Bone 40 (2007) 1244 – 1251
www.elsevier.com/locate/bone
The effects of hormone replacement therapy and resistance training on
spine bone mineral density in early postmenopausal women
Gianni F. Maddalozzo a,⁎, Jeffrey J. Widrick a , Bradley J. Cardinal a , Kerri M. Winters-Stone b ,
Mark A. Hoffman a , Christine M. Snow a
a
Department of Nutrition and Exercise Sciences, Bone Research Laboratory, Oregon State University, Corvallis, OR 97331, USA
b
School of Nursing, Oregon Health and Science University, Portland, OR, USA
Received 2 May 2006; revised 29 October 2006; accepted 19 December 2006
Available online 29 December 2006
Abstract
This study evaluated the additive effects of hormone replacement therapy (HRT) and a 1-year site-specific resistance-training (RT) program
involving two free weight exercises (i.e., squat and deadlift) 2 days/week as a strategy to reverse or attenuate bone loss at the lumbar spine in early
postmenopausal women. Participants from a group of self-selected HRT or non-HRT (N = 141) users were randomly assigned to RT (exercise) or
no training, creating four groups: 1) non-HRT plus RT [NHRT plus exercise (n = 35)]; 2) HRT plus RT [HRT plus exercise (n = 37)]; 3) HRT no
resistance training [HRT no exercise (n = 35)]; or 4) control [non-HRT no resistance training group (n = 34)]. Mean age and months past
menopause did not differ between groups (52.1 ± 3.0 years and 52.8 ± 9.9 months, respectively). Post-menopausal status was confirmed by folliclestimulating hormone levels ≥ 40 mIU/mL. Bone mineral density (BMD) of the spine was assessed by Dual Energy X-ray Absorptiometry
(Hologic), at baseline and month 12. Data were analyzed using a 4 (experimental condition) × 2 (time) repeated measures multivariate analysis of
variance to determine the effects of RT on HRT and non-HRT in early postmenopausal women. The main effects for group (P < 0.007), time
(P < 0.001), and the group by time interaction (P < 0.001) were each significant. Control participants experienced an average of −3.6% reduction of
BMD at the spine. HRT treatment with no exercise showed bone loss of − 0.66%. One year of RT produced increases in spine BMD of + 0.43%
and +0.70%, respectively for the NHRT plus exercise, and HRT plus exercise groups with no differences between the NHRT and HRT exercise
groups. In conclusion, RT alone was as effective as HRT in preventing bone loss at the spine and was more effective than HRT alone in attenuating
bone loss at the spine. Moreover, there was no additional benefit in combining HRT with RT for preventing bone loss at the spine in this group of
early postmenopausal women.
© 2006 Elsevier Inc. All rights reserved.
Keywords: Postmenopausal women; HRT; Resistance training; BMD; Osteoporosis
Introduction
Age-related changes to the musculoskeletal system are a
major health concern worldwide. In the United States alone, low
bone density and muscle loss that compromises bone and
muscle strength (i.e., osteopenia and sarcopenia, respectfully)
affects over 26 million people, predominantly postmenopausal
women, and results in excess of 1.5 million fractures a year [12].
Women are at a particular risk for osteoporosis due to an
accelerated bone loss of 2.0–6.5% per year within the first 3 to
⁎ Corresponding author. Fax: +1 541 737 2788.
E-mail address: [email protected] (G.F. Maddalozzo).
8756-3282/$ - see front matter © 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.bone.2006.12.059
5 years after menopause [28]. Evidence suggests that the loss of
ovarian function and decreased circulating 17beta-estradiol
levels accelerate the rate of bone loss. The loss of ovarian
function may also be involved in declines in muscle strength
and lean mass, as Walsh et al. [38] recently reported that 50% of
osteoporotic postmenopausal women have sarcopenia. Clearly,
menopause heralds the onset of changes that have far-reaching
consequences on women's future health [2].
Hormone replacement therapy (HRT) has been prescribed to
post-menopausal women to offset bone loss and to minimize the
symptoms of menopause, while anti-resorptive agents have
been accepted as the mainstay for osteoporosis treatment.
However, the pharmacological approach also has limitations,
G.F. Maddalozzo et al. / Bone 40 (2007) 1244–1251
including side effects and limited efficacy [8]. For example, the
Women's Health Initiative Randomized Trial reported significant adverse affects resulting from hormone replacement
therapy [37]. Currently, ultra-low-dose estrogens are being
explored and they appear to have affects on bone turnover and
bone mass consistent with active antiresorptive methods of
treatment for bone loss after menopause [25].
There is also evidence that physical activity of sufficient
loading can stimulate bone and muscle formation enough to
reduce or attenuate age related osteoporosis and sarcopenia
[17,40]. Identifying high compliance, low-risk exercise programs that improve bone mineral density (BMD), bone and
muscle strength, and physical function could add many quality
years to the lives of older adults.
In a recent meta-analysis of 19 studies, Martyn-St James and
Carroll [23] concluded that high intensity resistance training has
beneficial effects on lumbar spine BMD in postmenopausal
women. However, the potential interactive effects of hormone
replacement therapy (HRT) and resistance training as a method
of reversing bone loss in early postmenopausal women is not
well understood. Additionally, only a few studies, often with
limited sample sizes, have attempted to evaluate this topic and
they have yielded conflicting results [13,26,31,32]. Furthermore, positive change in BMD at the hip associated with
exercise plus HRT has only been reported in one study [13].
We recently reported that at the lumbar spine, men
performing high-intensity free-weight resistance exercise had
significant increases in BMD compared to men in the moderateintensity resistance exercise group (1.9% vs. − 0.52%). For
women, neither high, nor moderate-intensity resistance training
resulted in significant changes in lumbar spine BMD, although
there was a trend for maintenance of spine BMD in the free
weight, high intensity group [22].
Based on these findings we hypothesized that two exercises,
the squat and deadlift played a central role in reducing bone
resorption at the hip and spine. When performing these two
exercises, forces at the hip and spine approach 5–8 times body
weight at exercise intensities corresponding to 80–90% of one
repetition maximum (1RM) [14,15] thus providing sufficient
mechanical stress to increase lean muscle mass and BMD.
Table 1
Baseline characteristics of women who completed the study (N = 125), reported
as Mean ± S.D.
NHRT plus Ex HRT plus Ex HRT no Ex Control
(n = 29)
(n = 33)
(n = 34)
(n = 29)
Age (years)
52.3 ± 3.3
Months post25.3 ± 9.4
menopausal
Months on HRT
0.0
Calcium (g)
1089 ± 521
Protein (g)
57.1 ± 17.1
Energy expenditure 87.5 ± 55.4
(Mets)
52.1 ± 3.1
27.5 ± 10.5
25.0 ± 9.4
1269 ± 589
62.4 ± 21.3
107.8 ± 49.3
51.8 ± 2.9
27.8 ± 10.2
52.5 ± 3.0
23.4 ± 8.7
25.7 ± 8.6
0.0
1205 ± 598 1296 ± 558
58.3 ± 21.3 58.8 ± 20.2
80.9 ± 46.9 93.8 ± 61.2
Twelve-month follow-up results are not reported for calcium, protein
consumption, or energy expenditure since no significant differences was
observed (all P > 0.05).
1245
To investigate the independent and combined effects of
resistance training and estrogen, we examined the bone
response to two site specific free weight exercises (squat and
deadlift) performed 2 days per week plus HRT in early
postmenopausal women (Table 1).
Materials and methods
Study participants
Prior to the commencement of the study, sample size estimates were determined from formal power calculations. With a desired power ≥0.8, alpha = 0.05,
and an expected difference between groups of 4% increase in muscle mass and a
1% increase in spine BMD, 25 subjects per group were needed.
Early postmenopausal women (self selected HRT or non-HRT) were
recruited from a group of subjects who had previously participated in a one-year
follow-up observational study [21]. The Institutional Review Board at Oregon
State University approved the study. All subjects provided written consent after
being fully informed of the nature of the study. Following consent, the self
selected HRT and non-HRT users were randomly assigned to either 1) non-HRT
plus RT [NHRT plus exercise (n = 35)]; 2) HRT plus RT [HRT plus exercise
(n = 37)]; 3) HRT no resistance training [HRT no exercise (n = 35)]; or 4) control
[non-HRT no exercise group (n = 34)].
Acceptance into the original study [21] was based on the following inclusion
and exclusion criteria. Criteria for inclusion: (1) Women who had experienced
the menopause within the previous 0–36 months from the time of baseline
testing as determined retrospectively from questionnaire reports, (2) No
menstrual cycles within the previous 12 months without being pregnant, but
not longer than 36 months (based on questionnaire recall phone screening
interview), (3) Follicle-stimulating hormone levels ≥ 40 mIU/mL (obtained
from the subjects physician), (4) Body mass index (19–30 kg m− 2), (5)
36 months or less of being diagnosed as being post-menopausal by their general
physician, and (6) either taking HRT 0.625 mg conjugated equine estrogen,
(Premarin®) or non HRT use. Criteria for exclusion: (1) Non-HRT users who
had taken HRT for 12 consecutive months prior to applying to the study, (2)
Hypertension, (3) Metabolic disease that may affect bone or muscle metabolism
(including diabetes and thyroid disease), (4) statin medications for hypercholesterolemia), multiple sclerosis, and (4) osteoarthritis or other musculoskeletal
disorders that prevented participation in this study.
Participants were self selected as either HRT or non-HRT replaced. NonHRT replaced participants consisted of women who had never taken HRT either
prior to or during the study. HRT participants were women who began taking
HRT at the onset of menopause as determined by their personal physicians and
continued taking HRT for the duration of the study. No significant differences
were observed at baseline on any variable except for spine BMD. Prior to
randomization, significant differences were observed at the lumbar spine BMD
of 1.00 ± 0.12 and 0.96 ± 0.12 g/cm2 for the HRT and non-HRT groups. We
previously reported that this group of non-HRT replaced women lost 1.2 ± 3.0%
at the spine following a one-year observational study that preceded the present
study [21]. Mean scores (±S.D.) at baseline for the two study conditions (HRT
group and non-HRT group) are presented in Table 2.
All participants were given a log book and asked to record any changes to
their medication routine. The HRT groups were also asked to record compliance
of their HRT medication. Participants in the non-exercise groups were requested
quarterly to return their log books to our laboratory, at which time new log books
were issued to the participants via US mail. Participants in the exercise groups
dropped off their log books during one of their regular exercise sessions each
quarter of the study.
Measures and procedures
Assessment time line
After assignment into one of four groups, strength measurements were
obtained at baseline, 6 and 12 months (±7 days). Only participants who
completed all three assessments were included in analysis. Participants who
withdrew from the study were not invited back for follow-up analyses. Bone
1246
G.F. Maddalozzo et al. / Bone 40 (2007) 1244–1251
Table 2
Mean ± S.D. at baseline and month 12, for NHRT plus Exercise (Ex), HRT plus Exercise, HRT no Exercise and control groups (early postmenopausal women)
Spine: L1–L4
Trochanter: BMD
Femoral neck:
BMD
Total hip: BMD
Body weight (kg)
Lean mass (kg)
% Fat
A: NHRT plus Ex (n = 29) B: HRT plus Ex (n = 33)
C: HRT no Ex (n = 34)
D: Control (n = 29)
Results
Baseline
Baseline
Baseline
P Value Month 12
Month 12
Baseline
Month 12
Month 12
Month 12
0.972 ± 0.13 0.974 ± 0.13 1.01 ± 0.12 1.02 ± 0.12 0.995 ± 0.13 0.988 ± 0.12 0.958 ± 0.12 0.916 ± 0.09 0.01
0.680 ± 0.09 0.682 ± 0.09 0.671 ± 0.10 0.676 ± 0.09 0.667 ± 0.09 0.662 ± 0.09 0.687 ± 0.10 0.676 ± 0.09 0.05
0.756 ± 0.09 0.745 ± 0.09 0.747 ± 0.09 0.742 ± 0.08 0.765 ± 0.08 0.745 ± 0.08 0.752 ± 0.10 0.728 ± 0.08 0.05
A and B > C > D
A and B > C > D
A and B lost <D
0.895 ± 0.11 0.891 ± 0.10 0.877 ± 0.11 0.878 ± 0.10 0.885 ± 0.11 0.882 ± 0.10 0.880 ± 0.10 0.857 ± 10.0
70.0 ± 8.7
70.5 ± 8.0
64.9 ± 8.7
65.9 ± 9.5
69.1 ± 9.1
69.5 ± 10.2 67.1 ± 12.6 68.3 ± 14
44.8 ± 5.2
46.8 ± 5.2
41.3 ± 5.0
43.5 ± 5.4
43.0 ± 4.9
43.4 ± 5.3
43.1 ± 6
43.3 ± 6
32.0 ± 5.3
32.1 ± 5.6
30.0 ± 5.4
30.1 ± 5.3
33.0 ± 5.3
33.7 ± 5.1
31.9 ± 7
32.4 ± 7
D < A, B and D
No differences
A and B > C and D
No differences
mass and body composition were determined at baseline and month 12
(±7 days).
Bone mineral density and body composition
Dual-energy X-ray absorptiometry (DXA, Hologic QDR-4500 Elite A
Waltham, MA) was used to measure bone area (cm2), and bone mineral density
(BMD, g/cm2) of the lumbar spine (L1–L4), proximal femur (total hip, femoral
neck, and greater trochanter) and whole body composition. All scans were
performed and analyzed by a trained laboratory technician using Hologic
software version 9.80 D (Hologic, Inc., Waltham, MA). All follow-up scans
were analyzed using the compare mode. The coefficient of variation (CV) for
repeated DXA scans at the Oregon State University Bone Research Laboratory
are 1.0% for BMD of the hip and lumbar spine and 1.5% for whole body.
Isokinetic muscular strength
Peak force (kg) of the hip abductors and knee extensors and flexors, chest
and upper back was assessed by isokinetic dynamometry (KinCom 500H,
Chattex Corp.). Participants were seated in an adjustable chair and the upper
body was stabilized using straps secured across their hips and chest/shoulders.
All tests were conducted at a speed of 30° per second and were corrected for
gravitational forces. Following a brief warm-up, subjects performed 3–5
maximal efforts, the highest of which was recorded as peak force. Maximal
efforts were separated by approximately 60 s of rest. Protocols employed
previously in our laboratory have revealed good reliability with this population.
The coefficient of variation for muscle strength assessments in our laboratory is
between 4% and 8%. For analyses a mean total strength composite of dependent
variables was calculated for lower body strength [quadriceps + hamstring + hip
abduction/3 = mean total lower body strength score] and upper body strength
[chest + upper back force/2 = mean total upper body strength score] (Table 3).
Table 3
Percent change in Mean (±S.D.) upper and lower body strength following 6 and
12 months of training for the non hormone replaced plus resistance training
(NHRT plus Exercise), hormone replaced plus resistance training (HRT plus
Exercise), hormone replaced plus no resistance training (HRT no Exercise), and
non hormone replaced plus no resistance training (control)
Groups
Percent change upper
body strength
Percent change lower
body strength
6 Months
12 Months
6 Months
12 Months
NHRT plus
Exercise (n = 29)
HRT plus Exercise
(n = 33)
HRT no Exercise
(n = 34)
Control (n = 29)
27.6 ± 16.1⁎
37.4 ± 21.8⁎
26.2 ± 9.9⁎
35.4 ± 13.3⁎
30.4 ± 13.1⁎
41.1 ± 18.3⁎
18.8 ± 16.3⁎
25.4 ± 22.3⁎
8.7 ± 22.5
6.8 ± 30.4
− 1.7 ± 13.0
− 5.2 ± 9.3
5.8 ± 19.9
7.7 ± 26.9
− 2.2 ± 17.6
− 7.0 ± 13.6
⁎ P < 0.05 reflects significant increases in upper and lower body strength
following 6 and 12 months of training compared to the HRT no Exercise and
Controls groups. No significant differences were observed between the NHRT
plus Exercise and HRT plus Exercise groups (P > 0.05).
0.05
0.36
0.05
0.23
One repetition maximum (1RM)
1RM for the squat and deadlift was defined as the maximum amount of
weight each subject could lift once with proper technique. After two warm-up
sets of 10–12 repetitions using light weights, each participant performed a single
repetition with a weight she could lift through a complete range of motion. At the
conclusion of each successful lift, 5–20 lbs were added for the next attempt. This
procedure was repeated until the participant could no longer lift the weight
(generally achieved in 4–6 attempts), and the highest weight lifted successfully
was recorded as the 1RM. New 1RM values, re-testing and load adjustments
were conducted every 8 weeks. The 1RM results were used to determine
the amount of weight to be used by each participant during their exercise
session.
Mean training volume
Total mean training volumes of work for both exercise groups were
calculated by multiplying the amount of weight lifted by the number of sets and
repetitions performed for each exercise and then summed for each workout
(Volume = weight × sets × repetitions) [3]. These were then averaged for each
month over the course of the 12 month study.
Resistance training protocols
Both exercise groups performed free weight back squat and free weight
deadlift exercises. All exercise group participants were asked to perform
repetitions at a speed of 1–2 s for the concentric (lifting) and 2–3 s for the
eccentric (lowering) phases. Training sessions were conducted for 52 weeks
with a frequency of 2 times per week for approximately 50 min under close
supervision of a personal trainer (1–2 trainers per subject). This approach
ensured proper technique, provided motivation and encouragement, and
decreased risk of injury. All training sessions were conducted at Oregon State
University, Corvallis, OR. For every participant, trainers recorded all sets,
repetitions, intensity (% of 1RM); weight lifted and calculated volume of work
at each training session. Participants performed two warm-up sets of 10–12
repetitions at 50% of 1RM for each exercise. For each exercise, participants
performed three working sets at 60–75% of 1 RM (set 1 = 8 reps; set 2 = 10 reps
and set 3 = 12 reps). Each set was followed by approximately 60 s of rest. Prior to
and after each exercise session participants completed a 15–20 min warm-up
and 10 min cool down program consisting of exercises focusing on proper
alignment, posture, muscle engagement, abdominal strength and flexibility
designed to decrease the possibility of injury [9].
The training programs were well tolerated by all participants. None of the
participants developed any overuse or other types of injury sufficient to warrant
any program modifications, withdrawal, or interruption in their training
schedules.
Back squat exercise
The squat is a lower body exercise used in weight training where a barbell is
held across the upper back. The squat is performed by (squatting down) bending
the legs at the knees and hips, lowering the torso between the legs, and then
reversing direction to stand up straight again while maintaining spine alignment.
The torso remains relatively upright and completely uninvolved in the lift in any
capacity but as a supporting structure.
G.F. Maddalozzo et al. / Bone 40 (2007) 1244–1251
Deadlift exercise
The deadlift is an exercise where one lifts a loaded barbell off the ground
from a stabilized bent-over position with emphasis on the muscles of the lower
and upper back and legs. For a more detailed description of the squat and dead
lift exercises please refer to Baechle and Earle [3].
Dietary analysis: block food frequency questionnaire
At the conclusion of the study participants completed a computer-scored
100-item Block Food Frequency Questionnaire [5] designed by the National
Cancer Institute. Information regarding average daily calcium, protein and
grams of protein per kilogram of body weight were obtained using this tool at
baseline and month 12 of the study.
Physical activity
Habitual physical activity levels were assessed using a questionnaire
administered during a laboratory visit [1,4]. The questionnaire asked subjects to
recall the duration and intensity of their typical weekly recreational, leisure, and
household activities at baseline and at the conclusion of the study.
Statistical analysis
Data were entered and analyzed using SPSS statistical software (SPSS Inc.,
Version 11.0, Chicago, IL). A 4 (experimental condition) × 3 (time) repeated
measures multivariate analysis of variance (RM MANOVA) was used to analyze
the effects of training on lower body strength (quadriceps, hamstrings, hip
abduction) and upper body strength (pectorals and latissimus dorsi), and diet.
Where significant multivariate effects were detected, univariate F-tests were
used to isolate the specific dependent variables that were significant. Analysis of
covariance was conducted to determine the effect of group and training on
lumbar spine BMD while controlling for initial values of lumbar spine BMD. A
4 (experimental condition) × 2 (time) RM MANOVA was conducted to
determine the main effect of HRT and resistance training on femoral neck
BMD, trochanter BMD, total hip BMD, and body composition. Again, when a
significant multivariate effect was detected, univariate analyses were performed
to evaluate group-wise differences and Tukey post-hoc tests were used to
evaluate pair-wise differences. For all statistical tests, alpha was set at the
P < 0.05.
Results
Retention and compliance
1247
no HRT) for total hip BMD, femoral neck BMD or greater
trochanter BMD (all P > 0.05. Absolute values for BMD
(Mean ± S.D.) are presented in Table 2. However, to elucidate
interpretation, data are presented in terms of percent change in
the following Discussion section.
Lumbar spine BMD
The control group was significantly lower than the other
three groups, P ≤ 0.001. Over the course of our one-year
observation, the control group had a BMD loss of 3.60 ± 3.70%
at the lumbar spine whereas the HRT no exercise group lost
− 0.66 ± 3.2%. In contrast, HRT plus exercise group resulted
in an increase of 0.70 ± 2.2% in lumbar spine BMD, while
the NHRT plus exercise group had a 0.43 ± 4.3% increase
(Fig. 1).
Greater trochanter BMD
Over the course of 1 year of resistance training, the NHRT
plus exercise and HRT plus exercise groups increased BMD
0.43 ± 3.5% and 0.44 ± 2.6%, respectively P > 0.05. The HRT no
exercise group declined − 0.60 ± 4.6%, whereas the control
group lost − 1.5 ± 3.2% (Fig. 2).
Femoral neck BMD
The NHRT plus exercise and HRT plus exercise groups
lost − 1.2 ± 4.3% and − 0.61 ± 2.9%, respectively. The control
group lost the most bone over 1 year at a rate of − 3.9 ±
3.8%, P > 0.05 whereas the HRT no exercise group had a
similar loss as the NHRT plus exercise group of − 1.2 ± 3.3%
(Fig. 3).
Total hip BMD
The NHRT plus exercise and HRT plus exercise groups
declined at the total hip BMD by − 0.30 ± 3.1% and − 0.52 ±
3.6%, respectively. The control group declined the most at
At baseline, sample sizes were: NHRT plus exercise (n = 35);
HRT plus exercise (n = 37); HRT no exercise (n = 35); and 4)
control (n = 34). Overall, 122 of 141 (86.5%) of enrolled
participants completed the one-year study. Retention rates were:
83% NHRT plus exercise, 89% HRT plus exercise; 91% HRT
no exercise; and 82% control group. Women did not complete
the study for the following reasons: 1) stopped taking HRT
(n = 3) or started taking HRT (n = 2), 2) moved out of the area
(n = 5), 3) personal reasons, not related to the study (n = 4), and
4) did not enjoy the exercise program (n = 5). Attendance for the
exercise sessions for the non-HRT plus exercise (84.7 ± 12.8%)
and HRT plus exercise group (86.2 ± 11.4%) were similar. Log
books that were collected quarterly suggest that compliance for
all women taking daily HRT medication was reasonable with
participants only forgetting to take their daily medication on
average 2.47 ± 1.2 times per month.
Bone mineral density
At baseline independent t-test revealed no significant
differences between groups prior to randomization (HRT or
Fig. 1. Percent change (± S.E.) for lumbar spine (L1–L4) BMD at the completion
of the one-year exercise (Ex) interventions. *P < 0.005. At the conclusion of the
study, the NHRT and HRT exercise groups had greater BMD than the HRT and
control groups. The control group had lower BMD than the NHRT and HRT
exercise groups and HRT group P < 0.05.
1248
G.F. Maddalozzo et al. / Bone 40 (2007) 1244–1251
Fig. 2. Percent change (±S.E.) for trochanter BMD at the completion of the oneyear exercise (Ex) interventions. *P < 0.05. At the conclusion of the study, the
NHRT and HRT exercise groups had greater BMD than the HRT and control
groups. The control group had lower BMD than the NHRT and HRT exercise
groups and HRT group P < 0.05.
Fig. 4. Percent change (± S.E.) for total hip BMD at the completion of the oneyear exercise (Ex) interventions. *P < 0.05. At the conclusion of the study, the
control group lost significantly more bone than the HRT exercise, NHRT
exercise, or HRT groups.
At baseline, no between group differences on any of the
individual strength measures (quadriceps, hamstring, hip
abduction, pectoral (chest) and latissimus dorsi (upper back)
strength were observed. Six and 12 month results reflect percent
change from baseline.
respectively, (both P ≤ 0.01). After 6 months, the gain in upper
body strength in the HRT no exercise and control groups was not
significant (9 ± 23% and 6 ± 20%, respectively) and did not
change over the final 6 months of the study (P > 0.05).
Improvements in strength for the exercise groups were
significant when compared to the non-exercising groups at
months 6 and 12 regardless of HRT status. However, no
significant differences were observed between the HRT plus
exercise and NHRT plus exercise group responses (P > 0.05)
(Table 3).
Upper body strength
Lower body strength
At 6-month follow-up the HRT plus exercise and NHRT plus
exercise groups increased upper body strength by 28 ± 16% and
30 ± 13% respectively (both P ≤ 0.01). At 12 months the overall
increase in upper body strength was 41 ± 18% and 37 ± 22%
At 6-month follow-up both the HRT plus exercise and
NHRT plus exercise groups increased their lower body strength
19 ± 16% and 26 ± 10% respectively (both P ≤ 0.01). At
12 months the overall increase in lower body strength was
25 ± 22% and 35 ± 13%, respectively (both P ≤ 0.01). At
6 months, lower body strength in the HRT no exercise and
control groups decreased − 1.7 ± 13% and −5 ± 9%, respectively
(both P > 0.05). Overall the non-exercisers experienced a nonsignificant decrease in lower body strength from baseline of
− 2.2 ± 17% and − 7.0 ± 13%, respectively (both P > 0.05). In the
exercising groups, improvements in strength were significant
when compared to the non-exercising groups at months 6 and
12 regardless of HRT status. However, no significant differences were observed between the HRT plus exercise and NHRT
plus exercise groups (both P > 0.05) (Table 3).
− 2.4 ± 2.3%, P > 0.05 whereas the HRT no exercise group lost
− 0.79 ± 2.9% at the hip (Fig. 4).
Strength analyses
Mean training volumes of work
Fig. 3. Percent change (±S.E.) for femoral neck BMD at the completion of the
one-year exercise (Ex) interventions. *P < 0.05. At the conclusion of the study,
the control group lost significantly more bone than the HRT exercise, NHRT
exercise, or HRT groups.
A 2 (group: NHRT + exercise and HRT + exercise) × 12 (time)
repeated measures analysis of variance was employed to
compare mean training volume of work completed by the two
exercise groups over time. During the study mean training
volume of work significantly increased for both exercise groups
(P ≤ 0.001) however, no differences existed between the HRT
G.F. Maddalozzo et al. / Bone 40 (2007) 1244–1251
Fig. 5. Total volume of work performed. This figure presents the mean monthly
average of weight lifted by each of the two exercise groups (NHRT plus Exercise
and HRT plus Exercise). Total mean training volumes of work between both
exercise groups were calculated by multiplying the amount of weight lifted by
the number sets and repetitions performed for each exercise, and then summed
for each workout.
plus exercise and NHRT plus exercise groups over the 12 month
intervention (P = 0.84) (Fig. 5).
Body composition
At baseline no significant between group differences (Table
2) were detected for either total body lean mass or total body fat
mass (P ≤ 0.05). There was an overall increase in lean mass of
4.6 ± 2.7% and 3.9 ± 2.7%, respectively for the HRT plus
exercise and NHRT plus exercise groups after 12 months of
training (P ≤ 0.05). Improvements in lean mass in exercisers
were significant when compared to the non-exercising groups at
month 12 regardless of HRT status. However, no significant
differences were observed for lean mass and fat mass between
the HRT plus exercise and NHRT plus exercise groups
(P > 0.05) (Table 2) at the conclusion of the study. The nonexercising HRT and control groups experienced no changes in
lean mass non-significant increases in lean mass over the course
of the intervention. Improvements in lean mass in the HRT plus
exercise and NHRT plus exercise groups were significant when
compared to the non-exercising groups at month 12 regardless
of HRT status.
Dietary protein, calcium and energy expenditure analyses
No significant between group differences were observed on
any of the dietary variables (calcium and protein) assessed or
energy expenditure (P > 0.05) (Table 1).
Discussion
Our results demonstrate that regardless of hormone replacement therapy status, free weight squat and deadlift exercises
performed 2 days per week prevented bone loss at the spine in
this group of early postmenopausal women. Moreover,
resistance training regardless of the HRT status was more
effective than hormone replacement therapy alone in preventing
1249
bone loss at the spine. Additionally, resistance training
attenuated declines in bone at the femoral neck, greater
trochanter, and total hip in comparison to the control group,
and resistance training was as effective as HRT alone in
attenuating loss of bone at all sites reported. In this group of
women, resistance training may be as effective as HRT in
attenuating or preventing bone loss without the potential side
effects often associated with a pharmacological approach.
Furthermore, resistance training also increased lean mass and
muscle strength, both of which are important determinants of
functional independence and lower fall risk in older women.
Though the suppression of bone remodeling in response to
hormone replacement therapy is well documented [11,33] the
response to exercise and exercise plus HRT in postmenopausal
women is unclear. To date, several studies have reported that the
combination of weight-bearing exercise and estrogen replacement
therapy was more effective than either exercise alone or estrogen
replacement therapy alone in increasing total body and spine
BMD in older postmenopausal women [13,19]. Our findings with
early postmenopausal women do not support this hypothesis since
resistance training plus HRT did not produce greater gains in bone
mass, muscle strength and lean mass than resistance training
alone. However, we must point out that the women in the Kohrt et
al. [19] study were 60 to 72 years of age and HRT may have an
additive effect when combined with weight-bearing exercise in
women beyond early stages of menopause.
Even though we report small increases of +0.43% and
+0.70%, at the lumbar spine for the non-HRT plus resistance
training and HRT plus resistance training groups, these
increases were significant when compared to the declines
reported by the control and HRT no resistance training groups.
Additionally, the non-HRT plus resistance training and HRT
plus resistance training groups significantly increased trochanter BMD by + 0.44% and + 1.0%, respectively, compared to
decreases of − 0.60% and − 1.64% for HRT no resistance
training and control groups, respectively.
While the increases in BMD at these two sites (i.e., spine
and trochanter) were small, they are similar to changes
reported in previous studies involving postmenopausal women
[13,18,22,26,31,32,34] and the effect sizes reported in a metaanalysis [17]. These small increases, if maintained over time,
can translate into long-term preservation of bone mass
compared to the expected age-related losses. It is also
important to point out that resistance training significantly
attenuated bone loss at the trochanter when compared to the
control group as well as the hormone replaced no resistancetraining group. Importantly, the mean age of participants in
many of these previous studies ranged from 60 to 70 years of
age [13,19,26,27,29–31], whereas the average age of our
participants at baseline was 52 years. Our results indicate that
substantially younger women can derive benefits from
resistance training. If resistance training is continued, it may
be possible for women to attain greater BMD into their 60s
and 70s.
The onset of menopause is associated with the most rapid
period of postmenopausal bone loss due to the initial and rapid
withdrawal of estrogen [7,10,29]. Few studies have included
1250
G.F. Maddalozzo et al. / Bone 40 (2007) 1244–1251
early postmenopausal women as the study population and it is
possible that the competing affects of hypoestrogenism
diminish the efficacy of exercise to arrest bone loss. Our
population represents a critical point in time for intervention,
especially since this group of women is at the peak of
postmenopausal bone loss. Thus, halting this loss has the
potential to result in greater bone savings.
Our study was unique and designed to maximize changes
in bone and exercise compliance. The majority of resistance
training interventions that have focused on attenuating or
increasing bone mineral density in postmenopausal women
have employed traditional resistance training programs
involving 8–12 total body exercises 3 days per week
[13,22,26,32] and typically lasting 60–70 min per session.
If we eliminated the warm-up and cool down portions of our
resistance training program, the two site specific exercises,
free weight squats and deadlift, took approximately 10 min
per day, 2 days per week. The exercises required minimal
equipment (squat rack and weights) and is an exercise
program one could potentially do at home once the
participants are correctly instructed by an expert trainer on
how to perform each exercise correctly. Inclusive of all
program elements, 2 days per week, 45 min per session
(10 min of resistance exercises and 35 min of warm-up and
cool down exercises) is a reasonable amount of time to ask of
any individual to exercise especially given the potential
benefits one could derive from such an exercise program.
Findings from our study suggest that these two site specific
exercises (squat and deadlift) attenuated bone loss at the spine
and greater trochanter regardless of HRT status.
In addition to BMD improvements, resistance training can
impact other variables associated with decreased risk of
fractures and disability. Beginning at ∼ 40 years of age, humans
experience a gradual, but progressive, decline in the peak
voluntary isometric torque, dynamic torque, and maximal
contraction velocity of the lower limb skeletal muscles [20].
This age-related decline in muscle function appears to be due to
reductions in muscle quantity and to changes in muscle quality
[6]. Skelton and colleagues [35] have proposed that estrogen
confers a beneficial effect by altering muscle function at the
level of the cross-bridge, thus suggesting that the menopause
may also be associated with an increased decline in strength as
well as bone loss. However, our group observed no differences
in the cross sectional area (CSA), fiber type distribution, and
Ca2±-activated contractile properties of slow vs. fast vastus
lateralis muscle fibers obtained from early HRT and non-HRT
replaced post-menopausal women at baseline. These findings
provide additional evidence that HRT does not appear to
attenuate or preserve muscle strength or lean mass in HRT
replace postmenopausal women [39].
Our exercise program produced significant increases in both
lower and upper body strength regardless of HRT status.
Moreover, our participants did not experience a plateau of
strength increase after 6 months of training. We were also
surprised to see the large increases in upper body strength for
both resistance training groups. The isometric contraction of
upper body muscle groups required in stabilizing the body when
executing a free weight squat and deadlift could have indirectly
produced the upper body strength increases we observed. Since
women have relatively weak upper body strength when
compared to men [16,24,34,36] isometric stimulus from these
two exercises may have been sufficient to induce the gains in
upper body strength observed. Additionally, both exercise
groups experienced an increase in lean mass of approximately
4% regardless of HRT status.
To our knowledge, this is the first exercise intervention study
to examine the effects of a site specific resistance training
program using two very unique free weight exercises, squats and
deadlift, in this population of early postmenopausal women as a
strategy to reverse or attenuate bone loss at the spine. All
participants were considered postmenopausal for no more than
36 months by their physician (HRT: 27.16 ± 10.07 months and
no HRT: 24.6 ± 10.07 months post-menopause). Additionally,
participants in the HRT groups consumed 0.625 mg conjugated
equine estrogen (brand name Premarin®) as prescribed by their
physicians.
Several limitations of this study design must be noted. First,
this was not a randomized clinical trial design. Rather
participants were self selected as either HRT or non-HRT
replaced. Thus, in this group of women, those who chose HRT
may have had different socioeconomic status and education that
influenced their awareness of the importance of physical
activity and nutrition than those who did not chose HRT.
Although we attempted to control for some of these potential
confounding factors we may not have been able to account for
all of them. Self-selection may influence the baseline values or
the pre- and post-values of the non-trained groups. For instance
women not taking HRT may be more health conscience and
more likely to perform exercise that might increase or maintain
their bone mass than women not on HRT, or vice versa.
However, there is little reason to expect that women who chose
to take HRT would have different physiological responses to
strength training compared to women who chose not to take
HRT. Thus, self-selection would not be expected to confound
the interpretation of adaptations noted for the HRT-trained and
non HRT-trained groups in this study. This strengthens our
conclusion that strength training provides similar benefits as
HRT in this population of participants.
Within the limitations of this study, we have shown that two
site specific resistance training exercises, squat and deadlift can
be an effective substitute for HRT in preserving lumbar spine
BMD during the early stages of menopause. Regular participation in a resistance training program could potentially decrease
the dosage of drugs required to induce bone formation in ways
that enhance efficacy and reduce the risk of side effects
associated with drug therapies.
Acknowledgments
This research was supported by grants from The American
Federation of Aging the Erkkila Endowment for Health and
Human Performance, Good Samaritan Hospital Foundation,
Corvallis, OR and The American College of Sports Medicine.
Equipment was provided by LifeFitness, Schiller Park, IL.
G.F. Maddalozzo et al. / Bone 40 (2007) 1244–1251
References
[1] Ainsworth BE, Haskell WL, Whitt MC, Irwin ML, Swartz AM, Strath SJ,
et al. Compendium of physical activities: an update of activity codes and
MET intensities. Med Sci Sports Exerc 2000;32:S498–504.
[2] Aloia JF, McGowan DM, Vaswani AN, Ross P, Cohn SH. Relationship of
menopause to skeletal and muscle mass. Am J Clin Nutr 1991;53: 1378–83.
[3] Baechle TR, Earle RW, National Strength and Conditioning Association
(U.S.). Essentials of strength training and conditioning. Champaign, IL:
Human Kinetics; 2000.
[4] Blair SN. The aerobics center longitudinal study physical activity
questionnaire. Med Sci Sports Exerc 1997;29(6):10–4 [Supplement].
[5] Block G. Block food questionnaire. Berkley, CA: National Cancer
Institute; 1998.
[6] Brooks SV, Faulkner JA. Skeletal muscle weakness in old age: underlying
mechanisms. Med Sci Sports Exerc 1994;26:432–9.
[7] Cann CE, Genant HK, Ettinger B, Gordan GS. Spinal mineral loss in
oophorectomized women. Determination by quantitative computed
tomography. JAMA 1980;244:2056–9.
[8] Cauley JA, Robbins J, Chen Z, Cummings SR, Jackson RD, LaCroix AZ, et al.
Effects of estrogen plus progestin on risk of fracture and bone mineral density:
the Women's Health Initiative randomized trial. JAMA 2003;290:1729–38.
[9] Egoscue P, Gittines R. The Egoscue Method of health through motion:
revolutionary program that lets you rediscover the body's power to
rejuvenate it. New York: HarperCollins; 2001.
[10] Elders PJ, Netelenbos JC, Lips P, van Ginkel FC, van der Stelt PF.
Accelerated vertebral bone loss in relation to the menopause: a crosssectional study on lumbar bone density in 286 women of 46 to 55 years of
age. Bone Miner 1988;5:11–9.
[11] Felson DT, Zhang Y, Hannan MT, Kiel DP, Wilson PW, Anderson JJ. The
effect of postmenopausal estrogen therapy on bone density in elderly
women. N Engl J Med 1993;329:1141–6.
[12] Foundation NO, National Osteoporosis Foundation. Disease Statistics
February, 2004. hhtp://www.nof.org.osteoporosis/stats.htm; 2004.
[13] Going S, Lohman T, Houtkooper L, Metcalfe L, Flint-Wagner H, Blew R,
et al. Effects of exercise on bone mineral density in calcium-replete
postmenopausal women with and without hormone replacement therapy.
Osteoporos Int 2003;14:637–43.
[14] Granhed H, Jonson R, Hansson T. The loads on the lumbar spine during
extreme weight lifting. Spine 1987;12:146–9.
[15] Hattin HC, Pierrynowski MR, Ball KA. Effect of load, cadence, and
fatigue on tibio-femoral joint force during a half squat. Med Sci Sports
Exerc 1989;21:613–8.
[16] Heyward VH, Johannes-Ellis SM, Romer JF. Gender differences in
strength. Res Q 1986;4:265–7.
[17] Kelley GA, Kelley KS, Tran ZV. Resistance training and bone mineral
density in women: a meta-analysis of controlled trials. Am J Phys Med
Rehabil 2001;80:65–77.
[18] Kerr D, Morton A, Dick I, Prince R. Exercise effects on bone mass in
postmenopausal women are site-specific and load-dependent. J Bone
Miner Res 1996;11:218–25.
[19] Kohrt WM, Snead DB, Slatopolsky E, Birge Jr SJ. Additive effects of
weight-bearing exercise and estrogen on bone mineral density in older
women. J Bone Miner Res 1995;10:1303–11.
[20] Larsson L, Grimby G, Karlsson J. Muscle strength and speed of movement in
relation to age and muscle morphology. J Appl Physiol 1979;46:451–6.
[21] Maddalozzo GF, Cardinal BJ, Li F, Snow CM. The association between
hormone therapy use and changes in strength and body composition in
early postmenopausal women. Menopause 2004;11:438–46.
1251
[22] Maddalozzo GF, Snow CM. High intensity resistance training: effects on
bone in older men and women. Calcif Tissue Int 2000;66:399–404.
[23] Martyn-St James M, Carroll S. High-intensity resistance training and
postmenopausal bone loss: a meta-analysis. Osteoporos Int 2006;17:
1225–40.
[24] Miller AE, MacDougal J, Tarnopolsky MA, Sale DG. Gender difference in
strength and muscle fiber characteristics. Eur J Appl Physiol 1993;66:
254–62.
[25] Mizunuma H, Shiraki M, Shintani M, Gorai I, Makita K, Itoga S, et al.
Randomized trial comparing low-dose hormone replacement therapy and
HRT plus 1alpha-OH-vitamin D3 (alfacalcidol) for treatment of postmenopausal bone loss. J Bone Miner Metab 2006;24:11–5.
[26] Nelson ME, Fiatarone MA, Morganti CM, Trice I, Greenberg RA, Evans
WJ. Effects of high-intensity strength training on multiple risk factors for
osteoporotic fractures. A randomized controlled trial. JAMA 1994;272:
1909–14.
[27] Nichols JF, Omizo DK, Peterson KK, Nelson KP. Efficacy of heavyresistance training for active women over sixty: muscular strength, body
composition, and program adherence. J Am Geriatr Soc 1993;41:205–10.
[28] Pouilles JM, Tremollieres F, Ribot C. The effects of menopause on
longitudinal bone loss from the spine. Calcif Tissue Int 1993;52:340–3.
[29] Pouilles JM, Tremollieres F, Saint-Martin F, Ribot C. Postmenopausal
bone loss: results of a topographic study by X-ray absorptiometry. Rev
Rhum Ed Fr 1993;60:891–6.
[30] Preisinger E, Alacamlioglu Y, Pils K, Bosina E, Metka M, Schneider B,
et al. Exercise therapy for osteoporosis: results of a randomised controlled
trial. Br J Sports Med 1996;30:209–12.
[31] Pruitt LA, Jackson RD, Bartells RL, Lehnhard HJ. Weight-training effects
on bone mineral density in early postmenopausal women. J Bone Miner
Res 1992;7:179–85.
[32] Pruitt LA, Taaffe DR, Marcus R. Effects of a one-year high-intensity
versus low-intensity resistance training program on bone mineral density
in older women. J Bone Miner Res 1995;10:1788–95.
[33] Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C,
Stefanick ML, et al. Risks and benefits of estrogen plus progestin in
healthy postmenopausal women: principal results From the Women's
Health Initiative randomized controlled trial. JAMA 2002;288:321–33.
[34] Sipila S, Taaffe DR, Cheng S, Puolakka J, Toivanen J, Suominen H.
Effects of hormone replacement therapy and high-impact physical exercise
on skeletal muscle in post-menopausal women: a randomized placebocontrolled study. Clin Sci (Lond) 2001;101:147–57.
[35] Skelton DA, Phillips SK, Bruce SA, Naylor CH, Woledge RC. Hormone
replacement therapy increases isometric muscle strength of adductor
pollicis in post-menopausal women. Clin Sci (Lond) 1999;96:357–64.
[36] Vanderburgh PM, Kusano M, Sharp M, Nindl B. Gender differences in
muscular strength: an allometric model approach. Biomed Sci Instrum
1997;33:100–5.
[37] Vogel RA. The changing view of hormone replacement therapy. Rev
Cardiovasc Med 2003;4:68–71.
[38] Walsh MC, Hunter GR, Livingstone MB. Sarcopenia in premenopausal
and postmenopausal women with osteopenia, osteoporosis and normal
bone mineral density. Osteoporos Int 2005.
[39] Widrick JJ, Maddalozzo GF, Lewis D, Valentine BA, Garner DP, Stelzer
JE, et al. Morphological and functional characteristics of skeletal muscle
fibers from hormone-replaced and nonreplaced postmenopausal women.
J Gerontol A Biol Sci Med Sci 2003;58:3–10.
[40] Wolff I, van Croonenborg JJ, Kemper HC, Kostense PJ, Twisk JW. The
effect of exercise training programs on bone mass: a meta-analysis of
published controlled trials in pre- and postmenopausal women. Osteoporos
Int 1999;9:1–12.