Vol.29 (2006) No.6
Potassium Excretion in Healthy Japanese Women
Was Increased by a Dietary Intervention Utilizing
Home-Parcel Delivery of Okinawan Vegetables
Mallet K.-N. TUEKPE1), Hidemi TODORIKI1), Satoshi SASAKI2),
Kui-Cheng ZHENG1), and Makoto ARIIZUMI1)
Potassium, which is abundant in vegetables, is inversely related to blood pressure. Although the situation
has changed somewhat in recent years, the Okinawan diet has generally included a large amount of vege-
tables, and until recently Okinawans had the lowest rates of mortality due to stroke and coronary heart dis-
ease in Japan. Based on the hypothesis that these low mortality rates are partly attributable to increased
potassium intake resulting from the high vegetable consumption, this study examined whether increasing
the consumption of typical yellow-green Okinawan vegetables increases potassium intake. The purpose of
this investigation was to determine whether increased consumption of these vegetables should be one of
the dietary modifications recommended in public health promotion programs for Okinawans. The study
employed 56 healthy, normotensive, free-living Japanese women aged 18–38 years living in Okinawa. They
were randomized to a dietary intervention group (n=27) or a control group (n=29). Members of the dietary
intervention group received an average weight of 371.4 g/day of a combination of the following vegetables
twice weekly through an express home parcel deliver service for a period of 14 days: Goya (Momordica cha-
rantia), green papaya (Carica papaya), Handama (Gynura bicolor), Karashina (Brassica juncea), Njana (Cre-
pidiastrum lanceolatium), Fuchiba (Artemisia vulgaris) and Fudanso (Beta vulgaris); and they consumed an
average of 144.9 g/day, resulting in a 20.5% increase in their urinary potassium excretion over the baseline
(p=0.045). The members of the control group were asked to avoid these vegetables, and the change in
potassium excretion in this group was not significant (p=0.595). Urinary sodium and magnesium excretions,
systolic and diastolic blood pressures, folic acid, triglycerides and serum high density lipoprotein choles-
terol, low density lipoprotein cholesterol and total cholesterols changed non-significantly in both groups.
Also, post-intervention urinary potassium excretion correlated positively with vegetable consumption in
both the dietary intervention (p<0.0001) and control (p=0.008) groups and with Okinawan vegetable intake
in the dietary intervention group (p=0.0004). (Hypertens Res 2006; 29: 389–396)
Key Words: Okinawa, vegetable intake, urinary potassium excretion, randomized controlled dietary interven-
From the 1)Department of Environmental and Preventive Medicine, Faculty of Medicine, University of the Ryukyus, Okinawa, Japan; and 2)Project of
Scientific Evaluation of Dietary Reference Intakes, National Institute of Health and Nutrition, Tokyo, Japan.
This study was funded by a Grant-in-Aid from the Ministry of Agriculture, Forestry and Fisheries of Japan.
Address for Reprints: Hidemi Todoriki, Ph.D., Department of Environmental and Preventive Medicine, Faculty of Medicine, University of the Ryukyus,
207 Uehara, Nishihara-cho, Okinawa 903–0215, Japan. E-mail: firstname.lastname@example.org
Received October 6, 2005; Accepted in revised form February 21, 2006.
Hypertens Res Vol. 29, No. 6 (2006)
Although recent data shows that the prevalence of hyperten-
sion, stroke and myocardial infarction in Okinawa are cur-
rently not much different from those in the rest of Japan (1),
until recently Okinawans had the lowest rate of mortality due
to stroke and coronary heart disease in Japan (2–4). The rea-
sons for this are not entirely clear, but the dietary habits of
Okinawans have been suggested to be a major factor (5).
Although these dietary habits are changing (6), Okinawans
used to consume large amounts of vegetables (7, 8).
Vegetables are a rich source of dietary potassium (9).
Potassium inhibits platelet aggregation and arterial thrombo-
sis, increases vasomotricity and natriuresis (10–13), reduces
high blood pressure and helps prevent heart attacks (10, 14,
15). It also reduces both systolic and diastolic blood pressures
in normotensives and hypertensives (14, 15).
Although pharmacological methods are currently the main
treatment option for hypertension (16), non-pharmacological
treatments have also been proposed (17). Increasing the
intake of vegetables could be a less expensive non-pharmaco-
logical treatment not only for lowering high blood pressure in
hypertensives but also for promoting cardiovascular health.
This study examined whether increasing the consumption
of the yellow-green Okinawan vegetables typically used in
Okinawan dishes increases potassium intake. The purpose
was to determine whether dietary modifications involving
increased consumption of typical yellow-green Okinawan
vegetables could be proposed for public health promotion
programs for Okinawans.
In addition, this study employed the relatively new concept
of dietary intervention via home-parcel delivery, in which the
vegetables used in the intervention were delivered directly to
the homes of the subjects.
The subjects were healthy free-living female volunteers liv-
ing in Okinawa. They were recruited through posters and by
personal contacts. All subjects were between 18 and 38 years
of age, and none of them was undergoing treatment for any
disease at the time of the study. Fifty-six volunteers consented
to participate, and 53 successfully completed the study.
Twenty-five of these 53 subjects were randomized to the
dietary intervention group, and 29 to the control group.
Study Design, Vegetable Intake and Dietary
The study was carried out in March–April of 2005. It was
designed as a randomized controlled trial employing dietary
intervention. The aims, objectives, and intervention method
were carefully explained to all the subjects, and all of them
were informed that they were at liberty to withdraw from the
study at any time without providing an explanation. After pro-
viding their informed written consent, subjects were random-
ized to the dietary intervention or control group. A list of the
yellow-green Okinawan vegetables being used in the study
was given to all the subjects. Additionally, the dietary inter-
vention group was provided with instructions on how to cook
the Okinawan vegetables. The dietary intervention group was
strongly urged to consume a combination of these vegetables
as part of their diet, while the controls were asked to refrain
from consuming them as much as possible.
The dietary intervention lasted 14 days. Intervention days
were specifically tailored for each subject so as not to coin-
cide with their menstruation. During the intervention period
about 1.3 kg of a combination of five typical yellow-green
Okinawan vegetables were delivered twice a week to the
homes of the intervention group by an express home parcel
delivery service. The combination consisted of four primary
vegetables: about 400 g Goya (Momordica charantia), about
600 g green papaya (Carica papaya), 100 g Handama
(Gynura bicolor) and 100 g Karashina (Brassica juncea); and
100 g of one of three others: Njana (Crepidiastrum lanceola-
tium), Fuchiba (Artemisia vulgaris) or Fudanso (Beta vul-
garis). The 2.6 kg/week supply of vegetables was based on
the national recommended intake of 350 g/day and the results
of a 1 year preliminary nutritional survey carried out by us in
Ginowan City, Okinawa.
All subjects in both groups were provided with digital
scales which they used to weigh and record all vegetables,
fruits and juices they consumed during the intervention
period. Any of these foods consumed outside the home were
also recorded and their weights estimated later using a picture
booklet designed for this purpose (18). The potassium content
of vegetables was calculated from Standard Tables of Food
Composition in Japan (19). The Medical Ethics Committee of
the University of the Ryukyus approved the study protocol.
Blood Pressure and Anthropological Data
Pre- and post-intervention blood pressures were measured in
both groups using an automatic digital sphygmomanometer
(HEM-762; Omron Corp., Tokyo, Japan) after subjects had
rested for at least 5 min. These blood pressure values were
measured three times in the right arm with subjects in a seated
position. If any of the measurements differed by more than 5
mmHg from the others, it was repeated. The mean value was
then calculated for each subject. Anthropological data was
also collected at the start and end of the dietary intervention
period from all subjects in both intervention and control
Fasting blood samples (10 ml) were collected the day before
Tuekpe et al: Okinawan Vegetables and Urinary Potassium Excretion
intervention started and the day after intervention ended in all
subjects of both groups. The samples were centrifuged imme-
diately after collection and the serum was transferred into
stock tubes and stored at −80°C until measurement.
Plastic, opaque brown urine-collection containers containing
0.1 g pre-weighed boric acid were given to all subjects of both
groups after carefully instructing them on how to collect 24-h
urine samples. The importance of collecting complete sam-
ples was emphasized. Urine samples were collected the day
before intervention started and the day after intervention
ended. The plastic containers with the urine samples were
kept in black bags at room temperature during the collection
period. Subjects brought the urine samples to a pre-arranged
submission point. The weight, specific gravity and volume of
each sample were immediately determined after submission,
and approximately 50 ml of each sample was collected into
stock tubes and stored at −20°C until measurement.
Analysis of Samples
Analyses of both urine and blood samples were carried out at
the SRL Laboratory in Tokyo, Japan. The laboratory person-
nel were blinded to the subjects and the coding system used to
label the samples.
The outcome measures were urine electrolytes and serum
folic acid, triglycerides and high density lipoprotein (HDL)
Table 1. Pre-Intervention and Post-Intervention Changes in Variables
Body height (cm)
Body weight (kg)
Body mass index
Systolic blood pressure
Diastole blood pressure
Serum HDL cholesterol
Serum LDL cholesterol
Serum total cholesterol
Serum folic acid (ng/ml)
24-h urinary sodium
24-h urinary sodium-
−0.3 (0.068) 0.687
20.1±1.4 21.1±3.1 0.667
109.3±8.9112.3±10.00.365 0.3 (0.832)
−0.5 (0.689)0.5 (0.634)0.536
56.4±11.161.0±28.30.771 1.2 (0.879)
−208.9 (0.495)67.3 (0.772)0.465
1,600.9±474.0 1,630.4±451.60.843 363.5 (0.045)
78.1±21.1 77.0±21.5 0.8721.4 (0.800)
−0.4 (0.039) 0.1 (0.394)0.031
*Difference between pre-intervention and post-intervention values. Student’s paired sample t-test was used to compare normally distrib-
uted variables and Wilcoxon matched pairs test was used for those that could not be made normal by logarithm transformation. †Stu-
dent’s independent sample t-test was used to compare normally distributed variables and Mann-Whitney U test was used for those that
could not be made normal by logarithm transformation. HDL, high density lipoprotein; LDL, low density lipoprotein.
Hypertens Res Vol. 29, No. 6 (2006)
cholesterol, low density lipoprotein (LDL) cholesterol and
total cholesterols. Since a positive correlation exists between
potassium intake and its excretion in urine (20, 21), urinary
potassium excretion was set up as the primary endpoint.
For the purposes of this report, subjects with incomplete data
(only one urine or blood sample; n=6), and subjects with a
creatinine (mg)/body weight (kg) ratio less than 10.8 or
greater than 25.2 (22) (n=6) were excluded from the statisti-
cal analysis. Since current smokers excrete lower urinary
potassium than non-smokers (23) and since serum potassium
level is strongly associated with cigarette smoking (24),
smokers (n=2) were also excluded. The results for the 39
subjects (intervention group, n=19; control group, n=20)
who met the above criteria are reported here.
SAS statistical software (version 8.0; SAS Institute Inc.,
Cary, USA) was used for the analyses. For normally distrib-
uted variables, Student’s paired sample t-test was used to ana-
lyze changes within each group and Student’s independent
sample t-test was used to assess differences between the
groups. Wilcoxon matched pair and Mann-Whitney U tests,
respectively, were used for variables that could not be nor-
malized even after logarithmic transformation. Pearson corre-
lation procedures were used to explore relationships between
vegetable intake (total and Okinawan) and post-intervention
urinary sodium and potassium excretions. Also, relationships
between post-intervention changes in systolic and diastolic
pressures and changes in urinary sodium and potassium
excretions were explored. Since fruits are also a major source
of potassium, adjustments were made for fruit intake.
Unless otherwise stated, all p-values are two-tailed and
only differences with p<0.05 were considered significant.
Table 1 shows pre- and post-intervention characteristics of
the subjects. There were no statistically significant differ-
ences between the intervention and control groups pre-inter-
vention. The mean age for the intervention group was 24.4
years, vs. 25.7 years for the control group. The difference in
mean body mass index (BMI) was not significant (p=0.667).
Sodium and potassium excretion in the 24-h urine samples
were also not significantly different between the two groups
(p=0.852 and p=0.843, respectively). There was a non-sig-
nificant mean weight loss of 0.3 kg in each group at the end of
intervention. The post-intervention changes and differences
Table 2. Average Vegetable, Fruit and Potassium Intake (g/Day)
Intervention group (n=19)
Control group (n=20)
Average daily intake*
Average daily Okinawan vegetable intake
Average daily intake of other vegetables
Average daily intake
Average daily intake from all vegetables
Average daily intake from fruits
Average daily intake from Okinawan vegetables
Average daily intake from other vegetables
Potassium intake per 100 g vegetables
33.5±28.544.6±44.1 11.1 0.550
*Includes Okinawan vegetables. †Student’s independent sample t-test was used to compare differences between the two groups.
Table 3. Correlation between Urinary Sodium, Potassium and Change in Blood Pressure Post-Intervention
Change in systolic pressure
Change in diastolic pressure
*Pearson’s correlation coefficient.
Tuekpe et al: Okinawan Vegetables and Urinary Potassium Excretion
Fig. 1. Scatter plot of mean daily Okinawan, non-Okinawan and all vegetable consumption and urine potassium excretion in the intervention and control groups. r is
Pearson’s correlation coefficient.
Hypertens Res Vol. 29, No. 6 (2006)
in the mean urinary sodium and magnesium excretions, serum
HDL cholesterol, LDL cholesterol and total cholesterols,
folic acid and triglycerides of both groups were also not sig-
nificant. However, the post-intervention potassium excretion
was significantly different from the pre-intervention excre-
tion in the intervention group (p=0.045) but not in the control
group (p=0.595). While it increased by 363.5 mg in the inter-
vention group, it decreased in the control group by 68.6 mg.
The sodium-potassium ratio also changed significantly in the
intervention group and between the two groups post-interven-
tion (p=0.039 and p=0.031, respectively). When all 53 sub-
jects who successfully completed the study were included in
the statistical analyses no significant post-intervention change
in urinary potassium excretion of the control group
(p=0.656) was observed (data not shown). However, the
intervention group showed an increase, although this increase
did not reach the level of statistical significance (p=0.063).
An average of 371.4 g/day Okinawan vegetables was sup-
plied to the dietary intervention group during the intervention
period. However, they consumed only an average of 144.9 g/
day (Table 2). The control group, who were not supplied with
Okinawan vegetables during the intervention period, con-
sumed an average of 4.7 g/day of Okinawan vegetables. The
differences between the two groups in both average daily all
vegetable and Okinawan vegetable intakes were statistically
significant (p<0.001 for each). During the intervention
period, the control group consumed a higher amount of fruits
than the dietary intervention group (44.6 g/day and 33.5 g/
day, respectively), but the difference was not statistically sig-
nificant (p=0.550). Based on the Standard Tables of Food
Composition in Japan (19), the potassium intakes from fruits
and vegetables in the present study are shown in Table 2. For
every 100 g, Okinawan vegetables provided 0.382 g and
0.386 g of potassium, while other vegetables provided 0.273
g and 0.276 g in the dietary intervention and control groups,
respectively. Okinawan vegetables therefore provided about
40% more potassium than other vegetables. There was no sig-
nificant difference between the two groups in the potassium
intake from fruits (p=0.415).
Table 3 shows that there were no significant correlations
between changes in blood pressure and changes in urinary
sodium and potassium excretions in either group. Adjust-
ments for age and BMI did not materially change the
Figure 1 is a scatter plot of vegetable intake vs. potassium
excretion. The correlation between Okinawan vegetable con-
sumption and urinary potassium excretion in the dietary inter-
vention group was stronger than the correlation between the
intake of other vegetables and urinary potassium excretion in
the control group (r=0.73, p=0.0004 and r=0.56, p=0.010,
respectively), although the amount of Okinawan vegetable
intake in the intervention group (144.9 g) was similar to the
intake of other vegetables in the control group (149.0 g)
(Table 2). Also, Fig. 1 shows significant correlations between
average daily all vegetable intake and urinary potassium
excretion in both the intervention and control groups
(p<0.0001 and p=0.008, respectively).
This study found that increasing the intake of typical yellow-
green Okinawan vegetables significantly increased urinary
potassium excretion; a reflection of both the higher intake and
the higher bioavailability of potassium. The significant differ-
ence in intake of Okinawan vegetables (p<0.001) between
the intervention and control groups indicates that the dietary
intervention was successful. Although many studies have
been carried out on Okinawans and their diet, to our knowl-
edge there has been no randomized controlled study investi-
gating the effects of increased consumption of typical yellow-
green Okinawan vegetables in free-living healthy subjects. In
addition, this was the first study in which a dietary interven-
tion was accomplished by home-parcel delivery of vegeta-
The 251.3 g daily vegetable consumption by the interven-
tion group (of which 144.9 g were the yellow-green Oki-
nawan vegetables used in this study), is comparable to the 233
g daily vegetable consumption (82 g of which are yellow-
green) by Japanese in the 20–29 years age group (6), the same
age group to which most of our subjects belong. The yellow-
green Okinawan vegetables formed 57.7% and 3.1% of the
total vegetables consumed by the intervention and control
groups, respectively, with the intervention group consuming a
mean of 97.6 g/day (38.8%) more vegetables than the control
group. This resulted in a 20.5% increase in urinary potassium
excretion (p=0.045) over the baseline in the intervention
group. It has been reported that a daily intake of 0.6–0.8 kg of
fruits and vegetables is required to ensure a daily potassium
supply of 2.5–3.5 g (25).
All subjects in this study were normotensive and their
blood pressure was not significantly reduced post-interven-
tion. This finding is in agreement with another study which
found that neither potassium chloride nor potassium citrate
(the form of potassium found in vegetables) significantly
changed blood pressure in normotensives (26). Although
there was a significant post-intervention change in urinary
potassium excretion in the intervention group, it was not
strong enough to cause a significant reduction in their blood
pressure. In any case, increased potassium intake reduces
blood pressure to a lesser extent in normotensives as com-
pared to hypertensives (14).
There were several potential limitations in this study. First,
we relied on the subjects to correctly weigh and record their
entire vegetable intake and to submit their complete 24-h
urine samples. In addition, if the total daily amount of vegeta-
bles consumed in the intervention and control groups had
been made equal, this would have made it easier to evaluate
the specific effect of Okinawan vegetables on potassium
excretion. These limitations are related to the many difficul-
ties associated with dietary interventions in free-living sub-
Tuekpe et al: Okinawan Vegetables and Urinary Potassium Excretion
jects. At the same time, this is also one of the strengths of this
study, since the dietary habits of free-living subjects closely
resemble those of the general population.
One of the values of this study is that it is the first to employ
home-parcel delivery of vegetables as a dietary intervention
method. This intervention method is very convenient for
intervention studies in free-living subjects, and is expected to
prove useful in future dietary studies. Also, this study has
shown for the first time that the potassium content of the yel-
low-green Okinawan vegetables used here is higher than that
of other vegetables.
A larger study involving both normo- and hypertensives
with a longer intervention period and a higher vegetable con-
sumption is currently underway to further investigate the
effects of increasing consumption of typical yellow-green
Okinawan vegetables on blood pressure.
In conclusion, the present study has shown that increasing
the consumption of yellow-green Okinawan vegetables typi-
cally used in Okinawan dishes significantly increased urinary
potassium, a reflection of increased intake. Public health diet
modification programs aimed at promoting higher intakes of
dietary potassium in Okinawans (for example, for the purpose
of reducing high blood pressure) might consider recommend-
ing increases in the intake of typical yellow-green Okinawan
vegetables. In view of the fact that hypertension in particular
and cardiovascular diseases in general are major contributors
to the global disease burden in both developed and develop-
ing countries (27), the inclusion of non-pharmacological
methods (such as increased vegetable consumption) in the
clinical therapy as well as preventive programs for these dis-
eases would greatly reduce the cost of health care globally.
Also, the positive correlation found between urinary potas-
sium excretion and average daily vegetable intake indicates
that urinary potassium excretion could probably be a useful
biomarker for vegetable intake.
This study was supported by the Okinawa Prefectural Govern-
ment Department of Agriculture, Forestry and Fisheries, who
supplied all the Okinawan vegetables used in this study. We
would like to acknowledge all the volunteers who so willingly
devoted their time and energy to this study. We would also like to
thank Dr. Yusuke Ohya, Dr. Akio Ishida and Dr. Rieko Mano of
the 3rd Medical Department, Faculty of Medicine, University of
the Ryukyus, Okinawa, Dr. Kouame Koadio of the Department
of Environmental and Preventive Medicine, Faculty of Medicine,
University of the Ryukyus, Okinawa and Dr. Qimei He of the
Pacific Health Research Institute, Honolulu, for their enormous
1.Iseki K, Kimura Y, Wakugami K, et al: Comparison of the
effects of blood pressure on the development of stroke,
acute myocardial infarction, and end-stage renal disease.
Hypertens Res 2000; 23: 143–149.
Fukiyama K, Kimura Y, Wakugami K, Muratani H: Inci-
dence and long-term prognosis of initial stroke and acute
myocardial infarction in Okinawa, Japan. Hypertens Res
2000; 23: 127–135.
Kimura Y, Takishita S, Muratani H, et al: Demographic
study of first-ever stroke and acute myocardial infarction in
Okinawa, Japan. Intern Med 1998; 37: 736–745.
Cockerham WC, Hattori H, Yamori Y: The social gradient
in life expectancy: the contrary case of Okinawa in Japan.
Soc Sci Med 2000; 51: 115–122.
Willcox BJ, Willcox DC, Suzuki M: The Okinawa Pro-
gram: How the World’s Longest-Lived People Achieved
Everlasting Health⎯and How You Can Too. New York,
Clarkson Potter, 2001.
Todoriki H, Willcox DC, Willcox BJ: The effects of post-
war dietary change on longevity and health in Okinawa.
Okinawa J Am Stud 2004; 1: 52–61.
Japan Ministry of Health, Labor and Welfare: Results from
Ministry of Health, Labor and Welfare Nutrition Survey,
2002. Tokyo, Daiichi Publishers, 2004 (in Japanese).
Division of Environment and Health, Okinawa Prefectural
Government: Results of Okinawa Prefectural Government
1998 Nutrition Survey. Okinawa, Okinawa Prefectural
Cadenas E, Paker L (eds): Handbook of Antioxidants, Sec-
ond Edition Revised and Expanded. New York, Marcel
Dekker Inc, 2002.
10. Young DB, Lin H, McCabe RD: Potassium’s cardiovascu-
lar protective mechanisms. Am J Physiol 1995; 268: R825–
11. Gallen IW, Rosa RM, Esparaz DY, et al: On the mechanism
of the effects of potassium restriction on blood pressure and
renal sodium retention. Am J Kidney Dis 1998; 31: 19–27.
12. Liu S, Manson JE, Lee IM, et al: Fruit and vegetable intake
and risk of cardiovascular disease: the Women’s Health
Study. Am J Clin Nutr 2000; 72: 922–928.
13. He FJ, MacGregor GA: Fortnightly review: beneficial
effects of potassium. BMJ 2001; 323: 497–501.
14. Whelton PK, He J, Cutler JA, et al: Effects of oral potas-
sium on blood pressure. Meta-analysis of randomized con-
trolled clinical trials. JAMA 1997; 277: 1624–1632.
15. Sacks FM, Willett WC, Smith A, Brown LE, Rosner B,
Moore TJ: Effect on blood pressure of potassium, calcium,
and magnesium in women with low habitual intake. Hyper-
tension 1998; 31: 131–138.
16. Diamond JA, Phillips RA: Hypertensive heart disease.
Hypertens Res 2005; 28: 191–202.
17. Miura S, Yamaguchi Y, Urata H, et al: Efficacy of a multi-
component program (patient-centered assessment and coun-
seling for exercise plus nutrition [PACE+ Japan]) for
lifestyle modification in patients with essential hyperten-
sion. Hypertens Res 2004; 27: 859–864.
18. Pietinen P, Hartman AM, Haapa E, et al: Reproducibility
and validity of dietary assessment instruments. I. A self-
administered food use questionnaire with a portion size pic-
ture booklet. Am J Epidemiol 1988; 128: 655–666.
19. Science and Technology Agency: Standard Tables of Food
Composition in Japan, 5th ed. Tokyo, Printing Bureau of
the Ministry of Finance, 2000 (in Japanese).
396 Download full-text
Hypertens Res Vol. 29, No. 6 (2006)
20. Kesteloot H, Joossens JV: The relationship between dietary
intake and urinary excretion of sodium, potassium, calcium
and magnesium: Belgian Interuniversity Research on Nutri-
tion and Health. J Hum Hypertens 1990; 4: 527–533.
21. Tamaki J, Yoshita K, Kikuchi Y, et al: Applicability of the
stages of change model for analyzing fruit and vegetable
intake in relation to urinary potassium excretion: baseline
results from the High-Risk and Population Strategy for
Occupational Health Promotion (HIPOP-OHP) Study.
Hypertens Res 2004; 27: 843–850.
22. WHO Regional Office for Europe: Estimation of sodium
intake and output: review of methods and recommendations
for epidemiological studies, in WHO Regional Office for
Europe (ed): Report on a WHO Meeting by the WHO Col-
laborating Center for Research and Training in Cardiovas-
cular Diseases. Geneva, WHO, 1984.
23. Dyer AR, Elliott P, Stamler J, et al: Dietary intake in male
and female smokers, ex-smokers, and never smokers: the
INTERMAP study. J Hum Hypertens 2003; 17: 641–654.
24. Wannamethee SG, Lever AF, Shaper AG, et al: Serum
potassium, cigarette smoking, and mortality in middle-aged
men. Am J Epidemiol 1997; 145: 598–606.
25. Demigne C, Sabboh H, Remesy C, Meneton P: Protective
effects of high dietary potassium: nutritional and metabolic
aspects. J Nutr 2004; 134: 2903–2906.
26. Mullen JT, O’Connor DT: Potassium effect on blood pres-
sure: is the conjugate anion important? J Hum Hypertens
1990; 4: 589–596.
27. Reddy KS, Yusuf S: Emerging epidemic of cardiovascular
disease in developing countries. Circulation 1998; 97: 569–