Summary

Potassium and sodium are physiologically essential nutrients. Their functions are closely intertwined, and each has important roles in maintaining physiological homeostasis. Both nutrients have also been implicated in chronic disease risk, particularly cardiovascular disease, mainly through their effects on blood pressure. Additionally, a possible association of sodium intake with other adverse health outcomes has been suggested at low levels of intake. The unique nature of potassium and sodium—that is, the coexistence of their essentiality with a relationship to adverse health effects, including chronic disease risk—necessitated a new approach to the review of intake recommendations for these nutrients within the Dietary Reference Intakes (DRIs) context.

The DRIs are a set of quantitative reference values for the apparently healthy population, developed jointly for the United States and Canada. The DRIs are derived through an iterative process that was developed in response to recognition of the need for a safe and adequate range of intake for nutrients and other food substances, beyond meeting essential requirements to prevent deficiency diseases. Although the DRI model envisioned use of evidence for chronic disease risk, the model proved to be challenging and insufficient for that purpose. The relationships between diet and chronic disease risk are complex and are dependent on a variety of factors, both nutritional and nonnutritional, such as an individual’s baseline risk for the chronic disease, environmental factors, and nutrient–diet or nutrient–nutrient interactions, exposure time, and other lifestyle factors. The intake–response relationships between nutrient intakes and chronic

disease risk are often more complex than the relationships observed for adequacy and toxicity effects.

The evolution of both the DRIs and the definition of nutritional health to include more than the essential nutrients led to the reexamination of the DRI model for ways to consider inclusion of chronic disease in the process. A 2017 National Academies of Sciences, Engineering, and Medicine (the National Academies) report, Guiding Principles for Developing Dietary Reference Intakes Based on Chronic Disease (Guiding Principles Report), provides guidance and recommendations for expanding the DRI model to include a new category of reference values specific to chronic disease risk reduction. This study represents efforts to apply those recommendations to the process of deriving DRIs for potassium and sodium.

THE COMMITTEE’S TASK AND APPROACH

The inextricable link between potassium and sodium, in both biology and study designs, makes their concurrent DRI review both scientifically justified and efficient. An ad hoc committee of the National Academies was asked to review current evidence and update, as appropriate, the DRIs for potassium and sodium that were established in the 2005 report Dietary Reference Intakes for Water, Potassium, Sodium, Chloride, and Sulfate (2005 DRI Report). The committee’s Statement of Task is presented in Box S-1.¹

___________________

¹ The Statement of Task was abbreviated for this Summary. The complete Statement of Task is presented in Chapter 1, Box 1-1.

The committee was asked to incorporate the DRI organizing framework, which provides a structured process for establishing DRIs and consists of the following steps:

1. Review the evidence on indicators and select the indicator(s) that will inform the DRIs.²
2. Assess the intake–response relationships of the selected indicator(s) and establish DRI values.
3. Compare current population intake levels to DRI values to characterize risk.
4. Discuss public health implications and special considerations.

In addition to the DRI organizing framework, the committee was asked to apply the guidance from the Guiding Principles Report, which allows for evidence on chronic disease risk to be used to derive DRI values separate from the other DRI categories for adequacy and toxicity. With the existing DRI categories, which focused on essentiality and toxicity that can affect all individuals, the general assumption is that failure to derive a reference value is often not a viable public health option. In contrast, reference values for the DRI based on chronic disease are generally intended to be established only when the body of the evidence is sufficient to do so. In particular, the Guiding Principles Report recommended at least a moderate strength of evidence for both the causal and intake–response relationships between nutrient and chronic disease risk.³ In the context of this DRI review of potassium and sodium, the committee has called the specific category of DRIs based on chronic disease reference value the Chronic Disease Risk Reduction Intake (CDRR).⁴

The committee was provided with a systematic review prepared for this study, Sodium and Potassium Intake: Effects on Chronic Disease Outcomes and Risks (AHRQ Systematic Review), which served as a primary source of evidence. The AHRQ Systematic Review included risk-of-bias and strength-of-evidence assessments, along with meta-analyses of randomized controlled

___________________

² In context of the DRIs, an indicator broadly refers to clinical endpoints, surrogate markers, biomarkers, and chronic disease risk factors.

³ For consistency throughout this report and in alignment with the terminology used in the AHRQ Systematic Review, the committee uses the term strength of the evidence instead of quality of the evidence or certainty of the evidence when describing the grading of the evidence used to derive DRIs based on chronic disease.

⁴ Throughout the report, DRIs based on chronic disease is used when broadly describing the category. The committee uses CDRR to describe the specific category of values established. This aligns with the committee’s use of the phrases DRIs for adequacy, which broadly refers to the Estimated Average Requirement, Recommended Dietary Allowance, and Adequate Intake, and DRIs for toxicity, which refers to the Tolerable Upper Intake Level.

trial results.⁵ As a user of the AHRQ Systematic Review, the committee assessed its methodological quality and identified two domains that could be strengthened: exploring unexplained heterogeneity in meta-analyses and providing clear explanations of the process for grading the strength of the evidence. To interpret the evidence, the committee addressed these domains by conducting heterogeneity analyses and explaining in detail its grading of the evidence for assessments central to its decision making.⁶

The committee also undertook information-gathering activities that included hosting a workshop and public comment session, requesting information from the public and stakeholders, performing scoping searches to identify potential indicators, and conducting supplemental literature searches on selected indicators not included in the AHRQ Systematic Review.

The committee’s findings, conclusions, and resulting DRIs for potassium and sodium are presented in the following sections, organized by the steps in the DRI framework outlined above.

DIETARY REFERENCE INTAKES FOR POTASSIUM

Step 1: Review and Selection of Indicators

Indicators to Establish Potassium DRIs for Adequacy

The committee’s review of the evidence on potential indicators to inform the potassium DRIs for adequacy revealed the following:

• There is no sensitive biomarker that can be used to characterize the distribution of potassium requirements in the apparently healthy population.
• Limitations in the design of potassium balance studies—particularly the small sample size and incomplete measurement of intake and losses—precluded the committee from using such data to estimate median potassium requirements and the distribution of potassium requirements in the apparently healthy population.

___________________

⁵ A meta-analysis is a statistical analysis that combines the results of multiple scientific studies. Its interpretation is complicated by heterogeneity among the studies. Observed differences in the intervention effect between the studies could result from clinical diversity (the participants, interventions, and outcomes studied) and/or methodological diversity (differences in study design and risk of bias).

⁶ For DRIs based on chronic disease, the committee used the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system to assess the strength of the evidence as high, moderate, low, or insufficient depending on the level of confidence in the effect estimate related to potassium or sodium intake.

The committee concludes that none of the reviewed indicators for potassium requirements offer sufficient evidence to establish Estimated Average Requirement (EAR) and Recommended Dietary Allowance (RDA) values. Given the lack of evidence of potassium deficiency in the population, median intakes observed in an apparently healthy group of people are appropriate for establishing the potassium Adequate Intake (AI) values.

Indicators to Establish Potassium DRIs for Toxicity

The committee’s review of the evidence on potential indicators to inform the potassium DRIs for toxicity revealed the following:

• Case reports provided evidence that very large doses of potassium supplements can result in cardiac abnormalities and death. The doses of potassium in these case studies are generally imprecise and have been confounded by comorbidities and medication use.
• Potassium supplementation may slightly increase blood concentrations of potassium, although among adults with normal kidney function, there is no evidence that it results in hyperkalemia (serum potassium concentration > 5.5 mmol/L).
• No consistent patterns of reported adverse events were identified across the potassium supplementation trials included in the AHRQ Systematic Review and in the committee’s supplemental literature search.

The committee concludes that there is insufficient evidence of potassium toxicity risk within the apparently healthy population to establish a potassium Tolerable Upper Intake Level (UL).

Indicators to Establish Potassium DRIs Based on Chronic Disease

The committee’s review of the evidence on potential indicators to inform the potassium DRIs based on chronic disease revealed the following:

• The independent effect of potassium intake on all-cause mortality, cardiovascular disease, coronary heart disease, myocardial infarction, stroke, and chronic kidney disease has not been assessed in randomized controlled trials. Evidence from prospective cohort studies tended to be rated as having moderate or high risk of bias, and there was potential confounding of results attributable to dietary potassium’s strong correlation with other nutrients in the diet. These limitations precluded the determination of causal

• relationships and led to grading of low or insufficient strength of evidence for these indicators.

• The results of potassium supplementation trials on bone mineral density may differ by conjugate anion in the supplement (e.g., citrate, bicarbonate, or chloride), and therefore do not necessarily reflect the independent effect of potassium.
• There is insufficient evidence of an effect of potassium intake on kidney stones, and there is low strength of evidence that higher potassium intake may be associated with lower risk of kidney stones.
• There is insufficient evidence of a causal relationship between potassium intake and incident diabetes, glycemic control, and insulin sensitivity.
• There is moderate strength of evidence that potassium supplementation significantly reduces systolic and diastolic blood pressure. The effect was stronger among studies that included adults with hypertension. Still, considerable heterogeneity existed across trials and the committee was unable to determine its source. An intake–response relationship with dose of supplemental potassium could not be established.

The committee concludes that, although there is moderate strength of evidence for a causal relationship between potassium supplementation and reductions in blood pressure, heterogeneity across studies, lack of evidence for an intake–response relationship, and lack of supporting evidence for benefit of potassium on cardiovascular disease prevents the committee from establishing a potassium Chronic Disease Risk Reduction Intake (CDRR).

Step 2: Establishing Potassium Dietary Reference Intake Values

Data from the Canadian Community Health Survey–Nutrition 2015 (CCHS Nutrition 2015) and the National Health and Nutrition Examination Survey 2009–2014 were used to derive the potassium AIs. The committee sought to use intake data from apparently healthy survey participants, particularly those whose usual potassium intake would not be affected by illness, use of medications, or medical nutrition management. For adults, this consisted of normotensive males and females without a self-reported history of cardiovascular disease. The highest median potassium intake across the two surveys was selected as the AI for each of the DRI age and sex groups in children and adolescents, for adult females, and for adult males. The potassium AIs for infants were derived from estimates of potassium intakes in breastfed infants. The updated potas-

sium DRIs consist only of AIs for all age, sex, and life-stage groups (see Table S-1). There was insufficient evidence to establish EARs, RDAs, ULs, or CDRRs for potassium. A summary of the updated potassium DRIs is presented in Box S-2.

Steps 3 and 4: Characterization of Risk and Implications for Public Health

A comparison of the updated potassium AI values to distributions of potassium intakes in the United States and Canada revealed slight differences across population groups reviewed. For instance, in the United States, non-Hispanic blacks tended to have lower potassium intakes than their non-Hispanic white and Hispanic counterparts. Because it is unknown how the AI value relates to actual requirements, interpretation of intakes below the AI in terms of inadequacy cannot be made.

The committee cautions against misinterpretation of the revised potassium DRIs. The potassium AIs, although based on evidence from normotensive adult males and females, are intended to be applicable to the broader apparently healthy population. The previous potassium AI for adults was based on evidence from potassium supplementation trials that investigated chronic disease–related health outcomes. In the expanded DRI model, chronic disease risk reduction is characterized under a separate DRI category. The lack of a potassium CDRR does not necessarily indicate that there is a lack of an effect of potassium intake. Rather, the moderate strength of evidence for a blood pressure lowering effect of potassium supplementation, coupled with both a lack of an intake–response relationship and a lack of evidence of an effect on chronic disease endpoints, highlights

the need for further exploration of the effect of different doses and forms of potassium. Similarly, the absence of a potassium UL does not mean there is no risk from excessive intake either overall or for segments of the population. Caution against high intake through supplemental potassium is warranted for certain population groups, particularly those with or at high risk for compromised kidney function.

DIETARY REFERENCE INTAKES FOR SODIUM

Step 1: Review and Selection of Indicators

Indicators to Establish Sodium DRIs for Adequacy

The committee’s review of the evidence on potential indicators to inform the sodium DRIs for adequacy revealed the following:

• There is no sensitive biomarker that can be used to characterize the distribution of sodium requirements in the apparently healthy population.
• Sodium balance studies generally had small sample sizes and incomplete measurement of losses, which limits generalizability and accuracy. Furthermore, intra-individual variability and emerging evidence of infradian rhythms (i.e., lasting more than 24 hours) augment the uncertainty related to the duration needed to reach a steady state; recent evidence on potential skin and muscle sequestration may also affect the interpretation of data from balance studies.
• There is a limited and inconsistent body of evidence on the potential harms of low sodium intake. The inconsistency appears to be caused, in part, by methodological approaches used in observational studies.

The committee concludes that none of the reviewed indicators of sodium requirements offer sufficient evidence to establish Estimated Average Requirement (EAR) and Recommended Dietary Allowance (RDA) values. Adequate Intakes (AIs) are therefore established. Median population intakes are not suitable for establishing sodium AIs because they exceed the sodium Chronic Disease Risk Reduction Intake (CDRR) values. The committee concluded that the lowest levels of sodium intake evaluated in randomized trials and evidence from the best-designed balance study conducted among adults were congruent and are appropriate values on which to establish the sodium AIs.

Indicators to Establish Sodium DRIs for Toxicity

The committee’s review of the evidence on potential indicators to inform the sodium DRIs for toxicity revealed the following:

• Very high, acute intakes of sodium have resulted in hypernatremia (serum sodium concentration > 145 mmol/L) and death, but such intakes generally occur only under extreme circumstances.
• There is evidence to suggest that adverse effects could result when sodium is consumed in a concentrated form, but the evidence does not currently allow for quantification of a UL based on a specific toxicological effect.
• Some trials have reported headaches to be less prevalent during the lower-sodium period or in the lower-sodium group of participants, as compared to those in the higher-sodium interventions. Current evidence does not characterize the type, severity, duration, and frequency of headaches reported.

The committee concludes that there is insufficient evidence of sodium toxicity risk within the apparently healthy population to establish a sodium Tolerable Upper Intake Level (UL).

Indicators to Establish DRIs Based on Chronic Disease

The committee’s review of the evidence on potential indicators to inform the sodium DRIs based on chronic disease revealed the following:

• Few trials have assessed the effect of sodium intake reductions on the following indicators: cardiovascular mortality, myocardial infarction, left ventricular mass, stroke, osteoporosis, or kidney disease. The strength of evidence for causal relationships between sodium intake and these indicators was rated as low or insufficient.
• There is a moderate strength of evidence for a causal relationship between reductions in sodium intake and all-cause mortality. There are, however, more specific chronic disease endpoints with moderate or high strength of evidence.
• There is a moderate strength of evidence for a causal relationship between reductions in sodium intake and any cardiovascular event.⁷ Likewise, there was moderate strength of evidence from

___________________

⁷ In contrast to meta-analyses examining a single cardiovascular event as an outcome (e.g., a meta-analysis of cardiovascular mortality data), this meta-analysis combines data on cardiovascular events together (e.g., mortality, stroke, heart failure).

• randomized controlled trials to suggest that reducing sodium intake reduces hypertension incidence.

• There is a high strength of evidence from randomized controlled trials that reducing sodium intake reduces systolic and diastolic blood pressure. Much of the observed heterogeneity among trials examining systolic blood pressure could be explained by the net reduction in sodium (intake–response) and the baseline systolic blood pressure level. Among trials examining diastolic blood pressure, heterogeneity was mainly related to the difference in the size, rather than in the direction, of the effect. The effect of sodium reduction was greater among adults with hypertension, but it was also evident among nonhypertensive adults.

The committee concludes there is moderate to high strength of evidence for both a causal relationship and an intake–response relationship between sodium and several interrelated chronic disease indicators: cardiovascular disease, hypertension, systolic blood pressure, and diastolic blood pressure. Evidence from these indicators can be synthesized to inform the development of a sodium Chronic Disease Risk Reduction Intake (CDRR).

Step 2: Establishing Sodium Dietary Reference Intake Values

Establishing the Sodium AIs

To establish the sodium AIs, the committee reviewed the range of sodium intakes that have been assessed in sodium reduction trials included in the AHRQ Systematic Review. In a controlled feeding trial, the lowest levels of sodium intake ranged from 949 to 2,452 mg/d (41 to 107 mmol/d). The low sodium intake group or period in eight additional sodium reduction trials was below 1,800 mg/d (78 mmol/d). Across the trials, no deficiency symptoms were reported. Furthermore, there was insufficient evidence that low sodium intakes are associated with other potential harmful health effects. Taking these two types of evidence, together with evidence from the best-designed balance study, in which approximately neutral balance was achieved with daily heat stress⁸ at sodium intake of 1,525 mg/d (66 mmol/d), the committee established the sodium AIs for adults 19 years of age and older at 1,500 mg/d (65 mmol/d). The adult AI was extrapolated to children and adolescents 1–18 years of age based on sedentary Estimated Energy Requirements. The sodium AIs for infants were derived from estimates of sodium intakes in breastfed infants.

___________________

⁸ This text was revised since the prepublication release.

Establishing the Sodium CDRRs

The sodium CDRR for adults is based on a synthesis of evidence from sodium-reduction trials and outcomes of incident cardiovascular disease, incident hypertension, systolic blood pressure, and diastolic blood pressure. The sodium CDRR is the lowest level of intake for which there was sufficient strength of evidence to characterize a chronic disease risk reduction. Further reductions in sodium intake below the CDRR have demonstrated a lowering effect on blood pressure, but the effect on chronic disease risk could not be characterized.

Although there was insufficient evidence to establish a CDRR based on trials conducted in children and adolescents, there is evidence of blood pressure and cardiovascular disease risk tracking from early childhood into adulthood. Despite uncertainties about the long-term chronic disease benefits of reduced sodium intake beginning in childhood, the committee considered the risk of not setting a CDRR for children and adolescents to outweigh the risk of establishing a CDRR. In the absence of indicators for adverse effects, the adult CDRR is extrapolated to children and adolescents 1–18 years of age based on sedentary Estimated Energy Requirements.

Summary of the Sodium DRIs

The committee updated the sodium AIs across the DRI age, sex, and life-stage groups and introduced CDRRs for individuals 1 year of age and older (see Table S-2). There is insufficient evidence to establish EARs, RDAs, or ULs for sodium. A summary of the updated sodium DRIs is presented in Box S-3.

Steps 3 and 4: Characterization of Risk and Implications for Public Health

The vast majority of the U.S. and Canadian populations consume sodium above both the AI and CDRR values. There is no concern regarding sodium inadequacy in the population. Intakes above the CDRR, however, increase the risk of chronic disease in the population. Although larger effects of sodium reduction on blood pressure have been observed in adults with hypertension as compared with normotensive adults, the benefits of sodium intake reduction related to the CDRR are applicable to both. The evidence was insufficient to further define the applicable population (e.g., by age, weight status, race/ethnicity, comorbidities). As such, the committee notes that there are population groups with higher prevalence and risk for hypertension and cardiovascular disease. These include, but are not limited to, older individuals and certain race/ethnicity groups, particularly

TABLE S-2 Sodium Dietary Reference Intakes by Age, Sex, and Life-Stage Group

NOTE: AI = Adequate Intake; CDRR = Chronic Disease Risk Reduction Intake; mg/d = milligrams per day; ND = not determined; UL = Tolerable Upper Intake Level.

^aUpdated DRI value, as compared to the 2005 DRI Report.

^bNot determined owing to lack of a toxicological indicator specific to excessive sodium intake.

^cNot determined owing to insufficient strength of evidence for causality and intake–response.

^dExtrapolated from the adult CDRR based on sedentary Estimated Energy Requirements.

non-Hispanic blacks. Reducing intake toward the CDRR level will likely be particularly beneficial for these groups. Although there is evidence that further reductions in sodium intake below the CDRR can lower systolic and diastolic blood pressure, the effect on chronic disease risk cannot be characterized at this time.

The committee cautions against misinterpretation of the revision to the sodium DRIs. The sodium AI for adults 19–50 years of age that was established in 2005 DRI Report is reaffirmed. There remains limited evidence on sodium intakes below 1,500 mg/d (65 mmol/d), which prevented the committee from considering further reductions in the sodium AI. With the expansion of the DRI model, the UL now represents an intake above which toxicological risk increases. The risk that was formerly characterized in the sodium UL established in the 2005 DRI Report is now captured in the CDRR. The sodium CDRR, however, extends beyond the approach and evidence that informed the sodium UL in the 2005 DRI Report. The sodium CDRR is derived from evidence of risk of incident cardiovascular disease and incident hypertension, and reductions in systolic and diastolic blood pressure. It also

incorporates different methodologies, including use of a systematic review, committee-conducted meta-analyses, and grading of the evidence.

RESEARCH NEEDS AND FUTURE DIRECTIONS

The expansion of the DRI model afforded the committee the opportunity to specify the lowest level of sodium intake for which there was sufficient strength of evidence to characterize a chronic disease risk reduction in the population. This expansion also created challenges for the committee to define the potassium AIs and the sodium UL, which previously drew on evidence related to chronic disease endpoints. The refinement of the DRI categories brought to light the dearth of evidence on potassium and sodium requirements outside of the chronic disease context. Characterizing the toxicological effects of excessive potassium and sodium intake levels also posed challenges, as human studies are not designed to examine toxic effects because of ethical considerations.

Future potassium and sodium DRIs would benefit from additional research that identifies requirements for both nutrients and better characterizes negative health effects from high intake levels, to the extent that safety can be assessed. Future updates to the sodium CDRR would benefit from research that provides additional insight into population groups that have different responses to sodium intake. With the vast majority of the U.S. and Canadian populations consuming sodium at levels above the CDRR, opportunities exist to find novel solutions to reduce population sodium intakes, including technical innovations to decrease sodium in the food supply. Regarding potassium, the evidence on the relationships with chronic disease endpoints was of insufficient strength to establish a CDRR. Future trials that assess the long-term effects of different doses and forms of potassium are needed to characterize the intake–response relationship with blood pressure and chronic disease outcomes. Methodologically rigorous randomized controlled trials that study the effect of sodium on chronic disease endpoints are also still needed. Furthermore, additional research is needed on the interrelationship between potassium and sodium intakes.

Finally, as the first to implement the guidance in the Guiding Principles Report, the committee identified opportunties for improvement. These opportunties are related to defining applicable populations when prevalence of chronic disease is high, integrating a systematic review of the evidence into the DRI process, adapting the guidance and recommendations for establishing DRIs based on the expanded model, and providing additional guidance on the expanded DRI model as experience is gained over time.

This page intentionally left blank.