The effect of the menopausal transition on body composition and cardiometabolic risk factors: a Montreal-Ottawa New Emerging Team group study.

Abstract

Cardiovascular disease is the first cause of mortality in women in North America. The risk of cardiovascular disease increases sharply after middle age in women, especially after menopause. The aim was to investigate changes in body composition and cardiometabolic profile throughout the menopausal transition.
This was a 5-year observational, longitudinal study on the menopausal transition. The study included 102 premenopausal women at baseline (age, 49.9 ± 1.9 y; body mass index, 23.3 ± 2.2 kg/m). Outcome measures include menopause status, body composition by dual-energy x-ray absorptiometry (total fat mass [FM], trunk FM, and total fat-free mass), waist circumference, visceral and abdominal subcutaneous fat, fasting glucose and insulin levels, homeostasis model assessment of insulin resistance, plasma lipid levels (triglycerides, total cholesterol, and high- and low-density lipoprotein cholesterol), and resting blood pressure.
Repeated-measure analyses revealed significant increases for FM, percentage FM, trunk FM, visceral fat, plasma fasting glucose, and high-density lipoprotein cholesterol (0.05 > P < 0.01) and a significant decrease for plasma glucose levels after follow-up. Those who were in perimenopause or postmenopause by year 3 of the study showed a significant increase in visceral fat (P < 0.01) compared with baseline. Despite some significant changes in the metabolic profile among the menopause statuses, the women did not show any cardiometabolic deterioration by the end of the study.
Our results suggest that changes in body composition and fat distribution can occur in nonobese women as they go through the menopausal transition. However, these changes were not accompanied by cardiometabolic deteriorations in the present study.

Full text

Page 1

Menopause: The Journal of The North American Menopause Society
Vol. 19, No. 7, pp. 760/767
DOI: 10.1097/gme.0b013e318240f6f3
* 2012 by The North American Menopause Society
The effect of the menopausal transition on body composition and
cardiometabolic risk factors: a Montreal-Ottawa New Emerging
Team group study
Joseph Abdulnour, MSc,1,2E´ric Doucet, PhD,1,2Martin Brochu, PhD,3,4Jean-Marc Lavoie, PhD,5
Irene Strychar, EdD, RD,6,7,8Re ´mi Rabasa-Lhoret, MD, PhD,6,7,8,9and Denis Prud’homme, MD, MSc1,2
Abstract
Objective: Cardiovascular disease is the first cause of mortality in women in North America. The risk of car-
diovascular disease increases sharply after middle age in women, especially after menopause. The aim was to
investigate changes in body composition and cardiometabolic profile throughout the menopausal transition.
Methods: This was a 5-year observational, longitudinal study on the menopausal transition. The study included
102 premenopausal women at baseline (age, 49.9 T 1.9 y; body mass index, 23.3 T 2.2 kg/m2). Outcome measures
include menopause status, body composition by dual-energy x-ray absorptiometry (total fat mass [FM], trunk FM,
and total fat-free mass), waist circumference, visceral and abdominal subcutaneous fat, fasting glucose and insulin
levels, homeostasis model assessment of insulin resistance, plasma lipid levels (triglycerides, total cholesterol, and
high- and low-density lipoprotein cholesterol), and resting blood pressure.
Results: Repeated-measure analyses revealed significant increases for FM, percentage FM, trunk FM, visceral
fat, plasma fasting glucose, and high-density lipoprotein cholesterol (0.05 9 P G 0.01) and a significant decrease for
plasma glucose levels after follow-up. Those who were in perimenopause or postmenopause by year 3 of the study
showed a significant increase in visceral fat (P G 0.01) compared with baseline. Despite some significant changes in
the metabolic profile among the menopause statuses, the women did not show any cardiometabolic deterioration by
the end of the study.
Conclusions: Our results suggest that changes in body composition and fat distribution can occur in nonobese
women as they go through the menopausal transition. However, these changes were not accompanied by car-
diometabolic deteriorations in the present study.
Key Words: Menopause Y Body composition Y Cardiometabolic risk factors Y Visceral fat.
C
ardiovascular disease (CVD) is the leading cause of
death in women in many developed countries, and this
risk sharply increases after menopause.1Such an
increase in the incidence of CVD has been related to body
composition and cardiometabolic changes that occur through-
out the menopausal transition.2It has been reported that body
mass index (BMI) was found to be significantly higher in peri-
menopausal and postmenopausal women compared with pre-
menopausal women even after controlling for age, diet, physical
activity levels, and smoking habits.2Both fat mass (FM) and
waist circumference (WC) have also been shown to increase
throughout the menopausal transition in a group of premen-
opausal and perimenopausal women after a 6-year follow-up.3
It was also reported that postmenopausal women presented a
lower abdominal subcutaneous fat (AScF)YtoYvisceral fat (VF)
ratio compared with premenopausal women matched for age.4
VF, in turn, is associated with metabolic complications such
as insulin resistance (IR), dyslipidemia, type 2 diabetes, and
metabolic syndrome.5,6Menopause has also been reported to
be associated with a deleterious effect on plasma lipids, lipo-
proteins, and IR.7,8However, glucose homeostasis does not
seem to be affected by menopause per se but more by the
weight gain that occurs during menopause.7,9
Both cross-sectional and longitudinal studies have reported
that menopause isassociated withchanges inbody composition
and metabolic profile in overweight or obese women.3,10<15
Received September 26, 2011; revised and accepted November 8, 2011.
From the1School of Human Kinetics, Faculty of Health Science, Uni-
versity of Ottawa, Ottawa, Ontario, Canada;2Behavioural and Metabolic
Research Unit, University ofOttawa, Ottawa,Ontario,Canada;3Universite ´
de Sherbrooke, Sherbrooke, Quebec, Canada;4Research Centre on Aging,
Social Services and Health Centre, University Institute of Geriatrics of
Sherbrooke, Quebec, Canada;5Department of Kinesiology, Universite ´ de
Montre ´al, Montre ´al,Quebec, Canada;6DepartmentofNutrition,Universite ´
de Montre ´al, Montre ´al, Quebec, Canada;7Centre de Recherche du Centre
Hospitalier de l’Universite ´ de Montre ´al, Montre ´al, Quebec, Canada;
8Montre ´al Diabetes Research Centre, Montre ´al, Quebec, Canada; and
9Institut de Recherche Cliniques de Montre ´al Montre ´al, Quebec, Canada.
Funding/support: This research was supported by the Canadian Institutes
of Health Research (CIHR) (T 0602145.02). E´ric Doucet is a recipient of
a CIHR/Merck-Frosst New Investigator Award, a Canadian Foundation
for Innovation New Opportunities Award, and an Early Research Award
(Ontario). Re ´mi Rabasa-Lhoret and Martin Brochu holds a scholarship
from FRSQ (Fonds de Recherche en Sante ´ du Que ´bec). Re ´mi Rabasa-
Lhoret is the recipient of the J-A DeSe `ve chair in clinical research.
Financial disclosure/conflicts of interest: None reported.
Address correspondence to: Denis Prud’homme, MD, MSc, Faculty of
Health Science, University of Ottawa, 451 Smyth Road, Room 3028,
Ottawa, ON, Canada, K1N 6N5. E-mail: denisp@uottawa.ca
760
Menopause, Vol. 19, No. 7, 2012
Copyright © 2012 The North American Menopause Society. Unauthorized reproduction of this article is prohibited.

Page 2

However, cross-sectional studies2,4,16do not allow us to con-
clude that a cause-and-effect relationship exists or to detect
changes observed in the population’s characteristics compared
with longitudinal studies. Furthermore, none of the longitudinal
studies3,10<12,15published to date provide direct measures of
bothbody composition and fat distribution change ina group of
healthy nonobese (BMI, G30 kg/m2) women undergoing the
menopausal transition.
The main objective of this study therefore was to examine
changes in body composition and cardiometabolic risk factors
in nonobese women going through the menopausal transition.
Our first hypothesis was that women who will have become
postmenopausal at the end of the study will show increases in
total FM and VF. Our second hypothesis was that women who
will display increased VF after the follow-up will also present
a greater deterioration in their metabolic profile.
METHODS
Participants
The study included 102 healthy premenopausal women aged
between 47 and 55 years participating in the Montreal-Ottawa
New Emerging Team group study, which was a 5-year longi-
tudinal study (2004 to 2009) on the effects of the menopausal
transition on body composition and cardiometabolic risk factors.
This age range was selected, so the women wouldmost probably
become postmenopausal by the end of the study. Participants
were recruited using community advertising and referrals from
the Ob/Gyn clinics. Figure 1 presents the recruitment diagram
and sample size of the premenopausal women at baseline.
Premenopausal women were included if they met the fol-
lowing criteria: (1) premenopause status (two menstruations
in the last 3 months, no increase in cycle irregularity in the
12 months before testing, and a plasma follicular-stimulating
hormone level G30 IU/L as a mean of verification), (2) aged
between 47 and 55 years, (3) no surgically induced meno-
pause, (4) nonsmoker, (5) BMI between 20 and 29 kg/m2, and
(6) reported weight stability (T 2 kg) for 6 months or more
before enrollment in the study. Exclusion criteria were (1)
pregnancy or having plans to become pregnant, (2) medical
problems that could have interfered with outcome variables
including cardiovascular and/or metabolic diseases, (3) taking
oral contraceptives or hormone therapy, (4) high risk for hys-
terectomy, and (5) history of drug and/or alcohol abuse. This
study received approval from the University of Ottawa and
the Montfort Hospital ethics committees, and written consent
was obtained from each participant.
Design
This 5-year Montreal-Ottawa New Emerging Team study
was observational and consisted of several visits to the labo-
ratory (year 1: two screening visits [3-4 h each visit], and
3 half-days [3-6 h each visit]). Participants were invited to
the laboratory for the following tests and measures: anthro-
pometric and body composition (dual-energy x-ray absorpti-
ometry [DXA]) measurements; stress test for cardiorespiratory
fitness; CT scan for body fat distribution; and fasting blood
samples for measurement of fasting lipid, insulin, and glucose
levels. These measurements were performed annually with the
exception of (1) peak oxygen consumption (V˙O2max), which
was measured in years 1, 3, and 5 and (2) CT, which was not
measured in all participants in year 4 and was not performed in
year 5. The latter change regarding the study design is caused
by the theoretical concern of cancer risk associated with the
amount of radiation received from the CT scan, which was
considered unethical in a research setting for healthy nonobese
women.17As a means of standardization, all measurements
were performed in the early follicular phase (within 8 days) as
long as women still had a regular cycle. Participants were
contacted periodically by telephone and postcards throughout
the year to establish significant changes concerning their men-
strual cycle.
Status
Menopause status was determined yearly by self-reported
questionnaire aboutmenstrual bleeding andits regularity.Follicle-
stimulating hormone (FSH) levels were measured annually
during the early follicular phase to verify menopause status.
Women were classified as premenopausal if they reported no
change in menstrual cycle frequency and perimenopausal if they
reported changes in menstrual frequency and/or amenorrhea for
3 to 11 months. Finally, women were classified as postmeno-
pausal based on their final menstrual period (FMP) and con-
firmed by 12 months of amenorrhea.18
Anthropometric assessment
Body weight and height were measured using a BWB-
800AS digital scale and a Tanita HR-100 height rod, re-
spectively (Tanita Corporation of America Inc., Arlington
Heights, IL). Afterward, BMI was calculated (body weight in
FIG. 1. Recruitment process and sample size of premenopausal women
at baseline.
Menopause, Vol. 19, No. 7, 2012 761
MENOPAUSAL TRANSITION AND RISK FACTORS
Copyright © 2012 The North American Menopause Society. Unauthorized reproduction of this article is prohibited.

Page 3

kilograms/height in square meters). Waist circumference (mean
of two measures) was determined using a Gulick tape at the
middle distance between the lowest rib and the iliac crest
measured by four experienced research assistants who were
either certified exercise physiologists or registered nurses and
followed the guidelines according to the Canadian Society for
Exercise Physiology.19
FM, percentage FM, trunk FM, and fat-free mass (FFM)
were measured using DXA (GE-LUNAR Prodigy module;
GE Medical Systems, Madison, WI) as previously described.20
The coefficient of variation and the correlation for percentage
FM measured in 12 healthy participants tested in our labo-
ratory were 1.8% and r = 0.99, respectively.
VF and AScF were measured using CT (GE High Speed
Advantage CT Scanner; General Electric Medical Systems,
Milwaukee, WI). Participants had to lie down on an exami-
nation table while measurements of the abdomen at the L4
through L5 vertebrate level were performed. VF was quanti-
fied by delineating the intra-abdominal cavity at the internal-
most aspect of the abdominal and oblique muscle walls. The
AScF area was determined by highlighting the adipose tissue
located between the skin and the external-most aspect of the
abdominal muscle wall.21Interobserver coefficient of varia-
tion for CT scan is 2.8%. The participants wore a light hospital
gown without shoes during the procedure.
Cardiorespiratory fitness (V˙O2max)
A progressive exercise stress test was performed to measure
the participants’ V˙O2max on a treadmill. The participants were
asked to refrain from any vigorous exercise for 24 hours and
consumption of alcoholic beverages for 6 hours before the test.
They were also asked to abstain from eating and drink-
ing coffee for 2 hours before the test. The progressive test
consisted of 3-minute stages on a treadmill with an increasing
workload to the point of exhaustion. After a brief warm-up,
participants performed the test protocol. The test was ter-
minated when at least two of the following criteria were
achieved22: (1) predicted maximal heart rate was reached,
(2) respiratory quotient was greater than 1.1, (3) oxygen con-
sumption remained stable or decreased with an increase in
workload, or (4) rate of Borg-type scale reached 19 or higher.
Peak oxygen consumption was considered as the highest
V˙O2max reached during the test.
Heart rate, blood pressure, and the Borg scale23were taken
at rest and at the end of each stage during the test. Breath-by-
breath samples of expired air were collected through a
mouthpiece throughout the test, and measurements of oxygen
consumption, carbon dioxide consumption, and respiratory
exchange ratio were made automatically using a Vmax 229
series metabolic cart (SensorMedics Corporation, Yorba
Linda, CA). The indirect calorimetry unit was calibrated
according to the manufacturer’s specifications before every
test to further ensure the reliability of the data collected.
Resting blood pressure
Supine resting blood pressure was measured manually in
the left arm after the participants rested quietly for 5 minutes
using a standard stethoscope and a mercury sphygmoman-
ometer. An appropriate cuff size was selected for each par-
ticipant based on arm circumference. Assessment of resting
blood pressure was standardized according to the American
College of Sport Medicine.22
Physical activity energy expenditure
An accelerometer (Actical; Mini Mitter Co., Inc., Bend,
OR) was used to estimate mean physical activity energy
expenditure (PAEE). During 7 days, the participants wore the
accelerometer upon waking up and took it off just before
going to bed. Such duration was chosen because it is estimated
to result in 90% reliability for the measurement of physical
activity in both men and women.24The accelerometer was
worn on the right hip (anterior to the iliac crest), secured with
an elastic belt with the arrow pointing up. That placement,
when evaluated along with lower leg or foot, upper leg, head
and trunk, lower arm or hand, and upper arm placements, was
the best predictor of energy expenditure (r = 0.92-0.97).25The
accelerometers used in this study were validated previously
with the use of doubly labeled water measurements.26How-
ever, even if this tool has been shown to be a good predictor
of energy expenditure, we elected to use the time spent in
physical activity of varying intensities because, in comparison
with energy expenditure, it does not vary across body mass.
Blood sampling
Fasting samples were taken after a 12-hour overnight fast.
Plasma insulin concentrations were determined by radioimmu-
noassay using125I-labeled human insulin and a human insu-
lin antiserum (Millipore, St. Charles, MO). Plasma glucose
levels were determined using spectrophotometric analysis after
conversion of glucose to glucose-6-phosphate by hexokinase
(Sigma-Aldrich Canada Ltd., Oakville, Ontario, Canada; Fisher
Scientific Limited, Nepean, Ontario, Canada). IR was esti-
mated using homeostasis model assessment (HOMA) using the
following equation: HOMA-IR = (fasting glucose ? fasting
insulin)/22.5.
Total cholesterol (TC), high-density lipoprotein cholesterol
(HDL-C), and triglyceride (TG) levels were analyzed using
the Vitros 950 immunoassay analyzer (Ortho Clinical Diag-
nostics; Johnson & Johnson Company, Markham, Ontario,
Canada) at a wavelength of 540 nm. TC, HDL-C, and TG
were used in the Friedewald formula to calculate low-density
lipoprotein cholesterol (LDL-C) concentrations.27Finally, FSH
was measured on an automated immunoassay analyzer, the
Beckman Coulter DxI Unicell 800 (Beckman Coulter, Brea,
CA). All measures of blood profile presented a coefficient of
variation of less than 15%.
Statistical analysis
SPSS 17.0 for windows (SPSS Inc., Chicago, IL) was used
to perform statistical analyses. Variables were first checked
for normality, and only TG was log-transformed. Repeated-
measures analyses of variance (ANOVA) were used for the
determination of the main effects on various variables of in-
terest, having time (years 1, 2, 3, 4, and 5) as a within-subject
762
Menopause, Vol. 19, No. 7, 2012
* 2012 The North American Menopause Society
ABDULNOUR ET AL
Copyright © 2012 The North American Menopause Society. Unauthorized reproduction of this article is prohibited.

Page 4

factor and menopause status (premenopause, perimenopause,
and postmenopause) as a between-subjects factor, adjusted for
PAEE. A post hoc test with Games-Howel adjustment was
done if there was an effect of status. An ANOVA was used to
compare cardiometabolic risk profile between participants with
and without increase in VF. Analyses of covariance, adjusted
for age, were performed to compare variables of interest
between menopause status (premenopause, perimenopause,
and postmenopause). A post hoc test with Bonferroni adjust-
ment was used for multiple comparisons. Finally, one-way
ANOVAs with post hoc analysis using Bonferroni adjustment
were conducted to compare between years before and after
FMP in women who became postmenopausal by year 5. Year
0 is considered the year within FMP (menopause onset). Re-
sults are expressed as the mean T SD. P e 0.05 was considered
significant.
RESULTS
Women’s characteristics based on time and menopause
status are presented in Table 1. By the end of year 5, 4% (n = 4)
were still premenopausal (FSH, 26.5 T 24.7 IU/L), 29%
(n = 26) were perimenopausal (FSH,42.8 T 34.5IU/L),and67%
(n = 61) had become postmenopausal (FSH, 83.3 T 28.8 IU/L).
According to the repeated-measures analysis of variance
(Table 1), a significant main effect of time was observed for
total FM, percentage FM, trunk FM, and VF (0.05 9 P G 0.01),
showing an overall increase in time. Furthermore, women who
were perimenopausal or postmenopausal by year 3 showed a
significant increase in VF (P G 0.01) compared with their base-
line value (Fig. 2).
A significant effect of time was also observed for plasma
HDL-C (P G 0.05) and glucose levels (P G 0.01); where HDL-
C increased and glucose decreased in time (Table 2). Finally,
no significant effect of time, status or time ? status interaction
was observed for other variables of interest. After adjustment
for PAEE, percentage FM, VF, and total abdominal fat
remained significant, with the addition of FFM and WC.
However, FM, trunk FM, HDL-C, and fasting glucose no
longer presented a significant effect of time (data not shown).
To further analyze the effect of the menopausal transi-
tion on body composition and cardiometabolic risk factors, the
database was transformed into cases. For premenopause sta-
tus, only year 1 values were selected; for perimenopause status,
values were selected from the last year in which the participant
was in that status; and for postmenopause status, fifth-year
values were selected, and we performed an analysis of cova-
riance adjusted for age. As a result, postmenopausal women
showed significantly higher plasma TC compared with pre-
menopausal and perimenopausal women (P G 0.01). In addi-
tion, postmenopausal women presented significantly higher
plasma HDL-C than did premenopausal women (P G 0.05).
Finally, both perimenopausal and postmenopausal women pre-
sent a significantly lower plasma fasting glucose compared
with premenopausal women (P G 0.01; Fig. 3). Similar results
were observed when we performed the same analyses using
only values of women who became postmenopausal by the
end of the study (data not shown).
Further analysis revealed that women who increased their
VF (n = 63; 14.9 T 12.6 cm2) during the first 3-year period did
TABLE 1. Body composition characteristics of the population by time point and menopause status at year 5
PremenopausePerimenopause PostmenopauseRepeated-measures ANOVA P
BaselineYear 5 BaselineYear 5Baseline Year 5 Time (y)Status Interaction
n
Age, y
BW, kg
BMI, kg/m2
% FM
FM, kg
Trunk FM, kg
FFM, kg
WC, cm
n
VF, cm2
AScF, cm2
Total AF, cm2
Values are mean T SD.
The areas of abdominal fat are between baseline and year 3.
ANOVA, analysis of variance; AF, abdominal fat; AScF, abdominal subcutaneous fat; BW, body weight; BMI, body mass index; FM, fat mass; FFM, fat-free
mass; VF, visceral fat; WC, waist circumference; NS, not significant.
44 26 266161
48.3 T 0.5
56.4 T 4.5
21.4 T 0.7
22.9 T 5.6
12.7 T 1.9
5.94 T 0.7
40.3 T 3.4
74.3 T 2.3
18
53.2 T 28.0
208.8 T 77.1
262.1 T 99.7
52.3 T 0.5
55.7 T 3.8
21.0 T 0.8
26.4 T 5.6
14.7 T 3.0
6.71 T 1.8
38.6 T 4.7
75.0 T 4.5
18
58.8 T 27.1
225.9 T 99.5
284.7 T 122.4
49.1 T 1.5
62.1 T 6.1
23.3 T 2.0
31.4 T 6.0
19.5 T 4.7
9.65 T 2.7
39.7 T 4.5
78.9 T 6.1
48
45.7 T 23.3
225.0 T 71.2
270.7 T 83.9
53.2 T 1.4
61.7 T 6.6
23.3 T 2.2
32.6 T 7.4
20.2 T 6.0
9.45 T 3.7
38.7 T 4.4
76.8 T 6.7
48
56.6 T 30.4
218.0 T 73.1
274.6 T 94.8
50.3 T 2.0
60.4 T 6.4
23.3 T 2.3
31.3 T 6.7
19.0 T 5.3
9.07 T 3.1
38.2 T 4.0
77.7 T 7.1
13
49.1 T 23.0
233.7 T 62.7
282.8 T 78.9
54.4 T 2.0
61.0 T 7.3
23.5 T 2.9
33.7 T 7.5
20.8 T 6.5
9.40 T 3.8
37.5 T 3.5
78.1 T 7.0
13
63.3 T 28.8
231.6 T 63.4
294.9 T 83.6
NS
NS
G0.01
G0.01
G0.05
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
G0.01
NS
G0.05
NS
NS
NS
NS
NS
NS
FIG. 2. Abdominal adipose tissue accumulations by time point and
menopause status at year 3. The columns represent the total abdominal
fat. VF is delimited by a dashed box, and AScF is represented by the
value below the dashed box. Values are mean T SD. **Significant dif-
ference for VF. VF, visceral fat; AScF, abdominal subcutaneous fat.
Menopause, Vol. 19, No. 7, 2012 763
MENOPAUSAL TRANSITION AND RISK FACTORS
Copyright © 2012 The North American Menopause Society. Unauthorized reproduction of this article is prohibited.

Page 5

not show a significant increase in their cardiometabolic risk
profile compared with women who maintained or decreased
their VF (n = 16; j8.6 T 8.6 cm2) according to the ANOVA.
In addition, no significant correlations were observed between
changes in VF and changes in cardiometabolic risk factors
(data not shown).
One-way ANOVAs were performed to investigate the dif-
ferences between years relative to FMP in women who have
become postmenopausal by the end of the study. Year 0 is the
year within the FMP, or menopause onset; year 1 is considered
as 1 year after FMP, year j1 is considered as 1 year before
FMP, and so on. As a result, we did not observe any sig-
nificant difference between year 0 (menopause onset) and
any other year for measures of body composition (Table 3).
However, TC was significantly lower at years j3 and j2
(P G 0.01). HDL-C and LDL-C were significantly lower, and
TABLE 2. Metabolic and physiologic characteristics of the population by time point and menopause status at year 5
PremenopausePerimenopause Postmenopause Repeated-measures ANOVA P
Baseline Year 5BaselineYear 5 Baseline Year 5Time (y) Status Interaction
n
FSH, IU/L
TG, mmol/L
TC, mmol/L
HDL-C, mmol/L
LDL-C, mmol/L
TC/HDL-C ratio
Fasting glucose, mmol/L
Fasting insulin, KU/mL
HOMA-IR
n
V˙O2max, mL/kg per min
Resting blood pressure
Systolic, mm Hg
Diastolic, mm Hg
Values are mean T SD.
TG values were log-converted for statistical analysis and reconverted for presentation in table format.
ANOVA, analysis of variance; V˙O2max, peak oxygen consumption; HOMA-IR, homeostasis model assessment of insulin resistance; FSH, follicle-stimulating
hormone; TG, triglycerides; TC, total cholesterol; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; NS, not significant.
222525 5757
7.5 T 3.5
0.78 T 0.08
3.95 T 0.35
1.35 T 0.14
2.24 T 0.52
2.95 T 0.57
4.90 T 0.14
8.68 T 3.07
1.90 T 0.72
4
36.9 T 4.7
26.5 T24.7
0.77 T 0.25
3.91 T 0.32
1.42 T 0.16
2.14 T 0.27
2.77 T 0.09
4.90 T 0.00
8.45 T 4.24
1.84 T 0.92
4
33.3 T 5.6
9.1 T 8.0
0.93 T 0.40
4.36 T 0.76
1.52 T 0.37
2.42 T 0.68
3.00 T 0.77
4.83 T 0.36
12.5 T 3.02
2.70 T 0.71
20
33.7 T 7.0
42.8 T 34.5
0.96 T 0.42
4.67 T 0.74
1.66 T 0.38
2.57 T 0.60
2.91 T 0.65
4.60 T 0.30
10.7 T 2.10
2.18 T 0.44
20
33.1 T 6.5
13.1 T 13.0
0.81 T 0.24
4.49 T 0.69
1.62 T 0.36
2.49 T 0.60
2.89 T 0.73
4.79 T 0.39
11.8 T 6.22
2.57 T 1.47
50
33.9 T 6.2
83.3 T 28.8
0.85 T 0.31
5.07 T 0.80
1.83 T 0.37
2.83 T 0.66
2.87 T 0.70
4.57 T 0.44
11.3 T 4.05
2.37 T 0.92
50
32.8 T 6.5
G0.01
NS
NS
G0.05
NS
NS
G0.01
NS
NS
G0.001
NS
NS
NS
NS
NS
NS
NS
NS
G0.001
NS
NS
NS
NS
NS
NS
NS
NS
NSNSNS
114.0 T 2.8
74.0 T 5.7
117.0 T 9.9
72.0 T 2.8
120.2 T 11.4
76.3 T 7.5
121.1 T 10.5
76.6 T 7.4
115.7 T 12.3
72.8 T 7.5
119.3 T 12.1
73.8 T 7.2
NS
NS
NS
NS
NS
NS
FIG. 3. The effect of menopause status on metabolic profile in women. For premenopause status, only year 1 values were selected; for perimenopause
status, values were selected from the last year in which the participant was in that status; and for postmenopause status, the fifth-year values were
selected. Data are adjusted for age. *Significantly different from premenopause (P G 0.05); **Significantly different from premenopause (P G 0.01);
†Significantly different from perimenopause (P G 0.05). TC, total cholesterol; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density
lipoprotein cholesterol; HOMA-IR; homeostasis model assessment of insulin resistance.
764
Menopause, Vol. 19, No. 7, 2012
* 2012 The North American Menopause Society
ABDULNOUR ET AL
Copyright © 2012 The North American Menopause Society. Unauthorized reproduction of this article is prohibited.

Page 6

glucose was significantly higher at year j3 compared with
year 0 (P G 0.05; Table 4).
We did not observe any significant effect of time, meno-
pause status, and overall change for nonYHDL-C.
DISCUSSION
The major finding and perhaps the most novel contri-
bution of the present longitudinal study is that changes in
body composition and fat distribution can occur in nonobese
premenopausal women as they go through the menopausal
transition, but these are not necessarily accompanied by car-
diometabolic risk factor deteriorations. Such a finding sug-
gested that in healthy nonobese premenopausal women, the
menopausal transition would not necessarily have a negative
impact on the cardiometabolic risk factors.
Our results confirmed our primary hypothesis that women
going through the menopausal transition would show an in-
crease in VF even after adjustment for PAEE. These results
extend the findings from previous studies demonstrating that
postmenopausal women display higher body fat and abdomi-
nal fat compared with premenopausal women.3,12,28<31How-
ever, to our knowledge, none of the longitudinal studies to
date have investigated the effect of the menopausal transition
on nonobese women using gold standard techniques to mea-
sure body composition and body fat distribution. In fact, most
of the studies have either been cross-sectional or longitudi-
nal and used weight and/or BMI as a primary endpoint. How-
ever, one study conducted by Lovejoy et al.12reported similar
results to ours in 129 premenopausal women. They observed
significant increases in total abdominal fat and VF after a
4-year follow-up. Moreover, 40% of their women became
postmenopausal by the end of the study. However, com-
pared with the latter study cited above, our participants were
thinner (BMI, 23.3 T 2.2 kg/m2vs 27.1 T 0.6 kg/m2).
The body composition and fat distribution changes
observed during the menopausal transition can be explained,
in part, by the decline of estrogen levels that is associated with
menopause.12The direct effect of estrogen is through the
inhibition of adipose depot by decreasing lipogenesis and
lipoprotein lipase activity.32It could also relate to physio-
logic and/or behavioral changes in a woman’s life because of
changing ovarian hormone status.33For example, factors such
as decreased resting metabolic rate and/or alterations in energy
intake and appetite may play causative roles in total and central
fat gain during menopause.33,34Age can be another factor that
explains the changes in body composition and fat distribution
in women. Studies have reported that weight gain and increase
in BMI and WC are caused by the effect of aging and not by
the menopausal transition.35<37We did not observe such results
for weight, BMI, and WC in our group of women, probably
TABLE 3. Body composition changes since final menstrual period
Years since final menstrual period
j4
j3
j2
j1012
n
BW, kg
BMI, kg/m2
% FM
FM, kg
FFM, kg
WC, cm
Values are mean T SD.
None of the values were significantly different from 0.
BW, body weight; BMI, body mass index; FM, fat mass; FFM, fat-free mass; WC, waist circumference.
20 4460 59593816
58.6 T 6.7
23.0 T 2.3
27.9 T 6.9
16.5 T 5.6
38.3 T 4.5
76.3 T 8.0
59.8 T 7.0
23.0 T 2.5
30.3 T 6.9
18.3 T 5.6
38.7 T 3.7
76.8 T 6.6
60.4 T 6.7
23.3 T 2.3
31.4 T 6.6
19.1 T 5.4
38.5 T 3.8
77.8 T 7.3
60.6 T 7.2
23.4 T 2.6
31.9 T 7.5
19.5 T 6.1
38.2 T 3.9
77.7 T 7.2
60.6 T 7.0
23.4 T 2.7
32.7 T 7.3
19.9 T 6.1
37.8 T 3.6
78.1 T 6.8
61.8 T 7.2
23.6 T 2.9
35.2 T 7.4
21.9 T 6.3
37.1 T 3.7
79.4 T 7.2
61.5 T 6.4
23.6 T 2.9
35.6 T 6.2
21.9 T 5.6
36.8 T 2.9
78.4 T 8.3
TABLE 4. Fasting metabolic profile changes since the final menstrual period
Years since final menstrual period
j4
j3
j2
j1012
n
FSH, IU/L
TG, mmol/L
TC, mmol/L
HDL-C, mmol/L
LDL-C, mmol/L
TC/HDL-C ratio
Fasting glucose, mmol/L
Fasting insulin, KU/mL
HOMA-IR
Values are mean T SD.
TG was log-converted for statistical analysis and reconverted for presentation in table format.
FSH, follicle-stimulating hormone; HDL-C, high-density lipoprotein cholesterol; HOMA-IR, homeostasis model assessment of insulin resistance; LDL-C,
low-density lipoprotein cholesterol; TG, triglycerides; TC, total cholesterol.
aSignificantly different from 0 (P G 0.01).
bSignificantly different from 0 (P G 0.05).
20 4359 5857 3815
12.9 T 14.4a
0.77 T 0.25
4.45 T 0.54
1.66 T 0.33
2.44 T 0.46
2.76 T 0.51
4.80 T 0.31
9.69 T 2.50
2.06 T 0.56
13.6 T 8.2a
0.79 T 0.26
4.34 T 0.63a
1.58 T 0.34b
2.39 T 0.61b
2.87 T 0.73
4.79 T 0.37b
11.7 T 7.07
2.50 T 1.66
24.9 T 20.9a
0.78 T 0.27
4.52 T 0.66a
1.67 T 0.34
2.48 T 0.54
2.80 T 0.64
4.66 T 0.43
11.0 T 4.31
2.30 T 0.98
48.7 T 32.6a
0.80 T 0.45
4.74 T 0.66
1.76 T 0.33
2.62 T 0.57
2.79 T 0.68
4.62 T 0.48
11.0 T 5.41
2.28 T 1.12
85.2 T 31.7
0.82 T 0.38
5.04 T 0.82
1.82 T 0.34
2.85 T 0.72
2.86 T 0.71
4.51 T 0.36
11.1 T 3.97
2.29 T 0.84
89.1 T 27.4
0.85 T 0.48
5.04 T 0.80
1.76 T 0.36
2.90 T 0.64
2.97 T 0.79
4.52 T 0.53
11.8 T 4.63
2.40 T 1.09
90.3 T 33.3
0.93 T 0.37
5.20 T 1.04
1.77 T 0.39
2.92 T 0.77
3.03 T 0.85
4.55 T 0.52
11.1 T 2.89
2.27 T 0.72
Menopause, Vol. 19, No. 7, 2012 765
MENOPAUSAL TRANSITION AND RISK FACTORS
Copyright © 2012 The North American Menopause Society. Unauthorized reproduction of this article is prohibited.

Page 7

because of the differences in population studied. Our group of
women was relatively homogeneous, whereas the women in
the other studies had a wide range of BMI and multiple ethnic
groups as well as a larger sample of participants.35<37However,
we did observe a significant effect of time for certain variables
such as FM, total trunk FM, and VF and after adjustment for
PAEE for FFM and WC, which is concordant with results from
Lovejoy et al.12
Our second hypothesis, however, was not confirmed by our
results. We observed no significant difference in the changes
in the cardiometabolic risk factors between women with and
without increases in VF independent of menopause status as
well as no significant associations between changes in VF and
changes in cardiometabolic risk factors. On the other hand,
even if perimenopausal and postmenopausal women showed
increases in VF, they did not show clinical deterioration of
their cardiometabolic risk factors. In fact, some metabolic
modifications could be considered as protective, such as the
increase in HDL-C with time. This finding was unexpected;
however, it is consistent with other studies showing a gradual
increase in HDL-C up until menopause and then a progressive
decline in time.12,13,38In fact, our results seem to show that
HDL-C somewhat peaked at menopause onset (year 0) and
then decreased by 3% in the two following years, even if these
changes were not significant. In addition, such an increase in
HDL-C can probably be explained by the increase in TC as
well as the changes in its subclass profile seen with the men-
opausal transition as previously described.39Furthermore, a
decrease in glucose was observed, which is the opposite of
what was shown in an animal model study.40However, it is
consistent with the results reported in one Study of Women’s
Health Across the Nation study that showed a significant
decrease in glucose with time throughout the 9-year study of
949 participants.11
A possible reason that may explain why women did not
show metabolic deteriorations despite showing changes in
metabolic profile is probably related to the observation that
the mean VF value was still lower (72.4 T 40.9 cm2) than the
reported thresholds of 110 cm,2,41,42which is associated with
an increased risk of cardiometabolic disturbance in women. In
addition, our women were leaner compared with previous
studies,3,10<12based on BMI. It has been reported that BMI
and weight gain that occurs during the menopausal transition
are more closely related to changes in glucose homeostasis7,9;
in fact, in our cohort, the women who gained weight did show
higher change in glucose, but it was not significant (results
not shown). However, the mean weight gain in our sample is
0.28 T 3.4 kg, with individual variation, where 68% are weight
stable on a clinical point of view.43Furthermore, 67% of
women were classified as moderate to highly fit for their age
group based on their cardiorespiratory fitness, even after going
through the transition, based on the results of the Aerobics
Center Longitudinal Study44(Table 2). A high level of car-
diorespiratory fitness is known to be associated with lower VF
and lower rates of metabolic health disturbances, reducing the
risk of CVD altogether.45<47
The present study failed to show a significant effect of
the menopausal transition on resting blood pressure; such an
outcome can be explained by the fact that our population was
relatively lean at baseline and after follow-up.48Finally,
knowing that the duration of the postmenopausal period is
one of the determinants of cardiometabolic profile, a longer
follow-up may have been helpful to observe metabolic deteri-
orations in our sample of healthy nonobese women.49
Our study presents certain limitations. First, the population
consisted of healthy women with a BMI less than 30 kg/m2.
Therefore, our findings cannot be generalized to the whole
population. However, it is important to mention that 45% of
the women aged between 40 and 59 years in the Canadian
population present a BMI between 20 and 29 kg/m2. Second,
the amounts of VF and AScF were not measured at year 5
and were only measured in a small subgroup of participants
at year 4; thus, the interpretation should be taken cautiously.
Finally, the number of women in the premenopausal group
decreased throughout the study, slightly reducing power; how-
ever, this is expected because the women selected would
most probably become postmenopausal by the end of this pro-
spective study. Nevertheless, the present study is strengthened
by the well-characterized cohort of women. Throughout the 5
years, there was only a 10% dropout. We also used gold-
standard measuring methods (DXA and CT scan) for the
measurement of body composition and abdominal fat. Through-
out the study, 96% of the women are either in transition or
have become postmenopausal.
CONCLUSIONS
In conclusion, our results suggest that changes in body
composition and fat distribution can occur in nonobese women
as they go through the menopausal transition. However, these
changes were not accompanied by cardiometabolic deteriorations.
Acknowledgments: We thank Mrs. Ann Beninato, Dr. Genevie `ve
Leroux, Miss Ve ´ronique Bertrand, Miss Isabelle Gigue `re, Dr Pierre
Boulay, and Miss Karine Duval for their assistance and excellent
technical support throughout the study.
REFERENCES
1. Third Report of the National Cholesterol Education Program (NCEP)
Expert Panel on Detection. Evaluation, and Treatment of High Blood
Cholesterol in Adults (Adult TreatmentPanel III) final report. Circulation
2002;106:3143-3421.
2. PasqualiR,Casimirri LF, Labate AMM, et al. Body weight, fat distribution
and the menopausal status in women. Int J Obes 1994;18:614-621.
3. Sowers M, Zheng H, Tomey K, et al. Changes in body composition in
women over six years at midlife: ovarian and chronological aging. J Clin
Endocrinol Metab 2007;92:895-901.
4. Enzi G, Gasparo M, Raimondo Biondetti P. Subcutaneous and visceral
fat distribution according to sex, age, and overweight, evaluated by com-
puted tomography. Am J Clin Nutr 1986;44:739-746.
5. Van Gaal LF, Mertens IL, De Block CE. Mechanisms linking obesity
with cardiovascular disease. Nature 2006;444:875-880.
6. Despres JP. Abdominal obesity: the most prevalent cause of the meta-
bolic syndrome and related cardiometabolic risk. Eur Heart J Suppl
2006;8:B4-B12.
766
Menopause, Vol. 19, No. 7, 2012
* 2012 The North American Menopause Society
ABDULNOUR ET AL
Copyright © 2012 The North American Menopause Society. Unauthorized reproduction of this article is prohibited.

Page 8

7. Guthrie JR, Ball M, Dudley EC, et al. Impaired fasting glycaemia in
middle-aged women: a prospective study. Int J Obes Relat Metab Disord
2001;25:646-651.
8. Rosano GM, Vitale C, Mercuro G. The metabolic syndrome in women.
Womens Health (Lond Engl) 2006;2:889-898.
9. Mishra GD, Carrigan G, Brown WJ, Barnett AG, Dobson AJ. Short-term
weight change and the incidence of diabetes in midlife: results from
the Australian Longitudinal Study on Women’s Health. Diabetes Care
2007;30:1418-1424.
10. Guthrie JR, Dennerstein L, Taffe JR, et al. Central abdominal fat and
endogenous hormones during the menopausal transition. Fertil Steril
2003;79:1335-1340.
11. Janssen I, Powell LH, Crawford S, Lasley B, Sutton-Tyrrell K. Meno-
pause and the metabolic syndrome: the Study of Women’s Health Across
the Nation. Arch Intern Med 2008;168:1568-1575.
12. Lovejoy JC, Champagne CM, de Jonge L, Xie H, Smith SR. Increased
visceral fat and decreased energy expenditure during the menopausal
transition. Int J Obes (Lond) 2008;32:949-958.
13. MatthewsKA,CrawfordSL,ChaeCU, et al.Arechangesin cardiovascular
disease risk factors in midlife women due to chronological aging or to the
menopausal transition? J Am Coll Cardiol 2009;54:2366-2373.
14. Guthrie JR, Dennerstein L, Taffe JR, Lehert P, Burger HG. The meno-
pausal transition: a 9-year prospective population-based study. The Mel-
bourne Women’s Midlife Health Project. Climacteric 2004;7:375-389.
15. Guthrie JR, Dennerstein L, Dudley EC. Weight gain and the menopause:
a 5-year prospective study. Climacteric 1999;2:205-211.
16. Sternfeld B, Bhat AK, Wang H, Sharp T, Quesenberry CP Jr. Meno-
pause, physical activity, and body composition/fat distribution in midlife
women. Med Sci Sports Exerc 2005;37:1195-1202.
17. Mayo JR. Radiation dose issues in longitudinal studies involving com-
puted tomography. Proc Am Thorac Soc 2008;5:934-939.
18. Soules MR, Sherman S, Parrott E, et al. Executive summary: Stages of
Reproductive Aging Workshop (STRAW). Climacteric 2001;4:267-272.
19. Canadian Society for Exercise Physiology. The Canadian Physical
Activity, Fitness & Lifestyle Approach, 3rd ed. Ottawa, Canada: Canadian
Society for Exercise Physiology, 2003.
20. Doucet E, Laviolette M, Imbeault P, Strychar I, Rabasa-Lhoret R,
Prud’homme D. Total peptide YY is a correlate of postprandial energy
expenditure but not of appetite or energy intake in healthy women.
Metabolism 2008;57:1458-1464.
21. Brochu M, Starling RD, Tchernof A, Matthews DE, Garcia-Rubi E,
Poehlman ET. Visceral adipose tissue is an independent correlate of glu-
cose disposal in older obese postmenopausal women. J Clin Endocrinol
Metab 2000;85:2378-2384.
22. American College of Sports Medicine, Whaley MH, Brubaker PH, Otto
RM, Armstrong LE. ACSM’s Guidelines for Exercise Testing and Pre-
scription, 7th ed. Baltimore, MD: Lippincott Williams & Wilkins, 2006.
23. Borg GAV. Psychophysical bases of perceived exertion. Med Sci Sports
Exerc 1982;14:377-381.
24. Matthews CE, Ainsworth BE, Thompson RW, Bassett DR Jr. Sources of
variance in daily physical activity levels as measured by an accel-
erometer. Med Sci Sports Exerc 2002;34:1376-1381.
25. Bouten CV, Sauren AA, Verduin M, Janssen JD. Effects of placement
and orientation of body-fixed accelerometers on the assessment of energy
expenditure during walking. Med Biol Eng Comput 1997;35:50-56.
26. Goris AH, Meijer EP, Kester A, Westerterp KR. Use of a triaxial accel-
erometer to validate reported food intakes. Am J Clin Nutr 2001;73:549-553.
27. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concen-
tration of low-density lipoprotein cholesterol in plasma, without use of
the preparative ultracentrifuge. Clin Chem 1972;18:499-502.
28. Svendsen OL, Hassager C, Christiansen C. Age- and menopause-associated
variations in body composition and fat distribution in healthy women as
measured by dual-energy X-ray absorptiometry. Metabolism 1995;44:
369-373.
29. Garaulet M, Perez-Llamas F, Baraza JC, et al. Body fat distribution in
pre- and post-menopausal women: metabolic and anthropometric varia-
bles. J Nutr Health Aging 2002;6:123-126.
30. Zamboni M, Armellini F, Milani MP, et al. Body fat distribution in pre-
and post-menopausal women metabolic and anthropometric variables
and their inter-relationships. Int J Obes 1992;16:495-504.
31. Janssen I, Powell LH, Kazlauskaite R, Dugan SA. Testosterone and
visceral fat in midlife women: the Study of Women’s Health Across the
Nation (SWAN) fat patterning study. Obesity (Silver Spring) 2010;18:
604-610.
32. Cooke PS, Naaz A. Role of estrogens in adipocyte development and
function. Exp Biol Med (Maywood) 2004;229:1127-1135.
33. Heymsfield SB, Gallagher D, Poehlman ET, et al. Menopausal changes
in body composition and energy expenditure. Exp Gerontol 1994;29:
377-389.
34. Van Pelt RE, Jones PP, Davy KP, et al. Regular exercise and the age-
related decline in resting metabolic rate in women. J Clin Endocrinol
Metab 1997;82:3208-3212.
35. Matthews KA, Abrams B, Crawford S, et al. Body mass index in mid-life
women: relative influence of menopause, hormone use, and ethnicity.
Int J Obes Relat Metab Disord 2001;25:863-873.
36. Sternfeld B, Wang H, Quesenberry CP Jr, et al. Physical activity and
changes in weight and waist circumference in midlife women: findings
from the Study of Women’s Health Across the Nation. Am J Epidemiol
2004;160:912-922.
37. Wing RR, Matthews KA, Kuller LH, Meilahn EN, Plantinga PL. Weight
gain at the time of menopause. Arch Intern Med 1991;151:97-102.
38. Derby CA, Crawford SL, Pasternak RC, Sowers M, Sternfeld B, Matthews
KA. Lipid changes during the menopause transition in relation to age and
weight: the Study of Women’s Health Across the Nation. Am J Epidemiol
2009;169:1352-1361.
39. Woodard GA, Brooks MM, Barinas-Mitchell E, Mackey RH, Matthews
KA, Sutton-Tyrrell K. Lipids, menopause, and early atherosclerosis in
Study of Women’s Health Across the Nation Heart women. Menopause
2011;18:376-384.
40. Paquette A, Wang D, Jankowski M, Gutkowska J, Lavoie JM. Effects of
ovariectomy on PPARF, SREBP-1c, and SCD-1 gene expression in the
rat liver. Menopause 2008;15:1169-1175.
41. Williams MJ, Hunter GR, Kekes-Szabo T, et al. Intra-abdominal adipose
tissue cut-points related to elevated cardiovascular risk in women. Int J
Obes 1996;20:613-617.
42. Brochu M, Tchernof A, Turner AN, Ades PA, Poehlman ET. Is there a
threshold of visceral fat loss that improves the metabolic profile in obese
postmenopausal women? Metabolism 2003;52:599-604.
43. Crawford SL, Casey VA, Avis NE, McKinlay SM. A longitudinal study
of weight and the menopause transition: results from the Massachusetts
Women’s Health Study. Menopause 2000;7:96-104.
44. Barlow CE, LaMonte MJ, FitzGerald SJ, Kampert JB, Perrin JL, Blair
SN. Cardiorespiratory fitness is an independent predictor of hypertension
incidence among initially normotensive healthy women. Am J Epidemiol
2006;163:142-150.
45. Arsenault BJ, Lachance D, Lemieux I, et al. Visceral adipose tissue
accumulation, cardiorespiratory fitness, and features of the metabolic
syndrome. Arch Intern Med 2007;167:1518-1525.
46. Sui X, LaMonte MJ, Blair SN. Cardiorespiratory fitness as a predictor
of nonfatal cardiovascular events in asymptomatic women and men.
Am J Epidemiol 2007;165:1413-1423.
47. Lee S, Kuk JL, Katzmarzyk PT, Blair SN, Church TS, Ross R. Car-
diorespiratory fitness attenuates metabolic risk independent of abdominal
subcutaneous and visceral fat in men. Diabetes Care 2005;28:895-901.
48. Cifkova R, Pitha J, Lejskova M, Lanska V, Zecova S. Blood pres-
sure around the menopause: a population study. J Hypertens 2008;26:
1976-1982.
49. Carr MC. The emergence of the metabolic syndrome with menopause.
J Clin Endocrinol Metab 2003;88:2404-2411.
Menopause, Vol. 19, No. 7, 2012 767
MENOPAUSAL TRANSITION AND RISK FACTORS
Copyright © 2012 The North American Menopause Society. Unauthorized reproduction of this article is prohibited.