American Journal of Hypertension Advance Access published January 27, 2016
Review
A Radical Sodium Reduction Policy Is Not Supported by Randomized Controlled Trials or Observational Studies: Grading the Evidence Niels Graudal1 triglyceride; and (iv) observational studies show that sodium intakes below 2,645 and above 4,945 mg are associated with increased mortality. Given that 90% of the worlds’ population currently consumes sodium within the optimal range of 2,645–4,945 mg, there is no scientific basis for a public health recommendation to alter sodium intake.
The 2013 IOM report on sodium intake in populations1 has focused attention on whether the widely recommended 2,300 mg upper level for sodium intake is beneficial. The sodium intake in 95% of the world’s populations is between 2,620 and 4,830 mg sodium (6.6–12.2 g salt/114–210 mmol).2,3 In spite of this, several health institutions recommend sodium intake be reduced to below 2,000 mg4 to 2,300 mg5,6—meaning that 6–7 billion individuals should alter their diet to accommodate. Such a radical recommendation should be based on solid evidence that this would be safe and beneficial. The highest grade of evidence (GRADE system “High”) comes from randomized controlled trials (RCTs) and the second grade (GRADE “Moderate”) comes from prospective observational studies.7 The present review focuses on GRADE “high” and “moderate” evidence of health outcomes related to sodium intake. The IOM emphasized that outcomes of intermediate variables, specifically effect on blood pressure (BP), are inadequate to assess the health effect of sodium. Nevertheless, because the recommendations that sodium should be reduced to less than 2,300 mg/d are based upon its BP consequences and the denial of side effects,4–6 this
review will also assess BP effect and potential side effects on hormones and lipids.
Correspondence: Niels Graudal ([email protected]). Initially submitted November 25, 2015; date of first revision December 18, 2015; accepted for publication December 22, 2015.
Keywords: blood pressure; catecholamines; cholesterol; hypertension; mortality; observational study; randomized controlled trial; renin– angiotensin–aldosterone system; sodium; triglyceride. doi:10.1093/ajh/hpw006
HIGH GRADE EVIDENCE ON HEALTH OUTCOMES: THE EFFECT OF SODIUM REDUCTION ON CARDIOVASCULAR DISEASE EVENTS AND ALL-CAUSE MORTALITY The 2014 Cochrane review
A recently updated meta-analysis of 8 RCTs with follow-up data on morbidity and mortality found no effect of reduced sodium on all-cause mortality (ACM) (RR = 0.96 (0.83, 1.10)) in 6,603 individuals and a nonsignificant trend favoring reduced cardiovascular disease (CVD) morbidity (RR = 0.76 (0.57, 1.01)) in 3,397 individuals.8 Five of the studies were performed in hypertensive individuals and three in prehypertensive, obese individuals. None were performed in healthy individuals. Furthermore, no study has reduced sodium to less than 2,300 mg. In conclusion, there is no high GRADE evidence to support sodium reduction below 2,300 mg in either the healthy population or even in populations at increased CVD risk.
1Department of Rheumatology VRR4242, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark.
© American Journal of Hypertension, Ltd 2016. All rights reserved. For Permissions, please email: [email protected]
American Journal of Hypertension 1
Downloaded from http://ajh.oxfordjournals.org/ at Main Library of Gazi University on February 12, 2016
Several health institutions recommend sodium intake be reduced to below 2,300 mg, which means that 6–7 billion individuals should alter their diet to accommodate. Such a radical recommendation should be based on solid evidence. However, this review reveals that (i) there are no randomized controlled trials (RCTs) allocating individuals to below 2,300 mg and measuring health outcomes; (ii) RCTs allocating risk groups such as obese prehypertensive individuals and hypertensive individuals down to (but not below) 2,300 mg show no effect of sodium reduction on all-cause mortality; (iii) RCTs allocating individuals to below 2,300 mg show a minimal effect on blood pressure in the healthy population (less than 1 mm Hg) and significant increases in renin, aldosterone, noradrenalin cholesterol, and
Graudal MODERATE GRADE EVIDENCE ON HEALTH OUTCOMES: THE ASSOCIATION OF LOW SODIUM INTAKE WITH CVD EVENTS AND ACM The 2013 IOM report
The 2014 meta-analysis, updated
The IOM committee did not perform a formal meta-analysis of the available data.1 Such an analysis was published a year after the IOM report.13 This analysis confirmed that “higher levels” of sodium intake are associated with increased mortality but found that “higher level” was actually greater than 4,945 mg. This analysis also confirmed a U-shaped association of sodium intake to subsequent mortality, with significantly higher mortality at sodium intake below 2,645 mg. Optimal health outcomes were found when intakes were 2,645–4,945 mg and within this range event rates were constant. Figure 1 shows the previous meta-analysis of the risk
HIGH GRADE EVIDENCE ON BP: EFFECT OF SODIUM REDUCTION BELOW 2,300 MG ON BP
The traditional theoretical background for the sodium reduction policy was the assumption that a reduced sodium intake will lower BP and would therefore inevitably prevent CVD mortality. The potential effect of sodium reduction on BP will depend on the duration of achieved maximal efficacy (i.e., if too short, it may not show the true effect), the dose– response relationship (if the magnitude of sodium reduction is too small, it may not show maximal efficacy), as well as the baseline BP (the effect of sodium reduction is likely to increase with increasing baseline BP). In addition, it has been shown that BP increases with age. Previously, ethnicity was believed to have an impact on the effect of sodium reduction on BP, but factors like baseline BP, baseline sodium intake, study duration, age, and magnitude of sodium reduction can explain the major part of the ethnic association with BP.14 Therefore, taking these factors into consideration, adjustment for ethnicity should not have a meaningful impact. RCTs with longitudinal BP measurements have shown that the maximal reduction of BP produced by limiting sodium intake was reached at 1 week.15 Furthermore, in individuals with a BP below 130/80 mm Hg and a sodium intake below 250 mmol (5,750 mg, approximately corresponding to the 99% upper limit of the world’s populations), there was no dose–response relationship between sodium intake and systolic BP or diastolic BP in studies, which randomized individuals to different doses. In contrast, there was
Figure 1. All-cause mortality, usual sodium vs. low sodium. Results from 9 population representative observational studies adjusted for multiple confounders. Exchanging the first analyses of the NHANES studies1,7 with the re-analyses2,8: HR = 0.87 (0.82, 0.91), P < 0.00001. Further exchanging primary analyses5,11,13 with analyses of low-risk populations6,12,14: HR = 0.86 (0.81, 0.92), P < 0.00001. Abbreviations: CI, confidence interval; HR, hazard ratio; LRP, low-risk population.
2 American Journal of Hypertension
Downloaded from http://ajh.oxfordjournals.org/ at Main Library of Gazi University on February 12, 2016
The 2013 IOM committee reviewed observational studies directly relating sodium intake with morbidity and mortality.1 The committee concluded that there was a positive relationship between “higher levels” of sodium intake and risk of CVD, but “higher levels” was not defined. Furthermore, the previously defined recommended upper level of 2,300 mg of sodium intake was neither confirmed nor rejected. Consequently, this report left more questions than answers. The reason why the report questioned the 2,300 mg limit was that several (but not all) of the observational studies had indicated an increased mortality not only at “higher levels” but also below 2,300 mg, thus indicating a U-shaped association between sodium intake and mortality.9,10 Since then 2 additional studies have independently confirmed the U-shaped association between sodium intake and CVD events and ACM.11,12
of ACM in individuals within the usual sodium intake range compared vs. a low sodium intake below 2,645 mg13 updated with 2 studies.11,12 The present updated analysis is confined to include population samples supposed to be representative for the general population, adjusted for multiple confounders.
Radical Sodium Reduction Policy
a dose–response relationship in individuals with a BP higher than 130/80 mm Hg and a sodium intake above 250 mmol.15 The 2011 Cochrane review
HIGH GRADE EVIDENCE ON HORMONES AND LIPIDS: EFFECT OF SODIUM REDUCTION ON RENIN, ALDOSTERONE, CATECHOLAMINES, CHOLESTEROL, AND TRIGLYCERIDE
Many health institutions, which have issued sodium recommendations,4–6 have chosen to ignore the conclusions of
DISCUSSION
It is remarkable that health institutions think that 95% of the world’s population eat too much salt and recommend that they should reduce their sodium intake to less than 2,300 mg/d. The commitment to the 2,300 mg/d upper limit promulgated in 20055 persists, despite (i) no empiric evidence of a health benefit associated with intakes less than 2,300 mg/d, (ii) failure of the 2013 IOM report to support that recommendation, and (iii) the substantial body of observational data that sodium, like every other essential nutrient, bears a “U” or “J” shaped relation to health outcomes. Indeed, our 2014 metaanalysis of 25 population studies found that a sodium intake
Table 1. Effect of sodium reduction on SBP SBP (mm Hg)
0–25%
N participants
Mean
Sodium
(studies)
difference (95% CI)
reduction (mmol)
274 (3)
−0.34 (−1.51, 0.83)
110
0.57
2,077 (29)
−0.41 (−1.12, 0.30)
156
0.26
5,748 (33)
−2.72 (−3.41, −2.03)
91
0.00001
5,108 (69)
−5.62 (−6.64,−4.60)
104
0.00001
P value
131 mm Hg Abbreviations: CI, confidence interval; SBP, systolic blood pressure.
American Journal of Hypertension 3
Downloaded from http://ajh.oxfordjournals.org/ at Main Library of Gazi University on February 12, 2016
In agreement with data from the longitudinal studies and the dose–response studies,15 the 2011 Cochrane review of 167 studies showed a minimal effect of sodium reduction on BP in individuals with a normal BP (1.27/0.05 mm Hg) and a more meaningful effect in hypertensive individuals (5.18/2.59 mm Hg).16 Table 1 (systolic BP) and Table 2 (diastolic BP) display these results, but in the subgroup of studies with duration of at least 1 week, a sodium intake below 250 mmol and subdivided according to the 25% quartiles of BP of the American population.17 This analysis shows that about 83% of all individuals investigated have a BP above the 50% percentile. Therefore, the RCTs are biased in favor of a higher BP effect of sodium reduction. Consequently, to evaluate the potential effect of sodium reduction, it is necessary to estimate the results in the BP subgroups which actually reflect the general population. The results show that individuals with a BP below the median (119/71) have minimal effects on systolic BP and diastolic BP (Tables 1 and 2). Moreover, the effect up to the 75% percentile is small (2.72/0.75 mm Hg) (Tables 1 and 2). Individuals with a BP in the upper 25% percentile (above 131/78 mm Hg) of the BP distribution have a measurable, but still rather moderate effect of sodium reduction on BP (5.68/2.67 mm Hg)—significantly less than the effect that can be achieved by antihypertensive drugs (usually 10/5 mm Hg or more). It is important to note that all these BP effects are obtained by sodium reductions of 90 mmol or more—corresponding to a 60% sodium reduction for the American population.
the 2013 IOM report. Further, they have chosen to ignore the coherent results of multiple observational studies involving about 500,000 participants11–13 and rely instead on the sodium effect on the single variable of BP, and particularly on studies lasting for at least 4 weeks18,19 although the maximal efficacy of sodium reduction on BP is achieved at 1 week.15,16 By excluding studies shorter than 4 weeks, these meta-analyses exclude almost all studies investigating the effect on surrogate variables other than BP, and therefore these analyses lacked the power to assess the impact of other relevant variables. Moreover, the assumption that other surrogate variables have only “acute” and “transitional” associations18 has not been demonstrated. On the contrary, the permanently salt-deprived Yanomamo Indians had a 10-fold increased urinary excretion of aldosterone.20 The 2011 Cochrane review found that sodium reduction resulted in highly significant increases in hormones and lipids.16 This review has been criticized for the inclusion of “extreme” studies. Table 3 shows a subanalysis of results from studies, which randomize participants to usual sodium in the interval of 2,300 – 5,200 mg or to below 2,300 mg, i.e., the intervention recommended by health authorities. In this analysis, studies with extremely large sodium reductions and also studies with sodium reductions to below 40 mmol have been excluded. Furthermore, studies with duration less than 7 days were excluded. All associations are statistically significant, save for adrenalin. Similar findings concerning renin and aldosterone have been shown in a cross-sectional study.21
Graudal Table 2. Effect of sodium reduction on diastolic blood pressure DBP (mm Hg)
0–25%
N participants
Mean
Sodium
(studies)
difference (95% CI)
reduction (mmol)
1,530 (12)
(P value)
0.20 (−0.49, 0.90)
135
0.57
−0.57 (−1.77, 0.64)
143
0.36
3,081 (17)
−0.75 (−1.47, −0.03)
120
0.04
7,705 (78)
−2.69 (−3.25, −2.14)
92
78 mm Hg
Radical Sodium Reduction Policy
population (less than 1 mm Hg) and a moderate effect in individuals with a high baseline BP (about 5/2.5 mm Hg); (iv) RCTs allocating individuals to below 2,300 mg show significant increases in renin, aldosterone, noradrenalin cholesterol, and triglyceride; and (v) observational studies show that sodium intakes below 2,645 and above 4,945 mg are associated with increased mortality. Given that 90% of the worlds’ population currently consumes sodium within the optimal range, there is no scientific basis for a public health recommendation to alter sodium intake. Thus, a solution to the controversy could be to accept the present evidence and revise the recommendations to be in accordance with the present usual intake (2,645–4,945 mg sodium). Another— maybe impossible—solution could be to perform an RCT, which followed large groups of individuals on either a usual sodium diet above the present recommendations (2,300 mg sodium) or a low-sodium diet in accordance with the recommendations (below 2,300 mg) for a considerable time period recording health events and mortality.
DISCLOSURE
The author declared no conflict of interest.
REFERENCES 1. Institute of Medicine. Sodium Intake in Populations: Assessment of Evidence. National Academies Press: Washington, DC, 2013. 2. McCarron DA, Kazaks AG, Geerling JC, Stern JS, Graudal NA. Normal range of human dietary sodium intake: a perspective based on 24-hour urinary sodium excretion worldwide. Am J Hypertens 2013; 26:1218–1223. 3. Powles J, Fahimi S, Micha R, Khatibzadeh S, Shi P, Ezzati M, Engell RE, Lim SS, Danaei G, Mozaffarian D; Global Burden of Diseases Nutrition and Chronic Diseases Expert Group (NutriCoDE). Global, regional and national sodium intakes in 1990 and 2010: a systematic analysis of 24 h urinary sodium excretion and dietary surveys worldwide. BMJ Open 2013; 3:e003733. 4. WHO. Sodium Intake for Adults and Children. WHO: Geneva, Switzerland, 2012. 5. Institute of Medicine. Dietary Reference Intakes: Water, Potassium, Sodium, Chloride, and Sulfate. National Academies Press: Washington, DC, 2005. 6. Dietary Guidelines for Americans. health.gov 2010. 7. Higgins JPT, Green S (eds). Cochrane Handbook for Systematic Reviews of Interventions. Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011 2014. 8. Adler AJ, Taylor F, Martin N, Gottlieb S, Taylor RS, Ebrahim S. Reduced dietary salt for the prevention of cardiovascular disease. Cochrane Database Syst Rev 2014; 12:CD009217. 9. Thomas MC, Moran J, Forsblom C, Harjutsalo V, Thorn L, Ahola A, Wadén J, Tolonen N, Saraheimo M, Gordin D, Groop PH; FinnDiane Study Group. The association between dietary sodium intake, ESRD, and all-cause mortality in patients with type 1 diabetes. Diabetes Care 2011; 34:861–866. 10. O’Donnell MJ, Yusuf S, Mente A, Gao P, Mann JF, Teo K, McQueen M, Sleight P, Sharma AM, Dans A, Probstfield J, Schmieder RE. Urinary sodium and potassium excretion and risk of cardiovascular events. JAMA 2011; 306:2229–2238. 11. Pfister R, Michels G, Sharp SJ, Luben R, Wareham NJ, Khaw KT. Estimated urinary sodium excretion and risk of heart failure in men and women in the EPIC-Norfolk study. Eur J Heart Fail 2014; 16:394–402.
American Journal of Hypertension 5
Downloaded from http://ajh.oxfordjournals.org/ at Main Library of Gazi University on February 12, 2016
The present review shows that evidence in favor of a sodium intake above 2,300 mg is more compelling than the evidence in favor of a sodium intake below 2,300 mg. The only minimal evidence in favor of sodium reduction below 2,300 mg, namely the very small BP effect in healthy individuals, depend on the linear association between BP and mortality, which is derived from observational studies.26 Interestingly, the health institutions4–6,22,23 accept the outcome of observational studies linking BP with mortality,26 but ignore observational studies, which link sodium intake with mortality.9–13 An explanation could be that also RCTs have shown that lowering BP reduces mortality, but this is shown to be the case only in hypertensive individuals, not in normotensive individuals. Therefore, this inconsistency is remarkable, especially as the BP–mortality link is equivocal. First, the linear association between BP and mortality has been questioned, especially in the low end of the BP interval.27 Second, in the PURE study, a low sodium intake was associated with an increased mortality,12 although it was associated with a low BP,25 and in PURE, the healthiest people, without hypertension, had the greatest risk from low sodium. Third, although beta-blockers reduce BP in hypertensive individuals, they do not reduce mortality.28 This is probably due to side effects. The present review indicates that particularly side effects may explain the inverse association of a low sodium intake with mortality recorded in the observational studies.13 We have not systematically reviewed the association of a sodium intake with BP recorded in observational studies. However, 2 of the largest and most important of these studies25,29 confirm the small BP effects found in RCTs. The systolic BP effect in the PURE study25 was 1.3/0.58 mm Hg/1 g Na in individuals with a normal BP and 2.49/0.91 mm Hg/1 g Na in individuals with hypertension. Intersalt29 found a BP effect of 1.6/0.05 mm Hg in 10,079 individuals covering both the normal BP interval and the hypertensive BP interval. In addition to lack of general evidence for sodium reduction, the evidence selected by advocates for sodium reduction is questionable,30 as exemplified by the Trials of Hypertension Prevention, which developed from a correct intention to treat analysis of two RCTs, which showed no effect of sodium reduction on mortality (P = 0.52),31 to a statistically elaborated observational study in which all participants on reduced sodium were eliminated. The latter version showed a marginal “linear” effect between sodium intake and CVD events (P = 0.044).32 Twenty-three percent of events were not accounted for and the statistical weakness of this analysis is emphasized by the fact that the addition of a few events to the lowest sodium group would eliminate the significance. Data on ACM were not shown. The hypothesis that sodium reduction, by lowering BP, could improve health was raised 100 years ago, before any direct evidence linking sodium to health was available. Indeed, the first moderate GRADE evidence appeared in 1985.33 Since then, the robust body of empiric evidence supports the following conclusions: (i) there are no RCTs allocating individuals to below 2,300 mg and measuring health outcomes; (ii) RCTs allocating risk groups such as obese prehypertensive individuals and hypertensive individuals down to (but not below) 2,300 mg show no effect of sodium reduction on ACM; (iii) RCTs allocating individuals to below 2,300 mg show a minimal effect on BP in the healthy
Graudal
6 American Journal of Hypertension
23. Cobb LK, Anderson CA, Elliott P, Hu FB, Liu K, Neaton JD Whelton PK, Woodward M, Appel LJ. Methodological issues in Cohort studies that relate sodium intake to cardiovascular disease outcomes: a science advisory from the American Heart Association. Circulation 2014; 129:1173–1186. 24. MacMahon S, Peto R, Cutler J, Collins R, Sorlie P, Neaton J, Abbott R, Godwin J, Dyer A, Stamler J. Blood pressure, stroke, and coronary heart disease. Part 1, Prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet 1990; 335:765–774. 25. Mente A, O’Donnell MJ, Rangarajan S, McQueen MJ, Poirier P, Wielgosz A, Morrison H, Li W, Wang X, Di C, Mony P, Devanath A, Rosengren A, Oguz A, Zatonska K, Yusufali AH, Lopez-Jaramillo P, Avezum A, Ismail N, Lanas F, Puoane T, Diaz R, Kelishadi R, Iqbal R, Yusuf R, Chifamba J, Khatib R, Teo K, Yusuf S; PURE Investigators. Association of urinary sodium and potassium excretion with blood pressure. N Engl J Med 2014; 371:601–611. 26. Lewington S, Clarke R, Qizilbash N, Peto R, Collins R. Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet 2002; 360:1903–1913. 27. Mencken HL. There is a non-linear relationship between mortality and blood pressure. Eur Heart J 2000; 21:1635–1638. 28. Wiysonge CS, Bradley HA, Volmink J, Mayosi BM, Mbewu A, Opie LH. Beta-blockers for hypertension. Cochrane Database Syst Rev 2012; 11:CD002003. 29. Intersalt Cooperative Research Group. Intersalt: an international study of electrolyte excretion and blood pressure. Results for 24 hour urinary sodium and potassium excretion. BMJ 1988; 297:319–328. 30. Graudal N. Dietary sodium: where science and policy conflict: impact of the 2013 IOM Report on Sodium Intake in Populations. Curr Hypertens Rep 2015; 17:9. 31. Cook NR, Cutler JA, Obarzanek E, Buring JE, Rexrode KM, Kumanyika SK, Appel LJ, Whelton PK. Long term effects of dietary sodium reduction on cardiovascular disease outcomes: observational follow-up of the trials of hypertension prevention (TOHP). BMJ 2007; 334:885–888. 32. Cook NR, Appel LJ, Whelton PK. Lower levels of sodium intake and reduced cardiovascular risk. Circulation 2014; 129:981–989. 33. Kagan A, Popper JS, Rhoads GG, Yano K. Dietary and other risk factors for stroke in Hawaiian Japanese men. Stroke 1985; 16:390–396.
Downloaded from http://ajh.oxfordjournals.org/ at Main Library of Gazi University on February 12, 2016
12. O’Donnell M, Mente A, Rangarajan S, McQueen MJ, Wang X, Liu L, Yan H, Lee SF, Mony P, Devanath A, Rosengren A, Lopez-Jaramillo P, Diaz R, Avezum A, Lanas F, Yusoff K, Iqbal R, Ilow R, Mohammadifard N, Gulec S, Yusufali AH, Kruger L, Yusuf R, Chifamba J, Kabali C, Dagenais G, Lear SA, Teo K, Yusuf S; PURE Investigators. Urinary sodium and potassium excretion, mortality, and cardiovascular events. N Engl J Med 2014; 371:612–623. 13. Graudal N, Jürgens G, Baslund B, Alderman MH. Compared with usual sodium intake, low- and excessive-sodium diets are associated with increased mortality: a meta-analysis. Am J Hypertens 2014; 27:1129–1137. 14. Graudal N, Jürgens G. The blood pressure sensitivity to changes in sodium intake is similar in Asians, Blacks and Whites. An analysis of 92 randomized controlled trials. Front Physiol 2015; 6:157. 15. Graudal N, Hubeck-Graudal T, Jürgens G, McCarron DA. The significance of duration and amount of sodium reduction intervention in normotensive and hypertensive individuals: a meta-analysis. Adv Nutr 2015; 6:169–177. 16. Graudal NA, Hubeck-Graudal T, Jurgens G. Effects of low sodium diet versus high sodium diet on blood pressure, renin, aldosterone, catecholamines, cholesterol, and triglyceride. Cochrane Database Syst Rev 2011; 11:CD004022. 17. Wright JD, Hughes, JP, Ostchega Y, Yoon SS, Nwankwo T. Mean systolic and diastolic blood pressure in adults aged 18 and over in the United States, 2001–2008. Nat Health Stat Rep 2011; 35:1–24. 18. He FJ, MacGregor GA. Effect of longer-term modest salt reduction on blood pressure. Cochrane Database Syst Rev 2004; 4:CD004937. 19. Aburto NJ, Ziolkovska A, Hooper L, Elliott P, Cappuccio FP, Meerpohl JJ. Effect of lower sodium intake on health: systematic review and metaanalyses. BMJ 2013; 346:f1326. 20. Oliver WJ, Cohen EL, Neel JV. Blood pressure, sodium intake, and sodium related hormones in the Yanomamo Indians, a “no-salt” culture. Circulation 1975; 52:146–151. 21. Brunner HR, Laragh JH, Baer L, Newton MA, Goodwin FT, Krakoff LR, Bard RH, Bühler FR. Essential hypertension: renin and aldosterone, heart attack and stroke. N Engl J Med 1972; 286:441–449. 22. Gunn JP, Barron JL, Bowman BA, Merritt RK, Cogswell ME, Angell SY, Bauer UE, Frieden TR. Sodium reduction is a public health priority: reflections on the Institute of Medicine’s report, sodium intake in populations: assessment of evidence. Am J Hypertens 2013; 26:1178–1180.
Review
A Radical Sodium Reduction Policy Is Not Supported by Randomized Controlled Trials or Observational Studies: Grading the Evidence Niels Graudal1 triglyceride; and (iv) observational studies show that sodium intakes below 2,645 and above 4,945 mg are associated with increased mortality. Given that 90% of the worlds’ population currently consumes sodium within the optimal range of 2,645–4,945 mg, there is no scientific basis for a public health recommendation to alter sodium intake.
The 2013 IOM report on sodium intake in populations1 has focused attention on whether the widely recommended 2,300 mg upper level for sodium intake is beneficial. The sodium intake in 95% of the world’s populations is between 2,620 and 4,830 mg sodium (6.6–12.2 g salt/114–210 mmol).2,3 In spite of this, several health institutions recommend sodium intake be reduced to below 2,000 mg4 to 2,300 mg5,6—meaning that 6–7 billion individuals should alter their diet to accommodate. Such a radical recommendation should be based on solid evidence that this would be safe and beneficial. The highest grade of evidence (GRADE system “High”) comes from randomized controlled trials (RCTs) and the second grade (GRADE “Moderate”) comes from prospective observational studies.7 The present review focuses on GRADE “high” and “moderate” evidence of health outcomes related to sodium intake. The IOM emphasized that outcomes of intermediate variables, specifically effect on blood pressure (BP), are inadequate to assess the health effect of sodium. Nevertheless, because the recommendations that sodium should be reduced to less than 2,300 mg/d are based upon its BP consequences and the denial of side effects,4–6 this
review will also assess BP effect and potential side effects on hormones and lipids.
Correspondence: Niels Graudal ([email protected]). Initially submitted November 25, 2015; date of first revision December 18, 2015; accepted for publication December 22, 2015.
Keywords: blood pressure; catecholamines; cholesterol; hypertension; mortality; observational study; randomized controlled trial; renin– angiotensin–aldosterone system; sodium; triglyceride. doi:10.1093/ajh/hpw006
HIGH GRADE EVIDENCE ON HEALTH OUTCOMES: THE EFFECT OF SODIUM REDUCTION ON CARDIOVASCULAR DISEASE EVENTS AND ALL-CAUSE MORTALITY The 2014 Cochrane review
A recently updated meta-analysis of 8 RCTs with follow-up data on morbidity and mortality found no effect of reduced sodium on all-cause mortality (ACM) (RR = 0.96 (0.83, 1.10)) in 6,603 individuals and a nonsignificant trend favoring reduced cardiovascular disease (CVD) morbidity (RR = 0.76 (0.57, 1.01)) in 3,397 individuals.8 Five of the studies were performed in hypertensive individuals and three in prehypertensive, obese individuals. None were performed in healthy individuals. Furthermore, no study has reduced sodium to less than 2,300 mg. In conclusion, there is no high GRADE evidence to support sodium reduction below 2,300 mg in either the healthy population or even in populations at increased CVD risk.
1Department of Rheumatology VRR4242, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark.
© American Journal of Hypertension, Ltd 2016. All rights reserved. For Permissions, please email: [email protected]
American Journal of Hypertension 1
Downloaded from http://ajh.oxfordjournals.org/ at Main Library of Gazi University on February 12, 2016
Several health institutions recommend sodium intake be reduced to below 2,300 mg, which means that 6–7 billion individuals should alter their diet to accommodate. Such a radical recommendation should be based on solid evidence. However, this review reveals that (i) there are no randomized controlled trials (RCTs) allocating individuals to below 2,300 mg and measuring health outcomes; (ii) RCTs allocating risk groups such as obese prehypertensive individuals and hypertensive individuals down to (but not below) 2,300 mg show no effect of sodium reduction on all-cause mortality; (iii) RCTs allocating individuals to below 2,300 mg show a minimal effect on blood pressure in the healthy population (less than 1 mm Hg) and significant increases in renin, aldosterone, noradrenalin cholesterol, and
Graudal MODERATE GRADE EVIDENCE ON HEALTH OUTCOMES: THE ASSOCIATION OF LOW SODIUM INTAKE WITH CVD EVENTS AND ACM The 2013 IOM report
The 2014 meta-analysis, updated
The IOM committee did not perform a formal meta-analysis of the available data.1 Such an analysis was published a year after the IOM report.13 This analysis confirmed that “higher levels” of sodium intake are associated with increased mortality but found that “higher level” was actually greater than 4,945 mg. This analysis also confirmed a U-shaped association of sodium intake to subsequent mortality, with significantly higher mortality at sodium intake below 2,645 mg. Optimal health outcomes were found when intakes were 2,645–4,945 mg and within this range event rates were constant. Figure 1 shows the previous meta-analysis of the risk
HIGH GRADE EVIDENCE ON BP: EFFECT OF SODIUM REDUCTION BELOW 2,300 MG ON BP
The traditional theoretical background for the sodium reduction policy was the assumption that a reduced sodium intake will lower BP and would therefore inevitably prevent CVD mortality. The potential effect of sodium reduction on BP will depend on the duration of achieved maximal efficacy (i.e., if too short, it may not show the true effect), the dose– response relationship (if the magnitude of sodium reduction is too small, it may not show maximal efficacy), as well as the baseline BP (the effect of sodium reduction is likely to increase with increasing baseline BP). In addition, it has been shown that BP increases with age. Previously, ethnicity was believed to have an impact on the effect of sodium reduction on BP, but factors like baseline BP, baseline sodium intake, study duration, age, and magnitude of sodium reduction can explain the major part of the ethnic association with BP.14 Therefore, taking these factors into consideration, adjustment for ethnicity should not have a meaningful impact. RCTs with longitudinal BP measurements have shown that the maximal reduction of BP produced by limiting sodium intake was reached at 1 week.15 Furthermore, in individuals with a BP below 130/80 mm Hg and a sodium intake below 250 mmol (5,750 mg, approximately corresponding to the 99% upper limit of the world’s populations), there was no dose–response relationship between sodium intake and systolic BP or diastolic BP in studies, which randomized individuals to different doses. In contrast, there was
Figure 1. All-cause mortality, usual sodium vs. low sodium. Results from 9 population representative observational studies adjusted for multiple confounders. Exchanging the first analyses of the NHANES studies1,7 with the re-analyses2,8: HR = 0.87 (0.82, 0.91), P < 0.00001. Further exchanging primary analyses5,11,13 with analyses of low-risk populations6,12,14: HR = 0.86 (0.81, 0.92), P < 0.00001. Abbreviations: CI, confidence interval; HR, hazard ratio; LRP, low-risk population.
2 American Journal of Hypertension
Downloaded from http://ajh.oxfordjournals.org/ at Main Library of Gazi University on February 12, 2016
The 2013 IOM committee reviewed observational studies directly relating sodium intake with morbidity and mortality.1 The committee concluded that there was a positive relationship between “higher levels” of sodium intake and risk of CVD, but “higher levels” was not defined. Furthermore, the previously defined recommended upper level of 2,300 mg of sodium intake was neither confirmed nor rejected. Consequently, this report left more questions than answers. The reason why the report questioned the 2,300 mg limit was that several (but not all) of the observational studies had indicated an increased mortality not only at “higher levels” but also below 2,300 mg, thus indicating a U-shaped association between sodium intake and mortality.9,10 Since then 2 additional studies have independently confirmed the U-shaped association between sodium intake and CVD events and ACM.11,12
of ACM in individuals within the usual sodium intake range compared vs. a low sodium intake below 2,645 mg13 updated with 2 studies.11,12 The present updated analysis is confined to include population samples supposed to be representative for the general population, adjusted for multiple confounders.
Radical Sodium Reduction Policy
a dose–response relationship in individuals with a BP higher than 130/80 mm Hg and a sodium intake above 250 mmol.15 The 2011 Cochrane review
HIGH GRADE EVIDENCE ON HORMONES AND LIPIDS: EFFECT OF SODIUM REDUCTION ON RENIN, ALDOSTERONE, CATECHOLAMINES, CHOLESTEROL, AND TRIGLYCERIDE
Many health institutions, which have issued sodium recommendations,4–6 have chosen to ignore the conclusions of
DISCUSSION
It is remarkable that health institutions think that 95% of the world’s population eat too much salt and recommend that they should reduce their sodium intake to less than 2,300 mg/d. The commitment to the 2,300 mg/d upper limit promulgated in 20055 persists, despite (i) no empiric evidence of a health benefit associated with intakes less than 2,300 mg/d, (ii) failure of the 2013 IOM report to support that recommendation, and (iii) the substantial body of observational data that sodium, like every other essential nutrient, bears a “U” or “J” shaped relation to health outcomes. Indeed, our 2014 metaanalysis of 25 population studies found that a sodium intake
Table 1. Effect of sodium reduction on SBP SBP (mm Hg)
0–25%
N participants
Mean
Sodium
(studies)
difference (95% CI)
reduction (mmol)
274 (3)
−0.34 (−1.51, 0.83)
110
0.57
2,077 (29)
−0.41 (−1.12, 0.30)
156
0.26
5,748 (33)
−2.72 (−3.41, −2.03)
91
0.00001
5,108 (69)
−5.62 (−6.64,−4.60)
104
0.00001
P value
131 mm Hg Abbreviations: CI, confidence interval; SBP, systolic blood pressure.
American Journal of Hypertension 3
Downloaded from http://ajh.oxfordjournals.org/ at Main Library of Gazi University on February 12, 2016
In agreement with data from the longitudinal studies and the dose–response studies,15 the 2011 Cochrane review of 167 studies showed a minimal effect of sodium reduction on BP in individuals with a normal BP (1.27/0.05 mm Hg) and a more meaningful effect in hypertensive individuals (5.18/2.59 mm Hg).16 Table 1 (systolic BP) and Table 2 (diastolic BP) display these results, but in the subgroup of studies with duration of at least 1 week, a sodium intake below 250 mmol and subdivided according to the 25% quartiles of BP of the American population.17 This analysis shows that about 83% of all individuals investigated have a BP above the 50% percentile. Therefore, the RCTs are biased in favor of a higher BP effect of sodium reduction. Consequently, to evaluate the potential effect of sodium reduction, it is necessary to estimate the results in the BP subgroups which actually reflect the general population. The results show that individuals with a BP below the median (119/71) have minimal effects on systolic BP and diastolic BP (Tables 1 and 2). Moreover, the effect up to the 75% percentile is small (2.72/0.75 mm Hg) (Tables 1 and 2). Individuals with a BP in the upper 25% percentile (above 131/78 mm Hg) of the BP distribution have a measurable, but still rather moderate effect of sodium reduction on BP (5.68/2.67 mm Hg)—significantly less than the effect that can be achieved by antihypertensive drugs (usually 10/5 mm Hg or more). It is important to note that all these BP effects are obtained by sodium reductions of 90 mmol or more—corresponding to a 60% sodium reduction for the American population.
the 2013 IOM report. Further, they have chosen to ignore the coherent results of multiple observational studies involving about 500,000 participants11–13 and rely instead on the sodium effect on the single variable of BP, and particularly on studies lasting for at least 4 weeks18,19 although the maximal efficacy of sodium reduction on BP is achieved at 1 week.15,16 By excluding studies shorter than 4 weeks, these meta-analyses exclude almost all studies investigating the effect on surrogate variables other than BP, and therefore these analyses lacked the power to assess the impact of other relevant variables. Moreover, the assumption that other surrogate variables have only “acute” and “transitional” associations18 has not been demonstrated. On the contrary, the permanently salt-deprived Yanomamo Indians had a 10-fold increased urinary excretion of aldosterone.20 The 2011 Cochrane review found that sodium reduction resulted in highly significant increases in hormones and lipids.16 This review has been criticized for the inclusion of “extreme” studies. Table 3 shows a subanalysis of results from studies, which randomize participants to usual sodium in the interval of 2,300 – 5,200 mg or to below 2,300 mg, i.e., the intervention recommended by health authorities. In this analysis, studies with extremely large sodium reductions and also studies with sodium reductions to below 40 mmol have been excluded. Furthermore, studies with duration less than 7 days were excluded. All associations are statistically significant, save for adrenalin. Similar findings concerning renin and aldosterone have been shown in a cross-sectional study.21
Graudal Table 2. Effect of sodium reduction on diastolic blood pressure DBP (mm Hg)
0–25%
N participants
Mean
Sodium
(studies)
difference (95% CI)
reduction (mmol)
1,530 (12)
(P value)
0.20 (−0.49, 0.90)
135
0.57
−0.57 (−1.77, 0.64)
143
0.36
3,081 (17)
−0.75 (−1.47, −0.03)
120
0.04
7,705 (78)
−2.69 (−3.25, −2.14)
92
78 mm Hg
Radical Sodium Reduction Policy
population (less than 1 mm Hg) and a moderate effect in individuals with a high baseline BP (about 5/2.5 mm Hg); (iv) RCTs allocating individuals to below 2,300 mg show significant increases in renin, aldosterone, noradrenalin cholesterol, and triglyceride; and (v) observational studies show that sodium intakes below 2,645 and above 4,945 mg are associated with increased mortality. Given that 90% of the worlds’ population currently consumes sodium within the optimal range, there is no scientific basis for a public health recommendation to alter sodium intake. Thus, a solution to the controversy could be to accept the present evidence and revise the recommendations to be in accordance with the present usual intake (2,645–4,945 mg sodium). Another— maybe impossible—solution could be to perform an RCT, which followed large groups of individuals on either a usual sodium diet above the present recommendations (2,300 mg sodium) or a low-sodium diet in accordance with the recommendations (below 2,300 mg) for a considerable time period recording health events and mortality.
DISCLOSURE
The author declared no conflict of interest.
REFERENCES 1. Institute of Medicine. Sodium Intake in Populations: Assessment of Evidence. National Academies Press: Washington, DC, 2013. 2. McCarron DA, Kazaks AG, Geerling JC, Stern JS, Graudal NA. Normal range of human dietary sodium intake: a perspective based on 24-hour urinary sodium excretion worldwide. Am J Hypertens 2013; 26:1218–1223. 3. Powles J, Fahimi S, Micha R, Khatibzadeh S, Shi P, Ezzati M, Engell RE, Lim SS, Danaei G, Mozaffarian D; Global Burden of Diseases Nutrition and Chronic Diseases Expert Group (NutriCoDE). Global, regional and national sodium intakes in 1990 and 2010: a systematic analysis of 24 h urinary sodium excretion and dietary surveys worldwide. BMJ Open 2013; 3:e003733. 4. WHO. Sodium Intake for Adults and Children. WHO: Geneva, Switzerland, 2012. 5. Institute of Medicine. Dietary Reference Intakes: Water, Potassium, Sodium, Chloride, and Sulfate. National Academies Press: Washington, DC, 2005. 6. Dietary Guidelines for Americans. health.gov 2010. 7. Higgins JPT, Green S (eds). Cochrane Handbook for Systematic Reviews of Interventions. Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011 2014. 8. Adler AJ, Taylor F, Martin N, Gottlieb S, Taylor RS, Ebrahim S. Reduced dietary salt for the prevention of cardiovascular disease. Cochrane Database Syst Rev 2014; 12:CD009217. 9. Thomas MC, Moran J, Forsblom C, Harjutsalo V, Thorn L, Ahola A, Wadén J, Tolonen N, Saraheimo M, Gordin D, Groop PH; FinnDiane Study Group. The association between dietary sodium intake, ESRD, and all-cause mortality in patients with type 1 diabetes. Diabetes Care 2011; 34:861–866. 10. O’Donnell MJ, Yusuf S, Mente A, Gao P, Mann JF, Teo K, McQueen M, Sleight P, Sharma AM, Dans A, Probstfield J, Schmieder RE. Urinary sodium and potassium excretion and risk of cardiovascular events. JAMA 2011; 306:2229–2238. 11. Pfister R, Michels G, Sharp SJ, Luben R, Wareham NJ, Khaw KT. Estimated urinary sodium excretion and risk of heart failure in men and women in the EPIC-Norfolk study. Eur J Heart Fail 2014; 16:394–402.
American Journal of Hypertension 5
Downloaded from http://ajh.oxfordjournals.org/ at Main Library of Gazi University on February 12, 2016
The present review shows that evidence in favor of a sodium intake above 2,300 mg is more compelling than the evidence in favor of a sodium intake below 2,300 mg. The only minimal evidence in favor of sodium reduction below 2,300 mg, namely the very small BP effect in healthy individuals, depend on the linear association between BP and mortality, which is derived from observational studies.26 Interestingly, the health institutions4–6,22,23 accept the outcome of observational studies linking BP with mortality,26 but ignore observational studies, which link sodium intake with mortality.9–13 An explanation could be that also RCTs have shown that lowering BP reduces mortality, but this is shown to be the case only in hypertensive individuals, not in normotensive individuals. Therefore, this inconsistency is remarkable, especially as the BP–mortality link is equivocal. First, the linear association between BP and mortality has been questioned, especially in the low end of the BP interval.27 Second, in the PURE study, a low sodium intake was associated with an increased mortality,12 although it was associated with a low BP,25 and in PURE, the healthiest people, without hypertension, had the greatest risk from low sodium. Third, although beta-blockers reduce BP in hypertensive individuals, they do not reduce mortality.28 This is probably due to side effects. The present review indicates that particularly side effects may explain the inverse association of a low sodium intake with mortality recorded in the observational studies.13 We have not systematically reviewed the association of a sodium intake with BP recorded in observational studies. However, 2 of the largest and most important of these studies25,29 confirm the small BP effects found in RCTs. The systolic BP effect in the PURE study25 was 1.3/0.58 mm Hg/1 g Na in individuals with a normal BP and 2.49/0.91 mm Hg/1 g Na in individuals with hypertension. Intersalt29 found a BP effect of 1.6/0.05 mm Hg in 10,079 individuals covering both the normal BP interval and the hypertensive BP interval. In addition to lack of general evidence for sodium reduction, the evidence selected by advocates for sodium reduction is questionable,30 as exemplified by the Trials of Hypertension Prevention, which developed from a correct intention to treat analysis of two RCTs, which showed no effect of sodium reduction on mortality (P = 0.52),31 to a statistically elaborated observational study in which all participants on reduced sodium were eliminated. The latter version showed a marginal “linear” effect between sodium intake and CVD events (P = 0.044).32 Twenty-three percent of events were not accounted for and the statistical weakness of this analysis is emphasized by the fact that the addition of a few events to the lowest sodium group would eliminate the significance. Data on ACM were not shown. The hypothesis that sodium reduction, by lowering BP, could improve health was raised 100 years ago, before any direct evidence linking sodium to health was available. Indeed, the first moderate GRADE evidence appeared in 1985.33 Since then, the robust body of empiric evidence supports the following conclusions: (i) there are no RCTs allocating individuals to below 2,300 mg and measuring health outcomes; (ii) RCTs allocating risk groups such as obese prehypertensive individuals and hypertensive individuals down to (but not below) 2,300 mg show no effect of sodium reduction on ACM; (iii) RCTs allocating individuals to below 2,300 mg show a minimal effect on BP in the healthy
Graudal
6 American Journal of Hypertension
23. Cobb LK, Anderson CA, Elliott P, Hu FB, Liu K, Neaton JD Whelton PK, Woodward M, Appel LJ. Methodological issues in Cohort studies that relate sodium intake to cardiovascular disease outcomes: a science advisory from the American Heart Association. Circulation 2014; 129:1173–1186. 24. MacMahon S, Peto R, Cutler J, Collins R, Sorlie P, Neaton J, Abbott R, Godwin J, Dyer A, Stamler J. Blood pressure, stroke, and coronary heart disease. Part 1, Prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet 1990; 335:765–774. 25. Mente A, O’Donnell MJ, Rangarajan S, McQueen MJ, Poirier P, Wielgosz A, Morrison H, Li W, Wang X, Di C, Mony P, Devanath A, Rosengren A, Oguz A, Zatonska K, Yusufali AH, Lopez-Jaramillo P, Avezum A, Ismail N, Lanas F, Puoane T, Diaz R, Kelishadi R, Iqbal R, Yusuf R, Chifamba J, Khatib R, Teo K, Yusuf S; PURE Investigators. Association of urinary sodium and potassium excretion with blood pressure. N Engl J Med 2014; 371:601–611. 26. Lewington S, Clarke R, Qizilbash N, Peto R, Collins R. Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet 2002; 360:1903–1913. 27. Mencken HL. There is a non-linear relationship between mortality and blood pressure. Eur Heart J 2000; 21:1635–1638. 28. Wiysonge CS, Bradley HA, Volmink J, Mayosi BM, Mbewu A, Opie LH. Beta-blockers for hypertension. Cochrane Database Syst Rev 2012; 11:CD002003. 29. Intersalt Cooperative Research Group. Intersalt: an international study of electrolyte excretion and blood pressure. Results for 24 hour urinary sodium and potassium excretion. BMJ 1988; 297:319–328. 30. Graudal N. Dietary sodium: where science and policy conflict: impact of the 2013 IOM Report on Sodium Intake in Populations. Curr Hypertens Rep 2015; 17:9. 31. Cook NR, Cutler JA, Obarzanek E, Buring JE, Rexrode KM, Kumanyika SK, Appel LJ, Whelton PK. Long term effects of dietary sodium reduction on cardiovascular disease outcomes: observational follow-up of the trials of hypertension prevention (TOHP). BMJ 2007; 334:885–888. 32. Cook NR, Appel LJ, Whelton PK. Lower levels of sodium intake and reduced cardiovascular risk. Circulation 2014; 129:981–989. 33. Kagan A, Popper JS, Rhoads GG, Yano K. Dietary and other risk factors for stroke in Hawaiian Japanese men. Stroke 1985; 16:390–396.
Downloaded from http://ajh.oxfordjournals.org/ at Main Library of Gazi University on February 12, 2016
12. O’Donnell M, Mente A, Rangarajan S, McQueen MJ, Wang X, Liu L, Yan H, Lee SF, Mony P, Devanath A, Rosengren A, Lopez-Jaramillo P, Diaz R, Avezum A, Lanas F, Yusoff K, Iqbal R, Ilow R, Mohammadifard N, Gulec S, Yusufali AH, Kruger L, Yusuf R, Chifamba J, Kabali C, Dagenais G, Lear SA, Teo K, Yusuf S; PURE Investigators. Urinary sodium and potassium excretion, mortality, and cardiovascular events. N Engl J Med 2014; 371:612–623. 13. Graudal N, Jürgens G, Baslund B, Alderman MH. Compared with usual sodium intake, low- and excessive-sodium diets are associated with increased mortality: a meta-analysis. Am J Hypertens 2014; 27:1129–1137. 14. Graudal N, Jürgens G. The blood pressure sensitivity to changes in sodium intake is similar in Asians, Blacks and Whites. An analysis of 92 randomized controlled trials. Front Physiol 2015; 6:157. 15. Graudal N, Hubeck-Graudal T, Jürgens G, McCarron DA. The significance of duration and amount of sodium reduction intervention in normotensive and hypertensive individuals: a meta-analysis. Adv Nutr 2015; 6:169–177. 16. Graudal NA, Hubeck-Graudal T, Jurgens G. Effects of low sodium diet versus high sodium diet on blood pressure, renin, aldosterone, catecholamines, cholesterol, and triglyceride. Cochrane Database Syst Rev 2011; 11:CD004022. 17. Wright JD, Hughes, JP, Ostchega Y, Yoon SS, Nwankwo T. Mean systolic and diastolic blood pressure in adults aged 18 and over in the United States, 2001–2008. Nat Health Stat Rep 2011; 35:1–24. 18. He FJ, MacGregor GA. Effect of longer-term modest salt reduction on blood pressure. Cochrane Database Syst Rev 2004; 4:CD004937. 19. Aburto NJ, Ziolkovska A, Hooper L, Elliott P, Cappuccio FP, Meerpohl JJ. Effect of lower sodium intake on health: systematic review and metaanalyses. BMJ 2013; 346:f1326. 20. Oliver WJ, Cohen EL, Neel JV. Blood pressure, sodium intake, and sodium related hormones in the Yanomamo Indians, a “no-salt” culture. Circulation 1975; 52:146–151. 21. Brunner HR, Laragh JH, Baer L, Newton MA, Goodwin FT, Krakoff LR, Bard RH, Bühler FR. Essential hypertension: renin and aldosterone, heart attack and stroke. N Engl J Med 1972; 286:441–449. 22. Gunn JP, Barron JL, Bowman BA, Merritt RK, Cogswell ME, Angell SY, Bauer UE, Frieden TR. Sodium reduction is a public health priority: reflections on the Institute of Medicine’s report, sodium intake in populations: assessment of evidence. Am J Hypertens 2013; 26:1178–1180.
