To date, there have been only a few studies of anthropometric indicators of MetS recognition in non-overweight/obese people, and the results are not consistent with the latest anthropometric indicators. This cross-sectional study innovatively compared the identification ability of the novel central obesity index with traditional indicators for non-overweight/obese individuals with a large sample size and a larger population coverage. The results demonstrate that the novel anthropometric index can identify MetS in non-overweight/obese people. Among them, AVI had the best ability to identify MetS and low HDL-C in different genders.
A growing number of studies showed that cardiometabolic disease often occurs in people with normal weight(3–5, 8). Visceral adipose tissue (VAT) accumulation is a major cause (26). Clinicians often use anthropometric indices which reflect VAT to screen MetS in large populations. For a long time, BMI combined with WC has been extensively used to assess central obesity. But both two predicted all-cause mortality in the opposite way in some cases (27). The paradox occurs when the distinction of body fat is not made to predict cardiometabolic risk (28). Therefore, the assessment of central obesity and the prediction of cardiometabolic disease by BMI combined with WC are limited. In this study, people with BMI < 24 kg/m2 were taken as the study subjects, and MetS was taken as the disease, avoiding the assessment of BMI combined with WC.
WHtR, another widely used traditional anthropometric index, is superior to BMI and WC in the assessment of central obesity (10–12, 29). As a simple and effective anthropometric index, it has been recommended by many scholars as a screening tool for cardiometabolic risk factors. Even in people with normal BMI and/or WC, WHtR can effectively identify cardiometabolic disease (30, 31). However, meta-analysis based on Embase and Medline databases showed that WHtR was not superior to other anthropometric indicators in distinguishing MetS and other cardiometabolic factor (14). Moreover, these studies did not compare WHtR with sundry new anthropometric indexes such as WWI and AVI, and it was still uncertain whether the WHtR in non-overweight/obese people is the optimal anthropometric index to screen MetS.
Ulike the traditional anthropometric index, the new anthropometric index mostly started from the geometric model of the human body, reflected the VAT of the body in a three-dimensional way. Based on waist and hip circumference, AVI calculated the entire abdominal volume from symphysis of pubis to xiphoid appendix (21), theoretically including abdominal free fat and adipose tissue volumes, which are the main distribution areas of visceral fat (32). VAT has fewer insulin receptors distributed on the cell surface, decreased insulin receptor substrate protein-1 expression, and reduced insulin receptor affinity. Therefore, VAT becomes less sensitive to insulin and has lower sugar uptake and utilization (33). In addition, free fatty acid produced by VAT lipolysis affects insulin signaling pathway, reduces the sensitivity of liver and skeletal muscle to insulin, inhibits glucose uptake and oxidation, and aggravates glucose regulation disorders (34). Therefore, AVI demonstrated excellent predictive power for IGT and DM by fully evaluating VAT (21). Previous study found that the predictive ability of AVI for MetS was better than other new anthropometric indicators and traditional indicators (35, 36), and was the strongest predictor , which was similar to the results of this study. In this study, AVI's ability to identify MetS in non-overweight/obese people was lower than that in Spanish adolescents (AUC: 0.831 for males, 0.867 for females) (35), but higher than that in northern Iran (AUC: 0.72 for males, 0.73 for females). This was related to race, age and BMI range of the study population (37, 38).
This study found that BRI and WHtR had similar ability to screen MetS, which were similar to the results of previous studies (18, 39). WHtR reflects central obesity through simple numerical comparison, which overcomes the influence of height on VAT (40). However, BRI is a human body ellipse model, which evaluates body fat rate and VAT according to roundness and eccentricity (15) and quantifies individual body shape in an independent manner of height. Although the two anthropometric indicators have different principles, they are all derived from WC and height. VAT assessment is also based on WC and abdominal fat volume. Therefore, BRI and WHtR had the similar recognition capability for MetS, but they had no statistical difference with AVI. Compared with AVI and BRI, WHtR is easier to obtain and more suitable for early preliminary screening in a large population.
WWI is a recently developed unique obesity index based on weight and WC (24), which can predict the incidence and mortality of obesity-related diseases with a linear trend, avoiding the U-shaped relationship between BMI and CVD mortality. Until now, there have been no studies on the recognition capability of WWI for MetS and no studies on the prognosis of metabolic diseases in non-overweight/obese people. Our study found that WWI was less able to identify MetS in non-overweight/obese people than WHtR. It was speculated that WWI was based on body weight and WC and it could not distinguish fat distribution and body weight composition, so the fat content was underestimated. Since height was not taken into account, WWI usage may underestimate the VAT of short subjects and overestimate the VAT of tall subjects, thus misleading the diagnosis of central obesity and failing to predict the prevalence of MetS. ABSI is another body shape index based on WC, weight and height (16). At a given height and weight, high ABSI means WC is higher than expected, which is a good indicator of central obesity. ABSI changes the limitations of WC as a BMI dependent index (16). Studies have shown that ABSI can identify visceral obesity and sarcopenic obesity in overweight/obese adults with T2DM (41). However, this study showed that, similar to WWI, ABSI was also weak in predicting MetS in non-overweight/obese people. Similar findings have been found in other studies (42, 43). It is speculated that although these two indexes are new obesity indexes, both the establishment and verification of the prediction model are to make up for the deficiency of BMI in the prediction of obesity-related mortality risk, and the identification of MetS may affect its diagnostic efficacy. To realize the screening of MetS or cardiometabolic disease, further formulation and a large sample population studies are needed.
In addition, anthropometric indices were better at identifying MetS in females than in males after adjusting for confounding factors. This cannot be explained by conventional wisdom, which males have more visceral fat and females have more subcutaneous adipose tissue (37). This may be related to the large number of female subjects. Further studies on the predictive power of novel anthropometric indicators for MetS of different genders are needed.
Currently, the novel anthropometric indicators have shown advantages in various fields. They are worthy of clinical and public health promotion for their ability to predict obesity-related diseases and deaths at an earlier stage, but new longitudinal studies are needed in a broader population to further explore their predictive power.
Our study has several limitations. First, this was a cross-sectional study and it cannot reflect causality. Second, information on the lifestyle of the participants was not obtained in the study. These may confound the relationship between anthropometric indices and MetS. Third, the participants were volunteers who were more concerned about their health and who might have a history of CVD or family history. Finally, we did not measure 2 hours postprandial blood glucose, which may lead to under diagnosis in some diabetes patients.