| Introduction |
The case for reducing salt intake for the general community is as
compelling as that for stopping smoking. Before examining the
evidence to support this, I would like to ask the question: "Should we
have started an antismoking campaign?" There was no randomised,
double-blind controlled trial to show that smoking caused cancer in
humans. Nor was there a randomised, double-blind controlled study to
show that it caused heart disease. There was also no controlled study
to show that reducing smoking lowered the incidence of these
diseases. If you believe in the criterion of absolute proof in
this climate of evidence-based medicine, the answer to my question
would be a resounding "No!". The decision to implement an antismoking
campaign was based on epidemiological, observational evidence,
coupled with animal experiments and knowledge of disease mechanism.
Nevertheless, the results of the antismoking campaign have been
spectacular, reducing disease and death, and clearly the decision
was correct.
In the same way, it may not be possible to prove the benefits of reducing
salt intake with randomised controlled trials. Moreover, just as not
all smokers develop lung disease, cancer, heart disease or
associated problems, not all people with a high salt intake develop
high blood pressure, cardiovascular disease, osteoporosis and
other problems.
To support my advocacy for a general reduction in salt intake in the
community, I will first examine the evidence that a high salt intake is
harmful, and secondly look at whether reducing salt intake itself
causes harm.
|
|
|
Association of high sodium, low potassium intake with
cardiovascular death |
|
Cardiovascular disease is the major cause of death in Australia and
other Western countries. Hypertension of various degrees affects up
to 45% of older people in Australia and contributes to cardiovascular
death. As we all hope to reach old age, about 45% of us are at risk. We now
know that the so-called "physiological rise" in systolic and pulse
pressures with age is a reflection of vascular stiffness, and these
elevated pressures cause increased cardiac work and impaired
endothelial function, leading to increased cardiovascular
deaths.24
In indigenous populations with a low sodium, high potassium intake,
this "physiological rise" in blood pressure is not seen and
hypertension does not occur.25 Cardiovascular deaths
are virtually absent, despite many people living to old
age.25 The few indigenous
communities with a high salt intake (usually island or coastal
communities) do not experience these cardiovascular
advantages,26-28 and when indigenous
communities adopt a Western diet (a high sodium and low potassium
intake), reduce their physical activity and develop obesity, high
blood pressure occurs and cardiovascular deaths result.29 The strongest
correlation is with the alteration in sodium-potassium
intake.28
This concept has been verified experimentally in chimpanzees fed a
diet with a high sodium, low potassium content, with no change in other
aspects of their lifestyle. Hypertension developed in some but not
all animals.11 Similarly, the genetic
predisposition to hypertension in some humans is not expressed
unless the appropriate environment, including a high sodium, low
potassium diet, is present. Besides an effect on blood pressure, a
high sodium intake has been shown to cause cardiac
hypertrophy30 (associated with
increased cardiac death), vascular stiffness31 (a major cause
of systolic hypertension, and an independent predictor of sudden
death), and increased cardiovascular sympathetic
activity32 (which can lead to
elevated blood pressure, cardiac hypertrophy and sudden death).
These findings have frequently been based on experimental studies in
animals, and later confirmed by studies in human populations.
|
|
|
Effect of change in salt intake on blood pressure in Western
communities |
|
Most individual studies33,34 in hypertensive
patients have shown a fall in blood pressure with sodium restriction.
Meta-analysis has confused the issue, because studies have
frequently been included in which compliance with the therapy has not
been monitored, blood pressure has not been adequately assessed, or
the studies have been done in normotensive individuals.
An early study of salt restriction by MacGregor et al35 in 19 mildly
hypertensive patients on a low salt diet found greater falls in blood
pressure in those given placebo rather than sodium supplementation.
However, the largest studies with adequate urine assessment of
compliance with the salt restriction are two Australian
studies.36-38 One was a
double-blind study37 (with a similar protocol
to the MacGregor study) which showed a 5.5/2.9 mmHg greater
mean fall in blood pressure in the placebo group than in the group
supplemented with 80 mg slow release sodium tablets per day. This is a
similar blood pressure reduction to that seen in the Australian
National Heart Foundation study, in which the fall in diastolic blood
pressure among patients with mild hypertension given
antihypertensive medication was 4.7 mmHg more than in those given
placebo.39 In the Australian sodium
restriction studies, there were multiple blood pressure
measurements in patients taking placebo or salt supplementation,
and no patient had a rise in blood pressure level with sodium
restriction.37 The other Australian
study14
showed that a fall in blood pressure could be
achieved and maintained over a three-month period, with a change in
sodium intake from 159 mmol/day to 90 mmol/day obtained by giving
dietary advice to people in the community.
Studies which have shown no effect of sodium restriction on blood
pressure have frequently been done in normotensive
people,40,41 or sodium intake has
not been verified, or too few people have been included to provide
adequate power to address the question.41 Particularly in people
with normal blood pressure or those with borderline hypertension, it
is difficult to demonstrate a fall in blood pressure unless a large
number of subjects are included.
Increasing salt intake up to or above the normal intake in
normotensive people does cause an increase in blood
pressure.42,43 A salt increase from 70
to 200 mmol/day probably raises the blood pressure level by a
measurable extent (5 mmHg) in about 15% of young (< 50 years)
normotensive people, but this occurs in about 50% of the older
population.43 In many young people a
change of 1-3 mmHg may occur, but this small change cannot be
identified in individual subjects as the blood pressure
measurements have an SD of about 7 mmHg.
A change of 1-3 mmHg, however, could be of critical importance for two
reasons. It would shift the stroke and heart disease distribution
curve and would potentially provide more community benefit than
treating patients with severe hypertension. The second reason
relates to the "tracking" of blood pressure that takes place. Thus,
people in the top tenth percentile for blood pressure in our community
stay at that level as they age and eventually meet our definition of
hypertension. This "tracking" probably results from a positive
feedback mechanism acting on the heart and the blood vessels.
Reduction of blood pressure by even 1 mmHg might prevent this
occurrence, and thereby reduce the rise of blood pressure with age and
the development of systolic hypertension. An analogy can be made with
malignant hypertension, which is the end-result of a positive
feedback process. Even relatively ineffective methods of treating
blood pressure may prevent its occurrence.
|
|
|
Is a low salt intake harmful? |
|
Based on the evidence of Laragh's group (Brunner et al),44
Kincaid-Smith,5 Alderman et al14,22 and others
have stated that a low salt intake is harmful. This is an incorrect and
misleading interpretation of the evidence. The data that Laragh
present are that a renin level (by inference, angiotensin II) which is
inappropriately high in relation to the sodium level is harmful.
Patients with this high renin-sodium ratio are usually identified by
a short term study measuring renin at two levels of sodium. The renin
(angiotensin II) levels and the salt intake during the rest of the
patients' lives, or during the rest of the "prospective" study, bear
no relationship to the values recorded in the study. There is no
evidence that a high angiotensin II level, appropriate for the sodium
intake, is harmful. Most of the analyses by Alderman et al have used
this classification of people into high, normal or low renin groups,
and thus provide no objective or useful information as to whether a low
sodium intake is harmful. Alderman and Laragh's article,45 which
purports to be a prospective study but is in fact an observational
outcome study, does divide patients into four groups according to
their salt intake at the start of the study and claims to show a
relationship between initial salt intake and myocardial
infarction. However, this relationship was seen in men and not in
women (the trend in mortality in women was the reverse of that in men,
but was not significant owing to the smaller number of cardiovascular
events). Moreover, the salt intake status was classified from one
urine sample, whereas up to seven urine samples are needed for this
classification. No validation of the completeness of the urine
collection was provided. Thus, the low salt group may have included an
excess of people who were non-compliant with the instructions, and
perhaps with other aspects of their health. This could account for the
apparent excess of cardiovascular events in this group and may be
unrelated to their usual sodium chloride intake.
If, because of the rise in renin levels, a low salt intake is harmful for
hypertensive patients, as claimed by Alderman et al,14,22 it is
difficult to understand why diuretic-based therapy, which causes
greater disturbance in intravascular and extravascular volume and a
greater rise in renin levels, has reduced mortality from
cardiovascular disease.39
|
|
|
Targeted or community intervention? |
|
An argument can be made that salt restriction should be targeted to
specific groups (eg, people with a family history of hypertension,
strokes or heart disease; young people in the top 10% for blood
pressure level; and people with diabetes or renal disease). This is
one possible intervention strategy, but it ignores the fact that
potentially 45% of our community develop systolic hypertension and
this targeted strategy would miss most of these.
|
|
|
Conclusions |
Cardiovascular disease is a major epidemic in the Western world and
lifestyle alterations have unequivocally increased its incidence.
One of the contributing factors is undoubtedly elevated blood
pressure level, which virtually does not exist in communities with a
low sodium (< 50 mmol/day), high potassium (> 100 mmol/day)
intake.
Community intervention strategies have rarely been based on full
objective proof provided by double-blind clinical trials. If we had
waited for this we would not have introduced an antismoking campaign
or fluoridation for dental caries, as neither of these strategies was
based on full objective proof. To see the effects of stopping smoking
we would not randomly allocate people to a non-smoking and a smoking
group, and then, if 90% of the non-smoking group fail to stop smoking,
state that the results show no difference and that smoking is not
harmful. Likewise, we would not and did not randomly allocate
non-smokers to a smoking and non-smoking group to provide evidence of
a harmful effect. We would also not randomly allocate normotensive
people to a high (200-300 mmol) and a low (< 70 mmol) sodium intake,
as it would be unethical to subject the high salt group to the expected
blood pressure increase.
This experiment, with imperfections, has been provided by people
living a hunter-gatherer lifestyle who have come into contact with
and adopted a Western diet and lifestyle: in all cases this has
demonstrated the harmful effect of high salt intake
elevating blood pressure and increasing cardiovascular death.
At present the evidence for sodium intake and its harmful effects and
the idea of reducing salt intake are widely accepted by the public, but
many people find salt restriction difficult to achieve. This is
because of the excessive amount of sodium chloride in many common
foods, the relative lack of low sodium alternatives, and the
difficulties in adequately labelling foods.
If we want to ameliorate the worldwide epidemic of cardiovascular
disease, restriction of sodium chloride intake, together with
smoking reduction, obesity prevention, decreased saturated fat
intake and increased activity, is an essential part of achieving this
aim.
|
|
|
Conflict of interest |
|
None. I do not have any connection with or receive funds from the food
and salt industries or any related commercial interests.
|
|
| Authors' details |
|
Department of Physiology, University of Melbourne, Melbourne, VIC
3052.
Trefor O Morgan, MD, FRACP, Professor of Physiology.
Reprints will not be available from the author.
Correspondence:
Professor T O Morgan, Department of Physiology, University of
Melbourne, Parkville, VIC 3052.
Email: t.morganATphysiology.unimelb.edu.au
©MJA 1998
Make a
comment
 Link to salt debate references
|
|
|
Readers may print a single copy for personal use. No further
reproduction or distribution of the articles
should proceed without the permission of the publisher. For
permission, contact the
Australasian Medical Publishing Company.
Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au>".
<URL: http://www.mja.com.au/>
© 1998 Medical Journal of Australia.
We appreciate
your comments.
|
The 100-year conflict: Introduction