Maturitas 107 (2018) 33–38

Contents lists available at ScienceDirect

Maturitas
journal homepage: www.elsevier.com/locate/maturitas

Excess fat in the abdomen but not general obesity is associated with poorer MARK
metabolic and cardiovascular health in premenopausal and postmenopausal
Asian women
⁎
Victor Hng Hang Goh , William George Hart
Curtin Medical School,Building 410, Curtin University Kent Street, Bentley, WA 6102 Australia

A R T I C L E I N F O A B S T R A C T

Keywords: Objectives: To examine the associations of various metabolites and hormones and hormone replacement therapy
Abdominal obesity (HRT) with obesity.
General obesity Methods: This is a cross-sectional study of 1326 Singaporean women. A DXA-derived percent body fat (PBF) of
Metabolic syndrome ≥35% and percent abdominal fat (PAbdF) of > 21.8% were used, respectively, to define women with general
Hormones
(GOb) and abdominal (AbdOb) obesity.
Hormone replacement therapy
Results: Higher levels of insulin and glucose, lower levels of HDL, higher levels of TC/HDL and HOMA values,
Postmenopausal women
and different levels of some hormones were noted only in the women with abdominal, and not general obesity.
The incidence of general and abdominal obesity was higher in postmenopausal women with or without HRT,
except that those who were on conjugated estradiol-only HRT had no increase in the incidence of general obesity
compared with premenopausal women.
Conclusions: Abdominal obesity is associated with insulin resistance and with higher risks of metabolic syn-
drome and cardiovascular diseases, whereas general obesity is not. Abdominal obesity may predispose to a
higher risk of diabetes. The onset of the menopause tends to increase the incidence of general and abdominal
obesity, except that postmenopausal women on conjugated estradiol HRT appear to be relatively protected from
general obesity.

1. Introduction
An obesity pandemic is currently affecting most developed and
developing countries [1,2]. Obesity is a common predisposing factor for
elevated risk of several serious health conditions including insulin re-
sistance, type 2 diabetes mellitus, hypertension and other cardiovas-
cular disease, fatty liver disease and some types of cancer [3,4]. Me-
tabolic syndrome (MetSyn) has been associated with the endocrine,
metabolic and immunological functions of the adipose tissue [5].
Obesity is strongly associated with disorders of glucose, lipid metabo-
lism and insulin resistance [6,7]. Furthermore, high insulin levels lead
to an increase in bioavailability of IGF1 [8].
The high incidence of cardiometabolic comorbidities with obesity
has heightened the obesity crisis. A better understanding of the asso-
ciations among these cardiometabolic risk factors and obesity will lead
to more appropriate therapeutic treatment of obesity [9,10].
However, there is significant variability in studies of disease asso-
ciation with obesity. The differences may, in part, be due to the criteria
used to classify the different forms of obesity. Body Mass Index (BMI) of
≥30 kg/m2 is commonly used to define general obesity (GOb) [11],
while abdominal obesity (AbdOb) is commonly defined by either waist
circumference (W), waist-hip ratio (W/H) or waist-height ratio alone or
in combination [12]. These indices of adiposity have high degrees of
misclassification resulting in confusing inferences derived from their
use in obesity studies [13]. There is a need to use more appropriate
indices for the classification of obesity and to evaluate their association
with various hormonal and metabolic factors. A better understanding of
the etiologic factors of obesity could also lead to more appropriate
modalities for managing obesity.
In a previous study, DXA-derived percent body fat (PBF) and per-
cent abdominal fat (PAbdF) were used to define general (GOb) and
abdominal obesity (AbdOb) in men [14]. Therefore, the present study
explored whether the various forms of obesity exist in women and
evaluated their profiles of association with various metabolites and
hormones. In addition, the study explored the relationships between the
onset of menopause and different types of hormone replacement
therapy (HRT) and the incidence of obesity.

⁎
Corresponding author.
E-mail address: victor.goh@curtin.edu.au (V.H.H. Goh).

http://dx.doi.org/10.1016/j.maturitas.2017.10.002
Received 3 April 2017; Received in revised form 4 September 2017; Accepted 2 October 2017
0378-5122/ © 2017 Elsevier B.V. All rights reserved.
V.H.H. Goh, W.G. Hart
2. Materials and methods
2.1. Subjects
This study was approved by the Institutional Review Board of the
National University Hospital of Singapore and each volunteer gave her
written informed consent. One thousand three hundred and twenty-six
Singaporean Chinese women, aged between 29y and 71y, living in the
community were recruited through advertisements in the media and
through word of mouth. As the primary objective of the overall study
was to evaluate the determinants of the natural aging process, only
women without a history of medical illnesses such as cancer, hy-
pertension, thyroid dysfunction, diabetes, osteoporotic fracture, cardi-
ovascular events, major sleep disorders, or major joint surgery were
included in the study. Subjects were not paid for their participation.
The cohort of women represented the diverse spectrum of Chinese in
Singapore, ranging from those with low to high levels of education,
working and non-working women, and those in various types of voca-
tion. Their profiles were typical of women in Singapore, which is a
highly urbanized city-state with no rural population. The methodology
used was previously reported [15].
3. Methods
3.1. General questionnaire
Each subject answered a self-administered and investigator-guided
questionnaire. Questions asked included their medical, social, sex, ex-
ercise regime, and family history.
3.2. Biochemical and hormone measurements
An overnight 12 h fasting blood sample was collected in the
morning between 9.00am and 11.00am for premenopausal women
between Day 3–5 of their menstrual cycle and on any day for post-
menopausal women. The sera were stored at −80 °C until analysis.
Serum levels of total cholesterol (TC) and triglycerides (TG), high
density lipoprotein-cholesterol (HDL), low density lipoprotein choles-
terol (LDL) and fasting glucose level (GLU) were measured by methods
reported earlier [15]. Serum testosterone (T), dehydroepiandrosterone
sulphate (DHEAS), and sex hormone binding globulin (SHBG) were
measured by established radioimmunoassay methods reported earlier
[15]. Serum concentrations of insulin-like growth factor-1 (IGF1) and
insulin like growth factor binding protein-3 (BP3) were measured using
immunoradiometric assay kits (Diagnostic Systems Laboratories, Inc.,
Webster, TX) as reported earlier [16,17]. Serum concentrations of in-
sulin (INS) were measured in-house using the Axsym platform from
Abbott Laboratories (Irving, TX). Bioavailable testosterone (BioT) was
calculated using the computer formula of Vermeulen, which is available
on the ISSAM website (www.issam.ch).
3.3. Whole body DXA scan
Every woman had a whole body scan using the dual-energy x-ray
absorptiometry (DXA) (Hologic, Bedford, MA, USA). The percent total
body fat (PBF), total fat mass (TFM) and percent abdominal fat (PAbdF)
were calculated by the DXA machine based on the Siri formula.
3.4. Blood pressures
Brachial systolic (Sys) and diastolic (Dia) blood pressures were
measured by trained clinical researchers using a standardized manual
sphygmomanometer after subjects had five minutes of rest.
34
Maturitas 107 (2018) 33–38
3.5. Intensity of exercise (METmin)
The intensity of exercise, expressed as metabolic equivalent of task-
minutes (METmin), was shown in our earlier studies to correlate with
body composition, and with various metabolites and hormones. In
order to adjust for the effect of exercise, the METmin was used as a
covariate in all the analyses. Computation of the total exercise score
(METmin) was reported earlier [18]. In order to normalize the different
types of exercise/sport into a single common score, exercise intensity
was calculated using the Metabolic Equivalent of Task (MET) of each
exercise/sport type. A score was then calculated to denote the intensity
of exercise/sport per week in MET minutes (METmin) by taking into
account the duration of each exercise episode and the frequency of the
exercise per week in accordance with the exercise guidelines [19]. For
example, a participant reported that she walked for 30 min 5 times a
week, played tennis for 60 min twice a week and did line dancing for
60 min once a week. Walking is assigned a MET of 3; tennis, a MET of 7;
and line dancing, a MET of 5. Her total physical/sports activities in-
tensity per week was therefore 1590 METmin [(3 × 30 × 5) +
(7 × 60 × 2) + (5 × 60 × 1)]. These data were gathered using a self-
administered and investigator-guided questionnaire.
3.6. Definitions of general obesity (GOb), abdominal obesity (AbdOb)and
general and abdominal obesity (G + AbdOb)
As recommended by the American Council on Exercise (ACE), a
woman is considered to have obesity when her percent total body fat
(PBF) computed from the DXA-whole body scan is ≥32% [20]. Using
the frequency plot of all PBF of the 1326 women in the present study,
the cut off at the 95 percentile was 35% PBF. Hence, women were ca-
tegorised as having general obesity (GOb) when their PBF was ≥35%.
The regional distribution of body fat is not homogeneous in all
women. It is possible that there are women with significantly higher
accumulation of abdominal fat who may be considered to have ab-
dominal obesity (AbdOb). For the definition of abdominal obesity
(AbdOb), a frequency distribution of the DXA-derived PAbdF of the
1326 women was constructed. As with the definition in men reported
earlier (8), the PAbdF at the 95 percentile (21.8%) was used as the cut-
off value; any woman with a PAbdF above this cut-off value was con-
sidered to have abdominal obesity (AbdOb). Therefore, there are two
groups of women with abdominal obesity: those women with PBF
of < 35% and a PAbdF of > 21.8% were considered as having ab-
dominal obesity (AbdOb), while women with PBF of > 35% and a
PAbdF of > 21.8% were considered to have general as well as ab-
dominal obesity (G + AbdOb). However, it must be noted that the
DXA-derived PAbdF does not distinguish between abdominal sub-
cutaneous and abdominal visceral fat. It is the sum total of fat in the
abdominal area.
3.7. Indices of insulin resistance
3.7.1. Triglyceride and high density lipoprotein cholesterol ratio (TG/HDL)
as a marker for insulin resistance
Triglyceride/HDL ratio has been considered as a cardiovascular risk
factor [21]. Triglyceride/HDL ratio has also been used to identify in-
sulin-resistant individuals [22]. As suggested by McLaughlin et al. [22],
a TG/HDL of ≥1.80 was used as an indication that an individual was
insulin resistant.
Homeostasis model assessment (HOMA) was also used as a measure of
insulin resistance. As suggested by Matthews et al., HOMA is computed by
multiplying fasting insulin by fasting glucose levels and dividing by 22.5
[23]. A HOMA value of > 2.8 computed from a single fasting blood sample
correlated well with other measures of insulin resistance. The data in the
present study showed that TG/HDL and HOMA were positively and sig-
nificantly correlated. A HOMA value of > 2.8 was considered as indicative
of an individual with insulin resistance.
V.H.H. Goh, W.G. Hart Maturitas 107 (2018) 33–38

Table 1 Table 3
Distribution of women with general obesity (PBF ≥ 35%) and with abdominal obesity Prevalence of obesity in the different menopausal groups.
(PAbdF > 21.8%).
MenoGp1 MenoGp2 MenoGp3 MenoGp4 MenoGp5
PAbdF ≤21.8%, n = 1257 PAbdF > 21.8%, n = 69 N = 725 N = 460 N = 47 N = 10 N = 84

PBF < 35%, n = 1156 1098 58 NOb 653 345 29 8 63

PBF ≥35%, n = 170 159 11 GOb 53 (7.3%) 80 (17.4%) 13 (27.7%) 0 (0%) 13 (15.5%)
1v2,3,5
PBF – percent body fat. (< 0.040)
PAbdF – percent abdominal fat. AbdOb 15 (2.1%) 30 (6.5%) 4 (9.4%) 2 (20.0%) 7 (8.3%)
1v2,3,4,5
(< 0.030)
3.8. Statistical analysis G + AbdOb) 4 (0.6%) 5 (1.09%) 1(2.1%) 0 (0%) 1(1.2%)

Statistical analyses were performed using SPSS for windows version NOb − non-obesity.
21.0 (Armond, NY). Basic descriptive statistics and multivariate linear GOb − general obesity.
analyses coupled with the Bonferroni correction for multiple compar- AbdOb − abdominal obesity.
G + AbdOb − general and abdominal obesity.
isons were used on continuous measurements. To adjust for their ef-
MenoGp1–premenopausal women.
fects, age and METmin were used as co-variates in all multivariate MenoGp2–postmenopausal women not on any HRT.
analyses. Multivariate linear analyses were carried out on the four age MenoGp3–postmenopausal women on synthetic estrogen.
groups and the three obesity groups: women with no obesity (NOb), MenoGp4–postmenopausal women on conjugated estradiol.
women with general obesity (GOb), and those with abdominal obesity MenoGp5–postmenopausal women on combined estrogen/progestogen HRT.
(AbdOb). Fisher’s exact test was used to analyze the non-continuous
variables. on conjugated estrogen (MenoGp4) was not significantly different from
premenopausal women (MenoGp1) (Table 3). On the other hand, the
4. Results incidence of abdominal obesity (AbdOb) was significantly higher in all
postmenopausal women with or without HRT (Table 3). The incidence
4.1. Incidence of obesity of combined general and abdominal obesity was not significantly dif-
ferent in all groups of women (Table 3).
Table 1 shows that out of 1326 apparently healthy women 1098
were without obesity, 159 (12.0%) had general obesity (GOb) with PBF 4.4. Association of GOb and AbdOb with metabolic and hormones
≥ 35% but PAbdF < 21.8% and 58 (4.4%) had abdominal obesity
(AbdOb) with PAbdF of > 21.8% but PBF of < 35%. Eleven (0.83%) When adjusted for age, exercise intensity (METmin), total fat mass,
women had combined general and abdominal obesity with PBF ≥35% and percent body fat (PBF), Table 4A shows the associations mainly of
and PAbdF > 21.8%. the percent abdominal fat with the metabolic, hormonal and lipid and
lipoprotein factors in the four groups. On the other hand, when adjusted
4.2. Incidence of obesity with age for age, exercise intensity, total fat mass and percent abdominal fat
(PAbdF), Table 4B shows the associations mainly of the percent body
The incidence of general obesity (GOb) was significantly higher in fat (PBF) with the metabolic, hormonal and lipid and lipoprotein factors
the two older age groups (Table 2). While the incidence of abdominal in the four groups. Abdominal obesity was associated with significantly
obesity was significantly higher in the 51–60 age groups as compared to higher levels of INS and GLU, TC/HDL, BioT, BP3, DHEAS and HOMA
the younger age groups below 51y (Table 2). The incidence of com- value and lower levels of HDL and SHBG when compared to non-obese
bined general and abdominal obesity was independent of age (Table 2). women (Table 4A). Likewise women with general obesity (GOb), had
significantly lower INS, TC/HDL, BP3, BioT, and HOMA, and higher
4.3. Incidence of obesity in different obesity groups SHBG and HDL when compared with women with abdominal obesity
(Table 4A). Interestingly, women with combined general and abdom-
Postmenopausal women without hormone replacement therapy and inal obesity (G + AbdOb), as with those with abdominal obesity only,
those on synthetic estrogen and combined estrogen/progestogen had only significantly higher INS and HOMA levels than non-obese
therapy (MenoGp2, MenoGp3, MenoGp5) had a higher incidence of women (Table 4A). The levels of SHBG in women with general obesity
general obesity (GOb), while the incidence in postmenopausal women were significantly higher than those in non-obese women (Table 4A). In
addition, women with general obesity had significantly lower TC/HDL
Table 2 ratio and higher HDL than non-obese women (Table 4A). When the
Prevalence of types of obesity in four age groups.
analyses were adjusted for the percent abdominal fat, no significant
AgeGp1 AgeGp2 AgeGp3 AgeGp4 differences were noted for all parameters studied except that women in
(≤40 y) (41–50 y) (51–60 y) ( > 60 y) the abdominal obesity group had higher total cholesterol level when
N = 181 N = 533 N = 479 N = 133 compared to non-obese women (Table 4B).
NOb 170 463 372 93
GOb 10 (5.5%) 49 (9.2%) 68 (14.2%) 32 (24.1%) 5. Discussion
1v3,4
(< 0.005)
AbdOb 1 (0.55%) 16 (3.0%) 36 (7.5%) 5 (3.8%) Results of the present study show that, in women, the adverse ef-
1v3 2v3 (0.00311) fects of obesity depend on where in the body the accumulation of fat is.
( < 0.001) Women who were classified as having abdominal obesity in the present
G + AbdOb) 0 (0%) 5 (0.94%) 3 (0.63%) 3 2.25%) study were lean, therefore, may represent the group with sarcopenic
obesity.The present study demonstrates that only abdominal obesity
NOb – non-obesity.
GOb – general obesity. and not general obesity, is associated with impairment of carbohydrate
AbdOb – abdominal obesity. metabolism as shown by higher glucose and insulin levels compared to
G + AbdOb – general and abdominal obesity. women without obesity. Together with the significantly higher HOMA

35
V.H.H. Goh, W.G. Hart Maturitas 107 (2018) 33–38

Table 4
Metabolic and hormones in non-obesity (NOb), general (GOb), abdominal obesity (AbdOb) and general + abdominal obesity (G + AbdOb) groups.

A : multivariate linear analyses was carried out with age, and METmin, percent body fat (PBF) and total body fat as covariates.

Gp1 = NOb (n = 1098) Gp2 = Ob (n = 159) Gp3 = AbdOb (n = 58) Gp4 = G + AbdOb (n = 11)

Age 48.9 ± 0.24a,b 53.0 ± 0.63 53.6 ± 2.4 53.1 ± 1.0

METmin 450 ± 17 405 ± 46 363 ± 76 403 ± 174
PBF 27.4 ± 0.12b,c 36.6 ± 0.32a 28.7 ± 0.49 a
36.5 ± 1.20
PAbdF 18.4 ± 0.05a,c 18.6 ± 0.13a,c 22.8 ± 0.22 22.7 ± 0.50
T 0.68 ± 0.03 0.70 ± 0.10 0.57 ± 0.13 1.05 ± 0.32
SHBG 54.1 ± 0.85a 61.0 ± 2.70a 41.7 ± 3.50 43.8 ± 8.50
BioT 17.0 ± 0.29a 16.9 ± 0.90a 21.6 ± 1.18 23.0 ± 2.90
IGF1 183 ± 2.5 166 ± 7.7 197 ± 10.2 204 ± 24.6
BP3 3936 ± 35a 3896 ± 111a 4491 ± 145 4195 ± 352
DHEAS 1524 ± 29a 1395 ± 90a 1907 ± 118 2123 ± 286
INS 6.49 ± 0.11a,c 6.44 ± 0.34a,c 8.21 ± 0.45 9.57 ± 1.06

TC 5.65 ± 0.03a 5.57 ± 0.09a 5.98 ± 0.12 5.36 ± 0.27

TG 1.20 ± 0.07 0.900.23 1.59 ± 0.31 0.98 ± 0.72
LDL 3.51 ± 0.03 3.38 ± 0.08 3.76 ± 0.11 3.19 ± 0.24
HDL 1.65 ± 0.01b 1.77 ± 0.04a 1.52 ± 0.05 1.71 ± 0.11
TC/HDL 3.59 ± 0.03a,b 3.27 ± 0.09a 4.06 ± 0.12 3.20 ± 0.28
GLU 4.75 ± 0.02a 4.74 ± 0.05 4.92 ± 0.07 4.81 ± 0.16

Dias 77 ± 0.31 78 ± 0.96 77 ± 1.3 78 ± 3.0

Sys 121 ± 0.51 121 ± 1.60 121 ± 2.1 124 ± 5.0
TG/HDL 0.82 ± 0.06 0.56 ± 0.18 1.13 ± 0.24 0.61 ± 0.55
HOMA 1.40 ± 0.03a,c 1.42.09a,c 1.84 ± 0.12 2.22 ± 0.27

B : multivariate linear analyses was carried out with age, METmin, total body fat and abdominal fat as covariates to evaluate what differences in association with the various parameters
have when the contributions of percent abdominal fat have been adjusted.

Gp1 = NOb (n = 1098) Gp2 = Ob (n = 159) Gp3 = AbdOb (n = 58) Gp4 = G + AbdOb (n = 11)

Age
METmin
PBF
PAbdF
T 0.68 ± 0.03 0.71 ± 0.10 0.53 ± 0.15 1.03 ± 0.32
SHBG 53.8 ± 0.84 59.2 ± 2.7 50.1 ± 4.0 49.7 ± 8.5
BioT 17.1 ± 0.29 17.4 ± 0.91 19.1 ± 1.34 21.2 ± 2.90
IGF1 183 ± 2.4 169 ± 7.8 182 ± 11.6 194 ± 24.8
BP3 3941 ± 354 4036 ± 111 4085 ± 164 3936 ± 352
DHEAS 1526 ± 28 1430 ± 91 1798 ± 134 2052 ± 288
INS 6.57 ± 0.11 6.51 ± 0.36 7.51 ± 0.53 7.42 ± 1.13

TC 5.64 ± 0.03a 5.62 ± 0.09 6.01 ± 0.13 5.41 ± 0.28

TG 1.20 ± 0.07 0.99 ± 0.24 1.44 ± 0.35 0.89 ± 0.73
LDL 3.50 ± 0.03 3.44 ± 0.08 3.70 ± 0.12 3.18 ± 0.25
HDL 1.65 ± 0.01 1.74 ± 0.04 1.68 ± 0.05 1.82 ± 0.11
TC/HDL 3.60 ± 0.03 3.44 ± 0.08 3.70 ± 0.12 3.18 ± 0.25
GLU 4.75 ± 0.02 4.73 ± 0.05 4.92 ± 0.08 4.81 ± 0.16

Dias 77 ± 0.31 79 ± 0.97 75 ± 1.4 77 ± 3.0

Sys 122 ± 0.51 121 ± 1.60 119 ± 2.4 123 ± 5.1
TG/HDL 0.82 ± 0.06 0.61 ± 0.18 0.97 ± 0.27 0.52 ± 0.55
HOMA 1.41 ± 0.03 1.40 ± 0.09 1.69 ± 0.13 2.08 ± 0.27

Gp1–Non-obesity group.
Gp2–General obesity group.
Gp3–Abdominal obesity group.
Gp4–General and abdominal obesity group.
a
significantly different from Gp3 (p < 0.05).
b
significantly different from Gp2 (p < 0.05).
c
significantly different from Gp4 (p < 0.05).

values, the results suggest that abdominal obesity is associated with a
higher level of insulin resistance than in non-obese women. As sug-
gested in earlier studies, abdominal obesity was associated with lower
HDL and higher TC/HDL suggesting a higher risk of cardiometabolic
disorders [5–10]. On the other hand, high accumulation of fat generally
in the body except in the abdomen, appears not to have any adverse
effect, and in fact, may be have beneficial metabolic effect as they had
higher HDL and concurrently lower TC/HDL than women without
general obesity [24]. The combined presence of general together with
abdominal obesity appears not to have increased the risk factors for
metabolic and cardiovascular disorder associated with abdominal
obesity alone. Only INS and HOMA were raised when compared to non-
obese women.
The presence of general obesity was not associated with any change
in hormone levels. Only abdominal obesity was associated with lower
levels of SHBG and higher levels of bioavailable testosterone, insulin
growth factor-1, DHEAS and insulin growth factor binding protein-3.
The decrease in SHBG leading to higher levels of bioavailable testos-
terone may lead to higher accumulation of abdominal fat as was it has
previously suggested [25,26]. However, the mechanism of how de-
crease SHBG and increase in bioavailable testosterone affect body fat
accumulation and distribution is unclear. Higher levels of insulin

36
V.H.H. Goh, W.G. Hart
growth factor binding protein-3 in abdominal obesity is contrary to
earlier observation of a decrease in insulin growth factor binding pro-
tein-1 and 2 [8], leading to higher bioavailable insulin growth factor-1
[27,28]. It is unclear whether an increase in insulin growth factor-1
concurrent with an increase in insulin growth factor binding protein-3
has resulted in an increase or a decrease in bioavailable IGF-1 noted in
the present study. Further investigation is needed. Since this is a cross-
sectional study, the association of higher DHEAS levels in women with
abdominal obesity as compared to non-obese women is not definitive
and the significance of this increase in DHEAS in abdominal obesity is
unclear. These observations suggest that hormonal factors may have a
role, whether as the cause or effect, in the etiology of abdominal obe-
sity.
Age is a factor in the incidence of obesity. The incidence of both
general and abdominal obesity increases with age. Older adults with
abdominal obesity have been reported to be at increased risk for cardio
metabolic disorders compared with their non-obese counterparts, in-
dependently of the body mass index category [9]. The observations of
the present study support the notion that aging is associated by in-
creased adiposity and possibly a redistribution in the pattern of adip-
osity. With aging, there is a selective accumulation of visceral fat and
with its associated increased risk of cardiometabolic disorders [29,30].
The role of female hormones in the pathophysiology of obesity in
women is unclear. The present study suggests that conjugated estradiol
HRT is protective of postmenopausal women from increased risk of
general obesity. Other HRT including synthetic estrogen and combined
estrogen/progestogen HRT was not protective of postmenopausal
women from increased incidence of general obesity. Postmenopausal
women whether or not on any form of HRT, had a higher incidence of
abdominal obesity when compared to premenopausal women. The ef-
fect of natural estradiol/progesterone combination HRT on the in-
cidence of obesity needs to be explored.
The incidence of general and abdominal obesity in Singaporean
women were respectively, 12.0%, and 5.2%. The corresponding in-
cidence for Singaporean men were 6.2%, and 4.9% [20]. It must be
noted that the cohort of Singapore women represents a selected group,
those who were healthy with no major illness including diagnosed
morbid obesity, diabetes and cardiovascular diseases. Hence, the in-
cidence does not represent that of the total population in Singapore.
In the present study, the two forms of obesity were defined by the
use of DXA-derived ≥35% of PBF for peripheral obesity and a PAbdF
of > 21.8% for abdominal obesity (AbdOb). Clearly, the present study
shows that the selectively fat accumulation in the abdomen is asso-
ciated with higher risk of metabolic and cardiovascular disorders. On
the other hand, high accumulation of fat generally in the body except in
the abdomen, appears not to be adverse, and in fact, may be beneficial
to some aspects of metabolic health.
The major disadvantage of the present study is that as a cross-sec-
tional study, no definitive causal relationship could be attributed, but
only suggested. However, the advantages of the present study include
the use of DXA-derived total and regional distribution of body com-
position, avoiding the use of anthropometric indices, with their in-
herent risks of misclassification of obesity, to categorize general and
abdominal obesity. In addition, the use of age, and exercise intensity
(MET-min) as co-variates in the multivariate analyses ensured that the
potential confounding influences of these factors could be adjusted for.
In conclusion, the study showed that only excess fat accumulated in
the abdomen was associated with insulin resistance and with more
adverse metabolic syndrome, and cardiovascular disease risk factors. In
addition, only abdominal obesity was associated with different levels of
hormones when compared to non-obese women. Women with general
obesity appears to be metabolically healthier than women without
general obesity. Postmenopausal women, whether they were on or not
on any hormone replacement therapy, except for women on conjugated
estradiol HRT, had increased incidence of general and abdominal
obesity when compared to non-obese women. Postmenopausal women
37
Maturitas 107 (2018) 33–38
who were on conjugated estradiol therapy appear to have lower in-
cidence of general obesity, but not abdominal obesity.
Contributors
VHHG designed the study, and collected the data.
WGH was involved in data interpretation, and drafting and critical
revision of the article for submission.
Conflict of interest
The authors declare that they have no conflict of interest.
Funding
This study was supported, in part, by funds from the Academic
Research Fund of the National University of Singapore, Singapore.
Ethical approval
This study was approved by the Institutional Review Board of the
National University Hospital of Singapore and each volunteer gave her
written informed consent.
Provenance and peer review
This article has undergone peer review.
Acknowledgments
We would like to acknowledge the technical assistance from staff of
the Endocrine Research and Service Laboratory of the Department of
Obstetrics and Gynaecology, National University of Singapore.
References
[1] K.M. Flegal, M.D. Carroll, C.L. Ogden, L.R. Curtin, Prevalence and trends in obesity
among US adults, 1999–2008, J. Am. Med. Assoc. 303 (2010) 235–241.
[2] A.M. Prentice, The emerging epidemic of obesity in developing countries, Int. J.
Epidemiol. 35 (2006) 93–99.
[3] G. Whitlock, S. Lewington, P. Sherliker, et al., Body mass index and cause-specific
mortality in 900,000 adults: collaborative analysis of 57 prospective studies.
Prospective Studies Collaboration, Lancet 373 (2009) 1083–1096.
[4] G.M. Grundy, Obesity metabolic syndrome, and cardiovascular disease, J. Clin.
Endocrinol. Metab. 89 (2004) 2595–2600.
[5] S. Galic, J.S. Oakhill, G.R. Steinberg, Adipose tissue as an endocrine organ, Mol. Cell
Endocrinol. 316 (2010) 129–139.
[6] S. Fujioka, Y. Matsuzawa, K. Tokunaga, Contribution of intra-abdominal fat accu-
mulation to the impairment of glucose and lipid metabolism in human obesity,
Metabolism 36 (1987) 54–59.
[7] T.P. Nedungadi, D.J. Clegg, Sexual dimorphism in body fat distribution and risk for
cardiovascular diseases, J. Cardiovasc. Transl. Res. 2 (2009) 321–327.
[8] J.I. Jones, D.R. Clemmons, Insulin-like growth factors and their binding proteins:
biological actions, Endocr. Rev. 16 (1995) 3–34.
[9] S.S. Lim, T. Vos, A.D. Flaxman, et al., A comparative risk assessment of burden of
disease and injury attributable to 67 risk factors and risk factor clusters in 21 re-
gions, 1990–2010: a systematic analysis for the Global Burden of Disease Study
2010, Lancet 380 (2012) 2224–2260.
[10] E. Grieve, E. Fenwick, H.C. Yang, M. Lean, The disproportionate economic burden
associated with severe and complicated obesity: a systematic review, Obes. Rev. 14
(2013) 883–894.
[11] Obesity: preventing and, managing the global epidemic. Report of a WHO con-
sultation, World Health Organ. Tech. Rep. Ser. 894 (i–xii) (2000) 1–253.
[12] R.M. Parik, S.R. Joshi, K. Pandia, Index of obesity is better than waist circumference
in defining metabolic syndrome, Metab. Syndr. Relat. Disord. 7 (2009) 525–527.
[13] V.H.H. Goh, C.F. Tain, T.Y.Y. Tong, P.P. Mok, M.T. Wong, Are BMI and other an-
thropometric measures appropriate as indices for obesity? A study in an Asian
population, J. Lipid. Res. 45 (2004) 1892–1898.
[14] V.H.H. Goh, W.G. Hart, Association of general and abdominal obesity with age,
endocrine and metabolic factors in Asian men, Aging Male 19 (2016) 27–33.
[15] V.H.H. Goh, T.Y.Y. Tong, P.P. Mok, B. Said, Interactions among age adiposity,
bodyweight, lifestyle factors and sex steroid hormones in healthy Singaporean
Chinese men, Asian J. Androl. 9 (2007) 611–621.
[16] N.M. Probst-Hirsch, H. Wang, V.H.H. Goh, A. Seow, H.P. Lee, M.C. Yu,
Determinants of circulating insulin-like growth factor I and insulin-like growth
V.H.H. Goh, W.G. Hart
factor binding protein 3 concentrations in a cohort of Singapore men and women,
Cancer Epidemiol. Biomarkers Prev. 12 (2003) 739–746.
[17] V.H.H. Goh, S.C. Mu, F.K. Gao, K.S. Lim, Changes in body composition and endo-
crine and metabolic functions in healthy elderly Chinese men following growth
hormone therapy, Aging Male 1 (1998) 264–269.
[18] V.H.H. Goh, W.G. Hart, Association of physical exercise as a lifestyle habit with lean
and fat body mass and handgrip strength and age in Asian men, Aging Male 17
(2014) 131–135.
[19] US. Department of Health and Human Services, Physical Activity Guidelines for
American (PGA), (Accessed 5 April 2015) http:www.health.gov/PAguidelines/
guidelines/appendix.aspx.
[20] ACE, Fit Life. What Are the Guidelines for Percentage of Body Fat Loss? American
Council on Exercise (ACE), 2009.
[21] J. Jeppesen, H.O. Hein, P. Suadicani, F. Gyntelberg, Low triglycerides-high-density
lipoprotein cholesterol and risk of ischemic heart disease, Arch. Intern. Med. 161
(2001) 361–366.
[22] T. McLaughlin, F. Abbasi, K. Cheal, J. Chu, C. Lamendola, G. Reaven, Use of me-
tabolic markers to identify overweight individuals who are insulin resistant, Ann.
Intern. Med. 139 (2003) 802–809.
[23] D.R. Matthews, J.P. Hosker, A.S. Rudenski, B.A. Naylor, D.F. Treacher, R.C. Turner,
Homeostasis model assessment: insulin resistance and beta-cell function from
38
Maturitas 107 (2018) 33–38
fasting plasma glucose and insulin concentrations in man, Diabetologia 28 (1985)
412–419.
[24] A.D. Karelis, Metabolically healthy but obese individuals, Lancet 372 (2008)
1281–1283.
[25] X. Xu, G. De Pergola, P. Bjorntorp, The effects of androgens on the regulation of
lipolysis in adipose precursor cells, Endocrinol 126 (1990) 1229–1234.
[26] B.A. Evans, I.A. Hughes, Augmentation of androgen-receptor binding in vitro: stu-
dies in normal and patients with androgen insensitivity, Clin. Endocrinol. 23 (1985)
567–577.
[27] J. Frystyk, E. Vestbo, C. Skjaerbaek, et al., Free insulin like growth factors in human
obesity, Metabolism 44 (1995) 37–44.
[28] S.Y. Nam, E.J. Lee, K.R. Kim, et al., Effect of obesity on total and free insulin-like
growth factor (IGF)-1, and their relationship to IGF-binding protein (BP)-1, IGFBP-
2, IGFBP-3, insulin, and growth hormone, Int. J. Obes. (Lond.) 21 (1997) (1997)
355.
[29] J.L. Kuk, T.J. Saunders, L.E. Davidson, R. Ross, Age-related changes in total and
regional fat distribution, Ageing Res. Rev. 8 (2009) 339–348.
[30] I. Janssen, P.T. Katzmarzyk, R. Ross, Body mass index, waist circumference, and
health risk: evidence in support of current National Institutes of Health guidelines,
Arch. Intern. Med. (2002) 2074–2079.