Osteoporos Int (2001) 12:104–111
ß 2001 International Osteoporosis Foundation and National Osteoporosis Foundation
Original Article
Effects of Lifetime Volleyball Exercise on Bone Mineral Densities in
Lumbar Spine, Calcaneus and Tibia for Pre-, Peri- and Postmenopausal
Women
M. Ito1, T. Nakamura2, S. Ikeda2, Y. Tahara8, R. Hashmi1, K. Tsurusaki1, M. Uetani1 and K. Hayashi1
1
Department of Radiology, Nagasaki University School of Medicine; 2Department of Orthopedic Surgery, University of
Occupational and Environmental Health, Kitakyushu; and 3Department of Health and Physical Education, Faculty of Education,
Nagasaki University, Nagasaki, Japan
Abstract. To clarify the effects of habitual volleyball
exercise on bone in women during the menopausal
periods, we measured bone mineral densities (BMDs) of
the lumbar spine, calcaneus and tibia every 12 months
for 2 years and estimated factors related to the baseline
values and annual loss rates. Forty Japanese female
volleyball players 42–62 years of age, who had belonged
to the district non-professional club for more than 10
years on average, were recruited. Twenty women had
regular menstruation at the start, but 7 underwent
menopause during the study. Fifty-nine healthy women
who did not participate in habitual exercise, but were
otherwise comparable with the players, were recruited as
the controls. The lumbar and calcaneus BMDs were
measured by dual-energy X-ray absorptiometry (DXA),
and both the volumetric BMDs and cross-sectional
geometry at the diaphysis of the tibia were measured
by peripheral quantitative computed tomography. The
baseline BMDs of all measured sites and the values of
the cortical area and moment of inertia of the tibia in the
players were significantly higher than the values in the
control women. In the peri- and postmenopausal players,
the baseline values of the lumbar and calcaneus BMDs
related to total years of participating in regular exercises
during adulthood including volleyball (TYE), body mass
index (BMI) and years since menopause (YSM). Tibia
cortical area and moment of inertia values related to
TYE. Annual bone loss rates in the tibia and calcaneus of
Correspondence and offprint requests to: Masako Ito, MD,
Department of Radiology, Nagasaki University School of Medicine,
1-7-1 Sakamoto, Nagasaki, 852-8501 Japan. Tel: +81 958 49 7354.
Fax: +81 958 49 7357.
Osteoporosis
International
players were significantly smaller than those values in
the controls. However, the bone loss rates in the lumbar
spine did not differ significantly between the groups. The
bone loss rate in the calcaneus was significantly related
to the current number of training hours per week and
YSM. The rate of bone loss in the tibia was related to
BMI. These data indicated that the total number of years
participating in exercise activity during adulthood have
positive effects on lumbar and calcaneus BMDs and the
cortical structure of the tibia. Habitual volleyball
exercise apparently did not alleviate the menopause-
related bone loss in the lumbar spine.
Keywords: Age-related bone loss; DXA; Estrogen;
Exercise; pQCT
Introduction
High peak bone mass attained in young adulthood and a
reduction in the rate of bone loss in the postmenopausal
period are considered major factors in preventing
osteoporosis. Since bone minieral density (BMD) is
positively associated with the degree of physical activity
and body weight [1,2], high-impact exercises are
assumed to increase bone mass. Previous reports have
demonstrated that resistance exercise increases BMDs of
the axial and appendicular skeletons during adolescence
[3,4] and the premenopausal period [5].
For peri- and postmenopausal women, however, the
effect of the resistance exercise for 1–1.5 years seemed
to differ depending on the skeletal site and age of the
Effects of Lifetime Volleyball Exercise on BMD 105

subject [6–10]. Endurance exercise including walking, Materials and Methods

jogging and climbing stairs was able to increase the
lumbar spine and femoral neck BMD in postmenopausal
non-osteoporotic women 60–70 years old [6,7]. In
perimenopausal healthy women aged 52–53 years, a
similar type of endurance exercise maintained BMD of
the femoral neck but did not affect the lumbar spine and
calcaneus [8]. In a study of young premenopausal
volleyball players, the osteogenic effects of volleyball
appeared at weight-bearing bone sites as well as the
dominant arm [9]. In postmenopausal healthy women in
their fifties, jumping exercise with a height of
approximately 10 cm, producing a ground reaction
force of 4 times body weight, did not affect the
BMDs of the lumbar spine or femoral neck [10].
Resistance training in the limb with high load increased
the BMD of the hip at the trochanter and Ward’s
triangle, but not at the femoral neck, in 40- to 70-year-
old women [11]. These data suggest that the effects of
the loading exercise differ depending on the bones
involved [8–11].
Volleyball is a ground-reaction force exercise with
high strain rates and peak stresses on the skeleton. In
Japan, the non-professional national women’s volleyball
society was established to promote women’s health with
the support of the Ministry of Education, organizing
women’s volleyball clubs in various districts. Members
of the clubs participate regularly in training. The purpose
of this study was to clarify the site-specific effects of
habitual volleyball practice on bone mass and geometry
in women during the menopausal periods. We long-
itudinally measured their BMDs and geometry of the
lumbar spine, tibia and calcaneus for 2 years. Then,
factors related to the baseline values and annual loss
rates were analyzed. Results were compared with those
in control women who had not participated in habitual
exercise.
Table 1. Demographics of the volleyball players and the controls
Subjects
Forty female volleyball players, belonging to a district
club of the non-professional Japan national women’s
volleyball society, were included in the study. Twenty
premenopausal and 20 postmenopausal women were
randomly selected from the players. Menopause was
defined as no menstruation for at least 6 consecutive
months. The premenopausal women had regular
menstrual cycles at the start. Since 7 of 20 premeno-
pausal women experienced menopause during the study
period, they were classified at the end into a new group
described as the perimenopausal group. Thus, there were
three groups according to menstrual status: 13 pre-
menopausal women with a regular menstrual cycle, 7
perimenopausal women who underwent menopause
during the study period, and 20 postmenopausal women.
Fifty-nine healthy women who did not participate in
habitual exercise were selected for the controls. They
were all residents of the same district, not participating
in any kind of regular or organized sports activity in the
present or past during adulthood. Controls consisted of
24 premenopausal and 35 postmenopausal women at the
start. Eleven of the 24 premenopausal women experi-
enced menopause during the study period, and they were
classified into the perimenopausal group at the end.
There were thus three control groups: 13 premenopausal
women with a regular menstrual cycle, 11 perimeno-
pausal women and 35 postmenopausal women.
Demographics of the subjects and the controls,
including age at the start of the study, age at menopause,
years since menopause (YSM), body weight (BW, kg),
body height (BH, cm) and body mass index (BMI = BW/
BH2, kg/cm2), are presented in Table 1. Since there was
no significant difference in these parameters between the

Volleyball players Controls

Pre. Peri. Post. Pre. Peri. Post.

No. of subjects 13 7 20 13 11 35
Age (years) 43.4+0.7 50.3+1.5 54.4+1.2 40.1+1.6 49.1+0.7 56.3+0.7
(42–49 years) (51–54) (46–62) (33–48) (47–54) (45–62)
Age at menopause (years) – 50.4+0.8 47.1+1.0 – 49.3+0.8 50.2+0.5
YSM (years) – – 5.3+0.8 – – 6.1+0.6
BW (kg) 61.1+2.5 55.7+1.7 57.2+1.3 55.2+1.9 54.6+1.4 53.7+1.2
BH (cm) 157.6+2.1 157.0+1.1 156.5+1.1 158.0+1.1 156.0+1.4 154.0+0.8
BMI (kg/cm2) 0.24+0.01 0.24+0.01 0.23+0.01 0.22+0.01 0.23+0.01 0.22+0.01
TYP (years) 11.5+1.6 19.0+1.6* 19.8+1.8*
(6–29 years) (10–30) (7–30)
TYE (years) 17.1+2.3 25.4+2.7* 25.0+2.2*
(10–35 years) (16–36) (9–35)
Current training hours per week 4.4+0.3 3.2+0.1 3.5+0.4

Values are the mean and the standard error of the mean (SEM).
Pre., premenopausal women; Peri., perimenopausal women; Post., postmenopausal women; YSM, years since menopause; BW, body weight; BH,
body height; BMI, body mass index; TYP, total years of participation in volleyball; TYE, total years of participation in regular exercise including
periods of past activity and volleyball club participation.
*.Significant difference (p<0.05) versus premenopausal players by Student’s t-test.
106
players and control women of either menstrual status,
comparative analyses between the subjects and controls
appeared to be justified.
Regular menstrual cycles were confirmed in the
premenopausal women by questionnaire. The average
daily intake of milk and dairy products did not differ
significantly in each menstrual group between the player
and control groups. Women who were taking or had
taken estrogen or other drugs that influence bone
metabolism were excluded. Women with a history of
previous fracture were not included. Informed consent
for the study was obtained from each subject. Years
during which the players participated to the volleyball
club before the start of the study, total years of
participating in regular exercise including past activity
and volleyball, and current number of training hours per
week (52 week/year) are presented in Table 1.
Bone Density and Size Measurements
Measurements of bone density and size were performed
three times at intervals of 12 months between 1996 and
1998. Dual-energy X-ray absorptiometry (DXA) of the
lumbar spine and calcaneus and peripheral quantitative
computed tomography (pQCT: Densiscan 1000, Scanco
Medical, Zurich, Switzerland) of the tibia were
performed.
DXA Measurement of the Lumbar Spine and Calcaneus.
BMD (g/cm2) values and projection area (cm2) of the
lumbar spine (L2–L4) in anteroposterior (AP) projection
were determined using DXA (EXP 5000, Lunar,
Madison, WI). The lateral side projection area (cm2)
and BMD (g/cm2) values of the right calcaneus were
measured using DXA (HeelScan DX-2000, KDK,
Kyoto, Japan) [12]. Coefficient of variation (CV)
values of the BMD measurements for the lumbar spine
and calcaneus were 1.1% and 0.8%, respectively.
Peripheral QCT of the Tibia. Six consecutive slices of
tomographic measurements were obtained from the mid-
diaphysis. Each measurement included an area of 1 mm
thickness with 1.5 mm intervals [13,14]. Total volu-
metric BMD values were obtained from BMD values for
the total cross-sectional area. The cross-sectional cortical
area was obtained by subtracting the medullary cavity
area from the total cross-sectional area. The cross-
sectional moment of inertia (MI) was calculated
according to the following equation: MI = Sai(yi 7
ye)2, where ai is the voxel area and (yi 7 ye) is the
distance in the vertical direction from each voxel to the
horizontal axis passing through the center of the area.
Summation was performed over the voxels that were
defined as cortical bone by the attenuation threshold.
These parameter values were calculated on a pQCT
high-resolution image of the diaphysis of the tibia with
an attenuation threshold value of 120 using the NIH
image analysis system (version 1.61).
M. Ito et al.
Annual Rates of Bone Loss
Linear regression analyses were performed to obtain
annual rates of bone loss (%/year) on the BMD values at
the measured sites for each subject. For each site, the
measured values at the start and at the first and second
12-month periods were used to calculate the slope of the
regression line. The slope value was then divided by the
value at the start and multiplied by 100, estimating
changes in the percent values per year (%/year) [15–18].
Factors Related to the Baseline Values of BMDs and
Cortical Geometry of the Tibia in Premenopausal and
Menopausal Women
In the players, the total years of participation in regular
exercise including past periods of activity and playing
volleyball (TYE), current weekly training hours for
volleyball, YSM and BMI were selected for analyses. In
the controls, YSM and BMI were selected. Menopausal
women included both peri- and postmenopausal women.
Their relationships with the values of the initial BMDs,
cortical area and the moment of inertia of the tibia were
estimated by multiple regression analyses in the
premenopausal players.
Factors Related to Annual Rates of Bone Loss in the
Menopausal Women
In the players, TYE, current weekly training hours for
volleyball, the baseline value of bone measurements,
BMI and YSM were used to estimate the relationship
with the annual rates of loss of the measured bone mass
by multiple regression analyses in the peri- and
postmenopausal players. In the controls, the baseline
values of bone measurements, BMI and YSM were
examined in relation to the annual loss rates. Menopau-
sal women included both peri- and postmenopausal
women for these analyses.
Statistics
Data were expressed as the mean and standard error of
the mean (SEM). Statistical analyses were performed
using the StatView version 4.51.2 statistical package.
Linear regression analyses were used to obtain the
annual rates of bone loss (%/year). Differences in
baseline BMDs, geometric parameters and annual rates
of bone loss between the players and controls were
compared by Student’s t-test for each menstrual group,
Multiple regression analyses were used to obtain
regression coefficients, t values and the significance
level, to determine significant factors affecting the
baseline BMDs, geometry and annual rates of bone
loss. The significance level was set at less than 0.05.
Effects of Lifetime Volleyball Exercise on BMD 107

Results between the players and controls in either menstrual

Baseline Values of Bone Density and Geometry
Baseline BMD values of the lumbar spine, calcaneus and
tibia in the player groups were higher than the respective
values in the control groups (Table 2). The values of the
lumbar area or calcaneus area in DXA measurements did
not differ between the players and controls (Table 3). In
the tibia, the total cross-sectional area or medullary
cavity area did not differ significantly either. However,
the values of the tibia cortical area and cross-sectional
moment of inertia in the players were significantly
higher than the values in the controls.
Annual Loss rates for the Densities of the Lumbar
Spine, Calcaneus and Tibia
Mean percentage values of the annual loss rates in the
lumbar spine BMDs, did not differ significantly
group. In the premenstrual group, the respective values
for the players and control women were 70.05 and
0.94 (Table 4). The values were 2.74 and 2.87 in the
perimenopausal group, and 2.29 and 2.75 in the
postmenopausal group.
In the calcaneus and tibia, however, the bone loss
rates for players were significantly smaller than the
values for the controls in all menstrual groups. In the
calcaneus, the respective values for the players and
controls were 70.04 and 0.93 (p50.05) in the
premenopausal group. They were 0.83 and 1.97
(p50.05) in the perimenopausal group, and 0.62 and
2.29 (p50.0001) in the postmenopausal group. In the
tibia, the respective values for the players and controls
were 70.45 and 1.16 (p50.0001) in the premenopausal
group, 1.49 and 2.85 in the perimenopausal group
(p50.05) and 1.10 and 2.24 (p50.005) in the
postmenopausal group.

Table 2. Baseliine BMD values of lumbar spine, calcaneus and tibial diaphysis of the volleyball players and the controls

n Lumbar spine BMD Calcaneus BMD Tibial diaphyseal BMD

(g/cm2) (g/cm2) (pQCT) (mg/cm3)

Volleyball players
Pre. 13 1.330+0.042 0.920+0.020 1096.8+26.2
Peri. 7 1.311+0.046 0.901+0.024 1118.9+37.2
Post. 20 1.253+0.048 0.858+0.016 1051.0+30.6
Controls
Pre. 13 1.202+0.037 0.846+0.023 976.2+42.1
Peri. 11 1.121+0.084 0.791+0.021 1023.8+21.6
Post. 35 0.950+0.022 0.740+0.014 939.7+24.6
Comparisona
Pre. 0.05 0.05 0.05
Peri. 0.05 0.005 0.05
Post. 0.0001 0.0001 0.01

Values are the mean and the standard error of the mean (SEM) with significant difference by Student’s t-test.
Pre., premenopausal women; Peri., perimenopausal women; Post., postmenopausal women.
a
.Data of comparison: values for the significance by Student’s t-test, compared the values of the players and controls of the same menstrual status.

Table 3. Baseline values of bone geometries in the volleyball players and the controls

n Spinal area Calcaneal area Tibial cross- Tibial medullary Tibial cortical Moment of inertia
sectional area cavity area area

Volleyball player
Pre. 13 43.0+3.4 11.04+0.48 7942+246 3578+220 4363+109 0.854+0.143
Peri. 7 42.5+2.1 10.18+1.59 7728+327 3762+319 3966+94 0.789+0.162
Post. 20 40.7+7.5 11.24+0.37 7390+264 3374+211 4016+88 0.730+0.027
Controls
Pre. 13 43.0+2.9 10.74+0.36 7489+244 3913+207 3577+96 0.702+0.120
Peri. 11 42.3+4.6 10.38+0.51 6828+230 3358+609 3470+85 0.592+0.039
Post. 35 41.7+5.6 10.58+0.27 7009+157 3786+149 3223+78 0.606+0.021
Comparisona
Pre. ns ns ns ns 0.0001 0.01
Peri. ns ns ns ns 0.005 0.05
Post. ns ns ns ns 0.0001 0.001

Values are the mean and the standard error of the mean (SEM) with significant difference by Student’s t-test.
Pre., premenopausal women; Peri., perimenopausal women; Post., postmenopausal women.
a
.Data of comparison: values for the significance by Student’s t-test, compared the values of the players and controls of the same menstrual status.
108 M. Ito et al.

Table 4. Annual rates of bone loss of the volleyball players and the controls

n Lumbar spine BMD Calcaneus BMD Tibial diaphyseal BMD

(g/cm2) (g/cm2) (pQCT) (%/year)
(mg/cm3)

Volleyball players
Pre. 13 70.05+0.51 70.04+0.31 70.45+0.13
Peri. 7 72.74+0.88 70.83+0.54 71.49+0.67
Post. 20 72.29+0.34 70.62+0.26 71.10+0.18
Controls
Pre. 13 70.94+0.31 70.93+0.34 71.16+0.27
Peri. 11 72.87+0.98 71.97+0.16 72.85+0.26
Post. 35 72.75+0.57 72.29+0.27 72.24+0.26
Comparisona
Pre. 7ns 70.05 70.0001
Peri. 7ns 70.05 70.05
Post. 7ns 70.0001 70.005

Values are the mean and the standard error of the mean (SEM) with significant difference by Student’s t-test.
Pre., premenopausal women; Peri., perimenopausal women; Post., postmenopausal women.
a
.Data of comparison: values for the significance by Student’s t-test, compared the values of the players and controls of the same menstrual status.

Factors Related to the Baseline BMDs and Geometry
Values in the Players and Control Women of
Premenopausal and Menopausal Status
In premenopausal players, only tibial cortical area
was significantly related to BMI (t value 2.763, r =
0.71, p50.05). In menopausal players, however, the
baseline values of the lumbar and calcaneus BMDs
were related to TYE, BMI, and YSM (Table 5). Tibia
BMD values were related to BMI and YSM. Tibia
cortical area and moment of inertia values were
related to TYE. The cortical area was also related to
BMI,
In premenopausal control women, there was no
significant relation among these parameters. In meno-
pausal controls, BMDs of the lumbar spine and tibia and
the tibial cortical area were significantly related to YSM
(Table 6). Calcaneus BMD and the values of the tibia
cortical area and moment of inertia were significantly
related to BMI.

Table 5. Factors related to baseline values of BMDs and geometry in the menopausal volleyball players

Dependent variables Independent variables Coefficient t value p r

Lumbar spine BMD TYE (years) 7 0.01+0.01 73.285 0.005 0.71

Current training (h/week) 7 0.02+0.02 70.630 ns
BMI (kg/cm2) 7 3.18+1.21 72.626 0.05
YSM (years) 70.04+0.01 73.463 0.005
Calcaneus BMD TYE (years) 7 0.01+0.01 72.060 0.05 0.72
Current training (h/week) 7 0.01+0.01 70.157 ns
BMI (kg/cm2) 7 1.40+0.43 73.252 0.005
YSM (years) 70.01+0.01 73.975 0.001
Tibial diaphyseal TYE (years) 7 1.40+2.85 70.492 ns 0.50
BMD (pQCT) Current training (h/week) 7 3.29+16.80 70.196 ns
BMI (kg/cm2) 1883.4+883.5 72.132 0.05
YSM (years) 716.27+7.46 72.181 0.05
Tibial cortical area TYE (years) 714.79+8.11 72.123 0.05 0.49
Current training (h/week) 7 8.64+47.9 70.180 ns
BMI (kg/cm2) 4790.4+251.8 72.202 0.05
YSM (years) 712.25+21.27 70.576 ns
Moment of inertia TYE (years) 70.01+0.01 71.945 0.05 0.42
Current training (h/week) 70.02+0.02 70.888 ns
BMI (kg/cm2) 70.57+0.99 70.572 ns
YSM (years) 70.01+0.01 70.206 ns

Multiple regression analyses were performed with BMD or geometrical parameters as the dependent variables, and TYE, current training (h/
week), BMI and YSM as the independent variables.
Coefficient, the mean and the standard error of the mean (SEM) for multiple regression analyses; p, values for the significant level; r, regression
coefficients, TYE, the total years of exercise, BMI, body mass iindex; YSM, years since menopause.
Effects of Lifetime Volleyball Exercise on BMD 109

Table 6. Factors related to baseline values of BMDs and geometry in the menopausal controls

Dependent variables Independent variables Coefficient t value p r

Lumbar spine BMD BMI (kg/cm2) 7 1.77+0.91 71.939 ns 0.43

YSM (years) 70.02+0.01 72.298 0.05
Calcaneus BMD BMI (kg/cm2) 7 1.60+0.55 72.926 0.01 0.47
YSM (years) 70.01+0.01 71.426 ns
Tibial diaphyseal BMI (kg/cm2) 825.5+1022.17 70.808 ns 0.42
BMD (pQCT) YSM (years) 718.45+7.14 72.585 0.05
Tibial cortical area TBMI (kg/cm2) 6952.5+3198.2 72.174 0.05 0.43
YSM (years) 747.81+22.33 72.141 0.05
Moment of inertia BMI (kg/cm2) 72.56+0.84 73.031 0.005 0.48
YSM (years) 70.01+0.01 70.901 ns

Multiple regression analyses were performed with BMD or geometrical parameters as the dependent variable and TYE, current training (h/week),
BMI and YSM as the independent variables.
Coefficient, the mean and the standard error of the mean (SEM) for multiple regression analyses; p, values for the significant level; r, regression
coefficients, BMI, body mass iindex; YSM, years since menopause.

Table 7. Factors related to the annual rates of bone loss of the lumbar spine, calcaneus and tibial diaphysis in the menopausal volleyball players

Dependent variables Independent variables Coefficient t value p r

Calcaneus BMD TYE (years) 7 0.05+0.03 71.612 ns 0.64

Current training (h/week) 72.08+0.10 71.886 0.05
BMI (kg/cm2) 710.39+10.21 71.018 ns
YSM (years) 70.20+0.09 72.143 0.05
Baseline BMD of calcaneus 72.87+4.15 70.693 ns
(g/cm2)
Tibial diaphyseal TYE (years) 7 0.01+0.03 70.330 ns 0.47
BMD (pQCT) Current training (h/week) 7 0.06+0.15 70.371 ns
BMI (kg/cm2) 716.63+8.87 71.875 0.05
YSM (years) 70.03+0.08 70.300 ns
Baseline BMD of tibial 70.01+0.01 70.375 ns
diaphyseal BMD (mg/cm3)

Multiple regression analyses were performed with BMD or geometrical parameters as the dependent variables, and TYE, current training (h/
week), BMI, YSM and BMD at the initial measurement as the independent variables.
Coefficient, the mean and the standard error of the mean (SEM) for multiple regression analyses; p, values for the significant level; r, regression
coefficients, TYE, the total years of exercise, BMI, body mass iindex; YSM, years since menopause.

Factors Related to the Annual Bone Loss Rates in the
Menopausal Players and Control Women
In the players, the rates of bone loss in the lumbar spine
did not significantly relate to any of the factors
examined. The rate of bone loss in the calcaneus BMD
was significantly related to the current number of
training hours per week and YSM (Table 7). Values of
tibia BMD were related to BMI. The annual loss rates
did not significantly relate to baseline BMD values in
either measurement site. In the control women, none of
the factors was related to bone loss rate at any site.
Discussion
This study has demonstrated that female habitual
volleyball players participating in sports activity for 20
years on average obtained higher bone mass in the
lumbar spine, calcaneus and tibia than women who did
not participate in habitual exercise during adulthood. In
the volleyball players, the menopause-related increases
in the annual rates of bone loss in the calcaneus and tibia
were significantly alleviated, but the effect was not
significant in the lumbar spine.
The baseline BMD values of the measured sites were
apparently associated with BMI in the menopausal
women. In the players, the lumbar spine, calcaneus and
tibia BMDs were related to BMI. It has been well
recognized that lumbar BMD is associated with BMI in
women before and after menopause [12,16,19,20]. We
previously observed that BMI was related to lumbar
BMDs in a study of 67 menopausal women without
habitual exercise [11]. Thus, BMI seems to be a
determinant factor of lumbar spine, tibia and calcaneus
BMD in women participating in exercise as well as in
women without habitual exercise. However, since only
calcaneus BMD was related to BMI in the control
women in this study, the association could be stronger in
the players than in control women. Though part of this
association may be related to the influence of BMI on
peak bone mass, the effect of BMI on bone mass seems
to dominate even after menopause on BMDs in the trunk
and lower extremities. The values of tibia cortical area
110
were also related in the players and controls. Thus, it is
expected that the cortical bone architecture also depends
on BMI in menopausal women. The values of lumbar
spine, calcaneus and tibia BMDs were negatively
associated with YSM in the players, as often reported
in postmenopausal women [21–23].
The positive effects of the total years of participation
in exercise (TYE) were obvious on the baseline values of
lumbar spine and calcaneus BMDs in the players. The
data support the hypothesis that habitual exercise during
adulthood is effective in obtaining higher bone mass in
the lumbar spine and calcaneus. In the tibia, while the t
value for BMD was not significant, the values of the
cortical area and the moment of inertia were significantly
related to TYE, thus indicating that the diaphysis of the
tibia responded to exercise by improving the cross-
sectional cortical structure during adulthood. Since the
total cross-sectional areas were similar in the players and
control women, the periosteal surface is apparently less
sensitive to exercise than the endosteal surface in our
subjects. Since TYE and BMI were both significantly
related to BMDs of the lumbar spine and calcaneus and
the parameters of the geometry of tibia as well, history of
the exercise activity and BMI seem to affect these bone
parameters independently.
In women without habitual exercise, postmenopausal
bone loss rates increased in both the lumbar spine
(11,15,22–24] and calcaneus [11,25], as has often been
observed. Increases in the bone loss rate for the
diaphysis of tibia by pQCT measurements were also
confirmed in a previous report [11]. However, the annual
loss rates of lumbar BMD in the peri- and postmeno-
pausal control women in our study were larger than the
values previously reported for Japanese women [11,22].
In our previous study, the annual loss rates of the lumbar
BMD were 2.59%/year on average of perimenopausal
women and 1.88%/year for early postmenopausal
women [11]. The rates of bone loss in lumbar BMD
determined by a nationwide study in Japan were 1.91%/
year for perimenopausal women and 0.99%/year for
postmenopausal women [22]. The rates in Caucasian
women were 2.35%/year and 1.24%/year for peri- and
postmenopausal women, respectively [15]. Since the
annual bone loss rates of the calcaneus and tibia in the
controls were similar to the level in our previous report
and those of others [11,22], the controls in this study
may represent the loss rates in the lumbar spine of fast
bone losers. Baseline values for the lumbar BMD in
control women were above the level of osteoporosis in
this study. However, multiple regression analyses did not
indicate any relationship with factors including baseline
BMD values. Thus, the annual loss rates do not seem to
depend on the baseline BMD values in these women.
In the habitual volleyball players, the alleviation of
menopause-related bone loss was apparent in the BMDs
of the calcaneus and tibia when compared with women
without habitual exercise. Reportedly, resistance ex-
ercise including jumping and jogging was not able to
reduce lumbar bone loss in postmenopausal women
whose bone mass was above the osteoporotic level [8–
M. Ito et al.
11]. These reported data along with ours suggest that
impact loading by habitual volleyball exercise specifi-
cally counteracts bone loss in the calcaneus and tibia but
does not compensate for the estrogen-dependent bone
loss that is dominant in the lumbar spine.
This study has two limitations regarding extrapolating
the results to postmenopausal women with osteoporosis:
one is the selection of subjects with bone mass higher
than the level of osteoporosis and the other is the lack of
hip BMD data. In this study we selected the lumbar spine
and calcaneus as sites with predominantly trabecular
bone and the diaphysis of the tibia as a site with
predominantly cortical bone. Since the proximal femur is
assumed to contain approximately similar amounts of
trabecular and corticla bone in normal subjects [26], we
did not measure hip BMD. However, hip BMD has been
known to be a good predictor of fragility fracture [27].
Further, it may be expected that the loading pattern of
the skeleton differs markedly between healthy bone and
osteoporotic bone whose microarchitecture has deterio-
rated [28]. Therefore, this study does not necessarily
underscore the effect of impact exercise in the
prevention of fragility fracture in osteoporosis. To
resolve that question, a different study design is
necessary.
In conclusion, this study indicated that the total years
of participating in exercise activity including volleyball
during adulthood have a positive effect on the bone mass
of the lumbar spine and calcaneus and on tibia cortical
structure. Habitual volleyball exercise was apparently
effective in alleviating menopause-related bone loss in
the tibia and calcaneus, but not in the lumbar spine. The
effect on the tibia was BMI-dependent in the players.
Acknowledgement. This study was supported by a Grant-in-Aid for
Scientific Research from the Ministry of Education in Japan.
References
1. Pocock N, Eisman J, Gwinn T, et al. Muscle strength, physical
fitness, and weight but not age predict femoral neck bone mass. J
Bone Miner Res 1989;4:441–8.
2. Bevier WC, Wiswell RA, Pyka GA, Kozak KC, Newhall KM,
Marcus R. Relationship of body composition, muscle strength,
and aeroboic capacity to bne mineral density in older men and
women. J Bone Miner Res 1989;4:515–22.
3. Snow-Harter C, Bouzsein ML, Lewis BT, Carter DR, Marcus R.
Effects of resistance and endurance exercise on bone mineral
status of young men: a randomized exercise intervention trial. J
Bone Miner Res 1992;7:761–9.
4. Bass S, Peace G, Bradney M, et al. Exercise before puberty may
confer residual benefits in bone density in adulthood: studies in
active prepubertal and retired female gymnasts. J Bone Miner Res
1998;13:500–7.
5. Lohman T, Going S, Pamenter R, et al. Effect of resistance
training on regional and total bone mineral density in
premenopausal women: a randomized prospective study. J Bone
Miner Res 1995;10:1015–24.
6. Kohrt WM, Snead DB, Slatopolosky E, Birge SJ. Additive effects
of weight-bearing exercise and estrogen on bone mineral density
in older women. J Bone Miner Res 1995;10:1303–11.
7. Kohrt WM,Ehsani AA, Birge SJ. Effects of exercise involving
Effects of Lifetime Volleyball Exercise on BMD
predominantly either joint-reaction or ground-reaction forces on
bone mineral density in older women. J Bone Miner Res
1997;12:1253–61.
8. Heinonen A, Oja P, Sievanen H, Pasanen M, Vuori I. Effect of
two training regimens on bone mineral density in healthy
perimenopausal women: a randomized controlled trial. J Bone
Miner Res 1998;13:483–90.
9. Alfredson H, Nordstrom P, Lorentzen R. Bone mass in female
volleyball players: a comparison of total and regional bone mass
in female volleyball players and nonactive females. Calcif Tissue
Int 1997;60:338–42.
10. Bassey EJ, Rothwell MC, Littlewood JJ, Pye DW. Pre- and post-
menopausal women have different bone mineral density
responses to the same high-impact exercise. J Bone Miner Res
1998;13:1805–13.
11. 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.
12. Ito M, Nakamura T, Tsurusaki K, Uetani M, Hayashi K. Effects
of menopause on age-dependent bone loss in axial and
appendicular skeletons in healthy japanese women. Osteoporos
Int 1999;10:377–87.
13. Ito M, Tsurusaki K, Hayashi K. Peripheral QCT for the diagnosis
of osteoporosis. Osteoporos Int 1997;7(Suppl 3):S120–7.
14. Rugsegger P. The use of peripheral QCT in the evaluation of bone
remodeling. Endocrinologist 1994;4:167–76.
15. Pouilles JM, Tremollières R, Riot C. The effects of menopause on
longitudinal bone loss from the spine. Calcif Tissue Int
1993;52:340–3.
16. Tremollières FA, Pouikkes J, Ribot C. Vertebral postmenopausal
bone loss is reduced in overweight women: a longitudinal study in
155 early postmenopausal women. J Clin Endocrinol Metab
1993;77:683–6.
17. Riis BJ, Hansen MA, Jensen AM, Overgaard JK, Christiansen C.
Low bone mass and fast rate of bone loss at menopause: equal
risk factors for future fracture. A 15-year follow-up study. Bone
1996;19:9–12.
111
18. Riggs BL, Wahner HW, Melton LJ III, Richelson LS, Judd HL,
Offord KP. Rates of bone loss in the appendicular and axial
skeletons of women. J Clin Invest 1986;77:1487–91.
19. Ortolani S, Trevisan C, Bianchi ML, Gandolini G, Cherubini R,
Polli EE. Influence of body parameters on female peak bone mass
and bone loss. Osteoporos Int 1993;(Suppl 1):S61–66.
20. Harris S, Dallal GE, Dawson-Hughes B. Influence of body weight
in rates of change in bone density of the spine, hip, and radius in
postmenopausal women. Calcif Tissue Int 1992;50:19–23.
21. Ashizawa N, Nonaka K, Michikami S, et al. Tomographical
description of tennis-loaded radius: reciprocal relation between
bone size and volumetric BMD. J Appl Physiol 1999;86:1347–51.
22. Fujiwara S, Fukunaga M, Nakamura T, et al. Rates of changes in
spinal bone density among Japanese women. Calcif Tissue Int
1998;63:202–7.
23. Luckey MM, Wallenstein S, Lapinski R, Meier DE. A
prospective study of bone loss in African-American and white
women – a clinical research center study. J Clin Endocrinol
Metab 1996;81:2948–56.
24. Pouilles JM, Tremollières F, Ribot C. Effect of menopause on
femoral and vertebral bone loss. J Bone Miner Res
1995;10:1531–6.
25. Ensrud KE, Palermo L, Black DM, et al. Hip and calcaneal bone
loss increase with advancing age: longitudinal results from the
study of osteoporotic fractures. J Bone Miner Res 1995;10:1778–
87.
26. Kuiper JW, Kuijk CV, Grashuis JL. Distribution of trabecular and
cortical bone related to geometry. A quantitative computed
tomography study of the femoral neck. Invest Radiol 1997;32:83–
9.
27. De Laet CEDH, Van Hout BA, Burger H, Weel AEAM, Hofman
A, Pols HAP. Hip fracture prediction in elderly men and women:
validation in the Rotterdam study. J Bone Miner Res
1998;13:1587–93.
28. Hayes WC, Myers ER, Robinovitch SN, Kroonenberg AVD,
Courtney AC, McMahon TA. Etiology and prevention of age-
related hip fractures. Bone 1996;18:S77–86.
Received for publication 3 September 1999
Accepted in revised form 10 July 2000