Content uploaded by Chisayo Kozuka
Author content
All content in this area was uploaded by Chisayo Kozuka
Content may be subject to copyright.
Effects of the brown rice diet on visceral obesity and endothelial function:
the BRAVO study
Michio Shimabukuro
1,2,3,4
*, Moritake Higa
3,4
, Rie Kinjo
5
, Ken Yamakawa
3,4
, Hideaki Tanaka
4
,
Chisayo Kozuka
3
, Kouichi Yabiku
3
, Shin-Ichiro Taira
3
, Masataka Sata
2
and Hiroaki Masuzaki
3
1
Department of Cardio-Diabetes Medicine, The University of Tokushima Graduate School of Health Biosciences,
3-18-15 Kuramoto, Tokushima 770-8503, Japan
2
Department of Cardiovascular Medicine, The University of Tokushima Graduate School of Health Biosciences,
3-18-15 Kuramoto, Tokushima 770-8503, Japan
3
Division of Endocrinology, Diabetes and Metabolism, Haematology, Rheumatology, Second Department of Internal
Medicine, Graduate School of Medicine, University of the Ryukyus, Okinawa, Japan
4
Diabetes and Lifestyle-Related Disease Center, Tomishiro Central Hospital, Okinawa, Japan
5
Division of Clinical Laboratory, Tomishiro Central Hospital, Okinawa, Japan
(Submitted 28 January 2013 – Final revision received 27 June 2013 – Accepted 27 June 2013)
Abstract
Brown rice (BR) and white rice (WR) produce different glycaemic responses and their consumption may affect the dietary management of
obesity. In the present study, the effects of BR and WR on abdominal fat distribution, metabolic parameters and endothelial function were
evaluated in subjects with the metabolic syndrome in a randomised cross-over fashion. In study 1, acute postprandial metabolic parameters
and flow- and nitroglycerine-mediated dilation (FMD and NMD) of the brachial artery were determined in male volunteers with or without
the metabolic syndrome after ingestion of either BR or WR. The increases in glucose and insulin AUC were lower after ingestion of BR than
after ingestion of WR (P¼ 0·041 and P¼0·045, respectively). FMD values were decreased 60 min after ingestion of WR (P¼ 0·037 v. base-
line), but the decrease was protected after ingestion of BR. In study 2, a separate cohort of male volunteers (n 27) with the metabolic
syndrome was randomised into two groups with different BR and WR consumption patterns. The values of weight-based parameters
were decreased after consumption of BR for 8 weeks, but returned to baseline values after a WR consumption period. Insulin resistance
and total cholesterol and LDL-cholesterol levels were reduced after consumption of BR. In conclusion, consumption of BR may be
beneficial, partly owing to the lowering of glycaemic response, and may protect postprandial endothelial function in subjects with the
metabolic syndrome. Long-term beneficial effects of BR on metabolic parameters and endothelial function were also observed.
Key words: Glycaemic indices: Endothelial function: Obesity
Rice is generally consumed after it is refined to remove the outer
bran and germ portions of the intact grains (i.e. brown rice, BR)
and produce white rice (WR), consisting primarily of the starchy
endosperm. The postprandial blood glucose responses elicited
on consumption of WR or BR are substantially different
(1 – 3)
.
Whole grains have been shown to be associated with a low
fasting insulin level and a low glycaemic response after meal
ingestion
(4 – 6)
. The intake of whole grains, but not of refined
grains, has also been reported to be associated with low body
weight and adiposity
(5,6)
. Thus, the intake of whole-grain BR
might be beneficial for controlling weight and obesity. Recently,
we have reported that BR and g-oryzanol, one of its major
components, improve high-fat diet-induced metabolic derange-
ment and attenuate the preference for dietary fat, by decreasing
hypothalamic endoplasmic reticulum stress
(7)
. Therefore, we
hypothesised that BR may be useful for ameliorating obesity
and the metabolic syndrome, a constellation of obesity-based
metabolic abnormalities (glucose intolerance, insulin resist-
ance, dyslipidaemia and hypertension, all well-documented
risk factors of CVD), by regulating eating behaviour. Until
now, clinical trials have not been conducted to provide evidence
of a direct effect of BR on body weight and eating behaviour.
* Corresponding author: Professor M. Shimabukuro, fax þ 81 88 633 7894, email mshimabukuro-ur@umin.ac.jp
Abbreviations: BR, brown rice; FMD, flow-mediated dilation; GI, glycaemic index; GLP, glucagon-like peptide; HOMA-IR, homeostasis model assessment
of insulin resistance; hs-CRP, high-sensitivity C-reactive protein; NMD, nitroglycerine-mediated dilation; SFA, subcutaneous fat area; VFA, visceral fat area;
WR, white rice.
British Journal of Nutrition, page 1 of 11 doi:10.1017/S0007114513002432
q The Authors 2013
British Journal of Nutrition
Furthermore, an inverse relationship between the intake of
whole grains and the risk of IHD has been reported by two
cohort studies
(8,9)
. Although such protective effects are usually
explained by the presence of various constituents, such as
dietary fibre, phytic acid and vitamins, there is a lack of concrete
evidence supporting these assumptions
(10)
. A clinical meta-
analysis has shown an association between postprandial
glucose levels and macrovascular complications in non-diabetic
and diabetic individuals
(11)
. Our group
(12)
and others
(13)
have
reported that postprandial glucose elevation causes endothelial
dysfunction, an early marker of atherosclerotic changes and a
surrogate marker of future cardiovascular events. Such post-
prandial endothelial dysfunction has been suggested to be
involved in vascular complications in diabetic
(14)
and obese
patients
(15)
. The inhibition of postprandial glucose elevation
by whole grains might be beneficial for postprandial endothelial
function, but a study examining such a relationship has not
been reported.
The present investigative study was conducted to determine
the acute effects of BR, when compared with those of WR,
on postprandial metabolic parameters and postprandial
endothelial function as well as the chronic effects of BR and
WR consumption on abdominal fat distribution, metabolic
parameters and endothelial functions in subjects with the
metabolic syndrome.
Study design and methods
Study design
Healthy male volunteers were recruited by public advertisement
from 1 December 2008 to 31 January 2009, and they underwent
a 75 g oral glucose tolerance test
(16)
. In study 1, eleven subjects
participated, and in study 2, twenty-seven subjects participated,
as described below. A subject was defined as having the meta-
bolic syndrome
(17)
if he was obese (according to the modified
Japanese criteria, having a waist circumference $85 cm) and
any two of the following four factors: (1) hypertriacylglycerolae-
mia (serum TAG concentration $ 1500 mg/l (1·69 mmol/l)); (2)
low level of HDL-cholesterol (serum HDL-cholesterol concen-
tration , 400 mg/l (1·04 mmol/l)); (3) elevated blood pressure
(systolic blood pressure $ 130 mmHg and/or diastolic blood
pressure $85 mmHg); (4) high level of fasting glucose (serum
glucose concentration $ 1000 mg/l (5·6 mmol/l)). Anthropo-
metric measurements were taken while the subject was standing
erect, including those of subcutaneous fat area (SFA) and intra-
abdominal visceral fat area (VFA). These measurements were
taken at the level of the umbilicus using a standardised method
involving computed tomography, as described previously
(18)
.
The study protocol was approved by the Ethical Committee of
Tomishiro Central Hospital, and the study was carried out in
accordance with the principles of the Declaration of Helsinki as
revised in 2000. The subjects gave written informed consent
before the start of the study. The subjects were obligated to
report any serious or unexpected adverse events, whether or
not they appeared related to the intervention, immediately to
the principal investigator and/or the research ethics committee
in order to ensure appropriate management. For study 2,
in apriorianalysis made by a power analysis application
(G*Power version 3.1.7
(19)
), the required sample sizes to detect
differences in body weight per kg were as follows: fourteen in
each group at 0·8 of the power (1–
b
error probability) and the
5 % level of significance (
a
error probability) by the Wilcoxon–
Mann–Whitney test and twenty-eight matched pairs at 0·8 of
the power (1–
b
error probability) and the 5 % level of signifi-
cance (
a
error probability). Thus, we employed fourteen samples
in each group, a total of twenty-eight samples, in the present
study. The Consolidated Standards of Reporting Trials Statement
2010 checklist and a flow diagram of the progress through the
phases of a parallel randomised trial of two groups are provided
in Supplementary materials 1 and 2 (available online). The
study protocol was registered at the University Hospital Medical
Information Network Clinical Trials Registry (UMIN-CTR regis-
tration no. UMIN000009989).
Study 1: acute effects
On two mornings, at least 14 d apart, five participants without and
six with the metabolic syndrome ingested a 1883 kJ (450 kcal)
meal
(13)
, including an 837 kJ (200 kcal) meal of either BR or WR
of Japonica variety. Blood samples were collected before and
1, 2 and 4 h after meal ingestion. At the same time points, flow-
mediated dilation (FMD) and nitroglycerine-mediated dilation
(NMD) were determined using automated measurements of
the brachial arterial lumen, as described below
(20,21)
.
Study 2: chronic effects
Participants with the metabolic syndrome (n 27) were
instructed to ingest BR or WR of Japonica variety for 8 weeks
as described below. The participants were randomised by a
computer-generated random number table into either the BR
group followed by the WR group (BR-WR, n 14) or the WR
group followed by the BR group (WR-BR, n 13). The BR-WR
group consumed BR for the first 8 weeks and WR for the
next 8 weeks. The WR-BR group followed the reverse con-
sumption protocol. We instructed the participants to follow
regular dietary intake/exercise habits during the intervention
period, except for consumption of delivered rice. On each
day, after ingestion of BR or WR, the participants recorded
their compliance to the study requirement and their level of
satiety (scale 1–10: not satisfied to satisfied). Before and
after completion of both 8-week terms, blood and urine
samples were collected, and FMD and NMD were determined;
oral glucose tolerance test and abdominal computed tomogra-
phy were also performed. Because primary outcomes were
the effects of BR/WR on metabolic parameters and endothelial
function and the carry-over effect between BR-WR and WR-BR
groups could be minimised to these parameters after 2 months
of treatment, we had not included a washout period.
Assessment of vascular function
FMD and NMD were measured using a vascular ultrasound
system equipped with an automatic edge-tracking system for
two-dimensional lumen imaging (UNEXEF; UNEX), according
to the established guidelines
(20,21)
. The measurements of FMD
M. Shimabukuro et al.2
British Journal of Nutrition
and NMD were taken by a single laboratory technician using a
skilful and stable technique to avoid inter-technician variation.
The correlation coefficient between two FMD measurements
was 0·86, with a CV of 11·2 %
(21)
. The diameter of the brachial
artery was measured, at rest, in the cubital region. Sub-
sequently, the cuff was inflated to 50 mmHg above the systolic
blood pressure, held there for 5 min and deflated. The diameter
at the same point of the artery was monitored continuously,
and the maximum dilation, occurring 45–60 s after deflation,
was recorded. The measurement of NMD was taken after a
15 min interval to allow for vessel recovery. Sublingual gly-
ceryl trinitrate (75 mg) was administered, and the maximum
dilation of the brachial artery, at the same point as for the
measurement of FMD, was confirmed and measured by a pla-
teau in the diameter of the artery, using real-time monitoring
of the diameter of the artery, over a period of at least 1 min
after the point of maximum dilation. FMD and NMD values
were calculated as follows:
FMD or NMD value ð%Þ¼ðmaximum diameter
2 diameter at restÞ
£ 100=diameter at rest:
Biochemical measurements
Venous blood samples were collected in tubes without
an anticoagulant or in tubes with EDTA sodium (1 mg/ml) or
spray-dried K
2
EDTA and DPP-4 protease inhibitors (Becton
Dickinson) for the determination of the concentrations of
active glucagon-like peptide (GLP)-1 as well as plasma
glucose and serum total cholesterol, HDL-cholesterol, TAG,
creatinine and glucose. The concentration of LDL-cholesterol
was estimated using Friedewald’s method
(22)
. The con-
centration of glycosylated Hb was measured using HPLC and
that of insulin using chemiluminescent enzyme immunoassay.
The value for the concentration of glycosylated Hb (%) was
converted to the National Glycohemoglobin Standardization
Program levels
(23)
. The concentration of high-sensitivity
C-reactive protein (hs-CRP) (N Latex CRP II) was determined
using immunonephelometric methods. The concentration of
GLP-1 was measured using an ELISA kit (Millipore)
(24)
.The
plasma concentrations of amidated GLP-1(7–36) and GLP-1
(7–37) were measured using an antibody that was highly specific
for the N-terminus of GLP-1 and did not react with GLP-1(9–36),
GLP-2 or glucagon. Other blood component assays were
conducted using standard methods. Plasma was immediately
separated by centrifugation at 3000 rpm at 48Cfor10minand
serum by centrifugation at 1000 rpm at room temperature for
10 min. Urine samples were collected in light-resistant tubes and
used to measure the concentration of urinary 8-isoprostane,
using an enzyme immunoassay kit (Assay Designs).
Statistical analysis
Values are expressed as means and standard deviations, unless
otherwise indicated. Two-tailed paired/unpaired Student’s
t test or one-way factorial ANOVA, followed by Bonferroni’s
post hoc comparisons, was used to compare inter-group
or intra-group means for parametric data. For small-group
comparison, a non-parametric Wilcoxon–Mann–Whitney test
was employed. All analyses were performed using Jump
version 10.0.1.1 software (SAS Institute). Differences were
considered to be significant if the P value was , 0·05.
Results
Study 1: acute effects
The ages of the subjects with or without the metabolic syn-
drome were similar. Body weight, BMI, waist circumference
and blood pressure were higher in subjects with the metabolic
syndrome (Table 1). In subjects with the metabolic syndrome,
the concentrations of glucose and insulin at 120 min after
ingestion of the 1883 kJ (450 kcal) meal were lower when
the meal consisted of BR than when it consisted of WR
(Fig. 1). The increase in glucose AUC (DAUC
glucose
) after
meal ingestion was lower when BR was consumed than
when WR was consumed (253 (
SD 140) v. 346 (SD 116) min £
mmol/l, P¼0·006), and the increase in insulin AUC
(DAUC
insulin
) was also lower in participants consuming BR
than in those consuming WR (62 886 (
SD 32 251) v. 95 580
(
SD 55 310) min £ pmol/l, P¼0·044). The concentrations of
LDL, HDL and TAG were not different between the subjects con-
suming the different types of rice (Supplementary material 3,
available online), but that of NEFA, 240 min after meal ingestion,
was significantly lower in subjects ingesting BR. In subjects
without the metabolic syndrome, the concentrations of glucose,
insulin, NEFA, LDL-cholesterol or HDL-cholesterol and TAG,
DAUC
glucose
and DAUC
insulin
were all comparable between
those consuming BR and those consuming WR.
Among subjects without the metabolic syndrome, the
degree of FMD of the brachial arterial diameter during reactive
hyperaemia was not different at 60, 120 and 240 min after
ingestion of BR or WR, when compared with baseline values
(Fig. 2). However, in subjects with the metabolic syndrome,
FMD values were decreased at 60 min after consuming WR
(P¼0·037 v. baseline) and returned to baseline values by
120 min. FMD values were not decreased after ingestion of
BR. In both the groups, NMD values did not change after
ingestion of either BR or WR.
Study 2: chronic effects
There were no statistical differences in the baseline character-
istics between the subjects in the two groups (Table 2). Subject
adherence to the diets and post-meal satiety scores were com-
parable between the different diets (Supplementary material 4,
available online). In the BR-WR group, body weight, BMI and
waist circumference were decreased by the end of the 8-week
BR diet period and returned to baseline values by the end of
the WR diet period. In the WR-BR group, body weight, BMI,
waist circumference and systolic blood pressure were com-
parable with the baseline values by the end of the 8-week
WR diet period, but waist circumference and systolic blood
pressure were lower at the end of the 8-week BR diet period.
As shown in Table 3, the homeostasis model assessment
Brown rice, obesity and endothelial function 3
British Journal of Nutrition
of insulin resistance (HOMA-IR) and total cholesterol and
LDL-cholesterol concentrations of the BR-WR group were
decreased at the end of the 8-week BR diet period and
returned to baseline values over the course of the WR diet
period. In the WR-BR group, values of all the variables were
comparable with those at baseline at the end of the WR diet
period, but the HOMA-IR and LDL-cholesterol levels
decreased over the course of the BR diet period. The concen-
trations of markers of radical oxygen species (isoprostane),
inflammation (hs-CRP) and incretin (active GLP-1) were all
comparable between the periods of BR or WR consumption
(Table 2). As shown in Fig. 3(a) and (b), body weight and
waist circumference were significantly lower than those at
baseline, after consumption of the BR diet in both the
groups. VFA was also significantly smaller following the BR
diet phase of the study than following the WR phase of the
study in both the groups; SFA remained comparable between
the phases of the study. We did not observe significant differ-
ences in FMD values in the BR-WR and WR-BR groups, but
could find a difference in the combined group, indicating a
sample size effect. As shown in Fig. 4, FMD values increased
significantly from baseline by the end of the BR diet phase
of the study, but returned to baseline values during the WR
diet phase. We determined Pearson’s correlation coefficient, r,
a measure of the strength and direction of the linear relation-
ship between metabolic variables and FMD at baseline, after
the 8-week BR diet period and after the 8-week WR diet
period. After the 8-week BR diet period, there was a signifi-
cant negative correlation with HOMA-IR (r 0·652, P¼ 0·016)
and a borderline correlation with VFA (r 0·387, P¼0·062),
but not with other metabolic parameters including waist cir-
cumference, BMI, insulin levels at 0–120 min, glucose levels
at 0–120 min and lipid parameters. NMD values remained
comparable with the baseline values throughout the study.
Discussion
The present study revealed that a single daily meal that
included BR, when compared with that consisting of WR,
resulted in decreased postprandial concentrations of insulin
and glucose and prevented postprandial endothelial dys-
function in subjects with the metabolic syndrome. It also
revealed that switching the staple food of subjects with the
metabolic syndrome from WR to BR led to a decrease in
body weight, systolic blood pressure, HOMA-IR, and total
cholesterol and LDL-cholesterol levels and improved the
endothelial function.
Acute effects
In the present study, the DAUC
glucose
and DAUC
insulin
after
ingestion of the 1883 kJ (450 kcal) meal were lower in subjects
with the metabolic syndrome who consumed BR than in those
who consumed WR (Fig. 1). Glycaemic index (GI) is defined
as the response to 50 g available carbohydrates from a food
in relation to 50 g available carbohydrates from a control
food, i.e. it is a relative measure of the glycaemic response
to available carbohydrates in a food
(2)
. The high content of
viscous fibre and various enzymatic inhibitors are believed
to slow the digestion and absorption of whole grains such
as BR, compared with refined grains such as WR, and, there-
fore, elicit smaller postprandial glucose responses and a
reduced insulin demand
(25)
. Although GI vary in types of
rice
(3)
, there is a difference between BR and WR belonging
Table 1. General characteristics of study 1 subjects
(Mean values and standard deviations)
Metabolic syndrome – Metabolic syndrome þ
Parameters Mean
SD Mean SD
n 56
Age (years) 45 4 41 5
Body weight (kg) 67·4 6·1 79·2* 17·6
BMI (kg/m
2
) 23·1 1·7 28·1* 4·3
Waist circumference (cm) 81·3 7·5 93·8* 9·6
Systolic blood pressure (mmHg) 124 14 141* 10
Diastolic blood pressure (mmHg) 82 6 89* 6
Pulse (beats/min) 60 4 75 14
Glucose (mmol/l) 5·9 1·4 6·1 1·4
Insulin (pmol/l) 39 14 87* 37
HOMA-IR 1·43 0·51 3·48* 1·29
HbA
1c
(NGSP %) 5·58 0·68 6·15 1·76
Total cholesterol (mmol/l) 5·34 0·64 5·42 0·86
LDL-cholesterol (mmol/l) 3·55 0·24 3·27 0·62
HDL-cholesterol (mmol/l) 1·37 0·13 1·26 0·27
TAG (mmol/l) 1·41 0·62 2·48* 0·70
NEFA (mmol/l) 0·196 0·136 0·248 0·133
AST (IU/l) 23 6 31 12
ALT (IU/l) 30 12 49 30
g-GTP (IU/l) 34 25 80* 46
HOMA-IR, homeostasis model assessment of insulin resistance; HbA
1c
, glycosylated Hb; NGSP, National Glycohemoglobin
Standardization Program; AST, aspartate aminotransferase; ALT, alanine aminotransferase; g-GTP, g-glutamyl transpeptidase.
* Mean value was significantly different from that of the subjects without the metabolic syndrome (P, 0·05).
M. Shimabukuro et al.4
British Journal of Nutrition
to the same type of rice. It has been reported that in the
Japonica rice variety that we used, GI was significantly
lower in BR than in WR (61·5 (
SD 4·7) v.75·9(SD 6·6),
P, 0·05) when compared with a control 25 % weight/volume
glucose solution
(26)
. Notably, the effects of BR on postprandial
glucose and insulin levels were observed only in subjects
with the metabolic syndrome. A cross-over study involving
overweight subjects has reported that their insulin sensitivity
improved after being on a whole-grain diet for 6 weeks,
as opposed to a refined-grain diet, independent of body
weight
(27)
. Although BR has not been conclusively demon-
strated to improve postprandial insulin action after consump-
tion of a single meal, the observed decrease in postprandial
NEFA levels in subjects with the metabolic syndrome may
support this notion.
FMD, as assessed in the present study, mimics the NO-
mediated vasodilation produced by increased blood flow
after a period of ischaemia. Non-endothelium-dependent
dilation is useful in the measurement of the arterial changes
induced by the administration of a sublingual dose of
nitroglycerine, which predominantly reflects the smooth
muscle response
(20)
. In subjects with the metabolic syndrome,
FMD values of the brachial arterial diameter were decreased at
60 min after consumption of WR (P¼ 0·037 v. baseline). Since
NMD values did not change after WR ingestion, this change is
largely due to postprandial endothelial dysfunction and not
due to smooth muscle cell dysfunction. We
(13)
and others
(12)
have reported that patients with type 2 diabetes mellitus or
impaired glucose tolerance and endothelial function, but not
with impaired smooth muscle cell function, have worse endo-
thelial function after meal consumption. The endothelial
function in subjects with the metabolic syndrome may be
impaired, to some degree, by hyperglycaemia and a conse-
quent increase in reactive oxygen species levels
(14)
. We had
previously reported that a reduction in postprandial blood
glucose levels by a-glycosidase inhibitor, which delays
glucose release from complex carbohydrates, improves post-
prandial endothelial function in type 2 diabetes mellitus
patients
(28)
. Although the exact mechanisms of the beneficial
effects of BR on postprandial endothelial function are
unknown, part of the BR protective function may result from
the inhibition of postprandial glucose increase after consump-
tion of a single meal.
15
10
5
Glucose (mmol/l)Insulin (pmol/l)NEFA (mmol/l)
0
1500
1000
500
0
0·6
0·5
0·4
0·3
0·2
0·1
0·0
0·6
0·5
0·4
0·3
0·2
0·1
0·0
1500
1000
500
0
060
Metabolic
syndrome–
Metabolic
syndrome+
120 180 240
15
10
5
0
0 60 120 180 240
0
60 120 180
240 0
60 120 180
240
0
60 120 180
240
Time (min)
0
60 120 180 240
*
*
*
Fig. 1. Changes in biochemical parameters before and after ingestion of a
meal with either brown rice (BR,
) or white rice (WR, ). On two mornings,
participants with (BMI $ 25 kg/m
2
) or without obesity (BMI , 25 kg/m
2
)
ingested either a 1883 kJ (450 kcal) meal, including an 837 kJ (200 kcal) meal
of either BR or WR. Before and 60, 120 and 240 min after ingestion, blood
samples were collected. The concentrations of glucose, insulin and NEFA
are shown. For the concentrations of LDL-cholesterol, HDL-cholesterol and
TAG, see Supplementary material 3 (available online). Values are means,
with standard deviations represented by vertical bars. * Mean value was
significantly different from that at baseline (P, 0·05).
12
10
30
20
10
0
30
20
10
0
Metabolic
syndrome–
Metabolic
syndrome+
8
6
4
FMD (%)NMD (%)
2
0
0
60 120 180 240
12
10
8
6
4
2
0
0
60
*
120 180 240
0 60 120 180
240
Time (min)
0
60 120 180 240
Fig. 2. Changes in forearm flow-mediated dilation (FMD) and nitroglycerine-
mediated dilation (NMD) values before and after ingestion of a meal with
either brown rice (BR,
) or white rice (WR, ). On two separate mornings,
participants with (BMI $ 25 kg/m
2
) or without obesity (BMI , 25 kg/m
2
)
ingested a 1882·8 kJ (450 kcal) meal, including an 836·8 kJ (200 kcal) meal
of either BR or WR. Before and 60, 120 and 240 min after ingestion, FMD
and NMD were measured using a novel vascular ultrasound system
equipped with an edge-tracking system for two-dimensional imaging and
automatic measurement. FMD and NMD values were calculated as follows:
FMD or NMD value (%) ¼ (maximum diameter 2 diameter at rest) £ 100/
diameter at rest. Values are means, with standard deviations represented by
vertical bars. * Mean value was significantly different from that at baseline
(P, 0·05).
Brown rice, obesity and endothelial function 5
British Journal of Nutrition
Chronic effects
Metabolic parameters. A few trials have examined the differ-
ential effects of whole and refined grains on body weight and
weight changes. A 16-week clinical trial in Korean men with
CHD has shown that the isoenergetic replacement of WR
with whole grains and legume powder leads to significant
reductions in serum glucose and insulin concentrations,
whereas the body weight remains unchanged
(29)
. A study on
Chinese type 2 diabetes mellitus patients has found that sub-
stituting BR with WR for 16 weeks did not substantially
affect the concentrations of metabolic markers, although
HDL-cholesterol levels and diastolic blood pressure were
improved
(30)
. In the present study, switching the staple food
of the subjects to BR resulted in a significant decrease in
body weight, BMI and waist circumference in those with the
metabolic syndrome. There is a discrepancy between previous
reports
(29,30)
and the present one with regard to the effects of
whole and refined grains on body weights, which may have
been the result of differences in the study subjects. The ben-
eficial effect of BR on body weight has been confirmed by a
reduction in VFA, measured by abdominal computed tomogra-
phy. Thus, changes in VFA (DVFA%) were significantly lower
after 8 weeks of BR consumption than after a comparable
period of WR consumption among subjects in both the BR-
WR and WR-BR groups; however, changes in SFA (DSFA%)
were comparable among the subjects, regardless of the
staple food included in their diets. Previously, an analysis of
fat distribution by computed tomography has demonstrated
that visceral fat is decreased to a greater extent as a result
of a low-energy diet than abdominal subcutaneous fat,
particularly in subjects with visceral fat obesity
(31)
. These
observations suggest that subjects with visceral fat obesity
might be more susceptible to body weight reductions resulting
from a diet including whole grains, including BR.
There are three potential mechanisms by which BR
decreases body weight and visceral fat in subjects with the
metabolic syndrome. First, the lowered postprandial glucose
and insulin levels associated with BR intake may lead to
weight loss. Indirect evidence from both epidemiological
and short-term experimental studies suggests the potential
role of a high-GI diet, containing refined grains, in the
development of obesity
(32,33)
. Alternatively, the low postpran-
dial glucose and insulin levels associated with high whole
grain intake may lead to weight loss, especially among over-
weight or obese individuals
(34)
. Reductions in postprandial
glucose and insulin levels by an a-glycosidase inhibitor, migli-
tol, have been shown to be associated with reductions in body
weight, waist circumference and VFA in subjects with the
metabolic syndrome
(35)
. Together, the lower postprandial glu-
cose and insulin levels associated with BR intake may lead to
weight loss in subjects with the metabolic syndrome. Second,
alterations in hunger and/or increased satiety after BR con-
sumption may lead to voluntary energy intake reductions.
The consumption of low-GI foods has been reported to be
directly associated with reductions in subsequent hunger
and/or increased satiety, leading to low energy intake
(36)
.
Although most of these trials were conducted over only a
single meal or a single day, they collectively suggest that
long-term consumption of whole-grain products may increase
satiety and reduce energy consumption. Thus, a whole grain-
containing diet may contribute to weight loss, especially in
sedentary and overweight subjects. Epidemiological studies
on dietary fibre consumption have also suggested that intake
of whole grains is inversely associated with body weight
and fat distribution
(32,37)
. The inherent high fibre content of
whole-grain foods may help prevent weight gain by increasing
appetite control through delayed carbohydrate absorption
(38)
.
The correlation between dietary fibre and GI is modest,
suggesting that other factors are important in determining
the glycaemic effects of foods
(38)
. In the present study, the
satiety scores recorded after consumption of meals with
either BR or WR were comparable (Supplementary material 4,
available online). The effects of BR on eating behaviour
cannot be ruled out because the assessment of appetite and
eating behaviour in humans is complicated and difficult
(36)
.
The association between hunger and/or satiety and alteration
of metabolic parameters should be evaluated in future studies.
Third, microgradients, which are removed along with the
outer bran by refining, may affect eating behaviour and insulin
sensitisation independently by lowering the GI. Whole grains,
including BR, are generally low in saturated fat and high
in microgradients that might be associated with improved
insulin sensitivity in metabolic derangement
(4 – 6)
. Since the
Table 2. General characteristics of study 2 subjects
(Mean values and standard deviations)
BR-WR group (n 14) WR-BR group (n 13)
Baseline
8-week BR
phase
8-week WR
phase Baseline
8-week WR
phase
8-week BR
phase
Parameters Mean
SD Mean SD Mean SD Mean SD Mean SD Mean SD
Body weight (kg) 76·4 11·2 74·7* 10·6 76·6 10·8 76·8 15·4 77·2 15·6 76·4 16·1
BMI (kg/m
2
) 26·7 2·8 26·1* 2·7 26·1 2·7 26·7 4·2 26·9 4·3 26·6 4·5
Waist circumference (cm) 93·2 9·2 91·7* 8·8 91·5 8·8 92·1 9·3 92·3 11·0 90·3* 10·3
Systolic blood pressure (mmHg) 140 14 132* 11 139 21 141 13 140 16 134* 13
Diastolic blood pressure (mmHg) 84 11 86 9 88 12 85 8 89 9 88 8
Heart rate (beats/min) 66 8 70 8 70 7 72 10 73 9 75 12
BR, brown rice; WR, white rice.
* Mean value was significantly different from that at baseline (P, 0·05).
M. Shimabukuro et al.6
British Journal of Nutrition
Table 3. Changes in blood biochemical parameters (study 2)
(Mean values and standard deviations)
BR-WR group (n 14) WR-BR group (n 13)
Baseline 8-week BR phase 8-week WR phase Baseline 8-week WR phase 8-week BR phase
Parameters Mean
SD Mean SD Mean SD Mean SD Mean SD Mean SD
Glucose (mmol/l)
0 6·2 1·1 6 1 6 0·9 6 1·1 6·7 2·6 6·2 2·3
30 9·5 2 9·3 1·8 9·1 1·9 9·0 1·9 10·4 3·5 10·0 3·4
60 9·9 2·9 10·6 2·9 9·6 3·2 10·6 3·7 11·6 4·9 11·1 5·6
120 8·3 3·5 8·5 2·7 8·4 2·7 9·0 3·6 9·9 5·4 9·3 5·6
Insulin (pmol/l)
0 623559 2562265822825377 47
30 554 475 443 361 562 527 449 298 491 369 506 322
60 547 437 547 295 495 282 583 375 665 460 567 238
120 352 217 487 352 520 324 463 225 476 271 360 152
HOMA-IR 2·89 1·40 2·23* 0·95 2·51 1·33 2·79 1·24 3·83 8·10 2·34† 1·75
HOMA-b 75 43 77 50 77 46 83 41 96 63 106 83
HbA
1c
(NGSP %) 5·74 0·55 5·72 0·45 5·9 1·55 5·78 0·81 5·83 1·43 6·19 1·55
Total cholesterol (mmol/l) 5·85 0·54 5·31* 0·40 5·66 0·72 5·67 0·68 5·27 0·61 5·56 0·54
LDL-cholesterol (mmol/l) 3·45 0·40 3·15* 0·35 3·37 0·56 3·48 0·52 3·28 0·48 3·41 0·59
HDL-cholesterol (mmol/l) 1·44 0·30 1·37 0·23 1·47 0·24 1·26 0·24 1·21 0·19 1·20 0·14
TAG (mmol/l) 2·05 1·07 1·69 0·65 1·74 0·65 1·95 0·90 1·65 0·58 1·42 0·50
NEFA (mmol/l) 0·13 0·08 0·11 0·05 0·15 0·14 0·09 0·04 0·13 0·19 0·12 0·12
AST (IU/l) 25 8 27 21 24 8 23 6 24 9 25 9
ALT (IU/l) 32 19 40 38 31 16 36 16 42 21 43 21
g-GTP (IU/l) 53 24 64 31 54 29 67 36 68 35 70 45
Uric acid (mg/l) 66·3 9·7 70·1 10·8 68·2 9·5 68 15·7 65·8 16·4 64·6 13·0
High-sensitivity CRP (mg/l) 0·9 0·4 1·1 0·9 0·8 0·5 1·1 0·4 2·3 1·6 1·3 0·9
GLP-1 (pmol/l) 2·17 0·41 2·41 0·79 2·29 0·49 2·80 0·14 3·26 1·55 3·81 2·12
Urinary 8-isoprostane (ng/ml) 112 25 132 90 109 47 148 84 164 93 112 42
Urinary albumin excretion (mg/g creatinine) 4·04 2·39 4·59 4·66 5·16 8·24 3·73 1·72 5·42 4·05 6·30 8·13
BR, brown rice; WR, white rice; HOMA-IR, homeostasis model assessment of insulin resistance; HbA
1c
, glycosylated Hb; NGSP, National Glycohemoglobin Standardization Program; AST, aspartate aminotransferase; ALT, alanine
transaminase; g-GTP, g-glutamyl transpeptidase; CRP, C-reactive protein; GLP-1, glucagon-like peptide-1.
* Mean value was significantly different from that at baseline (P, 0·05).
† Mean value was significantly different from that of the 8-week WR phase.
Brown rice, obesity and endothelial function 7
British Journal of Nutrition
concentrations of markers of radical oxygen species (isopros-
tane), inflammation (hs-CRP) and incretin (active GLP-1) were
all comparable between the BR and WR diet groups in the
present study, other mechanisms must be involved in the
observed improvements in obesity and insulin sensitivity.
Recently, we have reported that BR and its component,
g-oryzanol, alter eating behaviour and fuel homeostasis in
mice
(7)
. When mice were allowed free access to both a BR-
containing chow diet and a high-fat diet, they preferred the
chow diet to the high-fat diet. BR and g-oryzanol improve
high-fat diet-induced metabolic derangement and attenuate
the preference for dietary fat by decreasing hypothalamic
endoplasmic reticulum stress. The relevance of microgradi-
ents, such as g-oryzanol, to eating behaviour needs to be
validated in future human studies.
Endothelial function. As shown in Fig. 4, FMD values (%)
were increased from baseline after consumption of the BR diet
for 8 weeks, but returned to baseline values after consumption
of WR for a similar period of time. NMD values (%) were com-
parable with those at baseline after consumption of diets
containing either type of rice. Decreased FMD has been
reported to be associated with cardiovascular risk factors,
including obesity and the metabolic syndrome, and with the
estimated 10-year risk of CHD
(20,21)
. Interventions to reduce
cardiovascular risk in adults have demonstrated a parallel
improvement in FMD
(20,21)
. These observations led to the
postulation of three mechanisms by which BR may improve
FMD. First, the effects of BR on postprandial glucose and insu-
lin levels may be associated with improved endothelial
function
(39)
, as suggested by the association of postprandial
6
(a) (c)
(b) (d)
5
4
3
2
1
0
–1
–2
–3
∆BW (%)
∆VFA (%)∆SFA (%)
∆WC (%)
–4
–5
–6
10·0
7·5
5·0
2·5
0·0
–2·5
–5·0
–7·5
BR
8 weeks
WR
8 weeks
WR
8 weeks
**
**
** **
BR
8 weeks
BR
8 weeks
WR
8 weeks
WR
8 weeks
BR
8 weeks
10·0
7·5
5·0
2·5
0·0
–2·5
–5·0
–7·5
640
30
20
10
0
–10
–20
–30
–40
40
30
20
10
0
–10
–20
–30
–40
40
30
20
10
0
–10
–20
–30
–40
40
30
20
10
0
–10
–20
–30
–40
5
4
3
2
1
0
–1
–2
–3
–4
–5
–6
Fig. 3. Percentage changes in (a) body weight (DBW), (b) waist circumference (DWC), (c) visceral fat area (DVFA) and (d) subcutaneous fat area (DSFA) after
ingestion of a brown rice (BR) or a white rice (WR) diet in obese subjects. Obese participants (BMI $25 kg/m
2
) were randomised to either a diet including BR fol-
lowed by a diet including WR group (BR-WR, n 14) or a WR-containing diet followed by the one containing BR group (WR-BR, n 13). Before and after completion
of the first and second 8-week terms, blood and urine samples were collected, and abdominal fat computed tomography scans were taken. Values represent
percentage change from baseline values of each 8-week term. Values are means, with standard deviations represented by vertical bars. (a) Mean value was
significantly different from that at baseline following the consumption of the BR diet: P¼ 0·009, P¼0·045 (paired t test). (b) Mean value was significantly different
from that at baseline following the consumption of the BR diet: P¼0·032, P¼ 0·047 (paired t test). (c) Mean value was significantly different from that following the
consumption of the BR diet: P¼ 0·018, P ¼ 0·003 (unpaired t test). Mean value was significantly different from that at baseline: *P, 0·05, **P, 0·01.
15·0
30
25
20
15
10
5
0
12·5
10·0
*
7·5
5·0
FMD (%)
NMD (%)
2·5
0·0
Baseline
BR 8
weeks
WR 8
weeks
Baseline
BR 8
weeks
WR 8
weeks
Fig. 4. Changes in forearm flow-mediated dilation (FMD) and nitroglycerine-
mediated dilation (NMD) values after brown rice (BR) or white rice (WR) diet
consumption in obese subjects. Obese participants (BMI $ 25 kg/m
2
) were
randomised into either a group consuming a diet that included BR for an
8-week period, followed by consumption of a diet containing WR for a similar
period (BR-WR, n 14), or a group consuming the diet in the reverse order
(WR-BR, n 13). Before and after completion of each 8-week term, FMD and
NMD were measured using a novel vascular ultrasound system equipped
with an edge-tracking system for two-dimensional imaging and automatic
measurement. FMD and NMD values were calculated as follows: FMD or
NMD value (%) ¼ (maximum diameter 2 diameter at rest) £ 100/diameter at
rest. Values are means with minimum and maximum values, with standard
deviations represented by vertical bars. * Mean value was significantly
different from that at baseline (P, 0·05; one-way ANOVA or paired t test).
M. Shimabukuro et al.8
British Journal of Nutrition
glucose levels with macrovascular complications
(11,14)
.
Postprandial glucose elevation has also been suggested to
elicit endothelial dysfunction
(12,13)
, a surrogate marker of
future cardiovascular events
(12 – 15)
. Second, a reduction in
body weight and/or improved insulin sensitivity may be associ-
ated with improved endothelial function. After 8 weeks of
BR consumption, there was a significant negative correlation
with HOMA-IR and a borderline correlation with VFA. How-
ever, the effects of body weight reduction on endothelial
function have been inconclusive. FMD was not improved
with drug-induced weight loss in overweight adults
(40)
,
whereas another study has shown increased FMD with the
use of a lipase inhibitor (orlistat) that prevents fat
absorption
(41)
. In a double-blind, placebo-controlled study
investigating the effect of orlistat, Bergholm et al.
(42)
demon-
strated that the lowering of LDL-cholesterol levels, rather
than moderate weight loss, improved endothelial function
in obese subjects. The weight loss associated with a very
low-energy, 2-week diet has been reported to improve
endothelium-dependent vasodilation in obese, hypertensive
subjects
(43)
. Moreover, dietary changes, combined with exer-
cise, have been found to elicit a multiplicative response,
and improvements in FMD appeared to be independent of
improvements in glucose tolerance
(44)
. The attenuation of
insulin resistance, rather than changes in carbohydrate toler-
ance, might be more important in affecting improvement in
endothelial function
(45)
. Third, a reduction in LDL-cholesterol
levels and blood pressure may directly affect endothelial
function. The concentrations of markers of radical oxygen
species (isoprostane), inflammation (hs-CRP) and incretin
(active GLP-1) were unchanged, but LDL-cholesterol levels
and systolic blood pressure were improved by BR ingestion.
As has been discussed above, the underlying mecha-
nism(s) by which BR improves endothelial function may
be multiactorial. Further studies are required to confirm our
observations and clarify the mechanism(s) underlying endo-
thelial function improvement. An inverse relationship between
the intake of whole grains and the risk of IHD has been
reported by large cohort studies
(9,10)
. If the above-mentioned
beneficial effects of BR on endothelium are elicited in a
clinical setting, then BR may provide protection against
atherosclerotic cardiovascular events, in part, by improving
endothelial function.
Limitations
The present study has several limitations. First, the number of
patients was too small for any definite conclusions to be
drawn, especially in study 1. Second, the study could not
determine a causal relationship between BR consumption
and endothelial function or weight loss. Third, the present
study includes potential confounding factors that may have
affected the observed effects of daily BR or WR consumption.
This was because it was not possible to closely and individu-
ally match side foods, based on macronutrient composition
and palatability, among the study subjects. Lastly, the assess-
ments of hunger and satiety were not precise. Although the
duration of satiety might be prolonged by BR, changes in
eating behaviour, manifest by hunger and satiety, could not
be quantified or characterised.
Conclusion
The results of the present study suggest that BR could be
beneficial, partly through lowering of the postprandial glycae-
mic response, and may provide a measure of protection to
postprandial endothelial function in subjects with the
metabolic syndrome. Long-term benefits of BR on metabolic
parameters and endothelial function were also observed.
Future studies with larger cohort sizes and longer durations
of follow-up are warranted to examine the effects of substitut-
ing BR with WR on metabolic risk for future cardiovascular
events.
Supplementary material
To view supplementary material for this article, please visit
http://dx.doi.org/0.1017/S0007114513002432
Acknowledgements
The authors cordially acknowledge the staff of Tomishiro
Central Hospital, Okinawa, Japan, especially Kaori Ichimatsu,
at the Nutrition Division, for the diet protocol and Hiroe
Shinjo, Saeko Nakamura, Ayano Kamiyama, Merina Ohta
and Megumi Yonaha, at the Diabetes and Lifestyle-related Dis-
ease Center, for their devoted secretarial work and Hiroyuki
Oshiro, Okinawa Shokuryo K.K., Okinawa, Japan for BR and
WR. The present study was supported by grants from the Min-
istry of Education, Culture, Sports, Science and Technology
(MEXT) and the Ministry of Health, Labour and Welfare
(MHLW), Japan. MEXT and MHLW had no role in the
design, analysis or writing of this article. The authors’ contri-
butions are as follows: Mi. S. designed the present study,
analysed the data and wrote the manuscript; M. H. was
involved in patient management, data collection and discussion;
R. K. performed the vascular function study; Ke. Y., H. T., C. K.,
Ko. Y., S. T., Ma. S. and H. M. contributed to the discussion.
None of the authors has any conflicts of interest.
References
1. O’Dea K, Nestel PJ & Antonoff L (1980) Physical factors influ-
encing postprandial glucose and insulin responses to starch.
Am J Clin Nutr 33, 760 –765.
2. Jenkins DJ, Wolever TM, Taylor RH, et al. (1981) Glycaemic
index of foods: a physiological basis for carbohydrate
exchange. Am J Clin Nutr 34 , 362–366.
3. Foster-Powell K, Holt SH & Brand-Miller JC (2002) Inter-
national table of glycaemic index and glycaemic load
values. Am J Clin Nutr 76, 5–56.
4. McKeown NM, Meigs JB, Liu S, et al. (2004) Carbohydrate
nutrition, insulin resistance, and the prevalence of the meta-
bolic syndrome in the Framingham Offspring Cohort.
Diabetes Care 27, 538– 546.
5. Tighe P, Duthie G, Vaughan N, et al. (2010) Effect of
increased consumption of whole-grain foods on blood
pressure and other cardiovascular risk markers in healthy
Brown rice, obesity and endothelial function 9
British Journal of Nutrition
middle-aged persons: a randomized controlled trial. Am J
Clin Nutr 92 , 733–740.
6. Nettleton JA, McKeown NM, Kanoni S, et al. (2010) Inter-
actions of dietary whole-grain intake with fasting glucose-
and insulin-related genetic loci in individuals of European
descent: a meta-analysis of 14 cohort studies. Diabetes
Care 33, 2684–2691.
7. Kozuka C, Yabiku K, Sunagawa S, et al. (2012) Brown rice
and its component, g-oryzanol, attenuate the preference
for high-fat diet by decreasing hypothalamic endoplasmic
reticulum stress in mice. Diabetes 61, 3084–3093.
8. Jacobs DR Jr, Meyer KA, Kushi LH, et al. (1998) Whole-grain
intake may reduce the risk of ischemic heart disease death in
postmenopausal women: the Iowa Women’s Health Study.
Am J Clin Nutr 68, 248– 257.
9. Liu S, Stampfer M, Hu F, et al. (1999) Whole grain consump-
tion and risk of coronary heart disease: results from the
Nurses’ Health Study. Am J Clin Nutr 70, 412–419.
10. Ma XY, Liu JP & Song ZY (2012) Glycemic load, glycemic
index and risk of cardiovascular diseases: meta-analyses of
prospective studies. Atherosclerosis 223, 491–496.
11. Coutinho M, Gerstein HC, Wang Y, et al. (1999) The relation-
ship between glucose and incident cardiovascular events. A
metaregression analysis of published data from 20 studies of
95,783 individuals followed for 12.4 years. Diabetes Care 22,
233–240.
12. Kawano H, Motoyama T, Hirashima O, et al. (1999) Hyper-
glycemia rapidly suppresses flow-mediated endothelium-
dependent vasodilation of brachial artery. J Am Coll Cardiol
34, 146 –154.
13. Shimabukuro M, Chinen I, Higa N, et al. (2007) Effects of
dietary composition on postprandial endothelial function
and adiponectin concentrations in healthy humans: a cross-
over controlled study. Am J Clin Nutr 86, 923–928.
14. Ceriello A, Hanefeld M, Leiter L, et al. (2004) Postprandial
glucose regulation and diabetic complications. Arch Intern
Med 164, 2090–2095.
15. Jonk AM, Houben AJ, Schaper NC, et al. (2011) Obesity
is associated with impaired endothelial function in the post-
prandial state. Microvasc Res 82, 423–429.
16. Seino Y, Nanjo K, Tajima N, et al. (2010) Report of the
Committee on the classification and diagnostic criteria of
diabetes mellitus. J Diab Invest 1, 212–228.
17. Alberti KG, Zimmet P & Shaw J (2006) Metabolic syndrome – a
new world-wide definition. A Consensus Statement from the
International Diabetes Federation. Diabet Med 23, 469–480.
18. Fujita K, Nishizawa H, Funahashi T, et al. (2006) Systemic
oxidative stress is associated with visceral fat accumulation
and the metabolic syndrome. Circ J 70, 1437 –1442.
19. Faul F, Erdfelder E, Buchner A, et al. (2009) Statistical
power analyses using G*Power 3.1: tests for correlation and
regression analyses. Behav Res Methods 41, 1149–1160.
20. Corretti MC, Anderson TJ, Benjamin EJ, et al. (2002) Guide-
lines for the ultrasound assessment of endothelial-dependent
flow-mediated vasodilation of the brachial artery: a report of
the International Brachial Artery Reactivity Task Force. JAm
Coll Cardiol 39, 257–265.
21. Inoue T, Matsuoka H, Higashi Y, et al. (2008) Flow-mediated
vasodilation as a diagnostic modality for vascular failure.
Hypertens Res 31, 2105–2113.
22. Friedewald WT, Levy RI & Fredrickson DS (1972) Estimation
of the concentration of low-density lipoprotein cholesterol
in plasma, without use of the preparative ultracentrifuge.
Clin Chem 18, 499 –502.
23. The Committee of Japan Diabetes Society on the Diagnostic
Criteria of Diabetes Mellitus (2010) Report of the Committee
on the classification and diagnostic criteria diabetes mellitus.
J Japan Diab Soc 53, 450–467.
24. Mannucci E, Ognibene A, Cremasco F, et al. (2001) Effect of
metformin on glucagon-like peptide 1 (GLP-1) and leptin
levels in obese nondiabetic subjects. Diabetes Care 24,
489–494.
25. Jenkins DJ, Wesson V, Wolever TM, et al. (1988) Wholemeal
versus wholegrain breads: proportion of whole or cracked
grain and the glycaemic response. BMJ 297, 958– 960.
26. Ito Y, Mizukuchi A, Kise M, et al. (2005) Postprandial blood
glucose and insulin responses to pre-germinated brown rice
in healthy subjects. J Med Invest 52, 159–164.
27. Pereira MA, Jacobs DR Jr, Pins JJ, et al. (2002) Effect of whole
grains on insulin sensitivity in overweight hyperinsulinemic
adults. Am J Clin Nutr 75, 848– 855.
28. Shimabukuro M, Higa N, Chinen I, et al. (2006) Effects of a
single administration of acarbose on postprandial glucose
excursion and endothelial dysfunction in type 2 diabetic
patients: a randomized cross-over study. J Clin Endocrinol
Metab 91, 837–842.
29. Jang Y, Lee JH, Kim OY, et al. (2001) Consumption of whole
grain and legume powder reduces insulin demand, lipid
peroxidation, and plasma homocysteine concentrations in
patients with coronary artery disease: randomized controlled
clinical trial. Arterioscler Thromb Vasc Biol 21, 2065–2071.
30. Zhang G, Pan A, Zong G, et al. (2011) Substituting white rice
with brown rice for 16 weeks did not substantially affect
metabolic risk factors in middle-aged Chinese men and
women with diabetes or a high risk for diabetes. J Nutr
141, 1685 –1690.
31. Fujioka S, Matsuzawa Y, Tokunaga K, et al. (1991) Improve-
ment of glucose and lipid metabolism associated with
selective reduction of intra-abdominal visceral fat in preme-
nopausal women with visceral fat obesity. Int J Obes 15,
853–859.
32. Ludwig DS, Pereira MA, Kroenke CH, et al. (1999) Dietary
fiber, weight gain, and cardio-vascular disease risk factors
in young adults. J Am Med Assoc 282, 1539–1546.
33. Thomas DE, Elliott EJ & Baur L (2007) Low glycaemic index
or low glycaemic load diets for overweight and obesity. The
Cochrane Database of Systematic Reviews 2007 issue 3,
CD005105.
34. McKeown NM, Meigs JB, Liu S, et al. (2004) Carbohydrate
nutrition, insulin resistance, and the prevalence of the
metabolic syndrome in the Framingham Offspring Cohort.
Diabetes Care 27, 538– 546.
35. Shimabukuro M, Higa M, Yamakawa K, et al. (2012) Miglitol,
a-glycosidase inhibitor, reduces visceral fat accumulation
and cardiovascular risk factors in subjects with the metabolic
syndrome: a randomized comparable study. Int J Cardiol
(epublication ahead of print version 19 June 2012).
36. Niwano Y, Adachi T, Kashimura J, et al. (2009) Is glycaemic
index of food a feasible predictor of appetite, hunger, and
satiety? J Nutr Sci Vitaminol 55, 201 –207.
37. Fukagawa NK, Anderson JW, Hageman G, et al. (1990) High-
carbohydrate, high-fiber diets increase peripheral insulin
sensitivity in healthy young and old adults. Am J Clin Nutr
52, 524 –528.
38. Koh-Banerjee P & Rimm EB (2003) Whole grain consump-
tion and weight gain: a review of the epidemiological
evidence, potential mechanisms and opportunities for
future research. Proc Nutr Soc 62, 25–29.
39. Adolphe JL, Drew MD, Huang Q, et al. (2012) Postprandial
impairment of flow-mediated dilation and elevated methyl-
glyoxal after simple but not complex carbohydrate
consumption in dogs. Nutr Res 32, 278 –284.
M. Shimabukuro et al.10
British Journal of Nutrition
40. Brook RD, Bard RL, Glazewski L, et al. (2004) Effect of
short-term weight loss on the metabolic syndrome and
conduit vascular endothelial function in overweight adults.
Am J Cardiol 93, 1012–1016.
41. Sekuri C, Tavli T, Avsar A, et al. (2003) The acute effect
of orlistat on endothelial function in young obese women.
Int J Clin Pharmacol Res 23, 111– 117.
42. Bergholm R, Tiikkainen M, Vehkavaara S, et al. (2003) Lowering
of LDL cholesterol rather than moderate weight loss improves
endothelium-dependent vasodilation in obese women with
previous gestational diabetes. Diabetes Care 26, 1667–1672.
43. Sasaki S, Higashi Y, Nakagawa K, et al. (2002) A low-calorie
diet improves endothelium-dependent vasodilation in obese
patients with essential hypertension. Am J Hypertens 15,
302–309.
44. Hamdy O, Ledbury S, Mullooly C, et al. (2003) Lifestyle
modification improves endothelial function in obese subjects
with the insulin resistance syndrome. Diabetes Care 26,
2119–2125.
45. Shimabukuro M (2009) Cardiac adiposity and global cardio-
metabolic risk: new concept and clinical implication. Circ
J 73, 27 –34.
Brown rice, obesity and endothelial function 11
British Journal of Nutrition
