Relationship between Generalized and Upper Body
Obesity to Insulin Resistance in Asian Indian Men*
MANISHA CHANDALIA, NICOLA ABATE, ABHIMANYU GARG,
JAMES STRAY-GUNDERSEN, AND SCOTT M. GRUNDY
The Center for Human Nutrition (N.A., S.M.G., A.G.), and the Department of Internal Medicine
(M.C., N.A., S.M.G., A.G.), University of Texas Southwestern Medical Center, and the Department of
Veteran Affairs Medical Center (M.C., S.M.G., A.G.), Dallas, Texas 75235
It has been proposed that excessive insulin resistance in Asian
Indians living in urban areas or migrated to western countries is
responsible for the higher incidence of type 2 diabetes and coronary
heart disease observed in this population. To evaluate whether Asian
Indians are more insulin resistant than Caucasians and to define the
role of generalized and truncal adiposity, we performed hydrodensi-
tometry, skinfold measurements, and euglycemic-hyperinsulinemic
clamps in 21 healthy Asian Indian men and 23 Caucasian men of
similar age and body fat content. The glucose disposal rate (Rd) was
significantly lower in the Asian Indians than in the Caucasians (3.7 6
1.3 vs. 5.3 6 2.0 mg/minzkg lean body mass, respectively; P 5 0.003).
Despite similar total body fat content, Asian Indians had higher
truncal adiposity than Caucasians (sum of truncal skinfolds, 117 6 37
and 92.4 6 38 mm, respectively). In both Asian Indians and Cauca-
sians, the insulin sensitivity index (Rd/plasma insulin concentra-
tions) was inversely correlated with both total body fat (r 520.49; P ,
0.03 and r 520.67; P , 0.001, respectively) and sum of truncal
skinfold thickness (r 520.55; P , 0.001 and r 520.61; P , 0.002,
respectively). After adjustment for total body fat and truncal skinfold
thickness, Asian Indians still had a significantly lower glucose dis-
posal rate (P 5 0.04). These results show that Asian Indian men are
more insulin resistant than Caucasian men independently of gener-
alized or truncal adiposity. The excessive insulin resistance in Asian
Indians is probably a primary metabolic defect and may account for
the excessive morbidity and mortality from diabetes and coronary
heart disease in this population. (J Clin Endocrinol Metab 84: 2329 –
PIDEMIOLOGICAL studies have shown that Asian In-
dians who have migrated to western countries as well
as those living in urban areas of the Indian subcontinent have
a higher prevalence of coronary heart disease (CHD) than do
Caucasians of European ancestry (1–7). However, measure-
ments of CHD risk factors, i.e. elevated serum cholesterol,
hypertension, and obesity, among Asian Indians reveal them
to be no more common than among Caucasians (8 –10). Thus,
other factors must account for the unusually high prevalence
of CHD in migrating and urbanized Asian Indians. One clue
to the cause of premature CHD in this population may a
concomitant propensity to type 2 diabetes (11–15). The prev-
alence of type 2 diabetes among Asian Indians is much higher
than would be anticipated from their degree of obesity (6, 14).
Recent reports indicate that Asian Indians in urban settings
are more insulin resistant than matched Caucasian controls
(16–20). This condition could contribute to a higher incidence
of type 2 diabetes and even to premature CHD.
One factor contributing to insulin resistance is obesity.
Very obese persons are almost uniformly insulin resistant
(21–23). Asian Indians, however, rarely have marked obesity;
nonetheless, they have been reported to have insulin resis-
tance in the urban setting with only mild obesity (10 –14).
Urban habits, their accompanying mild obesity, and limited
physical activity may be enough to induce insulin resistance
in this population. Moreover, studies in Caucasians reveal
that even a moderate degree of obesity can elicit insulin
resistance when fat is accumulated predominantly in the
trunk (24–28). Truncal obesity can be identified clinically by
an increase in waist circumference or an increase in truncal
skinfold thickness. Individuals with abnormal fat distribu-
tion, characterized by a high waist to hip circumference ratio
or a high truncal to peripheral skinfold thickness ratio appear
to be predisposed to developing insulin resistance (29). The
mechanistic basis of the association between truncal obesity
and insulin resistance is unknown. Some data, nonetheless,
suggest that Asian Indians are susceptible to developing
truncal obesity, which might account for their propensity to
insulin resistance. The present study was carried out to ad-
dress three questions. 1) Are Asian Indian men more insulin
resistant than Caucasian men matched for total body fat
content? 2) If so, are Asian Indian men more prone to pre-
dominantly truncal obesity fat distribution than are Cauca-
sian men? 3) Does insulin resistance occur independently of
obesity in Asian Indians?
Subjects and Methods
Two groups of men participated in this study: 23 were Caucasians of
European ancestry and 21 were Asian Indians from the Indian subcon-
tinent, temporarily living in the United States. They were recruited for
this study by public advertisement. The study was approved by the
Received January 7, 1999. Revision received March 12, 1999. Accepted
March 19, 1999.
Address all correspondence and requests for reprints to: Nicola Abate,
M.D., Department of Internal Medicine, Center for Human Nutrition, Uni-
versity of Texas Southwestern Medical Center, 5323 Harry Hines Boule-
vard, Dallas, Texas 75235-9052. E-mail: firstname.lastname@example.org.
* This work was supported by NIH Grants HL-29252, DK-42582,
DK-02700, and MO1-RR-00633 (NIH, DHS, DHHS); unrestricted grants
from Merck; Bristol-Myers-Squibb; the Southwestern Medical Founda-
tion; and the Moss Heart Foundation (Dallas, TX).
0021-972X/99/$03.00/0 Vol. 84, No. 7
The Journal of Clinical Endocrinology & Metabolism Printed in U.S.A.
Copyright © 1999 by The Endocrine Society
institutional review board of the University of Texas Southwestern Med-
ical Center (Dallas. TX). Volunteers were interviewed and screened for
hematological and blood chemistry abnormalities. Subjects with diabe-
tes mellitus and other endocrine disorders, coronary artery disease, liver
function tests abnormalities, and those receiving any form of therapy
were excluded. All subjects were weight stable before entering the study.
After obtaining written informed consent, the study subjects were ad-
mitted for 3 days to the General Clinical Research Center of the Uni-
versity of Texas Southwestern Medical Center. All subjects were pro-
vided with an isocaloric diet (calculated from height, weight, and age)
during the hospitalization period.
Oral glucose tolerance tests (OGTTs). A standard OGTT with 75 g glucose
(Tru-Glu100, Fisher Scientific, Pittsburgh, PA) was conducted after 12 h
of overnight fasting on the first day of admission. An iv catheter was
placed in a forearm vein, and blood was collected for determination of
glucose and insulin concentrations before glucose administration and at
30-min intervals thereafter for 180 min.
Anthropometric measurements. Height and weight were measured by stan-
dard procedures. Waist and hip circumferences were measured, using
a flexible measuring tape with a tension caliper at the extremity (Gulick-
Creative Health Product, Inc., Plymouth, MI), midway between the
xiphoid and the umbilicus during the midinspiratory phase and at the
maximum circumference in the hip area, respectively. The waist to hip
circumference (W/H) ratio was calculated for each subject. Skinfold
thickness was measured at nine different anatomical sites [subscapular
(diagonal and vertical), chest, midaxillary, abdominal (horizontal and
vertical), suprailiac (diagonal and vertical), triceps, biceps, thigh, and
calf], using a Lange skinfold caliper (Cambridge Scientific Industries,
Inc., Cambridge, MD). The same investigator (M.C.) performed all skin-
fold measurements to minimize interinvestigator variability. The means
of three repeat measurements at each site were used for calculations. The
horizontal/diagonal and vertical measurements of the subscapular, su-
prailiac, and abdominal skinfolds were averaged. The sum of truncal
skinfold thickness was calculated by adding the skinfold thickness of
subscapular, midaxillary, chest, abdomen, and suprailiac sites, and the
sum of peripheral skinfold thickness was calculated by adding skinfold
thickness of triceps, biceps, thigh, and calf regions. Body composition
was studied by determination of body density in a Whitmore volumeter
(Whitmore Enterprises, San Antonio, TX). Each subject was submerged
in water up to the chin in a seated position. Then he was given 3000 mL
gas to rebreathe (45% oxygen, 10% helium, and 45% nitrogen) and went
completely underwater. Total volume displacement was measured to
the nearest 50 mL. After resurfacing, the helium (He) concentration was
measured in the exhaled gas by mass spectrometry (model 1100, Perkin
Elmer, St. Louis, MO). Total submerged gas volume was calculated by
the formula: total gas volume (mL) 5 300 mL He/final He conc. 1 100
mL (for abdominal gas). Total gas volume was subtracted from total
displacement volume to give total body volume. Total body mass (ki-
lograms), measured to the nearest 0.1 kg, was divided by body volume
to obtain body density. Siri’s equation (30) was used to estimate the
percentage of total body fat, lean body mass, and total fat mass.
Euglycemic, hyperinsulinemic clamp study. Clamp studies were conducted
on the last day of admission after an overnight fast. A primed continuous
infusion of regular insulin (Humulin, Squibb-Novo, Princeton, NJ) was
given iv at a rate of 20 mU/m
zmin from 0–120 min. Blood samples were
obtained every 5 min from a catheter placed retrograde in a dorsal vein
of a hand kept in a radiant warmer at 70 C to arterialize venous blood.
Dextrose solution (20%) was infused iv to maintain plasma glucose at
the fasting levels throughout the clamp procedure, according to the
method of DeFronzo et al. (31). To study glucose turnover, a primed
continuous iv infusion of [3-
H]glucose (DuPont-NEN, Boston, MA) was
started at a rate of 2.36 nCi/kgzmin at 120 min before the initiation of
insulin infusion (2120 min) and was continued throughout the duration
of the clamp. To minimize the physiologically unacceptable negative
values of hepatic glucose output (HGO) during the hyperinsulinemic
phase of the study, the 20% dextrose solution was “spiked” with
H]glucose to maintain a constant specific activity according to the
method of Finegood et al. (32). Blood samples for measurement of glu-
cose, insulin, and [3-
H]glucose specific activity were collected at 10-min
intervals from 240 to 0 min before and from 80 –120 during the study.
The rate of glucose appearance (Ra) in plasma was calculated by mea-
suring specific activity of [3-
H]glucose in the plasma using the one-
compartment model described by Steele et al. (33) and modified, for
labeled variable glucose infusion, by Finegood et al. (32), assuming a
volume of distribution of 210 mL/kg. Endogenous glucose production
or HGO during the clamp was calculated as the difference between the
Ra and the glucose infusion rate for the time interval, and negative
values, if any, were assumed to be equal to zero. The rate of glucose
disposal (Rd) was calculated by subtracting the urinary glucose excre-
tion from the Ra and using space correction. The data for HGO and Rd
are presented in milligrams per min/kg lean body mass.
Biochemical analyses. Plasma and urinary glucose concentrations were
assayed using a glucose oxidase method (glucose analyzer, Beckman
Coulter, Inc., Fullerton, CA). The specific activity of glucose was deter-
mined from the plasma samples deproteinized by barium hydroxide and
zinc sulfate precipitation, according to the method of Meneilly et al. (34).
Plasma insulin levels were determined by a modification (35) of the RIA
described by Yalow and Berson (36).
Statistical analysis was performed using BMDP (SPSS, Inc., Chicago,
IL). Due to the skewness of some data (e.g. triglycerides), the nonpara-
metric Mann-Whitney U test was used to compare the Caucasians and
Asian Indians. Regression analysis was used to assess the relationship
between Rd values and regional adiposity and to compare the regression
lines between the two groups. Due to the curvilinear trend, the regres-
sion analysis was also performed using the log-transformed dependent
variable. As the transformed results were essentially the same as the
original units, the untransformed results are reported.
In Table 1, the general characteristics of Asian Indian sub-
jects are compared with those of the Caucasian subjects.
Despite the similar age range, the mean age of Caucasian
subjects was slightly higher than that of the Asian Indian
subjects. The average body mass indexes for the two groups
were similar. Fasting plasma glucose, total cholesterol, total
triglycerides, and low density lipoprotein cholesterol were
similar in both the groups. However, plasma high density
lipoprotein (HDL) cholesterol concentrations were signifi-
cantly lower in the Asian Indian group.
The plasma glucose and insulin concentrations during the
OGTT are illustrated in Fig. 1. Mean plasma glucose and
insulin levels were higher in the Asian Indian group at all
time points during the OGTT. The areas under the curve for
both plasma glucose and insulin were significantly higher in
the Asian Indian group (P 5 0.02 and P 5 0.004 for glucose
and insulin, respectively). The data for body composition
and fat distribution are summarized in Table 2. There was no
TABLE 1. General characteristics of Caucasian and Asian Indian
Variable Asian Indians Caucasians P
No. of subjects 21 23
Age (yr) 35 6 9406 9 0.08
) 24.4 6 2.1 25.2 6 3.4 0.36
Fasting plasma glucose (mg/dL) 97.5 6 5.8 94.5 6 6.8 0.12
Plasma total cholesterol (mg/dL) 167 6 26 182 6 37 0.13
Plasma total triglycerides (mg/dL) 118 6 52 136 6 78 0.39
Plasma HDL- cholesterol (mg/dL) 32 6 7376 10 0.02
Plasma LDL- cholesterol (mg/dL) 117 6 19 123 6 30 0.47
Data are the means 6 SD. P values were determined using the
Mann-Whitney U test.
2330 CHANDALIA ET AL.
Vol 84 • No 7
difference in the percentage of total body fat mass between
the two groups. Both waist and hip circumferences were
lower in Asian Indians compared to Caucasians, although
the difference did not reach statistical significance. However,
the sums of truncal skinfold thickness were significantly
higher in Asian Indians compared to Caucasians, whereas
the sums of peripheral skinfold thickness were similar be-
tween the groups (see also Fig. 2). Truncal to peripheral
skinfold ratios were significantly higher in Asian Indians, but
waist to hip ratios were similar in both groups. In Fig. 3, the
individual data on skinfold thickness in various anatomical
sites in the two study groups are compared. Overall, the
skinfold thickness of all truncal areas was higher in the Asian
Indian group, although a statistically significant difference
was observed only for the chest and subscapular areas. The
sum of the skinfold thickness in all peripheral areas was
similar for both groups.
Table 3 is a summary of the results of the euglycemic-
hyperinsulinemic clamp studies. Plasma insulin levels dur-
ing fasting and infusion were significantly higher in Asian
Indians compared to Caucasians. Fasting HGO were similar
in both groups and were suppressed to a similar extent in
both groups during the hyperinsulinemic clamp. The rate of
glucose disposal was significantly lower in Asian Indians
than that in Caucasians.
In the Caucasian group, but not in the Asian Indian group,
the rate of glucose disposal was significantly correlated with
total body fat (r 5 0.68; P 5 0.0003 and r 5 0.34; P 5 0.13,
respectively). However, the relationship between peripheral
glucose disposal and truncal skinfold thickness was highly
significant in both groups (r 5 0.65; P 5 0.0006 and r 5 0.47;
P 5 0.0003, for Caucasians and Asian Indians, respectively).
As the insulin levels during the clamp were higher in the
Asian Indians, the correlation analysis was performed using
the Rd/insulin concentration during clamp (insulin sensi-
tivity index) as the dependent variable. Figures 4 and 5
FIG. 1. Plasma glucose and insulin
concentrations during OGTT.
TABLE 2. Body composition and fat distribution in Caucasians and Asian Indians
Asian Indians Caucasians P
Fat content (% of total body mass) 20.3 6 5.7 19.0 6 6.6 0.51
Lean content (% of total body mass) 79.7 6 5.7 81.0 6 6.6 0.51
Waist (cm) 83.6 6 6.7 89.4 6 11.3 0.06
Hip (cm) 94.0 6 5.9 98.0 6 7.5 0.06
Sum of truncal skinfolds (mm) 117.0 6 37.0 92.4 6 38.0 0.03
Sum of peripheral skinfolds (mm) 42.0 6 13.0 42.0 6 15.0 1
Fat distribution indexes
Waist to hip circumference ratio 0.90 6 0.06 0.91 6 0.08 0.4
Truncal to peripheral skinfolds ratio 2.83 6 0.66 2.22 6 0.52 0.002
Data are the mean 6 SD. P values were determined using the Mann-Whitney U test.
INSULIN RESISTANCE IN ASIAN INDIAN MEN 2331
illustrate the relationships between total body fat and truncal
skinfold thickness and the insulin sensitivity index. The re-
gression curves shown in Figs. 4 and 5 illustrate that for any
given amount of adipose tissue in either the total body or
truncal area, the Asian Indians had lower insulin sensitivity
to insulin-mediated glucose disposal. This apparent differ-
ence was further examined by comparing the regression lines
between the insulin sensitivity index and truncal adipose
tissue for the two groups. The slopes of the regression lines
for total body fat were similar for the two groups (P 5 0.09).
However, the intercept of insulin sensitivity index for the
regression line was significantly lower for Asian Indian sub-
jects than for Caucasian subjects (P 5 0.002). Similarly, the
slopes of the regression lines for the sum of truncal skinfold
thickness were similar for the two groups (P 5 0.14), but the
intercept was significantly lower for the Asian Indians than
for the Caucasian subjects (P 5 0.04). A subgroup analysis
including the subjects with total body fat less than 20% of
total body mass confirmed that the insulin sensitivity index
was significantly lower in the Asian Indians compared to the
Caucasians (6.23 6 1.63 and 4.17 6 1.46 mg/minzkg lean
body mass, respectively; P 5 0.009). The insulin sensitivity
index in the Asian Indians remained significantly lower than
that in the Caucasians after adjustment for both total body fat
and truncal skinfold thickness (P 5 0.04).
Several previous reports indicate that Asian Indians carry
an increased susceptibility to both early-onset type 2 diabetes
(11–15) and premature CHD (1–7). For both conditions, this
augmented susceptibility cannot be explained by the usual
risk factors for these diseases. For example, the greater pro-
pensity to type 2 diabetes in Asian Indians cannot be ex-
plained by a high prevalence of marked obesity, such as
occurs in the Native American population in Arizona. Nei-
ther is susceptibility to premature CHD due to an unusually
high prevalence of major coronary risk factors, e.g. cigarette
smoking, hypertension, or high serum cholesterol (8 –10).
The relatively high prevalence of premature CHD in urban
and migrating Asian Indians, compared to that in subjects
living in rural areas, may be accentuated by a higher fre-
quency of type 2 diabetes; nevertheless, most cases of pre-
mature CHD occur in the absence of diabetes. Hence, other
factors must be sought as contributors to the high prevalence
of premature CHD in urban and migrating Asian Indians.
One hypothesis has been proposed that could account for
the high frequency of both type 2 diabetes and premature
CHD in Asian Indians. This hypothesis maintains that Asian
Indians are susceptible to a generalized metabolic condition
commonly called the insulin resistance syndrome (1). Pro-
longed insulin resistance confers an increased risk for the
development of type 2 diabetes (37), which is an independent
risk factor for CHD. In addition, insulin resistance is often
accompanied by other coronary risk factors, e.g. dyslipide-
mia and hypertension (38– 40). Finally, it is possible that
insulin resistance affects CHD risk status through other
mechanisms that are independent of the established risk
Several lines of evidence suggest that Asian Indians are
predisposed to developing insulin resistance, based on the
usual features of this syndrome. These include a relatively
high prevalence of type 2 diabetes (1–7), a tendency to truncal
obesity (12, 14, 20), an increased frequency of fasting hyper-
insulinemia (13, 16), and other metabolic indicators of insulin
resistance; among the latter are a hyperinsulinemic response
to an oral glucose challenge (13, 16) and abnormal steady
state concentrations of glucose during an insulin suppression
test with somatostatin (17). The present study was carried out
to determine whether a propensity to insulin resistance could
be confirmed by glucose clamp studies. The glucose clamp
technique provides quantitative data for one parameter of
insulin sensitivity, namely rates of glucose disposal at a given
level of plasma insulin.
The current study compared young adult Asian Indian
men living in the United States with Caucasian American
men. The subjects were of comparable age and body fat
content. As previously reported (13–17), the Asian Indian
men in this study had significantly higher fasting levels of
insulin than Caucasian men did; however, differences were
FIG. 3. Individual data on skinfold thickness in various anatomical
sites in the two study groups are compared.
FIG. 2. Body composition and fat distribution in the two groups.
There was no difference in the percentage of total body fat mass
between the two groups. Waist and hip circumferences were lower in
Asian Indians compared to Caucasians, although the difference did
not reach statistical significance. However, the sums of truncal skin-
fold thickness were significantly higher in Asian Indians compared to
Caucasians, whereas the sums of peripheral skinfold thickness were
similar between the groups.
2332 CHANDALIA ET AL.
Vol 84 • No 7
not as marked as has been reported previously. More striking
differences were noted for the area under the curve of plasma
insulin response to an oral glucose challenge. These differ-
ences strongly suggest that Asian Indian men as a group are
insulin resistant compared to matched Caucasian men. This
tendency was confirmed with euglycemic-hyperinsulinemic
glucose clamp studies. Average rates of glucose disposal
were markedly reduced in Asian Indians compared to those
in Caucasians. A significant decrease in insulin sensitivity
apparently was present in Asian Indian men regardless of the
level of total body fat. This finding raises the possibility that
the insulin resistance in Asian Indians can occur indepen-
dently of an increase in total body fat content. However,
when the glucose sensitivity index was plotted against the
percent total body fat, an increasing percentage of body fat
was accompanied by decreasing insulin sensitivity. Thus, it
appears that whereas relatively lean Asian Indian men may
be more insulin resistant than Caucasians of similar body fat
content, increasing obesity is still accompanied by a decrease
in insulin sensitivity.
An association between insulin sensitivity and body fat
distribution has been observed in many studies (24 –29, 40 –
43). Patients who have high waist to hip ratios usually are
more resistant to the action of insulin than those with low
ratios. Some investigators contend that ip (visceral) fat is the
compartment of adipose tissue most tightly linked with in-
sulin resistance (26–28). Recent studies from our laboratory
(41, 42) and others (43), however, indicate that sc fat in the
trunk is even better correlated with insulin sensitivity than
is ip fat. This is true in both nondiabetic subjects and those
with type 2 diabetes. The mechanism underlying this asso-
ciation is not known, although Jensen et al. (44) showed that
patients with upper body obesity (truncal obesity) have
higher levels of nonesterified fatty acids (NEFA) than do
those with lower body obesity (gluterofemoral obesity).
These high NEFA levels could lead to an increased fatty acid
content of muscle and to inhibition of glucose oxidation (45).
Possibly truncal adipose tissue more readily releases NEFA
into the circulation than does gluterofemoral adipose tissue
Several reports indicate that Asian Indians are predis-
posed to upper body obesity. This tendency is reflected in
reports of increased waist to hip ratios and increased truncal
skinfold thicknesses in Asian Indians compared to other
populations (10, 17, 20). The propensity to upper body obe-
sity was confirmed in the present study. Compared to Cau-
casian men, Asian Indian men in this study had significantly
thicker truncal skinfolds. In addition, the Asian Indians had
higher ratios of truncal to peripheral skinfold thickness. It is
interesting that in this group the thickness of skinfolds was
a better indicator of predominant truncal fat accumulation
than was the waist to hip ratio and the waist circumference.
The lack of difference in waist circumference between Asian
Indians and Caucasians suggests that a greater difference
existed between sc truncal fat than between ip fat; however,
FIG. 5. Relationship of total body fat and truncal skinfold thickness
to insulin sensitivity index.
TABLE 3. Euglycemic-hyperinsulinemic clamp study data in Caucasians and Asian Indians
Asian Indians Caucasians P
Plasma insulin (
Fasting 17 6 4146 4 0.03
During insulin infusion 62 6 12 55 6 10 0.04
Rd valuez(rate of glucose disposal; mg/minzlean body mass) 3.7 6 1.3 5.3 6 2.0 0.003
Hepatc glucose output (mg/minzkg of lean body mass)
Fasting 2.06 6 0.4 2.18 6 0.5 0.4
During insulin infusion 0.34 6 0.2 0.42 6 0.6 0.5
Data are the mean 6 SD. P values were determined using the Mann-Whitney U test. Fasting values are mean of 240 to 0 min values before
the insulin infusion. Rd value and hepatic glucose output represent the means of 80–120 min of insulin infusion at 20 mU/m
FIG. 4. Relationship of total body fat and truncal skinfold thickness
to insulin sensitivity index.
INSULIN RESISTANCE IN ASIAN INDIAN MEN 2333
definite proof of the latter would require studies that image
ip fat. As other workers have reported that waist circumfer-
ence tends to be greater in some groups of Asian Indians (14,
20), it is possible that more upper body obesity exists in all
of the adipose tissue compartments of Asian Indians. As we
previously observed in Caucasians, the sum of the truncal
skinfolds was highly (and inversely) correlated with insulin
One of the most interesting observations of the current
study was a strong tendency for insulin resistance in lean
Asian Indians. The latter were much more insulin resistance
than lean Caucasians. Although the curve of insulin sensi-
tivity against percent body fat was relatively steep in Cau-
casians, this was not the case in Asian Indians. Among the
latter, increasing adiposity was accompanied by some re-
duction in insulin sensitivity, but decrements were relatively
small. This finding thus strongly suggests that Asian Indian
men living in the United States have relatively low insulin
sensitivity even when their body fat content is in the normal
range. The mechanisms responsible for the low insulin sen-
sitivity in Asian Indians, whether due to physical inactivity,
dietary differences, or hereditary factors, remain to be
The current findings seemingly do not support the concept
that upper body fat distribution is the primary cause of
insulin resistance in Asian Indians. First, lean Asian Indians
were more insulin resistant than lean Caucasians, indicating
that Asian Indians are insulin resistant with little or no upper
body obesity. Second, although Asian Indians as a group
showed a significant trend toward more truncal fat than
Caucasians, Asian Indians appeared to be more insulin re-
sistant at any level of truncal skinfold thickness. Thus, trun-
cal obesity (or predominant upper body fat) seemingly can-
not be the primary cause of low insulin sensitivity in Asian
Current and previous data indicate that Asian Indians are
predisposed to truncal obesity. Previous studies have shown
that the degree of truncal obesity is inversely correlated with
insulin sensitivity (41, 42). According to current concepts,
truncal adipose tissue has a propensity to excessive release
of NEFA, which may impair insulin sensitivity. Thus, the
tendency of Asian Indians to develop truncal obesity could
accentuate insulin resistance in a population that inherently
has a low insulin sensitivity.
Alternatively, the propensity to truncal obesity in Asian
Indians could be secondary to an underlying insulin resis-
tance. Factors that determine the distribution of body fat are
not known; the possibility that abnormal insulin action at the
level of adipose tissue could promote the accumulation of
truncal fat cannot be excluded.
As discussed before, other investigators have indicated
that the established CHD risk factors can explain only a
portion of the increased risk for CHD in Asian Indians living
in urban settings. In the present study, the only established
risk factor that was more prevalent in the Asian Indians was
a lower level of HDL cholesterol. A low HDL cholesterol
concentration has been noted previously to be strongly as-
sociated with insulin resistance (48, 49). The mechanism is
not known, but may be related to an increase in hepatic lipase
activity accompanying insulin resistance in the liver (50).
However, a lower HDL cholesterol level and other major risk
factors for CHD in Asian Indians cannot fully account for
their increased risk for premature CHD. Instead, it appears
that insulin resistance in Asian Indians is associated with
causes of CHD that have yet to be elucidated. The current
study confirms that Asian Indians are susceptible to insulin
resistance. This condition is evoked by mild or even minimal
increases in body fat content. Whereas it is true that Asian
Indian men preferentially deposit body fat in the trunk, com-
pared to Caucasians, our results suggest that predominant
upper body fat is not a direct cause of the increased insulin
resistance in Asian Indians.
The authors express appreciation for the excellent technical assistance
of Margaret Arnecke and Jerry Payne. The assistance of Marjorie Whelan
and the nursing and dietetic services of the General Clinical Research
Center are gratefully acknowledged. Kay McKorkle, Lovie Peace, and
Margaret Haney in the laboratory of Dr. Roger H. Unger assisted with
the insulin RIA.
1. McKeigue PM, Miller GJ, Marmot MG. 1989 Coronary heart disease in south
Asians overseas: a review. J Clin Epidemiol. 42:597–609.
2. Sarvotham SG, Berry JN. 1968 Prevalence of coronary heart disease in an
urban population in northern India. Circulation. 37:939 –953.
3. Dewan BD, Malhotra KC, Gupta SP. 1974 Epidemiological study of coronary
heart disease in rural community in Haryana. Indian Heart J. 26:68–78.
4. Tunstall Pedoe H, Clayton D, Morris JN, Bridgen W, McDonald L. 1975
Coronary Heart-Attacks in East London. Lancet. 2(7940):833– 838.
5. Chadha SL, Radhakrishnan S, Ramachandran K, Kaul U, Gopinath N. 1990
Epidemiological study of coronary heart disease in urban population of Delhi.
Indian J Med Res. 92:424 – 430.
6. McKeigue PM, Marmot MG, Adelstein AM, et al. 1985. Diet and risk factors
for coronary heart disease in Asians in northwest London. Lancet. 2:1086 –1090.
7. Balarajan R. 1991 Ethnic differences in mortality from ischaemic heart disease
and cerebrovascular disease in England and Wales. Br Med J. 302:560–564.
8. Beckles GL, Miller GJ, Kirkwood BR, Alexis SD, Carson DC, Byam NT. 1986
High total and cardiovascular disease mortality in adults of Indian descent in
Trinidad, unexplained by major coronary risk factors. Lancet. 1:1298 –1301.
9. Miller GJ, Beckles GL, Maude GH, et al. 1989 Ethnicity and other charac-
teristics predictive of coronary heart disease in a developing community:
principal results of the St. James Survey, Trinidad. Int J Epidemiol. 18:808 – 817.
10. McKeigue PM, Shah B, Marmot MG. 1991 Relation of central obesity and
insulin resistance with high diabetes prevalence and cardiovascular risk in
South Asians. Lancet. 337:382–386.
11. Ramachandran A, Jali MV, Mohan V, Snehalatha C, Viswanathan M. 1988
High prevalence of diabetes in an urban population in south India. Br Med J.
12. Ramachandran A, Snehalatha C, Dharmaraj D, Viswanathan M. 1992 Prev-
alence of glucose intolerance in Asian Indians. Urban-rural difference and
significance of upper body adiposity. Diabetes Care. 15:1348–1355.
13. McKeigue PM, Marmot MG, Syndercombe Court YD, Cottier DE, Rahman
S, Riemersma RA. 1988 Diabetes, hyperinsulinaemia, and coronary risk factors
in Bangladeshis in east London. Br Heart J. 60:390 –396.
14. McKeigue PM, Pierpoint T, Ferrie JE, Marmot MG. 1992 Relationship of
glucose intolerance and hyperinsulinaemia to body fat pattern in south Asians
and Europeans. Diabetologia. 35:785–791.
15. Mather HM, Keen H. 1985 The Southall Diabetes Survey: prevalence of known
diabetes in Asians and Europeans. Br Med J. 291:1081–1084.
16. Knight TM, Smith Z, Whittles A, et al. 1992 Insulin resistance, diabetes, and
risk markers for ischaemic heart disease in Asian men and non-Asian in
Bradford. Br Heart J. 67:343–350.
17. Laws A, Jeppesen JL, Maheux PC, Schaaf P, Chen YD, Reaven GM. 1994
Resistance to insulin-stimulated glucose uptake and dyslipidemia in Asian
Indians. Arterioscler Thromb. 14:917–922.
18. Hughes LO, Cruickshank JK, Wright J, Raftery EB. 1989 Disturbances of
insulin in British Asian and white men surviving myocardial infarction. Br
Med J. 299:537–541.
19. Dhawan J, Bray CL, Warburton R, Ghambhir DS, Morris J. 1994 Insulin
resistance, high prevalence of diabetes, and cardiovascular risk in immigrant
Asians. Genetic or environmental effect? Br Heart J. 72:413– 421.
20. Singh RB, Niaz MA, Agarwal P, Beegum R, Rastogi SS, Singh NK. 1995
2334 CHANDALIA ET AL.
Vol 84 • No 7
Epidemiologic study of central obesity, insulin resistance and associated dis-
turbances in the urban population of North India. Acta Cardiol. 50:215–225.
21. Bogardus C, Lillijoia S, Mott DM, Hollenbeck C, Reaven GM. 1985 Rela-
tionship between degree of obesity and in vivo insulin action in man. Am J
Physiol. 248:E286 –E291.
22. Yki-Jarvinen H, Koivisto VA. 1983 Effects of body composition on insulin
sensitivity. Diabetes. 32:965–969.
23. DeFronzo RA. 1982 Insulin secretion, insulin resistance, and obesity. Int J
Obesity. 6:73– 82.
24. Kissebah AH, Vydelingum N, Murray R, et al. 1982 Relation of body fat
distribution to metabolic complications of obesity. J Clin Endocrinol Metab.
25. Pedersen SB, Borglum JD, Schmitz O, Bak JF, Sorensen NS, Richelsen B.
1993 Abdominal obesity is associated with insulin resistance and reduced
glycogen synthetase activity in skeletal muscle. Metabolism. 42:998–1005.
26. Sparrow D, Borkan GA, Gerzof SG, Wisniewski C, Silbert CK. 1986 Rela-
tionship of fat distribution to glucose tolerance. Results of computed tomog-
raphy in male participants of the Normative Aging Study. Diabetes.
27. Krotkiewski M, Bjorntorp P, Sjostrom L, Smith U. 1983 Impact of obesity on
metabolism in men and women. Importance of regional adipose tissue dis-
tribution. J Clin Invest. 72:1150 –1162.
28. Despres JP, Moorjani S, Ferland M, et al. 1989 Adipose tissue distribution and
plasma lipoprotein levels in obese women. Importance of intra-abdominal fat.
29. Lundgren H, Bengtsson C, Blohme G, Lapidus L, Sjostrom L. 1989 Adiposity
and adipose tissue distribution in relation to incidence of diabetes in women:
results from a prospective population study in Gothenburg, Sweden. Int J
30. Siri WE. 1961 Techniques for measuring body composition. Washington DC:
National Academy of Sciences; 223–224.
31. DeFronzo RA, Tobin JD, Andres R. 1979 Glucose clamp technique: a method
for quantifying insulin secretion and resistance. Am J Physiol. 233:E214–E223.
32. Finegood DT, Bergman RN, Vranic M. 1987 Estimation of endogenous glu-
cose production during hyperinsulinemic-euglycemic glucose clamps. Com-
parison of unlabeled and labeled exogenous glucose infusates. Diabetes.
33. Steele R. 1959 Influence of glucose loading and of injected insulin on hepatic
glucose output. Ann NY Acad Sci. 82:420 – 443.
34. Meneilly GS, Elahi D, Minaker KL, Rowe JW. 1987 Somatostatin does not
alter insulin-mediated glucose disposal. J Clin Endocrinol Metab. 65:364 –370.
35. Herbert V, Lau K, Gottlieb CW, Bleicher SJ. 1965 Coated charcoal immu-
noassay of insulin. J Clin Endocrinol Metab. 25:1375–1384.
36. Yalow RS, Berson SA. 1960 Immunoassay of endogenous plasma insulin in
man. J Clin Invest. 39:1157–1175.
37. Lillijoia S, Mott DM, Spraul M, et al. 1993 Insulin resistance and insulin
secretory dysfunction as precursor of non-insulin-dependent diabetes melli-
tus. N Engl J Med. 329:1988–1992.
38. Pyorala K, Savolainen E, Kaukola S, Haapakoski J. 1985 Plasma insulin as
coronary heart disease risk factor: relationship to other risk factors and pre-
dictive value during 91/2-year follow-up of the Helsinki Policemen Study
population. Acta Medica Scand. 701:38 –52.
39. Haffner SM, Fong D, Hazuda H, Pugh JA, Patterson JK. 1988 Hyperinsu-
linemia, upper body adiposity, and cardiovascular risk factors in non-diabet-
ics. Metabolism. 37:338 –345.
40. Zavaroni I, Bonora E, Pagliara M, et al. 1989 Risk factors for coronary artery
disease in healthy persons with hyperinsulinemia and normal glucose toler-
ance. N Engl J Med. 320:702–706.
41. Abate N, Garg A, Peshock RM, Stray-Gundersen J, Grundy SM. 1995 Re-
lationships of generalized and regional adiposity to insulin sensitivity in men.
J Clin Invest. 96:88–98.
42. Abate N, Garg A, Peshock RM, Stray-Gundersen J, Adams-Huet B, Grundy
SM. 1996 Relationship of generalized and regional adiposity to insulin sen-
sitivity in men with NIDDM. Diabetes. 45:1684 –1693.
43. Goodpaster BH, Thaete FL, Simoneau JA, Kelley DE. 1997 Subcutaneous
abdominal fat and thigh muscle composition predict insulin sensitivity inde-
pendently of visceral fat. Diabetes. 46:1579–1585.
44. Jensen MD, Haymond MW, Rizza RA, Cryer PE, Miles JM. 1989 Influence
of body fat distribution on free fatty acid metabolism in obesity. J Clin Invest.
45. Randle PJ, Garland PB, Hales CN, Newsholme EA. 1963 The glucose-fatty
acid cycle. Its role in insulin sensitivity and the metabolic disturbances of
diabetes mellitus. Lancet. 1:785–789.
46. Arner P, Engfeldt P, Lithell H. 1981 Site differences in the basal metabolism
of subcutaneous fat in obese women. J Clin Endocrinol Metab. 53:948–952.
47. Jansson PA, Smith U, Lonnroth P. 1990 Interstitial glycerol concentration
measured by microdialysis in two subcutaneous regions in humans. Am J
Physiol. 258:E918 –E922.
48. Garg A, Helderman JH, Koffler M, Ayuso R, Rosenstock J, Raskin P. 1988
Relationship between lipoprotein levels and in vivo insulin action in normal
young white men. Metabolism. 37:982–987.
49. Karhapaa P, Malkki M, Laakso M. 1994 Isolated low HDL cholesterol. An
insulin-resistant state. Diabetes. 43:411–417.
50. Nie L, Wang J, Clark LT, et al. 1998 Body mass index and hepatic lipase gene
(LIPC) polymorphism jointly influence postheparin plasma hepatic lipase
activity. J Lipid Res. 39:1127–1130.
INSULIN RESISTANCE IN ASIAN INDIAN MEN 2335