Dietary Sodium and Health: More Than Just Blood Pressure

Abstract

Sodium is essential for cellular homeostasis and physiological function. Excess dietary sodium has been linked to elevations in blood pressure (BP). Salt-sensitivity of BP varies widely, but certain subgroups tend to be more salt-sensitive. The mechanisms underlying sodium-induced increases in BP are not completely understood, but may involve alterations in renal function, fluid volume, fluid regulatory hormones, the vasculature, cardiac function, and the autonomic nervous system. Recent pre-clinical and clinical data support that even in the absence of an increase in BP, excess dietary sodium can adversely affect target organs, including the blood vessels, heart, kidneys, and brain. In this review, we address these issues and the epidemiological literature relating dietary sodium to BP and cardiovascular health outcomes, addressing recent controversies. We also provide information and strategies for reducing dietary sodium.

INTRODUCTION

Sodium is essential for fluid balance and cellular homeostasis. Claude Bernard was the first to highlight the “milieu intérieur.” Walter Cannon more explicitly defined homeostasis when he referred to the "fluid matrix" of the body and emphasized the role of sodium (1). In the last several decades, there has been a tremendous amount of work exploring dietary sodium and health. The amount of sodium needed to maintain homeostasis in adults is exceedingly low (<500 mg) compared to the average intake of most Americans (>3,200 mg)(2). We review the effects of dietary sodium on blood pressure (BP) and outcomes, emphasizing that excess sodium has direct adverse effects on target organs, beyond the increased risk of hypertension (HTN). We also review strategies for reducing sodium.

PATHOPHYSIOLOGY: SALT SENSITIVITY OF BP

BP responses to alterations in dietary sodium vary widely, leading to the concept of salt-sensitive (SS) BP (3,4). There are no standardized guidelines or firm BP cutoffs for classifying individuals as having SS BP. If BP increases during a period of high dietary sodium or declines during a period of low sodium, the individual is SS. If there is no change in BP with sodium restriction, an individual is salt-resistant (SR). Limited evidence supports the reproducibility of these responses (5,6). While individuals are commonly dichotomized as SS or SR (3), BP responses to sodium manipulation follow a Gaussian distribution (7,8). Table 1 lists groups that tend to be SR or SS (4,9-14). Salt-sensitivity in normotensive adults predicts future HTN (14,15) and SS BP has been associated with increased mortality (16). While there is interest in studying the pathophysiology of SS BP, there is less interest in its routine clinical assessment (17).

Table 1.

Salt Sensitivity in Various Groups^*

Salt-Resistant	Salt-Sensitive
Young	Aged
Middle-aged	Hypertensive
Normotensive	African-American
Caucasian	Chronic kidney disease
	History of pre-eclampsia
	Low birth weight

^*References 4, 9-14.

PATHOPHYSIOLOGY: MECHANISMS OF SODIUM-INDUCED INCREASES IN BP

The physiological mechanisms underlying SS BP are not fully elucidated, but involve alterations in renal function, fluid hormones, the vasculature, the heart, and/or alterations in central sympathetic outflow (Central Illustration). There are also genetic mechanisms related to the SS phenotype (18-20).

Central Illustration. Dietary Salt and Health: Mechanisms Mediating Dietary Salt-Induced Alterations in BP.

High dietary sodium can potentially exert its influence through various mechanisms to cause an increase in BP through alterations in cardiac output and total peripheral resistance. The change (Δ) in BP varies considerably, even within a given population (as depicted in the distribution). AT₁ = angiotensin II receptor, type1; BP = blood pressure.

Guyton’s studies demonstrated that sodium loading caused extracellular volume expansion and volume-loaded HTN in the context of induced renal dysfunction in dogs (21), consistent with clinical studies in patients with chronic kidney disease (10). The DOCA-salt model of experimental HTN in rats requires removal of 1 kidney, further supporting the kidney’s role in expression of salt-sensitivity (22).

Impaired hormonal (renin-angiotensin-aldosterone) responsiveness during a sodium manipulation is linked to a SS BP response. Indeed, African Americans have a blunted plasma renin response to a sodium manipulation (23). The molecular signaling pathways involved in sodium-induced increases in BP are not known, but likely involve angiotensin II, type 1 (AT₁) receptors (24), found in the renal and nonrenal vasculature, and the central nervous system, which are important for BP and/or fluid regulation. Mice lacking renal AT₁ receptors become SS (24).

Smooth muscle in the peripheral vasculature has also been implicated in SS BP responses. Studies suggest that elevated dietary sodium expands the extracellular volume and increases cardiac output, which will increase BP if there is no compensatory decline in peripheral resistance (25). Thus, an unchanged or increased peripheral resistance, coupled with a sodium-induced increase in cardiac output, results in a SS BP response, as observed in African Americans (25).

The autonomic nervous system may play an important role in SS BP. Rodent studies demonstrate modest plasma sodium elevations from high dietary sodium may signal to the brain, causing elevated sympathetic outflow (26). Other studies demonstrate a central interaction between sodium and angiotensin II, which increases sympathetic outflow, which targets splanchnic (27) and renal (28) circulations, and may be an important mechanism in SS BP. Studies in humans support a link between plasma sodium and BP (29,30) and osmolality and sympathetic outflow (31), but these findings are not consistent (32). Extremely low sodium diets (~230 mg) over a short time period (6 days) are associated with increased sympathetic outflow in humans (33).

The standard laboratory rat is SR, but SS strains have been bred (34). For example, Dahl fed Sprague-Dawley rats high-sodium chow and grouped them according to BP response, developing the Dahl SS and SR lineages (35). While SS BP is heritable, outside of monogenic renal tubular disorders causing sodium retention and HTN, the genetic underpinnings of “routine” salt sensitivity in humans are unknown (19). There has been recent progress in understanding genetic mechanisms, such as the GenSalt studies in China that identified AT₁ gene variants predictive of salt-sensitivity (18). Genome-wide association studies and overall BP studies have had limited success—genetic loci associated with a small effect on BP have been identified, (36), which is likely to be the case with SS BP responses.

PATHOPHYSIOLOGY: SODIUM AND TARGET ORGAN EFFECTS

There is evidence that in the absence of increased BP, elevated dietary sodium can adversely affect multiple target organs and tissues (19), including the vasculature, heart, kidneys, and areas of the brain that control autonomic outflow (Figure 1).

Figure 1. BP Independent Effects of High Dietary Sodium.

High dietary sodium can cause target organ damage and may have direct effects on the brain, heart, kidneys, and vasculature. These effects can be independent of changes in BP.

ARTERIES

Rodent studies demonstrated impaired endothelial function during sodium loading, without alterations in BP (37-40). Sodium loading in normotensive men reduced endothelial function (41), and sodium restriction in adults with elevated BP improved endothelial function (42). Additionally, high sodium impairs endothelial function in normotensive SR humans, providing support for a BP-independent effect of sodium on the endothelium (43,44). Sodium’s deleterious effects on endothelial function likely results from reactive oxygen species (38,44), such as superoxide (39,40), resulting in reduced nitric oxide bioavailability. Cell culture studies support that high sodium exposure stiffens endothelial cells and damages the glycocalyx (45).

Animal studies show that elevated dietary sodium can increase arterial stiffness independent of BP (46). In human studies, increased arterial stiffness was observed in groups consuming a higher sodium intake, independent of BP (47,48). This increased stiffness is likely related to the profibrotic effects of transforming growth factor-β (49). Thus, high sodium stiffens the arteries, and reducing dietary sodium lowers arterial stiffness in hypertensive patients (50,51).

HEART AND KIDNEYS

Increased BP is a major risk factor for left ventricular (LV) hypertrophy; high dietary sodium may increase LV wall thickness (52) and mass (53), independent of HTN status. For example, among a cohort of healthy adults with minimal HTN, those with the highest sodium excretion had greater LV mass (53). High aldosterone levels may be important in mediating the effect of dietary salt on LV mass (54). Also, a 12-month sodium restriction intervention in hypertensive patients has been shown to reduce LV hypertrophy (55).

There are a limited number of studies of subjects without kidney disease, but evidence suggests that high sodium is associated with reduced renal function (56). Sodium loading in spontaneously hypertensive rats increased renal vascular resistance, glomerular pressure, serum creatinine, and proteinuria; sodium loading also caused a decline in single-nephron plasma flow. This decline in renal function was observed with only a minimal additional increase in BP (57). Sodium restriction has been shown to reduce protein excretion and BP in black hypertensive patients (58). Similarly, in the LowSalt CKD study (59), low salt reduced proteinuria, albuminuria, and BP.

BRAIN

Sodium may affect brainstem nuclei that control BP (34). Chronically elevated dietary sodium may "sensitize" sympathetic neurons in the rostral ventral lateral medulla of rodents (60-62), causing a greater sympathetic response to a variety of stimuli (63), including skeletal muscle contraction (64). This increased responsiveness has been associated with increased BP variability, even without an elevation in average BP (65); this is relevant due to the association of BP variability with target organ damage (66). Even in the absence of increased BP, chronically increased sympathetic outflow may have deleterious target organ effects.

EPIDEMIOLOGY: DIETARY SODIUM, BP, AND CARDIOVASCULAR OUTCOMES

Studying the effect of salt restriction on clinical outcomes raises significant challenges, including: 1) assessment of sodium intake (best evaluated by multiple measurements of 24-h sodium urine excretion); 2) long-term maintenance on a defined salt intake regimen; and 3) the necessity for large numbers of patients and long-term follow-up to obtain enough outcomes for analysis. Randomized controlled clinical trials (RCTs) between groups with different amounts of sodium in the diet would reduce bias, but have largely been limited to short-term evaluation of the effect of salt restriction on BP, as larger studies with the longer time frames required to evaluate the effects of sodium on cardiovascular (CV) events have not been feasible.

SALT INTAKE AND BP

Multiple meta-analyses and systematic reviews of RCTs have shown a strong positive association between sodium intake and systolic BP (67-73) and a significant reduction in systolic BP with sodium restriction (74,75). A recent meta-analysis of 103 randomized interventions confirmed these results, showing a linear association between salt restriction and systolic BP. The reduction was larger with older age, among blacks, and among hypertensive patients (76,77). The incidence of HTN also decreased following a sodium intake reduction intervention in the Trials of HTN Prevention II RCT (78).

SALT INTAKE AND CV OUTCOMES

Few randomized trials have sufficient power and long enough follow-up to examine the effects of sodium restriction on CV outcomes (79). In a meta-analysis of 7 randomized sodium-reduction trials with follow-ups of at least 6 months, Taylor et al. did not find any effect of sodium restriction on all-cause mortality, CV mortality, or CV morbidity (80). However, He and colleagues replicated this analysis after excluding 1 trial with methodology issues related to the patient population (patients with heart failure) and showed that a modest reduction in salt intake resulted in a significant 20% decrease in CV and stroke events 73 (80,81).

Most of the studies examining the association between salt intake and CV events are observational cohort studies and, as described in an American Heart Association report (AHA)(82), are subject to multiple methodological issues. As listed in Table 2, errors with the greatest potential to alter the association in either direction are: 1) systematic errors in sodium assessment, most frequently related to measurements of sodium intake through food frequency questionnaires, 24-h recalls, spot or overnight urine collections, or 24-h urine collection without evidence of quality control measures; and 2) reverse causality related to recruiting sick patients who may consume less sodium as part of a therapeutic strategy or reduced overall food consumption, or to not excluding sick participants from general population studies (82). Cobb et al found evidence of systematic error in sodium assessment in 77% of studies that showed a direct association between salt intake and CV events, in 75% of those with an inverse association, in 100% (only 2 studies) of those with a J-shaped association, and in 100% of those that showed a null association. Reverse causality was found in 31% of those that showed a direct association, in 38% of those with an inverse association, in 50% of those with a J-shaped association, and in none of those with a null association (82). Considering these results, and that 3 to 4 methodological issues were identified in each study (82), systematic reviews and meta-analyses of observational studies should be evaluated with circumspection. Since AHA methodology report was published, O’Donnell et al, in an analysis of the Prospective Urban Rural Epidemiology (PURE) data, a prospective observational study of 101,945 adults, described a J-relationship between sodium excretion and CV events, with increased CV events at <3g/day and ≥7g/day (83). This was maintained, even after restricting the population to a low-risk cohort and excluding events occurring within 2 years to limit reverse causality. However, each patient’s sodium excretion was estimated from a single fasting morning urine specimen, overestimating the 24-h urinary excretion and likely introducing a systematic error in sodium assessment according to the AHA Science Advisory’s classification (82). There may also have been reverse causality. A subsequent analysis of the observational follow-up of the Trials of Hypertension Prevention, which, according to the AHA methodology report, has a low potential for reverse causality and for systematic error in sodium assessment, showed a 17% increase in CV events for every 1,000 mg/d increase in sodium (p = 0.054) and no evidence for a J-relationship (84).

Table 2.

Methodological Issues With Cohort Studies That Relate Sodium Intake to Cardiovascular Disease^*

Errors with the greatest potential to alter the direction of association in either direction

• Systematic error in sodium assessment

• Reverse causality

Errors with some potential to alter the direction of association in either direction

• Residual confounding

• Inadequate follow-up

Errors with potential to lead to a false null result

• Random error in sodium assessment

• Insufficient power

^*Reference 82.

In conclusion, a large body of evidence confirms the biological plausibility of the association between high sodium intake and increases in BP and CV events, whereas the evidence for an association between low salt intake and adverse events is unclear and largely conjectural. The association of low sodium intake with mortality has been speculated to result from elevated renin-aldosterone activity, sympathetic activation, and lipid abnormalities (85). However, meta-analyses suggest no significant effect of low sodium on lipids and no effect on catecholamines (74). Also, while dramatic short-term reductions in sodium may increase renin-angiotensin-aldosterone activity (86), modest-to-moderate longer-term reductions, as suggested by the AHA or IOM (87,88), may only produce minimal increases (69). Finally, because of the multiple methodological issues associated with cohort studies and the difficulty of organizing a trial to assess the association between sodium intake and CV events, the AHA recommends sodium on the basis of the large body of evidence linking sodium intake to BP (82).

SODIUM IN THE DIET: LIMITING SODIUM IN THE DIET

Sodium is ubiquitous in the diet of most developed countries. While some sodium intake is necessary, recommendations vary for adequate intake and tolerable upper intake levels (79,89,90). Physiological requirements for sodium are <500 mg/day in most healthy individuals, but the average consumption in the United States is over 3,200 mg/day (2,88,91,92). NHANES data reveal that sodium consumption increased between the early 1970s and the early 1990s (2,91). NHANES data from 2003 to 2008, using 24-h dietary recall, show that 99.4% and 97% of U.S. adults consumed more sodium than recommended by the AHA and the 2010 U.S. Department of Health and Human Services Dietary Guideline for Americans, respectively (79,90). While there is general consensus that current sodium consumption levels are excessive and contribute to CV risk (79,88), the U.S. Food and Drug Administration considers sodium added in food preparation to be “generally regarded as safe,” and there are no standards for its safe use in food (88).

Most people like some level of salt in food. Conceptually, there is a “bliss point” where the effect of sodium on flavor is optimum (93). However, this “bliss point” is malleable, and most people will adapt (94) to a reduction in dietary sodium (88). Sudden sodium changes are harder to accept, but if the United States gradually moves to a diet with less sodium, many people will likely make the transition with little difficulty (94-96).

Approximately 70% of sodium in the diet is in processed foods (97-99) and used to prepare foods as ubiquitous as bread (100). Sodium added in food preparation and at the table contributes less (88,98). Restaurants are also more likely to have saltier foods and more people are eating out in recent decades.

Market forces are a factor in the large amount of sodium in the diet, so without a societal approach, pressure on individual stakeholders is likely to be resisted. Food processors have marketed low-sodium alternatives without much success (101). A partnership of food processors and restaurant associations with groups like the AHA is more likely to successfully change diets (79). Efforts in Finland and the United Kingdom have successfully reduced sodium (102-104).

A number of approaches can decrease the dietary sodium: 1) decrease the sodium content of foods; 2) a switch by the consumer from high-sodium to low-sodium foods by avoiding processed foods and reading labels; 3) switch to substitute salts (105-107); 4) reduce sodium while increasing other flavors (96,100); and 5) use engineering approaches to provide salty taste with less sodium or food processing with less sodium (88,108). Importantly, flavor must be maintained because taste is the driving force behind salty foods (88,109). A decrease in sodium from current levels will represent a major change in our food supply and may be most successful as a series of small steps over several years. Coordination between food processors, restaurants, and advocacy groups is crucial and currently lacking (88). Importantly, increased dietary potassium intake may decrease salt sensitivity and favorably effect BP (110).

Altered dietary sodium intake targets should be considered for individuals engaging in high physical activity or exposed to heat stress (111). Although there may be some sweat loss accommodation in response to sodium restriction (112), sodium recommendations should be assessed to ensure that sodium intake matches sweat loss. While there is general consensus that the current sodium intake in the United States (3,200 mg per day) should be lowered, the Institute of Medicine cautioned against sodium diets <2,300 mg per day for selected groups (113).

CONCLUSIONS

BP correlates with sodium intake, with multiple mechanisms underlying this relation. Preclinical and clinical studies demonstrate that sodium adversely affects multiple target organs independent of BP. Clinical trials have shown decreased BP with decreased sodium intake, but the studies relating sodium consumption to CV events have significant limitations related to difficulty in assessment of sodium intake and confounding (79,82,114). Lack of power has been a barrier to demonstrating an effect of reduced sodium on hard outcomes in normotensive people. The difficulties of adhering to a sodium restriction diet over years may be an insurmountable hurdle for an RCT with enough power to detect a difference in CV events that could be generalizable to the entire population. Because of the weight of evidence in favor of salt reduction and the difficulties in organizing a clinical trial, the AHA recommends a population-wide reduction in sodium intake (87). Reducing sodium will take a coordinated effort involving organizations like the AHA, food producers and processors, restaurants, and public policy aimed at education.

ABBREVIATIONS

AHA

American Heart Association

BP

blood pressure

CV

cardiovascular

HTN

hypertension

LV

left ventricle

RCTs

randomized controlled clinical trials

SR

salt-resistant

SS

salt-sensitive