Introduction

Salt is unlikely to have been used as a food additive during human evolution, but over a relatively recent period of human history, it has been used to preserve food, especially over the winter period.1 The need for preservation of food with salt is now minimal due to the development of cold storage, including freezing, and the rapid transport of food from the site of production to that of consumption. Nevertheless, salt intake remains at very high levels in most regions of the world, as has been extensively documented over recent decades.2,3

Taste and salt intake

The major mechanism of salt taste occurs through amiloride-sensitive epithelial sodium channels on type 1 taste receptor cells in the fungiform papillae, although amiloride-insensitive pathways have also been identified in the circumvallate and foliate taste buds.4 There is considerable interindividual variation in the sensation and palatability of saltiness.5

When people reduce their dietary salt intake, they often report that food seems bland and tasteless to them, and this may discourage them from persisting with their attempts to consume less salt. A prolonged reduction in salt intake induces a change in taste perception, such that foods taste saltier than they do when consuming a higher-salt diet.6 In a small study including healthy volunteers, those who had been following a low-salt diet chose to add less salt to their food than those who had been on a higher-salt diet.7 Similarly, people on a low-salt diet rated foods as saltier than they had rated the same foods when they were on a high-salt diet prior to reducing their salt intake.8 This change in salt perception developed over weeks, but took several months to reach the maximal effect.

A more detailed understanding of salt taste perception may be useful in the design of strategies to help people reduce their salt intake. For example, soup was perceived as less salty when served at 70 °C than when served at lower temperatures.9

Epidemiologic studies of salt and health status

There are extensive and wide-ranging studies of the relationship between salt intake and disease.10 Salt intake is generally inferred from an analysis of urinary sodium excretion, as salt is the dominant source of dietary sodium. Some studies refer to salt and others to sodium, and Table 1 provides the relevant conversion information.

Table 1. Sodium and salt values with related conversion ratesa

Sodium, mmol

Sodium, g

Salt, g

25

0.6

1.5

50

1.2

2.9

75

1.7

4.4

87b

2b

5.1b

100

2.3

5.8

125

2.9

7.3

150

3.5

8.8

175

4

10.2

200

4.6

11.7

225

5.2

13.1

a Conversion factors: to convert mmol of sodium to grams of sodium, multiply by 0.023; mmol of sodium to grams of salt, by 0.058; grams of sodium to grams of salt, by 2.54; grams of sodium to mmol of sodium, by 43.5; grams of salt to grams of sodium, by 0.393; grams of salt to mmol of sodium, by 17.1.

b Maximal daily intake recommended by the World Health Organization

A combined study of surveys of global sodium intake and a meta-analysis of 107 randomized interventions concluded that globally, 1.65 million deaths from cardiovascular disease (CVD) each year were attributable to sodium intake above the recommended level.11 More recently, the Global Burden of Disease study12 concluded that sodium was the major dietary risk factor for death and disability, resulting in 3 million deaths as well as a loss of 70 million disability-adjusted life years in 2017. In a UK Biobank study13 of over 0.5 million individuals, adding salt to food at the Table was associated with a hazard ratio of 1.28 for premature mortality. In a pooled analysis of 4 large studies including over 10 000 participants with a median follow-up of 8.8 years, salt intake, as assessed by multiple 24-hour urine collections, was associated in a dose-response manner with cardiovascular risk.14

A key observation made in the Intersalt study2 of over 50 communities globally is that the magnitude of the increase in blood pressure (BP) that occurs with age correlates with sodium intake. Both systolic and diastolic BP values are positively associated with sodium intake in both hypertensive and normotensive individuals, and the correlation remains even after adjustment for age, sex, body mass, educational attainment, alcohol intake, current smoking status, and geographic region.15

Dietary salt intake may be declining in better-resourced countries, and has fallen by around 5% over the 13-year period to 2019 in the United Kingdom.16 A reduction in salt intake has also been reported in individuals over 71 years of age in the United States.17 These changes may reflect various actions documented in some countries by the 2023 World Health Organization (WHO) report, and these actions typically included mandatory labelling of the sodium or salt content of food.10

Controversy around very low dietary salt intake

Sodium intake is generally assessed by measuring the sodium content of urine. For this purpose, 24-hour urine collections can be used, but this method relies on the participant collecting all their urine during this time period. In some population studies, spot urine collections have been used to estimate sodium intake; however, substantial error can arise from employing this method.18

In meta-analyses of some population datasets, a J-shaped curve has been observed in the relationship between urinary sodium excretion and cardiovascular events or mortality.15 This has raised questions about whether a very low salt intake is causative of increased mortality, and the significance of this observation has been debated.19-21 The large population studies on which these analyses have been undertaken typically relied on estimates of sodium intake based on spot urine tests, which may not provide a reliable index of sodium intake.4 The potential for misunderstanding and overinterpretation of the J-shaped curve has been extensively critiqued, and potential problems with such conclusions have been identified.18,20 Notably, when sodium intake was estimated in the Trials of Hypertension Prevention follow-up data,22 a J-shaped relationship was seen between the sodium intake modelled on the standard spot urine method and mortality, but when the average multiple measured values for sodium excretion from 24-hour urine collections were used, the relationship was linear. Spot urine collections have been shown to perform poorly at estimating sodium intake in patients with CKD.23

When tested against the Hill criteria, a causative relationship between low salt intake and increased mortality was found to lack credibility for multiple reasons.20 Sodium intake is derived from food, and almost all food products contain a significant amount of sodium, so, at low levels, sodium intake can correlate with food intake. Therefore, increased mortality in individuals with very low sodium intake could represent reverse causation, with frail or unwell persons having a low food intake and, consequently, a low sodium intake. Evidence supporting this explanation comes from a detailed study from the Netherlands, which found that the increase in mortality with low salt intake only occurred in individuals with a low protein intake, and not in those with a healthy protein intake.24

Trials of lowering salt intake in the general population

Multiple studies have investigated the health outcomes of lowering salt intake. A meta-analysis of 103 trials concluded that a reduction of sodium intake by 100 mmol per day (2.3 g of sodium or 5.8 g of salt; Table 1) lowered systolic BP by 3.74 mm Hg in normotensive individuals. In patients aged 70 years, BP decreased by 5.84 mm Hg, and in hypertensive individuals, the BP reduction was by 1.87 mm Hg greater than in people without hypertension.11 In a further meta-analysis of prospective cohort studies and randomized controlled trials, a mean difference in systolic and diastolic BP of 3.39 and 1.54 mm Hg, respectively, was observed when sodium intake was higher vs lower than 2 g per day.25 In an analysis of over 133 studies including 12 197 participants, there was a significant correlation between the reduction in sodium intake and the reduction in BP, and this relationship was stronger for older populations, non-White populations, and individuals with higher BP values.26

A major challenge associated with dietary intervention studies is to ensure that the intervention is adhered to. Various approaches have been used to optimize the uptake of interventions, including direct provision of low-salt food. Within the DASH (Dietary Approaches to Stop Hypertension) trial,27 the participants were assigned to eat either standard or healthy diets that were provided for them, and within these groups, they spent time in random order consuming the allocated diet with either low, intermediate, or high sodium content (60–70, 100–110, or 140–150 mmol of sodium per day, respectively).27 With both the standard and the healthy diet, there were progressive reductions in both systolic and diastolic BP values as sodium intake was reduced to the intermediate and low levels. Kitchens catering for resident retired veterans in Taiwan were randomized to use normal salt or a potassium-enriched salt substitute.28 Over a mean of nearly 3 years, there was a significantly lower rate of cardiovascular deaths in the low-sodium group. In a large study from China,29 600 villages were cluster randomized to use regular salt or a potassium-enriched salt substitute, and over nearly 5 years of follow-up, systolic BP was reduced, and rates of stroke, major cardiovascular events, and death were all significantly lower in the low-sodium group.

Guidance on salt intake for the general population

On the basis of the current evidence, there is a clear consensus that in the general population, greater sodium intake is associated with increases in BP and higher rates of cardiovascular events and overall mortality. This has led to guidance from the WHO and other bodies to reduce the consumption of sodium. The WHO guidance is to reduce the intake to less than 2 g of sodium per day, which is equivalent to 87 mmol of sodium or 5 g of salt per day (Table 1).3 Guidance from the European Society of Hypertension, endorsed by the International Society of Hypertension and the European Renal Association,30 highlights the frequent presence of fluid volume retention in individuals with hypertension and the proven value for BP control associated with restricting sodium intake to less than 2 g of sodium or 5 g of salt per day. This dietary guidance to reduce the salt intake is likely to be particularly relevant for patients with resistant hypertension.31,32

Dietary salt intake in chronic kidney disease

A key question is whether epidemiologic studies of general populations, which demonstrate an association between increased cardiovascular risk and increased salt intake, are also applicable to individuals with chronic kidney disease (CKD). This issue has been addressed by multiple researchers. The CRIC (Chronic Renal Insufficiency Cohort) study33 enrolled nearly 4000 patients with CKD across 7 locations in the United States, and at around 10 years of follow-up there was a significant and relatively linear association between high 24-hour urinary sodium excretion and an increased risk of CVD.33 After adjustment for multiple variables associated with modifiable cardiovascular risk factors, the hazard ratios for the highest quartile of urinary sodium excretion, as compared with the lowest quartile, were 1.36, 1.34, and 1.81 for composite cardiovascular events, heart failure, and stroke, respectively. Analysis of CKD progression in the CRIC study showed a hazard ratio of 1.54 for CKD progression in the highest sodium excretion quartile, as compared with the lowest quartile.34 Analysis of nondiabetic participants in the REIN (Ramipril Efficacy in Nephropathy) trial35 estimated that there was a 1.61-fold increase in the risk of end-stage kidney failure with a 100 mmol/g increase in urinary sodium-to-creatinine ratio. High sodium intake also reduced the antiproteinuric effect seen with ramipril use.

Multiple studies have tested the benefits of lowering salt intake in CKD. A randomized trial of sodium restriction in patients with CKD and hypertension reported reductions of 10 mm Hg in systolic and 4 mm Hg in diastolic BP, indicating that the benefits in CKD might be substantial.36 By delivering intensive personalized dietary advice and providing low-sodium foods, this study achieved notable reductions in sodium excretion from 168 mmol/day (3.9 g of sodium or 9.8 g of salt) in the high-salt group to 75 mmol/day (1.7 g of sodium or 4.4 g of salt) in the low-salt group. In a nonblinded UK trial involving 56 participants, reductions in ambulatory systolic and diastolic BP values of 8 and 2 mm Hg, respectively, were associated with a reduction in sodium excretion values from 260 mmol/day (6 g of sodium or 15.2 g of salt) to 103 mmol/day (2.4 g of sodium or 6 g of salt).37 When moderate salt restriction was added to angiotensin-converting enzyme inhibitor (ACEI) treatment in patients with nondiabetic CKD (target daily salt intake, 2.9 g; achieved, 6.2 g), reduction of proteinuria and BP values was more pronounced than in the case of adding treatment with an angiotensin receptor blocker (ARB).38 A small study in transplant recipients who were already on an ACEI or ARB demonstrated a reduction in BP values with reduced sodium intake.39 A decrease in BP was also seen with reductions in sodium excretion in the ESMO (Effects of Self-monitoring on Outcome of Chronic Kidney Disease) trial.40 A study of 58 people with stage 3 or 4 CKD demonstrated significant reductions in both BP values and whole body extracellular volume when 24-hour sodium excretion was reduced by 57 mmol/day (1.3 g of sodium or 3.3 g of salt).41

The intensive and personalized nature of the interventions used to lower salt intake resulted in most of these trials being relatively small in size, often having only around 50 participants in total. Nevertheless, a Cochrane review and meta-analysis of 21 studies of lowering salt intake involving 1197 participants with CKD concluded that there was high certainty evidence that salt reduction lowered BP in CKD.42

Guidance on salt intake for individuals with chronic kidney disease

In light of the available evidence, guidelines for the management of the CKD population generally include recommendations around sodium intake.43-45 The Kidney Disease Outcome Quality Initiative guidelines43 recommend a reduction to less than 2 g per day of sodium or 5 g of salt, or 90 mmol of urinary sodium excretion. However, there is evidence that targets for salt intake may be the least achieved management targets in the care of individuals with CKD.46,47 This is borne out by the high sodium excretion seen in control groups in many of the studies considered above. In a cross-sectional analysis of potentially modifiable cardiovascular risk factors in Italian patients with CKD, urinary sodium excretion was elevated at around 150 mmol/day (3.5 g of sodium or 8.8 g of salt) across CKD stages 3 to 5.46 Similarly, in a study of cardiovascular risk factors in individuals with CKD in the Netherlands, levels of around 155 mmol/day (3.6 g sodium or 9.1 g salt) were identified.47

Challenges around reducing salt intake in chronic kidney disease

Overall, there is strong evidence demonstrating the benefit of lowering salt intake in CKD, but it is also evident that salt intake remains elevated in these patient populations. Why is this? In general populations, some studies on lowering salt intake have used potassium-enriched salt substitutes, but this approach is less suited to CKD, where problems with hyperkalemia may be more common and more serious. Studies that have sought to test the effect of lowering salt intake in CKD have used a wide range of interventions. These have typically been labor- and resource-intensive interventions, as illustrated in the following examples. The provision of information by physicians was combined with personalized dietary advice and individualized dietary counseling from dieticians in a study from the Netherlands.38 The ESMO study40 had a team of 4 personal coaches (3 health psychologists and 1 dietician) supporting an intervention group of 67 participants, and this was supplemented with further education, an online food diary, and a cookbook of appropriate recipes. In a UK study with 26 participants in the intervention group, there was input from a study dietician, follow-up telephone calls, as well as practical cooking sessions involving hands-on food preparation and education sessions, both of which were attended by a researcher.37 In a further study, feedback on salt intake based on the results of 24-hour urine sodium excretion analysis was combined with individualized dietary counselling in an intervention group of 22 participants.39 Unfortunately, CKD is a common long-term condition, and the cost and nature of these personalized intensive interventions make them neither affordable nor feasible for large-scale use in routine clinical care of people with CKD, even in countries with relatively well-funded health care services.

The OxSalt intervention

Although there is a significant body of evidence demonstrating the health benefits of reducing salt intake for people with CKD, there has been very little, if any, evidence about practical interventions that could be implemented in routine clinical practice to help these individuals lower their salt intake in accordance with guideline recommendations. This makes it extremely challenging for the CKD population to reduce their salt consumption. To address this evidence gap, we developed a simple, low-cost intervention designed to help people with CKD reduce their dietary salt intake. This intervention was developed by a multidisciplinary team involving nephrologists, primary care physicians, nurses, dieticians, patients, and members of the public; we termed the intervention the OxSalt care bundle.48,49 The underlying premise of the intervention was that it should empower people to make the changes in their diet that would help them lower their salt intake. Three principles guided the development of the intervention, and these were to empower people to 1) understand the health benefits of reducing salt intake, 2) understand how to evaluate the salt contents of food, and 3) understand how to select or prepare food that is appetizing and low in salt content. The participants were informed about the time it can take for taste to adapt to lower dietary salt intake. The intervention lasted only 4 weeks, and at the beginning, the participants were provided with a set of self-explanatory slides to look over on a device with a screen. The slides covered the benefits of reducing salt intake as well as advice on how to evaluate the salt content of foods and how to select or prepare appetizing food that is low in salt content. Written information was also provided, including a booklet about the topics covered in the slides, a credit card–sized reminder about food labelling and salt content, and tips on selecting lower-salt foods and using nonsalt flavoring. The intervention also included access to a website, making all this information easily accessible. This was supplemented at minimal cost with reminders about dietary salt intake that were sent as automated phone text messages and emails. Details of the intervention and its components are available online.49

The OxCKD1 trial to evaluate the OxSalt intervention

The randomized controlled OxCKD1 trial48 recruited 201 participants with CKD to evaluate the OxSalt care bundle, as compared with standard routine clinical care. Multiple 24-hour urine collections were used to assess sodium excretion and so salt intake. Baseline urinary sodium excretion was 130.2 mmol/24 hours (3 g of sodium or 7.6 g of salt). Recruitment was conducted at multiple centers and participants were randomized with minimization.50 The primary outcome was a reduction in salt intake as assessed by 24-hour urinary sodium excretion after 4 weeks of the intervention, and at this time point, the mean sodium excretion was 92.9 mmol/24 hours (2.1 g of sodium or 5.4 g of salt) in the intervention group and 118.9 mmol/24 hours (2.7 g of sodium or 6.9 g of salt) in the control group (P = 0.001). The mean change in individual 24-hour sodium excretion was –32.4 mmol/24 hours (–0.7 g of sodium or –1.9 g of salt) for the intervention group vs –6.3 mmol/24 hours (–0.1 g of sodium or 0.4 g of salt) for the control group. As shown in Figure 1, the reduction in 24-hour urinary sodium excretion remained significant in the intervention group, but not the control group, at 2 and 11 months after the end of the intervention. The aim of the study was not to evaluate the effect of lowering sodium intake, as this is already well established, but rather to determine whether the OxSalt intervention could empower people to lower their salt intake. As compared with the control group, the intervention group had lower mean systolic (P = 0.05) and diastolic BP (P = 0.04) at 1 month. The results demonstrate that this simple and low-cost intervention helps people with CKD cut their salt intake significantly. The intervention was only tested for 4 weeks, and the longer-term results indicate that there may be benefit from further “booster” intervention elements beyond the 4-week period.

Figure 1. Results of the OxCKD1 study.48 Salt intake was assessed by measuring the sodium content in 24-hour urine collections. After a simple, low-cost intervention lasting for 4 weeks, there was a significant reduction in salt intake in the intervention, but not in the control group, and this effect persisted beyond the period of the intervention.

Conclusions

There is strong evidence that salt intake is associated with increased BP values and a greater risk of cardiovascular events and mortality in the general population. Trials of decreased dietary salt intake have resulted in reduced rates of all these outcomes. It can take some time for the taste to adjust to a lower salt intake. In CKD, salt intake is also associated with increased BP and higher rates of cardiovascular events and mortality. Although trials of reducing salt intake in CKD show benefit, they have typically used resource-intensive approaches to achieve the study end points, and such approaches are not feasible for widespread use in routine clinical practice. The OxSalt care bundle is a simple low-cost intervention that lasts only 4 weeks. As shown in the OxCKD1 trial, it helps people with CKD lower their salt intake significantly; thus, it could be applied at a larger scale for this patient population.