heart-foundation-sodium-evidence-paper

An evidence-based nutrition statement from the National Heart Foundation of New Zealand’s Nutrition Advisory Committee

sodium

Author: D Roberts BPhEd.BSc.PGDipDiet.NZRD March 1999

contributors committee members Professor J Mann PhD, MD, FRACP, FRSNZ. Professor of Human Nutrition and Medicine, University of Otago, Dunedin, New Zealand.

Dr A Chisholm DipHSc, MCApSc, PhD, NZRD. Research Dietitian, Department of Human Nutrition, University of Otago, Dunedin, New Zealand.

Associate Professor L Eyres PhD, MBA. Technical and Developments General Manager, New Zealand Dairy Foods, Auckland, New Zealand.

M McKerchar DipSc, PGDipSc (Com Nutr), NZRD. Maori Health Promoter, Southern Public Health Services, Invercargill, New Zealand.

J Reid BSc (Hons), PGDipDiet, MPH, NZRD. Nutrition Advisor, Ministry of Health, Wellington, New Zealand.

D Roberts BPhEd, BSc, PGDipDiet, NZRD. National Dietitian, the National Heart Foundation of New Zealand, Auckland, New Zealand.

Associate Professor B Swinburn MB ChB, FRACP, MD. Medical Director, the National Heart Foundation of New Zealand, Auckland, New Zealand.

A Tuffin BA, DipTchg. Senior Lecturer Health Education, College of Education, Massey University, Palmerston North, New Zealand.

L Young BHSc, PGDipSc (Com Nutr), NZRD. Food Industry Manager, the National Heart Foundation of New Zealand, Auckland, Zealand.

other contributors J Bremer DipHSc, NZRD. Consulting Dietitian, Christchurch, New Zealand.

S Mackay BCApSc, PGDipPH, MSc. Nutritionist, Nelson, New Zealand.

M Seddon MB ChB, MPH. Public Health Registrar, the National Heart Foundation of New Zealand, Auckland, New Zealand.

reviewing organisations The Ministry of Health, Wellington, New Zealand.

The Nutrition and Metabolism Advisory Committee of the National Heart Foundation of Australia, Canberra, Australia.

1

table of contents

section summary

3

Explanatory notes

3

background paper

4

Examination of the evidence

4

Introduction

4

Observational studies

4

Intervention studies

6

Biological mechanism

7

Heterogeneous responses

7

Effect of lowering blood pressure on risk of Cardiovascular Disease

8

conclusion evidence tables

8 10

Evidence Table 3: Sodium reduction and blood pressure

10

Evidence Table 4: Sodium reduction and blood pressure

11

abbreviations

12

references

14

basic evidence-grading strategy Meta-analyses of, or single, well designed randomized controlled trials

Grade A

Non-randomized and/or controlled studies

Grade B

Historical cohort, population-based

Grade C

Case series

Grade D

Expert opinion

Grade E

2

section summary A high salt (sodium) intake has been linked to high blood pressure, which is an important risk factor for heart and blood vessel disease, especially stroke.

For the general population Choose a wide variety of low - or reduced-salt (sodium) products, and prepare meals with minimal added salt.

For those at high risk of cardiovascular disease As for the general population.

Explanatory notes •

The majority (up to 85%) of sodium comes in hidden form in processed and manufactured foods, while only 15% is accounted for when added in cooking or from the salt shaker at the table.

•

Foods that contribute a large amount of sodium to the NZ include bread, butter, cheese, biscuits, fish, processed meats, some breakfast cereals, most take-away food, many sauces, and much canned and processed food.

•

Avoiding salty foods and not adding salt in cooking or at the table reduces sodium intake by about 50 mmol.d-1 or approximately 3 g.d-1 of salt. To reduce sodium intake by 100 mmol.d-1 requires, in addition to avoiding salt added to the cooking, avoiding many common processed foods (Level of evidence A)

•

A reduction in sodium intake of 100 mmol.d-1 is associated with lower SBP and DBP by 3-6/0-3 mm Hg (Level of evidence C).

•

An average SBP lower by 2 mm Hg across a population is associated with significant and meaningful benefit: lower CHD and CVD death rates by 4-5% and lower all cause mortality by 3% (Level of evidence B).

•

Ordinary table salt is sodium chloride-40% sodium by weight. One half teaspoon of salt equals approximately 1,000 mg sodium (45 mmol).

•

The change in blood pressure for a given sodium intake will depend on a person’s age and existing blood pressure. The older the person and the higher their blood pressure, the greater is the decrease in blood pressure associated with a given decrease in sodium intake (Level of evidence A).

•

A higher intake of sodium from as early as six months of age can cause statistically significant increases in SBP in proceeding years (Level of evidence A).

•

Reduced sodium intake should be only one component of a comprehensive nutritional approach to lowering blood pressure which should also include prevention and treatment of overweight or obesity, limitation of alcohol intake, and strategies that ensure adequate intake of potassium, magnesium and calcium (Level of evidence A).

•

The heterogeneity in the response of individuals to sodium restriction suggests the possible existence of a group of hyper-responders, but there is no clear indication as to how these individuals might be defined.

•

High salt intake is associated with high calcium excretion and higher osteoporotic risk (Level of evidence C).

3

background paper Examination of the evidence Introduction Although sodium is the principal cation in the extracellular fluid and plays an essential role in the regulation of body fluids, it is thought that most New Zealanders consume more sodium and salt (sodium chloride) than their body needs. Numerous studies now exist to support the notion that habitually high sodium intakes may confer negative health outcomes.

There is evidence to suggest that higher sodium intakes proportionally increase calcium losses in the urine, leading to the possibility of long term negative calcium balance. This loss may significantly 1234 increase the risk of osteoporosis and contribute to this common problem in the NZ population , , , . In relation to CVD, high sodium intakes are considered to be one of the primary nutritional factors 5 responsible for raising blood pressure alongside low dietary potassium, alcohol, and obesity . Increased blood pressure can lead to hypertension, CHD, stroke and kidney disease.

It is estimated that the majority (up to 85%) of dietary sodium may come in hidden form in processed and manufactured foods, while only 15% is potentially accounted for when added in cooking or from the salt shaker at the table6,7. The Food and Nutrition Guidelines for New Zealanders recommend preparing meals and choosing pre-prepared foods, drinks and snacks that are low in salt8. The Nutrition Taskforce reported that the greatest sources of sodium in the diet were from prepared foods, including bread, butter, cheese, biscuits, cakes, some breakfast cereals and most take-away and canned foods 9.

Observational studies A positive across-population association between salt and blood pressure was first reported by Dahl in 1960, who found, over five population groups, a remarkable straight-line relation between the average 10 sodium intake of a population and the prevalence of hypertension . Dahl also reported that hypertension was uncommon in populations whose members consumed less than four or 5 g salt per day and hypothesised that salt intake increased the probability of elevated blood pressure in a group although not necessarily in the individual.

Publication of Dahl’s findings stimulated others to review the international literature for data on mean 11 12 13 sodium intake and mean blood pressure of populations , , . These studies generally confirmed the Dahl correlation but to a greater or lesser extent suffered from a number of uncertainties and biases. The major concern that arose from these studies was that the data were not derived from one standardised source but from a variety of studies published in the literature, in which unstandardised, and often unspecified methods, were used. Additionally, few data on confounding variables were available at the time of publication.

In 1988 the INTERSALT Cooperative Research Group published cross-sectional data illustrating the relationship between 24 hour urinary electrolyte excretion and blood pressure. The data came from 14 10,079 men and women, aged 20-59 years from 52 centres around the world . INTERSALT found significant positive relations between 24 hour urinary sodium excretion and SBP and DBP in individual participants. The populations with very low sodium excretion had low, median, blood pressure, a low prevalence of hypertension and virtually no increase in blood pressure with age. In the remaining centres, sodium was related to blood pressure in individuals and influenced the extent to which blood pressure increased with age.

4

Best estimates from INTERSALT data on the size of its significant sodium-blood pressure relations were as follows 15. From within-population analyses, SBP/DBP is 3-6/0-3 mm Hg on average per 100 mmol.d-1 lower 24 hour sodium excretion with and without BMI in the analyses, with multivariate adjustment for reliability, and with control for age, sex, alcohol intake and potassium; for subgroups, SBP/DBP is 5/2 mm Hg lower for non hypertensive persons, 8/4 mm Hg lower for those aged 20-39 years, 5/2 mm Hg lower for men, and 8/3 mm Hg lower for women, without BMI in the analyses. From estimates from across populations, controlled for age, sex, BMI, and alcohol intake, with population median sodium lower by 100 mmol.d-1, median SBP/DBP is lower by 5/2 mm Hg; for persons aged 40-59 years, SBP/DBP is lower by 7/4 mm Hg; for those aged 20-39 years, it is lower by 2/0.4 mm Hg. With sodium lower by 100 mmol.d-1, the average difference in the population sample SBP/DBP, age 55 years, is a decrease of 10-12/6-7 mm Hg. The authors acknowledge that these results were confounded by the different measurement errors of correlated variables. In multiple regression analyses against a dependent variable such as blood pressure, the well-measured variable BMI, may dominate over the poorly-estimated variable, 24 hour sodium excretion. Adjusting the association of 24 hour urinary sodium excretion with blood pressure for BMI may, therefore, result in over-adjustment. However, these within-population and cross-population 16 estimates generally agree with each other and with those from other analyses .

The Health Professionals Study, a prospective study conducted over four years, examined the relationship of various nutritional factors with hypertension in a cohort of 30,681, predominantly white, 17 US, male, health professionals, without diagnosed hypertension . Calculation of sodium intake, as measured by semi-quantitative food frequency questionnaire, included the amount of sodium added during cooking of staple foods (rice, pasta, potatoes), soup, vegetables and meat, and the number of shakes of salt added to food at the table each day. The authors found no significant association for the intake of dietary sodium and risk of hypertension, while age, relative weight, and alcohol consumption were found to be the strongest predictors.

In the Health Professionals Study, the authors used self-reported blood pressure and diagnosis of hypertension. This may have influenced the findings in the event that knowledge of borderline hypertension might have caused men to change their dietary habits during follow-up. The direction of change would be more likely to obscure relations with nutrients than to create them. For this reason, the results of this study do not exclude a possible positive association between sodium intake and risk of hypertension, however, because some men may have reduced their sodium intake in response to high blood pressure values.

In a more recent report, Beard et al used data derived from the Dietary and Nutritional Survey of British 18 Adults to record the association between blood pressure and dietary factors . A random sample of 2,635 adults aged 16-64 years (excluding pregnant women and those not living in private households in Great Britain) recorded seven-day food dairies and then compared these to electrolyte excretion rates and biochemical assays. For all subjects, sodium was a significant predictor of SBP and DBP. For an increase in sodium from 100-200 mmol.d-1, the authors predicted increases in SBP of 2.8 mm Hg (women) and 1.9 mm Hg (men). Regression dilution bias, from measurement error with sodium, may have reduced the effect of sodium. Correction for this suggests that the true increases in blood pressure would be 7.1 mm Hg (women) and 4.9 mm Hg (men). Other limitations of this survey that may have attenuated the blood pressure results were the effects of medications, behavioral factors and over adjustment for body fat.

5

Intervention studies Beginning in the 1970’s, trials were being undertaken to determine the effects of moderate dietary salt reduction in adults with less severe high blood pressure or blood pressure in the non-hypertensive range.

Several overviews consistently found significant reductions in SBP in both normotensive and hypertensive 19 20 21 (younger and older) adults because of a reduction in salt intake by trial intervention groups , , . These findings lead to the conclusion that: (i) habitual high salt intake adversely influences blood pressure of adult populations; and (ii) recommendations for reduced salt ingestion by the population, from postweaning on, are sound public health policy.

In this regard, trial data on infants merit attention. In the Rotterdam RCT of salt intake and blood pressure during the first six months of life, the group assigned to the higher salt intake had significantly higher 22 SBP by 2.1 mm Hg at six months, compared to the group assigned to lower salt intake . Fifteen years later, the group with more salt during the first six months of life had significantly higher SBP (adjusted for 23 confounding) by 3.6 mm Hg, even though there had been no known intervention since infancy .

These data lend further support to the concept of the primary prevention of blood pressure rise with age, and of adult adverse blood pressure levels, by means of improved lifestyles, including lower salt intake, 24 from post-weaning .

Law, Frost and Wald aggregated the results of 68 crossover trials and ten RCT's of dietary salt reduction25. They reported that the effects of salt reduction on blood pressure were larger than had previously been thought. They also found that salt reduction lowered blood pressure at all levels and not just in those with high blood pressure, though the extent of this reduction did depend on the initial blood pressure reading. This apparent difference was explained by one of two factors.

Firstly the subjects with high blood pressure were generally older than those with normal blood pressure and the response of blood pressure to salt reduction increases with age, and secondly, the duration of salt reduction in people with normal blood pressure was usually short (four weeks or less in 16 of 21 trials). Simple dietary manipulation, that is, avoiding salty foods and not adding salt in cooking or at the table reduces sodium intake by about 50 mmol.d-1 (about 3 g.d-1 of salt). In people aged 50-59 years, a reduction of daily sodium intake of 50 mmol.d-1 would, after a few weeks, lower SBP by an average of 5 mm Hg and by 7 mm Hg in those with high blood pressure; DBP would be lowered by half as much. It is estimated that such a reduction in salt intake by a whole western population would reduce the incidence of stroke by 26% and of CHD by 15%.

A reduction in sodium intake of 100 mmol.d-1 would reduce the incidence of stroke by 39% and that of CHD by 30%. This would require, in addition to avoiding salt added to cooking, avoiding many common processed foods and would be feasible only if manufacturers did not add salt to food in processing. It is, therefore, recommended that food manufacturers, caterers and individuals explore and grasp the opportunities for reducing the sodium content of foods and meals where possible26. The TOHP was a twopart project designed to provide experimental evidence regarding the value of non-pharmacological interventions in the prevention of hypertension19,27. Phase one tested the feasibility and short-term efficacy of several promising interventions in lowering SBP and DBP. Phase two was designed to provide a longer-term (36-48 months) test of the efficacy of the interventions that proved successful during phase one. Briefly, those enrolled in phase one were healthy, 30-54 year old men and women who were not being treated with anti-hypertensive medications and who had an average DBP of 80-89 mm Hg. Participants were randomized to one of three lifestyle interventions, including weight loss, sodium reduction and stress management, and compared to usual-care groups over an 18 month period. At 18 months, only weight loss and sodium reduction had produced significant reductions in blood pressure.

For the sodium reduction group, declines in SBP and DBP of 2.1 mm Hg and 1.2 mm Hg were recorded.

The core elements of the sodium-reduction, intervention programme were participant education and motivation, provision of information on nutrition and behavi our-change skills, goal-setting, monitoring of goal achievement, problem-solving and relapse management. Instruction in food preparation, meal selection and shopping choices, and participation in taste testing of appealing low-salt foods were important components of the programme. Participants were instructed on the use of herbs, curries, and spices as alternative flavouring to sodium.

In phase two of this trial, 2,382 men and women, aged 30-54 years, that were not taking antihypertensive drugs and were moderately overweight, were randomized to weight reduction, sodium

6

reduction, a combination of both, or usual-care. At 36 months, blood-pressure decreases remained greater in the active intervention groups than in the usual-care group (sodium-reduction group 1.2/0.7 mm Hg and combination group 1.1/0.6 mm Hg), although somewhat less than at six months. Through 48 months, the incidence of hypertension was significantly less in each active intervention group than the usual-care group. This study demonstrated that although the effects on average blood pressure declined over time, reductions in hypertension incidence are achievable with appropriate lifestyle modifications.

As recently as 1998, Whelton et al conducted a RCT on a total of 875 men and women to determine the feasibility, efficacy, and safety of sodium reduction and weight loss in older persons with hypertension28.

The main result from this study were that older patients with high blood pressure, receiving antihypertensive drugs and with well-controlled blood pressure, initiating moderate lowering of sodium intake and moderate reduction of obesity, singly and combined, effected further significant decreases in SBP and DBP while anti-hypertensive drug treatment was continued. While the results of this study are impressive, it should be noted that participants in the study were generally healthy prior to treatment and reasonably well-motivated. Despite this, the results are concordant with similar, recently conducted RCT's 29,30.

The preceding data needs to be considered in the context of mounting evidence that excess dietary salt has possible adverse effects on the population risks of several other major threats to health. These include gastric cancer, asthma, osteoporosis, cardiac left ventricular hypertrophy and 31 32 33 34 35 36 stroke , , , , , .

Biological mechanism Exactly how sodium contributes to a rise in blood pressure is not clear. An inherited or acquired defect in the kidney’s ability to excrete excess sodium may lead to elevated levels of sodium, chloride and water in the blood. Normally an increase in blood volume prompts secretion of natriuretic hormone, enabling the kidney to void some of the excess sodium in the urine. In some hypertensives, the kidneys may be unable to excrete normal amounts of sodium at normal blood pressures due to a natriuretic handicap’, one factor of which may be blood insulin, which has been associated with increased sodium reabsorption 37 by the kidney .

Another hypothesis proposes that increasing intracellular sodium could inhibit sodium-calcium exchange and cause accumulation of calcium in the vascular musculature, leading to increased muscle tone and 38 increased resistance, thus raising blood pressure .

Much of the evidence in favour of a role for sodium in the development of hypertension comes from epidemiological data. Epidemiological studies, mostly cross-sectional, have identified dietary sodium as well as other factors related to lifestyle, as a possible determinant of blood pressure levels.

Heterogeneous responses 39 40 41

Blood pressure responses to increases and decreases in dietary salt are heterogeneous , , . In some hypertensive individuals, decreases in blood pressure with salt restriction are clinically significant and approach that achieved with medication. In others, little or no change in blood pressure occurs, whereas in still others, blood pressure may actually increase with salt restriction. The heterogeneous responses are partly acquired and involve the influences of age, the intake of other electrolytes, and the influence of 42 43 certain medications , . Genetic predisposition may also play a substantial role because salt sensitivity is increased in coloured individuals and in persons with type 2 diabetes.

7

Effect of lowering blood pressure on risk of Cardiovascular Disease Even small differences in population average blood pressure relate importantly to CVD and all-cause 44 45 mortality risk , . Representative data, derived from multivariate analysis of 16 year prospective findings for the MRFIT cohort, are given in Table_1. The focus is on estimated favourable effects of lower SBP because from age 40 years on, SBP is more strongly related to risks than is DBP. As is evident, even SBP lower by 2 mm Hg is associated with significant and meaningful benefit: lower CHD and CVD death rates by 4-5% and lower all-cause mortality by 3%. For 10 mm Hg lower SBP, estimated from INTERSALT ecological data on sodium and SBP slope with age, CHD and CVD mortality rates were lower by approximately 20-21% and the all-cause death rate was lower by 14%.

conclusion Exactly how sodium contributes to a rise in blood pressure is not clear. Much of the evidence in favour of a role for sodium in the development of high blood pressure comes from epidemiological data. These studies have identified dietary sodium, as well as other factors related to lifestyle, as possible determinants of blood pressure levels. INTERSALT found significant positive relations between 24 hour urinary sodium excretion and SBP and DBP in individual participants within and across populations.

A number of clinical intervention trials have recently been able to establish a beneficial effect of sodium reduction, either alone or in combination with other non-pharmacologic interventions, in the treatment and prevention of high blood pressure. Most of these beneficial effects can simply be achieved by not adding salt to food and avoiding foods containing large amounts of sodium or salt. Blood pressure responses to sodium reduction are heterogeneous. This is partly acquired and involves the influences of age, the intake of other electrolytes, the influence of certain medications and genetic predisposition.

Table 1. Relationship of lower SBP to risk of mortality in MRFITz

46

16 year mortality lower by: Systolic blood pressure lower by:

CHD

All CVD

All causes

Percentage (%) 2 mm Hg

4.4

4.6

3.0

3 mm Hg

6.5

6.8

4.4 5.8

4 mm Hg

8.6

9.0

6 mm Hg

12.6

13.2

8.6

8 mm Hg

16.5

17.2

11.3

10 mm Hg

20.1

21.0

13.9

20 mm Hg

36.2

37.6

25.9

z

Cohort of 342,815 men free at baseline of history of hospitalisation for heart attack and drug-treated diabetes; data are based on coefficients for the relation of baseline SBP to 16 year risk of CHD, CVD, and all causes of death, from multivariate proportional regression analyses with baseline blood cholesterol, cigarettes smoked, age, sex, income, and race also in the model.

8

Table 2. Salt content of some common foods Low salt foods

High salt foods

Very high salt foods

Fresh fruits and vegetables

Salty crackers

Salt at the table

Fresh lean meats and poultry

Bacon

Anchovies

Fresh fish and seafoods

Salamis

Monosodium glutamate

Plain wholegrain products

Sausages

Kelp

Unprocessed breakfast cereals

Saveloys

Low -fat milk and milk products

Meat pastes/pates Cheddar cheese Salted peanuts Salted crisps Tinned meats Salted peanut butter Pizzas Dips using soup mix Corned and pickled meats Lean ham Meat/yeast extracts Soup and stock powders Tinned and instant meals Processed breakfast cereals Smoked and tinned fish Soy sauce Foods packed in brine Canned baked beans and spaghetti

9

evidence tables Evidence Table 3: Sodium reduction and blood pressure Key Words:

Sodium, blood pressure, clinical trials, meta-analysis

Reference :

Cutler JA, Follman D and Allender PS. Randomized trials of sodium reduction: an overview. Am J Clin 47 Nutr 1997; 65(suppl): S643-S51 .

Study Type/Grade

Meta-analysis Grade A

Outcomes

Primary: To determine the effects of sodium reduction on SBP and DBP.

Design

Number of studies: 32.

Number of subjects: 2,635.

Focused on a discrete clinical question: Yes. Explicit description of literature search: Yes, MEDLINE search, supplemented with a review of bibliographies.

State methodological standards used to select studies for inclusion in meta-analysis: Studies included if they were published and had, random allocation to experimental conditions, designs free of confounding, reporting of an objective measure of change in sodium intake, reporting of change in SBP, DBP or both, study subjects that were not prepubertal children, and sodium intake goals within usual levels for free-living adults in countries similar to the United States.

Demographics of study population: Not specified.

Validity

Is the study type appropriate for the question(s) being asked?: Yes. Data tested for homogeneity: Yes, reports selected were abstracted onto a standard form by two reviewers independently, and differences were then reconciled by consensus. Information abstracted included: design, nature and duration of intervention and follow -up, methods of blinding to intervention assignment, sample size, characteristics of the study population, and outcome data. Outcome measures abstracted were the differences between sodium reduction and control groups for mean change in blood pressure, 24 hour sodium excretion, and potential confounding variables.

Evidence of publication bias: No indication for DBP from graphic and regression analyses that small negative trials were underrepresented; for SBP, the graphic plot was more suggestive of bias, but the regression slope was not significant.

Summary: Valid methodology although no demographic data.

Results

Quantified results: Per 100 mmol.d-1 sodium reduction Hypertensives:

(± 95% CIs):

4.83 (± 1.04)/2.45 (± 0.68) mm Hg for SBP/DBP.

Normotensives:

(± 95% CIs):

1.90 (± 0.72)/1.09 (± 0.48) mm Hg for SBP/DBP.

All Trials: (± 95% CIs):

2.81 (± 0.58)/1.52 (± 0.38) mm Hg for SBP/DBP.

Authors’ Conclusions

"...moderate sodium reduction low ers SBP and DBP over periods of several weeks to a few years. An effect is seen in both hypertensive and normotensive subjects: about 5/3 mm Hg and 2/1 mm Hg, respectively".

Reviewer’s Summary

Agree with authors’ conclusions, however, more information is required on graded blood pressure responses to various dose levels of sodium. No demonstrable benefits for mortality or morbidity

10

Evidence Table 4: Sodium reduction and blood pressure Key Words:

Sodium, blood pressure, meta-analysis

Reference :

Law MR, Frost CD and Wald NJ. Analysis of data from trials of salt reduction. BMJ 1991; 302: 819-2425.

Study Type/Grade

Meta-analysis Grade depends on primary studies

Outcomes

Primary: To determine whether the reduction in blood pressure achieved in trials of dietary salt reduction are quantitatively consistent with estimates derived from blood pressure and sodium intake in different populations.

Design

Number of studies: 68 crossover trials and 10 randomized controlled trials of dietary salt reduction.

Number of subjects: Not specified.

Focused on a discrete clinical question: Yes. Explicit description of literature search: No. State methodological standards used to select studies for inclusion in meta-analysis: Subdivided the data from studies that recruited both subjects with high blood pressure and subjects with normal blood pressure. Only included trials with a crossover design or randomized parallel control group. Excluded trials that combined salt restriction with another intervention, trials in which patients were taking anti-hypertensive drugs, and trials in which low and high sodium intake were both not measured in at least one 24 hour urine collection.

Demographics of study population: Average age range 16-63 years.

Validity

Is the study type appropriate for the question(s) being asked?: Yes. Data tested for homogeneity: Not tested for but implied that data were heterogeneous.

Evidence of publication bias: Randomisation and crossover should minimise bias and possible confounding within trials. Exclusion criteria and search methods may limit external validity although this is unknown.

Summary: Without description of literature search there may be some bias that could influence validity of results.

Results

Quantified results: In people aged 50-59 years a reduction in daily sodium intake of 50 mmol.d-1 (3 g.d-1 salt) would after a few weeks lower SBP by an average 5 mm Hg, and by 7 mm Hg in those with high blood pressure (≥170 mm Hg); DBP would be lowered by half as much. It is estimated that such a reduction in salt intake would reduce the incidence of stroke by 26% and of ischaemic heart disease by 15%.

Authors’ Conclusions

"The effect of universal moderate dietary salt reduction on mortality from stroke and CHD would be substantial-larger, indeed, than could be achieved by fully implementing recommended policy for treating high blood pressure with drugs".

Reviewer’s Summary

Agree with authors’ conclusion based on the evidence that has been used. Question selection methods for trials and how this may affect the external validity of the findings.

11

abbreviations AHA

American Heart Association

ATBC

Alpha Tocopherol, Beta Carotene Cancer Prevention Study

BMI

Body Mass Index

CARE

Cholesterol and Recurrent Events Trial

CARET

Beta Carotene and Retinol Efficacy Trial

CCPT

Chicago Coronary Prevention Trial

CHAOS

Cambridge Heart Antioxidant Study

CHO

Carbohydrate

CHD

Coronary Heart Disease

CI

Confidence Interval

CVD

Cardiovascular Disease

DART

Diet and Reinfarction Trial

DASH

Dietary Approaches to Stop Hypertension

DBP

Diastolic Blood Pressure

DHA

Docosahexenoic Acid

DNSBA

Dietary and Nutrition Survey of British Adults

EPA

Eicosapentenoic Acid

EURAMIC

European Antioxidant Myocardial Infarction and Breast Cancer

GI

Glycaemic Index

HDL-C

High Density Lipoprotein Cholesterol

INTERSALT

International Study of SALT

LED

Low Energy Diet

LDL-C

Low Density Lipoprotein Cholesterol

LNA

Alpha Linolenic Acid

Lp (a)

Lipoprotein (a)

LRC-CPPT

Lipid Research Clinic Coronary Primary Prevention Trial

MEDLINE

Medical Literature Analysis and Retrieval System Online

MeSH

Medical Subject Heading

MI

Myocardial Infarction

MONICA

Monitoring of Trends and Determinants in Cardiovascular Disease

MRFIT

Multiple Risk Factor Intervention Trial

MUFA

Monounsaturated Fat

NAS

Normative Aging Study

NCEP

National Cholesterol Education Program

NHANES

National Health and Nutrition Examination Survey

NHLBI

National Heart, Lung and Blood Institute 12

NSP

Non-Starch Polysaccharide

NZ

New Zealand

NZDA

New Zealand Dietetic Association

OPPT

Oslo Primary Prevention Trial

PTCA

Percutaneous Transluminal Coronary Angioplasty

PUFA

Polyunsaturated Fat

RDA

Recommended Daily Allowance

RCT

Randomized Controlled Trial

4S

Scandinavian Simvastatin Survival Study

SBP

Systolic Blood Pressure

SFA

Saturated Fat

STARS

St Thomas’ Atherosclerosis Regression Study

TG

Triglyceride

tHcy

Total Homocysteine

TOHP

Trials of Hypertension Prevention

TONE

Trial of Nonpharmacologic Interventions in the Elderly

USDA

United States Department of Agriculture

UK

United Kingdom

US

United States of America

VLED

Very Low Energy Diet

VLDL-C

Very Low Density Lipoprotein Cholesterol

WC

Waist Circumference

WHO

World Health Organisation

WHR

Waist to Hip Ratio

WOSCOPS

West of Scotland Coronary Prevention Study

13

references 1.

Cappuccio FP. Dietary prevention of osteoporosis: Are we ignoring the evidence? Am J Clin Nutr 1996; 63: 787-88.

2.

Devine A, Criddle RA, Dick IM et al. A longitudinal study of the effect of sodium and calcium intakes on regional bone density in postmenopausal women. Am J Clin Nutr 1995; 62: 740-45.

3.

Matkovic V, Ilich JZ, Andon MB et al. Urinary calcium, sodium and bone mass of young females.

Am J Clin Nutr 1995; 62: 417-25.

4.

McParland BE, Goulding A and Campbell AJ. Dietary salt affects biochemical markers of resorption and formation of bone in elderly women. BMJ 1989; 299: 834-35.

5.

Bonita R and Beaglehole R. Primary prevention of cardiovascular disease in older NZers: A Report to the National Health Committee. Wellington, National Health Committee, 1998.

6.

Sanchez-Castillo CP, Warrender S, Whitehead TP et al. An assessment of the sources of dietary salt in a British population. Clin Sci 1987; 72: 95-102.

7.

Engstrom A, Tobelmann RC and Albertson AM. Sodium intake trends and food choices. Am J Clin Nutr 1997; 65: S704-S07.

8.

Department of Health. NZ Food and Nutrition Guidelines. Wellington, Department of Health, 1991.

9.

Nutrition Taskforce. Food for Health. Wellington, Department of Health, 1991.

10.

Dahl LK. Possible role of salt intake in the development of essential hypertension, in Cottier P and Bock KD (eds): Essential Hypertension-An International Symposium. Berlin, Springer-Verlag, 1960, 53-65.

11.

Gleibermann L. Blood pressure and dietary salt in human populations. Ecology and Food Nutrition 1973; 2: 142-56.

12.

McCarron DA, Henry HJ and Morris CD. Human nutrition and blood pressure regulation. An integrated approach. Hypertension 1982; 4 (suppl III): III-2-III-13.

13.

MacGregor GA. Sodium is more important than calcium in essential hypertension. Hypertension 1985; 127: 893-904.

14.

INTERSALT Cooperative Research Group. Intersalt: An international study of electrolyte excretion and blood pressure. Results from 24-hour urinary sodium and potassium excretion. BMJ 1988; 297: 319-28.

15.

Elliott P, Stamler J, Nicholas R et al. Intersalt revisited: Further analyses of 24-hour sodium excretion and blood pressure within and across populations. BMJ 1996; 312: 1249-52.

16.

Elliott P. Observational studies of salt and blood pressure. Hypertension 1991; 17 (suppl I): I-3-8.

17.

Ascherio A, Rimm EB, Giovannucci EL et al. A prospective study of nutritional factors and hypertension among US men. Circulation 1992; 86: 1475-84.

18.

Beard TC, Blizzard L, O’Brien DJ et al. Association between blood pressure and dietary factors in the dietary and nutritional survey of British Adults. Arch Intern Med 1997; 157: 234-38.

19.

Whelton PK, Kumanyika SK, Cook NR et al. Efficacy of non-pharmacologic interventions in adults with high-normal blood pressure: results from phase one of the Trials of Hypertension Prevention.

Am J Clin Nutr 1997; 65 (suppl): S652-S60.

20.

Midgley JP, Matthew AG, Greenwood CM et al. Effect of reduced dietary sodium on blood pressure. JAMA 1996; 275: 1590-97.

21.

Stamler J. Letter to the editor (on Midgley JP, Matthew AG, Greenwood CM et al. JAMA 1996; 275: 1590-1597). JAMA 1996; 276: 1467.

14

22.

Hofman, Hazebroek A, and Valkenburg HA. A randomized trial of sodium intake and blood pressure in newborn infants. JAMA 1983; 250: 370-73.

23.

Geleijnse JM. Sodium, potassium, and blood pressure. Studies in the young and the old. PhD dissertation. Rotterdam, Netherlands: Erasmus University, 1996.

24.

National High Blood Pressure Education Program Working Group. National High Blood Pressure Education Program Working Group report on primary prevention of hypertension. Arch Intern Med 1993; 153: 186-208.

25.

Law MR, Frost CD and Wald NJ. By how much does dietary salt reduction lower blood pressure.

BMJ 1991; 302: 811-24.

26.

Nutritional aspects of cardiovascular disease. Report of the Cardiovascular Review group, Committee on the Medical Aspects of Food Policy. London, HMSO, 1994.

27.

The Trials of Hypertension Prevention Collaborative Research Group. Effects of weight loss and sodium reduction intervention on blood pressure and hypertension incidence in overweight people with high-normal blood pressure. Arch Intern Med 1997; 157: 657-67.

28.

Whelton PK, Appel LJ, Espeland MA et al. Sodium reduction and weight loss in the treatment of hypertension in older persons: A randomized controlled trial of non-pharmacologic interventions in the elderly (TONE). JAMA 1998; 279 (11): 839-46.

29.

Cappuccio FP, Markandu ND, Carney C et al. Double-blind randomized trial of modest salt restriction in older people. Lancet 1997; 350: 850-54.

30.

Appel LJ, Moore TJ, Orbarzanek E et al. for the DASH Collaborative Research Group. A clinical trial of the effects of dietary patterns on blood pressure. N Engl J Med 1997; 336: 1117-24.

31.

Stamler J. Dietary salt and blood pressure. Ann NY Acad Sci 1993; 676: 122-56.

32.

Montes G, Cuello C, Correa P et al. Sodium intake and gastric cancer. J Cancer Res Clin Oncol 1985; 109: 42-45.

33.

Knox AJ. Salt and asthma. BMJ 1993; 307: 1159-60.

34.

MacGregor GA and Cappuccio FP. The kidney and essential hypertension: a link to osteoporosis?

J Hypertens 1993; 11: 781-85.

35.

Ferrara LA, De Simone G, Pasanisi F et al. Left ventricular mass reduction during salt depletion in arterial hypertension. Hypertension 1984; 6: 755-59.

36.

Sasaki S, Zhang X-H, and Kesteloot H. Dietary sodium, potassium, saturated fat, alcohol, and stroke mortality. Stroke 1995; 26: 783-89.

37.

Blaustein MP and Hamlyn JM. Sodium transport inhibition, cell calcium, and hypertension: The natriuretic hormone/Na Ca exchange/hypertension hypothesis. Am J Med 1984; 77: 45.

38.

Resnick LM. Dietary calcium and hypertension. J Nutr 1987; 117: 1806.

39.

Miller JZ, Weinberger MH, Daugherty SA et al. Heterogeneity of blood pressure response to dietary sodium restriction in normotensive adults. J Chronic Dis 1987; 40: 245-50.

40.

Weinberger MH, Miller JZ, Luft FC et al. Definitions and characteristics of sodium sensitivity and blood pressure resistance. Hypertension 1986; 8 (suppl II) II-127-34.

41.

Weinberger MH and Fineberg NS. Sodium and volume sensitivity of blood pressure: age and pressure change over time. Hypertension 1991; 18: 67-71.

42.

Luft FC, Miller JZ, Grim CE et al. Salt sensitivity and resistance of blood pressure: age and race as factors in physiological responses. Hypertension 1991; 17 (suppl I) I-102-I-108.

43.

Kotchen TA and Kotchen JM. Dietary sodium and blood pressure: Interactions with other nutrients. Am J Clin Nutr 1997; 65 (suppl): S708-S11.

15

44.

MacMahon S, Peto R, Cutler J et al. Blood pressure, stroke, and coronary heart disease. Part one. Prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet 1990; 335: 765-74.

45.

Stamler J, Stamler R and Neaton JD. Blood pressure, systolic and diastolic, and cardiovascular risks: US population data. Arch Intern Med 1993; 153: 598-615.

46.

Neaton JD, Kuller L, Stamler J et al. Impact of systolic and diastolic blood pressure on cardiovascular mortality. In: Laragh JH, Brenner BMI, eds. Hypertension: physiology, diagnosis, and management. 2nd ed. New York: Raven Press, 1995: 127-44.

47.

Cutler JA, Follman D and Allender PS. Randomised trials of sodium reduction: an overview. Am J Clin Nutr 1997; 65(suppl): S643-S51.

16