ARTICLE

TOP ARTICLE REFERENCES

A NUMBER OF HEALTH AGENCIES in the United States and abroad recommend a reduced NaCl intake for the general population (e.g., National Academy of Sciences, United States Surgeon General; National Heart, Lung, and Blood Institute; United States Departments of Agriculture and of Health and Human Services; World Health Organization; and American Heart Association). The primary rationale for this recommendation is to reduce the overall blood pressure level within the population and hence to reduce the overall incidence of cardiovascular disease. However, there are some vocal detractors from this recommendation. A recent feature article in Science, entitled "The (political) science of salt," called attention to the differing opinions concerning the recommendation for a reduced NaCl intake in the general population (23). Arguments against this recommendation include 1) a modest effect on blood pressure; 2) lack of a demonstrated impact on cardiovascular disease morbidity and mortality; and 3) concern about potential adverse health consequences of a lowered NaCl intake. The Science article suggests that the continuing "salt controversy" is being perpetuated by vested interests of the food industry and research funding agencies. The purpose of this report is to provide a brief overview of the salt-blood pressure relationship.

A high-NaCl diet convincingly contributes to elevated arterial pressure in a number of animal models of genetic and acquired hypertension. The chimpanzee is phylogenetically close to the human, and in a carefully controlled study, addition of NaCl to the usual diet of the chimpanzee over 20 mo resulted in a 33-mmHg and a 10-mmHg elevation of systolic and diastolic blood pressure, respectively (8). This increase was completely reversed within 6 mo of withdrawing the high NaCl intake. Animal studies, as well as limited clinical observations, also suggest that a high NaCl intake, independent of its effect on blood pressure, may contribute to cerebral arterial disease and stroke, left ventricular hypertrophy, and glomerular injury (16).

In humans, evidence for an association between NaCl consumption and blood pressure is based on both observational studies and intervention trials. Across populations, the level of blood pressure, the incremental rise in blood pressure with age, and the prevalence of hypertension are related to NaCl intake. The International Study of Salt and Blood Pressure (INTERSALT) is a cross-sectional study designed to evaluate both within-population and cross-population hypotheses on the relationship between blood pressure and sodium excretion in >10,000 adults at 52 centers around the world (21). The principal observations in INTERSALT are 1) among individuals, a difference of 100 meq/day in sodium intake (equivalent to 5.9 g NaCl) is associated on average with a difference of 3-6 mmHg in systolic blood pressure; and 2) among populations, a 100 meq/day lower sodium intake is associated with the attenuation of the rise in systolic blood pressure by 10 mmHg between the ages of 25 and 55 yr. Similar to observations in other isolated, preliterate populations, in four remote INTERSALT samples with low NaCl intakes, blood pressure levels were low, there was little or no hypertension, and blood pressure did not increase with age.

Randomized, controlled trials provide compelling evidence for a causal relationship between NaCl intake and blood pressure. Despite reservations about the limitations of meta-analyses, two recent meta-analyses document consistent reductions of blood pressure in response to generally short-term lowered intakes of NaCl (7, 17). In one meta-analysis of 32 trials, overall reductions of systolic and diastolic blood pressure were 1.9/1.1 mmHg in nonhypertensive subjects and 4.8/2.5 mmHg in hypertensive subjects (7). In a second meta-analysis of 52 trials, overall reductions of blood pressure were 1.6/0.5 mmHg in nonhypertensive subjects and 5.9/3.8 mmHg in hypertensive subjects (17). Greater reductions of blood pressure were observed in hypertensive than in nonhypertensive subjects and in trials lasting >5 wk than in shorter trials. Overall, 24% of the adult US population have hypertension (5).

The Trials of Hypertension Prevention (TOHP) is a longitudinal study that evaluated the efficacy of reduction of dietary NaCl and of weight loss on blood pressure in a cohort of moderately overweight adults with high-normal blood pressure. In phase I, systolic and diastolic blood pressures were significantly reduced in separate groups treated with a lowered NaCl intake over 18 mo or with weight loss (24). Phase II of TOHP more extensively evaluated the effects of reduced NaCl intake and weight loss, alone and in combination, on blood pressure over a 3- to 4-yr period in overweight adults with high-normal blood pressures (25). Compared with blood pressures in a usual care control group, at 6 mo systolic and diastolic blood pressures were significantly reduced by a lowered NaCl intake alone (5.1/4.4 mmHg) and by weight loss alone (6.0/5.5 mmHg), although the effects of the two interventions were not additive. Beyond 6 mo, the interventions were less effective for maintaining both lowered NaCl intake and weight loss, and the impact of these interventions on blood pressure was lessened.

The prevalence of hypertension increases with age, and among Americans aged 60 yr or greater the majority have hypertension. In a recently completed randomized trial, reduced NaCl intake and weight loss have recently been shown to be effective and safe approaches to treating hypertension in older people (27). A limited number of observational and intervention studies suggests that there is also an association between NaCl consumption and blood pressure level in children and adolescents (20).

In several animal models of NaCl-sensitive hypertension, selective dietary sodium loading provided with anions other than chloride fails to produce hypertension (4). Similarly, selective chloride loading without sodium fails to produce hypertension. Limited evidence suggests that salt-induced blood pressure increases in the human also depend on high intakes of both sodium and chloride. Thus, in experimental models and in the human, the full expression of salt-induced hypertension requires the concomitant provision of high dietary intakes of both sodium and chloride. These studies highlight the importance of the expansion of the extracellular fluid volume for the development of NaCl-sensitive hypertension, because extracellular fluid volume is expanded by dietary NaCl but not by nonchloride salts of sodium. Furthermore, chloride itself may act as a direct renal vasoconstrictor. Although most sodium in the usual diet is consumed as NaCl, studies with selective sodium loading and selective chloride loading should provide additional information about mechanisms by which dietary NaCl increases arterial pressure.

Among individuals there is considerable variability of blood pressure responsiveness to NaCl intake, and salt sensitivity of blood pressure should be considered a quantitative rather than a qualitative trait. On the basis of acute NaCl depletion and/or loading protocols, depending on arbitrary definitions of NaCl sensitivity, it has been estimated that ~30-50% of hypertensive persons and a smaller percentage of nonhypertensive persons are salt sensitive, i.e., arterial pressure is decreased by NaCl depletion and/or increased by NaCl loading (26). Although these designations of salt sensitivity may not reflect long-term blood pressure responses to dietary NaCl, these acute studies highlight the heterogeneity of the response to acute changes in NaCl balance.

In part, different blood pressure responses to dietary NaCl among individuals may have a genetic basis. Heritability of salt sensitivity and salt resistance of blood pressure is most convincingly documented in animal models. Family studies, including studies of twins, suggest that there is a heritable contribution of salt sensitivity of blood pressure in humans, and there is limited evidence for heritability of NaCl excretion and levels of hormones that regulate NaCl excretion (10, 18). In a number of relatively rare disorders, specific genetic alleles resulting in sodium retention and hypertension have been described, e.g., glucocorticoid remediable primary aldosteronism, Liddle's syndrome, and the syndrome of apparent mineralocorticoid excess (15). In these disorders, hypertension is the consequence of alterations of either adrenal steroid metabolism or direct renal tubular function resulting in antinatriuresis. Conversely, specific alleles have been identified that result in alterations of renal tubular function that promote natriuresis and consequently relatively low blood pressure levels. What relevance, if any, these polymorphisms may have to salt sensitivity of blood pressure in the general population remains to be determined.

In the TOHP trial, an attempt was made to determine whether blood pressure responsiveness to a reduced NaCl intake is linked to angiotensinogen genotype (11). The results highlight the limited contribution of a single polymorphism of the angiotensinogen gene to blood pressure responses to NaCl reduction and to weight loss. After a follow-up of over 36 mo, the reduction of blood pressure in response to a lowered NaCl intake and weight loss was greater in persons with the AA angiotensinogen genotype compared with those with the GG genotype. Blood pressure response in persons with the AG genotype was intermediate. Although different blood pressure responses to NaCl reduction and weight loss were observed by angiotensinogen genotype at 36 mo, there were no differences at 6 or 18 mo. Furthermore, although angiotensinogen genotype was associated with blood pressure responses to NaCl reduction alone and to weight loss alone, it was not associated with blood pressure responses to the combined NaCl reduction-weight loss intervention. The results would have been more credible if they had been consistent over time and if an effect of angiotensinogen genotype had also been observed in the combined intervention group. Nevertheless, the results raise the possibility that angiotensinogen genotype may have a modest influence on blood pressure responses to NaCl reduction and to weight loss. It is likely that additional genetic markers of salt sensitivity will be identified. Similar to the blood pressure level itself, in any individual the magnitude of the effect of dietary NaCl on blood pressure may reflect the culmination of a variable number of genetic polymorphisms.

Blood pressure responses to NaCl may also be modified by other components of the diet. Low dietary intakes of potassium or calcium augment NaCl-induced increases of blood pressure. Conversely, in several animal models, high dietary intakes of potassium or calcium attenuate NaCl-induced hypertension (14). The urine sodium-to-potassium ratio is a stronger correlate of blood pressure than either sodium or potassium alone. Additionally, simple carbohydrates potentiate NaCl sensitivity of blood pressure in normotensive Sprague-Dawley rats and in several rat models of hypertension (13).

It is appropriate to ask whether reduction of NaCl intake has any deleterious health consequences. In experimental animals, profound NaCl deprivation may stunt growth, may increase susceptibility to hemorrhage and renal damage, and may actually increase blood pressure (9, 19). Furthermore, in humans, intense acute NaCl depletion may have adverse consequences on plasma lipids and clotting factors. However, these observations are not relevant to the recommendation for a modest, long-term reduction of NaCl intake for the general population. Alderman et al. (1, 2) have recently published two papers which they believe show that habitual reduction of NaCl increases the risk of myocardial infarction in humans. These papers have been criticized on the basis of the inadequate assessment of habitual NaCl intake and concerns about confounding variables, including the fact that older individuals with cardiovascular disease were more likely to report consuming a lower NaCl intake. De Wardener (9) has recently summarized evidence for what he refers to as the "myth" that NaCl reduction increases cardiovascular disease risk, and he concludes: "Ruthless repetition of the myth, sometimes with the help of scientists who have not examined the data critically, provides splendid propaganda for those organizations whose commercial interests in salt outweigh their concern for the public good." In part prompted by the Alderman articles, the National Heart, Lung, and Blood Institute recently (Jan 28-29, 1999) convened a "Sodium and Blood Pressure" workshop to focus on the scientific evidence for the relationship of NaCl consumption to both blood pressure and cardiovascular disease risk. A summary of this workshop is currently being prepared for publication.

Unfortunately, a prospective clinical trial to evaluate the impact of a reduced NaCl intake on cardiovascular disease morbidity and mortality is probably not feasible. Any effort to mount a salt-health trial in the general population with specific cardiovascular end points would likely face formidable and insurmountable administrative, study design, and financial obstacles. Consequently, a recommendation for reduced NaCl consumption is based on the well-documented relationship between blood pressure level and the risk of cardiovascular disease. Blood pressure-associated risk for cardiovascular end points increases incrementally over a wide range of blood pressure levels, even among nonhypertensive persons (22). On a population basis, it has been estimated that a reduction in diastolic blood pressure of 2 mmHg would result in a 17% decrease in the prevalence of hypertension, a 15% reduction in risk of stroke and transient ischemic attacks, and a 6% reduction in risk of coronary heart disease (6). In Finland, between 1972 and 1992, there was a population-wide decrease in the dietary sodium-to-potassium ratio attributed to a decrease in the use of NaCl and replacement of common salt with a sodium-reduced, potassium-enriched salt and legislated upper limits of NaCl content for certain processed foods (12). During this time, there has been a 10-mmHg decrease in the population average diastolic blood pressure and a 60% decrease in deaths from both stroke and ischemic heart disease among 30- to 59-yr-old men and women.

Nutrients other than NaCl may also affect blood pressure levels. Observational studies document inverse associations of blood pressure with dietary potassium, calcium, and magnesium consumption (14). Oral potassium supplements lower blood pressure, particularly in hypertensive persons and in persons consuming a diet high in NaCl. Calcium supplementation preferentially lowers blood pressure in persons with NaCl-sensitive hypertension. In a randomized, multicenter study, the Dietary Approaches to Stop Hypertension (DASH) trial evaluated the effects on blood pressure of three dietary patterns over 8 wk in adults with high-normal blood pressure or mild hypertension (3). The dietary interventions were 1) a control diet with potassium, calcium, and magnesium levels close to the 25th percentile of US consumption; 2) a diet rich in fruits and vegetables; and 3) a "combination" diet rich in fruits, vegetables, and fat-free or low-fat dairy products. NaCl content was equivalent in all three diets (7.5 g/day). Systolic and diastolic blood pressure were significantly reduced by the diet enriched with fruits and vegetables (2.3/1.1 mmHg), and compared with the control diet these pressures were reduced to an even greater extent by the combination diet (5.5/3.0 mmHg). Although not designed to identify the effective nutrients of the diets, the DASH trial convincingly reaffirms the importance of multiple factors in the diet for blood pressure control. A follow-up trial, currently in progress, is evaluating the effects of different levels of NaCl intake on blood pressure responses to the DASH diets.

In summary, observational studies and interventional trials document a modest but consistent effect of NaCl consumption on blood pressure. The "salt controversy" is primarily related to the translation of these data into public policy. In the absence of a salt-health trial with definitive cardiovascular end points, a population-based recommendation for reduced NaCl consumption is justifiable and prudent. The observed effects of a reduced NaCl intake on blood pressure level within an overall population would be expected to translate into a lowered prevalence of cardiovascular disease. There is no evidence that limiting NaCl consumption to 6 g/day, as recommended by a number of health agencies, poses any health risk. A lower NaCl intake may be recommended for hypertensive patients. Both a genetic susceptibility as well as other components of the diet impact on the NaCl-blood pressure relationship. In the future, identification of specific genetic markers for "salt sensitivity" might permit more targeted strategies for the identification of those nonhypertensive individuals who would derive the most benefit from a reduced NaCl intake. Furthermore, blood pressure is also affected by other foods, and a reduced NaCl intake should be only one component of a nutritional strategy to lower blood pressure.