Clinical Science (1994) 87, 581-586 (Printed in Great Britain)
581
Regional fat distribution by dual-energy X-ray
absorptiometry: comparison with anthropometry and
application in a clinical trial of growth hormone
and exercise
Dennis R. TAAFFE, Barbara LEWIS and Robert MARCUS
Musculoskeletal Research laboratory, Aging Study Unit, Geriatric Research, Education and
Clinical Center, Veterans Affairs Medical Center, Palo Alto, CA, U.S.A., and Department of
Medicine, Stanford University, Stanford, CA, U.S.A.
(Received 10 March/l5 June 1994; accepted 4 July 1994)
1. The purpose of this study was to determine the
suitability of ratios derived from dual-energy X-ray
absorptiometry (DXA) whole body scans to assess
regional fat distribution in older men and women by
comparing them with the waist-to-hip ratio (WHR)
and to evaluate their clinical utility by applying them
in a clinical trial involving resistance exercise and
recombinant human growth hormone.
2. Sixty-four healthy older adults (39 women and 25
men), aged 65-82 years, served as subjects. The
ratios of trunk fat-to-total fat, trunk fat-to-body
weight, trunk fat-to-limb fat and trunk fat % were
determined by DXA. WHR was assessed on the same
day, as was the ratio of subscapular/triceps skinfolds
in men. Cardiovascular disease risk factors, functional capacity and serum lipids were also assessed.
3. A moderate relationship ( r = 0.360.54) between
the WHR- and DXA-derived ratios were observed for
both men and women. Both DXA and WHR showed
similar associations with cardiovascular disease risk
factors. However, in men, all DXA ratios were able
to detect subtle changes in regional fat distribution
resulting from daily administration of recombinant
human growth hormone in conjunction with resistance
exercise for 10 weeks, whereas the WHR or subscapular/triceps ratios did not.
4. This suggests that DXA-derived ratios may be
more sensitive than conventional anthropometric
methods in the assessment and categorization of body
fat distribution.
INTRODUCTION
It is well recognized for health risk that in
addition to adiposity, site of body fat deposition is
crucial. Central obesity is associated with an increased risk for cardiovascular disease (CVD) and
metabolic disease as well as overall mortality,
whereas gluteal-femoral pattern obesity is not [l81. Assessment of adipose tissue distribution is
conventionally undertaken anthropometrically using
girths such as the waist and hip or by skinfolds at
several sites on the trunk and extremities. These
measurements can then be used in ratios that
partition body fat into central and peripheral or
upper and lower body portions, such as the
subscapular-to-triceps [2] and waist-to-hip ratios
(WHRs).
Of the anthropometric methods, the WHR is the
most widely used in epidemiological and clinical
research and practice [8]. However, the WHR,
which is dependent on pelvic skeletal structure, does
not differentiate between fat and fat-free tissues and
is prone to several methodological problems [9-111.
Direct quantitative techniques, computed tomography [12, 131 and magnetic resonance imaging
[14, 151, have therefore been proposed to examine
regional fat distribution, however, the equipment is
expensive and not readily available. In addition,
computed tomography subjects the individual to a
relatively high radiation dose. Recent modification
of dual-energy X-ray absorptiometry (DXA) software to measure whole body and regional tissue
composition lends itself to examining fat distribution in individuals [16, 171, especially in populations unable to undergo standard anthropometric
assessment, such as the very old, the sick, and nonambulatory individuals.
With the expanded use and availability of DXA
to assess body composition, the purpose of this
study was to evaluate the clinical utility of ratios of
fat distribution derived from DXA by addressing the
following three issues: (1) the ability of these ratios
to simulate the WHR in older men and women, a
population with increased total and central fat mass
Key words: duaknergy X-ray absorptiometry, growth hormone, older men and women, regional fat distribution, resistance exercise.
Abbreviations: BMI, body mass index; CHOL, total cholesterol: CVD, cardiovascular disease; DXA. dual energy X-ray absorptiometry; GH. growth hormone; MET, metabolic
equivalent; LBM, lean body mass; PL, placebo; rhGH, recombinant human growth hormone; TrF, trunk fat; TrF/LimF. trunk-fat-telimb fat: TrF/BW. trunk-fat-to-body weight;
TrFjTotF. trunk fat-to-total fat: TRIG, triacylglycerol; WHR, waist-to-hip ratio.
Correspondence: Dr Dennis R. Taaffe. GRECC 182-8, Veterans Affairs Medical Center, 3801 Miranda Avenue, Palo Alto, CA 94304, U.S.A.
582
D. R. Taaffe et al.
[18, 191 and an increased risk for cardiovascular
disease [20, 213; (2) the relationship of DXA ratios
and the WHR to selected CVD risk factors (functional capacity and serum lipids); and (3) the sensitivity of DXA-derived ratios versus conventional
anthropometric methods in determining longitudinal
changes in body fat distribution. To evaluate the
last point we monitored the changes in body
composition of a group of older men who were
participating in a trial of resistance exercise and
recombinant human growth hormone (rhGH).
MET H0DS
Subjects
Sixty-four healthy older adults, 39 women (65-82
years) and 25 men (66-79 years) with a body mass
index (BMI) of < 30 kg/m*, recruited to participate
in exercise studies, served as subjects. All participants completed a screening procedure that consisted of a health history questionnaire, physical
examination, multiphasic laboratory profile, maximal exercise stress test, and lateral thoracic and
lumbar spine radiographs. All subjects were apparently healthy, free of disease and disorders that
would prohibit them from participating in a vigorous exercise programme. Activity history upon entry
revealed participants were sedentary to moderately
active. Eighteen of the women were taking exogenous oestrogen. The procedures were approved by
the Human Subjects Committee of Stanford University, and all subjects gave written consent.
Exercise protocol
The 25 men participated in a randomized doubleblind, placebo-controlled exercise trial to determine
whether the administration of rhGH enhanced the
muscle strength response to a programme of progressive resistance exercise. Individuals initially
underwent a 14 week training period to invoke a
trained state and achieve a levelling off in muscle
strength. Subjects then received daily injections of
either rhGH (0.02 mg/kg body weight; Somatropin;
Genentech, Inc., San Francisco, CA, U.S.A.) or an
equal volume of placebo (PL) (Genentech, Inc.)
while continuing to train for an additional 10 weeks.
Eighteen subjects completed the 24 week training
programme. Muscle strength results of this trial will
be separately reported.
Briefly, the training programme consisted of progressive resistance exercise, three sessions per week.
Each session consisted of a circuit of 10 exercises
involving major upper and lower body muscle
groups using Universal (Universal Gym Equipment,
Cedar Rapids, I A , U.S.A.) and Nautilus (Nautilus
Sports/Medical Industries, Inc, Independence, VA,
U.S.A.) equipment. Subjects performed three sets of
eight repetitions for each exercise at an initial
intensity equal to 75% of their individual one-
repetition maximum (1-RM), which is the maximal
weight an individual can lift one time with acceptable form. To assure the progressive nature of the
programme, 1-RM values for all exercises were
recorded every two weeks.
Body composition analysis
Lean body mass (LBM, kg), fat mass (kg), and
percentage body fat were assessed by DXA (Hologic
Q D R lOOO/W, Waltham, MA, U.S.A., software
version 5.47). In addition, trunk mass, limb mass,
and bone-free lean mass of the limbs and trunk
were derived from the whole body scan. The scan
time was approximately 14min with a whole body
radiation dose of < 1 mrem. The coefticient of
variation for replicate measurements in our laboratory is < I % for LBM, fat mass (kg), percentage
body fat, limb mass, trunk mass and trunk lean
mass, and 1.1% for limb lean mass.
Body fat distribution
The ratios of trunk fat-to-total fat (TrF/TotF),
trunk fat-to-body weight (TrF/BW), trunk fat-tolimb fat (TrF/LimF), and trunk fat % (TrFO/,;trunk
mass comprised of trunk fat) were derived from
DXA (Hologic, lOOO/W) whole body scans. Default
segmentation lines for whole body analysis were
placed at specific anatomic landmarks that divided
the body into upper limbs, lower limbs, trunk, and
head. To define the upper limb as a separate
segment a vertical line was extended between the
head of the humerus and the glenoid fossa of the
scapula. An oblique line placed through the femoral
neck demarcates the lower limbs from the trunk.
This segment includes the lateral portion of the hip
region and hence gluteal-femoral subcutaneous fat.
The head is separated from the trunk by a horizontal line placed inferior to the mandible, therefore the
trunk includes the thorax, abdomen, pelvis, and the
superior medial portion of the thigh. Limb fat
comprises the sum of both upper and lower limbs.
The precision error (coefficient of variation for
duplicate comparisons) in our laboratory for upper
limb fat mass is 2.3%, for lower limb fat mass 1.1%,
trunk fat mass 1.9%, and trunk percentage fat 1.4%.
In the men, we also assessed peripheral and central
fat distribution by measuring triceps and subscapular skinfolds in triplicate with Lange calipers
(Cambridge Instruments, Cambridge, MD, U.S.A.).
WHR
Circumference measures were performed in triplicate with the subject in the standing position. Waist
circumference was measured at the natural waistline
between the lower rib margin and the iliac crest.
When the natural waistline was not discernible the
waist was measured at the umbilicus. Hip circumference was measured at the level of widest circumfer-
Regional fat distribution and dual-energy X-ray absorptiometry
ence over the buttocks. Both measures were
rounded to the nearest 0.1 cm.
Table I.Subject characteristics. Values are mean f SEM. Statistical significance: *P<O.O5, tP<0.001 for women compared with men.
Functional exercise capacity
Peak functional capacity or physical fitness was
determined by a symptom-limited maximal graded
treadmill test using a modified Balke protocol [22].
Treadmill speed was kept constant at 2.2 miles/h
with the grade increasing l%/min from the initial
grade of 0%. Blood pressure, heart rate and ECG
were monitored throughout the test. Functional
capacity in metabolic equivalents (METs) was determined indirectly by treadmill speed and grade [23].
Lipid profiles
Total cholesterol (CHOL) and triacylglycerol
(TRIG) were determined from fasting serum
samples. Total cholesterol was measured using cholesterol esterase on a Technicom SMAC 3 System
and triacyglycerol was determined using an enzymatic method on the Kodak Ektachem 700.
Statistical analysis
Data were analysed with a statistical software
package (Statview 11, Abacus Concepts Inc.,
Berkeley, CA, U.S.A.). Differences between men and
women were determined by two-tailed unpaired t tests, as were differences between women taking
exogenous oestrogen and those not taking oestrogen. Linear regression was used to examine the
relationship between variables. Correlation coefficients were derived from the linear regression program. A two-way (group x time) repeated measures
analysis of variance was used to determine the
changes in body composition resulting from the
exercise and rhGH intervention programme. Where
appropriate, the Scheffe test was employed to locate
the source of significant differences. An alpha level
of 0.05 was required for significance. Results are
given as meansfSEM.
RESULTS
Characteristics of the study group are shown in
Table 1. There were no significant differences in any
measured variable between women taking exogenous oestrogen and those not taking oestrogen,
therefore results for these groups are pooled.
There was no difference between men and women
in age. However, men were taller, heavier, had a
higher LBM and lower percentage body fat than
women. Total body fat was significantly greater in
women, with the difference due to an increased limb
fat mass. As expected, the WHR was higher in men
as were the DXA ratios of TrF/LimbF and TrF/
TotF (Table 2). As men had a significantly higher
body weight and trunk mass than women, the ratio
of TrF/BW and TrF% were lower in men.
583
Age (yean)
Height (cm)
Weight (kg)
BMI (kg/mz)
LBM (kg)
Body fat (%)
Body fat (kg)
Trunk mass (kg)
Limb mass (kg)
Trunk fat (kg)
Limb fat (kg)
METs
CHOL (mmol/l)
TRIG (mmol/l)
Women
Men
(n = 39)
(n = 25)
68.5 f 0.6
161.7+ 1.0
65.4 k I.3
25.0 f 0.4
43.0 f 0.5
33.6k1.1
22.4+ 1.1
32.7 f 0.7
28.3 f0.7
9.9 f 0.6
11.7 0.6
6.6 0.3
5.90 f0.16
I .35 k 0.1 I
69.9 & 0.7
176.5+_1.6t
81.8 f 2.1 t
26.3 0.5
63.0 +_ I.4t
22.5 f 0.9t
18.5kl.l*
41.9+ I.2t
34.8 f 0.9t
9.4 f 0.6
8.4f0.5t
9.0 f 0.27
5.74 fO.18
1.66 k0.16
+
+
Table 2. Ratios of fat distribution by anthropometry and DXA.
Values are means SEM. Statistical significance: *P < 0.001 for women
compared with men.
WHR
TrF/LimbF
TrF/TotF
TrF/BW
TrF%
Women
(n = 39)
Men
(n = 25)
0.803 +O.OlO
0.837 k0.036
0.431 +O.OlO
0. I47 f0.007
29.4 f I.3
0.968 f0.010*
1.125 fO.M3*
0.497 f0.0IO*
0. II3 f 0.006'
22.0f I.o*
Table 3. Relationship of DXA ratios ( x variable) to W H R ( y variable). Abbreviations: SEE, standard error of the estimate, NS. not
significant.
x variable
Regression equation
R
SEE
Significance
-
Women (n = 39)
TrF/LimbF
TrF/TotF
TrF/BW
TrF%
y = 0.666 +0.164~
y =0.573 +0.533~
y=O.71 I + 0 . 6 2 5 ~
y =0.722 +0.003x
0.54
0.53
0.43
0.36
0.055
0.055
Men (n = 25)
TrF/Limb
TrF/TotF
TrF/BW
TrF%
y = 0.884
y = 0.792
y = 0.873
y = 0.873
+0.075~
+0.353~
+0.842~
+0.004~
0.36
0.38
0.50
0.46
0.045
0.045
0.059
0.061
0.042
0.043
P<0.001
P<O.Wl
P<O.Ol
P <0.05
NS
NS
P<O.Ol
P <0.05
Men had a significantly higher functional capacity
then women, however, there were no gender differences in total cholesterol or triacylglycerol.
DXA ratios and WHR
As older men and women represent two distinct
populations, comparisons between DXA-derived
ratios and the WHR were stratified by gender. As
shown in Table 3, a moderate relationship exists for
all DXA-derived ratios and the WHR. The strongest
relationship between the WHR and DXA ratio for
D. R. Taaffe et al.
584
Table 4. Correlation between ratios of fat distribution and CVD
risk factors in women and men. Statistical significance: *P <0.05.
tP<0.01. :P<O.005.
Ratio
METs
CHOL
TRIG
Table 5. Whole body and regional composition assessment at base
line and 14 and 24 weeks for rhGH and exercise trial. Values are
means fSEM. Scheffe test (P<0.05): ’baseline versus 14 weeks, bbaseline
versus 24 weeks, (14 weeks versus 24 weeks.
14 weeks
Baseline
Women (n = 39)
WHR
TrF/LimbF
TrF/TotF
TrF/BW
TrF%
Men (n = 25)
WHR
TrF/LimbF
TrF/TotF
TrF/BW
TrF%
-0.24
-0.13
-0.16
- 0.40*
- 0.44t
-0.35
- 0.08
-0.15
- 0.49*
- 0.49*
0.23
0.36*
0.36*
0.27
0.22
0.47:
0.33*
0.36*
0.35*
0.33*
Body weight (kg)
0.34
0.21
0.29
0.36
0.35
0.13
0.25
0.22
0.06
0.03
men was that of TrF/BW (r=0.50, P<O.Ol) and for
women TrF/LimbF ( r = 0.54, P < 0.001).
WHR, D X A ratios and CVD risk factors
Both the WHR- and DXA-derived ratios showed
similar associations with functional capacity, as
determined by MET level, and lipid levels for
women and men (Table 4). For both women and
men, all ratios were negatively correlated with MET
level; however, only TrF/BW and TrFX reached
statistical significance. All ratios were significantly
correlated with triacylglycerol levels in women;
however, only the DXA ratios of TrF/LimbF and
TrF/TotF were significantly related to total cholesterol. For lipid levels in men, none of the ratios
achieved statistical significance.
Clinical application of D X A ratios
Eighteen men completed the 24 week study
(rhGH = 10, PL = 8) which produced substantial increases in muscle strength in both groups. However,
two men in the rhGH group experienced oedema
and were omitted from the body composition analysis. Although body weight did not change in either
group, the rhGH group experienced a significant
increase in LBM and a significant decrease in fat
mass between baseline and 24 weeks (Table 5).
Standard anthropometric assessment using the
WHR and the ratio of subscapular/triceps skinfolds
indicated no change in regional body fat distribution following either 14 or 24 weeks of training
(Table 6). However, both trunk and limb fat mass
determined by DXA declined over the course of the
intervention period in the rhGH group. DXAderived ratios of TrF/LimbF, TrF/TotF, TrF/BW
and TrF% all decreased between baseline and 24
weeks, the difference occurring after rhGH administration for 10 weeks, that is, from week 14 to 24.
DISCUSSION
Results of this study indicate a moderate relation-
24 weeks
rhGH
PL
77.6 f4.6
83.5 f4.3
76.9 f4.5
83.9 f4.4
77.5 f4.4
83.3 f4.5
LBM (kg)’
rhGH
PL
60.5 f2.8
64.5 f2.8
61.1 f3.0
65.2f 2.7
62.5 f3.0b
64.7 f2.3’
Fat mass (kg)’
rhGH
PL
17. I f2.3
19.0 f I .8
15.8f 1.9
18.6 f2.0
I5.0f 1.91.b
18.6 f2.4
Trunk lean mass (g)
rhGH
PL
31028k 1491
32682 f I592
31410+ I430
33044 f 1472
31683f 1604
Trunk fat mass (g)*
PL
8613f 1405
9345 f981
7913f1193
9199f I100
7199f1178b
9129f 1286
Limb lean mass (g)’
rhGH
PL
23694 f I I35
25191 f 1080
23850 f 1250
25528 f I180
Lomb fat mass (g)’
rhGH
PL
7554 f978
8740 f890
6980 f708
8492 f927
32441 f I343
2-f
1253b,c
25500 k 901
6926 f781
8499fII50
*Main effect of time (F=6.21-3.96. P<O.O3).
Table 6. Body fat distribution at baseline and 14 and 24 weeks for
rhGH and exercise trial. Values are meanrf SEM. ScheHe test (P<O.O5):
’baseline versus 24 weeks, b14 versus 24 weeks.
Baseline
14 weeks
24 weeks
0.972 f0.022
0.970f0.016
0.974 f0.028
0.973 fO.018
0.972 f0.029
0.984f0.016
Subscap/Tri sk*
rhGH
PL
I .81 f0.25
2.36 k 0.39
1.78 f0. I4
2.22 f0.33
I .78 f0.16
2.33 f0.35
TrFiLimbFt
rhGH
PL
1.117fO.097
1.079f0.077
1.1 18fO.101
1.085f0.072
1.024f0.105a~b
1.078f0.083
TrF/TotFt
rhGH
PL
0.491 f0.024
0.488 f0.0 I7
0.489 f0.024
0.489 kO.015
0.465 f0.026a,b
0.484 f0.0 I8
TrF/BWt
rhGH
PL
0. I07 f0.012
0.1 10 fO.008
0.100fO.011
0.108 +0.008
0.05llfO.01 I2.b
0. I07 f0.010
TrF%t
rhGH
PL
20.6 & 2.2
21.6f 1.3
19.3f 1.9
21.0f 1.5
17.7 f2.0’,b
21.0f 1.8
WHR
rhGH
PL
*Subscapular/triceps skinfold.
tMain effect of time (F=9.33-5.84,
P<O.OI).
ship between the WHR- and DXA-derived ratios of
fat distribution obtained from a whole body scan.
However, DXA-derived ratios detected changes in
regional fat distribution resulting from a combination of resistance exercise and rhGH intervention,
Regional fat distribution and duaknergy X-ray absorptiometry
whereas the WHR or subscapular/triceps ratios did
not, suggesting DXA-derived ratios are more sensitive than conventional anthropometric methods to
assess and categorize body fat distribution.
Due to the importance of body fat distribution,
specifically central in relation to peripheral fat,
DXA-derived ratios in this study were aimed at
partitioning trunk fat from the remainder of the
body or in reference to the body as a unit. The ratio
TrF/LimbF segments the body into central and
peripheral portions, with LimbF a combination of
upper and lower extremity limb fat. TrF/TotF
examines central fat in relation to body fat mass
while TrF/BW assesses trunk fat in relation to total
body tissue mass. The index TrF%, is a localized
measurement of central fat mass, relating trunk fat
to trunk mass.
The moderate relationships found between DXAderived ratios and the WHR in this study are
similar to that reported by Fuller et al. [l 11 for the
DXA ratio of trunk fat/leg fat and the WHR in
young adults. By using WHR as a criterion, we do
not suggest that WHR is the gold standard for
assessing fat distribution and that alternative methods must agree highly with it. Nevertheless, we
recognize that the WHR is widely used and established in cross-sectional [19, 241, longitudinal [7,
251, and intervention [26, 271 studies and that it is
a robust predictor of disease risk and mortality [l,
2, 4-71.
Although a modest association exists between
DXA-derived ratios and WHR, both methods of
examining fat distribution are similarly related to
CVD risk factors in our elderly population. It is
apparent that a high central deposition of adipose
tissue leads to increased triacylglycerol and to a
lesser extent cholesterol levels in older women. The
adverse lipid profiles increase the risk of cardiovascular morbidity and mortality [28]. However, TrF/
BW and TrF% were the only measures of fat
distribution that correlated significantly to functional capacity, which independently predicts cardiovascular mortality in healthy older individuals
[20, 291. Although functional capacity or maximal
exercise capacity reflects the amount and intensity
of habitual physical activity, it also depends on
genetic factors [30]. Both ratios correlated inversely
with MET levels, indicating a more central distribution of body fat is related to a lower cardiovascular
fitness level.
Unlike the WHR and subscapular/triceps skinfold, the DXA-derived ratios were able to detect
subtle changes in regional fat distribution resulting
from administration of rhGH over a 10 week period. Both the WHR and subscapular/triceps skinfold
ratios were unable to detect a change in fat distribution, even though there was a decrease in trunk and
limb fat mass over the 24 week study period in the
rhGH group. All DXA-derived ratios indicate that
rhGH resulted in a preferential loss of central
compared with peripheral fat. Rosenbaum et al.
585
[31] also report a preferential reduction in abdominal adipose tissue as determined by adipocyte lipid
content subsequent to 3 months administration of
exogenous growth hormone (GH) to GH-deficient
children, which was similarly not detected by measuring body circumferences and skinfold thicknesses.
A decrease in body fat with administration of
rhGH has previously been observed in healthy older
men [32], young adults [33], and GH-deficient
adults [34, 351, reflecting the lipolytic characteristic
of GH. Although a preferential reduction in central
adipose tissue has been demonstrated in rhGH
treated GH-deficient adults [35], the present study
is the first to show a similar preferential loss of
central adipose tissue in healthy older adults.
Other recent studies have also added DXAderived ratios to other imaging techniques and
WHR in evaluating body fat distribution. Svendsen
et al. [26] used the DXA-derived ratio of
abdominal/total fat mass in addition to the WHR in
assessing regional changes in fat distribution in a
study of diet and exercise in postmenopausal
women. Pedersen et al. [36] used the DXA ratio of
trunk fat/leg fat and WHR to examine regional fat
distribution and insulin resistance in obese women.
Williams and colleagues [25] used DXA ratios of
trunk-to-total fat and leg fat-to-total fat, as well as
the WHR, to examine the association of fat distribution and circulating dehydroepiandrosterone sulphate, and suggest that circumference ratios are less
sensitive indexes of fat distribution than DXA
measurements.
Although DXA is viewed as a direct measure of
whole body and regional adiposity, it is subject to
limitations [37]. Unlike other imaging techniques,
its two-dimensional nature prevents differentiation
between intra- and extra-abdominal fat. In addition,
it does not directly assess soft tissue anteroposterior to bone (pixels containing soft tissue and
bone) but estimates it from soft tissue immediately
adjacent to the bone [38]. Nevertheless, it permits
an accurate and precise measure of whole body and
regional tissue composition, having the ability to
detect small changes, with a low radiation exposure
[ l l , 391. In addition, indices of fat distribution
apart from those examined in this study can also be
derived from DXA, depending on the software
version employed. Manually selected regions have
previously been utilized by Ley et al. [17] in
examining sex and menopause-related differences in
fat distribution, but provided no advantage over
regions derived from default settings.
We recognize that major variations in adiposity
and body thickness can influence the DXA measurement of both bone mineral and soft tissue [4&44].
In general, these confounding effects occur only
when adiposity is excessive or body thickness
exceeds 25cm [40, 441. This could potentially lead
to errors when used to follow changes in body
composition consequent to an intervention inducing
significant changes in body weight. In the present
586
D. R. Taaffe et al.
study, BMI for all subjects was below 30 kg/m2, and
there was no change in body weight for either the
rhGH or placebo group during the course of the
exercise trial. Therefore, it is unlikely that this
problem is an issue for the present analysis.
In conclusion, this study suggests that DXAderived ratios from a whole body scan may be more
sensitive
than
conventional
anthropometric
measurements in the assessment of regional fat
distribution changes. These ratios may prove useful
in assessing body fat distribution and changes
occurring as a result of exercise and/or dietary
intervention, especially in populations where
anthropometric assessment is difficult such as in the
obese, aged, and non-ambulatory individuals, and
where subtle changes in tissue composition may be
expected.
ACKN 0W LEDGMENTS
This study was supported by the Research Service
of the Department of Veterans Affairs.
REFERENCES
I. Kissebah AH, Vydelingum N, Murray R, et al. Relation of body fat distribution
to metabolic complications of obesity. J Clin Endocrinol Metab 1982; 54:
254-60.
2. Peiris AN, Hennes MI, Evans Dj, Wilson CR, Lee ME. Kissebah AH.
Relationship of anthropometric measurements of body fat distribution to
metabolic profile in premenopausal women. Acta Med Scand 1988; 723 (Suppl):
179-88.
3. Kaplan NM. The deadly quartet: upper-body obesity, glucose intolerance,
hypertriglyceridemia, and hypertension. Arch Intern Med 1989; 149 1514-20.
4. Chumlea WC. Baumgartner RN. Garry PI. Rhyne RL, Nicholson C, Wayne S.
Fat distribution and blood lipids in a sample of healthy elderly people.
Int J Obesity 1992; 16: 125-33.
5. Lapidus L, Bengtsson C, Larsson B, Pennert K. Rybo E. Sjostrom L.
Distribution of adipose tissue and risk of cardiovascular disease and death: a
12 year follow up of participants in the population study of women in
Gothenburg, Sweden. Br Med J 1984; 289: 1257-61.
6. Lanson 8. Svardwdd K, Welin L. Wilhelmsen L. Bjorntorp P, Tibblin G.
Abdominal adipose tissue distribution, obesity, and risk for cardiovascular
disease and death: I3 year follow-up of participants in the study of men born
1913. Br Med 1 1984, 11)8: 1401-4.
7. Folsom AR. Kaye SA. Sellers TA, et al. Body fat distribution and >year risk of
death in older women. j Am Med Asroc 1993; 269: 483-7.
8. Egger G. The case for using waist to hip ratio measurements in routine
medical checks. Med 1 Aust 1992; 156: 280-5.
9. Weststrate ]A. Deurenberg P. van Tinteren H. Indices of body fat distribution
and adiposity in Dutch children from birth to 18 years of age. Int J Obesity
1989; 13: 465-477.
10. Callaway CW, Chumlea WC, Bouchard C, e t al. Circumferences. In: Kohman
TG, Roche AF, Martorell R, eds. Anthropometric standardization reference
manual. Champaign, IL: Human Kinetics, 1988: 39-54.
I I. Fuller NJ. Laskey MA, Elia M. Assessment of the composition of major body
regions by duaknergy X-ray absorptiometry (DEXA), with special reference
to limb muscle mass. Clin Physiol 1992 II: 253-66.
12. Ashwell M, Cole TJ. Dixon AK. Obesity: new insight into the anthropometric
classification of fat distribution shown by computed tomography. Br
Med J 1985; 2%. 1692-4.
13. Borkan GA. Hults DE. Gerzof SG, Robbins AH, Silbert CK. Age changes in
body composition revealed by computed tomography. J Gerontol 1983; 38:
673-7.
14. van der Kooy K. Leenen R. Seidell IC. Deurenbern P. Droop A, Bakker CIG.
Waist-hip ratio is a poor predictor b f changes in ;isceral fa;. Am J Clin Nitr
1993; 5 7 327-33.
15. Gerard EL, Snow RC, Kennedy DN. e t al. Overall body fat and regional fat
distribution in young women: quantification with MR imaging. Am J Radio1
1991; 157: 99-104,
16. Dawson-Hughes B, Harris S. Regional changes in body composition by time of
year in healthy postmenopausal women. Am J Clin Nutr 1992; 5& 307-13.
17. Ley CJ, Lees B. Stevenson JC. Sex- and menopause-rrxrciated changes in
body-fat distribution. Am J Clin Nutr 1992; 55 950-4,
18. Novak LP. Aging, total body potassium, fat-free mass, and cell mass in males
and female between ages 18 and 85 years. J Gerontol 1972; 2E 43843.
19. Schwartz RS. Shuman WP. Brandbury VL, et PI. Body fat distribution in
healthy young and older men. J Gerontol 1990; 45 M181-5.
20. Kannel WE. Epidemiology of cardiovascular disease in the elderly: an
assessment of risk factors. Cardiovasc Clin 1992; 22: 9-22.
21. Wenger NK. Cardiovascular disease in the elderly. Curr Prob Cardiol 1992;
IT: &%90.
21. Hanson P. Clinical exercise testing. In: Blair SN, Painter P, Pate RR, Smith
LK, Taylor C8, eds. Resource manual for guidelines for exercise testing and
prescription. Philadelphia: Lea and Febiger, 1988: 205-22.
23. Guidelines for exercise testing and prescription. 4th ed. American College of
Sports Medicine. Philadelphia: Lea and Febiger, 1991.
24. Shimokata H, Tobin ID, Muller DC, et al. Studies in the distribution of body
I. Effects of age, sex, and obesity. J Gerontol 1989 44: M66-73.
Iliams DP, Boyden TW, Pamenter RW. Lohman TG, Going SB. Relationship
of body fat percentage and fat distribution with dehydroepiandrorterone
sulfate in premenopausal women. J Clin Endocrinol Metab 1993; 77.W-5.
26. Svendsen OL. Hassager C, Christiansen C. Effect of an energy-restrictive diet,
with or without exercise, on lean tissue mass, resting metabolic rate,
cardiovascular risk factors, and bone in overweight postmenopausal women.
Am J Med 1993; 95 1 3 1 4 .
27. Zamboni M, Armellini F, Turcato E, et al. Effect of weight loss on regional
body fat distribution in premenopausal women. Am J Clin Nutr 1993; 58:
29-34.
28. Simons LA. Triglyceride levels and the risk of coronary heart disease: a view
from Australia. Am J Cardiol 1992; 70: 14H-18H.
29. Sandvik L, Erikssen J, Thaulow E. Eriksren G, Mundal R, Rodahl K. Physical
fitness as a predictor of mortality among healthy, middle-aged Norwegian
men. N Engl j Med 1993; 328: 533-7.
30. Bouchard C. Lesage R, Lortie G. et al. Aerobic performance in brothers,
dizygotic and monozygotic twins. Med Sci Sports Exerc 1986, 18: 639-46.
31. Rosenbaum M. Gertner JM, Leibel R. Effects of systemic growth hormone
(GH) administration on regional adipose tissue distribution and metabolism in
GHdeficient children. j Clin Endocrinol Metab 1989; 68: 1274-81.
32. Rudman D, Deller AG, Nagraj HS. et al. Effects of human growth hormone in
men over 60 years old. N Engl J Med 1990; 3U: 1-6.
33. Crist DM, Peake GT. Egan PA, Waters DL. Body composition response to
exogenous GH during training in highly conditioned athletes. J Appl Physiol
1988; 6 5 579-04.
34. J9rgensen JOL, Pedenen SA, Thuesen L, et al. Beneficial effects of growth
hormone treatment in GHdeficient adults. Lancet 1989 i: 1221-5.
35. Salomon F. Cueno RC, Hesp R. Sonksen PH. The effects of treatment with
recombinant human growth hormone on body composition and metabolism in
adults with growth hormone deficiency. N Engl J Med 1989; 321: 1797403.
36. Pedersen SB, Wrglum JD, Schmitz 0.Bak IF, S9rensen NS, Richelsen B.
Abdomonal obesity is associated with insulin resistance and reduced glycogen
synthase activity in skeletal muscle. Metab Clin Exp 1993; 42: 998-1005.
37. Roubenoff R. Kehayias JJ, Dawson-Hughes B. Heymsfield SB. Use of
duaknergy X-ray absorptiometry in body composition studies: not yet a 'gold
standard'. Am 1 Clin Nutr 1993; 58. 589-91.
38. Svendson OL. Hassager C, Bergmann I, Christiansen C. Measurement of
abdominal and intra-abdominal fat in postmenopausal women by dual energy
X-ray absorptiometry and anthropometry: comparison with computerized
tomography. Int J Obesity 1993; 17: 45-51.
39. Going SB. Massett MP. Hall MC, Bare LA, Root PA, Williams DP. Lohman TG.
Detection of small changes in body composition by dual-energy X-ray
absorptiometry. Am J Clin Nutr 1993; 57 845-50.
40. Wahner HW. Dunn WL, Brown ML. Morin RL, Riggs EL. Comparison of
dual-energy X-ray absorptiometry and dual photon absorptiometry for bone
mineral measurements of the lumbar spine. Mayo Clin Proc 1988; 0: 107544.
41. Hangartner TN, Johnston CC. Influence of fat on bone measurements with
dual-energy absorptiometry. Bone Miner 1990; 9 71-81.
42. Haarbo J, Godredsen A, Hassager C, Christiansen C. Validation of body
Composition by dual energy X-ray absorptiometry (DEXA). Clin Physiol 1991;
II: 33141.
43. laskey MA, Lyttle KD, Flaxman ME, Barber RW. The influence of tissue depth
and composition on the performance of the Lunar duaknergy X-ray
absorptiometer wholebody scanning mode. Eur J Clin Nutr 1992; 46: 39-45.
44. )ebb SA, Goldberg GR. Elia M. DXA measurements of fat and bone mineral
density in relation to depth and adiposity. In: Ellis KJ, Eastman ID. eds.
Human body composition, New York: Plenum Press, 1993: 115-19.