Modifying dietary sodium and potassium intake: An end to the salt wars?

Abstract

Excessive salt intake raises blood pressure, but the implications of this observation for human health have remained contentious. It has also been recognized for many years that potassium intake may mitigate the effects of salt intake on blood pressure and possibly on outcomes such as stroke. Recent large randomized intervention trials have provided strong support for the benefits of replacing salt (NaCl) with salt substitute (75% NaCl, 25% KCl) on hard outcomes, including stroke. During the same period of time, major advances have been made in understanding how the body senses and tastes salt and how these sensations drive intake. Additionally, new insights into the complex interactions between systems that control sodium and potassium excretion by the kidneys and the brain have highlighted the existence of a potassium switch in the kidney distal nephron. This switch appears to contribute importantly to the blood pressure lowering effects of potassium intake. In recognition of these evolving data, the United States Food and Drug Administration is moving to permit potassium-containing salt substitutes in food manufacture. Given that previous attempts to reduce salt consumption have not been successful, this new approach has a chance of improving health and ending the ‘salt wars’.

Introduction

Hypertension is the major modifiable risk factor for cardiovascular disease ^1–3. Especially as the population ages, hypertension is projected to have an ever-increasing effect on society ^1,4,5. The relative risk for cardiovascular disease is directly associated with the absolute level of blood pressure (BP) ⁶. Small magnitude reductions of up to 5mmHg can have highly significant positive effects on risk for adverse cardiovascular events, especially at the population level ⁷. In fact, one model predicted that reducing salt intake by 3 grams per day could reduce yearly cardiovascular death by 50% ⁸.

Salt Consumption

Salt (sodium chloride) was the predominant food preservative until relatively recently. As far back as 5000 years before the present, salt-preserved meats and vegetables were important components of the diet because salt maintains food. Even after the advent of refrigeration, however, salted foods have been consumed, largely because salting increases taste appeal.⁹ Modern food processing includes a substantial amount of sodium salts, including sodium chloride, making processed and restaurant food high in Na⁺.

Saltiness is one of 5 primary taste sensations. Interestingly, however, salt can be either an appetitive or aversive stimulus, depending on its concentration; higher concentrations are typically aversive. Typical appetitive salt taste in rodents is mediated by tongue cells of the fungiform papillae that express the epithelial sodium channel (ENaC).¹⁰ Indeed, ENaC is widely expressed throughout body tissues.¹¹ The taste for salt and salt-seeking can be attenuated dramatically in rodents by the ENaC-blocking drug amiloride, or by deleting the alpha subunit of ENaC from these cells (Figure 1) ¹⁰. A separate receptor may be responsible for the aversive effects of high salt concentrations, perhaps via a mechanism involving Cl^{− 12}. Yet, human salt sensing appears to be less amiloride-sensitive and the mechanisms involved are not as clear. It likely does involve ENaC, or subunits of ENaC, but involves other channels, as well ¹². Of relevance to the goal of optimizing salt intake, salt taste is highly dependent on the accompanying anion, such that sodium chloride and sodium bicarbonate do not taste the same.

Figure 1. Taste mechanisms in the Tongue.

Appetitive salt taste is mediated in rodents by ENaC. Sodium enters cells, thereby depolarizing them. This creates an action potential that activates channels that transport ATP (Calcium homeostasis modulator (CALHM) channel 1/3). ATP signals other channels in gustatory nerves. Salt at high concentrations (center cell) is aversive. This appears to involve chloride and perhaps type 2 taste receptors (T2R). This signal also makes salt seem bitter. The cell on the right detects a sour taste, which is mediated by proton entry, perhaps through otopetrin 1 (OTOP1). Note that appetitive salt sensing in humans has not been proven to involve ENaC, and may be more complex.

Salt-craving is elicited when the extracellular fluid (ECF) volume is depleted, in part because aldosterone and angiotensin II, secreted or generated in response to ECF volume depletion, activate neurons in the nucleus tractus solitarius that express the enzyme 11-beta-hydroxysteroid dehydrogenase type 2 (HSD2). This enzyme is also expressed in the distal parts of the kidney tubule; in both brain and kidney, it allows aldosterone to activate mineralocorticoid receptors. 11βHSD2 knockout in the brain increases hunger for salt, modifying salt-resistance towards salt-sensitivity. As such, 11βHSD2-positive neurons may integrate salt appetite and the blood pressure response to dietary sodium through a mineralocorticoid receptor-dependent pathway.¹³ Activation of 11βHSD2 neurons increases the preference for salt, through neurons that express prodynorphin in the pre-locus coeruleus (pre-LC).¹⁴ Beautiful optogenetic work has highlighted that specific and yet intermingled neurons respond selectively to osmotic stress, to water depletion, and to ECF volume depletion, mandating specific behavioral responses that suppress the instigating aversive signal.¹⁵

Dietary Salt and Blood Pressure

A positive relationship between dietary salt intake and blood pressure has been known for more than 100 years, but the implications of that relationship for public health remain remarkably contentious; these disagreements have sometimes been called the ‘Salt Wars’ ¹⁶. These fundamental disagreements about the implications of the many observational and more limited interventional trials of salt restriction continue to be played out in the popular press, as well as within the confines of academia. Public health guidelines, both in the United States and elsewhere, have set ambitious goals for reducing salt intake; for example, the American Heart Association recommends that adults consume no more than 2.3 grams of sodium daily, with a desirable consumption of 1.5 grams (see Figure 2); the National Academies of Science, Engineering and Medicine updated their guidelines in 2019, also recommending that adults consume less than 2.3 grams ¹⁷. Yet, each time a guideline is issued, it is followed by a response, often from prominent hypertension experts, claiming that such levels are not just unfeasible, but may also be harmful ¹⁸. In fact, as shown in Figure 3, despite more than 20 years of recommendations to reduce salt intake, salt consumption remains remarkably stable. Rather than rehearse the voluminous, and often repetitive, data regarding the relationship between salt intake and blood pressure, this review will link new data from randomized controlled trials in humans with recent insights into mechanistic effects of dietary cation consumption, and suggest a new route to reduce population blood pressure and, we hope, calm the waters.

Figure 2. Salt and potassium intake in observational and interventional trials.

Estimated sodium and potassium intake across various trials, assuming that 7% of sodium and 23% of potassium appear in the urine. Note that the DASH ‘high NaCl arm’ is close to the median Na⁺ intake reported by Ma and colleagues²¹. In addition, the ‘control’ diet in the DASH was targeted to contain 25% less potassium (grey bar) than is typical in the United States. Note that the sodium intake recommended by several guidelines is well below average, and that the ‘ideal’ intake, recommended by the American Heart Association/ American College of Cardiology is below that used as the ‘low NaCl’ group in the DASH study. Based on the beneficial effects observed from the recent salt substitution trials, we propose (red star) that guidelines aim to increase potassium intake and reduce sodium intake by 20 mmol/day.

Figure 3. Typical sodium intake for males and females since 1999.

Note the comparison with current guidelines including recommended and ‘ideal’ intake of sodium. See text for more details about guidelines. Data from ¹⁰⁹.

While high intake of Na⁺ clearly can be detrimental to health, very low dietary Na⁺ intake has also been associated with adverse effects ^19,20. Although these results are widely suspected of being confounded ²¹, any reduction in ‘normal’ Na⁺ intake does activate the renin-angiotensin-aldosterone system. Hence, there remains debate over defining an optimal dietary Na⁺ intake as a worldwide public health measure ²². Yet recent interventional trials and recent insights into molecular physiology are now converging to justify at new strategy to ‘optimize’ diets in the United States.

While most guideline recommendations focus on salt (NaCl) intake, it has been clear since the early studies of Leaf and colleagues that the anion that accompanies Na+ has substantial effects on the Na⁺ distribution in the body.²³ This is reflected in the effects of different salts of sodium on blood pressure in hypertensive humans, in whom NaCl raise blood pressure but Na citrate did not.²⁴ Mechanisms that determine the relationship between salt intake and hypertension, and the role of anions, have been widely reviewed and will not be discussed here.^15,25

Dietary potassium and blood pressure

The concept that dietary potassium intake reduces blood pressure is not new. During the last half of the twentieth century, a number of investigators provided data supporting a beneficial role for potassium on cardiovascular health; in fact, dietary potassium was also shown in animal models to provide cardiovascular protection, independent of its effects on blood pressure, perhaps through direct effects on vascular smooth muscle.²⁶ While K⁺ supplementation alone can lower BP, the INTERSALT, Dietary Approaches to Stop Hypertension (DASH), and PURE studies reported that the BP-reducing effects of K⁺ are more notable in individuals consuming a high salt diet.^27,28,29

Urine Na⁺ excretion, as a relevant surrogate for dietary Na⁺ intake, positively associates with BP. Adjusting for covariates, for a 1-g increment in estimated sodium excretion SBP and DBP were found to be raised by 2.11 mmHg and 0.78 mmHg respectively ²⁹. Such findings have contributed to a public understanding that excessive salt consumption is adverse for cardiovascular health. However, despite also findings that K⁺ excretion inversely correlates with BP,²⁹ a good understanding of the effects of K⁺ for cardiovascular health seems quite limited in the general population. Indeed, the estimated potassium excretion was associated with a steeper inverse relationship with systolic and diastolic blood pressure among persons with increased levels of estimated sodium excretion, as well as among older persons, those with hypertension, and those with an increased body-mass index. Hence beneficial effects of K⁺ appear to be most pronounced in people with the greatest co-morbidities or risk factors for adverse cardiovascular health.³⁰

Comparing Modern and Ancestral diets

One approach to determining an ‘optimal’ human diet has been to use information concerning diets of early human ancestors. Interestingly, information about the palaeolithic diets is itself controversial. For years, studies relied predominantly on data from small populations of contemporary ‘hunter gatherers’, who can be studied contemporaneously ³¹; such studies typically suggest that most of the diet consists of fresh fruits and vegetables, with meat consumption intermittent and modest. Yet, more recent evolutionary approaches, leveraging evidence from human physiology and genetics, archaeology, paleontology, and zoology, have suggested that, until the advent of agriculture, humans subsisted primarily on meat ³². Despite these caveats, many investigators have used diets inferred for ‘hunter gatherers’ as models. Dietary Na/K intake ratio of hunter gatherers has been estimated to be about 0.07 (daily 768mg sodium and 10500mg potassium), whereas for the modern USA population, the Na/K intake ratio is about 1.6 (daily 4000 mg sodium and 2500 mg potassium) ^21,33.

One other assumption underlying the use of inferred ancient diets to make recommendations about optimal modern diets is that human dietary tolerance has been evolutionarily stagnant. This is likely untrue, as strong selection for certain genetic traits, such as lactase persistence, appears to have evolved much more recently;³⁴, Thus, approaches that suggest optimal diets are based on ancient diets are conjectural and should be supported by more contemporary approaches.

A major factor in the controversy regarding advisable sodium intake involves its appetitive character. More than 40 years ago, it was suggested that “low sodium diets are therapeutically effective but generally regarded as an impossible or an unnecessary nuisance”, and that assessing the dietary Na/K ratio could be clinically relevant ³⁵. Based on evolutionary theory it was considered that even the ‘normotensive’ level of BP in most Americans is abnormally high for our species, based on pressure values obtained from isolated tribes in ‘less developed’ parts of the world ³⁵.

Effects of dietary manipulation

The plasma potassium level needs to be maintained within a tight range. For the physiological is typically 3.5 – 5.5 mmol/L ³⁶. The absolute value is generally reproducible for most laboratories, whether drawn as a terminal or non-terminal procedure, although the method of collection and quantification (E.g. whole blood using I-Stat versus centrifuged plasma) can lead to variations. It is worth noting that many earlier studies using mouse models reported values as high as 6 mM as ‘normal’.^37,38; it is now recognized that those values are artifactual. For effects of dietary studies to be appropriately assessed and compared we propose that researchers also routinely present the change in Na⁺ and K⁺ following dietary feeding compared to a baseline value on a control diet. Further, we impress the importance of stating urine Na/K ratio value as well as levels of the ions individually.

It is worth noting that the relative amounts of sodium and potassium provided by most rodent laboratory chows differ greatly from human diets typical in the United States. Typical rodent ‘control’ diets contain 0.2–0.4% sodium, with the typical molar sodium/potassium ratio for such diets approximately 0.5, while the ratio is typically more than 1.5 in humans. This means that a ‘control diet’ for laboratory animals may correspond to a ‘high potassium diet’ for humans. The contrasting effect of high potassium intake in humans and animal models, described more fully below, may be at least in part, explained by this difference.

Salt-dependent hypertension has been attributed, at least in part, to a genetic defect in renal salt excretion requiring an elevation of BP to maintain salt balance on high dietary salt intake ³⁹; although excess sodium retention does not occur in Dahl sensitive rats, compared with control ⁴⁰. It was reported that K⁺ had a notable anti-hypertensive effect when combined in high Na⁺. Indeed, the change in BP was found to be inversely proportional to the dietary molar Na/K ratio.⁴¹ The beneficial effect of K⁺ with the concurrent feeding of Na⁺ was further confirmed in Dahl salt-sensitive rats ⁴² and also in the genetic model of spontaneously hypertensive Wistar-Koyto rats fed high sal.⁴³

Potassium effects on the vascular system

High extracellular K⁺ concentration depolarizes membranesand are routinely used in the study of isolated arterial function. In contrast, the effects of systemic potassium supplementation on the vasculature remain conflicting, which may be attributed at least partly to the various regimes administered. While high dietary K⁺ intake may have limited effect on human BP in salt-resistant conditions,⁴⁴ it can significantly increase flow mediated dilation of the brachial artery⁴⁵ and significantly increase pulse pressure.⁴⁶ Clinically advantageous effects towards vascular function following increased K⁺ intake have been reported to involve actions directly stimulating vasodilation,⁴⁷ thought to occur principally via stimulation of the K⁺ channel Kir2.1 and to a smaller extent increased Na/K-ase activity in vascular smooth muscle cells. Increased activity of such channels induces hyperpolarization of the membrane potential, favouring Ca2⁺ efflux ^48,49. Endothelial cell-VSMC signalling has also long been known to contribute to vasodilation, but the endothelial derived hyperpolarising (relaxing) factor (EDHF) appears to be a combination of factors, which can include K⁺ directly or in consort with other dilatory factors ^50–52.

Such dilatory effects may be, at least, influenced by nitric oxide (NO). High extracellular K⁺ has been reported to reduce the stiffness of bovine aortic endothelial cells in culture, which seems related to nitrite.⁵³ Also, high dietary K⁺ intake significantly reduced BP and increased the bioactivity of NO in normotensive salt-sensitive people dwelling in rural north China.⁵⁴ Potential antihypertensive effects of potassium supplementation may also be mediated through regulation of vascular sensitivity to catecholamines.^53,55 The direct effects of supplemental or high K⁺ seem in contrast to effects of Na⁺, which seems to have limited direct effect on arterial function ⁴⁷

Over a longer timeframe, arterial remodelling, including reduced lumen diameter of small resistance arteries, with VSMC proliferation or rearrangement, can drive hypertension ^56–58. Treating VSMCs in vitro with 7mmol/L extracellular K⁺ significantly attenuated cell proliferation ⁵⁹. Non renal effects of K⁺ can appear to be relatively slow, or at least an accumulation of beneficial effects is required with time, for a net positive physiological effect ⁴⁷. As such, while relatively short-term K⁺ feeding protocols can be informative on renal function, longer feeding periods of weeks should be studied towards understanding holistic effects of high K⁺ and potential reciprocal effects between different tissues. As shown by Little and colleagues, 4 days 10% KCl significantly reduced BP, but after 3 weeks BP was significantly higher than control fed animals.⁶⁰

Potassium effects on the kidney

Potassium supplementation can have quite a strong natriuretic effect in humans ^60–62 and rodents.^60,63 Natriuresis results from the inhibition of sodium chloride transport along the proximal tubule ⁶⁴, the thick ascending limb ⁶⁵, and distal convoluted tubule. These effects have recently been modelled. A model that incorporates tubuloglomerular feedback emphasizes that the distal tubule plays the dominant role in mediating these effects.⁶⁶ In this model, high potassium intake inhibits salt reabsorption along the proximal tubule thereby activating the tubuloglomerular feedback mechanism, which suppresses the glomerular filtration rate, limiting the contribution of proximal tubule.

Potassium intake modulates NCC

Part of the effect of potassium on blood pressure stems from its effects on the thiazide-sensitive sodium chloride cotransporter (NCC). Functional NCC activity is suppressed by high dietary NaCl intake.⁶⁷ In fact, the distal tubule, in concert with the adrenal gland, comprises a ‘potassium switch’,¹⁵ which involves NCC in the DCT and ENaC in the connecting tubule and collecting duct. NCC is strikingly sensitive to dietary potassium intake. A K⁺ deficient diet for as little as 12 hours.^68–71 increases the abundance of phosphorylated NCC. As phosphorylation activates NCC, this has profound, but indirect, effects to reduce K⁺ excretion. Indeed, plasma [K⁺] regulates NCC activity directly;^72,73 a low plasma K⁺ concentration activates NCC, which reduces Na⁺ delivery to distal segments that secrete K⁺. As increases in plasma [K⁺] also stimulate aldosterone secretion, which activates ENaC, the combination of increased Na⁺ delivery and activated ENaC combines to generate kaliuresis. The importance of this effect for potassium balance is evident from Gitelman syndrome patients (genetic loss of NCC function) and pseudohypoaldosteronism type II (PHA II, also known as Gordon’s syndrome) patients (genetic gain of function of NCC) who present with hypokalemia ⁷⁴ and hyperkalemia ⁷⁵ respectively. Yet this mechanism is also important for the blood pressure lowering effects of high potassium intake.

For rats fed a low K⁺ and Na⁺ replete diet (0.03% tri potassium citrate with 3% NaCl) for 10 days, BP was increased by about 15mmHg compared to animals on control diet (0.93% K⁺ and 0.24% Na⁺) while the low K⁺ and Na⁺ replete diet significantly increased total and pNCC expression ⁷⁶. Importantly, such increases in BP were shown to be dependent on NCC activity as elevated BP and plasma Na⁺ concentration were normalised following sustained HCTZ treatment ⁷⁶, while the low K⁺ and Na⁺ replete diet promoted no significant increase in BP in Slc12a3 knock out mice (global NCC negative) ⁷².

A reduced level of pNCC has also been reported in ex vivo isolated renal tubules incubated with high (8mM) [K⁺] compared to control solution of 3.5mM [K⁺] for 24 hours ⁷⁷. Rapid dephosphorylation of NCC, as early as 30 minutes after gastric gavage with 2% K⁺, is then sustained for at least 6 hours ⁷⁸. Additionally, when KCl was infused into the tail vein over 3 hours, raising plasma K⁺ concentration to approximately 5.4 mM, pNCC was lower (again by about 60%) than in control NaCl infused animals ⁷⁹.

Potassium intake modulates ENaC

Sodium reabsorption along more distal segments of the nephron occurs via the epithelial sodium channel (ENaC), in exchange for K⁺ secreted via the renal outer medullary K⁺ channel (ROMK; also termed Kir1.1) ^80,81. This channel also plays a crucial role in the response to variations in sodium and potassium intake. The importance of ENaC for regulation of sodium and potassium balance is evident in Liddle syndrome patients (gain of function mutation in ENaC), who present with hypertension and hypokalemia, and pseudohypoaldosteronism type 1 (PHA1, loss of ENaC function) patients, who present with hyperkalemia and hypotension.

ENaC is regulated by aldosterone ^82,83. Although other factors play important roles, the two most prominent aldosterone secretagogues are angiotensin II and potassium. Thus, there is a well-described linear relationship between plasma [K⁺] and aldosterone secretion. While ENaC is regulated directly by aldosterone, NCC is primarily regulated by the plasma [K+].^84–86. Thus, when plasma [K⁺] is low, NCC is active and, because aldosterone is low, ENaC is suppressed. This limits K+ excretion because Na⁺ is reabsorbed by NCC with chloride in the DCT, limiting Na+ delivery to the connecting tubule; in the connecting tubule, the low Na+ delivery combines with suppressed ENaC, generating little electrogenic Na⁺ reabsorption and therefore little K⁺ secretion.

Complex interplay between sodium and potassium

A recent meta-analysis of randomized controlled trials (duration ≥4 weeks) showed that SBP was elevated when dietary K⁺ intake was under 80 mmol/day and also tended to be higher when intake was high; thus, BP and K were best fitted as a ‘U shaped’ relationship (Figure 4) ⁸⁷. In mice a clear U shape relationship between BP and plasma K⁺ concentration has recently been reported for animals maintained on a control NaCl diet (0.3% Na⁺) and fed a KCl depleted diet or supplemented diet (10% KCl) for 3 weeks (Figure 4)⁶⁰. Another factor in these studies is plasma Cl^−, which can also promote vasoconstriction,⁸⁸ and hence affect BP. While in the study from Little and colleagues 3 weeks 10% KCl feeding did induce a significant increase in the plasma Cl⁻ concentration, the Cl⁻ load also appeared to be excreted, as fractional Cl⁻ excretion was significantly higher than for control fed animals. Plasma bicarbonate or pH may also contribute to BP but were not measured in the study.⁶⁰ In this study, the BP changed from being lower to being higher than that of control fed animals after about one week of exposure to dietary K⁺ excess. This occurred despite suppressed NCC activity. To our knowledge this is the first study to monitor the effect of high K⁺ feeding over an extended timeframe and to also relate BP or pNCC expression to plasma K⁺ concentration⁶⁰.

Figure 4. Impact of potassium intake on blood pressure in humans and rodents.

Left panels show results of controlled studies supplementing potassium intake in humans. Top left panel shows a U-shaped curve versus final achieved potassium intake. Note the blood pressure nadir near 100 mmol/day. The other panels deconstruct the data according to baseline sodium intake. Note that these graphs use the difference in potassium intake, rather than the total intake. When sodium intake is high (and aldosterone is presumed to be low), the effects of potassium are monotonic. In contrast, when sodium intake is lower, then high potassium consumption increases blood pressure. Data from Filippini ⁸⁷, with permission. Right panels show effects of potassium intake in mice and describe similar U-shaped curves. The lower panel shows the expected effects of potassium to stimulate aldosterone section. Data from Little⁶⁰, with permission. All panels show systolic blood pressure (SBP).

Figure 5 shows a proposed model for effects of dietary K⁺ intake on sodium handling by the kidney and on blood pressure. The model assumes that there is a linear negative relationship between plasma [K⁺] and NCC activity across the physiological range. The ability of K⁺ to suppress NCC is a result of both direct effects on Kir4.1 potassium channels and through increased bradykinin.^89,90 At the same time, aldosterone, which is secreted in response to elevations of plasma [K⁺], activates the sodium channel, ENaC, although this relationship is mitigated when dietary salt intake is high. The model suggests that when dietary K⁺ intake is low, NCC activity is stimulated, leading to NaCl retention and hypertension. Conversely, when dietary K⁺ intake is very high, ENaC activity is stimulated, which can also lead to Na⁺ retention and hypertension; according to the data in Figure 4, however, potassium intake in humans rarely reaches sufficient levels to increase blood pressure.

Figure 5. Proposed kidney mechanism for effects of both low potassium intake and very high potassium intake to raise blood pressure.

When potassium intake is low, NCC is stimulated, leading to NaCl retention and hypertension. In contrast, when dietary K⁺ intake is high, NCC is suppressed, both owing to direct effects and to bradykinin.^89,110 reducing blood pressure. High potassium intake also stimulates aldosterone secretion. When aldosterone is very high, indicated by the asterisks, its effects on ENaC may overcome the effects on NCC and raise blood pressure. The complex relations between salt and potassium are illustrated in Figure 4. Blue arrows indicate NaCl transport via NCC. Red arrows indicate Na⁺ reabsorption via ENaC. Purple arrows indicate potassium secretion. DCT1 is the first segment of the DCT. DCT2 Is the later segment of the DCT that expresses ENaC. CNT is the connecting tubule that strongly expresses ENaC and is under the control of aldosterone.

When dietary NaCl intake is moderately high and aldosterone is suppressed, the effects of K⁺ to suppress NCC activity remain dominant. The role of aldosterone is illustrated by the following. Firsts, when dogs were fed up to 200 mmol/day potassium, BP did not change; when dogs were adrenalectomized and given replacement amounts of mineralocorticoids at a fixed dose, however, the blood pressure did decrease.⁹¹ Second, when the effects of K⁺ supplementation on blood pressure in humans are parsed according to Na⁺ intake, the pressure-raising effects of K⁺ are absent, when dietary Na intake is high.⁸⁷ (Figure 4)

Implications for guideline development

It is widely accepted, even by the most skeptical observers, that most people consume more salt than is needed for excellent health. Yet, as noted, average salt consumption in the United States has remained largely unchanged. Even in England, where a nationwide effort was made to reduce salt consumption, intake has rebounded⁹². Two major developments are suggesting that more easily adopted guidelines can now be developed that both improve health and lengthen life. The first is the advent of large scale randomized dietary intervention trials.^93–95 The second is the understanding of the complex interactions between salt and K⁺ intake that provide a mechanistic understanding of how the balance between salt and K⁺ affects blood pressure.

Given the abundance of food choices, it is imperative that guidelines can easily be followed. New insights into salt-sensing by the tongue and brain complement efforts like the blind taste test of salt substitutes conducted recently by Consumer Reports. Their testers found that combinations of NaCl and KCl (as in the ‘salt substitutes’ used in controlled trials) “tasted and looked the most like the real thing…In rice and eggs, the testers had a hard time telling the difference between the light and regular salt.”⁹⁶ Similar results have been published in the scientific literature.⁹⁷ This important insight into palatability can leverage the results of randomized trials showing that replacing a modest amount of dietary salt with KCl can have substantial beneficial effects not only on blood pressure, but also on “hard outcomes”, such as stroke ⁹³.

The alignment of observational studies, intervention trials, and laboratory work allows us to recommend an approach to dietary cation consumption that is at variance with many current guidelines, but has a much greater chance of widespread adoption. As noted in ‘Futurity’ (https://www.futurity.org/restaurant-food-salty-sodium-1580062/), “Restaurant food is notoriously high in sodium-it’s one reason it tastes so good”; additionally surveys show people in the United States are eating out more, not less⁹⁸. Figure 2 suggests a reasonable goal to improve the diet of Americans, without substantially affecting the taste of food. We recommend widely introducing ‘salt substitute’ (75% NaCl/25% KCl); this approach could be implemented both by industry and in the home. Using the estimated baseline data from the meta-analysis,²¹ such a change would reduce the molar sodium/potassium ratio from 2.2 to 1.4.⁹⁹

By advocating for substituting a modest amount of KCl for NaCl, a change that might be imperceptible, the country can finally begin to move the needle on cation intake. This change might also help to reduce disparities in hypertension and salt-sensitivity. Black Americans have been found repeatedly to consume less potassium than white Americans.¹⁰⁰ Yet, Kurtz and colleagues provided evidence that the racial disparity in salt sensitivity disappears when dietary K⁺ intake exceeds recommended levels¹⁰¹. Additionally, although the American Heart Association suggests that “on average, blacks compared with whites have a greater BP response to a change in salt intake and that this finding is independent of baseline BP level”,¹⁰² salt-sensitivity should be mitigated when dietary potassium intake is high.^103,104

On March 24, 2023, the United States Food and Drug Administration announced plans to ‘amend the standards of identity to permit the use of salt substitutes in foods for which salt is a required or optional ingredient.’¹⁰⁵ The proposed rule would allow food manufacturers to make simultaneous changes to salt and potassium intake that would be expected to improve human health, without adversely affecting taste. This change, which is independent of guideline amendments, holds the potential to improve human health and reduce health disparities, without requiring individuals to make large dietary changes.

Sodium and potassium in chronic kidney disease

Some have argued that efforts to reduce the dietary Na⁺/K⁺ ratio might actually increase risk, for example in individuals with chronic kidney disease (CKD).¹⁰⁶. These and other patients are frequently taking drugs that raise plasma [K⁺], drugs widely used for cardiovascular and kidney disease prevention and treatment. Yet, owing to the potential for a variety of health benefits, the recommendations to restrict dietary potassium in patients with CKD have been questioned. In early follow up, potassium supplementation of K⁺ was well tolerated, although a few cases of hyperkalemia were observed. Interestingly, a reduction in blood pressure was not observed, but aldosterone concentrations rose.¹⁰⁶ If it is the plasma [K⁺] that drives both NCC activity and aldosterone secretion, then such individuals might be expected to show less benefit from supplementation, as they tend to have higher plasma [K⁺] at baseline.

Summary and Perspectives

The recent controlled clinical trials supporting long term dietary interventions to improve and prolong life should reduce the contentious arguments regarding salt, potassium and blood pressure. It is noteworthy that the two recent large trials both included interventions that increased potassium intake, while reducing sodium intake; in both, the increase in potassium was relatively larger than the decrease in sodium.^93,94,107 Despite the recent data, some continue to be skeptical regarding the advisability of moving forward with dietary changes.¹⁰⁸ One issue that has been emphasized is the nature of the ‘control’ diets in the various trials. For example, the potassium content of the ‘control’ diet in the DASH study was at the 25^th percentile.²⁸ Similarly, the basal potassium consumption amongst participants in the recent salt-substitute study in China, and the baseline consumption amongst the participants in the study from Peru, was quite low.^93,94 As shown on Figure 2, these values are well below the average consumption in the United States and many other countries. Yet, the data from the meta-analysis by Filippini⁸⁷ suggest that, when salt consumption is substantial, the benefits of potassium consumption are monotonic (see Figure 4).

Coupled with recent insights from the experimental laboratory, these results suggest that rather modest reductions in dietary salt intake, coupled with modest increases in potassium intake will provide substantial health benefits, while maintaining the palatability that humans appear to desire.