Content uploaded by Paul Whelton
Author content
All content in this area was uploaded by Paul Whelton on Jan 03, 2015
Content may be subject to copyright.
366 Am J C/in Nutr 1995;61:366-72. Printed in USA. © 1995 American Society for Clinical Nutrition
Oats and buckwheat intakes and cardiovascular disease
risk factors in an ethnic minority of China13
Jiang He, Michael J K/ag, Paul K Whelton, Jing-Ping Mo, Jun-Yun Chen, Ming-Chu Qian,
Pei-Sheng Mo, and Guan-Qing He
ABSTRACT The relationship of oats and buckwheat in-
take to cardiovascular disease risk factors was studied in 850
Yi people, an ethnic minority in southwest China. Blood pres-
sure was measured on 3 consecutive days. Serum total choles-
tenol, high-density-lipoprotein (HDL) cholesterol, and tniglyc-
enides were measured after a 14-h fast. Oats and buckwheat
intakes were assessed by questionnaire. In multiple-regression
analysis, oats intake (100 g/d) was associated with lower body
mass index (-0.25, in kg/m2; P < 0.05), systolic (-3.1 mm
Hg, P < 0.001) and diastolic (- 1.3 mm Hg, P < 0.01) blood
pressure, and HDL cholesterol (-0.13 mmol/L, P < 0.001).
Buckwheat intake (100 g/d) was associated with lower serum
total cholesterol (-0.07 mmol/L, P<0.01) and low-density-
lipoprotein cholesterol (-0.06 mmol/L, P < 0.05) and a higher
ratio of HDL to total cholesterol (0.01, P < 0.05). These
findings suggest a role for oats and buckwheat consumption in
the prevention and treatment of both hypertension and hypen-
cholesterolemia. Am J C/in Nutr 1995;61:366-72
KEY WORDS Oats, buckwheat, dietary fiber, blood pres-
sure, serum cholesterol
Introduction
Although mortality from cardiovascular disease has declined
progressively in the United States during the past three de-
cades, it is still the leading cause of death. In 1989 an estimated
944 688 US residents died of cardiovascular disease, almost as
many as from all other causes of death combined (1). In many
economically developing countries, including China, cardio-
vascular disease mortality has increased rapidly and has be-
come the leading cause of death (2).
Elevations of blood pressure (3-5) and serum cholesterol
(6-8) are widely recognized as major modifiable risk factors
for cardiovascular disease. In addition, pharmacologic therapy
of hypertension and hypenlipidemia has been well established
as a means to prevent cardiovascular disease (9-14). However,
it has also become increasingly apparent that altered quality of
life, medication toxicity, and costs of medical care limit the
usefulness of lifelong drug therapy for hypertension and hy-
perlipidemia (15). Lifestyle modifications, especially dietary
interventions, constitute an important and complementary ap-
proach to the therapy for hypertension and hyperlipidemia in
the individual (15, 16). Dietary interventions assume an even
more important role in the population-based approach to car-
diovascular disease risk reduction. Here, the goal is to achieve a
small downward shift in the population distribution of blood
pressure and cholesterol by altering environmental exposures (15).
Oats, a grain rich in water-soluble fiber, has been shown to
lower serum lipids in animal and human experiments. De Groot
et al (17) reported in 1963 that 21 male volunteers who ate 140
g rolled oats daily for 3 wk experienced a decrease in their
average serum cholesterol from 6.49 to 5.77 mmolfL. Since
then, most (18-30) but not all (31, 32) clinical trials have
demonstrated that intake of oats lowers serum cholesterol in
hyperlipidemic patients (18-26) as well as in healthy adults
(27-29) and children (30). The relationship between oats intake
and serum lipids has not been well studied in population-based
samples. In addition, the effect of buckwheat, another cereal
high in water-soluble fiber (33), on serum lipids and blood
pressure has not been established.
The present study investigated the relationship between oats
and buckwheat intakes and cardiovascular disease risk factors
in an ethnic minority in China-the Yi people.
Subjects and methods
Subjects
The Yi Migrant Study is a population-based epidemiologic
investigation of cardiovascular disease risk factors in the Li-
angshan Yi People Autonomous Prefecture in southwest China.
As part of this study, serum lipids and lipopnoteins, blood
pressure, body weight, dietary habits, physical activity, smok-
ing, and alcohol consumption were measured in 857 Yi men
randomly selected from the community. Details of the study
1From the Welch Center for Prevention, Epidemiology, and Clinical
Research, the Johns Hopkins University Medical Institutions, Baltimore; the
Department of Epidemiology, Peking Union Medical College and Chinese
Academy of Medical Sciences and the National Center for Clinical Labora-
tories, Beijing; and Liangshan Yi People Autonomous Prefecture Anti-epi-
demic Station, Xichang City, Sichuan Province, People’s Republic of China.
2Supported by the Ministry of Public Health, People’s Republic of
China; Outpatient General Research Center grant 5M01RR00722 from the
National Institutes of Health; the Quaker Oats Company; and National
Institutes of Health grant RR00035.
3Address reprint requests to J He, The Welch Center for Prevention,
Epidemiology, and Clinical Research, 2024 East Monument Street, Suite
2-600, Baltimore, MD 21205-2223.
Received April 5, 1994.
Accepted for publication August 29, 1994.
by guest on July 14, 2011www.ajcn.orgDownloaded from
OATS AND CARDIOVASCULAR DISEASE 367
population and methods have been described elsewhere (34).
Briefly, the Yi people are an ethnic minority of China who live
principally in a remote mountainous area in southwest China
and who are mainly engaged in subsistence agriculture. Yi
farmers are relatively isolated from the outside world and have
preserved their own language and life style. Traditionally, their
main crops have been oats, buckwheat, and potatoes. Because
of limited land resources, Yi farmers now grow and consume
rice and corn as well. Consumption of meat is limited to
weddings, funerals, and semiannual celebrations. Less salt is
consumed than in other areas of China. Starting in the 1950s,
Yi farmers began to migrate to Xichang City, the capital of the
Liangshan Yi People Autonomous Prefecture, and to the
county seats of counties in the Prefecture. These urban mi-
grants primarily eat rice, meat, and fresh vegetables with little
oats and buckwheat intake.
The protocol for this study was reviewed and approved by
the Chinese Academy of Medical Sciences, Beijing, People’s
Republic of China, and the Committee on Human Research,
Johns Hopkins School of Hygiene and Public Health, Balti-
more. In accord with customary practice in China at the time of
the study, informed consent was not obtained.
Methods
Our study was conducted during March, 1989. A blood
sample was obtained in the morning after a 14-h fast. Serum
was separated in the field and sent to the county hospital
laboratory where it was stoned at -20 #{176}Cuntil it was air-mailed
to the National Center for Clinical Laboratories, Beijing. Se-
rum total cholesterol, high-density-lipoprotein (HDL) choles-
terol, and triglycerides were measured by an enzymatic method
(35, 36) (IL Monarch 2000, Instrumentation Laboratory, Lax-
ington, MA). HDL cholesterol was separated from serum by
precipitation after the addition of phosphotungstic acid and mag-
nesium ion. These analytic procedures were standardized and met
the performance requirements of the Lipoprotein Standardization
Program of the Centers for Disease Control, Atlanta.
Starting on the day that the blood sample was collected,
blood pressure was measured on 3 consecutive days. Before the
blood pressure measurement, participants refrained from stren-
uous activity, smoking, and eating for 30 mm. After a mm-
imum of5 mm ofquiet sitting, blood pressure was measured on
the right arm and recorded to the nearest 2 mm Hg by specially
trained physicians using standard mercury sphygmomanome-
tens. Systolic blood pressure was recorded at the appearance of
the first sound and diastolic pressure at the disappearance of
sounds (Korotkoff phase 5). Three measurements of blood
pressure were obtained each day, and the mean of the nine
blood pressure readings obtained oven the 3-d period of study
was used in the analysis. The observers were trained by using
training tapes from the National Heart, Lung, and Blood Insti-
tute of the National Institutes of Health. After the training
period, the observers were certified by comparing their blood
pressure readings on eight persons (three readings per person)
to those taken by an experienced observer (JH). The differ-
ences in mean readings between observers were all <2 mm Hg
(F test, P>0.95). During the field work, quality control was
maintained by having a supervisor repeat the blood pressure
measurement on a 10% random sample of subjects every day.
If the mean of the three readings differed from those taken by
the first observer by 5 mm Hg for either systolic or diastolic
blood pressure, blood pressure determinations were repeated
for all the subjects measured by that observer on that day. Height
and weight were measured and body mass index [weight (kg)/
height (m2)] was calculated as an index of obesity (37).
Age, race, sex, education level, smoking, medical history
(including antihypertensive medication use), and intakes of
oats and buckwheat were ascertained by local physicians fluent
in both the Yi and Chinese languages. The intake of oats (and
buckwheat) was assessed by asking “How many Jin of oats
(buckwheat) did you eat in the past year?.” A Jin is a Chinese
unit of measure equivalent to 500 g. Information on diet was
also obtained by means of a 24-h dietary recall administered on
3 consecutive days. Agreement between the estimates obtained
by these two methods was moderate, with a correlation coef-
ficient of 0.41 (P < 0.001) for oats and 0.61 (P <0.001) for
buckwheat. This disagreement is primarily due to variability in
intake of these grains. Among the 424 participants who re-
ported eating oats during the preceding year, only 65 (15%)
consumed oats at the time of the 3-d dietar” recall. Likewise,
only 286 (54%) of 531 participants who att buckwheat in the
past year also included it in their diet at this time. The analyses
were based on annual intake of oats and buckwheat, instead of
24-h dietary recall, to reduce the variability in the estimates of
oats and buckwheat intakes. Intakes of total energy, protein,
fat, saturated fatty acids (SFAs), monounsatunated fatty acids
(MUFAs), polyunsaturated fatty acids (PUFAs), cholesterol,
and total fiber were calculated by using the Sichuan Province
section of the Chinese Food Composition Tables (33). In
addition, oats and buckwheat specimens were collected and
their fiber content analyzed at the John Stuart Research Labo-
ratonies, Quaker Oats Company, Barrington, IL. Assessment of
usual physical activity was based on occupational activity
because leisure-time activity was almost nonexistent.
The 857 Yi people in the present study included 515 Yi
farmers and 342 Yi migrants. Seven Yi migrants were excluded
from the analysis because they had been diagnosed as having
hyperlipidemia and took oats for therapy. Distributions of body
weight, blood pressure, and serum lipids were examined, by
intake of grains. Intakes of oats and buckwheat were divided
into four groups: no intake and low, middle, and upper tertiles
of intake. For example, oats intake was categorized as none,
<25 g/d, 25-90 g/d, and >9C g/d. Buckwheat intake was
categorized as none, <40 g/d, 40-200 g/d, and >200 g/d.
Statistical analysis
The differences in cardiovascular disease risk factors and
dietary nutrients among the oats and buckwheat intake groups
were examined by analysis of variance. Univaniate and multi-
variate linear-regression analyses were used to explore the
relation of oats and buckwheat intakes to cardiovascular dis-
ease risk factors. Because any beneficial effects of these grains
on cardiovascular disease risk factors were hypothesized to be
mediated through their soluble fiber content, the relation of risk
factors to fiber intake was also examined. To account for a
possible community effect, additional multivaniate analyses
were performed by adjusting for two urban areas: Xichang City
and the county seats and rural area of residence (Yi farmers).
All analyses were performed by using the SAS statistical anal-
ysis package (38).
by guest on July 14, 2011www.ajcn.orgDownloaded from
368 HE ET AL
Results
The ages of the 850 participants ranged from 15 to 77 y, with
a mean of 34 y. The study participants were lean, with an
average body mass index of 20.8. The mean systolic and
diastolic blood pressures were 107.2 and 66.9 mm Hg, respec-
tively. Only 21 (2.5%) participants had an elevated blood
pressure as defined by a mean systolic or diastolic blood
pressure 140 on 90 mm Hg, respectively, and none of the
study participants were taking antihypertensive medications.
Mean serum total cholesterol (3.86 mmol/L) and triglyceride
(1 .55 mmolfL) values were in the low normal range, by West-
ern standards. Seventy five (8.8%) participants had a total
cholesterol concentration >5.20 mmol/L. The mean HDL-
cholesterol concentration was 1 .43 mmol/L and the ratio of
HDL to total cholesterol was relatively high at 0.39.
Forty-four percent of the sample had consumed oats and
55% had consumed buckwheat during the preceding year. Of
those who had consumed these foods, the mean daily intake
was 70.3 g for oats and 137.7 g for buckwheat.
Distributions of the covariables for oats and buckwheat
consumption are summarized in Tables 1 and 2. Mean age and
dietary intakes of fat, cholesterol, and sodium tended to be
lower in persons who ate oats on buckwheat, whereas their
dietary ratio of polyunsaturated to saturated fatty acids, potas-
sium intake, and alcohol consumption were higher. In addition,
dietary energy intake was significantly greater in those who ate
buckwheat than in those who did not (Table 2).
Average body mass index, blood pressure, and lipid profile
by oats and buckwheat intake are presented in Tables 3 and 4.
At higher oats intake, mean body mass index and blood pnes-
sure were lower, in a dose-response pattern (Table 3). Mean
serum total cholesterol and low-density-lipopnotein cholesterol
were also lower in those who ate oats, although mean concen-
trations were somewhat higher (P > 0.05) in those who ate
>90 g/d compared with their counterparts who ate 26-90 g/d.
Consumption of 25 g oats/d was associated with lower mean
HDL-cholesterol concentrations. The ratio of HDL to total
cholesterol varied little by oats intake.
Average values for body mass index, systolic and diastolic
blood pressure, and HDL cholesterol also varied in a statisti-
cally significant fashion across categories of buckwheat intake,
but did not demonstrate a dose-response relationship (Table 4).
In contrast, mean low-density-lipoprotein (LDL) cholesterol con-
centrations were markedly lower, and the ratio of HDL to total
cholesterol higher, with successively higher buckwheat intakes.
The results of univaniate and multivaniate linear-regression
analyses of oats and buckwheat intakes on cardiovascular dis-
ease risk factors are shown in Table 5. After adjustment for the
covaniables listed in the table, oats intake continued to be
inversely associated with a lower body mass index, as well as
with lower systolic and diastolic blood pressure, serum HDL
cholesterol, ratio of HDL to total cholesterol, and serum tn-
glyceride. Buckwheat intake was associated with lower levels
of serum total and LDL cholesterol, as well as with a higher
ratio of HDL to total cholesterol. In an analysis adjusted for
community of residence, oats intake (100 g/d) was significantly
related to body mass index (/3 =-0.28, P=0.01), systolic
(13=-2.4 mm Hg, P < 0.0001) and diastolic (/3 --1.0 mm
Hg, P=0.04) blood pressure, HDL cholesterol concentration
(13 =-0.12 mmol/L, P < 0.0001), the ratio of HDL to total
cholesterol (/3 =-0.04, P < 0.0001), and triglyceride con-
centration (/3 =-0.09 mmol/L, P=0.05), whereas buckwheat
intake (100 g/d) was associated with total cholesterol concen-
tration (13 =-0.05 mmol/L, P<0.05), the ratio of HDL to
total cholesterol (/3 =0.01, P=0.004), and LDL cholesterol
concentration (/3 =-0.05 mmol/L, P=0.03).
Yi farmers tended to eat more oats and buckwheat than
migrants. For example, 49. 1% of those who did not eat oats
were Yi farmers compared with 91.1% of those who ate >90 g/d.
In separate analyses in the urban and rural groups, however,
similar patterns of associations of grain intake with cardiovascular
disease risk factors were seen in both groups (data not shown).
Because it has been suggested that a threshold effect may
exist for the effect of oats on the lipid profile, persons who
ate 25 g oats/d were compared with those who ate <25 g
oats/d. Findings were similar to the comparisons listed in
Tables 3-5. After adjustment for the covaniables listed in Table
5, systolic blood pressure was 5.3 mm Hg lower, diastolic
blood pressure was 1.8 mm Hg lower, total cholesterol was
0.21 mmol/L lower, HDL cholesterol was 0.20 mmolfL lower,
TABLE 1
Age and intake of dietary nutrients in 850 study subjects by oats intake’
Oats intakes
Group 1, 0 g/d Group 2, <25 g/d Group 3, 25-90 g/d Group 4, >90 g/d
Covariables (n = 426) (n = 176) (n 125) (n = 123) P
Age (y) 35 ±132 34 ± 122 34 ± 142 30 ± i3 0.001
Dietary energy (kJ/d) 13100 ± 4121 13515 ±4577 14481 ±4661 13042 ±5561 0.26
Dietary fat (g/d) 84.4 ± 51.425 85.8 ± 58.525 61.8 ± 5492 41.9 ± 40.2’ 0.001
P:S 1.37 ±1.1625 1.29 ±1.1125 2.14 ±1.12 2.08 ±1.04 0.001
Dietary cholesterol (mg/d) 214 ± 310 186 ± 21725 86 ± 204’ 38 ± 83’ 0.001
Dietary sodium (mgjd) 3539 ± 312725 3851 ± 341425 2304 ± 1974 2738 ± 181 0.001
Dietary potassium (mg/d) 4387 ± 293625 4518 ± 292925 7410 ± 38212 6651 ± 3616 0.001
Alcohol intake (g/d) 25.6 ± 40.124 35.6 ± 52.3 30.5 ± 51.5 41.0 ± 59.9’ 0.003
‘I ± SD. P:S, polyunsaturated to saturated fatty acid ratio.
2Significantly different from group 4, P<0.05.
3Significantly different from group 1, P<0.05.
4Significantly different from group 2, P<0.05.
5Significantly different from group 3, P<0.05.
by guest on July 14, 2011www.ajcn.orgDownloaded from
OATS AND CARDIOVASCULAR DISEASE 369
TABLE 2
Age and intake of dietary nutrients in 850 study subjects by buckwheat intake’
Buckwheat intakes
Group 1, 0 g1/d Group 2, <40 g/d Group 3, 40-200 g/d Group 4, >200 g/d
Covariables (n = 319) (i = 207) (n = 161) (n = 163) P
Age (y) 36 ±1323 34 ± 13’ 32 ± 13 30 ± l2 0.0()1
Dietary energy (kJ/d) 12531 ± 41 13235 13498 ± 4448 14017 ± 5059 14485 ± 4762 0.001
Dietary fat (g/d) 82.8 ± 58.723 91.7 ± 55523 60.4 ±44#{149}74.5 539 ± 41.9 0.001
P:S 1.44 ±1.17235 1.07 ± 1.0O” 1.86 ±1.15 2.17 ±1.06245 0.001
Dietary cholesterol (mg/d) 205 ± 27723 248 ± 33123 84 ± 145 49 ± I l6 0.001
Dietary sodium (mg/d) 3491 ± 3234235 4185 ± 32282 2919 ± 2224’’ 2231 ± 2030245 0.001
Dietary potassium (mg/d) 4020 ± 275923 4084 ± 276623 6547 ± 3288’ 7569 ± 3627245 0.()01
Alcohol intakes (g/d) 20.9 ± 33.2235 31.9 ± 46.824 41.9 ± 4J.545 39.3 ± 57.2 0.0()1
‘.t ± SD. P:S, polyunsaturated to saturated fatty acid ratio.
2Significantly different from group 3, P<0.05.
3Significantly different from group 4, P<0.05.
4Significantly different from group 1, P<0.05.
5Significantly different from group 2, P<0.05.
TABLE 3
Cardiovascular disease risk factors in 850 study subjects by oats intake’
Oats intakes
Group 1, 0 g/d Group 2, <25 g/d Group 3, 25-90 g/d Group 4, >90 g/d
Variables (n = 426) (n 176) (n 125) (n = 123) P
Body mass index 20.9 ± 2.323 20.9 ± 2.323 20.5 ± 20.1 ± 0.001
Systolic BP (mm Hg) 109.7 ± 12.42 108.5 ± 13.023 103.9 ± 10.6’s 100.4 ±9924.5 0.001
Diastolic BP (mm Hg) 68.3 ± 10.323 67.2 ± 10.1’ 65.9 ± 62.6 ± 8.9243 0.001
Serum total cholesterol (mmol/L) 4.03 ±0.9923 399 ± 1.0423 3.45 ± 0.91’ 3.59 ± 0.86’ 0.001
HDL cholesterol (mmol/L) 1.49 ± 0.452 1.48 ± 0.4423 1.34 ± 0.4O’ 1.23 ± 0.28245 0.001
LDL cholesterol (mmol/L) 1.85 ±0.942 1.78 ± 0.862 1.54 ± 0.85 1.76 ± 0.84 0.01
HDL:total cholesterol 0.39 ± 0.14 0.39 ± 0.14 0.40 ± 0.14 0.36 ± 0.12 0.1
Serum triglyceride (mmol/L) 1.59 ±0.97 1.67 ± l.01 1.49 ± 0.86 1.32 ± 0.61 0.007
‘I ± SD. BP, blood pressure. Body mass index in kg/m2.
2Significantly different from group 3, P<0.05.
3Significantly different from group 4, P<0.05.
4Significantly different from group 1, P<0.05.
SSignificantly different from group 2, P<0.05.
and the ratio of HDL to total cholesterol was 0.03 lower (all P <
0.01) in subjects who had eaten 25 g oats/d compared to those
who had not. Likewise, the adjusted differences between those
who had eaten 44J g buckwheat/d and had not or eaten buck-
wheat <40 g/d were -2.6 mm Hg for systolic blood pressure,
-0.22 mmol/L for total cholesterol, -0.24 mmol/L for LDL
cholesterol, and 0.03 for the ratio of HDL to total cholesterol (all
P<0.01).
Laboratory analysis demonstrated that each 100 g oats con-
tamed 10.2 g total fiber, 3.9 g soluble fiber, 13.9 g protein, and
9.9 g fat. For each 100 g buckwheat, there were 26.0 g total
fiber, I .7 g soluble fiber, 9.2 g protein, and 1 .9 g fat. The
results of the univaniate and multivariate linear-regression anal-
yses of total and soluble fibers from oats and buckwheat on
cardiovascular disease risk factors are shown in Table 6. After
adjustment for the covariables listed in the table, water-soluble
fiber was negatively and significantly related to systolic blood
pressure, serum total cholesterol, and HDL cholesterol,
whereas total fiber was related negatively to serum total and
LDL cholesterol. In the multivaniate model adjusted for com-
munity of residence, water-soluble fiber was negatively and
significantly associated with systolic blood pressure, total cho-
lesterol, and HDL cholesterol, but total fiber was not associated
with blood pressure or lipids.
Discussion
Because the population in the present study consumed fairly
high amounts of oats and buckwheat from an early age, the
results seen here probably represent the long-term effect of
these grains on cardiovascular disease risk factors. Special
strengths ofthe present study include its relatively large sample
size as well as the availability of detailed information on
dietary consumption and cardiovascular disease risk factors. In
addition, the fact that the results were from a random sample of
free-living men enhances the generalizability of the findings.
The present study identified a significantly lower mean sys-
tolic and diastolic blood pressure with progressively higher
intakes of oats. This association followed a dose-response
pattern and was independent of age, body mass index, alcohol
use, as well as dietary intakes of energy, cholesterol, sodium,
by guest on July 14, 2011www.ajcn.orgDownloaded from
370 HE ET AL
TABLE 4
Cardiovascular disease risk factors in 850 study subjects by buckwheat intake’
Buckwheat intakes
Group 1, 0 g/d Group 2, <40 g/d Group 3, 40-200 g/d Group 4, >200 g/d
Variables (n 319) (n 207) (n = 161) (n = 163) P
Body mass index 20.7 ± 2.22 21.3 ± 20.4 ± 2.12 20.5 ± 1.92 0.001
Systolic BP (mm Hg) 107.6 ± 12.3245 111.4 ± 14.1 104.7 ± 10.723 103.4 ± 9.623 0.001
Diastolic BP (mm Hg) 67.6 ± 68.3 ± i0.95 65.8 ± 10.723 64.6 ± 8.523 0.001
Serum total cholesterol (mmol/L) 4.03 ± 1.03 4.10 ± i.O75 3.71 ± 0.85235 3.38 ± 0.782 0.001
HDL cholesterol (mmol/L) 1.37 ± 0442.4 1.52 ± O.44 1.48 ± 0.43’ 1.38 ± 0.3624 0.001
LDL cholesterol (mmollL) 1.99 ± #{216}9545 1.86 ± 0.91 1.57 ± 0.8323 1.43 ± 0.7223 0.001
HDL:total cholesterol 0.35 ± 0.13245 0.39 ± 0.14 0.41 ± 0.14 0.43 ± 0.1523 0.001
Serum triglyceride (mmol/L) 1.50 ± 0.86 1.65 ± 1.06 1.63 ± 0.97 1.47 ± 0.78 0.1
‘g ± SD. BP, blood pressure. Body mass index in kg/m2.
2Significantly different from group 2, P<0.05.
3Significantly different from group 1, P<0.05.
4Significantly different from group 3, P<0.05.
5Significantly different from group 4, P<0.05.
TABLES
Unadjusted and adjusted differences (linear-regression analysis) in cardiovascular disease risk factors for 100-g/d intakes of oats and buckwheat’
Oats intakes Buckwhe at intakes
Unadjusted Adjusted Unadjusted Adjusted
Variables /3 SE /3 SE f3 SE j3 SE
Body mass index -0.47 0.112 -0.25 0.ii3 -0.09 0.06 0.04 0.06
Systolic BP (mm Hg) -4.3 0.62 -3.1 0.62 -1.2 0.32 -0.3 0.3
Diastolic BP (mm Hg) -2.5 #{149}52 -1.3 54 -0.7 0.2 0.1 0.2
Serum total cholesterol (mmol/L) -0.19 #{216}#{216}52 #{216}4 0.04 -0.17 0.022 -0.07 0.02
HDL cholesterol (mmol/L) -0.13 0.022 -0.13 0.022 -0.01 0.01 -0.00 0.01
LDL cholesterol (mmolIL) -0.03 0.05 0.07 0.05 -0.14 0.022 -0.06 0.02
HDL:total cholesterol -0.01 0.01 -0.03 0.012 0.02 0.00 0.01 0.00
Serum triglyceride (mmolfL) -0.14 #{216}#{149}#{216}54 -0.10 0.O4 -0.01 0.02 0.02 0.02
‘All analyses adjusted for age and dietary intakes of energy, cholesterol, and alcohol. In addition, body mass index (in kg/m2) was adjusted for dietary
fat and physical activity, blood pressure was adjusted for body mass index and dietary intake of sodium and potassium, and serum lipids were adjusted for
body mass intake, dietary intake of fat, and the ratio of polyunsaturated to saturated fatty acids. BP, blood pressure.
2-4 Statistically significant: 2p<0.001, p <0.05, p <0.01.
and potassium. Dietary buckwheat was associated with lower
blood pressure in univanate analysis, but this was not the case
after adjustment for other covaniables. Further analysis mdi-
cated that water-soluble fiber, but not total fiber, was indepen-
dently related to blood pressure. Previous studies of the asso-
ciation between dietary fiber and blood pressure have yielded
inconsistent results (39-41). In clinical trials, oats supplemen-
tation did not alter blood pressure (22, 23, 31). In all of these
trials, however, blood pressure was not the primary outcome
variable and hypertensive patients were excluded (22, 23, 31).
Apossible mechanism by which oats intake might lower blood
pressure is through alterations in insulin metabolism. Higher
insulin concentrations and insulin resistance have been linked to
higher blood pressures and hypertension (42, 43). In healthy
persons, consumption of oat bran has been demonstrated to lower
plasma insulin concentrations and to ameliorate insulin resistance
(44, 45).
Higher buckwheat intake was associated with lower total and
LDL serum cholesterol, with no effect on HDL cholesterol.
Thus, the ratio of HDL to total cholesterol was higher at higher
buckwheat intakes, independent of other variables. The asso-
ciation of oats intake with serum lipids seen in univariate
results in the present study is consistent with the results of most
clinical trials. Ripsin (46) pooled the results of 12 randomized,
controlled trials, which had evaluated the lipid-lowering effects
of oats in free-living subjects. The overall effect of oats intake
was to lower serum total cholesterol by 0.15 mmollL (95% CI:
-0.22 to -0.09), with a more pronounced effect in persons
with a higher intake of oats and a higher initial total serum
cholesterol concentration. In the present study, after adjustment
for the other covariables, oats intake was not significantly associ-
ated with either serum total or LDL-cholesterol concentrations,
but was associated with lower concentrations of HDL cholesterol
and triglyceride.
Whether dietary oats lower serum cholesterol concentrations
by substitution of carbohydrates fon saturated fats or by a direct
effect of the dietary fiber contained in oats remains an unan-
swered question. Swain et al (3 1) compared the effects of oats
and low-fiber wheat diets on serum lipids. Both types of
supplements lowered serum total cholesterol concentrations by
an average of 7-8% compared with baseline. However, a
decrease in LDL cholesterol in the oats group was accompa-
nied by a rise in HDL cholesterol. The low-fiber diet, on the
other hand, lowered HDL-cholestenol concentrations. The pan-
ticipants ate less saturated fat and cholesterol and more poly-
unsaturated fat during both periods of supplementation than at
by guest on July 14, 2011www.ajcn.orgDownloaded from
OATS AND CARDIOVASCULAR DISEASE 371
TABLE 6
Unadjusted and adjusted differences (linear-regression analysis) in cardiovascular disease risk factors for 10-g/d intakes of water-soluble or total fiber
from oats and buckwheat’
Water-sol uble fiber Total fiber
Unadjusted Adjusted Unadjusted Adjusted
Variables f3 SE /3 SE 3 SE 3 SE
Body mass index -0.59 0.172 -0.19 0.18 -0.04 0.02’ 0.00 0.02
Systolic BP (mm Hg) -6.0 0.92 -3.7 1.0 -0.5 0.12 -0.21 0.11
Diastolic BP (mm Hg) -3.5 0.82 1.2 0.7 -0.3 0.i 0.02 0.08
Serum total cholesterol (mmollL) -0.46 0.072 -0.16 0.07 -0.06 0.012 -0.02 0.01
HDL cholesterol (mmolfL) -0.13 0.032 -0.12 0.042 -0.01 0.00 -0.00 0.00
LDL cholesterol (mmolfL) -0.26 0.072 -0.01 0.07 -0.04 0.012 -0.02 0.0i3
HDL:total cholesterol 0.02 0.01 -0.01 0.01 0.005 0.0012 0.002 0.001
Serum triglyceride (mmol/L) -0.14 0.07 -0.07 0.07 -0.01 0.01 0.00 0.01
IAll analyses adjusted for age and dietary intake of energy, cholesterol, and alcohol. In addition, body mass index (in kg/rn2) was adjusted for dietary
fat and physical activity, blood pressure was adjusted for body mass index and dietary intake of sodium and potassium, and serum lipids were adjusted for
body mass intake, dietary intake of fat, and the ratio of polyunsaturated to saturated fatty acids. BP, blood pressure.
2-4 Statistically significant: 2p<0.001, ?p < 0.05, p<0.01.
baseline (31). Most (80%) of the study participants were
women and their average serum cholesterol concentration was
in the desirable range, 4.81 mmollL. Studies in hypercholes-
terolemic men, on the other hand, have shown a greater reduc-
tion in serum lipids with oats than with wheat, rice, on corn
supplementation (21-24). Recent studies suggest that -glucan,
a water-soluble fiber, is the hypocholesterolemic component in
oats (26, 47, 48). In our study, both total and water-soluble
fiber from oats and buckwheat were significantly and indepen-
dently associated with lower serum total cholesterol, even
though the average serum cholesterol in the study population
was low. Soluble fiber may lower serum cholesterol concen-
trations through several mechanisms. It has been suggested that
soluble fiber binds strongly to bile acids and increases fecal
bile acid excretion (27, 49, 50). The loss of bile acids creates a
demand for cholesterol to facilitate the synthesis of bile acids
and diverts the available cholesterol from lipoprotein synthesis.
It also decreases the availability of bile acids for optimal fat
digestion and absorption (27, 47, 48, 51). In addition, fiber delays
gastric emptying, slowing access of nutrients to digestive enzymes
and to absorptive surfaces of the small intestine (52). Colonic
bacteria also ferment soluble fiber to form short-chain fatty acids,
which decrease hepatic cholesterol synthesis (53). In addition to
fiber, short-chain fatty acids present in oats also may play a role in
lowering serum lipids (54).
In the present study, oats intake was associated with a lower
mean HDL-cholesterol concentration. Most clinical trials, in
contrast, have shown that oats intake selectively lowers serum
LDL cholesterol without affecting HDL cholesterol (18, 19,
21-26, 28). Only one study identified a decrease in HDL
cholesterol with oat bran supplementation (20). The results
may differ in the present study because of the low average
concentrations of LDL cholesterol in the study population.
Even though a higher oats intake was associated with a lower
ratio of HDL to total cholesterol in multiple analysis, the
absolute ratio was very high. Thus, the HDL cholesterol low-
ening effect of oats intake was associated with minimal, if any,
additional cardiovascular disease risk. Because higher concen-
tnations of HDL cholesterol protect against coronary heart
disease (47, 48), the long-term effect of oats intake on HDL
cholesterol deserves more study.
Our study suggests that intake of grains rich in water-soluble
fiber in the presence of reasonable amounts of total fiber, like
oats and buckwheat, are associated with lower serum choles-
terol and blood pressure values. These results imply that a
high-fiber diet may be a useful part of a lifestyle modification
to prevent cardiovascular diseases. The effect of dietary oats on
blood pressure and the association of buckwheat intake with
cardiovascular disease risk factors should be investigated fun-
ther in clinical trials. U
We express our appreciation to the Department of Public Health and
Anti-epidemic Stations of Liangshan Yi People Autonomous Prefecture,
Butuo, Meigu, and Zhaojue counties, the People’s Republic of China, for
their help in performing field work, and the National Center for Clinical
Laboratories, People’s Republic of China, for their help in performing
laboratory work.
References
1.National Center for Health Statistics and the American Heart Associ-
ation. Facts about cardiovascular disease. Circulation 1992;85:A103
(abstr).
2. World Health Organization. 1990 World health statistics annual. Ge-
neva: World Health Organization, 1991.
3. Stokes J III, Kannel WB, Wolf PA. Blood pressure as a risk factor for
cardiovascular disease. The Framingham study-30 years of follow-
up. Hypertension 1989;12(suppl 1):I-13-8.
4. MacMahon 5, Peto R, Cutler J, et al. Blood pressure, stroke, and
coronary heart disease, part 1: prolonged differences in blood pressure:
prospective observational studies corrected for the regression dilution
bias. Lancet 1990;335:765-74.
5. Stamler J, Stamler R, Neaton JD. BlOOd pressure, systolic and dia-
stolic, and cardiovascular risks: U.S. population data. Arch Intern Med
1993;153:598-615.
6. Castelli WP, Anderson K, Wilson PWF, Levy D. Lipids and risk of
coronary heart disease. The Framingham Study. Ann Epidemiol 1992;
2:23-8.
7. Reed D, Benfante R. Lipid and lipoprotein predictors of coronary heart
disease in elderly men in the Honolulu Heart program. Ann Epidemiol
1992;2:29-34.
8. KIag Mi, Ford DE, Mead LA, He J, Whelton PK, Levine DM. Serum
cholesterol in youth and subsequent cardiovascular disease. NEngI J
Med 1993;328:313-8.
9. SHEP Cooperative Research Group. Prevention of stroke by antihy-
by guest on July 14, 2011www.ajcn.orgDownloaded from
372 HE ET AL
pertensive drug treatment in older persons with isolated systolic hy-
pertension. Final results of the systolic hypertension in the elderly
program. JAMA 1991;265:3244-64.
10. Hypertension Detection and Follow-up Program Cooperative Group.
Persistence of reduction in blood pressure and mortality of participants
in the Hypertension Detection and Follow-up Program. JAMA 1988;
259:2113-22.
11.Collins R, Peto R, MacMahon 5, et al. Blood pressure, stroke, and
coronary heart disease, II: short-term reductions in blood pressure:
overview of randomized drug trials in their epidemiological context.
Lancet 1990;335:827-38.
12. Hebert P, Moser M, Mayer J, Glynn Ri, Hennekens CH. Recent
evidence on drug therapy of mild to moderate hypertension and de-
creased risk of coronary heart disease. Arch Intern Med 1993;153:
578-81.
13. Lipid Research Clinics Program. The Lipid Research Clinics Coronary
Primary Prevention Trial results, I: reduction in incidence of coronary
heart disease. JAMA 1984;251:351-64.
14. Fnick MH, Elo 0, Haapa K, et al. Helsinki Heart Study: primary-
prevention trial with gemfibrozil in middle-aged men with dyslipi-
demia: safety of treatment, change in risk factors, and increased
incidence ofcoronary heart disease. N EngI J Med 1987;317:1237-45.
15. Whelton PK, Adams-Campbell LL, Appel Li, et al. National High
Blood Pressure Education Program Working Group report on primary
prevention of hypertension. Arch Intern Med 1993;153:186-208.
16. Joint National Committee on Detection, Evaluation, and Treatment of
High Blood Pressure. The fifth report of the Joint National Committee
on the Detection, Evaluation, and Treatment of High Blood Pressure
(JNC V). Arch Intern Med 1993;153:154-83.
17. Dc Groot AP, Luyken R, Pikaar NA. Cholesterol-lowering effect of
rolled oats. Lancet 1963;2:303-4.
18. Kirby RW. Oat-bran intake selectively lowers serum low-density
lipoprotein cholesterol concentrations of hypercholesterolemic men.
Am J Clin Nutr 1981;34:824-9.
19. Anderson JW, Story L, Siding B, Chen W-JL, Petro MS. Story J.
Hypocholesterolemic effects of oat-bran or bean intake for hypercho-
lesterolemic men. Am J Clin Nutr 1984;40:1 146-55.
20. Demark-Wahnefried W, Bowering J, Cohen PS. Reduced serum cho-
lesterol with dietary change using fat-modified and oat bran supple-
mented diets. J Am Diet Assoc 1990;90:223-9.
21 .Anderson JW, Spencer DB, Hamilton CC, et al. Oat-bran cereal lowers
serum total and LDL cholesterol in hypercholesterolemic men. Am J
Clin Nutr 1990;52:495-9.
22. Kestin M, Moss R, Clifton PM, Nestel PJ. Comparative effects of three
cereal brans on plasma lipids, blood pressure, and glucose metabolism
in mildly hypercholesterolemic men. Am J Clin Nutr 1990;52:661-6.
23. Horn LV, Moag-Stahlberg A, Liu K, et al. Effects on serum lipids of
adding instant oats to usual American diets. Am J Public Health
1991;81: 183-8.
24. Anderson JW, Gilinsky NH, Deakins DA, et al. Lipid response of
hypercholesterolemic men to oat-bran and wheat bran intake. Am J
Clin Nutr 1991;54:678-83.
25. Spiller GA, Farquhar JW, Gates TE, Nichols SF. Guar gum and plasma
cholesterol. Effect of guar gum and oat fiber source on plasma lipopro-
tein and cholesterol in hypercholesterolemic adults. Arteniosclen
Thromb 1991;11:1204-8.
26. Davidson MH, Dugan LD, Burns JM, Bova J, Story K, Drennan KB.
The hypocholesterolemic effects of j3-glucan in oatmeal and oat bran.
A dose-controlled study. JAMA 1991;265:1833-9.
27. Judd PA, Truswell AS. The effect of rolled oats on blood lipids and
fecal steroid excretion in man. Am J Clin Nutr 1981;34:2061-7.
28. Horn LW, Liu K, Parker D, et al. Serum lipid response to oat product
intake with a fat-modified diet. J Am Diet Assoc 1986;86:759-64.
29. Horn LW, Emidy LA, Liu K, et al. Serum lipid response to a fat-
modified, oatmeal-enhanced diet. Prey Med 1988;17:377-86.
30. Gold K, Wong N, Tong A, et al. Serum apolipoprotein and lipid profile
effects of an oat-bran-supplemented, low-fat diet in children with
elevated serum cholesterol. Ann NY Acad Sci 1991;623:429-31.
31. Swain iF, Rouse IL, Curley CB, Sacks FM. Comparison of the effects
of oat bran and low-fiber wheat on serum lipoprotein levels and blood
pressure. N Engl J Med 1990;322: 147-52.
32. Birkeland 1G. Gullestad L, Falch D, Torsvik H. Oat bran and serum
cholesterol. N EngI J Med 1990;332: 1748-9 (letter).
33. Institute of Nutrition of Chinese Academy of Medical Sciences. Chi-
nese Food Composition Tables. 3rd ed. Beijing: People’s Health Press,
1985.
34. He J, Klag Mi, Whelton PK, et al. Migration, blood pressure pattern,
and hypertension: the Yi Migrant Study. Am J Epidemiol 1991;134:
1085-101.
35. Allain CC, Poon LS, Chan CSG, Richmond W, Fu PC. Enzymatic
determination of total serum cholesterol. Clin Chem 1974;20:470-5.
36. McGowan MW, Artiss JD, Strandbergh DR, Zak B. A peroxidase-
coupled method for the calorimetric determination of serum triglycer-
ides. Clin Chem 1983;29:538-42.
37. Keys A, Fidanza F, Karvonen Mi, Kumura N, Taylor HL. Indices of
relative weight and obesity. J Chron Dis 1972;25:329-43.
38. SAS Institute Inc. SAS/STAT user’s guide, release 6.03. Cary, NC:
SAS Institute mc, 1990.
39. Margetts BM, Beilin U, Vandongen R, Armstrong BK. A randomized
controlled trial of the effect of dietary fiber on blood pressure. Clin Sci
1987;72:343-50.
40. Schlamowitz P, Halberg T, Warnoe 0, Ryttig K. Treatment of mild to
moderate hypertension with dietary fiber. Lancet 1987;2:622-3.
41. Rossner 5, Anderson I-L, Ryttig K. Effects of a dietary fiber supple-
ment to a weight reduction programme on blood pressure. A random-
ized, double-blind, placebo-controlled study. Acta Med Scand 1988;
223:353-7.
42. Ferrannini E, Shieh SM, Fuh MT. et al. Insulin resistance in essential
hypertension. N Engl J Med 1987;317:350-7.
43. Ferrannini E, Haffner SM, Stern MO. Essential hypertension: an
insulin-resistance state. J Cardiovasc Parmacol 1990;15(suppl 5):s18-
25.
44. Braaten IT, Wood PJ, Scott FW, Riedel KD, Poste LM, Collins MW.
Oat gum lowers glucan and insulin after an oral glucose load. Am J
Clin Nutr 1991;53:1425-30.
45. Wood PJ. Physicochemical properties and physiological effects of the
(1-*3) (1-’4)--D-Glucan from oats. In: Furda I, Brine Ci, eds. New
developments in dietary fiber. New York: Plenum Press, 1990.
46. Ripsin CM. Oat products and lipid-lowering: a meta-analysis. Circu-
lation 1992;85:864(abstr).
47. Gordon T, Castelli WP, Hjortland MC, Kannel WB, Dawber TR. High
density lipoprotein as a protective factor against coronary heart dis-
ease. The Framingham Study. Am J Med 1977;62:707-14.
48. Shinnick FL, Ink SL, Marlett JA. Dose-response to a dietary oat bran
fraction in the cholesterol-fed rat. J Nutr 1990;120:561-86.
49. Story JA, Kritchevsky D. Comparison of the binding of various bile
acids and bile salts in vitro by several types of dietary fiber. J Nutr
1976;106:1292-4.
50. Vahouny GV, Tombes R, Cassidy MM, Knitchevsky D, Gallo LL.
Dietary fiber V: Binding of bile salts, phospholipids and cholesterol
from mixed micelles by bile acid sequestrants and dietary fibers.
Lipids 1980;15:1012-8.
51. Story JA. Dietary fiber and lipid metabolism. Proc Soc Exp Biol Med
1985;180:447-52.
52. Anderson JW. Hypocholesterolemic effects of oat products. In: Furda
I, Brine CI, eds. New developments in dietary fiber. New York:
Plenum Press, 1990.
53. Nyman MG-L, Asp N-GL. Fermentation of oat fiber in the rat intes-
tinal tract: a study of different cellular areas. Am J Clin Nutr 1988;
48:274-8.
54. Anderson JW, Gustafson NJ. Hypocholesterolemic effects of oats and
bean products. Am J Clin Nutr 1988;48:749-53.
by guest on July 14, 2011www.ajcn.orgDownloaded from
