Previous studies have identified ethnic specific differences in visceral adipose tissue (VAT), which may account for ethnic differences in cardio-metabolic risk. However, two distinctive sub-compartments of abdominal subcutaneous adipose tissue (SAT) have been recently identified that may also differ among ethnic groups. Therefore, the relationship between SAT compartments and body fat mass (BFM) between Aboriginal, Chinese, and South Asian cohorts compared to Europeans was investigated.
Design and Methods
Healthy Aboriginal, Chinese, European, and South Asian (n = 822) men and women (30-65 years) were assessed for BFM via dual energy X-ray absorptiometry, and SAT areas using computer tomography. SAT was subdivided into superficial SAT (SSAT) and deep SAT (DSAT) via the fascia-superficialis. Linear regression was performed using DSAT and SSAT as separate dependent variables and BFM and ethnicity as primary independent variables adjusting for confounders.
Aboriginal (181.0 cm2; p = 0.045) and South Asians (178.3 cm2; p = 0.013) had significantly higher amounts of DSAT, whereas the Chinese cohort had significantly less when compared with Europeans (114.3 cm2; p = <0.001). The Aboriginal cohort had a significantly higher amount of SSAT than Europeans (123.13 cm2 vs. 108.7 cm2; p = 0.04). Ethnicity modified the relationship between DSAT and BFM (p < 0.001 for interaction) such that Aboriginals and majority of South Asians had a significantly greater DSAT.
These data further demonstrate ethnic differences in body fat distribution such that Aboriginals and South Asians have greater amounts of DSAT. This may contribute to the increased cardio-metabolic risk in these groups.
Historically, non-Caucasian populations such as Aboriginal, Chinese, and South Asians were considered to be relatively free of obesity; however, the prevalence of overweight and obesity and the resulting cardio-metabolic consequences is increasing in these populations and worldwide ([1, 2]). In North American Aboriginal populations, the rates of obesity are nearly two-fold that of Europeans (). Obesity rates in China have also risen in both rural and urban populations, pushing the prevalence of overweight and obesity to over 60% amongst men and women between the ages of 18-44 (); furthermore, the prevalence of the metabolic syndrome has been reported to be 38.0% and 32.3% for Chinese men and women, respectively () Although in India the rate of obesity has been estimated to range between 13.3% and 45.9% in men and between 23.7% and 47.4% in women ([6, 7]), the prevalence of metabolic syndrome for South Asians living in locations worldwide is between 13.1% and 29% for men and between 11% and 32% for women (). Similarily, all three of these populations are experiencing increased rates of type 2 diabetes coincident with the rise in obesity ([2, 9]). Furthermore, with increasing migration from rural to urban, in addition to the advent of economic development and westernization of these cultures, the prevalence of obesity has risen significantly ([8, 10-13]).
It has also been suggested that Aboriginals, Chinese, and South Asians may have a particular senstivity to adverse effects of obesity, specifically abdominal obesity, as the prevalence of metabolic complications associated with obesity (dyslipidemia, insulin resistance, and type 2 diabetes) manifest at smaller BMI (calculated as weight in kilograms divided by height in meters squared [kg/m2]), waist circumferences (WCs) and total body fat masses (BFMs) ([14-18]). Although this notion has been recognized with the introduction of lower WC recommendations for South Asians and Chinese by the International Diabetes Federation in the context of metabolic syndrome (), the etiology behind the susceptibility of these ethnic populations to these metabolic disturbances is not clearly understood.
The higher propensity to develop cardiovascular disease (CVD) and type 2 diabetes is often explained by excessive accumulation of visceral adipose tissue (VAT). We have previously reported that Chinese and South Asians have a greater proportion of VAT for a given total body fat and WC compared with Europeans; in contrast, we found no significant differences between the Aboriginals and the Europeans ([20, 21]). However, in addition to VAT, a select few studies have recently identified that abdominal subcutaneous adipose tissue (SAT) sub-compartments may also play an integral role in the metabolic disturbances associated with excessive adipose tissue ([22-24]). The SAT compartment can be differentiated into subcutaneous abdominal adipose tissue: superficial SAT (SSAT) and deep SAT (DSAT) segregated by the abdominal wall fascia (scarpa's fascia), respectively ([25, 26]). The rationale for the division of these compartments stems from studies indicating that in addition to being anatomically different, they are also histologically and morphologically different, as there is an apparent progression toward irregularity of the adipose cells and greater vascularization as you proceed from the outer most compartment, SSAT, to DSAT and then to VAT (). Therefore, differences in these two adipose tissue depots may be important to understanding cardiometabolic risk in different populations. Given the previous findings of increased cardiometabolic risk in Aboriginal, Chinese, and South Asian populations, the aim of this study is to compare the SAT compartments of a European cohort with that of Aboriginal, Chinese, and South Asian cohorts. We hypothesize that the amount of DSAT assessed by computer tomography (CT) and corrected for total BFM is greater in men and women of Aboriginal, Chinese, and South Asian origin than Europeans whereas there would be no difference in SSAT among these groups.
Participants were recruited (n = 822) as a part of the Multicultural Health Assessment trial (M-CHAT), which was designed to investigate body fat distribution in four different ethnic groups and has been described in detail (). Healthy men and woman (aged between 30 and 65) with exclusive ancestry of either Aboriginal (reserve and nonreserve residents), Chinese (China, Hong Kong, and Taiwan), European (continental Europe, Ireland, and the United Kingdom), or South Asian (Bangladesh, India, Nepal, Pakistan, and Sri Lanka) origin were potential candidates and were screened for eligibility. Ethnicity was determined by self-report. Participants within each ethnic group were purposely recruited to ensure equivalent numbers of men and women across a range of BMI values: <24.9, 25-29.9, and ≥30 kg/m2. Chinese, European, and South Asian participants must have resided in Canada for more than 3 years and be third-generation Canadian or less. Participants were excluded if they had self-reported CVD and/or diabetes, undergone recent weight change (greater than 2 kg in 3 months prior to assessment date), if participants were taking medications known to affect type 2 diabetes and CVD risk factors (lipid lowering, antihypertensive, hypoglycemic, hormone replacement therapy, or insulin), or if they had significant prosthetics or amputations. Women who were pregnant or lactating were also excluded from the study (). All study participants gave informed consent. The study was approved by the Simon Fraser University Research Ethics board.
Participants were assessed for sociodemographics, family history of CVD and type 2 diabetes, anthropometry, and lifestyle factors. Weight and height were assessed with participants in light street clothing (or hospital gowns), footwear removed, and pockets emptied. WC was recorded in centimeters as the average of two measures taken against the skin at the point of maximal narrowing from the anterior view following a normal expiration. Hip circumference was recorded in centimeters, as the average of two measures taken at the point of maximal gluteal protuberance from the lateral view over undergarments. Humerus breadth was the distance between the medial and lateral epicondyles of the humerus and used as a proxy for body frame size. Smoking status and alcohol consumption were assessed by self-report. Leisure-time physical activity was also assessed via self-report and was recorded in minutes per week of activity over the previous year using the Modifiable Activity Questionnaire ().
Body composition assessment
Abdominal adipose tissue was assessed using CT from a single cross-sectional CT scan centered at the L4/L5 intervertebral disc. All scans were performed with a CTi Advantage Scanner (General Electric, Milwaukee, WI). Scan parameters were set at 120 kVp, 300 mA for 1 second, 10 mm thickness, 512 × 512 matrix, using a 48 cm field of view. Computation of surface areas was conducted using medical imaging software, SliceOmatic v.4.2 (Tomovision, Montreal), using the attenuation range of −190 to −30 Hounsfield units for adipose tissue. Total abdominal fat was calculated as total pixels and area in square centimeters (cm2) in this attenuation range in the abdominal slice image. VAT was calculated as total pixels and area within this attenuation range that falls within the visceral peritoneum (abdominal wall), and SAT was calculated as the difference between total abdominal adipose tissue and VAT. The SAT was then subdivided into SSAT and DSAT via identifying the fascia superficialis (scarpa's fascia) (). The coefficients of variation for the analysis from the same observer and between observers were 1.2% and 1.1%, respectively, for the VAT compartment, 4.9% and 5.7% respectively, for the SSAT compartment, and 4.5% and 4.3%, respectively, for the DSAT compartment. Total BFM and free-fat mass was measured using dual-energy X-ray absorptiometry with a Norland XR-36 scanner (Norland Medical Systems, White Plains, NY) using Host Software Version 2.1.0. Percentage body fat was calculated from the above data by dividing BFM by total body mass.
Normally distributed continuous variables are reported as means ± SD. Nonnormal variables were log-transformed and presented in the form of medians and quartile distributions. Categorical data were represented in the form of frequencies and percentages. To test for differences between groups, we used analysis of variance for continuous variables and the χ2 test for categorical variables. Bivariate Pearson correlation analyses were used to assess the association between between each of SSAT and DSAT with physical activity within each ethnicity. A DSAT/SSAT ratio was also derived to describe the propensity to store adipose tissue in the DSAT compartment.
Age adjusted and segregated by sex, values for DSAT and SSAT were calculated for BMI values of 25 kg/m2 and 30 kg/m2 as well as for WC cut-off values of 90 cm and 102 cm for men, and 80 cm and 88 cm for women for each ethnic group. We also investigated BMI × ethnicity and WC × ethnicity interactions to determine whether the relationship between DSAT or SSAT with the anthropometric measures differed by ethnicity; however, neither of these interactions were significant (Supporting Information Table 1). Separate linear regression analyses were performed using DSAT and SSAT as the dependent variable with BFM and ethnicity as primary independent variables adjusting for age, sex, household income, educational status, physical activity, humerus breadth, and smoking status. To test the association between DSAT and BFM, we analyzed the interaction effect for ethnicity X BFM. We also tested for ethnicity × sex, sex × BFM, and sex × ethnicity × BFM interactions, none of which were significant (Supporting Information Table 1). The associations of SSAT with BFM were also modeled in the similar manner. The final models were based on 793 participants as 29 CT scans were eliminated as the fascia line or the CT scan itself was unreadable. All hypothesis tests for regression analyses were two-sided with significance level of p = 0.05. We used the Bonferroni adjustment for multiple comparisons between each of Aboriginal, Chinese, and South Asian groups with the European group (p set at 0.0167). All statistical analyses were conducted using SPSS Windows 15.0 (SPSS, Chicago, IL).
Demographics are provided in Table 1 for the 201 European, 195 Aboriginal, 219 Chinese, and 207 South Asian participants in this study. Smoking status, alcohol intake, and physical activity differed significantly among all four ethnic groups. Although there were no diabetic subjects in the study cohort, family history of diabetes differed significantly, whereas family history of CVD did not differ between ethnic groups. There were no differences in the menopausal status of the women (26% postmenopausal). Physical activity was weak, but significantly associated with SSAT in all four ethnic groups and was weak but significantly associated with DSAT in all ethnic groups except for the Aboriginal group (Supporting Information Table 2).
Table 1. Study demographics stratified by ethnicity
In unadjusted analysis, significant differences were observed among the four ethnic groups with respect to anthropometric and adipose tissue measurements in Table 2. The Chinese cohort had a lower mean BMI than did the other groups. Aboriginals had a greater mean WC and waist-to-hip ratio than Europeans, whereas the Chinese had a lower WC than did the Europeans. In addition, Chinese and South Asians had smaller humerus breadth than their European counterparts. Aboriginals and South Asians also had a significantly greater percentage body fat than did the Europeans. Aboriginals and South Asians had significantly greater total adipose tissue (TAT), SAT, SSAT, and DSAT than Europeans, and Chinese had less TAT, SAT, SSAT, and DSAT than Europeans. The mean DSAT/SSAT was lower in the Chinese cohort compared with that in Europeans. VAT was higher in South Asians compared with that in Europeans.
Table 2. Body composition stratified by ethnicity
Aboriginal (n = 195)
European (n = 201)
Chinese (n = 219)
South Asian (n = 207)
PP for overall comparison
PP for Aboriginal vs. European
PP for Chinese vs. European
PP for South Asian vs. European
avalues are presented as median; 25th and 75th percentiles in parentheses.
BMI, = body mass index,; SSAT, = superficial subcutaneous adipose tissue,; DSAT, = deep subcutaneous adipose tissue,; VAT, = vis ceral adipose tissue.
a Values are presented as median: 25th and 75th percentiles in parentheses.
Figure 1 shows the age- and sex-adjusted differences for DSAT (Figure 1A) and SSAT (Figure 1B) at BMI targets for overweight and obese for Aboriginal, Chinese, and South Asian cohorts compared with the European cohort. At both BMI values for overweight and obese, the South Asian and Aboriginal cohort had higher DSAT, whereas the Chinese had a lower DSAT than the European cohort. There were no significant differences in SSAT at either BMI targets for both Aboriginal and Chinese cohorts compared with the Europeans; however, South Asians had a significantly higher amount of SSAT at both BMI targets compared with Europeans.
Figure 2 depicts age-adjusted differences in DSAT in men (Figure 2A) at WC targets of 90 cm and 102 cm and in women (Figure 2B) at WC targets of 80 cm and 88 cm. At both WC values, South Asian men (Figure 2A), and women (Figure 2B) had significantly greater amounts of DSAT when compared with Europeans counterparts. Similar results were found with respect to SSAT, in that at both WC values, South Asian men and women also had significantly greater amounts of SSAT than Europeans (data not shown).
After adjustment for age, sex, BFM, and smoking status, Aboriginal and South Asian men and women had more DSAT compared with European men and women (Figure 3), whereas there were no significant differences in SSAT between any of the non-European groups with the Europeans (Figure 3 B).
Figure 4 indicates the distribution of DSAT over a range of BFM adjusted for ethnicity, age, sex, maximum educational level, humerus breadth, smoking status, and physical activity (Figure 4). In the fully adjusted model, ethnic background was a significant modifier of relationship between DSAT and total body fat (overall ethnicity × BFM interaction; p < 0.001). Aboriginal participants had greater amounts of DSAT at levels of BFM below 50 kg compared with European participants (Figure 4A). No differences were observed for the relation of DSAT to total BFM between Chinese and European participants (Figure 4B). The slopes of the South Asian and European groups (Figure 4C) crossed at approximately 47.4 kg BFM such that South Asians had a greater amount of DSAT than Europeans below this value but less DSAT above this value. The majority of South Asian participants in this study (95.6%) had a total body fat <47.4 kg, indicating that most of the South Asians in our study had a greater proportion of DSAT at a given total BFM than their European counterparts.
This study was undertaken to examine the hypothesis that the amount of superficial and deep subcutaneous abdominal adiposity, defined anatomically via the scarpa's facia that divides the two depots, differs between Aboriginal, Chinese, and South Asian ethnic groups compared with a European group. Results of this study clearly support the notion that body fat distribution, as measured by SSAT and DSAT, differs according to ethnicity. Overall, South Asians had greater amounts of DSAT than Europeans at the same BMI and WC targets, and over a range of BFM. In addition, Aboriginals also had greater DSAT at the BMI targets and across a range of BFM. Although Chinese participants had lower DSAT at the BMI targets, there was no difference between the Chinese and European participants in DSAT regardless of BFM.
Previous research has demonstrated that individuals of Aboriginal, Chinese, and South Asian descent have a higher prevalence of cardio-metabolic complications associated with obesity (type 2 diabetes and CVD) compared with Europeans ([15, 16, 18, 29, 30]). It has been postulated that this increased risk may be because of an increase in VAT, which has been demonstrated to be more metabolically active than other fat depots (). Indeed, the M-CHAT study indicated that VAT in South Asian and Chinese men and women was greater, even at the same level of body fat; when compared with Europeans, in contrast, no differences where observed between Aboriginals and Europeans (). However, a handful of studies have begun to identify, that in addition to the VAT, the distribution of adipose tissue within the DSAT and SSAT regions may also have a role with respect to an individual's cardiometabolic risk associated with obesity ([23, 32]).
A small number of studies indicate that the DSAT region is associated with greater risk ([23, 32, 33]), such that DSAT has been reported to be associated with lipid abnormalities in healthy women, which is consistent with that reported in individuals with excessive VAT (). A second study conducted by Kelley et al (), demonstrated that DSAT strongly correlated with measures of insulin resistance and triglyceride levels, thereby further contributing to the cluster of lipid and glycemic abnormalities seen in those with visceral adiposity in both men and women. Furthermore, Walker et al. () demonstrated that DSAT reflected similar expression profile to VAT in terms of 11 beta-hyroxysteriod dehydrogenase type 1 (an enzyme in the cortisol synthesis pathway), leptin, and resitin. Together, these studies support the notion that excessive accumulation of DSAT results in adverse cardiometabolic risk profiles.
In Aboriginals, the prevalence of type 2 diabetes is greater than that of other populations, and it has been speculated that this may be due in part to the greater prevalence of abdominal obesity, and specifically VAT ([9, 34]). Unexpectedly, others and we have found no difference in VAT between Aboriginals and Europeans ([27, 35]). In this study, Aboriginals had greater amounts of DSAT at the BMI cut-offs for overweight and obesity, as well as at any given level of BFM than the European participants, whereas there were no differences in SSAT. To our knowledge, no previous studies have been conducted to demonstrate differences in DSAT in Aboriginal populations. Although we did not report on measures of glucose control and insulin resistance, the DSAT fat depot has shown to be correlated with increased levels of insulin resistance (). Therefore, elevated DSAT levels may in part explain elevated rates of type 2 diabetes in Aboriginal populations.
In contrast to the findings in Aboriginals, DSAT levels were actually lower in our Chinese participants than in Europeans; however, this difference was no longer apparent in the fully adjusted model. In addition, we also found there were no differences in SSAT between these two groups. This was an unexpected finding as previous studies have found that individuals of Chinese descent have greater cardiometabolic risk factors at smaller BMI and WC than that of Europeans ([15, 18]). However, it is possible that the higher cardiometabolic risk is attributed to the greater amount of VAT at a given value of body fat than Europeans ().
It has been well-documented that South Asians have a greater risk for type 2 diabetes and CVD, and this may be the result of having greater cardiometabolic risk factors than other populations (). In addition, South Asians have higher amounts of body fat and VAT than other populations as well as greater risk factors at a given BMI and/or WC ([15, 16, 18, 20, 29]). Our finding that South Asians have greater amounts of DSAT than Europeans adds to this unique high-risk body composition phenotype of South Asians. Specifically, South Asians had greater amounts of DSAT and SSAT at the BMI thresholds for overweight and obesity, and at the WC thresholds of 90 cm and 102 cm than Europeans. In addition, South Asians had a greater amount of DSAT across a range of lower BFM, but not at higher BFM compared with Europeans. However, as we oversampled for obese participants, it is likely that the majority of South Asians are likely to have BFM values that fall in the ranges with relatively greater amounts of DSAT than the Europeans, and therefore are at greater cardiometabolic risk.
The mechanisms by which body fat distribution may differ amongst ethnic groups are presently unknown. However, a number of hypotheses exist that suggest the development of dysfunctional adipocytes through intrauterine origins (including epigenetics) and a lower threshold for storing adipose tissue in subcutaneous stores ([22, 37]). Although these hypotheses have been predominantly put forward to address the unique phenotype of South Asians, they may also be relevant in explaining the differences in body fat in other ethnic groups. Our results that Aboriginal and South Asians have greater amounts of DSAT than Europeans lend support to these hypotheses.
We purposely recruited across a range of BMI values and not a random population sample as this study was designed to investigate differences across a range of body fat distribution. However, there is no reason to believe that these results are not applicable to the general population given that the primary focus of this study was to investigate superficial adipose tissue depots, and therefore, a self-selection bias is doubtful. Furthermore, the four cohorts were defined based on commonly used definitions to distinguish different ethnic populations. It is recognized that within each ethnic cohort, cultural and possibly physiological differences may occur. However, the differences within the group are likely to be minor in comparison to differences between the two ethnic cohorts and therefore should not have affected the main hypothesis and the internal validity. Furthermore, participants self-described their ethnic origin, and therefore, we were unable to rule out possible inaccuracies that can only be confirmed through genetic testing. We must also acknowledge the lack of consideration of dietary information in our analyses as diet is a relevant factor with respect to body composition. Lastly, as this study was cross-sectional in design, it cannot give insight into changes over time in these ethnic groups. Follow-up studies would is needed to determine whether accumulation of DSAT and SSAT differs over time.
This study demonstrates that the distribution of SSAT depots differs by ethnic background. We found that Aboriginal and South Asians have greater amounts of the more cardiometabolically active DSAT region compared with Europeans, whereas there was no difference in DSAT levels between the Chinese and European participants. Specifically, Aboriginals and South Asians tend to have greater amounts of DSAT than Europeans, which may suggest a preferential deposition of adipose tissue in the DSAT dept. In addition, there were relatively little differences in the more metabolically benign SSAT depot between the non-European and European groups. This study provides novel insight into a complicated puzzle as it contributes to a body of evidence investigating the understanding of the unique susceptibility of Aboriginals and South Asians to complications of central adiposity and high prevalence of CVD and type 2 diabetes. These differences merit considerable study in order to further our understanding of ethnic and gender health disparities. Future studies will need to be conducted to determine whether these higher levels of DSAT do indeed result in greater cardiometabolic risk. If true, then SAT in the abdominal region cannot be viewed as a homogeneous region and must be considered along with VAT when assessing the risk associated with abdominal obesity.
Dr. Lear is the Pfizer/Heart and Stroke Foundation Chair in Cardiovascular Prevention Research at St. Paul's Hospital.