Simple anthropometric indices in relation to cardiovascular risk factors in Chinese type 2 diabetic patients.

Abstract

To determine which is the best anthropometric index among body mass index (BMI), waist circumference (WC), waist to hip ratio (WHR) and waist to height ratio (WHtR) in type 2 diabetic patients, we examined the relationship between these indices and cardiovascular risk factors using partial correlation analysis, chi-square test, logistic regression analysis and Receiver Operator Characteristic (ROC) curves. Partial correlation analysis showed that among the 4 obesity indices, WHtR had the highest r values for all the cardiovascular risk factors in both sexes, followed by WC. Chi-square analysis which revealed that an increased WHtR was more strongly associated with hypertension, hypertriglyceridemia (high TG) and low high-density lipoprotein cholesterol (HDL-C) than the other indices. Logistic regression analysis showed that, after controlling for age, the hypertension, high TG and low HDL-C odds ratios of WHtR > or = 0.5 were 2.56 (95% CI: 1.24, 5.29), 2.87 (95% CI: 1.43, 5.78), 2.59 (95% CI: 1.03, 6.59) in men and 3.75 (95% CI: 1.75, 8.05), 3.21 (95% CI: 1.52, 6.79), 3.62 (95% CI: 1.43, 9.21) in women, respectively. In ROC analysis, the areas under curve of WHtR were the largest for at least one risk factor in both men and women. These results indicated that WHtR had a higher correlation with cardiovascular risk factors than WC, WHR or BMI in newly diagnosed type 2 diabetes. We proposed the measurement of WHtR as a screening tool for cardiovascular risk factors in this population.

Full text

Page 1

Corresponding author: Dr. Lu-Lu Chen, Department of Endocrinology, Union Hospital, 1277 Jiefang Avenue, Wuhan 430022, Hubei, P.R.
China. Tel: +86-27-85801744, Fax: +86-27-85356365, E-mail: hongyzn_wu2003@yahoo.com.cn
Received: November 22, 2006; Revised: January 25, 2007; Accepted: March 15, 2007.
Chinese Journal of Physiology 50(3): 135-142, 2007
135
Simple Anthropometric Indices in Relation to
Cardiovascular Risk Factors in Chinese Type 2
Diabetic Patients
Hong-Yan Wu1, 2, Lu-Lu Chen1, Juan Zheng1, Yun-Fei Liao1, and Min Zhou1
1Department of Endocrinology, Union Hospital, Tongji Medical School,
Huazhong Technology University,
Wuhan 430022, Hubei
and
2Department of Endocrinology, Jingzhou No.1 People’s Hospital,
Yangtze University,
Jingzhou 434000, Hubei, P.R. China
Abstract
To determine which is the best anthropometric index among body mass index (BMI), waist
circumference (WC), waist to hip ratio (WHR) and waist to height ratio (WHtR) in type 2 diabetic
patients, we examined the relationship between these indices and cardiovascular risk factors using
partial correlation analysis, chi-square test, logistic regression analysis and Receiver Operator
Characteristic (ROC) curves. Partial correlation analysis showed that among the 4 obesity indices,
WHtR had the highest r values for all the cardiovascular risk factors in both sexes, followed by WC. Chi-
square analysis which revealed that an increased WHtR was more strongly associated with hypertension,
hypertriglyceridemia (high TG) and low high-density lipoprotein cholesterol (HDL-C) than the other
indices. Logistic regression analysis showed that, after controlling for age, the hypertension, high TG
and low HDL-C odds ratios of WHtR ≥ ≥ ≥ ≥ ≥ 0.5 were 2.56 (95%CI: 1.24, 5.29), 2.87 (95%CI: 1.43, 5.78),
2.59 (95%CI: 1.03, 6.59) in men and 3.75 (95%CI: 1.75, 8.05), 3.21 (95%CI: 1.52, 6.79), 3.62 (95%CI:
1.43, 9.21) in women, respectively. In ROC analysis, the areas under curve of WHtR were the largest for
at least one risk factor in both men and women. These results indicated that WHtR had a higher
correlation with cardiovascular risk factors than WC, WHR or BMI in newly diagnosed type 2 diabetes.
We proposed the measurement of WHtR as a screening tool for cardiovascular risk factors in this
population.
Key Words: obesity, waist to height ratio, diabetes mellitus, cardiovascular risk factors
Introduction
The leading cause of mortality in type 2 diabetes
is atherosclerotic vascular disease. Obesity is a well-
known risk factor for type 2 diabetes mellitus and
cardiovascular disease. Simple anthropometric
measurements have been used as surrogate measurements
of obesity and have more practical value in both clinical
practice and for large-scale epidemiological studies.
Body mass index (BMI) is the most widely used and
a simple measure of body size, and is frequently used
to estimate the prevalence of obesity within a population.
It is now established that central adiposity shows stronger
associations with cardiovascular disease risk and type
2 diabetes mellitus than overall adiposity in general
population (5, 25). Thus, measurements of waist
circumference (WC) and waist to hip ratio (WHR)
have been viewed as alternatives to BMI and are regularly
used in the clinical and research settings. WC has
been shown to be the best simple measure of both
intra-abdominal fat mass and total fat (8, 17), but it
was reported (13) that the metabolic risk between people
of similar WC with different heights is different. Recent
epidemiological studies (10, 14) have shown that waist

Page 2

136
WU, CHEN, ZHENG, LIAO AND ZHOU
to height ratio (WHtR) is an effective abdominal obesity
index in predicting the risk of diabetes and coronary
heart disease in the general population. Its predicting
power is better than WC and WHR. However, few
studies have examined the feasibility and effectiveness
of WHtR as an abdominal obesity index in predicting
the cardiovascular risk factors in Chinese adults with
type 2 diabetes mellitus. The aim of this study was to
compare the four indices in relation to cardiovascular
risk factors in Chinese adults with newly diagnosed
type 2 diabetes mellitus to determine which would be
the best.
Materials and Methods
Patients
The study group was composed of patients who
were newly diagnosed with type 2 diabetes at the
Diabetic Clinic of Union Hospital (a teaching hospital
of Tongji Medical School, Huazhong Technology
Univerity in Wuhan City, Hubei, China). This clinic
provides treatment to approximately 15% of the
diabetic population of Wuhan and its surrounding areas
of central south of China, including both rural and
urban area. Type 2 diabetes was diagnosed using the
ADA (1997) criteria (fasting glucose ≥ 7.0 mM or
postload glucose ≥ 11.1 mM)1. Patients with any of
the following criteria were excluded: age < 40, diabetes
history, or diabetic complications such as coronary
artery disease, peripheral arterial disease or previous
stroke, for the reason that a previous diagnosis of
diabetes or macrovascular complications could have
induced a modification of life style or specific
therapeutic interventions interfering with the assessed
parameters of adiposity. No patient had previously
received insulin or oral hypoglycemic drugs or
diuretics therapy that would also influence weight.
Four hundred and forty-eight Chinese patients were
recruited from January 2003 to March 2006. Further
exclusion of the patients lacking complete
anthropometrical data left 411 patients for the analysis.
In general, the subjects had a middle-to-high education
level. All the patients had signed informed consent.
Measurements
Participants were seen after a 12-h fast. The
interviewers were trained to measure weight and height
according to the World Health Organisation (WHO,
1987) standards2. Height and weight were measured
after the removal of shoes and with the patients wearing
light clothing. BMI was calculated as weight (kg)/
[height (m)]2. Both WC and hip circumference were
measured to the nearest 0.1 cm in triplicate with a
flexible tape before the average value was calculated.
WC was measured at the mid-point between the distal
border of the ribs and the top of the iliac crest with
subjects standing at the end of a normal expiration.
Hip circumference was measured at the widest point
over the buttocks. The ratio of waist to hip (WHR)
and waist to height (WHtR) was calculated. Obesity
was defined as a BMI value ≥ 25 kg/m2 according to
the Asia Pacific criteria3 or WC ≥ 85 cm for men and
≥ 80 cm for women (19, 28) or WHR ≥ 0.9 for men and
≥ 0.85 (1) for women or WHtR ≥ 0.5 in either sex (10,
16).
Sitting blood pressure was measured twice,
and the mean reading was used. Participants were
classified as hypertensive if they were on treatment
for hypertension, had a mean systolic reading (SBP)
≥ 140 mmHg or a mean diastolic reading (DBP) ≥ 90
mmHg.
Biochemical Analyses
Levels of fasting plasma glucose (FPG), serum
triglyceridemia (TG), cholesterol, and high-density
lipoprotein cholesterol (HDL-C) were measured by
enzymatic methods with an autoanalyzer (Hitachi
7170A, Tokyo, Japan) and manufacture’s reagent kits.
Dyslipidemia was defined as self-reported current
treatment with TG-lowering or HDL-C-raising
medication or having one or more of the following:
TG ≥ 1.7 mM, HDL-C ≤ 0.9 mM for men and ≤ 1.0
mM for women. Glycosylated hemoglobin (HbA1c)
was measured by liquid chromatographic methodology
using DiaSTAT(tm) (Bio-Rad Laboratories, Hercules,
CA, USA).
Statistical Analysis
Partial correlation analysis was performed
between cardiovascular risk factors and anthropometric
indices after adjusting for age as a continuous variable.
Comparison between genders was performed using
Student’s t-tests. Chi square test was used to compare
the prevalence of cardiovascular risk factors with BMI,
WC, WHR and WHtR as indices of obesity. Logistic
regression models were applied to calculate odds ratios
1Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care 20:1183-1197, 1997.
2Measuring obesity-classification and description of anthropometric data. Report on a WHO consultation of the epidemiology of obesity, Warsaw,
21-23. October, 1987.
3The Asia-Pacific Perspective: Redefining Obesity and its Treatment. World Health Organization, Western Pacific Region. Geneva: World Health
Organization, 2000.

Page 3

ANTHROPOMETRIC INDICES AND CARDIOVASCULAR RISK FACTORS
137
(OR) for the presence of cardiovascular risk factors
(dependent variables) after adjustment for age, obesity
indices treated as categorical variables using the selected
cut-off points were considered as independent variables.
To further evaluate the accuracy of obesity indices for
assessment of at least one metabolic risk, we calculated
sensitivity and specificity of obesity indices using
receiver operating characteristic (ROC) curves. The
area under the ROC curve (AUC) provides a single
measure of overall accuracy that is not dependent upon
a particular threshold. AUC for the ROC curves and
comparison of ROC curves were performed using
MedCalc (MedCalc Software, Mariakerke, Belgium).
Other statistical analyses were performed with SPSS
13.0. The level of significance for all statistical tests
of hypotheses was set at P < 0.05. TG values were
transformed to the natural logarithm to normalize
skewed distribution for statistical testing; however,
actual values are displayed.
Results
Characteristics of the Study Population
Subjects’ characteristics regarding age, body
composition and metabolic risk profile are given in
Table 1. According to the BMI cut-off of 25 kg/m2, 97
males (49.0%) and 113 females (53.1%) were obese,
whereas the prevalence of central obesity was 74.2%
in men and 72.8% in women using the WHtR cut-off
of 0.5. The distribution of BMI, WC, WHR and WHtR
were normal. Table 2 shows the 25th, 50th, 75th, 90th
percentile values of the four obesity indexes for men
and women. Anthropometric indices for women were
Table 1. Clinical and metabolic characteristics
Variables Male (n =198)
53.9 ± 8.3
72.4 ± 10.0
169.6 ± 5.6
25.1 ± 2.9
149 (75.3%)
97 (49.0 %)
90.6 ± 8.1
144 (72.7%)
0.53 ± 0.05
147 (74.2%)
0.92 ± 0.05
20 (60.6%)
8.11 ± 1.64
8.65 ± 2.64
131.4 ± 19.8
82.3 ± 13.0
77 (38.9%)
2.21 ± 1.19
90 (45.5%)
1.22 ± 0.40
44 (22.2%)
Female (n = 213)
54.5 ± 8.5
61.4 ± 8.9*
157.6 ± 4.9*
24.7 ± 3.2
146 (68.5%)
113 (53.1%)
85.0 ± 8.3*
157 (73.7%)
0.54 ± 0.06
155 (72.8%)
0.87 ± 0.06*
130 (61.0%)
7.79 ± 2.06
8.44 ± 2.39
131.6 ± 16.0
79.2 ± 9.9*
82 (38.5%)
2.03 ± 1.39
86 (40.4%)
1.21 ± 0.60
61 (28.6%)
Age (yr)
Weight (kg)
Height (cm)
BMI (kg/m2)
BMI ≥ 23
BMI ≥ 25
WC (cm)
WC ≥ 85/80 cm
WHtR
WHtR ≥ 0.5
WHR
WHR ≥ 0.9/0.85 1
FPG (mM)
HbA1C (%)
SBP (mmHg)
DBP (mmHg)
BP ≥ 140/90 mmHg
TG (mM)
TG ≥ 1.7 mM
HDL-C (mM)
HDL-C ≤ 0.9/1.0 mM
Data are presented as means ± SD or n (%). *P < 0.05 for comparison between male and female groups.
Table 2. Distribution of obesity indices among newly diagnosed type 2 diabetics
Percentile, Male
50th
Percentile, Female
50th 25th75th 90th25th75th90th
BMI (kg/m2)
WC (cm)
WHtR
WHR
23.0
84.4
0.50
0.89
24.9
90.3
0.53
0.92
26.9
95.6
0.57
0.96
28.7
100.5
0.60
0.98
22.2
79.5
0.49
0.84
24.7
85.2
0.54
0.87
26.9
91.0
0.58
0.92
28.7
95.0
0.61
0.95

Page 4

138
WU, CHEN, ZHENG, LIAO AND ZHOU
always lower than those for men for height, weight,
WC and WHR, except for WHtR and BMI for which
the gender ratio was closest to 1. The prevalences of
hypertension, high TG and low HDL-C were 38.9%,
45.5%, 22.2% in men and 38.5%, 40.4%, 28.6% in
women, respectively. Sex differences in metabolic
risk factors were observed only for DBP (with lower
values in women).
Correlation between Obesity Measurements and
Cardiovascular Risk Factors
Table 3 shows the age-adjusted partial
correlation coefficients between anthropometric
indices and cardiovascular risk factors. WHtR had
the highest coefficients for all the risk factors in both
sexes, followed by WC. The correlations between
WHR and both of blood pressure and lipidemia were
weakest. HbA1c and FPG showed a significant inverse
correlation with WHtR, WC and BMI. None of the
four indices of adiposity showed a significant
correlation with TC and LDL-C (data not shown).
Prevalence of Cardiovascular Risk Factors According to
BMI, WC, WHR and WHtR
Table 4 shows the prevalence of cardiovascular
risk factors according to the four obesity indices,
respectively. An increased WHtR was significantly
associated with all 3 risk factors (hypertension,
high TG and low HDL-C) in both sexes (all P < 0.01,
except for low HDL-C P < 0.05). An increased WC
was also significantly associated with all 3 risk factors
(in men, all P < 0.05, except for low HDL-C P > 0.05;
in women, all P < 0.01, except for low HDL-C P <
0.05). BMI was only significantly associated with
hypertension and high TG in both sexes (all P < 0.05),
whereas an increased WHR was only significantly
associated with hypertension and high TG in men (all
P < 0.05).
Table 4. Prevalence of cardiovascular risk factors according to BMI, WC, WHR and WHtR
Male
High -TG
n (%)
Female
High-TG
n (%)
Variables Hypertention
n (%)
Low-HDL-C
n (%)
Hypertentionn
n (%)
Low-HDL-C
n (%)
BMI ≥ 25kg/m2
BMI < 25 kg/m2
P value
WC ≥ 85/80 cm
WC < 85/80 cm
P value
WHtR ≥ 0.5
WHtR < 0.5
P value
WHR ≥ 0.9/0.85
WHR < 0.9/0.85
P value
45(46.4)
32(31.7)
0.034
63(43.8)
14(25.9)
0.022
65(44.2)
12(23.5)
0.009
55(45.8)
22(28.2)
0.013
52(53.6)
38(37.6)
0.024
73(50.7)
17(31.5)
0.016
76(51.7)
14(27.5)
0.003
62(51.7)
28(35.9)
0.029
26(23.2)
18(20.9)
0.224
37(25.7)
7(13.0)
0.055
38(25.9)
6(11.8)
0.037
30(25.0)
14(17.9)
0.244
51(45.1)
31(31.0)
0.034
69(43.9)
13(23.2)
0.006
72(46.5)
10(17.2)
< 0.001
56(43.1)
26(31.3)
0.086
54(47.8)
32(32.0)
0.019
73(46.5)
13(23.2)
0.002
75(48.4)
11(19.0)
< 0.001
56(43.1)
30(36.1)
0.315
38(33.6)
23(23.0)
0.087
51(32.5)
10(17.9)
0.038
55(35.5)
6(10.3)
< 0.001
41(31.5)
20(24.1)
0.241
Table 3. Partial correlation coefficients between obesity measurements and cardiovascular risk factors
VariablesMale Female
BMIWC WHtRWHRBMIWCWHtR WHR
SBP
DBP
lgTG
HDL-C
HbA1c
FPG
0.16*
0.22**
0.23**
-0.07
-0.15*
-0.16*
0.22**
0.23**
0.31**
-0.15*
-0.20**
-0.16*
0.27**
0.24**
0.35**
-0.18**
-0.22**
-0.19**
0.09
0.11
0.09
-0.10
-0.11
-0.12
0.12
0.25**
0.29**
-0.09
-0.24**
-0.22**
0.16*
0.25**
0.33**
-0.23**
-0.25**
-0.24**
0.18**
0.25**
0.35**
-0.28**
-0.26**
-0.23**
0.09
0.10*
0.14*
-0.07
-0.21**
-0.18**
Adjusted for age. *P < 0.05, **P < 0.01

Page 5

ANTHROPOMETRIC INDICES AND CARDIOVASCULAR RISK FACTORS
139
Multivariate Logistic Regression Analysis
Table 5 shows adjusted ORs and 95% confidence
intervals (CIs) of the 4 obesity indices for the presence
of cardiovascular risk factors. Subjects above the
selected cut-offs had higher risk for hypertension and
dyslipidaemia. Both men and women with WHtR ≥
0.5 had the highest ORs for hypertension, high TG
and low HDL-C.
Accuracy of Obesity Indices for Assessment of One or
More Risks Using ROC Curve Analysis
Tables 6 and 7 display the AUC for each
anthropometric index as a predictor of at least one
risk factor (hypertension, high TG or low HDL-C)
and comparison of the AUC. One hundred and forty-
three males (72.2%) and 150 females (70.4%) had at
least one risk factor. The AUC for WHtR was the
largest in both sexes, although the difference between
WHtR and WC is marginal in men (P = 0.057). The
differences of AUC between BMI and WC and between
WC and WHR were not significant. The optimal cut-
off values in men and women were 24.5 and 24.9 kg/
m2 for BMI, 83 and 81 cm for WC, 0.89 and 0.86 for
WHR, 0.51 and 0.50 for WHtR, respectively. The
ROC curves of the four anthropometric indices in
relation to one or more risk factors in men and women
are given in Figures 1 and 2.
Discussion
Diabetic patients tend to have an android pattern
of fat distribution, with accumulation of fat in the
abdomen regardless of sex (22). Increased intra-
abdominal adipose tissue has been suggested to
contribute both to the development of type 2 diabetes
(7) and to the increased risk of cardiovascular events
Table 5. Adjusted odds ratios and 95% confidence intervals for the presence of hypertension and dyslipidaemia
according to the obesity status
Male
High -TG
Female
High-TG Variables
WHtR ≥ 0.5
WC ≥ 85/80 cm
BMI ≥ 25 kg/m2
WHR ≥ 0.9/0.85
Adjusted for age.
HypertentionLow-HDL-CHypertentionnLow-HDL-C
2.56(1.24,5.29)
2.30(1.14,4.64)
1.96(1.08,3.52)
2.21(1.19,4.09)
2.87(1.43,5.78)
2.16(1.11,4.21)
1.83(1.03,3.24)
1.86(1.03,3.35)
2.59(1.03,6.59)
2.47(1.02,6.01)
0.69(0.34,1.36)
1.38(0.68,2.81)
3.75(1.75,8.05)
2.59(1.29,5.20)
1.86(1.051,3.29)
1.66(0.93,2.96)
3.21(1.52,6.79)
2.95(1.46,5.96)
2.25(1.25,4.04)
1.12(0.63,1.98)
3.62(1.43,9.21)
2.26(1.04,4.88)
2.00(1.05,3.78)
1.31(0.68,2.53)
Table 6. Area under the receiver operating characteristic curve for anthropometric indices as a predictor of at
least one cardiovascular risks factor
Male
AUC
Female
AUC Cut-off95%CI Cut-off95%CI
BMI
WC
WHR
WHtR
24.5
83
0.89
0.51
0.587
0.658
0.616
0.690
0.515, 0.657
0.588, 0.724
0.545, 0.684
0.621, 0.754
24.9
81
0.86
0.50
0.663
0.716
0.687
0.754
0.595, 0.726
0.651, 0.776
0.620, 0.749
0.690, 0.810
AUC = Area under the receiver operating characteristic curve. 95% CI, 95% confidence interval.
Table 7. Difference test of the areas under the ROC curves for each anthropometric index
Male
WHR
Female
WHR WCWHtRWCWHtR
BMI
WC
WHR
P = 0.087
~
P = 0.572
P = 0.277
~
P = 0.016
P = 0.057
P = 0.040
P = 0.100
~
P = 0.603
P = 0.348
~
P = 0.002
P = 0.014
P = 0.035

Page 6

140
WU, CHEN, ZHENG, LIAO AND ZHOU
(25). WC could be a better predictor of visceral
adipose tissue mass (15) and total body fat (27) when
compared to WHR. Moreover, the improvement of
cardiovascular risk factors observed during a weight
loss program is correlated to the reduction of WC, but
not of WHR (9). This suggests that the association
between WC and cardiovascular risk factors was also
stronger than that of WHR. Therefore, various
organizations have proposed WC to assess central
obesity. However, differences in cutoff values
between men and women and among various ethnic
groups (21) may limit its usefulness around the world.
Even within the same population, people with identical
WC but different heights were observed to have
dissimilar risk for hyperglycemia, hypertension and
fatty liver (13). For a given WC a short person will
obviously look more centrally obese than a taller
person. Therefore, height is an important parameter
that should not be ignored before adopting an obesity
index. WHtR which takes into account differences in
height may make up the deficiency and may best
predict the risks of cardiovascular disease.
The correlation between WHtR and cardiovascular
disease risk factors have been studied extensively
worldwide in general population. Various studies (10,
14, 24) have confirmed that WHtR have the potential
to be globally applicable to different ethnic populations
and to children as well as adults and has the highest
predictive value for the detection of cardiovascular
disease risk factors. Though type 2 diabetes mellitus
is tightly correlated with cardiovascular disease, few
studies report the association of WHtR with
cardiovascular disease risk factors in Chinese population
with type 2 diabetes. A previous diagnosis of diabetes
1.0
0.8
0.6
0.4
0.2
0.0
0.00.2 0.40.6
1-Specificity
0.81.0
WHtR
WHR
WC
BMI
Sensitivity
1.0
0.8
0.6
0.4
0.2
0.0
0.00.20.4 0.6
1-Specificity
0.8 1.0
WHtR
WHR
WC
BMI
Sensitivity
Fig. 1. ROC curves for one or more risk factors in men.Fig. 2. ROC curves for one or more risk factors in women.
mellitus could induce a modification of life style or
specific therapeutic interventions which will interfere
with the assessed parameters of adiposity. To avoid
this interference, the subjects enrolled were those newly
diagnosed of type 2 diabetics with no acute or chronic
complications. In this population, there were no
significant differences between sexes for distribution
of WHtR and BMI while the means and quartiles of
WC and WHR for females were significantly different
from that of the males. Additionally, the values of
height for men were significantly higher than that of
women. We speculated that the differences of WC
between sexes and ethnicity may be partly caused by
height.
The present analyses showed that WHtR was
the only anthropometric index which was consistently
among the best in association with cardiovascular
risk factors from four methods of analysis. The
results of the partial correlation analysis show that in
both sexes after adjustment for age, WHtR, WC and
BMI are significantly correlated with SBP, DBP and
TG, but only WHtR and WC are inversely correlated
with HDL-C. It is interesting to note the relatively
high correlation coefficients between WHtR and SBP,
DBP, TG and HDL-C among the four anthropometric
indexes. WC was comparable to WHtR in partial
correlation analysis but was clearly inferior in chi-
square and ROC curve analysis. The optimal cut-off
values in men and women for assessment of at least
one risk factor from our ROC curve analysis were
24.5 and 24.9 kg/m2 for BMI, 83 and 81 cm for WC,
0.89 and 0.86 for WHR, 0.51 and 0.50 for WHtR,
respectively. Obviously these cut-off values were
close to the proposed criteria for obesity.

Page 7

ANTHROPOMETRIC INDICES AND CARDIOVASCULAR RISK FACTORS
141
We chose WHtR level of 0.5 for both males and
females as the indicator of central obesity because
this WHtR cutoff has previously been demonstrated
to correlate with adverse health outcomes for an
Asian population (10-12, 16, 18). It has also been
suggested for use in European (3, 23) and North
American subjects (4). Our results showed that
although most of the subjects (about 75%) studied
had WHtR ≥ 0.5, only about half were obese according
to BMI cut-point (≥ 25kg/m2). This suggests that
visceral obesity was more common than overall obesity
in this group. Chi-square analysis revealed that in
men, all of the abnormal indices were significantly
associated with hypertension and high TG but the
significance level was P < 0.01 only for an increased
WHtR. None of the indices was significantly
associated with low HDL-C except for an increased
WHtR (P < 0.05). In women, both elevated WC and
WHtR were strongly associated with hypertension,
high-TG (P < 0.01) but elevated WHtR was more
strongly associated with low HDL-C (P < 0.01) than
elevated WC (P < 0.05). This result confirmed that
WC was a better anthropometric index than BMI or
WHR in relation to cardiovascular disease risk factors,
but WHtR was shown to be a preferred index over
WC. In addition, the associations of WHtR with
cardiovascular disease risk factors for females (all P
< 0.01) were a little stronger than that for males (for
Low-HDL-C, P < 0.05). The difference suggested
that the risk factors might be slightly different between
men and women and that some factors other than
central obesity might exist in the diabetic males.
Logistic regression analysis also showed WHtR was
most related to hypertension, high-TG and low HDL-
C in both men and women. These results are consistent
with that from the DAI Study (20), although in that
study the samples were composed of Caucasian
populations living in southern Europe. The subjects
enrolled in the present study were from Asian and
with relatively low levels of BMI and WC. These
findings indicate that WHtR may have the potential to
be globally applicable to different ethnic populations
for predicting cardiovascular disease risk factors in
type 2 diabetes. Our results were partially inconsistent
with that of the study by Xiao et al. (26) in which the
use of BMI, WC, WHR and conicity index (CoI) in
combination was more accurate than using them alone
in the prediction of trunk fat mass measured by DXA
in Chinese males. Although height also entered the
calculation of CoI but WHtR was not taken into
consideration in that study. Moreover, DXA can not
distinguish between visceral and subcutaneous adipose
tissue and is less accurate compared with computed
tomography or magnetic resonance imaging for the
measurement of fat mass. However WHtR has been
shown to be a better indicator of intra-abdominal fat
measured by computed tomography than BMI, WC or
WHR (2), which supports our results.
There was no correlation between both of TC
and LDL-C and WHtR which is in line with that TC
and LDL-C are not listed in insulin resistance
syndrome. Although hyperglycemia is a strong risk
factor for cardiovascular disease, our results show
that there is a negative correlation between HbA1c
and WHtR, BMI and WC. This seems to be contrary
to the conclusion that WHtR had good predictive
value for diabetes in general population (10, 13), but
similar to that of a previous analysis from Chan (6)
concerning the association between obesity and
glycemia in diabetic patients. We speculated that it is
probably due mainly to insulin deficiency in non-
obese diabetic patients in the present study because
the mean BMI value in WHtR < 0.5 group (22.6 ±
2.2 kg/m2) was significantly lower than that in WHtR
≥ 0.5 group (25.7 ± 2.9 kg/m2).
In conclusion, the present study shows that
WHtR, which is cheaper and easier to measure and
calculate than BMI, could be a better predictor of
abnormalities associated with the metabolic syndrome
than WC in Chinese adults affected by type 2 diabetes.
The use of WHtR could be an important screening
tool to identify cardiovascular disease risk factors in
Chinese type 2 diabetic patients.
References
1. Alberti, K.G. and Zimmet, P.Z. Definition, diagnosis and classifi-
cation of diabetes mellitus and its complications. Part 1: Diagnosis
and classification of diabetes mellitus provisional report of a WHO
consultation. Diabet. Med. 15: 539 - 553, 1998.
2. Ashwell, M., Lejeune, S. and McPherson, K. Ratio of waist
circumference to height may be better indicator of need for weight
management. Br. Med. J. 312: 377, 1996.
3. Ashwell, M. and Hsieh, S.D. Six reasons why the waist-to-height
ratio is a rapid and effective global indicator for health risks of
obesity and how its use could simplify the international public
health message on obesity. Int. J. Food Sci. Nutr. 56: 303-307,
2005.
4. Bertsias, G., Mammas, I., Linardakis, M. and Kafatos, A. Over-
weight and obesity in relation to cardiovascular disease risk factors
among medical students in Crete, Greece. BMC Public Health 3: 3,
2003.
5. Bosy-Westphal, A., Geisler, C., Onur, S., Korth, O., Selberg, O.,
Schrezenmeir, J. and Muller, M.J. Value of body fat mass vs.
anthropometric obesity indices in the assessment of metabolic risk
factors. Int. J. Obes. (Lond). 30: 475-483, 2006.
6. Chan, W.B., Tong, P.C., Chow, C.C., So, W.Y., Ng, M.C. and Ma,
R.C. The associations of body mass index, C-peptide and metabolic
status in Chinese Type 2 diabetic patients. Diabet. Med. 21: 349-
353, 2004.
7. Haffner, S.M., Stern, M.P., Hazuda, H.P., Pugh, J. and Patterson,
J.K. Do upper-body and centralized adiposity measure different
aspects of regional body fat distribution? Relationship to non-
insulin-dependent diabetes mellitus, lipids, and lipoproteins. Dia-
betes 36: 43-51, 1987.
8. Han, T.S., McNeill, G., Seidell, J.C. and Lean, M.E. Predicting
intraabdominal fatness from anthropometric measures: the influ-

Page 8

142
WU, CHEN, ZHENG, LIAO AND ZHOU
ence of stature. Int. J. Obes. Relat. Metab. Disord. 21: 587-593,
1997.
9. Han, T.S., Richmond, P., Avenell, A. and Lean, M.E. Waist
circumference reduction and cardiovascular benefits during weight
loss in women. Int. J. Obes. Relat. Metab. Disord. 21: 127-134,
1997.
10. Ho, S.Y., Lam, T.H. and Janus, E.D. Hong Kong Cardiovascular
Risk Factor Prevalence Study Steering Committee. Waist to stature
ratio is more strongly associated with cardiovascular risk factors
than other simple anthropometric indices. Ann. Epidemiol. 13: 683-
691, 2003.
11. Hsieh, S.D. and Yoshinaga, H. Abdominal fat distribution and
coronary heart disease risk factors in men – waist/height ratio as a
simple and useful predictor. Int. J. Obes. 19: 585-589, 1995.
12. Hsieh, S.D. and Yoshinaga, H. Waist/height ratio as a simple and
useful predictor of coronary heart disease risk factors in women.
Intern. Med. 34: 1147-1152, 1995.
13. Hsieh, S.D. and Yoshinaga, H. Do people with similar waist
circumference share similar health risks irrespective of height?
Tohoku. J. Exp. Med. 188: 55-60, 1999.
14. Hsieh, S.D., Yoshinaga, H. and Muto, T. Waist-to-height ratio, a
simple and practical index for assessing central fat distribution and
metabolic risk in Japanese men and women. Int. J. Obes. Relat.
Metab. Disord. 27: 610-616, 2003.
15. Jia, W.P., Lu, J.X., Xiang, K.S., Bao, Y.Q., Lu, H.J. and Chen, L.
Prediction of abdominal visceral obesity from body mass index,
waist circumference and waist-hip ratio in Chinese adults: receiver
operating characteristic curves analysis. Biomed. Environ. Sci. 16:
206-211, 2003.
16. Ko, G., Chan, J., Cockram, C. and Woo, J. Prediction of hypertension,
diabetes, dyslipidemia or albuminuria using simple anthropometric
indexes in Hong Kong Chinese. Int. J. Obes. Relat. Metab. Disord.
23: 1136-1142, 1999.
17. Lemieux, S., Prud’homme, D., Bouchard, C., Tremblay, A. and
Despres, J.P. A single threshold value of waist girth identifies
normal weight and overweight subjects with excess visceral adi-
pose tissue. Am. J. Clin. Nutr. 64: 685-693, 1996.
18. Lin, W.Y., Lee, L.T., Chen, C.Y., Lo, H., Hsia, H.H., Liu, I.L., Lin,
R.S., Shau, W.Y. and Huang, K.C. Optimal cut-off values for
obesity: using simple anthropometric indices to predict cardiovas-
cular risk factors in Taiwan. Int. J. Obes. Relat. Metab. Disord. 26:
1232-1238, 2002.
19. Lu, B., Yang, Y., Song, X., Dong, X., Zhang, Z., Zhou, L., Li, Y.,
Zhao, N., Zhu, X. and Hu, R. An evaluation of the International
Diabetes Federation definition of metabolic syndrome in Chinese
patients older than 30 years and diagnosed with type 2 diabetes
mellitus. Metabolism 55: 1088-1096, 2006
20. Mannucci, E., Alegiani, S.S., Monami, M., Sarli, E. and Avogaro,
A. DAI (Diabetes and Informatics) Study Group. Indexes of
abdominal adiposity in patients with type 2 diabetes. J. Endocrinol.
Invest. 27: 535-540, 2004.
21. Molarius, A., Seidell, J.C., Sans, S., Tuomilehto, J. and Kuulasmaa,
K. Varying sensitivity of waist action levels to identify subjects
with overweight or obesity in 19 populations of the WHO MONICA
Project. J. Clin. Epidemiol. 52: 1213-1224, 1999.
22. Resnick, H.E. and Howard, B.V. Diabetes and cardiovascular
disease. Annu. Rev. Med. 53: 245 -267, 2002.
23. Sargeant, L.A., Bennett, F.I., Forrester, T.E., Cooper, R.S. and
Wilks, R.J. Predicting incident diabetes in Jamaica: the role of
anthropometry. Obes. Res. 10: 792 - 798, 2002.
24. Sayeed, M.A., Mahtab, H., Latif, Z.A., Khanam, P.A., Ahsan, K.A.
and Banu, A. Waist-to-height ratio is a better obesity index than
body mass index and waist-to-hip ratio for predicting diabetes,
hypertension and lipidemia. Bangladesh. Med. Res. Counc. Bull.
29: 1-10, 2003.
25. Thomas, G.N., Ho, S.Y., Lam, K.S., Janus, E.D., Hedley, A.J. and
Lam, T.H. Hong Kong Cardiovascular Risk Factor Prevalence
Study Steering Committee. Impact of Obesity and Body Fat Distri-
bution on Cardiovascular Risk Factors in Hong Kong Chinese.
Obes. Res. 12: 1805-1813, 2004.
26. Xiao, S.M., Lei, S.F., Chen, X.D., Liu, M.Y., Jian, W.X., Xu, H.,
Tan, L.J., Deng, F.Y., Yang, Y.J., Wang, Y.B., Sun, X., Jiang, C.,
Guo, Y.F., Guo, J.J., Li, Y.N., Jiang, H., Zhu, X.Z. and Deng, H.W.
Correlation and prediction of trunk fat mass with four anthropomet-
ric indices in Chinese males. Br. J. Nutr. 96: 949-955, 2006.
27. Yang, F., Lv, J.H., Lei, S.F., Chen, X.D., Liu, M.Y., Jian, W.X., Xu,
H., Tan, L.J., Deng, F.Y., Yang, Y.J., Wang, Y.B., Sun, X., Xiao,
S.M., Jiang, C., Guo, Y.F., Guo, J.J., Li, Y.N., Zhu, X.Z., Papasian,
C.J. and Deng, H.W. Receiver-operating characteristic analyses of
body mass index, waist circumference and waist-to-hip ratio for
obesity: Screening in young adults in central south of China. Clin.
Nutr. 25: 1030-1039, 2006.
28. Zhou, B.F. Co-operative Meta Analysis Group of the Working
Group on Obesity in China. 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. 15: 83-96, 2002.