Change in Salt Intake Affects Blood Pressure
Implications for Human Populations
Paul Elliott, MB, PhD; Lesley L. Walker, BSc; Mark P. Little, MA, DPhil;
John R. Blair-West, MSc, PhD†; Robert E. Shade, PhD; D. Rick Lee, DVM; Pierre Rouquet, DVM;
Eric Leroy, DVM, PhD; Xavier Jeunemaitre, MD, PhD; Raymond Ardaillou, MD;
Francoise Paillard, MD; Pierre Meneton, PhD; Derek A. Denton, MBBS
Background—Addition of up to 15.0 g/d salt to the diet of chimpanzees caused large rises in blood pressure, which
reversed when the added salt was removed. Effects of more modest alterations to sodium intakes in chimpanzees, akin
to current efforts to lower sodium intakes in the human population, are unknown.
Methods and Results—Sodium intakes were altered among 17 chimpanzees in Franceville, Gabon, and 110 chimpanzees
in Bastrop, Tex. In Gabon, chimpanzees had a biscuit diet of constant nutrient composition except that the sodium
content was changed episodically over 3 years from 75 to 35 to 120 mmol/d. In Bastrop, animals were divided into 2
groups; 1 group continued on the standard diet of 250 mmol/d sodium for 2 years, and sodium intake was halved for
the other group. Lower sodium intake was associated with lower systolic, diastolic, and mean arterial blood pressures
in Gabon (2-tailed P?0.001, unadjusted and adjusted for age, sex, and baseline weight) and Bastrop (P?0.01,
unadjusted; P?0.08 to 0.10, adjusted), with no threshold down to 35 mmol/d sodium. For systolic pressure, estimates
were ?12.7 mm Hg (95% confidence interval, ?16.9 to ?8.5, adjusted) per 100 mmol/d lower sodium in Gabon and
?10.9 mm Hg (95% confidence interval, ?18.9 to ?2.9, unadjusted) and ?5.7 mm Hg (95% confidence interval,
?12.2 to 0.7, adjusted) for sodium intake lower by 122 mmol/d in Bastrop. Baseline systolic pressures higher by
10 mm Hg were associated with larger falls in systolic pressure by 4.3/2.9 mm Hg in Gabon/Bastrop per 100 mmol/d
Conclusions—These findings from an essentially single-variable experiment in the species closest to Homo sapiens with
high intakes of calcium and potassium support intensified public health efforts to lower sodium intake in the human
population. (Circulation. 2007;116:1563-1568.)
Key Words: blood pressure ? diet ? hypertension ? sodium
trials, and genetic, epidemiological, and anthropological find-
ings.1–15In preliterate societies in which sodium excretion is low
(1 to 10 mmol/d) and potassium excretion is high (80 to
200 mmol/d), blood pressure does not rise with age, and
incidence of cardiovascular disease is low.1,16–19When popula-
tions migrate to a more urbanized environment, blood pressure
rises over a period of months,20associated with an increase in
dietary sodium and other dietary and lifestyle changes.21,22
vidence for the role of dietary sodium in high blood
pressure comes from animal and clinical studies, clinical
Editorial p 1530
Clinical Perspective p 1568
National and international agencies have recommended
dietary sodium intakes of no more than 100 mmol/d sodium
(6 g/d salt).3,4,6Well-conducted short-term trials have found
greater blood pressure lowering for sodium reductions to ?50
to 60 mmol/d.11,12Studies of higher primates provide the
opportunity to alter dietary sodium experimentally for pro-
longed periods in the species genetically closest to humans.
Received November 23, 2006; accepted July 20, 2007.
From the Department of Epidemiology and Public Health, Imperial College London, London, UK (P.E., M.P.L.); Howard Florey Institute of
Experimental Physiology and Medicine (L.L.W.) and Department of Physiology (J.R.B.-W.), University of Melbourne, Melbourne, Victoria, Australia;
Department of Physiology and Medicine, Southwest Foundation for Biomedical Research, San Antonio, Tex (J.R.B.-W., R.E.S., D.A.D.); University of
Texas MD Anderson Cancer Center, Bastrop, Tex (D.R.L.); Centre International de Recherches Médicales de Franceville, Franceville, Gabon (P.R., E.L.);
Universite ´ Paris Descartes, Faculte ´ de Me ´decine, Paris (X.J.); INSERM U772, Collège de France, Paris (X.J.); Hôpital Tenon, Paris, France (R.A., F.P.);
INSERM U872, Département de Santé Publique et d’Informatique Médicale, Faculté de Médecine René Descartes, Paris, France (P.M.); and Baker
Medical Research Institute, Melbourne, Victoria, Australia (D.A.D.). Dr Lee currently is at the Alamogordo Primate Pacility, Holloman AFB, NM.
Correspondence to Paul Elliott, Department of Epidemiology and Public Health, Imperial College London, Faculty of Medicine, St. Mary’s Campus,
Norfolk Place, London W2 1PG, UK. E-mail firstname.lastname@example.org
© 2007 American Heart Association, Inc.
Circulation is available at http://circ.ahajournals.orgDOI: 10.1161/CIRCULATIONAHA.106.675579
We previously reported large increases in blood pressure
among 26 chimpanzees in Gabon fed increasing amounts of
salt.7We report here the effects on blood pressure of more
modest alterations of salt intake in 2 sets of chimpanzees, a
cohort of 17 animals in Gabon and 110 in Bastrop, Tex.
This study was performed at the Centre International de Recherches
Médicales, Franceville, Gabon (CIRMF). The 17 chimpanzees (11 of
which were included in the study by Denton et al7) lived in
long-standing, socially stable, small groups and were on a vegetable
and fruit diet with low sodium and high potassium intake supple-
mented by a biscuit diet (25 g/kg body weight; Mazuri Primate
Chunks, SDS, Essex, UK). This provided 75 mmol/d sodium at
baseline in early 1997. The study was divided into 3 periods
according to sodium content of the biscuits. After 2 years at baseline,
the biscuit supplement was changed to 35 mmol/d sodium in
December 1999 and to 120 mmol/d sodium from August 2000 until
October 2001. The 120-mmol/d period was divided into 2: Measure-
ments were taken on average 5 months apart in these final 2 periods.
Calcium content of these biscuits was reported as 2.48%. Daily
calcium intake from biscuits was therefore 15.5 mmol/kg (body
weight). The other constituents of the biscuits were constant.
Blood pressure measurements (per Denton et al7) were taken at the
end of each of the first 2 periods and in April to May 2001 and
September to October 2001 (2 of the animals were pregnant in the
last period). The study was approved by the CIRMF Animal Ethics
Committee and International Advisory Committee.
The study was performed at the University of Texas MD Anderson
Cancer Center in Bastrop, Tex. The 138 chimpanzees at the facility
were housed in socially stable groups of 3 to 15 per den; 28 were
excluded because of old age; because of specified medical condi-
tions, including leprosy, obesity, hepatitis, and hypertension (with
some of these animals under treatment); or because the chimpanzees
were needed for other projects. The remaining 110 animals were
allocated by den into 2 experimental groups. This allocation was
done from a list of dens stratified as either large dens (housing ?8
animals) or small dens (?7 animals); the dens in each stratum were
numbered and randomized to either the experimental or the control
group. The female chimpanzees of breeding age had levonorgestrel
implants. The chimpanzees had regular health checks while sedated,
formalized for the purpose of this study. Sedation was by intramus-
cular injection of Telazol (Fort Dodge Labs, Fort Dodge, Iowa) 3.5
mg/kg body weight. Times of injection, immobilization of the
animal, and recovery were recorded. Animals were weighed and
placed on an examination table. An adult large pneumatic cuff was
placed on the right arm, and blood pressures were measured with a
calibrated Propac automated sphygmomanometer ?12 to 24 minutes
after injection (measurements at 16,18, and 20 minutes were used).
If required, a further 100 mg Telazol was given intravenously, and
blood pressure readings were repeated 4, 6, and 8 minutes later. The
automated device gave readings of systolic and diastolic blood
pressure, as well as mean arterial pressure (MAP). Collection of
blood sample, physical examination, tuberculosis, and other tests
were then done.
After 2 years of baseline observations, 60 chimpanzees (controls)
remained on the standard dietary regime, Teklad Chimpanzee Diet
No. 7775, for 2 more years. The other 50 chimpanzees (intervention
group) were presented with a reduced-sodium diet (Teklad Low
Sodium Chimpanzee Diet No. 7188, Harlan Teklad, Madison, Wis)
for 2 years, from October 1999 to October 2001. The biscuits had
identical composition except for sodium content; calcium content is
reported as 0.98%. Chimpanzees were fed in their groups at morning
and afternoon sessions. “Produce” (various vegetables, eg, cucum-
bers, carrots, and cabbage quarters, and fruit, eg, apples, oranges, and
bananas) was presented individually. One large scoop, ?0.9 kg per
animal, of the Teklad biscuit diet (No. 7775 or 7188) was then
presented in feeders to the group, time regulated to ensure that each
chimpanzee had access. The “alpha male” chimpanzee was rewarded
with a piece of fruit if he allowed others to access the food. Usually,
no food was left at the end of the period.
Most chimpanzees received ?1.8 kg/d of the biscuit diet (2.0
kcal/g, ?3600 kcal/d). Sodium and potassium content of Teklad diet
No. 7775 was determined 5 times during 4 years and of No. 7188 diet
3 times during the last 2 years, with duplicate samples for each
analysis. The high-sodium diet contained 137.7?3.0 mmol/kg
(range, 127 to 152 mmol/kg) sodium and 200.8?2.5 mmol/kg
(range, 191 to 217 mmol/kg) potassium. The reduced-sodium diet
contained 69.8?2.1 mmol/kg (range, 64 to 76 mmol/kg) sodium and
194.3?4.4 mmol/kg (range, 186 to 208 mmol/kg) potassium. Thus,
daily sodium intakes averaged ?248 and 126 mmol/d, respectively;
potassium intakes averaged ?361 and 350 mmol/d; and calcium
intakes averaged ?440 mmol/d with both biscuit diets.
Nearly 6 months was required to survey the entire colony. Most
chimpanzees had at least 2 blood pressure measurements in each of
the first and second years of the study; blood pressures were not
measured during the first 6-month intervention period to exclude
early, possibly incomplete, changes and to attend to the 60 control
chimpanzees. A difference in timing existed between the 2 groups in
blood pressure measurements over the 2 phases of the study. On
average, the 2 blood pressure measurements were 2.14 years apart in
the control group and 2.52 years apart in the intervention group. The
study was approved by the University of Texas MD Anderson
Cancer Center Institutional Animal Care and Use Committee.
For descriptive statistics, means and percentages were compared
between groups. Given the close genetic similarity between humans
and chimpanzees,23we defined hypertension as systolic blood
pressure ?140 mm Hg or diastolic blood pressure ?90 mm Hg, as in
human studies. Multiple linear regression was used to assess rela-
tionships of systolic, diastolic, and MAP to sodium intake, with time
period, baseline weight, sex, and attained age included in the model;
for the Bastrop data, phase (first versus second period) and group
assignment, ie, controls (high-high sodium) versus intervention
(high-low sodium), were also included. For the Bastrop data, the age
adjusted to was 19 years, and the baseline weight adjusted to was 56
kg. In the Gabon data, the age adjusted to was 21 years, and the
baseline weight adjusted to was 49 kg. Supplementary analyses also
included baseline blood pressure and weight change. We used data
from all time periods to construct the regression models; ie, in
Gabon, all 4 periods (75 mmol/d, 35 mmol/d, both periods at
120 mmol/d) were included simultaneously. In addition, we tested
for interactions of the effect of sodium on blood pressure by age, sex,
baseline weight, and baseline blood pressure by including all these
interaction terms simultaneously. Implicitly, we are assuming the
same effect of sodium on blood pressure per animal; in particular, we
assume independence of normal (gaussian) measurement errors. We
fitted a random-effects model to the Gabon data, taking into account
the component of error in common for the measurements made for
each animal24; for the Bastrop data, such a model is not appropriate
because the model we fitted—effectively looking at the difference in
blood pressure differences (first versus second period) between the
control and intervention groups—results in the cancellation of the
within-animal errors, giving a singular design matrix. Singularities
also arose in fitting the descriptive models used in Figure 1, so these
analyses were conducted without the use of random effects. Param-
eter estimates are presented with 95% confidence intervals (CIs);
probability values are 2 tailed. Analysis was performed (by M.P.L.)
in S-Plus (Insightful Corp, Seattle, Wash).25
The authors had full access to and take responsibility for the
integrity of the data. All authors have read and agree to the
manuscript as written.
Descriptive statistics at baseline are shown in the Table.
Compared with Gabon, the Bastrop chimps had higher body
October 2, 2007
weight (P?0.02) and blood pressures (P?0.001); 47 (42.7%)
animals in Bastrop had hypertension compared with 2
(11.8%) in Gabon (P?0.02). Among the Bastrop chimps, the
controls were younger (P?0.001) with lower blood pressures
(P?0.05) than the intervention group.
With sodium reduced from 75 to 35 mmol/d, systolic blood
pressure (without adjustment) fell by 5.3 mm Hg (95% CI,
?0.5 to 11.2); with intake increased to 120 mmol/d sodium,
it rose by 10.3 mm Hg (95% CI, 4.3 to 16.3) after 9 months
and a further 0.7 mm Hg (95% CI, ?5.6 to 6.9) after 5 more
months. Results after adjustment for age, sex, and baseline
weight are shown in Figure 1. In random-effects multiple
regression analysis with adjustment, sodium intake lower by
100 mmol/d was associated with systolic blood pressure
lower by 12.7 mm Hg (95% CI, 8.5 to 16.9), diastolic blood
pressure lower by 7.5 mm Hg (95% CI, 5.1 to 10.0), and
MAP lower by 9.9 mm Hg (95% CI, 6.5 to 13.2) (P?0.001).
Significant interactions existed between sodium intake and
sex (P?0.001), baseline weight (P?0.038), and baseline
blood pressure (P?0.001) for systolic blood pressure. Per
were larger in females than males by 21.0 mm Hg (95% CI, 10.5
to 31.5), and by 5.9 mm Hg (95% CI, 0.3 to 11.5) per 10 kg
higher baseline body weight, and by 4.3 mm Hg (95% CI, 2.7 to
5.9) per 10 mm Hg higher baseline systolic pressure.
Over the 2 phases of the study, blood pressure fell for the
control group (standard diet), but the intervention group
(reduced sodium) experienced a larger fall. Without adjust-
ment, the differences in falls between the 2 groups (interven-
tion minus control) were ?10.9 mm Hg (95% CI, ?18.9 to
?2.9) systolic, ?9.4 mm Hg (95% CI, ?16.1 to ?2.8)
diastolic, and ?9.3 mm Hg (95% CI, ?15.9 to ?2.6) MAP
for a sodium reduction of 122 mmol/d (P?0.007, P?0.005,
and P?0.006, respectively). Adjusted for age, sex, and
baseline weight (Figure 2), the analogous numbers were
?5.7 mm Hg (95% CI, ?12.2 to 0.7) systolic, ?4.4 mm Hg
(95% CI, ?9.6 to 0.8) diastolic, and ?4.8 mm Hg (95% CI,
?10.2 to 0.7) MAP (P?0.08 to 0.10).
Significant interactions existed between sodium intake and
sex (P?0.01), baseline weight (P?0.002), and baseline blood
pressure (P?0.001) for all measures of blood pressure. Per
100 mmol/d lower sodium intake, estimated falls in systolic
pressure were larger in females than in males by 9.7 mm Hg
(95% CI, 5.1 to 14.4), and 2.6 mm Hg (95% CI, 1.0 to 4.2)
per 10 kg higher baseline body weight, and by 2.9 mm Hg
(95% CI, 2.0 to 3.9) per 10 mm Hg higher baseline systolic
These studies of chimpanzees, the animal species phyloge-
netically closest to humans, allow conclusions of relevance to
Mean arterial BP
Blood pressure (mm Hg)
Figure 1. Variation in blood pressure (BP; mean, 95% CI) by
dietary sodium intake adjusted for age, sex, and baseline
weight, Gabon data (P?0.001 for trend with sodium).
Summary of Gabon and Bastrop Chimpanzee Data at Baseline
Bastrop, mean (SD)†
Variable Gabon, mean (SD)*Control‡ Intervention§
Animals (males), n
Systolic blood pressure, mm Hg
Diastolic blood pressure, mm Hg
MAP, mm Hg
Hypertensive, n (%)?
*Mean (SD) at baseline (75 mmol/d sodium).
†Mean (SD) in phase 1 (248 mmol/d sodium).
‡248 mmol/d sodium in phases 1 and 2.
§248 mmol/d sodium in phase 1 and 126 mmol/d in phase 2.
?Systolic blood pressure ?140 mm Hg or diastolic blood pressure ?90 mm Hg.
Elliott et al Salt Intake and Blood Pressure of Chimpanzees
the blood pressure problem in human populations. First,
blood pressure falls were as large as or larger for sodium
intakes at or below current guidelines (range, 35 to
120 mmol/d) as they were in the range of “usual” sodium
intakes of humans (120 to 250 mmol/d),14,26(ie, ?12.5/
?7.5 mm Hg per 100 mmol versus ?5.7/?4.4 mm Hg for
122-mmol difference in sodium). Second, they occurred in a
vegetarian “high-potassium” (?350 mmol/d) and “high-
calcium” (?350 mmol/d) environment in contradistinction to
claims that sodium intake is not relevant to blood pressure in
mineral-replete states.27,28Unlike the human studies, these
were single-variable experiments uncomplicated by other
lifestyle exposures such as diet change, alcohol drinking, or
cigarette smoking and were prolonged over years rather than
days or weeks as in most human trials.8–10,12
The chimpanzees in Bastrop were substantially heavier
than the Gabon group, possibly reflecting higher daily energy
intake. The large difference in baseline blood pressure might
be explained by the different baseline levels of daily sodium
intake, 248 versus 75 mmol/d, and the weight difference;
human trial data indicate that the effects of intervention on
weight and sodium intakes are additive.29The larger associ-
ation of sodium intake with blood pressure among females
than males is also found in humans30; it might relate to the
hormonal milieu among females or to their smaller body and
hence kidney size, with reduced ability to deal with a sodium
load many times above physiological need.30It is unlikely
that the levonorgestrel implants in the Bastrop females played
a role because human data indicate that contraceptives con-
taining progestogens alone (ie, in the absence of estrogens) do
not affect blood pressure.31,32
One limitation, because of logistic constraints, was that the
chimpanzees in Gabon acted as their own controls, in contrast
to the earlier study of Denton et al.7By including an extended
2-year run-in period, first a fall then a rise in sodium intakes,
we aimed to minimize any possible “period” effect (ie, blood
pressure changes occurring over time that were unrelated to
sodium). The fact that the blood pressure changes closely
mirrored those of sodium (ie, a fall followed by a rise) argues
against a strong period effect, unrelated to sodium, in explain-
ing our findings.
A further limitation in Bastrop was that despite random-
ization of the chimps by den (ie, not individually randomized
for logistic reasons because the chimps live together in social
groups), they were not well matched at baseline for age,
weight, and blood pressure; therefore, we adjusted for age
and weight (and sex) in the analysis. The control animals had
an unexplained fall in blood pressure over the course of study,
a well-known phenomenon in human trials, although in that
setting blood pressure measurements are made in conscious
individuals. By including a control group, we were able to
adjust for this trend with time, although it reduced the power
of the study to detect falls in blood pressure in the interven-
tion compared with the control group. After adjustment, the
differences in blood pressure between the two groups were
not statistically significant.
Our findings extend the results of the study by Denton et
al,7which considered a group of 26 chimpanzees allocated
into 2 age- and sex-matched groups; baseline sodium intake
was 2 to 25 mmol/d (ie, the diet of a preliterate society in
which essential hypertension is rare). The treatment group
was given a diet with increasing additions of salt (5, 10, and
15 g/d; ie, 85, 170, and 257 mmol/d sodium), associated with
a progressive increase in blood pressure in most animals. The
highest intake significantly increased mean systolic pressure
by 33 mm Hg, diastolic pressure by 10 mm Hg, and MAP by
15 mm Hg. Twenty weeks after the cessation of added salt,
the blood pressures of the treatment group had fallen to
baseline and control group values.7
The changes in blood pressure in the Denton et al7study
are greater than those observed here. The highest sodium
intake in the earlier study was almost 20 times the low
baseline intake. In the present Gabon study, the increase in
sodium intake was only 3- to 4-fold from the lowest level of
35 to 120 mmol/d and is more relevant to present-day
discussions about optimal targets for dietary sodium reduc-
tion in humans.33In Bastrop, sodium intake was halved from
248 to 126 mmol/d (ie, a decrease from the mean sodium
intake of ?200 mmol/d found in a number of populations
worldwide14,26to below the mean values of ?150 mmol/d
currently seen in the United States and the United Kingdom).
The results of this study are qualitatively and quantitatively
consistent with results of human epidemiological studies and
clinical trial data. The International Cooperative Study on the
Relation of Sodium and Potassium to Blood Pressure
(INTERSALT) estimated that per 100 mmol/d lower sodium
intake, systolic pressure of individuals was lower by 3 to
6 mm Hg at an average of 40 years of age, and the rise in
Figure 2. Variation in blood pressure (BP; mean, 95% CI) by
dietary sodium intake adjusted for age, sex, and baseline
weight, Bastrop data. For presentational purposes only, the
control and intervention groups are separated on the x axis; this
does not represent differences in timing of measurements
between the 2 groups.
October 2, 2007
systolic pressure between 25 and 55 years of age would be
smaller by 9 to 11 mm Hg.15The Dietary Approaches to Stop
Hypertension (DASH-Sodium) study was a randomized feed-
ing trial of 412 volunteers with and without hypertension (about
two thirds had blood pressure levels in the prehypertension
range).12For a month, participants were fed diets containing 1 of
3 levels of sodium—141, 106, and 64 mmol/d—in combination
in vegetables, fruit, and low-fat dairy products (the so-called
DASH diet).12Reducing sodium intake from the highest to the
lowest level (ie, by 77 mmol/d sodium) lowered blood pressure
by 3 to 7/2 to 3 mm Hg systolic/diastolic (the range depending
on whether the DASH diet was used), with larger decreases
among those with hypertension.12In DASH, blood pressure falls
were greater when sodium was reduced from 106 to 64 mmol/d
than when it was reduced from 141 to 106 mmol/d, suggesting
possible nonlinearity in the blood pressure response to reduced
sodium. In Gabon, the fall in blood pressure from 75 to
35 mmol/d was proportionately larger than the rise from 35 to
120 mmol/d. However, we were unable to assess possible
nonlinearity of response in Bastrop because we tested only 2
levels of sodium.
Our results have important policy implications for human
populations. Despite the plethora of studies implicating high
salt intake in the origin of high blood pressure in humans,2
some commentators have continued to question its impor-
tance.27,28Their arguments have been strongly refuted34but
include the notion that it is dietary quality rather than dietary
sodium that is important for blood pressure control, particu-
larly the need for adequate mineral intake (calcium and
potassium).28As noted, we have shown here that in mineral-
replete states, lowered dietary sodium is associated with
substantial reductions in blood pressure, without threshold,
down to at least 35 mmol/d sodium. It also is suggested that
reducing dietary salt would be appropriate only for hyperten-
sive people.10Although the effects were greater for animals
with higher baseline pressures, at least in Gabon, most of the
animals had blood pressures in the normal or prehypertensive
range (?140 mm Hg systolic, ?90 mm Hg diastolic), unlike
the human studies, which have been carried out predomi-
nantly among hypertensive individuals.8–10This is important
in considering dietary guidelines for the millions of people
with blood pressures below the clinical criterion for hyper-
tension but who nonetheless incur increased risk of heart
disease and stroke.35
Current public health guidelines to reduce sodium intakes
in human populations to 100 mmol/d (6 g/d salt) are a
compromise between what might be readily achieved given
the high amounts of salt added to food in manufacture and the
benefits of more extensive reductions in sodium intake.3For
such policies to be effective, investment by the food industry
is needed to increase consumer choice by providing a far
greater range of low-sodium and sodium-free products with
informative and easily understood labeling of the sodium
content of foods.
We would like to acknowledge the skillful assistance of Tim Fleming
of the University of Texas MD Anderson Cancer Center during the
study of the Bastrop chimpanzees. We dedicate this article to the
memory of Dr John Blair-West, who passed away before this article
could be published.
Sources of Funding
This work was supported by the Robert J. Jr. and Helen C. Kleberg
Foundation, the Harold and Leila Mathers Charitable Trust, and the
Dr Lee reports that he is funded by Charles River Laboratories, Inc,
which receives government support for the work on chimpanzees.
The other authors report no conflicts.
1. Denton D. The Hunger for Salt: An Anthropological, Physiological and
Medical Analysis. Berlin, Germany: Springer Verlag; 1982.
2. Elliott P. The INTERSALT study: an addition to the evidence on salt and
blood pressure, and some implications. J Hum Hypertens. 1989;3:
3. WHO Expert Committee on the Prevention of Coronary Heart Disease.
Prevention of Coronary Heart Disease. Geneva, Switzerland: World
Health Organization; 1982. Technical Report Series No. 678.
4. National Research Council, Committee on Diet and Health, Food and
Nutrition Board, Commission on Life Sciences. Diet and Health: Impli-
cations for Reducing Chronic Disease. Washington, DC: National
Academy Press; 1989.
5. Scientific Advisory Committee on Nutrition. Salt and Health. London:
The Stationery Office; 2003.
6. Institute of Medicine. Dietary Reference Intakes: Water, Potassium,
Sodium, Chloride, and Sulfate. Washington, DC: National Academies
7. Denton D, Weisinger R, Mundy NI, Wickings EJ, Dixson A, Moisson P,
Pingard AM, Shade R, Carey D, Ardaillou R, Paillard F, Chapman J,
Thillet J, Michel JB. The effect of increased salt intake on blood pressure
in chimpanzees. Nat Med. 1995;1:1009–1016.
8. Midgley JP, Matthew AG, Greenwood CMT, Logan AG. Effect of
reduced dietary sodium on blood pressure: a meta-analysis of randomised
controlled trials. JAMA. 1996;275:1590–1597.
9. Cutler JA, Follmann D, Allender PS. Randomized trials of sodium
reduction: an overview. Am J Clin Nutr. 1997;65(suppl):643S–651S.
10. Graudal NA, Galloe AM, Garred P. Effects of sodium restriction on blood
pressure, renin, aldosterone, catecholamines, cholesterols, and triglycer-
ide: a meta-analysis. JAMA. 1998;279:1383–1391.
11. Meneton P, Jeunemaitre X, de Wardener HE, MacGregor GA. Links
between dietary salt intake, renal salt handling, blood pressure, and
cardiovascular diseases. Physiol Rev. 2005;85:679–715.
12. Sacks FM, Svetkey LP, Vollmer WM, Appel LJ, Bray GA, Harsha D,
Obarzanek E, Conlin PR, Miller ER III, Simons-Morton DG, Karanja N,
Lin P-H, for the DASH-Sodium Collaborative Research Group. Effects
on blood pressure of reduced dietary sodium and the Dietary Approaches
to Stop Hypertension (DASH) diet. N Engl J Med. 2001;344:3–10.
13. Elliott P. Observational studies of salt and blood pressure. Hypertension.
14. INTERSALT Cooperative Group. INTERSALT: an international study of
electrolyte excretion and blood pressure: results for 24 hour urinary
sodium and potassium excretion. BMJ. 1988;297:319–328.
15. Elliott P, Stamler J, Nichols R, Dyer AR, Stamler R, Kesteloot H, Marmot
M, for the INTERSALT Cooperative Research Group. INTERSALT
revisited: further analysis of 24 hour sodium excretion and blood pressure
within and across populations. BMJ. 1996;312:1249–1253.
16. Truswell AS, Kennelly BM, Hansen JDL, Lee RB. Blood pressure of
Kung bushmen in Northern Botswana. Am Heart J. 1972;84:5–12.
17. Sinnett PF, Whyte HM. Epidemiological studies in a total highland
population, Tukisenta, New Guinea, Cardiovascular disease and relevant
clinical, electrocardiographic, radiological and biochemical findings.
J Chron Dis. 1973;26:265–290.
18. Page LB, Damon A, Moellering RC Jr. Antecedents of cardiovascular
disease in six Solomon Islands societies. Circulation. 1974;49:
19. Oliver WJ, Cohen EL, Neel JV. Blood pressure, sodium intake and
sodium related hormones in the Yanomamo: a “no salt” culture. Circu-
Elliott et al Salt Intake and Blood Pressure of Chimpanzees
20. Poulter NR, Khaw KT, Hopwood BE, Mugambi M, Peart WS, Rose G, Download full-text
Sever PS. The Kenyan Luo migration study: observations on the initiation
of a rise in blood pressure. BMJ. 1990;300:967–972.
21. Klag MJ, He J, Coresh J, Whelton PK, Chen JY, Mo JP, Qian MC, Mo
PS, He GQ. The contribution of urinary cations to the blood pressure
differences associated with migration. Am J Epidemiol. 1995;142:
22. Shaper AG. Cardiovascular disease in the tropics. 3. Blood pressure and
hypertension. Br Med J. 1972;3:805–807.
23. Wildman DE, Uddin M, Liu G, Grossman LI, Goodman M. Implications
of natural selection in shaping 99.4% nonsynonymous DNA identity
between humans and chimpanzees: enlarging genus Homo. Proc Natl
Acad Sci U S A. 2003;100:7181–7188.
24. Laird NM, Ware JH. Random-effects models for longitudinal data. Bio-
25. S-Plus 6.0 Professional Release 2 for Windows. Seattle, Wash: Insightful;
26. Elliott P, Brown I. Sodium intakes around the world. Presented at: WHO
Forum and Technical Meeting on Reducing Salt Intake in Populations,
October 5–7, 2006; Paris, France.
27. McCarron DA. Diet and blood pressure: the paradigm shift. Science.
28. McCarron DA. The dietary guideline for sodium: should we shake it up?
Yes! Am J Clin Nutr. 2000;71:1013–1019.
29. Whelton PK, Appel LJ, Espeland MA, Applegate WB, Ettinger WH Jr,
Kostis JB, Kumanyika S, Lacy CR, Johnson KC, Folmar S, Cutler JA.
Sodium reduction and weight loss in the treatment of hypertension in
older persons: a randomized controlled trial of nonpharmacologic inter-
ventions in the elderly (TONE). TONE Collaborative Research Group.
30. Elliott P, Dyer AR, Stamler R. The INTERSALT Study: results for
24-hour sodium and potassium, by age and sex. J Hum Hypertens.
31. Wilson ES, Cruickshank J, McMaster M, Weir RJ. A prospective con-
trolled study of the effect on blood pressure of contraceptive preparations
containing different types and dosages of progestogen. Br J Obstet
32. Dong W, Colhoun HM, Poulter NR. Blood pressure in women using oral
contraceptives: results from the Health Survey for England 1994.
J Hypertens. 1997;15:1063–1068.
33. He FJ, MacGregor GA. How far should salt intake be reduced? Hyper-
34. Stamler J, Appel L, Cooper R, Denton D, Dyer AR, Elliott P, Greenland
P, Kesteloot H, Kumanyika S, Liu K, Marmot M, Van Horn L, Whelton
P. Dietary sodium chloride (salt), other dietary components and blood
pressure: paradigm expansion, not paradigm shift. Acta Cardiol. 2000;
35. Elliott P, Stamler J. Primary prevention of high blood pressure. In:
Marmot M, Elliott P, eds. Coronary Heart Disease Epidemiology: From
Aetiology to Public Health. 2nd ed. Oxford, UK: Oxford University Press;
Considerable evidence exists relating high dietary sodium (salt) intake to raised blood pressure coming from animal studies,
anthropology, clinical observations, clinical trials, and epidemiological studies. In the United States, with high prevalence
of high blood pressure at middle and older ages and high rates of coronary heart disease and stroke, mean sodium intake
is ?180 mmol/d (10.5 g/d salt) in men and 140 mmol/d (8 g/d salt) in women. In contrast, in preliterate societies in which
sodium intake is low (1 to 10 mmol/d) and potassium intake is high (80 to 200 mmol/d), blood pressure does not rise with
age, and incidence of cardiovascular disease is low. National and international agencies have recommended dietary sodium
intakes of no more than 100 mmol/d (6 g/d salt) or 65 mmol/d (4 g/d salt) in high-risk groups. Studies of higher primates
provide the opportunity to alter dietary sodium experimentally for prolonged periods in the species genetically closest to
humans. Chimpanzees in the wild have low sodium intakes and high intakes of other minerals. Large increases in blood
pressure were previously found among a group of chimpanzees fed increasing amounts of salt up to 250 mmol/d. In this
article, effects of more modest alterations of salt intake, akin to the levels recommended for human populations, on blood
pressure are reported. Two sets of chimpanzees were studied, a cohort of 17 animals in Gabon and 110 in Bastrop, Tex.
Systolic blood pressures were lower by an estimated 6 to 13 mm Hg for sodium intakes lower by 100 to 120 mmol/d. These
findings from the species closest to Homo sapiens support intensified public health efforts to lower sodium intake in the
October 2, 2007