Dietary Reference Intakes of sodium for Koreans: focusing on a new DRI component for chronic disease risk reduction.

Sodium is a physiologically essential nutrient, but excessive intake is linked to the increased risk of various chronic diseases, particularly cardiovascular. It is, therefore, necessary to accomplish an evidence-based approach and establish the Korean Dietary Reference Intakes (KDRIs) index, to identify both the nutritional adequacy and health effects of sodium. This review presents the rationale for and the process of revising the KDRIs for sodium and, more importantly, establishing the sodium Chronic Disease Risk Reduction Intake (CDRR) level, which is a new specific set of values for chronic disease risk reduction. To establish the 2020 KDRIs for dietary sodium, the committee conducted a systematic literature review of the intake-response relationships between the selected indicators for sodium levels and human chronic diseases. In this review, 43 studies published from January 2014 to December 2018, using databases of PubMed and Web of Science, were finally included for evaluating the risk of bias and strength of evidence (SoE). We determined that SoE of the relationship between dietary sodium and cardiovascular diseases, cerebrovascular disease, and hypertension, was moderate to strong. However, due to insufficient scientific evidence, we were unable to establish the estimated average requirement and the recommended nutrient intake for dietary sodium. Therefore, the adequate intake of sodium for adults was established to be 1,500 mg/day, whereas the CDRR for dietary sodium was established at 2,300 mg/day for adults. Intake goal for dietary sodium established in the 2015 KDRIs instead of the tolerable upper intake level was not presented in the 2020 KDRIs. For the next revision of the KDRIs, there is a requirement to pursue further studies on nutritional adequacy and toxicity of dietary sodium, and their associations with chronic disease endpoint in the Korean population. ©2022 The Korean Nutrition Society and the Korean Society of Community Nutrition.

INTRODUCTION

Sodium is an essential element that maintains homeostasis and physiological functions. Being a major cation in the extracellular fluid, sodium is involved in osmotic pressure regulation and water balance, and also plays a role in nerve impulse transmission, muscle contraction, and acid-base balance [1]. Therefore, insufficient sodium intake is usually associated with adverse health outcomes. Contrarily, excessive sodium intake is a risk factor for several chronic diseases [23], particularly cardiovascular, mainly through the effect of sodium intake on blood pressure (BP) [2456]. It is well known that a reduction in the dietary sodium decreases BP and the incidence of hypertension, which is also associated with reduced morbidity and mortality resulting from cardiovascular diseases (CVDs) [237].

Dietary Reference Intakes (DRIs) are defined as a set of quantitative reference values for the apparently healthy population. They are traditionally used in planning and assessing diets based on nutrient deficiencies and/or toxicities [8]. Previously, an attempt was made for the Korean population to set the Estimated Average Requirement (EAR) and the recommended nutrient intake (RNI) for sodium adequacy, and the tolerable upper intake level (UL) for preventing health problems arising from sodium toxicity [9], leading to the establishment of the sodium adequate intake (AI) instead of the EAR and RNI, and the sodium intake goal instead of the UL, in the 2015 DRIs for Koreans (KDRIs). This was based taking into account the lack of appropriate indicators of sodium exposure, and insufficient scientific evidence on sodium adequacy or toxicity [9]. However, considering the impact of nutrition on the prevention, treatment and health promotion of chronic diseases, the use of chronic disease endpoints to establish the DRIs has been addressed. Population aging and the increased prevalence of chronic diseases worldwide, along with accumulating scientific evidence on the health effect of sodium intake, further potentiated the establishment of dietary recommendations aimed at reducing chronic diseases [8]. As such, there has been a consensus to include a new DRI category for the prevention of chronic diseases, in addition to the traditional framework of DRIs [10].

Very recently, chronic disease risk reduction intake (CDRR) for dietary sodium (a new category of reference values specific for chronic disease risk reduction) has been proposed in the Consensus Study Reports published by the committee of the National Academies of Sciences, Engineering, and Medicine [8]. In this report, the CDRR is defined as the lowest level of intake where there is sufficient strength of evidence (SoE) to characterize a chronic disease risk reduction, which was established in the DRIs for US and Canada [8]. This implies that reducing the sodium intake above the CDRR established would decrease the risk of chronic disease [8]. For the 2020 KDRIs, the committee comprehensively reviewed the available evidence, and considered the update of current KDRIs for sodium by focusing on the CDRR. A thorough systematic literature review of the relationship between dietary sodium and the risk of chronic diseases resulted in the committee establishing the sodium CDRR to prevent chronic diseases as a new category in the 2020 KDRIs.

This review presents the rationale and processes of establishing the KDRIs for adequacy and toxicity of dietary sodium, as well as specific values of the 2020 KDRIs for dietary sodium, using the 4-step evidence-based approach suggested by the US sodium committee [8], comprising the following steps: Step 1. Review and selection of indicators; Step 2. Assessment of the intake–response relationships of the selected indicators and establishment of DRI values; Step 3. Comparison of current population intake levels and DRI values to characterize the risk; Step 4. Discussion of public health implications and special considerations. This report provides the sodium intake level as an AI and, more importantly, we herein report a new CDRR value for dietary sodium for Koreans.

INDICATORS TO ESTABLISH THE KDRIs FOR DIETARY SODIUM

Indicators of sodium adequacy

A critical initial step of the DRIs framework is selection of the indicators for adequate and excessive dietary intakes of sodium. As reported earlier, no sensitive biomarkers are available to help characterize the sodium requirements in a healthy population [8]. For the purpose of the literature review, sodium balance, urinary sodium excretion, and sodium level in blood were selected as indicators of sodium exposure in the 2020 KDRIs. In addition, intermediate biomarkers selected for chronic disease indicators include insulin resistance, plasma renin activity, and BP.

Indicators of sodium toxicity

The UL is defined as the highest intake level at which there is no risk of any adverse health effect in apparently healthy individuals. To set the UL for dietary sodium in the US and Canada, the lowest toxic sodium level was calculated on the basis of an increase in BP due to sodium intake, as the toxicity endpoint in the 2005 DRIs [8]. However, the approach to establish the sodium UL in the report of 2019 jointly developed by the US and Canada differs in that health effects of excessive sodium intakes on BP, stroke and coronary heart need to be considered when establishing the sodium CDRR. Under the synthesis of evidence for the 2020 KDRIs, sodium toxicity should be considered based on toxicity, and not on chronic disease endpoint related indicators.

Chronic disease indicators to establish sodium CDRR

While the sodium UL is established based on the toxic effect derived from excessive intake of sodium, the CDRR indicates the decrease in the risks of chronic diseases with reductions in dietary sodium intake [8]. For the 2020 KDRIs, the committee reviewed the causal relationship between sodium intake and indicators that could potentially inform the sodium CDRRs, including chronic disease endpoints and surrogate markers. The scientific evidence and conclusions previously accumulated in the 2015 KDRIs was also considered for this deliberation. To review the causal relationship between sodium intake and chronic disease, we finally selected CVD (cerebrovascular disease and coronary heart disease), hypertension, cancer (gastric cancer and renal cancer), osteoporosis, obesity, diabetes, and chronic kidney disease as potential health outcomes, and renin-angiotensin-aldosterone system (RAAS), high BP, bone mineral density, bone mineral content, body mass index (BMI), and glomerular injury as surrogate markers (Fig. 1). Evidence from these indicators can be accumulated to inform the development of a sodium CDRR. More specifically, if there is moderate to high SoE for a causal relationship and an intake–response relationship between sodium intake and chronic disease indicators, a sodium CDRR would be established [11].

Fig. 1

Indicators of exposure and health outcomes in a systematic review on the causal relationship between sodium intake and chronic disease for establishing the sodium Chronic Disease Risk Reduction Intake level.

INTAKE–RESPONSE RELATIONSHIP AND SPECIFICATION OF KDRI VALUES

The EAR, RNI, and AI for sodium adequacy

In general, the EAR and RNI are established based on available sufficient scientific evidence for nutrient requirements, in the absence of which the AI is established [12]. Due to insufficient scientific evidence of the intake–response relationship for selected adequacy indicators to establish the EAR and RNI in the 2020 KDRIs (which was similar to the 2015 KDRIs), the committee established the AI of dietary sodium for the 2020 KDRIs. It is generally accepted that AI can be established using the average sodium intake of a healthy population [8] when it is not possible to establish the EAR and RNI. However, the average daily intake of dietary sodium for Koreans was determined to be 3,287 mg according to the 2019 Korea National Health and Nutrition Examination Survey (KNHANES) [13], which is too high to be considered as the sodium AI value. Therefore, the AI for sodium in the 2020 KDRIs was established based on the results of balance studies involving healthy subjects and evidence of the adverse physiological effects of low sodium consumption, such as toxic effects or the occurrence of various diseases [8]. It was established that a diet containing approximately 1,500 mg/day of sodium met the recommended intakes for other nutrients [14]. Increase in plasma levels of total cholesterol was found in the range 460–690 mg/day of sodium intake [15]. When considering the association of sodium intake with insulin resistance, sodium intake lower than 700 mg/day increased insulin resistance [16], while a sodium intake of ≥ 1,200 mg/day was not associated with insulin resistance in non-obese and normotensive men [17]. In addition, evidence from a well-designed balance study conducted among adults [18] reported that sodium intake of 1,525 mg/day among adult men was equivalent to the excreted sodium. Taken together, the appropriateness of the sodium AI value of 1,500 mg/day for adults aged 19–64 years in the 2015 KDRIs was reviewed, and it was decided to recommend this value in the 2020 KDRIs. Since there was no available evidence enough to establish the sodium AI for other age groups, the sodium AI for infants, children, adolescents, and the elderly was calculated by extrapolating from the adult sodium AI value, based on the energy intake involved, as follows [19]:

For males aged 6–8 years, 9–11 years, 12–64 years, 65–74 years, and ≥ 75 years, the sodium AI values were established at 1,100 mg/day, 1,300 mg/day, 1,500 mg/day, 1,300 mg/day, and 1,100 mg/day, respectively. For females aged 6–8 years, 9–64 years, 65–74 years, and ≥ 75 years, the sodium AI values were established at 1,400 mg/day, 1,500 mg/day, 1,400 mg/day, and 1,100 mg/day, respectively. For children aged 1–2 years and 3–5 years, the AI for sodium was 800 mg/day and 1,000 mg/day, respectively, for both genders. The sodium AI for infants was estimated based on the sodium intakes from breastfeeding, and were determined to be 110 mg/day for infants aged 0–5 months, and 370 mg/day for infants aged 6–11 months (Supplementary Table 1).

The UL for sodium toxicity

In the 2020 KDRIs, no potential indicators for adverse toxicological effects of sodium were identified subsequent to reviewing the available evidence after excluding the association of sodium intake with chronic disease related indicators. Lethal levels of dietary sodium, primarily due to the ingestion of acute doses, have been previously reported [202122]. However, they were unsuitable for establishing a sodium UL since the levels did not necessarily reflect the toxicological effects of high sodium intake in our habitual diets. Furthermore, several sodium studies reported that sodium intakes affect the occurrence of headaches [232425]. However, since the evidence for sodium toxicity was insufficient to determine sodium toxicity risk, the UL for sodium was not established in the 2020 KDRIs. In fact, sodium UL for Koreans has been replaced by intake goal, set at 2,000 mg/day for adults since 2015 KDRIs. The sodium CDRR for reducing the risk of chronic disease is now included as a new category in the 2020 KDRIs; hence, the intake goal of dietary sodium could not be identified in the 2020 KDRIs.

The CDRR for sodium based on chronic disease risk reduction

A systematic literature review was thoroughly conducted to identify scientific evidence of the potential detrimental health effects of excessive intake of dietary sodium. The methods and results of the review are as follows.

Selection of studies for the systematic review

Randomized controlled trials, cohort studies, nested case-control studies, case-control studies, and cross-sectional studies on the relationship between sodium and the risk of several chronic diseases, published between January 2014 and December 2018, were identified using the PubMed (https://pubmed.ncbi.nlm.nih.gov) and Web of Science (https://www.webofknowledge.com) databases. The search keywords used were “salt,” “sodium intake,” “dietary sodium,” and “urinary sodium” for the independent variables, and “CVD,” “cerebrovascular disease,” “coronary heart disease,” “hypertension,” “BP,” “RAAS,” “obesity,” “BMI,” “osteoporosis,” “bone mineral density,” “bone mineral content,” “osteoporotic fracture,” “diabetes,” “chronic kidney disease,” “proteinuria,” “gastric cancer,” and “renal cancer” for dependent variables. Studies written in both Korean and English were included for this review.

Inclusion/exclusion criteria

Articles were extracted based on their titles and abstracts. The inclusion and exclusion criteria were as follows: 1) inclusion of only human studies, 2) in cases of multiple publications involving the same study population, only the most recent study was included, 3) reviews or meta-analyses were excluded. Using these criteria, the studies were reviewed independently by 2 reviewers, and initial disagreements between the reviewers were resolved by eventual consensus between them.

Data collection

The following information was collected from each study: author name, publication year, country, diseases examined, study design, study subjects (health status, life stage, age, and sex), confounding or effect-modifying variables, measures of sodium (dietary sodium intake or urinary sodium excretion), category of sodium intake, outcomes (level of incidence, and mortality and controls), adjusted relative risk/odds ratio with 95% confidence intervals, and trends (P).

Quality evaluation of the study

The quality of each study was evaluated using the Cochrane RoB 2.0 template and the STROBE checklist for intervention and cross-sectional studies, and the Newcastle-Ottawa scale for cohort, nested case-control, and case-control studies. Using these tools, the risk of bias (RoB) level was classified as “low,” “some concerns,” or “high” for each study. Moreover, the overall SoE was classified into grades I (strong), II (moderate), III (limited), and IV (grade not assignable), based on the criteria employed by the US Department of Agriculture’s Nutrition Evidence Library. These grading criteria include quality in relation to RoB, quantity (number of studies and participants), consistency of findings across studies, public health impact (directness and magnitude), and generalizability (references). To establish the sodium CDRR, the SoE for both the causal relationship between sodium intake and chronic disease indicators and their intake–response relationship should be at least moderate, using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) system [11].

Results of the systematic literature review

A total of 1,033 articles were identified via the initial computerized literature search. Based on the titles and abstracts of the articles, 248 articles were selected independently by the 2 reviewers. Randomized controlled trials, cohort studies, nested case-control studies, case-control studies, and cross-sectional studies were selected, whereas reviews or meta-analyses were excluded. Moreover, multiple publications that included the same population, insufficient study population, or unhealthy subjects, were excluded. Finally, through a full text review of the remaining articles, 43 studies (7 randomized controlled trials, 18 cohort studies, 3 nested case-control studies, and 15 cross-sectional study) were included for evaluating the RoB and SoE of the harmful effects of sodium intake levels on several chronic diseases (Fig. 2 and Table 1). Details of the eligible studies for the systematic review are presented in Tables 2 and 3, Supplementary Tables 2 and 3. We found that SoE of the relationship between dietary sodium and CVD, cerebrovascular disease, hypertension, and their surrogate marker systolic/diastolic BP, was at least moderate. Specifically, there was strong evidence for BP lowering the effect of sodium. Furthermore, there was a moderate SoE for the intake–response relationship between sodium intake and the risk of CVD or cerebrovascular disease. However, evidence associating sodium intake with the risk of coronary heart disease, cancer, bone mineral density, RAAS, and obesity were insufficient (limited or grade not assignable) (Table 4).

Fig. 2

Study selection process for the systematic review.

Table 1

List of studies included in the systemic literature review

Study design	Author	Year	Title
RCT-parallel	Jenkins et al. [38]	2015	The effect of a dietary portfolio compared to a DASH-type diet on blood pressure
	Diaz et al. [39]	2014	The effects of weight loss and salt reduction on visit-to-visit blood pressure variability: results from a multicenter randomized controlled trial
	Reidlinger et al. [40]	2015	How effective are current dietary guidelines for cardiovascular disease prevention in healthy middle-aged and older men and women? A randomized controlled trial
	Zhou et al. [41]	2016	Intake of low sodium salt substitute for 3years attenuates the increase in blood pressure in a rural population of North China - A randomized controlled trial
	Juraschek et al. [42]	2017	Effects of sodium reduction and the DASH diet in relation to baseline blood pressure
RCT-crossover	Muth et al. [43]	2017	Central systolic blood pressure and aortic stiffness response to dietary sodium in young and middle-aged adults
	Nielsen et al. [44]	2016	Changes in the renin-angiotensin-aldosterone system in response to dietary salt intake in normal and hypertensive pregnancy. A randomized trial
Cohort	Merino et al. [45]	2015	Is complying with the recommendations of sodium intake beneficial for health in individuals at high cardiovascular risk? Findings from the PREDIMED study
	Kalogeropoulos et al. [46]	2015	Dietary sodium content, mortality, and risk for cardiovascular events in older adults: the Health, Aging, and Body Composition (Health ABC) Study
	Okayama et al. [47]	2016	Dietary sodium-to-potassium ratio as a risk factor for stroke, cardiovascular disease and all-cause mortality in Japan: the NIPPON DATA80 cohort study
	Mente et al. [48]	2018	Urinary sodium excretion, blood pressure, cardiovascular disease, and mortality: a community-level prospective epidemiological cohort study
	Cook et al. [49]	2014	Lower levels of sodium intake and reduced cardiovascular risk
	Willey et al. [36]	2017	Dietary sodium to potassium ratio and risk of stroke in a multiethnic urban population: the Northern Manhattan Study
	Prentice et al. [50]	2017	Associations of biomarker-calibrated sodium and potassium intakes with cardiovascular disease risk among postmenopausal women
	Li et al. [51]	2018	Longitudinal change of perceived salt intake and stroke risk in a Chinese population
	Joosten et al. [52]	2014	Sodium excretion and risk of developing coronary heart disease
	Voortman et al. [53]	2017	Adherence to the 2015 Dutch dietary guidelines and risk of non-communicable diseases and mortality in the Rotterdam Study
	Takase et al. [54]	2015	Dietary sodium consumption predicts future blood pressure and incident hypertension in the Japanese normotensive general population
	Buendia et al. [55]	2015	Longitudinal effects of dietary sodium and potassium on blood pressure in adolescent girls
	Timpka et al. [56]	2017	Lifestyle in progression from hypertensive disorders of pregnancy to chronic hypertension in Nurses’ Health Study II: observational cohort study
	Bertoia et al. [57]	2014	Mediterranean and Dietary Approaches to Stop Hypertension dietary patterns and risk of sudden cardiac death in postmenopausal women
	Horikawa et al. [58]	2014	Dietary sodium intake and incidence of diabetes complications in Japanese patients with type 2 diabetes: analysis of the Japan Diabetes Complications Study (JDCS)
	Umesawa et al. [59]	2016	Salty food preference and intake and risk of gastric cancer: The JACC Study
	Wang et al. [60]	2017	Composite protective lifestyle factors and risk of developing gastric adenocarcinoma: the Singapore Chinese Health Study
	Carbone et al. [61]	2016	Sodium intake and osteoporosis. Findings from the Women’s Health Initiative
Nested case control	Lee et al. [62]	2018	Hyponatraemia and its prognosis in acute heart failure is related to right ventricular dysfunction
	Deckers et al. [63]	2014	Long-term dietary sodium, potassium and fluid intake; exploring potential novel risk factors for renal cell cancer in the Netherlands Cohort Study on diet and cancer
	Deckers et al. [64]	2017	Promoter CpG island methylation in ion transport mechanisms and associated dietary intakes jointly influence the risk of clear-cell renal cell cancer
Cross sectional	Kim et al. [65]	2014	The relationship of dietary sodium, potassium, fruits, and vegetables intake with blood pressure among Korean adults aged 40 and older
	Tabara et al. [66]	2015	Descriptive epidemiology of spot urine sodium-to-potassium ratio clarified close relationship with blood pressure level
	Noh et al. [67]	2015	Association between high blood pressure and intakes of sodium and potassium among Korean adults: Korean National Health and Nutrition Examination Survey, 2007–2012
	Xu et al. [68]	2014	Estimation of salt intake by 24-hour urinary sodium excretion: a cross-sectional study in Yantai, China
	Hu et al. [69]	2017	Prevalence, awareness, treatment, and control of hypertension among Kazakhs with high salt intake in Xinjiang, China: a community-based cross-sectional study
	Park et al. [70]	2016	The effect of the sodium to potassium ratio on hypertension prevalence: a propensity score matching approach
	Navia et al. [71]	2014	Sodium intake may promote weight gain; results of the FANPE study in a representative sample of the adult Spanish population
	Ge et al. [72]	2016	Are 24 h urinary sodium excretion and sodium:potassium independently associated with obesity in Chinese adults?
	Huh et al. [73]	2015	Gender-specific association between urinary sodium excretion and body composition: analysis of the 2008-2010 Korean National Health and Nutrition Examination Surveys
	Murakami et al. [74]	2015	Ability of self-reported estimates of dietary sodium, potassium and protein to detect an association with general and abdominal obesity: comparison with the estimates derived from 24 h urinary excretion
	Ma et al. [75]	2015	High salt intake: independent risk factor for obesity?
	Grimes et al. [76]	2016	24-h urinary sodium excretion is associated with obesity in a cross-sectional sample of Australian schoolchildren
	Oh et al. [77]	2017	Associations of sodium intake with obesity, metabolic disorder, and albuminuria according to age
	Nam et al. [78]	2017	Association between 24-h urinary sodium excretion and obesity in Korean adults: a multicenter study
	Zhang et al. [79]	2018	A positive association between dietary sodium intake and obesity and central obesity: results from the National Health and Nutrition Examination Survey 1999–2006

RCT, randomized controlled trial.

Table 2

Association between sodium levels and the risk of chronic diseases (event outcomes)

Author (Year), Nation		Study design	Study subjects			Sodium levels		No. of cases/No. of category	RR/OR (95% CI)	P for trend	Confounding variables considered	RoB
			Source of subjects, Mean age (yrs)/Sex (%)	Outcome (incidence or death rate)	Follow-up duration	Measurement unit	Categories
Cardiovascular disease
	Merino et al. (2015) [45], Spain	Cohort	Adults, NA (50–88 yrs)/Males 42.0%	Incidence (125/3,982; 0.031)	1 yr	Intake amounts	Decrease in sodium intake	33/1,199	0.66 (0.38, 1.15)	0.040	Demographic, anthropic, and medical factors, lifestyles, and foods intake	Low
							No change in sodium intake	62/2,016	1.00
							Increase in sodium intake	30/767	1.72 (1.01, 2.91)
	Kalogeropoulos et al. (2015) [46], US	Cohort	Adults, 73.6 (70–79 yrs)/Males 48.8%	Incidence (572/1,981; 0.289)	10 yrs	Intake amounts	< 1,500 mg/day	63/217	1.02 (0.79, 1.41)	0.470	Demographic, anthropic, and medical factors, lifestyles	Low
							1,500–2,300 mg/day	161/576	1.00
							> 2,300mg/day	348/1,188	1.02 (0.84, 1.24)
	Okayama et al. (2016) [47], Japan	Cohort	Adults, NA (30–79 yrs)/Males 44.5%	Death rate (579/8,283; 0.070)	24 yrs	Intake dietary Na-K ratio	1st quintile	110/1,581	1.00	0.005	Demographic, anthropic, and medical factors, lifestyles, and foods intake	Low
							2nd quintile	114/1,652	NA
							3rd quintile	100/1,686	NA
							4th quintile	113/1,684	NA
							5th quintile	142/1,681	1.39 (1.20, 1.61)
	Mente et al. (2018) [48], 18 countries	Cohort	Adults, 50.4 (35–70 yrs)/Males 42.1%	Incidence (NA)	8.1 yrs	Intake amounts	1st tertile	NA	−1.00 (−2.00, −0.01)	< 0.001	Demographic, anthropic, and medical factors, lifestyles	Low
							2nd tertile	NA	0.24 (−2.12, 2.61)
							3rd tertile	NA	0.37 (−0.03, 0.78)
	Cook et al. (2014) [49], US	Cohort	Prehypertensive adults, NA (30–54 yrs)/Males 69.5%	Incidence (193/2,312; 0.083)	10–15 yrs	24-h urine	< 2,300 mg/24 h	17/236	0.68 (0.34, 1.37)	0.130	Demographic, anthropic, and medical factors, lifestyles, and foods intake	Low
							2,300–< 3,600 mg/24 h	61/893	0.75 (0.50, 1.11)
							3,600–< 4,800 mg/24hr	74/768	1.00
							> 4,800mg/24hr	41/415	1.05 (0.68, 1.62)
	Prentice et al. (2017) [50], US	Cohort	Postmenopausal women, NA (50–79 yrs)/Female 100%	Incidence (5,897/86,444; 0.068)	12 yrs	Intake	20% increase in intake	5,897/86,444	1.06 (0.92, 1.23)	NA	Demographic, and medical factors, lifestyles	Low
	Bertoia et al. (2014) [57], US	Cohort	Postmenopausal women, NA (50–79 yrs)/Female 100%	Death rate (237/93,122; 0.003)	10.5 yrs	Intake	1st quintile	52/18,465	1.00	0.460	Demographic, anthropic, and medical factors, lifestyles, and foods intake	Low
						DASH diet pattern score	2nd quintile	56/18,216	1.09 (0.75, 1.60)
							3rd quintile	57/20,220	1.11 (0.75, 1.63)
							4th quintile	41/17,808	0.95 (0.62, 1.45)
							5th quintile	31/18,413	0.86 (0.54, 1.38)
	Horikawa et al. (2014) [58], Japan	Cohort	T2DM adults, NA (40–70 yrs)/Males 47.5%	Incidence (132/1,414; 0.09)	8 yrs	Intake amounts	1st quartile	23/354	1.00	0.030	Demographic, anthropic, and medical factors, lifestyles, and foods intake	Low
							2nd quartile	36/350	1.70 (0.98, 2.93)
							3rd quartile	32/351	1.47 (0.82, 2.62)
							4th quartile	41/359	2.07 (1.16, 3.71)
	Lee et al. (2018) [62], Korea	Nested case-control	Adults, 67.8 ± 14.9 yrs/Males 55.7%	Prevalence (NA)	2 yrs	Serum level	With hyponatraemia	94/116	1.00	< 0.001	Baseline characteristics	Low
							Without hyponatraemia	78/232	8.00 (4.50, 14.22)
Cerebrovascular disease
	Okayama et al. (2016) [47], Japan	Cohort	Adults, NA (30–79 yrs)/Males 44.5%	Death rate (273/8,283; 0.033)	24 yrs	Intake dietary Na-K ratio	1st quintile	45/1,581	1.00	0.002	Demographic, anthropic, and medical factors, lifestyles, and foods intake	Low
							2nd quintile	46/1,652	NA
							3rd quintile	55/1,686	NA
							4th quintile	53/1,684	NA
							5th quintile	74/1,681	1.43 (1.17, 1.76)
	Mente et al. (2018) [48], 18 countries	Cohort	Adults, 50.4 (35–70 yrs)/Males 42.1%	Incidence (NA)	8.1 yrs	Intake amounts	1st tertile	NA	−0.12 (−0.67, 0.43)	< 0.001	Demographic, anthropic, and medical factors, lifestyles	Low
							2nd tertile	NA	0.28 (−1.26, 1.82)
							3rd tertile	NA	0.54 (0.12, 0.96)
	Willey et al. (2017) [36], US	Cohort	Adults, 69 ± 10 yrs/Males 36%	Incidence (274/2,496; 0.110)	12 yrs	Intake dietary Na-K ratio	Increase in Na-K ratio	274/2,496	1.58 (1.20, 2.06)	NA	Demographic, anthropic, and medical factors, lifestyles, and foods intake	Low
	Prentice et al. (2017) [50], US	Cohort	Postmenopausal women, NA (50–79 yrs)/Female 100%	Incidence (2,843/86,444; 0.033)	12 yrs	Intake	20% increase in intake	2,843/86,444	0.98 (0.85, 1.13)	NA	Demographic, and medical factors, lifestyles	Low
	Li et al. (2018) [51], China	Cohort	Adults, 53.6 ± 12 yrs/Males 77.8%	Incidence (1,564/79,490; 0.020)	5 yrs	Intake changing pattern	Moderate-stable	1,225/59,241	1.00	NA	Demographic, anthropic, and medical factors, lifestyles	Low
							Moderate-decreasing	141/9,268	0.77 (0.64, 0.91)
							Moderate-increasing	72/2,975	1.04 (0.82, 1.32)
							Low-increasing	54/2,879	0.92 (0.70, 1.22)
							High-decreasing	72/3,242	1.01 (0.79, 1.28)
	Voortman et al. (2017) [53], Netherlands	Cohort	Adults, 64.1 (49–83 yrs)/Males 41.9%	Incidence (979/29,442; 0.104)	10.2 yrs	Intake Dutch dietary guidelines	1st quintile	NA	1.00	0.520	Demographic, and anthropic factors, lifestyles, and foods intake	Low
							2nd quintile	NA	0.93 (0.78, 1.12)
							3rd quintile	NA	0.88 (0.73, 1.06)
							4th quintile	NA	0.97 (0.80, 1.17)
							5th quintile	NA	0.92 (0.75, 1.13)
Hypertension
	Takase et al. (2015) [54], Japan	Cohort	Adults, 54.1 ± 10.9 yrs/Males 64.2%	Incidence (1,027/4,523; 0.227)	3.1 yrs	Intake amounts	Lower intake	NA	1.00	< 0.001	Demographic, anthropic, and medical factors, and lifestyles	Low
							Higher intake	NA	1.25 (1.04, 1.50)
	Timpka et al. (2017) [56], US	Cohort	Adults, NA (32–39 yrs)/Females 100%	Incidence (572/90,887 PY)	NA	Intake dietary Na-K ratio	1st quartile	NA	1.00	0.650	Demographic, and anthropic factors, lifestyles, and foods intake	Low
							2nd quartile	NA	1.07 (0.82, 1.40)
							3rd quartile	NA	0.98 (0.75, 1.27)
							4th quartile	NA	1.07 (0.83, 1.38)
			Adults, NA (40–49 yrs)/Females 100%	Incidence (5,716/334,976 PY)	NA	Intake dietary Na-K ratio	1st quartile	NA	1.00	0.030
							2nd quartile	NA	1.04 (0.96, 1.13)
							3rd quartile	NA	1.10 (1.02, 1.19)
							4th quartile	NA	1.09 (1.00, 1.18)
			Adults, NA (50–59 yrs)/Females 100%	Incidence (5,366/20,207 PY)	NA	Intake dietary Na-K ratio	1st quartile	NA	1.00	0.006
							2nd quartile	NA	1.07 (0.99, 1.15)
							3rd quartile	NA	1.14 (1.05, 1.23)
							4th quartile	NA	1.11 (1.02, 1.20)
	Noh et al. (2015) [67], Korea	Cross-sectional	Adults, NA (> 19 yrs)/Males 50.3%	Prevalence (2,812/24,096; 0.120)	-	Intake combinations of Na and K intakes	Low Na/High K	9.50%/4,516	1.00	< 0.001	Demographic, and anthropic factors, lifestyles, and foods intake	Low
							High Na/High K	10.30%/7,532	0.99 (0.84, 1.18)
							Low Na/Low K	11.80%/7,532	1.19 (1.01, 1.40)
							High Na/Low K	12.40%/4,516	1.21 (1.02, 1.44)
	Hu et al. (2017) [69], China	Cross-sectional	Adults, 46.5 (> 30 yrs)/Males 46.7%	Prevalence (NA/1,668; 0.455)	-	Intake amounts	1st quartile	NA	1.00	< 0.001	Demographic factors	Low
							2nd quartile	NA	NA
							3rd quartile	NA	NA
							4th quartile	NA	1.74 (1.26, 2.39)
	Park et al. (2016) [70], Korea	Cross-sectional	Adults, 46.1 (20–79 yrs)/Males 31.8%	Prevalence (NA/30,206; 0.196)	-	Intake dietary Na-K ratio	1st quartile	19.27%/2,356	0.00	NA	Propensity score matching	Low
							2nd quartile	18.00%/2,356	1.02% point
							3rd quartile	19.44%/2,356	2.74% point
							4th quartile	21.52%/2,356	3.44% point

A moderate to strong strength of evidence was determined for the relationship between sodium and cardiovascular disease, cerebrovascular disease, and hypertension.

RR, relative risk; OR, odds ratio; CI, confidence interval; RoB, risk of bias; NA, not available; PY, person years; T2DM, type 2 diabetes mellitus.

Table 3

Association between sodium levels and the risk of chronic diseases (continuous outcomes): hypertension

Author (Year), Nation	Study design	Study subjects			Sodium levels			No. of cases/No. of category	Mean (95% CI)	P-values	Confounding variables considered	RoB
		Source of subjects, Mean age (yrs)/Sex (%)	Outcome	Follow-up duration	Measurement unit	Categories
Jenkins et al. (2015) [38], Canada	RCT	Hyperlipidemia adults, NA (20–85 yrs)/Males 39.0%	Mean arterial pressure	24 wks	Intake	Dietary portfolio		159	−2.1 (−3.0, −1.3)	0.026	-	Low
						DASH-type diet		82	−0.3 (−1.5, 1.0)
			Systolic blood pressure			Dietary portfolio		159	−2.5 (−3.7, −1.2)	0.045
						DASH-type diet		82	−0.4 (−2.1, 1.4)
			Diastolic blood pressure			Dietary portfolio		159	−2.0 (−2.8, −1.2)	0.026
						DASH-type diet		82	−0.2. (−1.4, 0.9)
Diaz et al. (2014) [39], US	RCT	Overweight adults, 43.9 ± 6.1 yrs/Males 66.0%	Visit-to-visit blood pressure	36 mon	Intake	Sodium light lifestyle		452	7.1 ± 3.0	0.290	-	Low
						Usual care control		463	6.9 ± 2.9
Reidlinger et al. (2015) [40], UK	RCT	Adults, 53 (40–70 yrs)/Males 39.0%	Systolic blood pressure	12 wks	Intake	Restriction on salt and sugar		80	−4.1 (NA)	0.003	-	Low
						Control diet		82	−0.5 (NA)
			Diastolic blood pressure			Restriction on salt and sugar		80	−2.9 (NA)	0.002
						Control diet		82	−0.2 (NA)
Zhou et al. (2016) [41], China	RCT	Families, 46.4 ± 13.6 yrs/Males 49.0%	Systolic blood pressure	3 yrs	Intake	Low salt		224	−8.9 (NA)	NA	-	Low
						Normal salt		238	−5.8 (NA)
			Diastolic blood pressure			Low salt		224	−4.7 (NA)	NA
						Normal salt		238	−2.4 (NA)
Juraschek et al. (2017) [42], US	RCT	Pre- or stage 1 hypertension adults, 49.1 ± 10.4 yrs/Males 43.0%	Systolic blood pressure	4 wks	Intake	DASH diet		204	−10.4 (−15.5, −5.3)	0.020	-	Low
						Control diet		200	−7.0 (−12.9, −1.2)
Muth et al. (2017) [43], US	RCT	Young adults, 27.0 ± 1.0 yrs	Systolic blood pressure	14 days	Intake	Low sodium diet		85	Young adults: −4.0 (NA)	0.012	-	Some concerns
		Middle-aged adults, 52.0 ± 1.0 yrs/Males 51.0%							Middle-aged: −9.0 (NA)	< 0.001
						High sodium diet		85	Young adults: NA	NA
									Middle-aged: NA	NA
Buendia et al. (2015) [55], US	Cohort	Adolescent girl, NA (9–10 yrs)/Females 100%	Systolic blood pressure	10 yrs	Intake dietary Na-K ratio	< 2,000 mg/day		425	108.8 (108.0, 109.7)	0.550	Demographic, and anthropic factors, lifestyles, and foods intake	Low
						2,500–3,000 mg/day		644	109.3 (108.3, 109.6)
						3,000–4,000 mg/day		905	108.9 (108.6, 109.7)
						> 4,000 mg/day		211	108.1 (108.0, 110.4)
			Diastolic blood pressure			< 2,000 mg/day		425	65.6 (64.8, 66.4)	0.560
						2,500–3,000 mg/day		644	65.6 (64.9, 66.2)
						3,000–4,000 mg/day		905	65.5 (65.0, 66.0)
						> 4,000 mg/day		211	64.9 (63.8, 66.1)
Kim et al. (2014) [65], Korea	Cross-sectional	Adults, Males 61.5 yrs, Females 59.7 yrs/Males 38.9%	Systolic blood pressure	-	Intake amounts	Male					Demographic, and anthropic factors, lifestyles, and foods intake	Low
							1st quintile	488	124.6 (123.0, 126.1)	0.600
							2nd quintile	489	124.4 (123.0, 125.9)
							3rd quintile	489	124.3 (122.8, 125.7)
							4th quintile	489	125.5 (124.1, 127.0)
							5th quintile	488	125.7 (124.1, 127.3)
						Female
							1st quintile	768	121.5 (120.2, 122.8)	0.610
							2nd quintile	768	122.5 (121.3, 123.7)
							3rd quintile	768	121.9 (120.7, 123.1)
							4th quintile	768	121.7 (120.5, 123.0)
							5th quintile	768	121.2 (119.7, 122.4)
			Diastolic blood pressure			Male
							1st quintile	488	78.8 (77.8, 79.7)	0.020
							2nd quintile	489	79.7 (78.8, 80.6)
							3rd quintile	489	80.5 (79.6, 81.4)
							4th quintile	489	80.7 (79.9, 81.6)
							5th quintile	488	80.6 (79.7, 81.6)
						Female
							1st quintile	768	76.8 (75.9, 77.3)	0.005
							2nd quintile	768	78.3 (77.5, 79.0)
							3rd quintile	768	77.4 (76.7, 78.1)
							4th quintile	768	78.1 (77.4, 78.9)
							5th quintile	768	78.4 (77.7, 79.1)
Tabara et al. (2015) [66], Japan	Cross-sectional	Adults, 54 (30–74 yrs)/Males 52.2%	Blood pressure	-	Spot urine	Linear increase in urinary Na-K ratio		9,144	0.112 (NA)	< 0.001	Demographic, anthropic, and medical factors, lifestyles	Low
Noh et al. (2015) [67], Korea	Cross-sectional	Adults, NA (> 19 yrs)/Males 50.3%	Systolic blood pressure	-	Intake combinations of Na and K intakes	Low Na/High K		4,516	113.0 ± 0.30	< 0.001	Demographic, and anthropic factors, lifestyles, and foods intake	Low
						High Na/High K		7,532	113.7 ± 0.22
						Low Na/Low K		7,532	114.2 ± 0.24
						High Na/Low K		4,516	114.8 ± 0.31
			Diastolic blood pressure		Intake combinations of Na and K intakes	Low Na/High K		4,516	73.9 ± 0.21	< 0.001
						High Na/High K		7,532	75.0 ± 0.17
						Low Na/Low K		7,532	74.6 ± 0.17
						High Na/Low K		4,516	75.8 ± 0.22
Xu et al. (2014) [68], China	Cross-sectional	Adults, 42.3 (18–69 yrs)/Males NA	Systolic blood pressure	-	24-h urine	Linear increase in urinary sodium excretion		191	0.16 (NA)	0.010	Demographic, and anthropic factors	Low
			Diastolic blood pressure			Linear increase in urinary sodium excretion		191	0.12 (NA)	0.060
Park et al. (2016) [70], Korea	Cross-sectional	Adults, 46.1 (20–79 yrs)/Males 31.8%	Average treatment effects on systolic blood pressure	-	Intake dietary Na-K ratio	1st quartile		2,356	0.00	NA	-	Low
						2nd quartile		2,356	1.10 ± 0.51
						3rd quartile		2,356	0.90 ± 0.52
						4th quartile		2,356	1.40 ± 0.54
			Average treatment effects on diastolic blood pressure			1st quartile		2,356	0.00
						2nd quartile		2,356	0.80 ± 0.32
						3rd quartile		2,356	0.20 ± 0.33
						4th quartile		2,356	0.90 ± 0.34

Moderate to strong strength of evidence was determined for the relationship between sodium and hypertension.

CI, confidence interval; RoB, risk of bias; NA, not available; RCT, randomized controlled trial.

Table 4

Level of risk of bias and strength of evidence for the relationship between sodium intakes and chronic diseases

Health outcomes	Authors (Year)	RoB	Overall SoE
Cardiovascular disease	Bertoia et al. (2014) [57], Cook et al. (2014) [49], Horikawa et al. (2014) [58], Kalogeropoulos et al. (2015) [46], Okayama et al. (2016) [47], Merino et al. (2015) [45], Prentice et al. (2017) [50], Lee et al. (2018) [62], Mente et al. (2018) [48]	Low: 8	Moderate
Cerebrovascular disease	Okayama et al. (2016) [47], Prentice et al. (2017) [50], Voortman et al. (2017) [53], Willey et al. (2017) [36], Li et al. (2018) [51], Mente et al. (2018) [48]	Low: 6	Moderate
Coronary heart disease	Joosten et al. (2014) [52], Kalogeropoulos et al. (2015) [46], Prentice et al. (2017) [50], Voortman et al. (2017) [53]	Low: 4	Limited
Hypertension	Diaz et al. (2014) [39], Kim et al. (2014) [65], Xu et al. (2014) [68], Buendia et al. (2015) [55], Jenkins et al. (2015) [38], Noh et al. (2015) [67], Reidlinger et al. (2015) [40], Tabara et al. (2015) [66], Takase et al. (2015) [54], Timpka et al. (2015) [56], Park et al. (2016) [70], Zhou et al. (2016) [41], Hu et al. (2017) [69], Muth et al. (2017) [43], Juraschek et al. (2017) [42]	Low: 14	Strong
		Some concerns: 1
Gastric cancer	Umesawa et al. (2016) [59], Wang et al. (2017) [60]	Low: 2	Limited
Renal cancer	Deckers et al. (2014) [63], Deckers et al. (2017) [64]	Low: 2	Limited
Bone mineral density	Carbone et al. (2016) [61]	High: 1	Grade not assignable
RAAS	Nielsen et al. (2016) [44]	Some concerns: 1	Grade not assignable
Obesity	Navia et al. (2014) [71], Ge et al. (2016) [72], Huh et al. (2015) [73], Ma et al. (2015) [75], Murakami et al. (2015) [74], Grimes et al. (2016) [76], Nam et al. (2017) [78], Oh et al. (2017) [77], Zhang et al. (2018) [79]	Low: 8	Limited
		Some concerns: 1

RoB, risk of bias; SoE, strength of evidence; RAAS, renin-angiotensin-aldosterone system.

In a meta-analysis of 39 randomized clinical trials on the intake–response, there was a strong SoE that sodium reduction decreased the risk of CVD, hypertension, and systolic/diastolic BP in subjects with a sodium intake of 2,300–4,100 mg/day. In addition, there was a moderate SoE that sodium reduction in subjects with a sodium intake of 4,100–5,000 mg/day decreased the effect of systolic/diastolic BP. However, the SoE was weak for intake levels ≤ 2,300 mg/day [8]. Taken together, we concluded that there is a moderate-to-high SoE for reduced sodium intake decreasing the risk of CVD, hypertension, and BP in persons with sodium intake levels of 2,300–5,000 mg/day. Accordingly, the DRIs jointly developed by US and Canada set 2,300 mg/day as the CDRR of sodium for adults [8]. Based on a systematic review and the CDRR established in the DRIs for US and Canada, the 2020 KDRIs established the CDRR for sodium at 2,300 mg/day for Koreans aged 19–64 years. Moreover, the CDRR for sodium according to age, sex, and life-stage were extrapolated from the adult values by considering the energy intake involved, as follows [19]:

The sodium CDRR for children aged 1–2 years and 3–5 years was established as 1,200 mg/day and 1,600 mg/day, respectively, for both genders. The CDRR for children aged 6–8 years and 9–11 years was set at 1,800 mg/day and 2,000 mg/day for males, and 2,100 mg/day and 2,300 mg/day for females, respectively, and for adolescents aged 12–18 years the CDRR was set at 2,300 mg for both genders. For the elderly aged 65–74 years and ≥ 75 years, the sodium CDRR values were established as 1,900 mg/day and 1,800 mg/day for males, and 2,200 mg/day and 1,700 mg/day for females, respectively (Supplementary Table 1).

INTAKE ASSESSMENT TO CHARACTERIZE RISK

Currently, majority of the populations are reported to consume dietary sodium above the recommended WHO levels [7], while countries consuming less than 2,000 mg of sodium per day are the poor with malnutrition problems [26]. Especially in Asian countries, the mean sodium intakes are higher than 4,600 mg/day [27]. According to the 2019 KNHANES, the average daily sodium intake was 3,287 mg/day (males, 3,851 mg/day; females, 2,699 mg/day) in South Korea, and the percentage of excessive sodium intake of ≥ 2,000 mg/day was 74.0% (84.5% for males and 63.5% for females) among Koreans aged ≥ 9 years [28]. In both developed and developing countries, including South Korea, where various processed foods produced by the modern food supply system are consumed, the feasibility of achieving a sodium intake goal of 2,000 mg/day is constantly questioned. Therefore, it is necessary to set a flexible goal for sodium intake for Koreans, based on the eating habits and food system in this population. Considering the food sources of dietary sodium, the predominant sources of dietary sodium for the Korean population are reported to be salt and kimchi, followed by soy sauce, ramen and soybean paste [1329]. Moreover, the highest sodium source per serving was ramen (1,563 mg per serving), followed by salted seafood (jeotgal) (956 mg per serving), buckwheat noodles (891 mg per serving), and sandwich/hamburger/pizza (830 mg per serving) [1329].

PUBLIC HEALTH IMPLICATIONS AND SPECIAL CONSIDERATIONS

Contrary to the conventional idea that a lower sodium intake is healthy, recent studies have reported a U-shaped or J-shaped curve for the association between sodium intake and CVD risks, rather than a continuously increasing positive association [530]. Although there is a continuous positive linear relationship between sodium intake and BP, a sodium intake of < 3 g or > 7 g/d increases the risk of CVD in a U-shaped curve, when compared to an intake of 4–5 g/d [30]. Future studies on the effect of very low sodium intakes, and balance studies on adequate sodium intake, are required to establish the AI for sodium in the Korean population. It should be acknowledged that the absence of reference values of UL or intake goal for dietary sodium does not indicate that there are no adverse toxicological effects of excessive sodium intake. Rather, it means that there is a lack of evidence for the toxicological effects of sodium in healthy populations. Therefore, further studies using the toxicological risk assessment approach are needed in order to set the UL in the future.

It has also been reported that the ratio between dietary sodium and potassium (Na:K ratio) is associated with health outcomes [313233]. Specifically, a high Na:K ratio is reported to be associated with increased BP, and risk of hypertension and stroke [343536]. Despite national efforts of reducing sodium intake at the population level, the Na:K ratio remains high among Koreans (ratio of 2.20 in 2019) [37]. Therefore, a health policy that targets lowering of the Na:K ratio is needed to improve public health in the Korean population.

The main limitation is insufficient evidence necessary to assess the relationship between sodium intake and chronic diseases in the Korean population. However, there is considerable evidence worldwide, for understanding the association between sodium intake and chronic disease. Furthermore, a meta-analysis on intake–response conducted by the US committee studying the DRIs of sodium, included Japanese and Chinese studies. Therefore, the CDRR established by the US committee can be used as the provisional CDRR for Koreans. Considering the increasing prevalence of hypertension and CVDs in South Korea, establishing and promoting this CDRR would be beneficial. Further studies on the relationship between sodium intake and chronic diseases are required to establish the CDRR for sodium in the Korean population. Moreover, since the majority of Koreans consume sodium at levels above the CDRR, there is a need to find new solutions, including technological innovations, that can reduce sodium in the food supply and in consumption.

CONCLUSIONS

The 2020 KDRIs established the AI value for dietary sodium intake at 1,500 mg/day for adults aged 19–64 years, which is the same as in the 2015 KDRIs. Moreover, on the basis of a systematic review, the CDRR for sodium was set at 2,300 mg/day for adults aged 19–64 years as a new reference value, specific to chronic disease risk reduction. Intake goal for dietary sodium established in the 2015 KDRIs (instead of the UL) was not presented in the 2020 KDRIs. For the next revision of the KDRIs, further studies are required on sodium adequacy and toxicity, and their associations with chronic diseases, in the Korean population.