Is Sodium Restriction Important to Hypertension?

Cheryl L. Laffer, MD, PhD; Fernando Elijovich, MD

In This Article

The Argument For

The data showing that sodium is an important contributor to hypertension derive from a variety of studies. Epidemiologic studies have shown that sodium intake is directly related to cardiovascular mortality and to the increase in blood pressure with aging. Guidelines recommending sodium reduction for persons with hypertension were originally based on literature demonstrating a decrease in blood pressure with reduction in sodium intake. However, there have now been multiple large treatment and outcome trials demonstrating that a combination of sodium reduction and other nutritional/hygienic interventions is as effective as single-agent pharmacologic treatment of hypertension. Long-term studies of salt-sensitivity of blood pressure have demonstrated its adverse prognostic implications in both hypertensive and normotensive populations. These results support the contention that sodium restriction is a mainstay in the treatment of hypertension.

What is the relationship between sodium intake and mortality? The INTERnational study of SALT and blood pressure (INTERSALT) was a multinational study of more than 10,000 persons in 52 centers in 32 countries. Urine sodium excretion (UNaV), blood pressure (BP), and body mass index (BMI) were measured. Using only the data for the 12 European countries, Perry and Beevers[1] plotted each country's stroke mortality against mean UNaV, which reflects average sodium intake. The data were normalized for age, and the relationship was direct and significant, independent of BP and BMI. In the United States, the National Health and Nutrition Examination Survey (NHANES) has conducted periodic examinations in representative communities since 1971. The NHANES I follow-up study examined mortality in 9500 people who were aged 25-74 years at the time of the original survey. Sodium and energy intakes were estimated from dietary recall using food models and the frequency of adding salt; from these, the ratio of sodium to energy at the baseline visit was derived. In the non-overweight subjects, those with a higher ratio were older and had higher blood pressure and prevalence of hypertension or diabetes at baseline. Subsequently, He et al.[2] looked only at the overweight subjects, in whom these characteristics, as well as BMI, did not vary across the range of sodium to energy ratios, and who were without a history of coronary disease or stroke at baseline. After adjusting for age, sex, race, systolic BP, cholesterol level, and BMI, there was a direct relationship between increasing quartiles of the sodium to energy ratio and cumulative mortality. The relationship was highly significant for all-cause mortality as well as mortality from stroke and cardiovascular disease (CVD), although not mortality from coronary heart disease (CHD). All four mortality rates directly correlated with sodium to energy ratio in the group as a whole. Similar, although not statistically significant, findings were seen in the Multiple Risk Factor Intervention Trial (MRFIT),[3] in which mortality rates from CHD and CVD were greater in the highest quintile of sodium intake estimated from annual diet recalls compared with the lowest. The relationship was stronger when the results were adjusted for total energy intake.

These large multicenter studies appear to conflict with the results of Alderman et al.,[4] who followed 2937 participants in a work site program in Bronx, NY. Subjects collected urine 3-4 weeks after stopping antihypertensive therapy and after 4-5 days of avoiding sodium. They were then followed for an average of 3.5 years. Those in the lowest quartile of UNaV had the highest rate of cardiovascular events. However, they also had slightly but significantly higher BPs during the follow-up despite treatment, were thinner, and had lower potassium excretion. Furthermore, the relationship was found only in men. In contrast, a positive relationship between sodium intake and mortality was definitively demonstrated by a Finnish study.[5] In that community survey, UNaV was measured in 2436 subjects at baseline. Using causes of death from the national health registry, the authors demonstrated a direct relationship between increasing sodium intake and all-cause/CHD/CVD mortality, as well as with coronary events. The implication of these studies is that in the Western world, higher sodium intake is associated with increased mortality, particularly from cardiovascular causes.

A variety of population-based surveys have shown an increase in BP with increasing sodium intake. INTERSALT was the largest such study, and it also demonstrated a direct relationship between average sodium intake in a population and the degree to which BP increases with aging.[6] The increase was seen in all societies except nonindustrialized ones such as indigenous people of the Amazon rain forest, who eat <1 mEq sodium per day, have very low energy intake, and die young, largely from noncardiovascular causes. Hence, at least in the industrialized world, there seems to be a direct relationship between sodium intake and increasing BP with age.

But is there any evidence that lowering sodium intake decreases BP? Meta-analyses[7] of randomized trials of sodium restriction in subjects with hypertension predict BP decreases of 3-6 mm Hg systolic and 1-3 mm Hg diastolic for a decrease of 100 mmol in sodium intake, but this decrease in consumption is difficult to achieve. In the Dietary Approaches to Stop Hypertension (DASH) trial,[8] subjects were given either a typical US diet or the DASH diet. In persons with hypertension, this intervention lowered BP by 11/6 mm Hg. The DASH-Sodium trial further randomized subjects to one of three targets: 150 mmol, 100 mmol, or 50 mmol sodium, and actually achieved 142 mmol, 107 mmol, and 65 mmol, respectively. The lowest sodium intake, compared with the highest, significantly reduced BP on both diets, although the effect was greater on the typical US diet. DASH-Sodium reduced BP regardless of race and sex and had the biggest effect in subgroups with hypertension.

The final question, and the most controversial, is whether decreasing sodium intake improves outcomes. From epidemiologic data, a reduction of 2 mm Hg in systolic BP would be expected to reduce mortalities from stoke by 6%, CHD by 4%, and all causes by 3%.[9] To date, there have only been two trials that have provided some evidence for such benefit. The Treatment of Mild Hypertension Study (TOMHS)[10] included 900 adults aged 45-70 years with diastolic BP <100 mm Hg. All received "nutritional-hygienic" advice to reduce weight and intake of alcohol and sodium (goal <70 mmol) and to increase physical activity. They were then randomized to placebo or one of five antihypertensive monotherapies and followed for 4 years. The placebo group sustained a reduction of BP of 9/9 mm Hg, whereas the treated groups sustained a reduction of 16/12 mm Hg. This difference was sufficient to decrease the combined end point of major and other cardiovascular outcomes (i.e., hospitalization for transient ischemic attack, angina or claudication, and peripheral arterial disease) in the treated vs. the placebo groups (11% vs. 16%; p<0.05). However, there was no significant difference in the primary outcome of cumulative rates of death or nonfatal cardiovascular events between the groups (5.1% vs. 7.3%; p=0.21). The Trial of Nonpharmacologic interventions in the Elderly (TONE)[11] studied 975 persons aged 60-80 years with BP <145/85 mm Hg on monotherapy. They were randomized to reduction in sodium (to 80 mmol), weight loss (if obese), both, or usual care, and then drug therapy was withdrawn after 3 months. The primary end point was a combination of the recurrence of high BP, need for antihypertensive treatment, or a cardiovascular event. The result was that persons randomized to sodium reduction alone were significantly more likely to be free of the end point after 30 months than those who received usual care. Further, sodium reduction was as effective in reducing the primary outcome as was weight loss in the obese subjects. In subjects with hypertension, reducing sodium intake both lowers BP and helps prevent cardiovascular events.

Salt sensitivity of BP is defined as an increase in a person's BP due to a sodium load, and most subjects with hypertension exhibit it. We have studied one of the factors that may be involved, the eicosanoid 20-hydroxyeicosatetraenoic acid (20HETE), which is known to be deficient in inbred, salt-sensitive strains of rats. We demonstrated an effect of sodium balance in subjects with hypertension on the urinary excretion of 20-HETE, which doubled from sodium-depleted to sodium-loaded states. While levels of 20-HETE did not differ between salt-sensitive and salt-resistant subjects with hypertension, there was a difference in the relationship between 20-HETE and natriuresis during sodium loading in these two groups.[12] In salt-resistant subjects there is a direct relationship between 20-HETE and UNaV, which is lacking in salt-sensitive subjects. Conversely, in salt-sensitive subjects there is a direct relationship between BP and UNaV, which is not found in salt-resistant subjects. It is as if a deficit in action of 20-HETE in salt-sensitive subjects makes their natriuresis pressure dependent. Furthermore, this deficit was worsened by obesity, which is associated with salt sensitivity of BP.

Whatever the mechanism, most subjects with hypertension are salt sensitive and should restrict their sodium intake to reduce BP and to possibly improve outcomes. Salt sensitivity of BP is related to obesity, which is also increasing in prevalence. Recommendations to lose excess weight have also been shown to improve control of hypertension. Lower BP reduces cardiovascular morbidity and mortality; hence, the public health advice to restrict sodium intake and avoid weight gain can be applied to society as a whole.

---Cheryl L. Laffer, MD, PhD

  1. Perry LJ, Beevers DG. Salt intake and stroke: a possible direct effect. J Hum Hypertens. 1992;6:23-25.

  2. He J, Ogden LG, Vupputuri S, et al. Dietary sodium intake and subsequent risk of cardiovascular disease in overweight adults. JAMA. 1999;282:2027-2034.

  3. Stamler J, Cohen J, Cutler JA, et al. Sodium intake and mortality from myocardial infarction: multiple risk factor intervention trial (MRFIT). Can J Cardiol. 1997;17:272B.

  4. Alderman MH, Madhavan S, Cohen H, et al. Low urinary sodium is associated with greater risk of myocardial infarction among treated hypertensive men. Hypertension. 1995;25:1144-1152.

  5. Tuomilehto J, Jousilahti P, Rastenyte D, et al. Urinary sodium excretion and cardiovascular mortality in Finland: a prospective study. Lancet. 2001;357:848-851.

  6. Stamler J, Rose G, Elliott P, et al. Findings of the international cooperative INTERSALT study. Hypertension. 1991;17:I9-I15.

  7. Cutler JA, Follman D, Allender PS. Randomized trials of sodium reduction: an overview. Am J Clin Nutr. 1997;65(suppl 2):643S-651S.

  8. Sacks FM, Svetkey LP, Vollmer LM, et al. Effects on blood pressure of reduced dietary sodium and the dietary approaches to stop hypertension (DASH) diet. N Engl J Med. 2001;344:3-10.

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

  10. Neaton JD, Grimm RH, Prineas RJ, et al. Treatment of Mild Hypertension Study. Final results. Treatment of Mild Hypertension Study Research Group. JAMA. 1993;270:713-724.

  11. 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 nonpharmacologic interventions in the elderly (TONE). JAMA. 1998;279:839-846.

  12. Laffer CL, Laniado-Schwartzman M, Wang MH, et al. Differential regulation of natriuresis by 20-hydroxyeicosa-tetraenoic acid in human salt-sensitive versus salt-resistant hypertension. Circulation. 2003;107:574-578.