0.047 | 0.210 | 0.095 |
BMI | 0.160 | 0.033 | <0.001 |
Waist circumference | 0.166 | 0.011 | <0.001 |
Systolic blood pressure | 0.043 | 0.006 | <0.05 |
Diastolic blood pressure | 0.033 | 0.009 | 0.168 |
Total cholesterol | 0.351 | 0.003 | <0.001 |
HDL cholesterol | −0.210 | 0.007 | <0.001 |
Triglycerides | 0.786 | 0.001 | <0.001 |
Uric acid | 0.016 | 0.073 | <0.001 |
Glucose | 0.051 | 0.011 | <0.05 |
Insulin | 0.151 | 0.018 | <0.001 |
HOMA index | 0.139 | 0.067 | <0.001 |
A1C | 0.116 | 0.277 | <0.001 |
Alcohol intake | −0.057 | 0.254 | 0.804 |
Current smoking | 0.086 | 0.278 | <0.05 |
Antihypertensive medication | −0.002 | 0.278 | 0.939 |
. | β . | SE . | P value . |
---|---|---|---|
Age | −0.064 | 0.010 | <0.05 |
Sex | −0.047 | 0.210 | 0.095 |
BMI | 0.160 | 0.033 | <0.001 |
Waist circumference | 0.166 | 0.011 | <0.001 |
Systolic blood pressure | 0.043 | 0.006 | <0.05 |
Diastolic blood pressure | 0.033 | 0.009 | 0.168 |
Total cholesterol | 0.351 | 0.003 | <0.001 |
HDL cholesterol | −0.210 | 0.007 | <0.001 |
Triglycerides | 0.786 | 0.001 | <0.001 |
Uric acid | 0.016 | 0.073 | <0.001 |
Glucose | 0.051 | 0.011 | <0.05 |
Insulin | 0.151 | 0.018 | <0.001 |
HOMA index | 0.139 | 0.067 | <0.001 |
A1C | 0.116 | 0.277 | <0.001 |
Alcohol intake | −0.057 | 0.254 | 0.804 |
Current smoking | 0.086 | 0.278 | <0.05 |
Antihypertensive medication | −0.002 | 0.278 | 0.939 |
β indicates standardized regression coefficients: men = 0, women = 1.
Age-, sex-, and antihypertensive medication–adjusted means of components of the metabolic syndrome stratified by the number of the components
. | 0 . | 1 . | 2 . | ≥3 . | P for trend . |
---|---|---|---|---|---|
n | 356 | 503 | 276 | 126 | |
Systolic blood pressure (mmHg) | 117.5 ± 16.5 | 135.8 ± 15.9 | 139.6 ± 15.9 | 143.9 ± 15.9 | <0.001 |
Waist circumference (cm) | 72.1 ± 7.5 | 75.2 ± 7.3 | 80.9 ± 7.2 | 84.6 ± 7.3 | <0.001 |
Triglycerides (mg/dl)* | 74.9 ± 51.3 | 95.0 ± 49.5 | 122.3 ± 49.4 | 193.9 ± 49.6 | <0.001 |
HDL cholesterol (mg/dl) | 64.2 ± 12.7 | 58.3 ± 12.3 | 49.7 ± 12.2 | 44.1 ± 12.3 | <0.001 |
Glucose (mg/dl)* | 91.5 ± 28.9 | 92.9 ± 28.6 | 96.2 ± 28.6 | 99.5 ± 28.7 | <0.001 |
. | 0 . | 1 . | 2 . | ≥3 . | P for trend . |
---|---|---|---|---|---|
n | 356 | 503 | 276 | 126 | |
Systolic blood pressure (mmHg) | 117.5 ± 16.5 | 135.8 ± 15.9 | 139.6 ± 15.9 | 143.9 ± 15.9 | <0.001 |
Waist circumference (cm) | 72.1 ± 7.5 | 75.2 ± 7.3 | 80.9 ± 7.2 | 84.6 ± 7.3 | <0.001 |
Triglycerides (mg/dl)* | 74.9 ± 51.3 | 95.0 ± 49.5 | 122.3 ± 49.4 | 193.9 ± 49.6 | <0.001 |
HDL cholesterol (mg/dl) | 64.2 ± 12.7 | 58.3 ± 12.3 | 49.7 ± 12.2 | 44.1 ± 12.3 | <0.001 |
Glucose (mg/dl)* | 91.5 ± 28.9 | 92.9 ± 28.6 | 96.2 ± 28.6 | 99.5 ± 28.7 | <0.001 |
Data are means ± SD.
Log-transformed values were used.
Article Information
This study was supported in part by the Kimura Memorial Heart Foundation, Fukuoka, and by a grant for Science Frontier Research Promotion Centers from the Ministry of Education, Science, Sports and Culture, Japan.
We are grateful to members of the Japan Medical Association of Ukiha and the elected officials and residents of Tanushimaru town.
References
A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances.