Content uploaded by Yukihiro Ikeda
Author content
All content in this area was uploaded by Yukihiro Ikeda on Feb 22, 2018
Content may be subject to copyright.
Nutritional Epidemiology
Intake of Fermented Soybeans, Natto, Is Associated with Reduced Bone Loss
in Postmenopausal Women: Japanese Population-Based Osteoporosis
(JPOS) Study
1
Yukihiro Ikeda,*
2
Masayuki Iki,* Akemi Morita,* Etsuko Kajita,
y
Sadanobu Kagamimori,**
Yoshiko Kagawa,
z
and Hideo Yoneshima
yy
*
JPOS Study Group, Department of Public Health, Kinki University School of Medicine, 377-2 Oono-Higashi,
Osaka-Sayama, Osaka, Japan;
y
Department of Public Health and Home Nursing, Nagoya University School
of Health Sciences, Nagoya, Aichi, Japan;
**
Department of Welfare Promotion and Epidemiology,
Faculty of Medicine, University of Toyama, Toyama, Japan;
z
Kagawa Nutrition University, Tokyo, Japan;
and
yy
Syuuwa Sougoh Hospital, Kasukabe, Japan
ABSTRACT Japanese fermented soybeans (natto in Japanese), which contain a large amount of menaquinone-7,
may help prevent the development of osteoporosis. We assessed the possibility of an association between habitual
natto intake and bone mineral density (BMD) and BMD change over time in healthy Japanese women who
participated in a large representative cohort study (Japanese Population-based Osteoporosis Study: JPOS study).
The BMD was measured at the spine, hip, and forearm in 944 women (20–79 y old) at baseline and at a follow-up
conducted 3 y later. Dietary natto intake was assessed by a FFQ on both occasions. Additional covariates including
age, height, weight, lifestyle factors, dietary calcium intake, and the intake of other soybean products, were also
measured. The total hip BMD at baseline increased (Pfor trend ¼0.0034) with increasing habitual natto intake in the
postmenopausal women, although this was not the case at other skeletal sites. There were significant positive
associations between natto intake and the rates of changes in BMD at the femoral neck (P,0.0001) and at the distal
third of the radius (P¼0.0002) in the postmenopausal women. The association in the femoral neck persisted even
after adjusting for covariates. No significant association was observed between the intake of tofu or other soybean
products and the rate of BMD change in the postmenopausal women. Natto intake may help prevent postmeno-
pausal bone loss through the effects of menaquinone 7 or bioavailable isoflavones, which are more abundant in natto
than in other soybean products. J. Nutr. 136: 1323–1328, 2006.
KEY WORDS: fermented soybean menaquinone isoflavone bone density Japanese women
Osteoporosis is one of the most important diseases affecting
elderly people worldwide (1). Hip fractures resulting from
osteoporosis cause many elderly people to become bedridden,
and osteoporosis is a great burden on society today. In the
prevention of osteoporosis, nutrition plays a key role in addition
to other factors such as genetics, physical activity, cigarette
smoking, and alcohol intake (1–4). Many studies have assessed
the role of calcium, vitamin D, magnesium, and other micro-
nutrients, as well as macronutrients such as protein, in the
prevention of osteoporosis (3–6). There is emerging evidence,
however, that vitamin K may also play a protective role against
age-related bone loss (7–9).
Vitamin K is a cofactor of g-carboxylase, which mediates the
conversion of undercarboxylated osteocalcin to carboxylated
osteocalcin by transforming the glutamyl residues of osteocalcin
into carboxyglutamic acid residues; the last-mentioned have a
high affinity for calcium ions in hydroxyapatite and regulate the
growth of these crystals (4). Vitamin K is classified into 2
groups: phylloquinone, which is the major form and exists in
plants, and menaquinone, which is synthesized by bacteria and
exists in dairy products.
Natto is a kind of fermented soybean that is consumed very
widely and frequently in Japan. This food is rich in menaquinone-7
and contains .100 times more menaquinone-7 than various
kinds of cheese (10).
The incidence of hip fracture in Japan is lower than in
European countries and the United States. The reason for this
difference is not known, but it is thought to be due to the
differences in dietary (11), bone-geometric (12), and lifestyle
factors (13) between these populations. Natto is one of the
candidates that may account for this difference in hip fracture
rate. Furthermore, there is some evidence suggesting that natto
is effective in maintaining bone stiffness (10), increasing serum
levels of menaquinone-7 and g-carboxylated osteocalcin (14),
and maintaining the bone mineral density (BMD) of middle-
aged women (15). However, these studies were conducted
1
Supported by the Japan Milk Promotion Board, the Japan Dairy Council
(1995, 1996), and the Japan Society for the Promotion of Science (Grant-in-Aid
for Scientific Research (B) 10470114, 1998–2000).
2
To whom correspondence should be addressed. E-mail: yuyu@med.
kindai.ac.jp.
0022-3166/06 $8.00 Ó2006 American Society for Nutrition.
Manuscript received 12 August 2005. Initial review completed 16 September 2005. Revision accepted 8 February 2006.
1323
Downloaded from https://academic.oup.com/jn/article-abstract/136/5/1323/4669999
by guest
on 22 February 2018
among subjects with a limited age distribution and selected
from limited areas (14,15); the sample sizes were small (14) and
the studies were performed cross sectionally (10,14,15).
The purpose of the present study was to assess whether
habitual natto intake has a positive association with BMD and
with change in BMD over time in healthy Japanese women of
various ages who participated in a large representative cohort
study.
SUBJECTS AND METHODS
Subjects. The present study was conducted as a part of a larger
study (Japanese Population-based Osteoporosis Study) (16), and in-
volved 3 cohorts, 1 from the northeastern part of Honshu island (Nishiaizu
Town, Fukushima Prefecture), 1 from Shikoku island (Sanuki City,
Kagawa Prefecture), and 1 from Miyako island in the subtropical zone
(Miyakojima City, Okinawa Prefecture). We recruited healthy women
aged 20–79 y, randomly selected from the resident registry of each
municipality. Of the 1628 women who participated in the baseline
study, 1280 participated in the follow-up study 3 y later. The subjects
were interviewed and were excluded if they reported current in-
volvement or history of any disease that might affect bone metabolism,
such as amenorrhea, oligomenorrhea, bilateral oophorectomy, diseases
of the parathyroid gland, rheumatoid arthritis, or diabetes mellitus.
Subjects with abnormal serum calcium, inorganic phosphorus, and
parathyroid hormone concentrations in the present study were also
excluded. Those who were being administered drugs affecting bone
turnover, such as calcium, estrogens, vitamin D, calcitonin, or gluco-
corticoids, as determined through the interviews conducted in the
present study, were also excluded. According to the menstrual infor-
mation obtained from the interviews, subjects who menstruated reg-
ularly at baseline were judged to be premenopausal and those who had
entered menopause at least 6 mo before the survey were classified as
postmenopausal. Subjects with hysterectomy-induced menopause or
whose age at menopause could not be determined were classified as
postmenopausal women if they were .57 y old at the time of the
survey. In Japan, 97.5% of women $57 y old are reported to be
postmenopausal (17). The remaining subjects were classified as
perimenopausal and were excluded from further analysis. In total,
336 women were excluded from further analysis; the remaining 944
women served as the representative sample of Japanese women free
from apparent abnormalities affecting bone mass.
We obtained written informed consent for all study procedures
from each subject. The study protocol was approved by the Ethical
Committee of the Kinki University School of Medicine.
Bone mass. The BMD was measured at baseline and 3 y
thereafter by dual-energy X-ray absorptiometry at the lumber spine
(L
2–4
), the right hip (QDR 4500A, Hologic), and at the distal third of
the radius (pDXA, Norland/Stratec). The short-term CV of the BMD
measurement in vivo, as calculated on the basis of 5 measurements on
different days for each of 5 volunteers, were 1.2, 1.6, 1.2, and 1.2% for
the spine, femoral neck, total hip, and distal third of the radius,
respectively (16). The BMD of the spine and forearm phantoms were
measured every day for quality assurance; they did not drift signifi-
cantly during the study period.
Body size. The height (cm) and weight (kg) of the subjects were
measured with an automatic scale (TK-11868h, Takei Kagaku). The
BMI (kg/m
2
) was calculated.
Interviews. Detailed interviews were conducted during the sur-
vey by trained nurses both at baseline and at follow-up. The interviews
were conducted to confirm the information given on a questionnaire
that was delivered to the subjects by mail and completed beforehand.
The questionnaire included points on menstrual history and on the
past history and present involvement of gynecological and other dis-
eases or medications that may affect bone metabolism.
Information on the habitual intake of natto, bean curd (tofu), and
other soybean products was obtained through an interview conducted
by trained dietitians. We determined the portion size and weekly
frequency of intake using a FFQ that was developed and validated
previously to estimate dietary calcium intake (18). The FFQ was given
to the subjects during the baseline and follow-up surveys. Natto is sold
in a pack that usually contains 40 g of natto, a quantity considered to
be suitable for 1 meal in Japan. One pack of natto contains 35 mg of
calcium and 350 mg menaquinone-7 (19). Natto intake can be deter-
mined quite precisely by counting the number of natto packs consumed
by the subject over a period of 1 wk. The usual portion size of tofu or
other soybean products was determined through the interview by
comparison with the portion size of the reference food samples. The
total calcium intake, which included calcium intake from mineral
supplements, was estimated on the basis of the same FFQ.
Statistical data analysis. The subjects were divided into premen-
opausal and postmenopausal groups; each of these 2 groups was then
classified into 3 subgroups according to their mean habitual intake of
natto at baseline or the mean intakes at baseline and at follow-up.
Because the premenopausal women did not consume large amounts of
natto compared with the postmenopausal women, we used different
intervals to classify the weekly natto intake for these 2 groups: no
intake, 1 or 2 packs, and ,2 packs for the premenopausal women, and
no intake, 1–4 packs, and .4 packs for the postmenopausal women.
The rate of BMD change was calculated as follows:
The rate of change 5f½ðBMD at follow up2BMD at baselineÞ=BMD
at baseline=3g3100:
The results of the indices are expressed as means and SD or SEM.
Student’s ttest was used to test the significance of the difference in
mean BMD and other continuous variables between 2 groups. For
comparisons among 3 groups with different habitual natto intake or
among 3 cohorts, ANOVA was performed and, if significant, pairwise
comparisons were done using ttests. The effects of confounding
variables, such as age, height, weight, smoking habits, physical activity,
years since menopause, and intake of calcium, tofu, and other soybean
products, were adjusted using analysis of covariance (ANCOVA) where
necessary. The linear regression analysis for BMD or its rate of change
was performed after introducing the ordinal variable for natto intake,
which was coded as 0, 1 or 2 according to the amount of habitual
natto intake. The significance for the regression coefficient for the
ordinal variable for natto intake was interpreted as the significance of
trend of BMD or its rate of change with the increase in natto intake.
The x
2
test was used to compare the prevalence rates of lifestyle fac-
tors, such as smoking, exercise, or milk consumption. Differences were
considered significant at P,0.05. Scheffe
´’s multiple comparison pro-
cedure was applied to adjust the level of statistical significance to 5%
when 3 groups were compared. All statistical analyses were performed
using the SASÒsystem for personal computers (release 8.2, SAS
Institute).
RESULTS
Characteristics of the subjects at baseline and at follow
up. The pre- and postmenopausal women differed in height
(P,0.0001) and BMI (P,0.0001) at baseline. Calcium
intake and the amount of natto, tofu, and other soybean prod-
ucts consumed by the postmenopausal women were higher
(P,0.0001) than in the premenopausal women at baseline
(Table 1). BMD at the spine, total hip, femoral neck, and distal
third decreased (P,0.0001) from baseline to follow-up in the
postmenopausal women. None of the variables listed in Table
1 differed among the 3 cohorts in either the premenopausal or
postmenopausal group.
Habitual natto intake and BMD at baseline. The BMD at
baseline and the potential confounding variables were com-
pared among the groups classified according to their menopau-
sal status and the habitual natto intake at baseline (Table 2).
Habitual natto intake was associated with age in the premen-
opausal women (Pfor trend 50.0019). The calcium intake
increased with increasing habitual natto intake in both the
premenopausal and the postmenopausal women (Pfor trend ,
0.0001). The intakes of other soybean products increased with
increasing habitual natto intake in the postmenopausal women
1324 IKEDA ET AL.
Downloaded from https://academic.oup.com/jn/article-abstract/136/5/1323/4669999
by guest
on 22 February 2018
(Pfor trend ,0.0001). In the postmenopausal women, the
total hip BMD increased with increasing habitual natto intake
(Pfor trend 50.0034), although no association existed for
other skeletal sites (Table 2).
Habitual natto intake and the rate of BMD change. The
premenopausal and postmenopausal women were grouped into
3 subgroups according to the mean habitual natto intake at
baseline and at follow-up. The postmenopausal women showed
a positive trend (P,0.0001) across subgroups in the rate of
femoral-neck BMD change (Table 3). Even after adjusting for
covariates such as age, height, weight, physical activity, smok-
ing habits, and years since menopause, as well as for the intake
of calcium, tofu, and other soybean products, these associations
persisted (Pfor trend 50.0094) (Table 4). The BMD change
at the distal third of the radius in the postmenopausal women
showed a positive trend (Pfor trend 50.0002) across cate-
gories of natto intake (Table 3). However, the association was
not significant after adjusting for the covariates (Table 4). A
similar result was obtained for the postmenopausal women
grouped by age in 10-y increments. The rate of BMD change
was positively associated with increasing natto intake at the
femoral neck in the subjects in their 60s (Pfor trend 50.0019)
TABLE 1
Characteristics of pre- and postmenopausal women at baseline and at the 3-y follow-up
1
Premenopausal Postmenopausal
Baseline Follow-up Baseline Follow-up
n394 394 550 550
Age, y34.0 67.1 37.1 67.1* 64.2 68.4
#
67.2 68.4*
Age at menopause, y— — 49.5 63.3 49.5 63.3
Time since menopause, y— — 14.7 68.8 17.7 68.8*
Height, cm 156.8 65.1 156.9 65.1 148.5 65.3
#
148.0 65.6*
Weight, kg 53.6 67.7 53.7 68.1 52.9 68.1
#
51.7 68.4*
BMI, kg/m
2
21.8 63.0 21.8 63.1 24.0 63.3 23.6 63.4*
Dietary intakes
2
Calcium, mg/d 564 6211 574 6229 685 6260
#
699 6300
Natto,packs/wk 1.4 61.8 1.8 62.2* 2.0 62.6
#
2.1 62.7
Tofu, servings/wk 1.1 61.3 3.6 62.2* 1.5 61.6
#
5.1 63.5*
Other soybean products, servings/wk 0.8 61.5 1.3 61.7* 2.2 62.5
#
2.8 63.3*
Prevalence of smokers, %10.2 6.6* 2.4
#
1.8*
Prevalence of exercise, %32.1 32.6* 40.3
#
49.1*
Prevalence of milk intake, %71.3 78.7* 77.9 83.6*
BMD, g/cm
2
Spine 1.034 60.115 1.033 60.121 0.800 60.138
#
0.782 60.136*
Total hip 0.887 60.108 0.885 60.111 0.740 60.117
#
0.723 60.117*
Femoral neck 0.798 60.105 0.795 60.106 0.650 60.102
#
0.627 60.104*
Distal third of the radius 0.754 60.066 0.755 60.066 0.570 60.110
#
0.548 60.106*
1
Values are means 6SD or %. *Different from baseline, P,0.05;
#
different from premenopausal women at that time.
2
Dietary intake: natto (fermented soy beans, 40 g/pack); tofu (bean curd, 75 g/serving); other soy bean products: ground soybeans, 17.5 g/serving;
green soybeans, 140 g/serving; boiled soybeans, 45 g/serving; and milk intake (3800 mL/wk).
TABLE 2
Baseline anthropometry, dietary intakes and BMD in pre- and postmenopausal women stratified by habitual natto intake
1
Premenopausal women Pfor
trend
Postmenopausal women Pfor
trend
Natto intake, packs 0 1–2/wk .2/wk 0 1–4/wk .4/wk
n190 119 85 267 197 85
Age, y32.7 67.0
b
34.9 67.2
a
35.4 66.3
a
0.0019 64.9 68.5
a
62.9 68.4
b
64.7 68.3
a
0.3039
Height, cm 156.9 65.1 156.4 65.0 157.4 65.4 0.6147 148.7 65.3 148.2 65.4 148.5 65.4 0.5221
Weight, kg 53.5 67.2 53.2 67.8 54.4 68.6 0.4623 52.6 67.7 53.0 68.4 53.6 68.3 0.3262
BMI, kg/m
2
21.8 62.9 21.8 63.0 22.0 63.2 0.6428 23.8 63.2 24.1 63.4 24.3 63.5 0.1531
Calcium intake, mg/d 509 6191
c
570 6179
b
679 6247
a
,0.0001 594 6213
c
721 6248
b
892 6284
a
,0.0001
Tofu,
2
servings/wk 1.1 61.4 1.1 61.3 1.2 61.1 0.4674 1.5 61.7 1.5 61.5 1.7 61.4 0.5155
Other soy bean products,
3
servings/wk
0.7 61.3 0.9 61.7 1.0 61.6 0.0731 1.7 62.2
c
2.4 62.2
b
3.5 63.3
a
,0.0001
BMD at baseline, g/cm
2
Spine 1.034 60.110 1.031 60.114 1.039 60.126 0.8149 0.795 60.139 0.807 60.139 0.800 60.133 0.5779
Total hip 0.882 60.110 0.893 60.115 0.889 60.095 0.4778 0.724 60.113
b
0.752 60.114
a
0.760 60.134
a
0.0034
Femoral neck 0.796 60.104 0.801 60.112 0.801 60.095 0.6551 0.642 60.100 0.657 60.098 0.655 60.116 0.1481
Distal third of the radius 0.760 60.066 0.750 60.067 0.750 60.064 0.2059 0.567 60.112 0.567 60.111 0.564 60.103 0.7603
1
Values are means 6SD. Within a group, means in a row with superscripts without a common letter differ, P,0.05.
2
Tofu (bean curd 75 g/serving).
3
Other soybean products: ground soybeans, 17.5 g/serving; green soybeans, 140 g/serving; boiled soybeans, 45 g/serving.
1325JPOS STUDY
Downloaded from https://academic.oup.com/jn/article-abstract/136/5/1323/4669999
by guest
on 22 February 2018
and 70s (Pfor trend 50.0002) and at the distal third of the
radius in the subjects in their 50s (Pfor trend 50.0069) and
60s (Pfor trend 50.0002) (Table 5).
Natto intake and the rate of BMD change were not
associated at any skeletal site in the premenopausal women
(Table 3). Furthermore, the results did not change even when
the same classification for natto intake as that for postmeno-
pausal women was applied.
We also tested whether there was a relation between the
intake of tofu or other soybean products and the rate of BMD
change after adjusting for the effects of the covariates, but there
were no associations (Table 4).
DISCUSSION
The present study showed that habitual natto intake was
associated with reduced bone loss at the femoral neck and at
the distal third of the radius in postmenopausal women. An
ordinary 40-g pack of natto contains ;350 mg of menaquinone-
7, a remarkably high amount for a food. Menaquinone-7 may
have mediated the association found in the present study.
Two epidemiologic studies derived from the Framingham
cohort examined the association between phylloquinone intake
and BMD (20,21). In the Framingham Offspring Study (20),
greater dietary phylloquinone intake was associated with higher
BMD at the hip and spine in women aged 29–83 y (mean age:
58.9 y). However, there was no association between dietary
phylloquinone intake and femoral-neck BMD or its change
during the 4 y of follow-up in postmenopausal women aged 68–
94 y (mean age: 75.3 y) in the Framingham osteoporosis study
(21). The authors of these studies suspected that the lack of
association in the latter study was due to the relatively small
age-related bone loss because the subjects were older. The
present findings support this hypothesis because the significant
association was present both at the femoral neck and the distal
third of the radius in subjects in their 60s but only at the
femoral neck in subjects in their 70s. The mean annual BMD
change at the femoral neck in the subjects in their 70s, 1.4%,
was greater than that of women in their 60s, 1.2%. At the
distal third of the radius, however, the mean annual change in
BMD in the subjects in their 70s was 0.9%, which was smaller
than that of women in their 60s, 1.1%; this change was so
much less than the measurement error that the statistical power
for detecting a significant association in this age group was
reduced. A similar mechanism may have operated in the
Framingham osteoporosis study.
For Asian populations, several clinical trials were conducted
to investigate the effect of menaquinone in pharmacologic
doses on BMD (22,23) and fracture risk (24–26). Shiraki, et al.
(24) reported that 45 mg/d of menatetrenone maintained BMD
at the spine more effectively than the control regimen (0.4%
vs. 2.6% for 24 mo, respectively) and reduced the incidence
of fracture in patients with osteoporosis. That study showed
that menaquinone had a preventive effect on bone loss and
supports the findings of the present study. The menaquinone-7
content in an ordinary pack of natto is ,1% of the pharma-
cologic dose. Therefore, the preventive effects of dietary natto
intake on bone loss may be small compared with that of
menaquinone medication. The preventive effect of phylloqui-
none or menaquinones on fractures was also previously de-
scribed (14,21,27). These studies support our speculation that
the habitual intake of phylloquinone or menaquinones atten-
uates bone loss at the femoral neck, and may lead to reduction
in the incidence of hip fractures in postmenopausal women. We
should follow the present subjects further to assess whether
natto intake reduces the risk of hip fracture by maintaining the
BMD at the femoral neck.
TABLE 3
The rate of change in BMD in pre- and postmenopausal women stratified by habitual natto intake at baseline and follow-up
1
Premenopausal women Postmenopausal women
0 1–2/wk .2/wk Pfor trend 0 1–4/wk .4/wk Pfor trend
n129 144 73 210 226 81
%/y %/y
Spine 0.2 61.2 0.0 60.9 0.1 61.0 0.4914 0.7 61.5 0.7 61.7 0.2 61.3 0.0930
Total hip 0.1 60.9 0.0 61.1 0.1 61.1 0.9609 0.9 61.4 0.9 61.4 0.5 61.4 0.0651
Femoral neck 0.3 61.3 0.0 61.5 0.3 61.5 0.8394 1.6 61.9
c
1.2 61.8
b
0.5 62.1
a
,0.0001
Distal third of the radius 0.1 60.9 0.1 61.1 0.0 61.1 0.3177 1.4 61.7
b
1.4 61.8
b
0.3 61.5
a
0.0002
1
Values are means 6SD. Within a group, means in a row with superscripts without a common letter differ, P,0.05.
TABLE 4
The rate of change in BMD in postmenopausal women stratified
by habitual intakes of natto, tofu, and other soybean products at
baseline and at follow-up adjusted for other covariates
1
Soybean
products
Intake
classification
Adjusted mean rate of change in BMD
Femoral neck Distal third of radius
%/y
Natto
2
01.5 60.2
b
1.2 60.2
ab
1–4/wk 1.3 60.2
b
1.5 60.2
b
.4/wk 0.4 60.3
a
0.6 60.3
a
Pfor trend 0.0094 0.2261
Tofu
3
00.6 61.1 1.5 61.0
1–4/wk 1.2 60.1 1.2 60.1
.4/wk 1.5 60.3 1.4 60.3
Pfor trend 0.2990 0.3601
Other soybean
products
4
01.1 60.3 1.5 60.3
1–4/wk 1.4 60.1 1.2 60.1
.4/wk 0.8 60.3 0.9 60.3
Pfor trend 0.4320 0.2283
1
Values are means 6SEM. Means are adjusted by ANCOVA for the
covariates measured at baseline and follow-up, including age, height,
weight, calcium intake, exercise, smoke, years since menopause, and
intakes of natto, tofu, other soybean products. Within a group, means in
a column with superscripts without a common letter differ, P,0.05.
2
Natto (fermented soybeans 40 g/pack).
3
Tofu (bean curd 75 g/serving).
4
Other soy bean products: ground soybeans, 17.5 g/serving; green
soybeans, 140 g/serving; boiled soybeans, 45 g/serving.
1326 IKEDA ET AL.
Downloaded from https://academic.oup.com/jn/article-abstract/136/5/1323/4669999
by guest
on 22 February 2018
Isoflavones also may be effective components of natto. Natto
contains large amounts of isoflavones, which were reported to
reduce bone resorption through estrogenic mechanisms (28–
30). One portion of tofu also contains an amount of isoflavones
similar to that in 1 pack of natto. However, tofu intake was not
associated with bone loss prevention in our subjects. This may
be due to the difference in the types of isoflavones present in
natto and tofu. Isoflavone aglycones are absorbed faster in hu-
mans and are more bioavailable than isoflavone glycosides (31).
Fermented soy products contain larger amounts of aglycones
than other soy products (32–34). Morabito et al. (35) con-
ducted a randomized placebo-controlled trial to evaluate the
effect of genistein, one of the aglycones, on BMD in women
aged 47–57 y. The administration of 54 mg/d of genistein
significantly increased BMD at the femoral neck (genistein: 3.6
63.0% vs. placebo: 0.7 60.1%) after 1 y of treatment. In the
present study, only natto decreased bone loss. We assume, there-
fore, that this may be due to the greater amount of isoflavone
aglycone contained in natto in addition to menaquinone-7.
The present study had several advantages over previous
studies with respect to study design. It used a larger represen-
tative sample of the general population, the ages of the subjects
varied widely, and the follow-up rate was acceptable. However,
its limitations should be addressed. First, our FFQ listed only
calcium-rich foods. Therefore, we could not adjust the results
for potential confounding due to energy intake and protein
intake. Subjects who consumed a large amount of natto may
also have consumed large amounts of energy and protein, which
may have resulted in part in the protective effect on bone loss in
the present study. This possible confounding could be con-
trolled for through the adjustment for height and weight in the
present analysis, but may still have been present. Second, natto
contains large amounts of menaquinone-7, but we did not
examine the effects of any other food containing menaquinone-
7. Therefore, the results should be interpreted carefully and the
conclusion should be restricted to the effect of natto. Third, the
information on natto intake was obtained through interviews
based on a questionnaire, and there may have been individual
misclassifications of natto intake. Fourth, we did not observe
any association between tofu or other soybean products and
BMD in this study. One reason for this is that natto intake could
be determined easily and precisely through the FFQ (because
natto is sold in packs containing a similar amount of natto). This
information may yield a more accurate estimate of the con-
sumption of this soybean product compared with any others.
Fifth, the 3 study areas were not randomly selected from all
municipalities in Japan; thus, the subjects might not be rep-
resentative of the Japanese female population. However, there
were no differences in body size or the BMD at the skeletal sites
between the present subjects and the Japanese female popu-
lation. It is not likely, therefore, that the present findings
originated from a sampling bias.
The present representative cohort study suggests that natto
intake may decrease the loss of bone mass at the femoral neck
and possibly at the distal third of the radius in postmenopausal
women. Further studies on this topic should be designed to
allow for possible confounding effects due to protein and energy
intakes and to address whether the effect of menaquinone-7 or
of the isoflavones is the primary mechanism.
ACKNOWLEDGMENTS
This study was conducted by the JPOS Study Group comprising
F. Marumo, T. Matsuzaki, T. Yamagami, T. Matsukura, and H. Nishino
in addition to the authors.
LITERATURE CITED
1. Cooper C. Melton, LJ. Magnitude and impact of osteoporosis and
fractures. In: Marcus R, Feldman D, Kelsey JL, editors. Osteoporosis. Academic
Press: San Diego, CA; 1996. p. 409–34.
2. Ralston SH. Genetic control of susceptibility to osteoporosis. J Clin
Endocrinol Metab. 2002;87:2460–6.
3. Tucker KL. Dietary intake and bone status with aging. Curr Pharm Des.
2003;16:139–50.
4. Heaney RP. Nutrition and risk for osteoporosis. In: Marcus R, Feldman D,
Kelsey JL, editors. Osteoporosis. Academic Press: San Diego, CA; 1996. p. 483–
509
5. Nieves JW. Osteoporosis: the role of micronutrients. Am J Clin Nutr.
2005;81:S1232–9.
6. Weaver CM, Fleet JC. Vitamin D requirements: current and future. Am J
Clin Nutr. 2004;80:1735S–9.
7. Weber P. Vitamin K and bone health. Nutrition. 2001;17:880–7.
8. Zittermann A. Effects of vitamin K on calcium and bone metabolism. Curr
Opin Clin Nutr Metab Care. 2001;4:483–7.
9. Ryan-Harshman M, Aldoori W. Bone health. New role for vitamin K? Can
Fam Physician. 2004;50:993–7.
10. Katsuyama H, Ideguchi S, Fukunaga M, Saijoh K, Sunami S. Usual
dietary intake of fermented soybeans (natto) is associated with bone mineral
density in premenopausal women. J Nutr Sci Vitaminol (Tokyo). 2002;48:207–15.
11. Greendale GA, FitzGerald G, Huang MH, Sternfeld B, Gold E, Seeman T,
Sherman S, Sowers M. Dietary soy isoflavones and bone mineral density: results
from the study of women’s health across the nation. Am J Epidemiol.
2002;155:746–54.
12. Nakamura T, Turner CH, Yoshikawa T, Slemenda CW, Peacock M, Burr
DB, Mizuno Y, Orimo H, Ouchi Y, Johnston CC Jr. Do variations in hip geometry
TABLE 5
The rate of change in BMD in postmenopausal women stratified by age group and habitual natto intake
1
Age Natto intake classification nSpine Total hip Femoral neck Distal third of the radius
y %/y
50–59 0 1.0 61.4 0.8 61.2 1.6 61.6 1.9 61.7
b
1–4/wk 53 1.3 61.4 0.7 61.1 1.2 61.9 1.8 61.7
b
.4/wk 77 0.8 61.2 0.4 60.9 1.0 61.4 0.7 61.5
a
Pfor trend 26 0.9093 0.0842 0.0779 0.0069
60–69 0 0.6 61.3
b
0.6 60.9 1.4 61.6
b
1.4 61.8
b
1–4/wk 71 0.4 61.7 0.7 61.3 1.1 61.6 1.1 61.5
b
.4/wk 86 0.5 61.2
a
0.4 61.2 0.2 61.9
a
0.1 61.3
a
Pfor trend 28 0.0043 0.6486 0.0019 0.0002
70–79 0 0.4 61.6 1.1 61.8 1.8 62.3
b
1.0 61.7
1–4/wk 80 0.1 61.5 1.0 61.8 1.2 61.9 1.0 62.2
.4/wk 48 0.3 61.2 0.8 62.1 0.5 62.9
a
0.4 61.5
Pfor trend 25 0.4216 0.4268 0.0076 0.2081
1
Values are means 6SD. Within a group, means in a column with superscripts without a common letter differ, P,0.05.
1327JPOS STUDY
Downloaded from https://academic.oup.com/jn/article-abstract/136/5/1323/4669999
by guest
on 22 February 2018
explain differences in hip fracture risk between Japanese and white Americans?
J Bone Miner Res. 1994;9:1071–6.
13. Hirota T, Nara M, Ohguri M, Manago E, Hirota K. Effect of diet and lifestyle
on bone mass in Asian young women. Am J Clin Nutr. 1992;55:1168–73.
14. Kaneki M, Hedges SJ, Hosoi T, Fujiwara S, Lyons A, Crean SJ, Ishida N,
Nakagawa M, Takechi M, et al. Japanese fermented soybean food as the major
determinant of the large geographic difference in circulating levels of vitamin K2:
possible implications for hip-fracture risk. Nutrition. 2001;17:315–21.
15. Tsuchida K, Mizushima S, Toba M, Soda K. Dietary soybeans intakes and
bone mineral density among 995 middle-aged women in Yokohama. J Epidemiol.
1999;9:14–9.
16. Iki M, Kagamimori S, Kagawa Y, Matsuzaki T, Yoneshima H, Marumo F.
Bone mineral density of the spine, hip and distal forearm in representative samples
of the Japanese female population. Japanese Population-based Osteoporosis
(JPOS) Study. Osteoporos Int. 2001;12:529–37.
17. Kajita E, Iki M, Nishino H. Bone mineral density of the lumbar spine and its
relation to biological and lifestyle factors in middle-aged and aged Japanese
women. Jpn J Hyg. 1994;49:674–83.
18. Sato Y, Tamaki J, Kitayama F, Kusaka Y, Kodera Y, Koutani A, Iki M.
Development of a food-frequency questionnaire to measure the dietary calcium
intake of adult Japanese women. Tohoku J Exp Med. 2005;207:217–22.
19. Sakano T, Notsumoto S, Nagaoka T, Morimoto A, Fujimoto K, Matsuda S,
Suzuki Y. Hirauchi K. Measurement of K vitamins in food by high-performance
liquid chromatography with fluorometric detection. Vitamins (Kyoto). 1988;62:393–
8.
20. Booth SL, Broe KE, Gagnon DR, Tucker KL, Hannan MT, McLean RR,
Dawson-Hughes B, Wilson PW, Cupples LA, Kiel DP. Vitamin K intake and bone
mineral density in women and men. Am J Clin Nutr. 2003;77:512–6.
21. Booth SL, Tucker KL, Chen H, Hannan MT, Gagnon DR, Cupples LA,
Wilson PW, Ordovas J, Schaefer EJ, et al. Dietary vitamin K intakes are associated
with hip fracture but not with bone mineral density in elderly men and women.
Am J Clin Nutr. 2000;71:1201–8.
22. Orimo H, Shiraki M, Fujita T, Onomura T, Inoue T, Kushida K. Clinical
evaluation of menatetrenone in the treatment of involutional osteoporosis—a
double blind multicenter comparative study with 1-a-hydroxyvitamin D3. J Bone
Miner Res. 1992;7: suppl 1:S122.
23. Iwamoto J, Takeda T, Ichimura S. Effect of combined administration of
vitamin D3 and vitamin K2 on bone mineral density of the lumbar spine in
postmenopausal women with osteoporosis. J Orthop Sci. 2000;5:546–51.
24. Shiraki M, Shiraki Y, Aoki C, Miura M. Vitamin K2 (menatetrenone)
effectively prevents fractures and sustains lumbar bone mineral density in
osteoporosis. J Bone Miner Res. 2000;15:515–21.
25. Iwamoto J, Takeda T, Ichimura S. Effect of menatetrenone on bone
mineral density and incidence of vertebral fractures in postmenopausal women
with osteoporosis: a comparison with the effect of etidronate. J Orthop Sci.
2001;6:487–92.
26. Ozuru R, Sugimoto T, Yamaguchi T, Chihara K. Time-dependent effects
of vitamin K2 (menatetrenone) on bone metabolism in postmenopausal women.
Endocr J. 2002;49:363–70.
27. Feskanich D, Weber P, Willett WC, Rockett H, Booth SL, Colditz GA.
Vitamin K intake and hip fractures in women: a prospective study. Am J Clin Nutr.
1999;69:74–9.
28. Branca F. Dietary phyto-oestrogens and bone health. Proc Nutr Soc.
2003;62:877–87.
29. Dalais FS, Ebeling PR, Kotsopoulos D, McGrath BP, Teede HJ. The
effects of soy protein containing isoflavones on lipids and indices of bone
resorption in postmenopausal women. Clin Endocrinol (Oxf). 2003;58:704–9.
30. Yamori Y, Moriguchi EH, Teramoto T, Miura A, Fukui Y, Honda KI, Fukui
M, Nara Y, Taira K, Moriguchi Y. Soybean isoflavones reduce postmenopausal
bone resorption in female Japanese immigrants in Brazil: a ten-week study. J Am
Coll Nutr. 2002;21:560–3.
31. Izumi T, Piskula MK, Osawa S, Obata A, Tobe K, Saito M, Kataoka S,
Kubota Y, Kikuchi M. Soy isoflavone aglycones are absorbed faster and in higher
amounts than their glucosides in humans. J Nutr. 2000;130:1695–9.
32. Wang HJ, Murphy PA. Isoflavone content in commercial soybean foods.
J Agric Food Chem. 1994;42:1666–73.
33. Fukutake M, Takahashi M, Ishida K, Kawamura H, Sugimura T,
Wakabayashi K. Quantification of genistein and genistin in soybeans and soybean
products. Food Chem Toxicol. 1996;34:457–61.
34. Toda T, Tamura J, Okuhira T. Isoflavone content in commercial soybean
foods. FFI (Fish Farming Int). 1997;172:83–9.
35. Morabito N, Crisafulli A, Vergara C, Gaudio A, Lasco A, Frisina N, D’Anna
R, Corrado F, Pizzoleo MA, et al. Effects of genistein and hormone-replacement
therapy on bone loss in early postmenopausal women: a randomized double-blind
placebo-controlled study. J Bone Miner Res. 2002;17:1904–12.
1328 IKEDA ET AL.
Downloaded from https://academic.oup.com/jn/article-abstract/136/5/1323/4669999
by guest
on 22 February 2018
