salt intake and cvd - bmj

7/28/2019 Salt Intake and CVD - BMJ

1/9

RESEARCH

Salt intake, stroke, and cardiovascular disease: meta-analysis of prospective studies

Pasquale Strazzullo, professor of medicine,1 Lanfranco DElia, clinical lecturer in medicine,1 Ngianga-Bakwin

Kandala, principal research fellow in medical statistics,2 Francesco P Cappuccio, professor of cardiovascular

medicine and epidemiology2

ABSTRACT

Objective To assess the relation between the level of

habitual salt intake and stroke or total cardiovascular

disease outcome.Design Systematic review and meta-analysis of

prospective studies published 1966-2008.

Data sources Medline (1966-2008), Embase (from 1988),

AMED (from 1985), CINAHL (from 1982), Psychinfo (from

1985), and the Cochrane Library.

Review methods For each study, relative risks and 95%

confidence intervals were extracted and pooled with a

random effect model, weighting for the inverse of the

variance. Heterogeneity, publication bias, subgroup, and

meta-regression analyses were performed. Criteria for

inclusion were prospective adult population study,

assessment of salt intake as baseline exposure,

assessment of either stroke or total cardiovasculardisease as outcome, follow-up of at least three years,

indication of number of participants exposed and number

of events across different salt intake categories.

Results There were 19 independent cohort samples from

13 studies, with 177 025 participants (follow-up 3.

5-19 years) and over 11 000 vascular events. Higher salt

intake was associated with greater risk of stroke (pooled

relative risk 1.23, 95% confidence interval 1.06 to 1.43;

P=0.007) and cardiovascular disease (1.14, 0.99 to 1.32;

P=0.07), with no significant evidence of publication bias.

For cardiovascular disease, sensitivity analysis showed

that the exclusion of a single study led to a pooled

estimate of 1.17 (1.02 to 1.34; P=

0.02). The associationsobserved were greater the larger the difference in sodium

intake and the longer the follow-up.

Conclusions High salt intake is associated with

significantly increased risk of stroke and total

cardiovascular disease. Because of imprecision in

measurement of salt intake, these effectsizes arelikely to

be underestimated. These results support the role of a

substantial population reduction in salt intake for the

prevention of cardiovascular disease.

INTRODUCTION

During the past century, the evidence for the risks

imposed on human health by excess salt consumptionhas become compelling. The causal relation between

habitual dietary salt intake and blood pressure hasbeen established through experimental, epidemiologi-cal, migration, and intervention studies. Most adult

populations around the world have average daily saltintakeshigherthan6g,andformanyineasternEuropeand Asia higher than 12 g. International recommenda-tionssuggest thataveragepopulation salt intake shouldbe less than 5-6 g per day. Population based inter-vention studies and randomised controlled clinicaltrials have shown that it is possible to achieve signifi-cant reductions in blood pressure with reduced saltintake in people with and without hypertension.1

Based on the effects of high salt intake on blood pres-sure and on the prominent role of high blood pressurein promoting cardiovascular diseases, it has been sug-gested that a population-wide reduction in salt intake

could substantially reduce the incidence of cardio-vascular disease.2 On the basis of the results of ameta-analysis of randomised controlled trials of saltreduction,3 it was estimated that a reductionin habitualdietary salt intakeof 6 g a day would be associated withreductions in systolic/diastolic blood pressure of7/4 mm Hg in people with hypertension and 4/2 mmHg in those without hypertension. At the populationlevel these reductions in blood pressure could predictan average lower rate of 24% for stroke and 18% forcoronary heart disease.4 Validation of these predic-tions by a randomised controlled trial of the effects oflong term reduction in dietary salt on morbidity and

mortality from cardiovascular disease would providedefinite proof. At present, a study of this kind is notavailable and, in fact, it is extremely unlikely that itwill ever be performed because of practicaldifficulties,the long duration required, and high costs. Neverthe-less, prospective cohort studies performed in the pastthree decades that measured the levels of dietary saltintake at baseline and recorded the incidence of vascu-lar events have provided important indirect evidence.Most of these studies found evidence of such relation,although few hadenough power to attain statistical sig-nificance.

We performed a systematic review and meta-analy-

sis of the prospective studies of habitual dietary saltintake and incidence of stroke and total cardiovascular

1Department of Clinical andExperimental Medicine, FedericoII University of Naples MedicalSchool, Naples, Italy2

University of Warwick, WHOCollaborating Centre for Nutrition,Warwick Medical School, ClinicalSciences Research Institute,Coventry CV2 2DX

Correspondence to: P [email protected], FP [email protected]

Cite this as: BMJ2009;339:b4567doi:10.1136/bmj.b4567

BMJ | ONLINE FIRST | bmj.com page 1 of 9


2/9

disease using strictly predetermined criteria for inclu-sionorexclusion.Weassessedwhetherornottheover-all evidence in prospective studies supports thepresence of a relation between levels of dietary saltintake and both stroke and cardiovascular outcomesand calculated an estimate of the risk.

METHODS

Data sources and searches

We performed a systematic search for publicationsusing Medline (1966-2008), Embase (from 1988),AMED (from 1985), CINAHL (from 1982), Psychinfo(from 1985), and the Cochrane Library. Search strate-gies used subject headings and key words with no lan-guage restrictions. Further information was retrievedthrough a manual search of references from recentreviews and relevant published original studies. Weexamined reference lists of the relevant reviews, iden-tified studies, and reviewed the cited literature.5

Study selection

Two reviewers (LD and N-BK) independently extra-cted the data. Discrepancies about inclusion of studiesand interpretation of data were resolved by arbitration(PS or FPC), and consensus was reached after discus-sion. In the case of missing data for potentially suitablestudies, we contacted authors and asked them to pro-vide the necessary information. To be included in themeta-analysis a published study had to be an originalarticle published from January 1966 to December2008, be a prospective population study, assess saltintake as baseline exposure, determine either strokeor total cardiovascular disease prospectivelyas the out-

come, follow participants for at least three years,include an adult population, and indicate the numberof participants exposed and the rate or number ofevents in different categories of salt intake.

Of the 3246 publications retrieved, we identified 15studies that met the inclusion criteria. Onewas a dupli-cate analysis of a singlecohort previously describedbythe same authors6 7 and another8 referred to the samecohort (national health and nutrition examination sur-vey(NHANES) I) analysed by otherauthors with morestringent criteria.9 We therefore included 13 studies inthe meta-analysis that provided suitable data on 19population samples610-21 (tables 1 and 2).

Data extraction

From the identified studies and respective populationswe recorded publication reference, total number ofparticipants, country, sex, age (mean, median, orrange), recruitment time, follow-up (years), outcomereported (stroke, cardiovascular disease) and methodof outcome assessment, number (rate) of events,method of assessing salt intake, and level of salt intakein different categories.

Categorisation of salt intake differed among studies.Some reported thenumber of subjects exposed andtherate (number) of events across the distribution of salt

intake; others reported differences in the event rate fora 100 mmol/day difference in sodium intake, as in the

studies by He et al9 and Tuomilehto et al.13 In the lasttwo cases we used the relative risk or hazard ratioreported by the authors for the analysis. In all thecases in which categorisation of the study participantsby level of salt intake was available, we calculated therelative risk of higher versus lower salt intake by com-paringthe event rate in the twocategories with a differ-ence in average salt intake closest to 100 mmol ofsodium or about 6 g of salt a day.

Statistical analysis

We evaluated the quality of the studies included in themeta-analysis with the Downs and Black scoresystem.21 We extracted relative risks or hazard ratiosfrom the selected publications and calculated theirstandard errors from the respective confidence inter-vals. Thevalue from each study andthe correspondingstandard error were transformed into their naturallogarithms to stabilise the variances and to normalisetheir distribution. The pooled relative risk (and 95%confidence interval) was estimated with a randomeffect model, weighting for the inverse of thevariance.22 The heterogeneity among studies wastested by Q statistic and quantified by H statistic and I2

statistic.23 The influence of individual studies, fromwhich the meta-analysis estimates are derived, wasexamined by omitting one study at a time to see theextent to which inferences depend on a particularstudy or group of studies (sensitivity analysis). Sub-group or meta-regression analyses were used to iden-tify associations between risk of stroke orcardiovascular disease and relevant study characteris-tics (age and sex of participants, year of publication,

durationof follow-up,method of assessment of sodiumintake, difference in sodium level, control for baselineblood pressure) as possible sources of heterogeneity.We used funnel plot asymmetry to detect publicationbiasand applied Eggers regression test to measure anyasymmetry.2425 All statistical analyses were performedwith MIX software version 1.726 and Stata software formeta-regression analysis.

RESULTS

Characteristics of the study cohorts

We included in the meta-analysis 13 studies reportingon 19 independent cohorts (table 1). There were

177 025 participants from six different countries (sixstudies from the United States, two each from Finlandand Japan, one each from the Netherlands, Scotland,and Taiwan). Eleven studies recruited both male andfemale participants, while two studies included onlymen. Follow-up ranged from 3.5 to 19 years. Four stu-dies reported only strokeevents (either total strokerateor stroke deaths), three only cardiovascular disease(total cardiovascular disease rate or cardiovascular dis-ease deaths), and six reported both. Salt intake wasassessed by 24 hour dietary recall (n=4), food fre-quency questionnaire (n=4), 24 hour urine excretion(n=4), and questionnaire (n=1). In total there were

5346 strokes reported and 5161 total cardiovasculardisease events. Of the 11 studies that included both

RESEARCH

page 2 of 9 BMJ | ONLINE FIRST | bmj.com


3/9

men and women, five reported outcomesseparately.69 13-15TheTOHPstudyincludedtwodiffer-ent cohorts (I and II) 17 and the study by He and cow-orkers provided separate findings for men and womenor, alternatively, for normal weight and overweightparticipants.9 Overall, data on the relation betweensalt consumption and stroke were available from 14cohorts and on the relation between salt intake andcardiovascular disease from 14 cohorts.

The overall study quality, evaluated by the Downsand Black score,averaged15.5 (range12-18) on a scaleof 19 (table 1).

Salt intake and risk of stroke

Table 2 provides data on the relation between saltintake and risk of stroke in each of the 14 cohortsincluded in our study. Figure 1 shows the results ofthe pooled analysis. In the pooled analysis, higher saltintake was associated with greater risk of stroke (rela-tive risk 1.23, 95% confidence interval 1.06 to 1.43;P=0.007). Therewas significant heterogeneity betweenstudies (P=0.04; I2=61%). The funnel plot did not show

asymmetry, thus excluding publication bias (Eggerstest P=0.26; see appendix on bmj.com). As shown in

figure 1 for the individual cohorts included in the ana-lysis, we found a trend towards a direct associationbetween salt intake and risk of stroke in nine cohorts,which was significant in four. We observed a non-sig-nificant inverse trend in three cohorts.

Sensitivity analysis showed that the pooled estimateof the effect of salt intake on risk of stroke did not varysubstantially with the exclusion of any one study; inparticular, the exclusion of the study by Umesawa et

al,19

which accounted for about 40% of all participantsin the meta-analysis and nearly 20% of all strokes,resulted in a pooled relative risk of 1.19 (1.03 to 1.39),P=0.022.

Salt intake and risk of cardiovascular disease

Table 2 provides data on the association between saltintake and the risk of cardiovascular disease in 14cohorts. In the pooled analysis, there was an associa-tion between higher salt intake and risk of cardio-vascular disease (1.14, 0.99 to 1.32; P=0.07) (fig 2).The heterogeneity between studies was significant(P


4/9

evaluation of individual studies showed a trendtowards a direct association between salt intake andrisk of cardiovascular disease in 10 cohorts, with signif-

icantly higher relative risk in six. An inverse trend wasobserved in four cohorts and was significant in one.Sensitivity analysis showed that the exclusion of theonly study showing a significant inverse trend 6 led toa pooled estimate of relative risk of 1.17 (1.02 to 1.34),P=0.02 (fig 2). Further exclusion of the study by Ume-sawa et al,19 which accounted for over 50% of all parti-cipants and about 40% of all cardiovascular diseaseevents led to a pooled relative risk of 1.14 (0.99 to1.31), P=0.06.

Sources of heterogeneity

AgeMeta-regression analyses indicated no associa-

tion between mean age of study participants and effectof sodium intake on the risk of stroke: exp(b)=1.01

(0.99 to 1.03). Likewise, meta-regression showed noassociation between age and effect of sodium intakeon the risk of cardiovascular disease: exp(b)=0.99

(0.97 to 1.02).SexThree studies reported data for men and

women separately for incidence of stroke.61314 Thepooled estimates from these three studies were 1.30(0.64 to 2.65; P=0.47) and 1.56 (1.14 to 2.13; P


5/9

urine collection the pooled risk estimate was 1.16 (0.94to1.44;P=0.17).61316 In five cohorts that used food fre-quency questionnaires or dietary recall the pooled esti-mate for cardiovascular disease was 1.21 (0.92 to 1.59;P=0.17),9151920 and in nine cohorts that used 24 hoururine collection the pooled risk estimate for cardio-vascular disease was 1.10 (0.92 to 1.31; P=0.32).612131617

Baseline blood pressureor hypertension statusInthestu-dies that provided relative risk estimates adjusted forbaseline blood pressure or hypertension status, thepooled relative risk was 1.22 (1.02 to 1.45; P=0.03)for stroke (nine cohorts) and 1.25 (0.99 to 1.57;P=0.06) for cardiovascular disease (seven cohorts).

Baseline body mass index (BMI) or body weightIn thestudies that provided relative risk estimates adjustedfor baseline BMI or body weight, the pooled relative

risk was 1.20 (1.02 to 1.40; P=

0.02) for stroke (10cohorts) and 1.22 (1.00 to 1.49; P=0.05) for cardio-vascular disease (10 cohorts).

Length of follow-upMeta-regression analysisshowed a significant association between duration offollow-up andthe effectof sodiumon the risk of stroke.The log relative risk was estimated to increase by 0.07per increase of one year of follow-up: exp(b)=1.07(1.04 to 1.10). The estimated variance between studies(heterogeneity) was reduced from 0.05 to 0.02. In con-trast, however, we found no association between dura-tion of follow-up and effect of sodium on the risk ofcardiovascular disease: exp(b)=0.98 (0.95 to 1.02).

Dose-response analysis

Variance weighted leastsquares regression of the risk of stroke on the study

specific difference between higher and lower cate-gories of sodium intake (differences in sodium intakevalues reported in table 2) provided evidence of a sig-nificant direct association (exp(b)=1.06 (1.03 to 1.10)),indicating a 6% increase in the rate of stroke for every50 mmol/daydifference in sodiumintake. There wasasimilar trend for the risk of cardiovascular disease (exp(b)=1.19 (0.69 to 2.07)), that was not significant.

Time trend (year of publication)Starting with the firstpublished study10 we calculated the cumulative pooledrelative risk by stepwise addition of the results of theother available studies up to the last one published inJuly 2008.19 Figure 3 shows the results of these cumu-lative meta-analyses. The pooled relative risk forstroke stabilised early in the 1.20-1.30 interval andachieved significance starting in 2001. Similar resultswereobtainedin the analysisof cardiovascular disease,

for which the pooled relative risk estimate also stabi-lised early and close to the final value, achievingsignif-icance starting in 1999.

DISCUSSION

This meta-analysis shows unequivocally that highersalt intake is associated with a greater incidence ofstrokes and totalcardiovascular events. Oursystematicreview identified 13 relevant and suitable studies pub-lished from 1996 to 2008. These studies provided evi-dence from 170 000 people contributing overall morethan 10 000 vascular events.

Cardiovascular diseasesare the major cause of death

among people aged over 60 and second among thoseaged 15-59. According to the World Health

Kagan 198510

Hu 199211

Alderman 19956

Men

Women

He 19999

Normal weight

Overweight

Tuomilheto 200113

Men

Women

Nagata 200414

Men

Women

Cohen 200615

Geleijnse 200716

Larsson 200818

Umesawa 200819

Combined effect: P=0.007

Heterogeneity: P=0.04, Eggers test: P=0.26

0.92 (0.60 to 1.42)

1.79 (1.18 to 2.70)

0.59 (0.10 to 3.43)

2.10 (1.01 to 4.33)

0.99 (0.81 to 1.20)

1.39 (1.10 to 1.76)

1.00 (0.68 to 1.48)

1.34 (0.87 to 2.06)

2.34 (1.23 to 4.47)

1.70 (0.96 to 3.00)

0.56 (0.28 to 1.11)

1.08 (0.81 to 1.45)

1.04 (0.93 to 1.17)

1.55 (1.20 to 2.00)

1.23 (1.06 to 1.43)

100

150

120

100

100

100

100

110

92

92

69

84

85

0.1 1 10

Study

Favourable Adverse

Higher salt intake

Relative risk(95% CI)


Sodium difference(mmol/day)

7895

8562

1900

1037

6797

2688

1173

1263

13 355

15 724

7154

1448

26 556

58 730

154 282

Samplesize

238

104

17

6

430

250

43

41

137

132

79

181

2702

986

5346

Events

10

4

13.7

5

13.6

12.7

3.5

19

13

7

Follow-up(years)

Fig 1 | Risk of incident stroke associated with higher compared with lower salt intake in 14 population cohorts from 10

published prospective studies including 154 282 participants and 5346 events

RESEARCH



6/9

Organization, 62% of all strokes and 49% of coronaryheart disease events are attributable to high bloodpressure.27 The direct causal relation between levelsof dietary salt intake and blood pressure at the popula-tion level has also been recognised.12829 Given thegraded causal relation between blood pressure andcardiovascular disease, beginning at around 115 mmHg systolic pressure,30 it is reasonable to expect con-siderable benefit on the rate of cardiovascular diseasefrom a reduction in salt intake.

Association between salt intake, stroke, and cardiovascular

disease

The results of this meta-analysis provide evidence of adirect association between high dietary salt intake and

risk of stroke. Despite the considerable heterogeneitybetween the 14 cohorts available for the analysis, theresults are strengthened by the lack of major publica-tionbiasandbytheobservationofasignificantassocia-tion in four individual cohorts included in the analysis,whereas in none was an inverse statistical associationapparent. The pooled relative risk indicates a 23%greater risk of stroke for an average difference insodium intake (weighted for the population size ofeach study) of 86 mmol (equivalent to about 5 g ofsalt a day). Sensitivity analysis with the exclusion of asingle study, on the basis of its particular weight withregard to both number of participants and events, only

moderately reduced thedifference in risk (from 23%to19%), which remained significant.

Likewise, the pooled analysis of the 12 cohorts forwhich data on cardiovascular disease outcome wereavailable(aftertheexclusionofasingleoutlier)showeda direct association between higher salt intake and riskof cardiovascular disease, with a pooled relative risk of1.17.4 A trend in this direction occurred in as many asnine of the 12 cohorts and was significant in six. Therewas an inverse trend in three cohorts.151620 The studyby Alderman et al,6 showing a relative risk in men of0.37, has been challenged because of the low numberof events recorded and several methodological inade-quacies, the most important being the evaluation ofhabitual salt consumption on the basis of 24 hoururine collection obtained shortly after the study parti-cipants had been instructed to reduce their usual levelof sodium intake.31 The results of sensitivity analysisindicate that the exclusion of this single study fromour meta-analysis strengthens the estimate. The addi-tional exclusion of a large Japanese cohort providing ahigh proportion of participants and events overall19

only slightly reduced the pooled relative risk estimate(from 1.17 to 1.14) and the level of significance (to0.06).

Evaluation of main sources of heterogeneity

We used subgroup and meta-regression analyses toassess the influenceof several factors on the associationbetween habitual sodium intake and risk of stroke or

cardiovascular disease. For both stroke and cardio-vascular disease outcomes, separate analyses of the

Alderman 19956

Men

Women

Tunstall-Pedoe 199712

Men

Women

He 19999

Men

Women

Tuomilheto 200113

Men

Women

Cohen 200615

Geleijnse 200716

Cook (I) 200717

Cook (II) 200717

Umesawa 200819

Cohen 200820

Combined effect: P=0.07

Combined effect (without Alderman): P=0.02

Heterogeneity: P


7/9

male and female cohorts suggest that the associationsare consistent and not significantly different betweenthe sexes. Similar results were obtained with respectto the method of assessment of habitual sodium intakeused in the various studies.

Eight studies provided data adjusted for baselineblood pressure or hypertension status. Separate eva-luations of these studies provided relative risk esti-mates for both stroke and cardiovascular diseasesimilartothoseobtainedforthetotalnumberofstudiesincluded in the meta-analysis. This finding seems atvariance with the hypothesis that the effect of salt oncardiovascular risk is substantially mediated by itsunfavourable action on blood pressure. Adjustmentfor baseline blood pressure or hypertension status

only partially corrects for the overall influence ofblood pressure on the study results in as much as it

does not account for changes in blood pressure occur-ring during the observation period, a problem morerelevant the longer the follow-up period. Part of theassociation observed, however, might be mediated byfactors otherthan blood pressure,and there is evidencein the literature of deleterious effects of high salt intakeon left ventricular mass,32-34 arterial stiffness,35 andrenal function,3637 which are not totally explained byits effect on blood pressure.

Overweight and obesity are often associated withhigh blood pressure and are causally involved in thedevelopment of hypertension.38 Nine out of 13 studiesincluded in the meta-analysis provided relative riskestimates adjusted for BMI or body weight at entryinto the study.9 14-21 Therefore, as for blood pressure,the association between habitual sodium intake andrisk of stroke and cardiovascular disease seems partlyindependent from theinfluence of excessbody weight.Two studies, however, reported a significant inter-action between overweight and habitual sodium intakeon the risk of cardiovascular events.9 13 This finding isconsistent with the description of alterations in renaltubular sodium handling in obese individuals, makingthem particularly sensitive to the effects of high saltintake.39

Study limitations

The studies included in our meta-analysis were hetero-geneous regarding sample size, number of events, andduration of follow-up, with a few cohorts having smallnumbers. In the calculation of the pooled relative riskwe weighted the results of the individual studies forsample size but did not account for the duration of fol-

low-up. Our meta-regression analysis indicated thatthe longer the follow-up the greater the effect of habi-tual sodium intake on the risk of stroke but not, appar-ently, on the risk of total cardiovascular events.Possible explanations for this discrepancy are thehigher mean age at occurrence of stroke, whichwould increase the chances of an event the longer thefollow-up, and the closer relation of high blood pres-sure to stroke compared with other types of vascularevents.

Theestimateof thebaselinepopulation salt intakeineach study was based on a single measurement(whether through 24 hour urine collection or dietary

assessment). We were therefore unable to correct forregression dilution bias. Because of thelarge dayto dayvariability within people in salt consumption and theconsequent diluting effect imposed on the average esti-mate of exposure, our estimates of risk are probablyunderestimated.

Categorisation of saltintake wasalsoheterogeneous:some studies stratified the population by categories ofsodium intake and comparedcardiovascular outcomesacross categories, other studies gave a difference inoutcome for a given difference (for example,100 mmol/24 h) in sodium intake or excretion. Tostandardise our comparison between higher and

lower salt consumption we sought to refer to a differ-ence as close as possible to 100 mmol or 6 g a day

Stroke

Kagan 198510

Hu 199211

Alderman 19956

Men

WomenHe 19999

Normal weight

Overweight

Tuomilheto 200113

Men

Women

Nagata 200414

Men

Women

Cohen 200615

Geleijnse 200716

Larsson 200818

Umesawa 200819

Cardiovascular disease

Tunstall-Pedoe 199712

Men

Women

He 19999

Men

Women

Tuomilheto 200113

Men

Women

Cohen 200615

Geleijnse 200716

Cook (I) 200717

Cook (II) 200717

Umesawa 200819

Cohen 200820

1

P=0.045

P=0.007

P=0.013

P=0.020

5


0.5

Fig 3 | Cumulative meta-analysis. Evaluation of time trends

(year of publication) in relation between habitual sodium

intake and risk of stroke or cardiovascular disease

RESEARCH



8/9

between high and low salt intake. Nevertheless, thereremained appreciable differences in this respectbetween studies. We tried to overcome the problemwith meta-regression analysis, which provided evi-dence of a highly significant dose-response relationbetween the difference in sodium intake and theincrease in risk of both stroke and cardiovascular dis-ease.

Implications

The habitual salt intake in most Western countries isclose to 10 g a day (and much higher in many EasternEuropean and Asian countries), and we calculated thatthe average difference between higher and lower saltintake across the study cohorts included in our meta-analysis was 5 g a day. Given this approach, we believethat, despite the inherent inaccuracies, the results ofour meta-analysis are applicable to real life conditions.A reduction of 5 g (about one teaspoon) of salt wouldbring consumption close to the WHO recommendedlevel (5 g a day at the population level).

AccordingtoarecentreportoftheWorldHeartFed-eration there are over 5.5 million deaths a year fromstroke throughout the world and close to 17.5 milliondeaths a yearfromcardiovascular disease.40 Given thatthe case fatality rates for stroke is estimated at one inthree and those for total cardiovascular disease at one

in five, a 23% reduction in the rate of stroke and a 17%overall reduction in the rate of cardiovascular diseaseattributable to a reduction in population salt intakecould avert some one and a quarter million deathsfrom stroke and almost three million deaths fromcardiovascular disease each year. Many studies havealso shown that a reduction in salt intake is costeffective,41-43 arguing for the more widespread intro-duction of national programmes to reduce dietary saltconsumption. In recent years, a few countries havemade some progress towards reduction of habitualsalt intake through a voluntaryapproach1 or by regula-tion, as in Finland,42 but levels of salt consumption are

still far from the WHO recommended targets. Thereare many reasons for these delays. One barrier to a

more effective implementation of public health poli-cies has been the historical opposition of the foodindustry,44 based on the arguments that the availableevidence does not show significant benefits on hardend points at a population level from a moderatereduction in salt intake. Our study now clearly

addresses those doubts. Some progress has beenmade in the past few years by closer collaborationbetween governments, public health bodies, andsome sectors of the industry on avoluntary basis, asin the UK, with the reformulation of many food itemstowardsalowersaltcontentandproposalsofimprovedlabelling.Theseeffortshaveledtoareductionof0.9gaday (or about 10%) in population salt intake in fouryears (from 9.5 to 8.6 g a day), still far from the recom-mended 6 g a day initial targets thatwereset in the UK.While the voluntary approach is the preferred choicefor many governments, the regulatory approach hasadvantages,45 sometimes being the most efficient,effective, and cost effective way of achieving publichealth targets.41-43 For population salt intake toapproach the recommended targets within a reason-able time frame, anupstream approach is nowneces-sary alongside the traditional downstream publichealth approach based on health promotion and beha-vioural changes.

Contributors: PS and FPC conceived the study aims and design,contributed to the systematic review and data extraction, performed the

analysis, interpreted the results, and drafted the manuscript. LDE and N-

BK contributed to the data extraction, interpretation of results, and

revision of the manuscript. PS is guarantor.

Funding: This study was funded in part by an EC Grant (FP7-HEALTH-2007-201550). The publication does not necessarily represent the

decisions or the stated policy of WHO and the designations employ ed and

the presentation of the material do not imply the expression of anyopinion on the part of WHO.

Competing interests: None declared.Ethical approval: Not required.Data sharing: Search strategy and flow chart are available from thecorresponding authors by e-mail.

1 HeFJ, MacGregor GA.A comprehensivereview onsaltand healthandcurrent experience of worldwide saltreduction programmes.J HumHypertens2009;23:363-84.

2 World Health Organization. Reducing salt intake in populations:report of a WHOforumand technicalmeeting. WHO, 2007:1-60.

3 HeFJ, MacGregor GA. Effectof modestsalt reduction onbloodpressure: a meta-analysis of randomized trials. Implications forpublic health.J HumHypertens2002;16:761-70.

4 HeFJ, MacGregor GA. How far shouldsalt intakebe reduced?Hypertension2003;42:1093-9.

5 Moher D, CookDJ, EastwoodS, Olkin I,Rennie D,Stroup DF.Improving the quality of reports of meta-analyses of randomisedcontrolled trials: the QUOROM statement. Quality of reporting ofmeta-analyses. Lancet1999;354:1896-900.

6 AldermanMH, MadhavanS, Cohen H,Sealey JE, Laragh JH. Lowurinary sodium is associated with greater riskof myocardialinfarction among treated hypertensive men.Hypertension1995;25:1144-52.

7 AldermanMH, Sealey JE, Cohen H,Madhavan S and LaraghJH.Urinary sodium excretion and myocardial infarction in hypertensivepatients: a prospective cohort study.Am J ClinNutr1997;65(suppl):682-6S.

8 Alderman MH,Cohen H, Madhavan S. Dietary sodiumintake andmortality: the national health and nutrition examination survey(NHANES I). Lancet1998;351:781-5.

9 HeJ, OgdenLG, VupputuriS, BazzanoLA, LoriaC,WheltonPK.Dietarysodium intake and subsequent riskof cardiovascular disease inoverweight adults.JAMA1999;282:2027-34.

10 KaganA, Popper JS,RhoadsGG, Yano K.Dietaryandotherrisk factorsfor stroke in HawaiianJapanese men.Stroke 1985;16:390-6.

WHAT IS ALREADY KNOWN ON THIS TOPIC

Experimental, epidemiological, migration, and intervention studies have shown a causalrelation between habitual dietary salt intake and blood pressure

Population based interventionstudies and meta-analyses of randomised controlled trialshave shown that it is possible to achieve significant reductions in blood pressure withreduced salt intake in both hypertensive and normotensive individuals

WHAT THIS STUDY ADDS

Higher salt intake is associated with significantly greater incidence of strokes and totalcardiovascular events, with a dose dependent association

A difference of 5 g a day in habitual salt intake is associated with a 23% difference in the rateof stroke and 17% difference in the rate of total cardiovascular disease

Each year a 5 g reductionin dailysaltintakeat thepopulationlevel could avert someone anda quarter million deaths from stroke and almost three million deaths from cardiovasculardisease worldwide

RESEARCH

page 8 of 9 BMJ | ONLINE FIRST | bmj.com


9/9

11 HuHH, Sheng WY, Chu FL, Lan CF, ChiangBN. Incidence ofstroke inTaiwan.Stroke 1992;23:1237-41.

12 Tunstall-Pedoe H, Woodward M, TavendaleR, ABrook R,McCluskey MK. Comparison of the prediction by 27 different factorsof coronary heart disease anddeath in menand women of theScottish heart health study: cohort study. BMJ1997;315:722-9.

13 Tuomilehto J, Jousilahti P, RastenyteD, Moltchanov V, TanskanenA,Pietinen, P, et al. Urinary sodium excretion and cardiovascularmortality in Finland: a prospective study. Lancet2001;357:848-51.

14 Nagata C, Takatsuka N, Shimizu N, Shimizu H. Sodiumintake andrisk of death from stroke in Japanesemen andwomen.Stroke2004;35:1543-7.

15 Cohen HW, Hailpern SM,Fang J, Alderman MH.Sodium intake andmortality in theNHANES II follow-upstudy.Am J Med2006;119:275 e7-14.

16 GeleijnseJM, Witteman JC,Stijnen T, Kloos MW,Hofman A,Grobbee DE.Sodium andpotassiumintake andriskofcardiovascular events and all-cause mortality: the Rotterdam study.Eur J Epidemiol2007;22:763-70.

17 Cook NR,Cutler JA,Obarzanek E, Buring JE, Rexrode KM,KumanyikaSK, et al.Long term effects of dietary sodiumreductionon cardiovascular disease outcomes: observational follow-up of thetrials of hypertension prevention (TOHP).BMJ2007;33:885-8.

18 Larsson SC,Virtanen MJ,MarsM, Mannisto S, Pietinen P, Albanes D,et al. Magnesium, calcium, potassium, and sodium intakes and riskof stroke in male smokers.ArchIntern Med2008;168:459-65.

19 Umesawa M, IsoH, Date C, YamamotoA, ToyoshimaH, Watanabe Y,et al. Relations between dietary sodium and potassium intakes and

mortality fromcardiovasculardisease: theJapan collaborative cohortstudy for evaluation of cancer risks.AmJ ClinNutr2008;88:195-202.

20 CohenHW, HailpernSM, AldermanMH. Sodiumintake andmortalityfollow-up in the third national health and nutrition examinationsurvey (NHANES III).J Gen Intern Med2008;23:1297-302.

21 Downs SH,BlackN. The feasibility of creatinga checklistfor theassessment of the methodological quality both of randomised andnon-randomised studies of health care interventions.J EpidemiolCommunity Health 1998;52:377-84.

22 Dersimonian R, Laird N. Meta-analysis in clinical trials.Control ClinTrials1986;7:177-88.

23 Higgins JP,ThompsonSG, Deeks JJ,Altman DG.Measuringinconsistency in meta-analyses. BMJ2003;327:557-60.

24 Egger M, Davey SG,Schneider M, MinderC. Bias in meta-analysisdetected by a simple, graphical test. BMJ1997;315:629-34.

25 SuttonAJ, Duval SJ,Tweedie RL, Abrams KR,Jones DR.Empiricalassessment of effect of publication bias on meta-analyses.BMJ2000;320:1574-7.

26 Bax L,Yu LM, Ikeda N,Tsuruta H,MoonsKG. Development andvalidation of MIX: comprehensive free software for meta-analysis ofcausal research data. BMCMedRes Methodol2006;6:50.

27 World HealthOrganization.The world health report 2002reducingrisks promoting healthy life. WHO, 2002.

28 Scientific Advisory Committee on Nutrition. Salt and health.Stationery Office, 2003:1-134.

29 World Health Organization. Reducing salt intake in populations.Report of a WHOforumand technicalmeeting. WHO, 2007:1-60.

30 Lewington S, ClarkeR, Qizilbash N, Peto R, Collins R. Age-specificrelevance of usual blood pressure to vascular mortality: a meta-analysis of individual data forone million adultsin 61 prospectivestudies. Lancet2002;360:1903-13.

31 Karppanen H, Mervaala E. Sodiumintake and mortality. Lancet1998;351:1509.

32 KupariM, Koskinen P, Virolainen J. Correlates of left ventricularmass

in a population sample aged 36 to 37 years. Focuson lifestyle andsalt intake.Circulation1994;89:1041-50.

33 SchmiederRE, MesserliFH, Garavaglia GE, Nunez BD. Dietary saltintake. A determinant of cardiac involvement in essentialhypertension. Circulation1988;78:951-6.

34 Du CalairG, RibsteinJ, Daures JP, MimranA. Sodiumandleftventricular mass in untreated hypertensive and normotensivesubjects.Am J Physiol 1992;263:H177-81.

35 AvolioAP, Clyde KM, Beard TC, Cooke HM, HoKK, ORourkeMF.Improved arterial distensibility in normotensive subjects on a lowsalt diet.Arteriosclerosis 1986;6:166-9.

36 Du CalairG, Ribstein J, MimranA. Dietary sodium andtarget organdamage in essential hypertension.Am J Hypertens2002;15:222-9.

37 Verhave JC,Hillege HL, Burgerhof JG,Janssen WM,Gansevoort RT,Navis GJ,et al.Sodium intake affects urinary albumin excretionespecially in overweight subjects.J Intern Med2004;256:324-30.

38 National Institutes of Health. Clinicalguidelineson the identification,evaluation, and treatment of overweight and obesity in adultsthe

evidence report.Obes Res1998;(suppl 2):51-209S.39 Strazzullo P, BarbaG, Cappuccio FP,SianiA, Trevisan M, Farinaro E,

et al.Alteredrenalsodiumhandling inmen withabdominaladiposityand insulin resistance: a linkto hypertension.J Hypertens2001;19:2157-64.

40 World Health Organization. Theworldhealth report2004changinghistory.WHO, 2004.

41 Selmer RM,Kristiansen IS,Haglerod A, Graff-IversenS, LarsonHK,Meyer HE,et al.Cost and healthconsequences of reducing thepopulation intake of salt.J EpidemiolCommunity Health2000;54:697-702.

42 MurrayCJ,Lauer JA,HutubessyRC, NiessenL, TomijimaN, Rodgers A,et al. Effectiveness and costs of interventions to lowersystolic bloodpressureand cholesterol:a global andregional analysison reductionof cardiovascular-disease risk. Lancet2003;361:717-25.

43 AsariaP, Chisholm D, Mathers C, EzzatiM, Beaglehole R. Chronicdiseaseprevention: health effects andfinancial costs of strategies toreduce salt intake and control tobacco use.Lancet2007;370:2044-53.

44 Godlee F. The food industry fightsfor salt.BMJ1996;312:1239-40.

45 Cappuccio FP. Salt and cardiovascular disease.BMJ2007;334:859-60.

Accepted: 22 October 2009

RESEARCH


salt intake and cvd - bmj

Documents