Three novel obese indicators perform better in monitoring management of metabolic syndrome in type 2 diabetes

The present study evaluated the performance of three novel obese indicators, visceral adiposity index (VAI), lipid accumulation product (LAP) and waist circumference-triglyceride index (WTI), for identifying metabolic syndrome(MetS) in type 2 diabetes. A cross-sectional study was conducted on 711 type 2 diabetes in Qinhuangdao. The MetS was defined as the definition of Chinese Diabetes Society. Receiver operating characteristic curve analyses were performed to assess the accuracy of three obese indicators as diagnostic tests for MetS. The prevalence of MetS was 71.3%. In men, among all three obese indicators, the LAP had the highest area under curve (AUC) value (AUC = 0.894), followed by VAI (AUC = 0.860) and WTI (AUC = 0.855). In women, among all three obese indicators, the LAP had the highest AUC value (AUC = 0.906), followed by WTI (AUC = 0.887) and VAI (AUC = 0.881). However. there was no significant difference between the three obese indicators(P > 0.05). Three obese indicators were effective indicators for the screening of MetS, LAP and WTI are more simple.

Measurements. Anthropometric measurements, including height, weight and waist circumference(WC) were obtained while the subjects were in light clothing and barefoot. Height and weight were measured to the nearest 0.1 cm and 0.1 kg, respectively. WC was accurately measured at the level of midway between the lowest rib and the top of the iliac crest. All measurements were taken twice, and the two measurements were averaged for analysis. Blood pressure was measured three times with a mercury sphygmomanometer while the subjects were seated after 10 min of rest, and the three measurements were averaged for analysis. Body mass index(BMI) was calculated by dividing weight (kg) by height squared (m 2 ). Waist-to-height ratio(WHtR) was calculated by dividing the WC by height.
After a 10-hour overnight fast, blood samples were collected from an antecubital vein into heparinised tubes. Fasting plasma glucose(FPG) concentration was measured using the glucose oxidase method, and serum lipid levels were measured using enzymatic assays with an autoanalyzer (Hitachi, Tokyo, Japan).
Statistical Analyses. All analyses were performed using the SPSS 11.5 statistical software (SPSS 11.5 for Windows; SPSS, Inc., Chicago, IL). Numerical variables were reported as mean ± standard deviation. Comparisons were conducted between groups using the t test. Comparison of prevalence data was performed by χ 2 analysis. For each obese indicators, we divided them into increasing sex-specific quartile values and used logistic regression analysis to calculate odds ratios (ORs) and 95% confidence intervals (CIs) for MetS across quartiles, with quartile 1 as reference group, adjusting for potentially confounding variables such as age, BMI, drink, smoke and history of diabetes. By using receiver operating characteristic (ROC) analysis, ROC curves of each obese indicators were drawn to show how well they could separate subjects into groups with or without MetS. A test with an area under curve (AUC) of ≥0.85 is considered an accurate test 27 . Sensitivity and specificity of each obese indicators have been calculated at all possible cutoff points to find the optimal cutoff values. The optimal sensitivity and specificity were the values yielding maximum sums from the ROC curves. The comparisons of AUCs were performed with MedCalc 11.4.2.0 software (Ostend, Belgium). P < 0.05 was considered statistically significant.

Results
Among these subjects, 71.3% were characterized by MetS. The prevalence of MetS was similar between men and women (men 72.1% vs women 70.3%, χ 2 = 0.261, P = 0.610). Characteristics of the study population are presented in Table 1. The age, history of diabetes, drink, smoke and FPG between the two groups were similar (P > 0.05). The levels of BMI, WC, WHtR, SBP(systolic blood pressure), DBP(diastolic blood pressure), TG, VAI, LAP and WTI were all significantly higher in MetS group than in non-MetS group (P < 0.001). The level of HDL-C was significantly lower in MetS group than in non-MetS group (P < 0.001).
The VAI, LAP and WTI were strongly associated with the odds of having MetS in both men and women, after adjustment for age, BMI, drink, smoke and history of diabetes. ORs for MetS increased with increasing quartiles of all variables(P < 0.001) (Tables 3, 4 and 5).

Discussion
In this population, we found that 71.3% of participants had MetS. The prevalence of MetS was very common in type 2 diabetes, which is similar to previous study 5,6 . The performances of WC and WHtR were similar in our study. Several visceral adiposity indicators, such as LAP in men and LAP, WTI, VAI in women appeared to be better indices to predict presence of MetS, relative to WC and WHtR. Our study presents evidence that the LAP may simple and useful tool for the screening of MetS in type 2 diabetes.
It is now widely acknowledged that, the core of MetS is central obesity (also known as visceral adiposity), overweight with adipose tissue accumulation particularly around the waist. The evaluate of visceral adipose tissue, such as magnetic resonance imaging and computed tomography, is very expensive, may involve exposure to radiation, and their availability is limited. WC is the most accurate surrogate marker of visceral adiposity in adults,

Variables
Non-MetS group (n = 204) MetS group (n = 507) t or χ 2 P  and are good indicators of insulin resistance 28,29 . However, two pepole are the same in WC but different in height, so the cardiovascular risk factors maybe different. WHtR is defined as their WC divided by their height and has been widely recommended 30 . But the performances of WC and WHtR were similar in our study. Robust statistical evidence from studies involving 12 studies(men) and 13 studies(women) in several ethnic groups, shows the improvement in AUC (WHtR > WC) for detecting MetS is slightly in men (0.003) and women(0.009) 31 . WC may be not appropriate for identifying MetS on its own. First, metabolically obese, normal-weight (MONW) individuals are very common in type 2 diabetes 32 . Second, obesity is a heterogeneous condition, thus metabolic abnormalities and cardiometabolic risk vary among obese individuals, with a significant proportion considered to be "metabolically healthy" 33 . Kahn et al. developed the LAP as an accurate index for identifying United States adults at cardiovascular risk 25 . LAP is an index based on a combination of WC and TG. In our study, LAP is preferred to WC and WHtR, both men and women. Dual metabolic defects are required to produce hypertriglyceridemia in obese subjects with similar levels of visceral adiposity. Mirmiran et al. found that LAP is associated with insulin resistance, lipid peroxidation, and systemic inflammation in type 2 diabetic patients 34 . Sambataro et al. found that LAP in relationships with insulin sensitivity (evaluated by the quantitative insulin sensitivity check index) in overweight type 2 diabetic subjects 35 . These results emphasize the clinical importance of assessing LAP in type 2 diabetes to identify subjects at high cardiovascular risk. WTI is similar to LAP 22 . Compared with LAP, WTI has better specificities but poor sensitivities.
VAI is an index estimated with the use of both anthropometric (BMI and WC) and metabolic (TG and HDL-C) parameters. This index is more complicated, but is not superior to LAP and WTI. It is consistent with previous results. In elderly, the VAI showed the lowest discriminatory power for MetS 16 . In the Chinese rural adults, the AUC of VAI for the screening of MetS was less than that of LAP in both men and women 36 .
There were differences between the results of VAI and WTI in men and women. The AUCs of WC and WHtR in women were slightly lower than men. The AUCs of LAP and WTI in women were slightly higher than men. Therefore, LAP and WTI appeared to be better indices to predict presence of MetS only in women, relative to WC and WHtR. The results suggest that WC may be more practical in men. Kim et al. found that the diagnostic values of the visceral fat area and WC for predicting the presence of multiple metabolic risk factors are influenced by gender. Although the WC and visceral fat area are similar in men, the WC is inferior to the visceral fat area in women 37 . WC correlates with hepatic fat accumulation in male Japanese patients with non-alcoholic fatty liver disease, but not in females 38 .
There are limitations to our study. First, the LAP was derived from studies of white, non-Hispanic blacks and Mexican Americans and the VAI was established in Caucasian populations 24,25 . The suitability for other populations need to be further investigated. Second, the relationship between LAP, VAI, WTI and diabetic complication should be studied in the futrure. Third, the study population was relatively small. A larger sample size may be helpful in further strengthen the conclusion. Fourth, the levels of insulin were not measured in our study. We can not compare the performances of LAP, VAI and insulin resistance. Fifth, I did not record the information of excluded patients. This maybe cause some selective bias.
In summary, the prevalence of MetS was high in type 2 diabetes. Three novel obese indicators were effective indicators for the screening of MetS, LAP and WTI are more simple.