Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Epidemiology and Population Health

Visceral adiposity index, hypertriglyceridemic waist and risk of diabetes: the China Health and Nutrition Survey 2009



The visceral adiposity index (VAI) and hypertriglyceridemic waist phenotype (the simultaneous presence of waist circumference (WC)90/80 cm for men/women and plasma triglyceride (TG) concentration1.7 mmol l−1 for both genders) have been identified as good indicators of visceral adiposity, which is an independent risk factor for diabetes. The Chinese population is characterized by a predominance of visceral fat accumulation despite having comparatively low weight. These two surrogate markers of visceral adiposity might effectively identify Chinese adults who are at risk of getting diabetes. We aimed to examine the association between VAI and risk of diabetes or between the hypertriglyceridemic waist phenotype and diabetes risk.


We conducted a cross-sectional analysis of 7639 Chinese men and women aged 18 years using data from the China Health and Nutrition Survey 2009. Logistic regression was used to evaluate the associations.


For men, compared with participants in the lowest quartile of VAI scores, the multivariable-adjusted odds ratios (ORs) (with 95% confidence intervals) for diagnosed diabetes were 1.1 (0.7–1.7), 1.9 (1.3–2.8) and 3.6 (2.5–5.3) for those in the second, third, and top quartile of VAI scores, respectively. For women, the corresponding figures were 0.9 (0.5–1.4), 1.7 (1.1–2.6) and 2.8 (1.9–4.2), respectively. The multivariate-adjusted ORs (with 95% confidence intervals) for diabetes in men with the hypertriglyceridemic waist phenotype compared with men with both WC and TG measurements below the defined cut points were 3.7 (2.6–5.4). For women, the corresponding figure was 3.7 (2.4–5.5). For both men and women, the associations between the 4th quartile of VAI scores and risk of diabetes or between the hypertriglyceridemic waist phenotype and risk of diabetes were consistently seen in various subgroups.


Among Chinese adults, high VAI scores and the hypertriglyceridemic waist phenotype are strongly associated with diabetes risk.


With the rapid economic growth and an accelerated pace of nutrition transition, the increasing rate of diabetes in China is especially alarming. According to data from the China National Diabetes and Metabolic Disorders Study, 92.4 (9.7%) million adults 20 years of age or older suffer from diabetes.1 As a result of the continuous increase in the prevalence of overweight and obesity in China,2 diabetes is expected to be even more prevalent in the future, which is of concern as China remains one of the world’s most populous regions. Given that diabetes is a major risk factor for cardiovascular disease as well as premature mortality3 and that there is an enormous burden of diabetes in China, it is important to identify and treat risk factors for diabetes.

Visceral adiposity is independently associated with incident diabetes.4 The magnetic resonance imaging (MRI) and computed tomography (CT) constitute the gold standard for quantitative evaluation of visceral adiposity, but the costs and radiation exposure associated with these two methods represent major barriers to their widespread use in clinical practice. Waist circumference (WC), a surrogate marker of central adiposity, has been identified as a risk factor for the development of diabetes.5 However, WC cannot sufficiently discriminate between visceral and subcutaneous fat. Chinese individuals are characterized by a greater amount of visceral adipose tissue than Europeans at a given body mass index (BMI) or WC.6 Markers of visceral obesity may thus be particularly useful to the Chinese. The visceral adiposity index (VAI), a mathematical model that uses both anthropometric (BMI and WC) and metabolic (triglyceride (TG) and high-density lipoprotein cholesterol (HDL-C)) parameters, is a recently proposed marker of both visceral fat distribution and dysfunction.7 The hypertriglyceridemic waist phenotype defined as an enlarged WC along with an elevated TG concentration was firstly introduced in the Quebec Cardiovascular Study as a marker of excess visceral adiposity.8 Few studies, however, have examined whether the VAI and hypertriglyceridemic waist phenotype are related to diabetes risk in Chinese adults.

We took advantage of the large representative sample of Chinese adults who participated in the China Health and Nutrition Survey (CHNS) 2009 to examine the association between the two indicators of visceral adiposity and diabetes risk.

Materials and methods

Study design

The CHNS is the only large-scale, longitudinal, household-based survey in China,9 which was designed to represent a set of large provinces with significant variation in terms of geography, economic development and health status, covering approximately 56% of China’s population, including Liaoning, Heilongjiang, Jiangsu, Shandong, Henan, Hubei, Hunan, Guangxi and Guizhou. Currently, data were available for eight rounds (CHNS 1989, 1991, 1993, 1997, 2000, 2004, 2006 and 2009). For each round, a multistage random cluster process was employed to draw study samples in each of the provinces. Counties in the nine provinces were stratified by income (low, middle and high), and a weighted sampling scheme was used to randomly select four counties in each province. Full details of the CHNS have been described elsewhere.9 Each participant provided written informed consent, and the study was approved by the institutional review committees of the University of North Carolina at Chapel Hill, the National Institute of Nutrition and Food Safety, Chinese Center for Disease Control and Prevention, and the China-Japan Friendship Hospital, Ministry of Health.

Study population

Since fasting blood samples were initially collected in 2009, this study examined data from CHNS 2009. Participants aged 18 years were included in the present analysis. From the 2009 examination, data for 10 039 adult respondents were available. All participants were asked to complete a structured questionnaire that provided information on age, sex, urban/rural settings, regions (Southern (Jiangsu, Hubei, Hunan, Guangxi, Guizhou)/Northern (Liaoning, Heilongjiang, Shandong, Henan)), histories of current and previous illness, and medical treatment. Exclusion criteria included pregnancy, a self-reported diabetes diagnosis or diabetes medication use, anemia (hemoglobin<13 g dl−1 in men and<12 g dl−1 in women), chronic kidney disease (estimated glomerular filtration rate<15 ml min−1 per 1.73 m2) or no information on age, BMI, WC, blood pressure (BP), fasting plasma glucose (FPG), hemoglobin A1c (HbA1c), TG, or HDL-C. Ultimately, 7639 (76.1%) adults with anthropometry and clinical examination information were included in our analysis.


Weight, height, WC and BP were measured following standardized protocols from the World Health Organization (WHO). BMI was calculated as weight (in kilograms) divided by the square of height (in meters). WC was measured with an inelastic tape to the nearest 0.1 cm at a midpoint between the bottom of the rib cage and the top of the iliac crest at the end of exhalation. BP was measured by trained examiners using a mercury sphygmomanometer at three different consecutive times at 3–5 min intervals on one visit. The three readings were averaged as the BP values in our data analysis. All physical examinations were performed at the same location and followed the same protocol at each study visit.

Biochemical measurements

Blood was collected after an overnight fast. Samples for FPG and HbA1c measurements were tested immediately. Serum samples for determination of cardiovascular risk factors were stored at −86 °C for later laboratory analysis. All blood samples were analyzed in a national central lab in Beijing. Glucose was measured with a Hitachi 7600 analyzer by GOD-PAP method (Randox Laboratories Ltd, Crumlin, UK). HbA1c was measured with a high-performance liquid chromatography system (model HLC-723 G7; Tosoh Corporation, Tokyo, Japan). Lipids (total cholesterol (TC), TG, low density lipoprotein cholesterol, HDL-C, apolipoprotein A1 and apolipoprotein B), uric acid and hypersensitive C-reactive protein (hs-CRP) were measured using a biochemical autoanalyzer (Hitachi 7600 automated analyzer, Tokyo, Japan). hs-CRP was determined by the immunoturbidimetric method. Fasting insulin concentration was measured using the radioimmunology assay (Gamma counter XH-6020, Xi'an, China). HOMA-IR (homeostasis model assessment of insulin resistance) was calculated by the formula: HOMA-IR=fasting insulin (micro-international units per milliliter) × FPG (millimoles per liter)/22.5. The VAI score was calculated by using the published formula:7 Men: [WC/39.68+(1.88 × BMI)] × (TG/1.03) × (1.31/HDL); Women: [WC/36.58+(1.89 × BMI)] × (TG/0.81) × (1.52/HDL), where both TG and HDL levels are expressed in mM.


According to the criteria recommended by the Working Group on Obesity in China,10 subjects were classified as normal (BMI of 18.5–24.0 kg m−2) or overweight (BMI of 24.0–27.9 kg m−2) or obese (BMI28.0 kg m−2).

According to the 2010 American Diabetes Association criteria,11 diabetes is defined as having FPG7.0 mmol l−1 or HbA1c6.5%, and prediabetes is defined as having FPG of 5.66.9 mmol l−1 or HbA1c of 5.76.4%. The hypertriglyceridemic waist phenotype was defined as the simultaneous presence of WC90/80 cm in men/women and TG1.7 mmol l−1 for both genders.

Statistical analysis

All statistical analyses were conducted using SPSS software (version 12.0 for windows; SPSS, Chicago, IL, USA). Both men and women were classified on the basis of the gender-specific quartiles of VAI scores (Table 1). Both men and women were also categorized into four mutually exclusive groups by the presence or absence of enlarged WC and elevated TG (Table 2). Continuous variables were presented as medians (25th to 75th percentiles (IQR)) due to their skewed distribution. Categorical variables were expressed as percentages. Kruskal–Wallis analysis of median test was used followed by the Mann–Whitney U test for pairwise comparisons. Bonferroni correction was applied to adjust P-values for multiple comparisons. Chi-square test was performed to assess differences in proportions across groups. For each gender, adjusted odds ratios (ORs) and corresponding 95% confidence intervals (CI) were estimated with the use of logistic regression analysis. The four models were as follows: Model 1 was adjusted for age. Model 2 was adjusted for age, socioeconomic status (rural/urban settings, region and education level), smoking status and alcohol use. Model 3 was adjusted for all the variables in model 2 plus TC, systolic and diastolic BP. Model 4 was additionally adjusted for white blood cell count, uric acid and hs-CRP. We chose these variables because of their potential role as confounders from the clinical point-of-view. We assessed collinearity between independent variables using variance inflation factor.12 Variance inflation factor >10 warrants caution.13 With the collinearity diagnosis, it was observed that all models presented were free from collinearity (the variance inflation factor for all variables were<2.5). We examined whether the association between VAI and diabetes was linear or non-linear by performing regression models with the VAI fourths as three-indicator variables (non-linear) and comparing this model with one in which the fourths were included as a single variable (a score from1 to 4; linear model across fourths). We compared these two models using the Akaike Information Criterion. The lower the Akaike Information Criterion value, the better the model fit. We also estimated ORs of diabetes associated with the hypertriglyceridemic waist phenotype with the subgroup with both WC and TG measurements below the defined cut points served as the reference group. For each evaluated subgroup, the unadjusted ORs between the exposure variables (the VAI score and the hypertriglyceridemic waist phenotype) and diabetes were determined by univariate logistic regression analysis. A two-tailed P value of<0.05 was considered to be significant.

Table 1 Characteristics represented across visceral adiposity index (VAI) quartiles in each gender
Table 2 Characteristics of participants by waist circumference (WC) and triglyceride (TG) status in each gender


VAI scores and the prevalence of hypertriglyceridemic waist phenotype in BMI categories

In both genders, there was remarkable variation in VAI scores for any given BMI value. For men, the VAI medians (IQR) were 1.0 (0.6–1.6) for those with normal weight, 1.6 (1.0–2.7) for those with overweight and 2.1 (1.4–3.8) for those with obesity. For women, the VAI medians (IQR) were 1.3 (0.9–2.0) for those with normal weight, 1.9 (1.3–3.1) for those with overweight and 2.4 (1.5–4.0) for those with obesity. For men, the prevalence of hypertriglyceridemic waist phenotype was 2.8% for those with normal weight, 24.5% for those with overweight and 54.8% for those with obesity. For women, the corresponding figures were 8.4, 31.1 and 49.0%, respectively.

Characteristics of the study participants classified according to the VAI quartiles

In both genders, there were clear dose-response relationships of VAI scores with all of the listed variables listed in Table 1 and Figure 1, except the percentages of current smokers and current drinkers. BMI, WC, systolic and diastolic BP, TC, TG, TC/HDL-C, apolipoprotein B/apolipoprotein A1, FPG, HbA1c, HOMA-IR, uric acid, white blood cell count, hs-CRP (Table 1), apolipoprotein B and fasting serum insulin (Figure 1) associated positively, whereas HDL-C associated inversely with elevated VAI scores. Age progressively increased across VAI quartiles in women, whereas it decreased linearly in men (both P for trend<0.001). The proportion of both men and women with prediabetes and diabetes increased progressively with increasing VAI scores.

Figure 1

Apolipoprotein B and fasting insulin levels across quartiles of visceral adiposity index in men and women. Note medians and box (P25-P75) and whisker (minimum-maximum) plots of apolipoprotein B and fasting insulin levels in men (white) and women (gray).

Characteristics of the study participants stratified by WC and TG levels

For both genders, individuals with both WC and TG measurements below the defined cut points were characterized by the most favorable cardiovascular risk profile (Table 2). By contrast, those with either abdominal obesity or hypertriglyceridemia, or both had higher levels of BMI, diastolic BP, TC, TG, TC/HDL-C, apolipoprotein B, fasting insulin, apolipoprotein B/apolipoprotein A1, FPG, HbA1c, HOMA-IR, uric acid, hs-CRP and VAI score and had lower levels of HDL-C. Both men and women in the hypertriglyceridemic waist phenotype subgroups had a more deteriorated cardiovascular risk profile than those who were not characterized by this phenotype. Similarly, they had the highest prevalence of prediabetes and diabetes.

ORs for VAI- or for hypertriglyceridemic waist-related diabetes

Among both men and women, the ORs for diabetes increased with increasing quartiles of VAI scores (P for trend<0.001), but the associations across quartiles of VAI scores seemed to be nonlinear (the Akaike Information Criterion values for the models with the VAI fourths as three-indicator variables were lower than those for the models with VAI fourths were included as a single variable). Although cut-points for VAI quartiles were higher in women than men, an independent association of VAI with diabetes was stronger in men than in women. For men, the age-adjusted ORs (95% CIs) for diabetes (model 1) were 1.1 (0.7–1.6) for the second, 1.9 (1.3–2.8) for the third, and 3.9 (2.8–5.6) for the fourth VAI quartile, in comparison with the first quartile (Table 3). For women, the corresponding figures were 0.9 (0.6–1.5), 1.9 (1.3–2.8) and 3.6 (2.5–5.3), respectively. For both men and women, results remained essentially unchanged after additionally adjusting for socioeconomic status, smoking, and alcohol consumption (model 2). The associations persisted, although they were slightly attenuated, after additional adjusting for potential intermediate variables, including TC, systolic and diastolic BP (model 3), and further adjusting for inflammatory biomarkers (model 4).

Table 3 Adjusted odds ratios (OR) (with 95% confidence intervals (CI)) for the visceral adiposity index-related diabetes risk and for the hypertriglyceridemic waist phenotype-related diabetes risk

For both men and women, the ORs for diabetes increased with either solely-elevated WC or solely-increased TG levels (Table 3). The strongest positive association, however, was found in subgroups with hypertriglyceridemic waist phenotype. Moreover, the hypertriglyceridemic waist phenotype was a stronger risk factor for diabetes in women compared to men. For men, the age-adjusted OR (95% CI) of diabetes associated with hypertriglyceridemic waist phenotype was 4.6 (3.3–6.2). For women, the corresponding figure was 5.2 (3.6–7.6). The associations remained statistically significant after additional adjustment for potential intermediate variables. For both men and women, the hypertriglyceridemic waist phenotype seemed to be associated stronger with diabetes compared to the VAI score.

As shown in Figure 2, in each gender, the greater ORs for diabetes comparing the presence versus the absence of the hypertriglyceridemic waist phenotype were consistently significant in all evaluated subgroups (all P<0.001). The relations between the 4th VAI quartile (because of the infrequency of diabetes among those with VAI in the three lowest quartiles, we collapsed the 3 lowest VAI quartiles into a single reference group) and risk of diabetes were also consistently seen in each evaluated subgroup. For men, both the hypertriglyceridemic waist phenotype and the 4th VAI quartile were associated with similar ORs in those with TC<5.2 mmol l−1 and those with TC5.2 mmol l−1. The hypertriglyceridemic waist phenotype and the 4th VAI quartile were associated with significantly greater ORs in those with low levels of other cardiovascular risk factors, including systolic/diastolic BP<140/90 mm Hg, and HDL-C 1.0 mmol l−1 (all P<0.01); Only the hypertriglyceridemic waist phenotype associated stronger with diabetes in those with hs-CRP<3 mg l−1, whereas only the 4th VAI quartile associated stronger with diabetes in those with BMI<24 kg m−2 and younger than 65 years of age. For women, the hypertriglyceridemic waist phenotype and the 4th VAI quartile were associated with significantly greater ORs in those with low levels of cardiovascular risk factors including younger age (<65 years), BMI<24 kg m−2, HDL-C 1.3 mmol l−1, TC<5.2 mmol l−1 and hs-CRP<3 mg l−1 (all P<0.01); Only the hypertriglyceridemic waist phenotype associated stronger with diabetes in those with systolic/diastolic BP<140/90 mm Hg.

Figure 2

Associations of the hypertriglyceridemic waist phenotype with diabetes risk or associations of the visceral adiposity index with diabetes risk within subgroups. Data are expressed as odds ratio and 95% confidence interval.


In the present study, we found a graded positive association between VAI scores and the prevalence of diabetes for both men and women. Our results are consistent with the limited data showing that the VAI is an independent predictor of diabetes.14 We also found a strong positive association between the hypertriglyceridemic waist phenotype and diabetes risk for both men and women. In each gender, the relationships between the VAI and diabetes risk or between the hypertriglyceridemic waist phenotype and diabetes risk were independently of age, socioeconomic status, and other potential risk factors for diabetes and persisted in each evaluated subgroup. This is, to our best knowledge, the first Chinese population-based study to report a strong relationship between VAI and the risk for diabetes.

A marked increase in the prevalence of obesity and physical inactivity has contributed to a tripling in diabetes incidence over the past 15 years in China.1, 15 Accumulating evidence has suggested that easily measurable markers of adiposity (such as BMI) were associated with diabetes.5, 16 However, BMI has been proved to inadequately discriminate diabetes risk among obese individuals.17 Our data further supported this notion (data not shown). Moreover, our findings that there is remarkable variation in VAI scores for any given BMI value and that a considerable proportion of participants without obesity have the hypertriglyceridemic waist phenotype support the notion that the performance of BMI in distinguishing visceral from subcutaneous adiposity is limited.

Obesity is a heterogeneous disorder. A recent study indicates that 51% of overweight and 32% of obese adults are metabolically healthy and a substantial number of individuals with metabolic disturbance do not conform to general obesity.18 Prior studies have indicated that measurement of visceral fat offer a powerful tool to discriminate between normal-weight individuals with metabolic abnormalities and metabolically healthy obese individuals.19, 20 Different fat compartments associate with differential metabolic risk, as evidenced by the fact that visceral fat is more metabolically deleterious than subcutaneous fat.21, 22, 23 Moreover, despite the fact that studies have shown the hazard of subcutaneous fat for cardiometabolic risk factors,24 two prospective studies have suggested that subcutaneous fat was not associated with the development of diabetes in obese individuals.4, 25 Excess visceral adiposity increases diabetes risk through several potential mechanisms. Visceral adiposity, having greater endocrine activity than does subcutaneous fat,26 has been proposed as a marker of dysfunctional adipose tissue and ectopic fat deposition with excessive free fatty acid and resistine release, leading to lipotoxicity and insulin resistance in muscle, liver and pancreatic β cells, which in turn can inhibit glucose uptake. In addition, compensatory hyperinsulinemia through renal or other mechanisms would contribute to hyperuricemia and hypertension,27 leading to an increased risk of diabetes. Furthermore, the increased delivery of lipolytic products to the liver would also promote hepatic gluconeogenesis.28 Hence, visceral adiposity is useful in diabetes risk discrimination. Although imaging techniques, such as CT and MRI, are required for measurement of visceral adiposity, they cannot be used in daily practice due to practical, ethical and economic reasons. In light of recent data about the VAI and hypertriglyceridemic waist phenotype, simple and inexpensive alternative approaches may serve as surrogate markers of visceral adiposity for the quantitative evaluation of fat mass and for assessing viscerally obese individuals at risk for cardiometabolic disease.

The Alkam Metabolic Syndrome Study has shown that the VAI is inversely associated with insulin sensitivity (evaluated with a euglycemic-hyperinsulinemic clamp) and is positively associated with visceral adipose tissue (measured with MRI).7 The study also shows that the VAI associates with an increased cardiometabolic risk.7 Recent studies provide evidence that VAI associates with substantially elevated risk of coronary heart disease29 and relates inversely with adiponectin.14 Although the VAI mathematical model was developed in Caucasian population, one study conducted in Korean population showed that the VAI can replace visceral CT scanning as a marker for visceral adiposity,30 indicating that the VAI mathematical model can be also suitable for Asian populations. Our present study, for the first time, observed a significant relationship between increased VAI scores and diabetes risk in Chinese adults. Since excess visceral fat, but not general adiposity, is independently associated with incident diabetes in obese adults,4 our findings that the positive relationships between increased VAI scores and diabetes risk persisted in both BMI<24 kg m−2 and BMI24 kg m−2 subgroups and that apoprotein B and insulin concentrations, two markers sensitive to increasing levels of visceral adipose tissue, progressively increased with increasing VAI scores support the notion that VAI is a good indicator of visceral adiposity and adipose tissue dysfunction.

The Quebec Cardiovascular Study group firstly introduced the ‘hypertriglyceridemic waist phenotype’ as a marker of excess visceral adiposity and atherogenic metabolic triad (i.e., hyperinsulinemia, hyperapolipoprotein B and small, dense LDL particles) in men and demonstrated that the hypertriglyceridemic waist phenotype was a stronger marker of cardiovascular risk and a better predictor of 5-year risk for cardiovascular disease than the metabolic syndrome.8, 31 The CHICAGO cohort showed that the hypertriglyceridemic waist phenotype could represent a simple marker of excess visceral fat in persons with type 2 diabetes.32 The consistent positive associations between the hypertriglyceridemic waist phenotype and the risk for coronary heart disease persisted in various studies.29, 33, 34 Our results show a close relation between the hypertriglyceridemic waist phenotype and diabetes risk. Studies showed that individuals’ cardiovascular disease risk depends jointly on their body size and metabolic profile.35, 36 Our finding that the hypertriglyceridemic waist phenotype, which represents the simultaneous presence of abdominal obesity and metabolic abnormality, confers a higher diabetes risk than does solely-increased WC or solely-increased TG further supports this notion. The outcomes between the metabolically healthy obese individuals (one body size phenotype) and metabolically abnormal obese individuals (another body size phenotype) are different, with the former tending to be resistant to the development of the adiposity-associated cardiometabolic abnormalities.19 The visceral adiposity is the distinguishing factor separating metabolically healthy obese individuals from metabolically abnormal obese individuals,20 implicating that the concomitant of elevated WC and elevated TG (the hypertriglyceridemic waist phenotype) is a good indicator of visceral adiposity.

Interestingly, in our study, increasing VAI scores associate stronger with diabetes in men than women, whereas the hypertriglyceridemic waist phenotype is a stronger risk factor for diabetes in women compared to men. Although explanations for these issues remain to be elucidated, it is probably related to gender differences in regional adipose tissue distribution as well as patterns of visceral fat deposition.37 Also, the mechanisms by which visceral adiposity can lead to diabetes may be different in men versus women. On the other hand, in our study, both the 4th VAI quartile and hypertriglyceridemic waist phenotype have stronger relations with diabetes in those with low levels of cardiovascular risk factors. Age and BMI are well-established risk factors for diabetes.16 Other cardiometabolic traits such as hypertension, inflammation and lipid levels have also been identified as independent variables in clinical predication models.38, 39 Hence, diabetes risk may be partly explained by older age, obesity, hypertension, unfavorable lipid profile or increased hs-CRP. It is probably that early changes associated with visceral fat deposition and reflected by increased VAI scores or the hypertriglyceridemic waist phenotype trigger a worse glycemic profile; however, once the cardiovascular risk factors are onset, other factors rather than visceral adiposity might become more important. In our opinion, both the VAI and hypertriglyceridemic waist phenotype may serve as early markers of diabetes risk for both men and women.

Our findings have important clinical and public health implications because Chinese individuals are prone to have a greater amount of visceral adiposity than Westerners at a given BMI or WC.6 The increased visceral adiposity suggests that Chinese individuals are predisposed to an increased risk of diabetes and metabolic abnormalities. Traditional approaches, such as BMI and WC, reflect both subcutaneous and visceral fat mass. The current study presents evidence that the VAI and hypertriglyceridemic waist phenotype could serve as simple markers of visceral adiposity that would help to identify persons with increased risk for diabetes. Prevention of visceral fat accumulation or interventions that reverse this accumulation might substantially reduce the societal burden of diabetes. In addition, both the VAI and hypertriglyceridemic waist phenotype may be useful tools for tailoring bariatric surgery to obese individuals with diabetes.

Our study has several limitations. First, the sample is partially representative as only nine of China’s 31 provinces are included, and therefore, extrapolating results to the whole of China should be interpreted cautiously. Second, the cross-sectional study design makes it difficult to infer causality between the VAI and diabetes risk or between the hypertriglyceridemic waist phenotype and diabetes risk. Third, estimates across subgroups should also be interpreted with caution because of limited sample size. Nevertheless, our study has several strengths including a vigorous quality assurance program and the same strict methodology used to ensure the quality of the data collection over the entire study period, and the centralization of laboratory measurements.

In conclusion, our study documents that both the VAI and hypertriglyceridemic waist phenotype, simple and convenient markers of visceral obesity, are strong and independent risk factors for diabetes.


  1. 1

    Yang W, Lu J, Weng J, Jia W, Ji L, Xiao J et al. Prevalence of diabetes among men and women in China. N Engl J Med 2010; 362: 1090–1101.

    CAS  Article  Google Scholar 

  2. 2

    Xi B, Liang Y, He T, Reilly KH, Hu Y, Wang Q et al. Secular trends in the prevalence of general and abdominal obesity among Chinese adults, 1993-2009. Obes Rev 2012; 13: 287–296.

    CAS  Article  Google Scholar 

  3. 3

    Selvin E, Steffes MW, Zhu H, Matsushita K, Wagenknecht L, Pankow J et al. Glycated hemoglobin, diabetes, and cardiovascular risk in nondiabetic adults. N Engl J Med 2010; 362: 800–811.

    CAS  Article  Google Scholar 

  4. 4

    Neeland IJ, Turer AT, Ayers CR, Powell-Wiley TM, Vega GL, Farzaneh-Far R et al. Dysfunctional adiposity and the risk of prediabetes and type 2 diabetes in obese adults. JAMA 2012; 308: 1150–1159.

    CAS  Article  Google Scholar 

  5. 5

    Hu D, Xie J, Fu P, Zhou J, Yu D, Whelton PK et al. Central rather than overall obesity is related to diabetes in the Chinese population: the InterASIA study. Obesity (Silver Spring) 2007; 15: 2809–2816.

    CAS  Article  Google Scholar 

  6. 6

    Nazare JA, Smith JD, Borel AL, Haffner SM, Balkau B, Ross R et al. Ethnic influences on the relations between abdominal subcutaneous and visceral adiposity, liver fat, and cardiometabolic risk profile: the International Study of Prediction of Intra-Abdominal Adiposity and Its Relationship With Cardiometabolic Risk/Intra-Abdominal Adiposity. Am J Clin Nutr 2012; 96: 714–726.

    CAS  Article  Google Scholar 

  7. 7

    Amato MC, Giordano C, Galia M, Criscimanna A, Vitabile S, Midiri M et al. Visceral Adiposity Index: a reliable indicator of visceral fat function associated with cardiometabolic risk. Diabetes Care 2010; 33: 920–922.

    Article  Google Scholar 

  8. 8

    Lemieux I, Pascot A, Couillard C, Lamarche B, Tchernof A, Almeras N et al. Hypertriglyceridemic waist: A marker of the atherogenic metabolic triad (hyperinsulinemia; hyperapolipoprotein B; small, dense LDL) in men? Circulation 2000; 102: 179–184.

    CAS  Article  Google Scholar 

  9. 9

    Popkin BM, Du S, Zhai F, Zhang B . Cohort Profile: The China Health and Nutrition Survey—monitoring and understanding socio-economic and health change in China, 1989–2011. Int J Epidemiol 2010; 39: 1435–1440.

    Article  Google Scholar 

  10. 10

    Zhou BF . Predictive values of body mass index and waist circumference for risk factors of certain related diseases in Chinese adults—study on optimal cut-off points of body mass index and waist circumference in Chinese adults. Biomed Environ Sci 2002; 15: 83–96.

    Google Scholar 

  11. 11

    American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2010; 33: S62–S69.

    Article  Google Scholar 

  12. 12

    Bagley SC, White H, Golomb BA . Logistic regression in the medical literature: standards for use and reporting, with particular attention to one medical domain. J Clin Epidemiol 2001; 54: 979–985.

    CAS  Article  Google Scholar 

  13. 13

    Abbate LM, Stevens J, Schwartz TA, Renner JB, Helmick CG, Jordan JM . Anthropometric measures, body composition, body fat distribution, and knee osteoarthritis in women. Obesity (Silver Spring) 2006; 14: 1274–1281.

    Article  Google Scholar 

  14. 14

    Al-Daghri NM, Al-Attas OS, Alokail MS, Alkharfy KM, Charalampidis P, Livadas S et al. Visceral adiposity index is highly associated with adiponectin values and glycaemic disturbances. Eur J Clin Invest 2013; 43: 183–189.

    CAS  Article  Google Scholar 

  15. 15

    Yoon KH, Lee JH, Kim JW, Cho JH, Choi YH, Ko SH et al. Epidemic obesity and type 2 diabetes in Asia. Lancet 2006; 368: 1681–1688.

    Article  Google Scholar 

  16. 16

    Mokdad AH, Ford ES, Bowman BA, Dietz WH, Vinicor F, Bales VS et al. Prevalence of obesity, diabetes, and obesity-related health risk factors, 2001. JAMA 2003; 289: 76–79.

    Article  Google Scholar 

  17. 17

    Ferrannini E, Natali A, Bell P, Cavallo-Perin P, Lalic N, Mingrone G . Insulin resistance and hypersecretion in obesity. European Group for the Study of Insulin Resistance (EGIR). J Clin Invest 1997; 100: 1166–1173.

    CAS  Article  Google Scholar 

  18. 18

    Wildman RP, Muntner P, Reynolds K, McGinn AP, Rajpathak S, Wylie-Rosett J et al. The obese without cardiometabolic risk factor clustering and the normal weight with cardiometabolic risk factor clustering: prevalence and correlates of 2 phenotypes among the US population (NHANES 1999-2004). Arch Intern Med 2008; 168: 1617–1624.

    Article  Google Scholar 

  19. 19

    Stefan N, Kantartzis K, Machann J, Schick F, Thamer C, Rittig K et al. Identification and characterization of metabolically benign obesity in humans. Arch Intern Med 2008; 168: 1609–1616.

    Article  Google Scholar 

  20. 20

    Primeau V, Coderre L, Karelis AD, Brochu M, Lavoie ME, Messier V et al. Characterizing the profile of obese patients who are metabolically healthy. Int J Obes (Lond) 2011; 35: 971–981.

    CAS  Article  Google Scholar 

  21. 21

    Smith JD, Borel AL, Nazare JA, Haffner SM, Balkau B, Ross R et al. Visceral adipose tissue indicates the severity of cardiometabolic risk in patients with and without type 2 diabetes: results from the INSPIRE ME IAA study. J Clin Endocrinol Metab 2012; 97: 1517–1525.

    CAS  Article  Google Scholar 

  22. 22

    Liu J, Fox CS, Hickson DA, May WD, Hairston KG, Carr JJ et al. Impact of abdominal visceral and subcutaneous adipose tissue on cardiometabolic risk factors: the Jackson Heart Study. J Clin Endocrinol Metab 2010; 95: 5419–5426.

    CAS  Article  Google Scholar 

  23. 23

    Hayashi T, Boyko EJ, Leonetti DL, McNeely MJ, Newell-Morris L, Kahn SE et al. Visceral adiposity and the prevalence of hypertension in Japanese Americans. Circulation 2003; 108: 1718–1723.

    Article  Google Scholar 

  24. 24

    Fox CS, Massaro JM, Hoffmann U, Pou KM, Maurovich-Horvat P, Liu CY et al. Abdominal visceral and subcutaneous adipose tissue compartments: association with metabolic risk factors in the Framingham Heart Study. Circulation 2007; 116: 39–48.

    Google Scholar 

  25. 25

    Boyko EJ, Fujimoto WY, Leonetti DL, Newell-Morris L . Visceral adiposity and risk of type 2 diabetes: a prospective study among Japanese Americans. Diabetes Care 2000; 23: 465–471.

    CAS  Article  Google Scholar 

  26. 26

    Hajer GR, van Haeften TW, Visseren FL . Adipose tissue dysfunction in obesity, diabetes, and vascular diseases. Eur Heart J 2008; 29: 2959–2971.

    CAS  Article  Google Scholar 

  27. 27

    Reaven GM . The kidney: an unwilling accomplice in syndrome X. Am J Kidney Dis 1997; 30: 928–931.

    CAS  Article  Google Scholar 

  28. 28

    Gastaldelli A, Miyazaki Y, Pettiti M, Matsuda M, Mahankali S, Santini E et al. Metabolic effects of visceral fat accumulation in type 2 diabetes. J Clin Endocrinol Metab 2002; 87: 5098–5103.

    CAS  Article  Google Scholar 

  29. 29

    Zhang X, Shu XO, Li H, Yang G, Xiang YB, Cai Q et al. Visceral adiposity and risk of coronary heart disease in relatively lean Chinese adults. Int J Cardiol 2013 e-pub ahead of print 2013 doi:10.1016/j.ijcard.2013.01.275.

    Article  Google Scholar 

  30. 30

    Oh JY, Sung YA, Lee HJ . The visceral adiposity index as a predictor of insulin resistance in young women with polycystic ovary syndrome. Obesity (Silver Spring) 2012; 21: 1690–1694.

    Article  Google Scholar 

  31. 31

    Tanko LB, Bagger YZ, Qin G, Alexandersen P, Larsen PJ, Christiansen C . Enlarged waist combined with elevated triglycerides is a strong predictor of accelerated atherogenesis and related cardiovascular mortality in postmenopausal women. Circulation 2005; 111: 1883–1890.

    CAS  Article  Google Scholar 

  32. 32

    Sam S, Haffner S, Davidson MH, D'Agostino RB Sr., Feinstein S, Kondos G et al. Hypertriglyceridemic waist phenotype predicts increased visceral fat in subjects with type 2 diabetes. Diabetes Care 2009; 32: 1916–1920.

    CAS  Article  Google Scholar 

  33. 33

    Czernichow S, Bruckert E, Bertrais S, Galan P, Hercberg S, Oppert JM . Hypertriglyceridemic waist and 7.5-year prospective risk of cardiovascular disease in asymptomatic middle-aged men. Int J Obes (Lond) 2007; 31: 791–796.

    CAS  Article  Google Scholar 

  34. 34

    Arsenault BJ, Lemieux I, Despres JP, Wareham NJ, Kastelein JJ, Khaw KT et al. The hypertriglyceridemic-waist phenotype and the risk of coronary artery disease: results from the EPIC-Norfolk prospective population study. CMAJ 2010; 182: 1427–1432.

    Article  Google Scholar 

  35. 35

    Kip KE, Marroquin OC, Kelley DE, Johnson BD, Kelsey SF, Shaw LJ et al. Clinical importance of obesity versus the metabolic syndrome in cardiovascular risk in women: a report from the Women's Ischemia Syndrome Evaluation (WISE) study. Circulation 2004; 109: 706–713.

    Article  Google Scholar 

  36. 36

    Katzmarzyk PT, Janssen I, Ross R, Church TS, Blair SN . The importance of waist circumference in the definition of metabolic syndrome: prospective analyses of mortality in men. Diabetes Care 2006; 29: 404–409.

    Article  Google Scholar 

  37. 37

    Kahn HS, Cheng YJ . Longitudinal changes in BMI and in an index estimating excess lipids among white and black adults in the United States. Int J Obes (Lond) 2008; 32: 136–143.

    CAS  Article  Google Scholar 

  38. 38

    Wilson PW, Meigs JB, Sullivan L, Fox CS, Nathan DM, D'Agostino RB Sr. . Prediction of incident diabetes mellitus in middle-aged adults: the Framingham Offspring Study. Arch Intern Med 2007; 167: 1068–1074.

    Article  Google Scholar 

  39. 39

    Sattar N, Gaw A, Scherbakova O, Ford I, O'Reilly DS, Haffner SM et al. Metabolic syndrome with and without C-reactive protein as a predictor of coronary heart disease and diabetes in the West of Scotland Coronary Prevention Study. Circulation 2003; 108: 414–419.

    CAS  Article  Google Scholar 

Download references


We thank the China Health and Nutrition Survey, supported by the NIH (R01-HD30880, DK056350 and R01-HD38700), and the National Institute of Nutrition and Food Safety, China Center for Disease Control and Prevention, Carolina Population Center, the University of North Carolina at Chapel Hill and the Fogarty International Center for providing the data used here. We also thank the China-Japan Friendship Hospital and the Ministry of Health for support for the CHNS 2009 survey. The study was supported by National Natural Science Foundation of China 30772207.

Author information



Corresponding author

Correspondence to X Yu.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Du, T., Sun, X., Huo, R. et al. Visceral adiposity index, hypertriglyceridemic waist and risk of diabetes: the China Health and Nutrition Survey 2009. Int J Obes 38, 840–847 (2014).

Download citation


  • visceral adiposity index
  • hypertriglyceridemic waist phenotype
  • diabetes

Further reading


Quick links