Participants were recruited at the time of their annual health examination. The sample population compromised 3,164 middle-aged to elderly residents. Information on medical history, present conditions, and drugs were obtained by interview. Other characteristics, e.g., smoking and alcohol habits, and medication, were investigated by individual interviews using a structured questionnaire. Subjects taking medications for hypertension, diabetes, or dyslipidemia were excluded. The final study sample included 614 men and 779 women. This study was approved by the ethics committee of Ehime University School of Medicine, and written informed consent was obtained from each subject.

Information on demographic characteristics and risk factors was collected using clinical files. Body mass index (BMI) was calculated by dividing weight (in kilograms) by the square of the height (in meters). We measured blood pressure in the right upper arm of participants in a sedentary position using an automatic oscillometric blood pressure recorder (BP-103i; Colin, Aichi, Japan) while the subjects were seated after having rested for at least 5 min. Smoking status was defined as the number of cigarette packs per day multiplied by the number of years smoked (pack · year), and the participants were classified into never smokers, past smokers, light smokers (<30 pack · year) and heavy smokers (≥30 pack · year). The daily alcohol consumption was measured using the Japanese liquor unit in which a unit corresponds to 22.9 g of ethanol, and the participants were classified into never drinkers, occasional drinkers (<1 unit/day), light drinkers (1-1.9 unit/day), and heavy drinkers (≥2 unit/day). T-C, TG, HDL-C, fasting blood glucose (FBG), creatinine (enzymatic method), immunoreactive insulin (IRI), and serum HMW adiponectin (FUJIREBIO, Tokyo, Japan) were measured during fasting. LDL-C levels were calculated using the Friedewald formula [¹⁷]. Participants with TG levels ≥400 mg/dl were excluded. Homeostasis model assessment of insulin resistance (HOMA-IR) was calculated from FBG and IRI levels using the following formula; {FBG (mg/dL) × IRI (mU/mL)}/405 [¹⁸].

We applied condition-specific cutoff points for MetS based on the modified guidelines for the diagnosis of MetS in Japan [¹⁹]. Metabolic syndrome was defined as obesity with at least two of the following three conditions: hypertension, dyslipidemia, and impaired fasting glucose (IFG). Obesity was defined as a BMI of ≥25.0 kg/m² [²⁰]. Hypertension was defined as systolic blood pressure (SBP) ≥130 mmHg and diastolic blood pressure (DBP) ≥85 mmHg. Dyslipidemia was defined as TG concentrations ≥150 mg/dL and low HDL cholesterolemia (HDL-C <40 mg/dL). IFG was defined as a FBG level ≥110 mg/dL.

Statistical analysis was performed using PASW Statistics 17.0 (Statistical Package for Social Science Japan, Inc., Tokyo, Japan). All values are expressed as mean ± standard deviation (SD), unless otherwise specified. Data for TG, FBG, HOMA-IR, and serum HMW adiponectin were skewed, and log-transformed for analysis. Differences between the two groups were determined by Student's t-test and χ² test. Subjects were divided into four groups based on the quartile of HOMA-IR and serum HMW adiponectin levels and the cutoff points for metabolic disorder were defined as the fourth quartile of HOMA-IR and first quartile of serum HMW adiponectin levels, respectively. Multiple linear regression analysis was used to evaluate the contribution of each lipid profile for the number of MetS components, HOMA-IR, and serum HMW adiponectin. In addition, areas under the receiver operating characteristic (ROC) curves were determined for each variable to identify the predictors of MetS-associated variables. The area under the curve (AUC) of ROC curves is a summary of the overall diagnostic accuracy of the test. The best markers have ROC curves that are shifted to the left with areas under the curve near unity. Nondiagnostic markers are represented by diagonals with the AUC of ROC curves close to 0.5. A value of P < 0.05 was considered significant.

Table 1 shows the value of each background factor categorized by sex. The subjects comprised 614 men aged 58 ± 14 (range, 20-89) years and 779 women aged 60 ± 12 (range, 21-88) years. The mean BMI was 23.3 ± 3.0 kg/m² and 22.9 ± 3.2 kg/m² in men and women, respectively. BMI, smoking status, alcohol consumption, systolic blood pressure (SBP), diastolic blood pressure (DBP), TG, and TG/HDL-C ratio were significantly higher but age, T-C, HDL-C, LDL-C, and Non-HDL-C were significantly lower in men than in women. There were no inter-group differences regarding T-C/HDL-C ratio, LDL-C/HDL-C ratio, eGFR, and prevalence of CVD.

Prevalence of MetS and FBG were significantly higher in men than in women, but IRI, HOMA-IR and serum HMW adiponectin were significantly lower in men (Table 2).

Figure 1 showed synergistic effect of increased HOMA-IR and reduced serum HMW adiponectin. Mean accumulated number of MetS components was the highest in the highest HOMA-IR-lowest serum HMW adiponectin quartile group and lowest in the lowest HOMA-IR-highest serum HMW adiponectin quartile group, while HOMA-IR appears to be a more dominant determinant for MetS components than serum HMW adiponectin. HOMA-IR and serum HMW adiponectin were independently associated with the number of MetS components. We assessed the statistical significance of the relationship using a general linear model with the following confounding factors: age, sex, BMI, smoking status, alcohol consumption, SBP, DBP, and eGFR. The interaction between increased HOMA-IR and reduced serum HMW adiponectin was a significant and dependent determinant for the accumulated number of MetS components (F = 18.7, P < 0.001), in addition to their direct association (HOMA-IR, F = 77.0, P < 0.001; serum HMW adiponectin, F = 9.71, P = 0.002).

To further investigate whether lipid profiles can explain the MetS-associated variables independent of other known confounding factors, multiple linear regression analyses for the number of MetS components, HOMA-IR, and serum HMW adiponectin were performed (Table 3). In both men and women, TG/HDL-C and T-C/HDL-C ratios as well as TG showed significantly strong associations with all 3 of the MetS-associated variables. Non-HDL-C and LDL-C/HDL-C ratio also showed significant correlations; however almost their β estimates were lower than the TG/HDL-C ratio.

In men, the ROC curve analyses showed that the best marker of the MetS-associated variables (Presence of MetS, HOMA-IR§ >1.98, and serum HMW adiponectin <2.97 μg/mL) was TG/HDL-C ratio, with the AUC for presence of MetS (AUC, 0.82; 95% CI, 0.77-0.87), HOMA-IR (AUC, 0.75; 95% CI, 0.70-0.80), and serum HMW adiponectin (AUC, 0.67; 95% CI, 0.63-0.71), respectively (Table 4). The T-C/HDL-C ratio, TG, HDL-C, non-HDL-C, and LDL-C/HDL-C ratio also discriminated these markers; however all their AUC estimates were lower than the TG/HDL-C ratio. These results were similar in women. On the other hand, each AUC of LDL-C for these three variables was near 0.50 in both women and men.

HDL, high-density lipoprotein; LDL, low-density lipoprotein; eGFR, estimated glomerular filtration rate. Data presented are mean ± standard deviation. Data for triglycerides and triglycerides/HDL cholesterol ratio were skewed and were presented as median (interquartile range). Data for triglycerides and triglycerides/HDL cholesterol ratio were log-transformed for analysis. * Student's t-test or χ² test.

HOMA-IR, homeostasis model assessment of insulin resistance; HMW, high molecular weight. Data for fasting blood glucose, fasting insulin, HOMA-IR, and serum HMW adiponectin were skewed and presented as median (interquartile range) values. §HOMA-IR was calculated using the following formula; {fasting blood glucose (FBG) (mg/dL) X fasting insulin (mU/mL)}/405. Data for fasting blood glucose, immuno-reactive insulin, HOMA-IR, and serum HMW adiponectin were log-transformed for analysis. * Student's t-test or χ² test.

MetS, Metabolic syndrome; HOMA-IR, homeostasis model assessment of insulin resistance; HMW, high molecular weight; HDL, high-density lipoprotein; LDL, low-density lipoprotein. §HOMA-IR was calculated using the following formula; {fasting blood glucose (mg/dL) X immuno-reactive insulin (mU/mL)}/405. Adjusted for age were BMI, smoking status, alcohol consumption, SBP, DBP, and eGFR. Data for triglycerides, triglycerides/HDL cholesterol ratio, and HOMA-IR were skewed, and log-transformed for analysis.

ROC, receiver operating characteristics; MetS, metabolic syndrome; HOMA-IR, homeostasis model assessment of insulin resistance; HMW, high molecular weight; AUC, area under ROC curve; CI, confidence interval; HDL, high-density lipoprotein; LDL, low-density lipoprotein. §HOMA-IR was calculated using the following formula; {fasting blood glucose (mg/dL) X immuno-reactive insulin (mU/mL)}/405. Insulin resistance was defined as values exceeding the fourth quartile of HOMA-IR, and less than the first quartile of serum HMW adiponectin, respectively.