weight-loss diet alone or combined with resistance ... · original article weight-loss diet alone...

ORIGINAL ARTICLE

Weight-loss diet alone or combined with resistancetraining induces different regional visceral fat changesin obese women

F Idoate1, J Ibanez2, EM Gorostiaga2, M Garcıa-Unciti3, C Martınez-Labari2 and M Izquierdo2

1Department of Radiology. Clinic of San Miguel, Navarra, Spain; 2Studies, Research and Sport Medicine Center, Institute ofSport, Government of Navarra, Navarra, Spain and 3Department of Nutrition and Food Sciences, Physiology and Toxicology,University of Navarra, Pamplona, Spain

Background: Quantification of abdominal fat and its regional distribution has become increasingly important in assessing thecardiovascular risk.Objective: To examine the effects of 16 weeks of a hypocaloric diet with a caloric restriction of 500 Kcal per day (WL) or thesame dietary intervention plus resistance training (WLþ RT) on regional variation of abdominal visceral (visceral adipose tissue(VAT)) and subcutaneous (subcutaneous adipose tissue (SAT)) fat loss. Second, to identify the single-image that best representstotal magnetic resonance imaging measurements of total VAT and SAT volume before and after WL or WLþ RT intervention.Design: A total of 34 obese (body mass index: 30–40 kg m�2) women, aged 40–60 years, were randomized to three groups:a control group (C; n¼9), a diet group (WL; n¼12) and a diet plus resistance training group (WLþ RT; n¼13) with the samecaloric restriction as group WL and a 16-week supervised whole-body RT of two sessions per week.Results: WLþ RT programs lead to significant changes in the location of highest mean VAT area from L3-L4 to L2-L3 discal levelfrom pre- to post- intervention, whereas after WL the greatest relative VAT losses were located at L5-S1. Similar decreases in theSAT areas at all discal levels were observed after WL and WLþ RT.Conclusion: Different weight loss regimes may lead to different distribution of VAT. Sites located significantly above (cranial to)L4-L5 (that is, B5–6 cm above L4-L5 or at L2-L3 discal level) provided superior prediction of total abdominal VAT volume,whereas more caudal slices provide better prediction of subcutaneous fat, not only before but also after either WL or WLþ RT.

International Journal of Obesity advance online publication, 7 September 2010; doi:10.1038/ijo.2010.190

Keywords: visceral adipose tissue; subcutaneous adipose tissue; exercise; MRI

Introduction

Bodyweight loss induced by either dietary intervention,

alone or in combination with exercise (that is, aerobic or

resistance training) lead to decreases in total visceral depot.

These changes in abdominal fat and in its regional distribu-

tion have also been accompanied by significant improve-

ments in different cardiovascular risk factors (for example,

insulin sensitivity, baseline insulin levels, total colesterol or

low-density lipoprotein-C).1–13 Magnetic resonance imaging

(MRI) has been extensively used to measure adipose tissue-

system level components in vivo.13–22 Limited by scanning

time and the relatively high cost of image analyses obtained,

several studies have attempted to precisely identify the best

single-image location to best represent total volume of visceral

adipose tissue (VAT) and subcutaneous adipose tissue (SAT).

Previous epidemiologic and intervention studies have com-

monly used the L4-L5 image to examine whole-body MRI

visceral fat measurement.23–25 Other studies, however, have

considered the L2-L3 site and/or images located B5–10cm

above the L4-L5 image to be more appropriate as they have

been very highly correlated with total VAT volume.26–28,25,29–32

However, exercise program (that is, aerobic or resistance

training), hypocaloric diet and/or combined interventions

may lead to differential regional alterations of adipose tissue

metabolism and regional VAT depot loss, possibly by mobili-

zing free fatty acids from VAT at different regions of the

abdomen. Whether or not dietary intervention or a combined

diet and resistance training changes the optimum anatomic

site to prediction total VAT volume is a hypothesis that should

be taken into consideration. Therefore, quantification of the

degree of intra-subject regional variation in intra-abdominalReceived 3 January 2010; revised 4 August 2010; accepted 5 August 2010

Correspondence: Professor M Izquierdo, Studies, Research and Sport Medicine

Center, Government of Navarra, C/Sanguesa 34, 31005 Pamplona (Navarra),

Spain.

E-mail: [email protected]

International Journal of Obesity (2010) 1–14& 2010 Macmillan Publishers Limited All rights reserved 0307-0565/10 $32.00

www.nature.com/ijo

http://dx.doi.org/10.1038/ijo.2010.190

mailto:[email protected]

http://www.nature.com/ijo

VAT and SAT content (from xiphoid to symphysis using

contiguous image data without gaps) need further attention.

In addition, the ability to precisely identify the single-image

that best represents total MRI measurements of VAT and SAT

volume within the abdominal region in response to different

interventions (that is, diet or diet combined with resistance

training) has not been extensively examined.

The purpose of this study was to examine the effects of

either 16 weeks of a hypocaloric diet with a caloric restriction

of 500Kcal per day (WL) or the same dietary intervention

plus resistance training (WLþRT) on regional variation

of abdominal VAT and SAT loss. A second objective was to

compare the ability to precisely identify the single-image

that best represents total MRI measurements of total VAT

and SAT volume within the abdominal region following

either a location relative to L4-L5 method,29 and a discal

level approach before and after a diet or combined diet

and resistance training intervention. We hypothesized

differential effects on abdominal tissue distribution pattern

of VAT and SAT depots depending on the diet and/or resis-

tance training weight-loss intervention performed. Further-

more, after either WL or WLþRT interventions the ability

to predict total VAT and SAT mass using single-slices images

will be modified.

Methods

Subjects

A total of 34 sedentary, non-smoking, obese (body mass

index (BMI): 30–40 Kg m�2) women, aged 40–60 years, were

recruited through an advertisement in a local newspaper.

Before inclusion in the study all candidates were thoroughly

screened using an extensive medical history, resting and

maximal exercise electrocardiogram and blood pressure

measurements. Subjects with cardiovascular, neuromuscular,

arthritic, pulmonary or other debilitating diseases were

excluded from the study. None of the subjects received any

medication. All the subjects were informed in detail about

the possible risks and benefits of the project, and signed a

written consent form before participating in the study. This

project was approved by the ethical committee of the

regional health department. Participants were randomized

to three groups: a control group (C; n¼9); a diet group

(WL; n¼12) with a caloric restriction of 500 Kcal per day;

and a diet plus resistance training group (WLþRT; n¼ 13)

with the same caloric restriction as group WL and a 16-week

supervised whole-body resistance training program of two

sessions per week. Peri-menopausal women were balanced

among groups. The subjects were tested on two different

occasions (weeks 0 and 16) using identical protocols. During

the 16 weeks of the study the subjects maintained their

customary recreational physical activities (for example,

walking). The baseline characteristics of the subjects are

presented in Table 1.

MRI acquisition and image analysis

The MRI of the abdomen was performed with a 1Tesla

magnet (Magnetom Impact Expert, SIEMENS, Erlangen,

Germany) using a body coil. The subjects were examined

in supine position with arms positioned parallel along the

lateral sides of the body. The whole body fat was imaged

using a ‘multi station’ approach at two different table

positions within 5–10 min. We obtained a spoiled T1-

weighted gradient-echo sequence; imaging parameters in-

cluded field of view¼500 mm, matrix¼512�128, acquired

in-plane resolution¼3.90�1.97 mm, reconstructed slice

thickness¼10 mm (acquired: 10 mm), repetition time/echo

time¼127/6 ms and number of slices¼10. Because of the

limitation of the table-top translation, the subjects were

imaged in two half-body volumes. Sagittal, coronal and

transverse localizers of the abdomen from the diaphragm to

the symphysis pubis were obtained to determine the precise

anatomical sites for image acquisition, allowing location of

each image to discal reference. Each half-body volume was

scanned using two stacks, each containing 10 contiguous

10 mm-thick slices. Each stack was acquired in 20 s. and

interleaved slice order was used. All the stacks were acquired

with breath holding during expiration. The total research

time was about 5–10 min.

Because of the limitation of the table-top translation, the

subjects were imaged in two half-body volumes. Sagittal,

coronal and transverse localizers of the abdomen from the

diaphragm to the symphysis pubis were obtained to

determine the precise anatomical sites for image acquisition,

allowing location of each image to discal reference. Each

half-body volume was scanned using two stacks, each

containing 10 contiguous 10 mm thick slices. Each stack

was acquired in 20 sec and interleaved slice order was

used. A field of view of 500 mm was used, and all the stacks

were acquired with breath holding. The total research time

was about 5–10 min.

The segmentation of SAT and VAT was performed by the

same experienced operator. VAT was defined as adipose

tissue, including the intra-abdominal cavity, region enclosed

by the inner aspect of the abdominal wall and the anterior

margin of the vertebral body. SAT was defined as the area

enclosed by the innermost aspect of the abdominal muscle

wall and the skin surface. Total abdominal fat (TAT) was

defined as the summation of VAT and SAT areas.33 The

acquired axial MR images were transferred to an external

personal computer running Windows XP. We used a specially

designed image analysis software (SliceOmatic 4.3, Tomo-

vision, Montreal, QC, Canada) for quantitative analysis of

the images. The model used to segment the various tissues is

fully described and illustrated elsewhere.34,35 We used a

region-based thresholding method (that is, ‘region growing’

function) to segment the fat regions. This method is based

on the property of our acquired MR images in which the

signal intensities of fat pixels at T1 images are higher than

those of muscle and visceral organ pixels. Briefly, a multiple-

step procedure was used to identify tissue area (cm2) for a

Abdominal tissue distribution pattern after diet or combined with resistance trainingF Idoate et al

2

International Journal of Obesity

Tab

le1

Base

line

an

dfo

llow

-up

an

thro

pom

etr

icch

ara

cteri

stic

sp

rean

dp

ost

16

weeks

of

inte

rven

tion

(week

0an

dw

eek

16)

Cgro

up

(n¼

9)

WL

gro

up

(n¼

12

)W

L+RT

gro

up

(n¼

13

)

Age

(yea

rs)

51

.5±

7.2

51

.6±

6.6

47

.7±

6.5

Pre

Post

Pre

Post

Pre

Post

Antr

opom

etry

Weig

ht

(kg

)88.6

8±

12.7

88.1

5±

11.7

487.2

2±

18.0

180.8

1±

16.6

4*

88.0

8±

12.6

380.3

6±

9.1

1**

*

BM

I(k

gm�

2)

35±

3.6

35±

3.3

34.6

±3.4

32.4

±3.7

***

35±

3.1

32.3

±3**

*

Wais

t(c

m)

100±

7.4

99.9

±8

101.1

±6.5

94.5

±9.9

***

99.7

±8.3

93.4

±8.8

***

WH

R0.9

±0.0

50.9

±0.0

30.9

±0.0

30.9

±0.0

2**

0.9

±0.0

30.9

±0.0

2**

Abdom

inal

MRI

volu

me

VA

T(c

m3)

3370.7

8±

1228.4

3329.2

6±

1187.2

3243.7

5±

10

85.3

2557.5

1±

1171.2

***

3211.3

0±

1232.1

2528.4

3±

1039.9

***

SA

T(c

m3)

14

225.2

1±

3347.6

14

021.2

4±

3127.2

13

340.5

7±

3411.1

11

204.7

6±

3326.8

***

15

093.7

8±

2644.5

12

343.6

6±

2318.1

***

TA

T(c

m3)

17

399.8

1±

3735.2

17

180.8

5±

3455.9

16

672.0

7±

3889.1

13

819.7

5±

4078.9

***

18

327.6

4±

3035.5

14

921.6

2±

2821.1

***

Base

line

met

abolic

vari

able

s

Fast

ing

pla

sma

glu

cose

(mg

dl�

1)

98.4

±9.8

100.7

±11.2

98.6

±14.9

Insu

lin(m

cUm

l�1)

16.8

±5.5

16.7

±9.2

15.1

±6.9

HO

MA

ind

ex

(10�

14�mU�

1�

ml�

1)

4.2

±1.4

4.2

±2.4

3.7

±1.9

Base

line

lipopro

tein

pro

file

s

Tri

gly

ceri

des

(mg

. per

100

ml)

140.2

±61.3

140.8

±48

111.2

±27.8

TC

(mg

. per

100

ml)

262.6

±35.3

250.1

±42.4

248.6

±33.5

LDL

(mg

. per

100

ml)

155±

29

147.8

±30.1

143.3

±27.6

HD

L(m

g. p

er

100

ml)

64.6

±9.9

63.7

±14.5

69.3

±8.6

VLD

L(m

g. p

er

100

ml)

28.1

±12.4

28.1

±9.5

22.2

±5.4

TC

/HD

L(r

atio)

4.2

±0.8

4.1

±0.9

3.6

±0.6

Ab

bre

viati

on

s:BM

I,b

od

ym

ass

ind

ex;

HD

L,h

igh

-den

sity

lipop

rote

in;

HO

MA

,h

om

eost

asi

sm

od

elass

esm

en

t;LD

L,lo

w-d

en

sity

lipop

rote

in;

MRI,

mag

neti

cre

son

an

ceim

ag

ing

;SA

T,

sub

cuta

neous

ad

ipose

tiss

ue;

TA

T,

tota

lad

ipose

tiss

ue;

TC

,to

talch

ole

stero

l;V

AT,

visc

era

lad

ipose

tiss

ue;

VLD

L,ve

rylo

w-d

en

sity

lipop

rote

in;

WH

R,

wais

t-to

-hip

-ratio.

Valu

es

are

exp

ress

ed

as

mean

s±s.

d.

Diffe

ren

ces

betw

een

gro

up

saft

er

inte

rven

tion

*Po

0.0

5,

**Po

0.0

1,

***Po

0.0

01

betw

een

valu

es

inw

eeks

0an

d16.


3


given MRI image. In the first step, the observer labelled the

different tissues by assigning them different codes on the basis

of each image’s histogram histogram illustrating separate

peaks for adipose tissue and lean tissue. The determination of

an adequate threshold of the signal intensity was performed

in each image individually to distinguish between adipose

and lean tissues on MR images. Within the ‘Region Growing/

Painting’ control panel, the upper threshold was set to the

maximum possible setting and the lower threshold was set to

a value within the range of 300–650, with the exact value

dependent on the pixel intensity of the SAT and VAT in the

particular series of images, referred to the histogram. A region

of interest describing the VAT area were created. Each

consecutive slice was tagged using the same region growing

procedure. The computer program connected and grouped

automatically the pixels of similar signal intensities at VAT

and SAT areas created by the regions of interest. At the second

step, SAT and VAT tags resulting from the ‘Region Growing/

Painting’ mode were reviewed by an interactive slice-editor

program, which allowed for verification and, where necessary,

correction of the segmented results. The original gray-level

image was superimposed on the segmented (for example,

colour coded) image using a transparency mode to facilitate

the corrections and edited as needed in the ‘Edit’ mode if

image artefacts are identified, and to manually tag adipose

tissue not previously tagged (adipose tissue areas below the

chosen threshold). High-intensity non-fat pixels arising from

bone marrow, liver and fatty intestinal contents within the

VAT were avoided or manually removed. The area (mm2) of

SAT and VAT in each image was computed automatically by

summing the AT pixels and multiplying by the individual

pixel surface. Summation of all the 1-cm slices provided total

fat volume.

The discal level analysis was conducting labelling each

image referred to discal spaces using sagittal scout images.

Image related to L4-L5 analysis was conducted after labelled

each image referred to L4-L4 discal level. Intra-observer

reliability for calculation of total VAT, SAT and TAT volumes

was of 0.99 with a coefficient of variation of 5–8%.

Hypocaloric diet

Each subject in WL and WLþRT groups received a varied

and well-balanced hypocaloric diet (55% of calories as

carbohydrates, 15% as proteins and the rest as fat) of

500 Kcal per day, according to the previous analysis of

individual daily energy expenditure by accelerometry. This

diet was designed to elicit a 0.5 kg weight loss per week. The

control group was asked to maintain body weight. Through-

out the 16-week intervention period body weight was

recorded once every 2 weeks in both WL and WLþRT

groups. Also, every 2 weeks each subject in the intervention

groups participated in a series of 1-hour seminars in which

the dietician taught proper food selection and preparation,

eating behaviour, control of portion sizes and modification

of binge eating and other adverse habits.

Strength testing and training protocol

The strength training program used in the present study was

similar to that reported previously.11 Briefly, the subjects

were asked to report to the training facility twice a week

to perform dynamic resistance exercise for 45–60 min per

session. A minimum of 2 days elapsed between two

consecutive training sessions.

Each training session included two exercises for the leg

extensor muscles (bilateral leg press and bilateral knee

extension exercises), one exercise for the arm extensor

muscle (the bench-press) and four to five exercises for the

main muscle groups of the body. Only resistance machines

(Technogym, Gambettola, Italy) were used throughout the

training period. Resistance in this study was progressively

increased or decreased every week for the 16-week training

period using a repetition maximum approach, so that the

loads that brought about a given relative intensity remained

unchanged from week to week.

During the first 8 weeks of the training period the subjects

trained with loads of 50–70% of the individual 1-RM

(repetition maximum), and during the last 8 weeks of the

training period the loads were 70–80% of the maximum. In

addition, from week 8 to week 16 the subjects performed a

part (20%) of the leg extensor and bench-press sets with

loads ranging from 30 to 50% of the maximum. In all the

individual exercise sessions performed one of the researchers

was present to direct and assist each subject towards

performing the appropriate work rates and loads. For all

subjects average compliance with the diet classes and

exercise sessions was above 95%.

Statistical tests

Standard statistical methods were used for the calculation of

the means, s.d. One-way analysis of variance was used to

determine any differences between the three groups’ initial

measurements. The resistance training and/or diet-related

effects were assessed using a two-way analysis of variance

with repeated measures (groups� time). When a significant

F-value was achieved, Bonferroni post hoc procedures were

performed to locate the pairwise differences between the

means. Selected relative changes were analyzed by one-way

analysis of variance. Analyses of covariance were used to

adjust post-interventional values to compare the data

between the groups. For this purpose, pre-interventional

values were used as covariates so that the effects of

covariance could be observed. Pearson product–moment

correlation was applied to identify the relationship among

total VAT and SAT volumes and individual SAT and VAT

images, areas calculated from each MRI method (that is,

relative to discal level and to L4-L5 MRI methods. In

addition, bivariate correlation was also applied to identify

the relationship between magnitude of changes of total VAT

and SAT volumes and magnitude of changes of single-image

areas relative to discal level and to L4-L5 MRI methods as a

result of WL or WLþRT. To test the similarity of slopes and

intercepts of these relationships, the corresponding Student’s


4


t-test was applied for the model: Yij¼ aiþ biXijþ eij for i ¼1,2

(1¼Baseline; 2¼ POST WL or WLþRT interventions) and

j¼1, y ,n1 being eij i.i.d. random variables following a

distribution N(0.s1). The data were log transformed and

Bland–Altman plot was used to highlight the differences

between methods to access changes in VAT distribution after

WL or WLþRT interventions. Statistical power calculations

for this study ranged from 0.75 to 0.80. The Po0.05 criterion

was used to ascertain statistical significance. SPSS (version

11.0, SPSS, Chicago, IL, USA) were used for descriptive

statistics and analyses.

Results

Baseline characteristics were similar in all groups. After

16 weeks, no significant changes were observed in the

different parameters evaluated in the control group (Table 1).

No significant differences were observed between WL and

WLþRT groups in the magnitude of decrease in body

weight, BMI and waist-to-hip-ratio after the 16 weeks

intervention (Table 1).

Abdominal adipose tissue distribution after WL or WLþRT

At baseline, relative abdominal fat distribution for the whole

group of patients was 18.82% for total VAT and 81.18% for

total SAT volumes. After intervention, no significant differ-

ences were observed in relative abdominal VAT and SAT

fat distribution in WL (18.92 and 81.08%) and WLþRT

(17.23 and 82.77%), respectively.

Before intervention, the single image with the highest

mean SAT area relative to discal level method was located at

L5-S1 discal level and at 3 cm below L4-L5 discal level. As a

result of WL or WLþRT, no significant differences were

observed in the location of the single image with the highest

mean SAT area with any of the MRI analysis methods

performed compared with those reported before interven-

tion. For VAT values the overall pattern was ‘slightly’

reversed from that observed for SAT (Figure 1). Before

intervention, the single image with the highest mean VAT

area was located at L3-L4 discal level and at 4 cm above L4-L5

for the whole group of subjects. As a result of WL, no

significant differences were observed in the location of the

highest single image with the highest mean VAT area within

the abdomen with any of the MRI analysis methods

performed. As a result of WLþRT, the location of the

highest mean VAT area within the abdomen differ (Po0.01)

to that reported before intervention. Thus, the highest

single-image VAT location reached L2-L3 and 8 cm above

L4-L5 discal level, respectively. The general pattern for TAT

value distribution was similar to that observed for SAT. The

single image with the highest mean TAT area within the

abdomen was located at L5-S1 discal level and at 16 cm

below L4-L5. As a result of WL or WLþRT, no significant

differences were observed in the location of the single image

with the highest mean TAT area.

Abdominal fat changes following each MRI analyze method

Image level relative to discal level method. Total TAT, VAT and

SAT were significantly reduced for WL and WLþRT groups

(Figure 2). No significant differences were observed in the

magnitude of decrease between WL and WLþRT groups in

total VAT, SAT and TAT volumes. After intervention, VAT and

SAT areas significantly decreased in most discal levels

analyzed (Po0.05) except for VAT at L2-L3 in WL group,

and for VAT at L5-S1 in WLþRT group. The magnitude of

decrease in VAT at L5-S1 level was significantly greater in WL

(24.8 %) compared with that observed in WLþRT (14.84%)

(Figure 3). The magnitude of decrease in VAT at L2-L3 level

was greater in WLþRT (22.36%) than that observed in WL

group (18.05%). No significant differences were observed in

the magnitude of decrease between WL and WLþRT in the

SAT volumes at all discal levels analyzed.

Image level relative to L4-L5 method. After intervention, VAT

volume decreased significantly (Po0.05) in most images

levels referred to L4-L5 in WL group (except for L4-L5þ8 cm,

L4-L5þ7 cm, L4-L5þ3 cm, L4-L5–4 cm, L4-L5–5 cm, L4-L5–

6 cm and L4-L5 –9 cm image levels) and in WLþRT group

(except for L4-L5 þ8 cm) (Figure 4). No significant differ-

ences were observed in the magnitude of decrease between

WL and WLþRT groups in VAT volume in all images relating

to the L4-L5 method. Following the image relating to the

L4-L5 method, the SAT volume significantly decreased at all

image levels (Po0.05) in both WL and WLþRT groups

(Figure 4). No significant differences were observed in the

magnitude of decrease between WL and WLþRT groups in

the SAT volume in all images analyzed related to the L4-L5

method (Figure 4).

Relationships between total VAT and SAT volumes and singleimage areas relative to discal level and image location relativeto L4-L5 MRI method before and after WL or WLþRTinterventions

Discal level method. The variation in correlations by discal

level location for SAT and VAT are illustrated in Figure 5.

Before intervention, the correlation between total SAT and

VAT volumes and individual discal levels was moderate

(generally r40.81), whereas these were improved as a result

of WL (r40.92 ) or WLþRT (r40.85). Before intervention,

the discal level with the highest correlation with total SAT

volume was located at L3-L4 discal level (r¼0.92). After

16 weeks of intervention the corresponding correlation as a

result of WL was changed to L5-S1 discal level (r¼0.96),

whereas as a result of WLþRT was located at L4-L5 (r¼0.93)

discal level. Before intervention, the discal level with the

highest correlation with total VAT volume was located at

L2-L3 discal level (r¼0.90). As a result WL or WLþRT

intervention, no significant changes were observed in the

location of the discal level area with the highest correlation

with total VAT volume (for example, at L2-L3; r¼0.96 and

r¼0.96, respectively).


5


Location of image relative to L4-L5 method. Before interven-

tion, the correlation between total SAT and VAT volume and

individual image areas relative to L4-L5 was moderate

(generally r40.74), whereas these improved as a result

of WL (r¼0.93) or WLþRT (r¼0.81). In contrast, the

corresponding correlation coefficient for SAT volume

decreased (r¼0.79), compared with previous intervention.

Relationships between magnitude of changes of total VAT andSAT volumes and magnitude of changes of single-image areas,relative to discal level and to L4-L5 MRI methods as a result ofWL or WLþRT

Discal level method. Correlation coefficients between indi-

vidual changes of total abdominal SAT volume and indivi-

dual changes of single-slice changes in relation to the discal

level method were generally high (mean r¼ 0.83) with a

regional variation after WL (from r¼0.73 to r¼ 0.93), but

with greater relative variability as a result of WLþRT (from

r¼0.63 to r¼0.91) (Figure 6). The corresponding correlation

coefficients for changes of total abdominal VAT volume and

individual single image area changes in relation to the discal

level method were lower than those observed for SAT as a

result of WL (mean r¼0.62) and WLþRT (mean r¼0.59).

Regional variations were also observed in both intervention

groups so that the highest correlation coefficient between

individual abdominal VAT area changes at different discal

level slices and corresponding individual global VAT volume

changes was located at L2-L3 as a result of WL (r¼0.79) and

WLþRT (r¼0.76).

Location of image relative to L4-L5 method. The correlation

coefficients between individual changes in total abdominal

SAT volume and the corresponding individual changes of

Figure 1 Left: sagittal scout image; middle: distribution of cross-sectional area (mm2) from xiphoid process (top) to symphisis pubis (bottom); right: cross sectional

images at L2-L3 y L4-L5. The scout and the cross sectional images corresponded to a woman before (a) and after (b) WLþRT intervention. The distribution of

cross-sectional area corresponds to the mean values of WLþ RT group; in green VAT areas, in red SAT areas. We re-examined the sagittal scout images in all the

subjects to determine the anatomical location of the image levels related to L4-L5 in reference to dorsolumbar vertebrae. The L4-L5 þ 7 crossed the L2-L3 discal

space in 54.5% of patients (n¼ 18).


6


Figure 2 Absolute volumes of visceral vat (VAT), subcutaneous fat (SAT) and total abdominal fat (TAT) before and after control, diet or diet + resistance training

interventions. *Significant differences (Po0.05) compared with the corresponding baseline point (before vs after); ysignificant differences (Po0.05) compared with

the corresponding relative magnitude of change of the control group.

Figure 3 Absolute volumes of visceral vat (VAT) and subcutaneous fat (SAT) within the abdominal region following a discal level approach before and after WL

(a) or WLþRT (b) interventions. *Significant differences (Po0.05) compared with the corresponding baseline point (before vs after); ySignificant differences

(Po0.05) compared with the corresponding relative magnitude of change in the control group. wSignificant differences (Po0.05) compared with the corresponding

relative magnitude of change of the WL group.


7


single-slice areas in relation to L4-L5 were generally high as a

result of WL (mean r¼0,84), but lower as a result of WLþRT

(mean r¼ 0,62) (Figure 7). The corresponding correlation

coefficients for individual changes in total abdominal VAT

volume and single-slice area changes relative to L4-L5 were

lower than those observed for SAT as a result of WL

(mean r¼0,69) and WLþRT (mean r¼0,51) (Figure 7).

Further, highlighting the difference between the methods,

the Bland-Altman plots demonstrate that there are a ±4.4

and 5.8% coefficients of variance between log of the

difference between total abdominal VAT and L2-L3 slice

changes and changes in total abdominal VAT volume with a

total error (log) of 8.8 cm3 and 10.6 cm3 as a result of WL and

WLþRT, respectively (Figure 8).

Discussion

The findings of the present study show that either a diet or a

combined diet and resistance 16-week training intervention

lead to similar relative decreases in MRI measurements of

total VAT, SAT and TAT volumes. However one of the key

findings in the present study was that the abdominal tissue

distribution pattern after WL differs to that observed after

WLþRT. Therefore, our results show that 1) the WLþRT

programs lead to significant changes in the location of

highest mean VAT area from L3-L4 to that of L2-L3 discal

level from PRE to POST intervention, whereas the overall

VAT distribution pattern was not altered as a result of WL

and 2) after WL the relative VAT loss at L5-S1 was greater

than that in WLþRT and differed to the greater relative VAT

loss observed at L2-L3 as a result of WLþRT. In contrast, to

the differences in regional variations in VAT areas, the

location of the highest mean SAT (that is, at L5-S1) before

and after intervention, as well as the magnitudes of decrease

in the SAT areas at all discal levels were similar in both WL

and WLþRT. Also this is the first study to demonstrate that

pre-intervention (that is, without any diet or exercise

training effect), the relationship between total VAT and SAT

volumes was moderate (generally r40.81), whereas the

ability to predict VAT and SAT mass using single images

was improved after both WL and WLþRT intervention as

compared with subjects at baseline with overall higher BMI.

Figure 4 Absolute volumes of visceral vat (VAT) and subcutaneous fat (SAT) within the abdominal region relative to L4-L5 MRI method PRE and POST WL (a) or

WLþRT (b) interventions. *Significant differences (Po0.05) compared to the corresponding baseline point (before vs after); ySignificant differences (Po0.05)

compared with the corresponding relative magnitude of change in the control group. aSignificant differences (Po0.05) compared with the corresponding relative

change of VAT.


8


Before and after either WL or WLþRT interventions the

single-slice VAT area with the highest correlation with either

the corresponding individual total abdominal VAT volume

or with the loss of abdominal VAT was located at L2-L3 discal

level (that is, B5–6 cm above L4-L5). Therefore, these

findings suggest that different weight loss regimes may lead

to different distribution of VAT. Sites located significantly

above (cranial to) L4-L5 (that, B5–6 cm above L4-L5 or at

L2-L3 discal level) seem to be more suitable to predict total

visceral fat, whereas more caudal slices provide better

prediction of subcutaneous fat, not only before but also

after either WL or WLþRT.

In line with previous studies,36,37 our data showed that

8 weeks intervention of a hypocaloric diet with a caloric

restriction of 500 Kcal per day or a WL plus resistance

training led to similar total VAT and SAT volume loss.

This could be explained in part by the allometric relation-

ship between changes of VAT and total fat mass during a

weight-loss diet alone or combined with a resistance training

program. Thus, Hallgreen and coworkers38 suggested that

changes of VAT mass are determined primarily by fat-mass

changes, as well as the initial ratio of VAT to fat-mass, and

this is independent from weight-loss intervention. This

relationship may explain the absence of differences in VAT

and SAT volume reduction.

As seen in previous studies,39,40 however, our data also

demonstrated at baseline (that is, before any diet or exercise

intervention effect) regional variation in fat distribution in

the whole group of subjects with the largest depot of

absolute VAT and SAT at L3-L4 and L5-S1 discal level,

respectively. Kanaley et al.40 showed that although the

largest volume of abdominal fat was found in region at

0.65

0.7

0.75

0.8

0.85

0.9

0.95

1

Co

rrel

atio

n (

r) w

ith

tota

l vo

lum

e

SATVATTAT

0.7

0.75

0.8

0.85

0.9

0.95

1

Co

rrel

atio

n (

r) w

ith

tota

l vo

lum

e

SATVATTAT

0.7

0.75

0.8

0.85

0.9

0.95

1

Co

rrel

atio

n (

r) w

ith

tota

l vo

lum

e

Discal level

SATVATTAT

T11-T12 L5-S1L4-L5L3-L4L2-L3L1-L2T12-L1

Discal levelT11-T12 L5-S1L4-L5L3-L4L2-L3L1-L2T12-L1

Discal levelT11-T12 L5-S1L4-L5L3-L4L2-L3L1-L2T12-L1

Figure 5 Correlations between SAT, VAT and TAT total volumes and single image areas relative to discal level analyze method at baseline (a) and PRE and POST WL

(b) or WLþ RT (c) interventions.


9


L2-L3, this critical region had greater SAT, but smaller

absolute VAT, than other regions located superiorly. The

observation that ranking of subjects for VAT is influenced by

measurement site has also been reported in previously

premenopausal women28,30 and Caucasian men.41 However,

a unique finding of the present study was that a weight-loss

diet alone or combined with a resistance training program

induce different regional changes in VAT volume, but not in

SAT volume distribution patterns. Yet, after a combined

weight-loss diet and resistance training intervention the

location of the largest volume of VAT changed from L3-L4 to

L2-L3 discal level, whereas the overall VAT distribution

pattern was not altered as a result of a similar weight-loss diet

alone. In contrast, to the regional variations in the overall

VAT distribution pattern after WLþRT, the location of the

highest depot of absolute SAT remained at the same position

(that is, at L5-S1 discal level) after both a weight-loss diet or

a combined diet and resistance training intervention. It is

also likely that both intervention programs (that is, WL or

WLþRT) induced a similar magnitude of SAT area decrease

at all discal levels examined.

Taking our data as a whole, the present study showed

differences in abdominal fat decreases depending on the

diet and/or resistance training weight-loss intervention

performed. This may suggest that WL alters adipose

visceral tissue from L5-S1 abdominal level, whereas WL plus

resistance training reduces this fat depot in higher scans of

this region (for example, L2-L3), probably related in part to

regional metabolic difference. This was true despite similar

weight and waist circumference loss. Moreover, similar

relative total VAT and SAT loss were obtained after both

intervention programs. In agreement with our findings,

Christiansen et al.,36 demonstrated that 12-week randomized

intervention led to comparable reduction of the VAT depot

after a hypocaloric diet-induced weight loss (for example,

8 weeks of very low energy diet with a caloric restriction of

600 kcal per day, followed by 4 weeks of weight maintenance

diet) with or without aerobic exercise in obese subjects.

Similarly, Janssen et al.12 reported comparable reductions in

VAT at L4-L5 after either a hypocaloric energy-restrictive diet

(for example, to reduce their weight maintenance energy

intake by 1,000 kcal per day with dietary fat intake restricted

Figure 6 Correlations between magnitudes of changes of total VAT, SAT and TAT volumes and corresponding magnitude of changes of single image areas relative

to discal level method as a result of WL (a) or WLþRT (b) interventions.


10


to less than 30%) or the combination of a energy-restrictive

diet with aerobic or resistance training on abdominal fat in

obese women after a 16-week intervention period. Along

these same lines, Redman et al.37 suggest that the changes

in body composition and abdominal fat distribution are

comparable after either 6 months of dietary caloric restric-

tion (that is, 25% reduction in energy intake) only or dietary

restriction in combination with endurance exercise. Kanaley

et al.40 reported that after a 14-week weight loss intervention

with diet and/or exercise the fat loss from the abdominal

region is not uniform and greater relative VAT loss was found

in the lower regions (that is, at L2-L3 and four slices below)

of the abdomen than in the upper regions of the abdomen

(that is, at four and eight slices above L2-L3).

Unfortunately, the differential effects on regional VAT loss

of diet alone, exercise alone and diet plus resistance exercise

were not compared separately in this study. Thus, resistance

training had no additional effects on reduction of the VAT

depot, compared with the major effects of hypocaloric diet

alone. It is also likely that, by using the present resistance

training program, energy expenditure was not substantial

and would likely not have a significant additional impact on

weight loss compared with WL intervention alone. However,

a primary finding of the present study was that the effects of

combined resistance training and a hypocaloric weight-loss

diet per se on VAT does not seem to be uniform, with greater

relative regional VAT loss in upper regions of the abdomen.

Janssen et al.12 observed similar reductions in VAT and SAT fat

depots in healthy men and women with either the combina-

tion of 16-week treatment of a combined diet and aerobic

exercise, diet and resistance exercise or a hypocaloric diet with

dietary fat intake restricted to less than 30% intervention

alone. In contrast to our finding, Giannopoulou et al.10

demonstrated in women with type 2 diabetes that 14-week

intervention of both a hypocaloric monounsaturaded fat diet

and a hypocaloric monounsaturaded fat diet plus aerobic

exercise reduced total and SAT abdominal fat, however,

aerobic exercise is required for a reduction in VAT, because

the hypocaloric monounsaturaded fat diet alone did not

reduce this fat depot. Thus, exercise program (that is, aerobic

vs resistance training), hypocaloric diet and/or combined

interventions may alter adipose tissue metabolism and

regional VAT depot loss, possibly by mobilizing free fatty

acids from VAT at different regions of the abdomen.

In agreement with previous studies,26–28,25,29–32 our results

show that sites located significantly above (cranial to) L4-L5

(that is, L4-L5 þ6 cm discal level or L2-L3 discal level)

provided higher prediction of total abdominal VAT volume,

compared with the association with VAT reported at L4-L5

discal level. A noteworthy finding of the present study was

also that after both the intervention programs the measure-

ment site located between L2-L3 discal level (or at B5–6 cm

above the L4-L5) provided even higher prediction of total

VAT than observed at baseline level (that is, before WL or

WLþRT). Therefore, this may suggest that quantification of

intervention induced (that is, exercise and/or diet) abdo-

minal total VAT changes may sub-estimate local abdominal

fat changes at specific discal sites. Previous epidemiologic and

intervention studies have commonly used the L4-L5 image to

examine whole-body MRI visceral fat measurement.23–25

Several studies, however, have also considered more appro-

priate the L2-L3 site and/or images located B5–10 cm above

the L4-L5 image as they were very highly correlated with

total VAT volume.26,42,27–30,32 An advantage of the present

study was to examine the regional variation in intra-

abdominal VAT and SAT content, as well as comparing the

0.00

0.20

0.40

0.60

0.80

1.00a

b

Co

rrel

atio

n (

r) w

ith

to

tal

% v

aria

tio

n

Image level related to L4-L5

SATVATTAT

0.00

0.20

0.40

0.60

0.80

1.00

14 cm 13 cm 12 cm 11 cm 10 cm 9 cm 8 cm 7 cm 6 cm 5 cm 4 cm 3 cm 2 cm 1 cm L4-L5 -1 cm -2 cm -3 cm -4 cm -5 cm -6 cm -7 cm -8 cm -9 cm -10 cm -11 cm

14 cm 13 cm 12 cm 11 cm 10 cm 9 cm 8 cm 7 cm 6 cm 5 cm 4 cm 3 cm 2 cm 1 cm L4-L5 -1 cm -2 cm -3 cm -4 cm -5 cm -6 cm -7 cm -8 cm -9 cm -10 cm -11 cm

Co

rrel

atio

n (

r) w

ith

to

tal

% v

aria

tio

n

Image level related to L4-L5

SATVAT

TAT

Figure 7 Correlations between magnitudes of changes of total VAT, SAT and TAT volumes and corresponding magnitude of changes of single image areas relative

to L4-L5 MRI method as a result of WL (a) or WLþ RT (b) interventions.


11


ability to precisely identify the single-image that best

represent total MRI measurements of VAT and SAT volume

within the abdominal region following either a location

relative to L4-L5 method29 and one following a discal level

approach. In the present study, therefore, we used conti-

guous image data without gaps from xiphoid to symphysis

allowing more precisely indentify the best single-image

location to best represent total VAT and SAT volume. In

doing so, we evaluated the degree of correspondence

between multiple-image protocols to estimate abdominal

VAT and SAT volumes performed in many previous

studies.29,31,40 In addition, the present study examined

separately the interventions effect of either a WL with a

caloric restriction of 500 Kcal per day or a WL plus resistance

training, which allowed us to compared the differential

effects on regional variation of VAT and SAT distribution, as

well as differences in abdominal region decreases in fat

depending on the diet and/or resistance training weight-loss

intervention performed. A limitation of the present study

was that it included only a special clinical population

(that is, perimenopausal obese women), so that our results

should be applied cautiously in healthy adults of white,

Figure 8 Bland–Altman plots comparing the log total abdominal fat mean changes and the mean bias ± 2 s.d. of total abdominal VAT and corresponding changes

of single image at L2-L3 as a result of WL (a) and WLþRT (b).


12


black or Asian origin. It is also likely that the prediction of

total VAT and SAT mass from single-slices maybe also

compared with a group with similar weight and/or BMI to

those reported after different weight loss regimes examined

(that is, either a diet or a combined diet and resistance 16-wk

training intervention).

In agreement with previous studies,28,23,29 our results

also revealed that in pre-intervention the relationship

between single-SAT areas and total SAT volume were

uniformly high at all image locations with the highest

relationship located at site measurement L3-L4 discal level

(r¼0,92). In these instances, the ability to predict total SAT

volume using single images after both WL or WLþRT

intervention were improved, however, the location of site

measurement changed to L5-S1 after WL and at L4-L5 discal

level as a result of WLþRT. Therefore, at baseline status the

optimum measurement site for prediction of SAT volume

will be slightly different to that reported for VAT volume

prediction, whereas it will change after WL or WLþRT

interventions.

In conclusion, the findings of the present study show that,

different weight-loss regimes may lead to different distribu-

tion of VAT. Thus, after a combined weight-loss diet and

resistance training intervention the location of the largest

volume of VAT changed from L3-L4 to L2-L3 discal level but,

the overall VAT distribution pattern was not altered as a

result of a similar weight-loss diet alone. The location of the

highest depot of absolute SAT remained at the same position

(that is, at L5-S1) after both WL or WLþRT. Sites located at

L2-L3 discal level (that is, B5–6 cm above L4-L5) seem to be

more suitable to predict total visceral fat, whereas more

caudal slices provide better prediction of subcutaneous fat.

In these instances, the ability to predict VAT and SAT

mass using single images was improved after both WL

or WLþRT as compared with subjects at baseline with

overall higher BMI. This should be taken into consideration

to accurately estimate SAT and VAT volume before and after

diet, an exercise program or a combined diet and exercise

program.43,44,45

Conflict of interest

The authors declare no conflict of interest.

References

1 World Health Organization. Obesity: Preventing and Managing theGlobal Epidemic. World Health Org: Geneva, 1998. (WHO/NUT/NCD/98.1).

2 Bouchard C, Bray GA, Hubbard VS. Basic and clinical aspects ofregional fat distribution. Am J Clin Nutr 1990; 52: 946–950 .PubMed PMID: 2239774.

3 Kuk JL, Katzmarzyk PT, Nichaman MZ, Church TS, Blair SN,Ross R. Visceral fat is an independent predictor of all-causemortality in men. Obesity (Silver Spring) 2006; 14: 336–341.

4 Lapidus L, Bengtsson C, Larsson B, Pennert K, Rybo E, Sjostrom L.Distribution of adipose tissue and risk of cardiovascular diseaseand death: a 12-year follow-up of participants in the populationstudy of women in Gothenburg, Sweden. Br Med J (Clin Res Ed)1984; 289: 1257–1261.

5 Poirier P, Despres JP. Waist circumference, visceral obesity, andcardiovascular risk. J Cardiopulm Rehabil 2003; 23: 161–169.

6 Ryan AS. Insulin resistance with aging: effects of diet andexercise. Sports Med 2000; 30: 327–346.

7 Fox CS, Massaro JM, Hoffmann U, Pou KM, Maurovich-Horvat P,Liu CY et al. Abdominal visceral and subcutaneous adiposetissue compartments: association with metabolic riskfactors in the Framingham Heart Study. Circulation 2007; 116:39–48.

8 Piche ME, Lapointe A, Weisnagel SJ, Corneau L, Nadeau A,Bergeron J et al. Regional body fat distribution and metabolicprofile in postmenopausal women. Metabolism 2008; 57:1101–1107.

9 Fujioka S, Matsuzawa Y, Tokunaga K, Kawamoto T, Kobatake T,Keno Y et al. Improvement of glucose and lipid metabolismassociated with selective reduction of intra-abdominal visceral fatin premenopausal women with visceral fat obesity. Int J Obes1991; 15: 853–859.

10 Giannopoulou I, Ploutz-Snyder LL, Carhart R, Weinstock RS,Fernhall B, Goulopoulou S et al. Exercise is required for visceralfat loss in postmenopausal women with type 2 diabetes. J ClinEndocrinol Metab 2005; 90: 1511–1518.

11 Ibanez J, Izquierdo M, Martinez-Labari C, Ortega F, Grijalba A,Forga L et al. Resistance training improves cardiovascularrisk factors in obese women despite a significative decreasein serum adiponectin levels. Obesity (Silver Spring) 2010; 18:535–541.

12 Janssen I, Fortier A, Hudson R, Ross R. Effects of an energy-restrictive diet with or without exercise on abdominal fat,intermuscular fat, and metabolic risk factors in obese women.Diabetes Care 2002; 25: 431–438.

13 Ryan AS, Pratley RE, Goldberg AP, Elahi D. Resistive trainingincreases insulin action in postmenopausal women. J Gerontol ABiol Sci Med Sci 1996; 51: M199–M205.

14 Abate N, Burns D, Peshock RM, Garg A, Grundy SM. Estimation ofadipose tissue mass by magnetic resonance imaging: vali-dation against dissection in human cadavers. J Lipid Res 1994;35: 1490–1496.

15 Brennan DD, Whelan PF, Robinson K, Ghita O, O’Brien JM,Sadleir R et al. Rapid automated measurement of body fatdistribution from whole-body MRI. AJR Am J Roentgenol 2005;185: 418–423.

16 Eustace SJ, Nelson E. Whole body magnetic resonance imaging.BMJ 2004; 328: 1387–1388.

17 Fowler PA, Fuller MF, Glasbey CA, Cameron GG, Foster MA.Validation of the in vivo measurement of adipose tissue bymagnetic resonance imaging of lean and obese pigs. Am J ClinNutr 1992; 56: 7–13.

18 Ludescher B, Machann J, Eschweiler GW, Vanhofen S, Maenz C,Thamer C et al. Correlation of fat distribution in whole body MRIwith generally used anthropometric data. Invest Radiol 2009; 44:712–719.

19 Ross R, Kimberley DS, Martel Y, De Guise J, Avruch L. Adiposetissue distribution measured by magnetic resonance imaging inobese women. Am J Clin Nutr 1993; 57: 470–475.

20 Siegel MJ, Hildebolt CF, Bae KT, Hong C, White NH. Totaland intraabdominal fat distribution in preadolescents andadolescents: measurement with MR imaging. Radiology 2007;242: 846–856.

21 Staten MA, Totty WG, Kohrt WM. Measurement of fat distributionby magnetic resonance imaging. Invest Radiol 1989; 24: 345–349.

22 Thomas EL, Saeed N, Hajnal JV, Brynes A, Goldstone AP, Frost Get al. Magnetic resonance imaging of total body fat. J Appl Physiol1998; 85: 1778–1785.


13


23 Kuk JL, Katzmarzyk PT, Nichaman MZ, Church TS, Blair SN,Ross R. Visceral fat is an independent predictor of all-causemortality in men. Obesity (Silver Spring) 2006; 14: 336–341.

24 Kvist H, Chowdhury B, Grang ard U, Tylen U, Sjostrom L. Totaland visceral adipose-tissue volumes derived from measurementswith computed tomography in adult men and women: predictiveequations. Am J Clin Nutr 1988; 48: 1351–1361.

25 Ross R, LAger L, Morris D, de Guise J, Guardo R. Quantificationof adipose tissue by MRI: relationship with anthropometricvariables. J Appl Physiol 1992; 72: 787–795.

26 Abate N, Garg A, Coleman R, Grundy SM, Peshock RM. Predictionof total subcutaneous abdominal, intraperitoneal, and retro-peritoneal adipose tissue masses in men by a single axialmagnetic resonance imaging slice. Am J Clin Nutr 1997; 65:403–408.

27 Ellis KJ, Grund B, Visnegarwala F, Thackeray L, Miller CG,Chesson CE et al. Visceral and subcutaneous adiposity measure-ments in adults: influence of measurement site. Obesity (SilverSpring) 2007; 15: 1441–1447.

28 Thomas EL, Bell JD. Influence of undersampling on magneticresonance imaging measurements of intra-abdominal adiposetissue. Int J Obes Relat Metab Disord 2003; 27: 211–218.

29 Demerath EW, Shen W, Lee M, Choh AC, Czerwinski SA,Siervogel RM et al. Approximation of total visceral adipose tissuewith a single magnetic resonanceimage. Am J Clin Nutr 2007; 85:362–368.

30 Greenfield JR, Samaras K, Chisholm DJ, Campbell LV. Regionalintra-subject variability in abdominal adiposity limits usefulnessof computed tomography. Obes Res 2002; 10: 260–265.

31 Shen W, Punyanitya M, Wang Z, Gallagher D, St-Onge MP, Albu Jet al. Visceral adipose tissue: relations between single-slice areasand total volume. Am J Clin Nutr 2004; 80: 271–278.

32 Warren M, Schreiner PJ, Terry JG. The relation between visceralfat measurement and torso levelFis one level better thananother? The Atherosclerosis Risk in Communities Study,1990–1992. Am J Epidemiol 2006; 163: 352–358.

33 Shen W, Wang Z, Punyanita M, Lei J, Sinav A, Kral JG et al.Adipose tissue quantification by imaging methods: a proposedclassification. Obes Res 2003; 11: 5–16.

34 Mitsiopoulos N, Baumgartner RN, Heymsfield SB, Lyons W,Gallagher D, Ross R. Cadaver validation of skeletal muscle

measurement by magnetic resonance imaging and computerizedtomography. J Appl Physiol 1998; 85: 115–122.

35 Ross R, Rissanen J, Pedwell H, Clifford J, Shragge P. Influence ofdiet and exercise on skeletal muscle and visceral adipose tissue inmen. J Appl Physiol 1996; 81: 2445–2455.

36 Christiansen T, Paulsen SK, Bruun JM, Overgaard K, Ringgaard S,Pedersen SB et al. Comparable reduction of the visceral adiposetissue depot after a diet-induced weight loss with or withoutaerobic exercise in obese subjects: a 12-week randomizedintervention study. Eur J Endocrinol 2009; 160: 759–767.

37 Redman LM, Heilbronn LK, Martin CK, Alfonso A, Smith SR,Ravussin E, Pennington CALERIE Team. Effect of calorie restric-tion with or without exercise on body composition and fatdistribution. J Clin Endocrinol Metab 2007; 92: 865–872.

38 Hallgreen CE, Hall KD. Allometric relationship betweenchanges of visceral fat and total fat mass. Int J Obes (Lond) 2008;32: 845–852.

39 Han TS, Kelly IE, Walsh K, Greene RM, Lean ME. Relationshipbetween volumes and areas from single transverse scans of intra-abdominal fat measured by magnetic resonance imaging.Int J Obes Relat Metab Disord 1997; 21: 1161–1166.

40 Kanaley JA, Giannopoulou I, Ploutz-Snyder LL. Regional differ-ences in abdominal fat loss. Int J Obes (Lond) 2007; 31: 147–152.

41 Lee S, Janssen I, Ross R. Interindividual variation in abdominalsubcutaneous and visceral adipose tissue: influence of measure-ment site. J Appl Physiol 2004; 97: 948–954.

42 Shen W, Punyanitya M, Chen J, Gallagher D, Albu J, Pi-Sunyer Xet al. Heshka Visceral adipose tissue: relationships between singleslice areas at different locations and obesity-related health risks.Int J Obes (Lond) 2007; 31: 763–769.

43 Liou KH, Chan YL, Chan JC, Chan WB, Kong MO, Poon MY. Thepreferred magnetic resonance imaging planes in quantifyingvisceral adipose tissue and evaluating cardiovascular risk. DiabetesObes Metab 2005; 7: 547–554.

44 Kullberg J, Johansson L, Ahlstrom H, Courivaud F, Koken P,Eggers H et al. Automated assessment of whole-body adiposetissue depots from continuously moving bed MRI: a feasibilitystudy. J Magn Reson Imaging 2009; 30: 185–193.

45 Machann J, Thamer C, Schnoedt B, Haap M, Haring HU, ClaussenCD et al. Standardized assessment of whole body adipose tissuetopography by MRI. J Magn Reson Imaging 2005; 21: 455–462.


14


weight-loss diet alone or combined with resistance ... · original article weight-loss diet alone...

Documents