ABSTRACT
Background
Hypertension (HTN) is a leading risk factor for cardiovascular disease, and dietary sodium reduction is a cornerstone strategy for blood pressure (BP) control, particularly in vulnerable populations such as individuals with chronic kidney disease (CKD), who often present with HTN. This study aims to quantify the dose–response relationship between changes in sodium intake and changes in BP, through a meta-analysis of recent randomized controlled trials (RCTs) in individuals with CKD and with or without HTN.
Methods
A systematic literature search was conducted in PubMed, Cochrane Central and Embase databases to identify RCTs published since 2000 that evaluated the isolated effect of sodium intake modification on systolic and diastolic. Only trials estimating sodium intake via 24-h urinary sodium excretion were included to ensure accurate measurement of dietary intervention. A random-effects model was used to pool effect sizes, accounting for between-study heterogeneity. Linear, quadratic and cubic dose–response models were fitted and compared with identify the best-fitting relationship between sodium change and BP change.
Results
The analysis included 43 RCTs with 51 distinct population groups, involving 1759 subjects. The linear model best described the dose–response relationship in subjects with CKD, with or without HTN. A reduction in sodium intake of 40 mmol/day (approximately 2.4 g of salt) was associated with a mean systolic BP/diastolic BP reduction of −4.5 mmHg (95% confidence interval −5.8, −3.1)/−2.2 mmHg (−3.0, −1.3) in patients with CKD, −2.3 mmHg (−3.0, −1.6)/−1.1 mmHg (−1.5, −0.6) in patients with HTN and −0.3 mmHg (−0.5, −0.1)/−0.1 mmHg (−0.2, −0.1) in patients without HTN.
Conclusion
Reducing sodium intake has a pronounced antihypertensive effect in patients with CKD and/or HTN. Regardless of the variability in effect according to BP status, adopting a low-salt diet represents good clinical practice and should be considered a standard prescription, particularly for individuals with CKD, as it supports better BP control and helps reduce cardiovascular risk.
Keywords: blood pressure, dose–response, meta-analysis, randomized clinical trials, sodium
GRAPHICAL ABSTRACT
GRAPHICAL ABSTRACT.
KEY LEARNING POINTS.
What was known:
Sodium intake is a well-established modifiable factor influencing blood pressure (BP).
Dietary sodium reduction is known to lower BP, particularly in individuals with hypertension (HTN).
Patients with chronic kidney disease (CKD) are frequently hypertensive and considered at high cardiovascular risk, making sodium diet control a critical component of their clinical management.
This study adds:
This study provides a quantified dose–response relationship between sodium diet reduction and BP change using a meta-analysis of recent randomized controlled trials.
Importantly, only trials that estimated sodium intake via 24-h urinary sodium excretion were included to ensure accurate measurement of the dietary intervention.
It distinguishes the effects in three distinct populations: subjects with CKD, subjects with HTN and subjects without HTN.
The findings highlight a significantly greater BP-lowering effect of sodium reduction in subjects with CKD and HTN compared with subjects without HTN.
Potential impact:
These results confirm the priority given to sodium reduction strategies in clinical recommendations, particularly in patients with CKD and HTN.
They underscore that, in line with international guidelines, a reduced sodium intake should be regarded as an essential component of good clinical practice for everybody, and even more so for high-risk populations.
This approach strengthens the relevance of clinical and public health interventions aimed at promoting a low-salt diet, in order to optimize BP control and reduce cardiovascular risk.
INTRODUCTION
Hypertension (HTN), a major risk factor for cardiovascular diseases, can be influenced by various dietary factors, sodium intake being a major contributor. The relationship between sodium intake and blood pressure (BP) has been extensively studied, with numerous guidelines recommending sodium reduction to prevent and manage HTN [1–3]. Despite this, the precise dose–response relationship between changes in sodium intake and BP changes remains an area of ongoing research, particularly when aiming to refine public health recommendations in the general population and in populations at high risk of HTN and cardiovascular disease like chronic kidney disease (CKD) populations [4, 5].
Many meta-analyses have been conducted to estimate the effects of various sodium modification interventions on BP, but only few studies have attempted to quantify the dose–response relationship between sodium and BP [6–8]. In addition, advances in statistical methods now make it possible to assess dose–response relationships with greater precision [9]. By using flexible models that do not require a priori assumptions about the shape of the dose–response curve, researchers can develop more valid models to predict the effect of sodium intake on BP [10]. Most studies attempted to estimate a linear slope, potentially oversimplifying the true nature of the relationship [6, 8]. More recently, Filippini et al. have described the dose–response relationship as approximately linear using a one-stage cubic spline model, a sophisticated statistical technique that accommodates complex, non-linear relationships starting at two dose levels [7].
In the meta-analysis presented here, only recent studies (post-2000) were included to consider recent BP measurement practices and to ensure that results are based on the most reliable and up-to-date evidence. In addition, only studies that estimated changes in salt intake using 24-h urinary sodium excretion were included in the meta-analysis to have an objective assessment of the intervention. We excluded studies that used salt substitutes as a means of sodium reduction (to avoid cross-effects with another nutrient, notably potassium), those with confounding dietary regimens such as calorie reduction or fluid restriction, and those with confounding interventions like physical exercise. The meta-analyses were carried out in subjects with CKD, subjects with HTN (without CKD) and subjects without HTN, who are at particularly high risk of cardiovascular complications. The pathophysiology of CKD involves impaired sodium handling and volume regulation, potentially leading to a heightened sensitivity to sodium intake [11]. Understanding this dose–response relationship in individuals with CKD is crucial, as tailored dietary recommendations could improve BP management and overall health outcomes in this vulnerable group [12]. Studies that selected normotensive subjects based on salt sensitivity were excluded to avoid selection bias. Dose–response analyses compared linear and non-linear approaches, to identify the best fit and ensure the accuracy of our predictions. This careful selection process aimed to mitigate the biases and limitations that could be present in previous analyses [13].
MATERIALS AND METHODS
The protocol was registered a priori on PROSPERO (CRD42023440909). The methodology complied with the Cochrane Handbook for Systematic Reviews of Interventions [14] and the study was reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist (Supplementary data, S1) [15].
Database search and study selection
We conducted a comprehensive search of three electronic databases (PubMed, Cochrane Central and Embase) until November 2024. Our PICO criteria were: (i) population: humans; (ii) intervention: sodium intake modification only; (iii) comparator: no or other modification of sodium intake; (iv) outcome: systolic and diastolic BP (SBP and DBP); and (v) type of publication: systematic reviews and meta-analyses of randomized controlled trials (RCTs). The search strategy was a combination of different keywords for each criterion (for details, see Supplementary data, S2). In addition, a manual search was performed. The publication date was restricted to the period after 2000. The search was restricted to articles published in English and French. Two independent reviewers (M.G. and V.S.) performed the title and abstract screening, thereafter, the full text screening for eligibility (M.G. and O.I.). Any disagreements were resolved through consultation of two researchers (A.G. and J.P.F.). We used a two-step process for our study selection. Initially, we identified systematic reviews and meta-analyses that met our inclusion criteria based on our PICO criteria. Afterwards we extracted all included post-2000 RCTs from the eligible systematic reviews and meta-analyses, using a pre-defined mapping table to avoid overlapping studies and screened them for eligibility based on our predefined criteria. The studies were included if they met the following criteria: (i) RCTs; (ii) sodium intake modification intervention; (iii) measured an SBP and DBP outcome; and (iv) estimated sodium intake from 24-h urinary sodium excretion.
We excluded studies that used salt substitutes as a means of sodium reduction (to avoid cross-effects with another nutrient, notably potassium), those with confounding dietary regimens such as calorie reduction or fluid restriction, and those with confounding interventions like physical exercise. Studies with missing outcome report or involving children as participants were also excluded. For the purposes of this review, sodium intake modification was defined as an increase or decrease in sodium intake through supplementation or diet. Analysis was carried out in subjects with CKD, with HTN or without. Studies selecting subjects without HTN on the basis of salt sensitivity were excluded to avoid selection bias. For studies mixing populations, the distribution was not representative of the general population, particularly in terms of BP, and were therefore not retained.
Risk of bias assessment
The quality of the included studies was assessed by O.I. and M.G. using the revised Cochrane risk-of-bias tool for randomized trials (ROB2) [16]. A risk of bias assessment was performed in five domains: randomization process, blinding, missing outcome data, outcome measurement and outcome reporting. The risk of bias judgments for each domain are ‘low risk of bias’, ‘some concerns’ or ‘high risk of bias’. Judgments are based on responses to signalling questions. Studies assessed as having a high risk of bias were excluded from the meta-analysis.
Data extraction
Three independent reviewers (O.I., M.G. and J.P.F.) extracted the data using a standardized form. When some data of a clinical study were missing in the included systematic reviews, we extracted them from the original clinical study. Extracted data included: study ID, author, publication date, country, study design, washout period, number of participants in the intervention and control groups, duration of intervention, age, proportion of female participants, pathology population, CKD population, estimated glomerular filtration rate, HTN status, body mass index (BMI), BP measurement method, baseline and follow-up SBP, baseline and follow-up DBP, and estimated 24-h urinary sodium excretion.
BMI when not available was calculated from weight and height data. When available, ambulatory BP was preferred over clinic-based BP measurements. In the absence of 24-h data, diurnal BP was preferred to nocturnal BP. Sodium intake represented in milligrams was converted to millimoles by dividing by 23, the molar mass of sodium.
For studies that used a crossover design, the effect of the intervention was taken as the difference between endpoint measurements in the two groups. In studies that used a parallel design, we did not adjust for baseline differences in BP. It was assumed that randomization would balance baseline characteristics between groups, and any differences would be minimal and unlikely to significantly affect the measured treatment effects. Studies with substantial baseline imbalances were excluded from the analysis. The duration of the intervention was defined as the period from the start of the intervention to the point at which the outcome was measured.
For studies that did not report the standard deviation of the mean for each group, this was converted from the standard error, median (interquartile range) or 95% confidence interval (95% CI) using meta-converter [17], a meta-analysis conversion tool with available references.
Authors were contacted for missing reports on outcome variables, and no imputation was performed on these outcome variables.
Statistical analysis
We performed a random-effects dose–response meta-analysis to examine the effects of sodium intake modification on SBP and DBP modification. The dose was expressed as the mean difference in 24-h sodium excretion between the low and high sodium groups, the high sodium group referenced at ‘zero’. The response was defined as the mean difference in SBP and DBP changes between groups at the end of each intervention.
We examined the dose–response relationship using three regression models: linear, quadratic and one-stage cubic spline. The cubic spline model included knots at the 10th, 50th and 90th percentiles of the sodium intake modification distribution [18]. Model selection was based on the Akaike Information Criterion (AIC), with the model having the lowest AIC considered the best fit for the data. The selected model was used to predict the dose–response relationship for sodium intake values ranging from 40 to 240 mmol/day. We performed sensitivity analyses in each study population: subjects with CKD or HTN, or without HTN to assess the robustness of our results based on different study characteristics. These included dietary interventions versus supplement interventions, ambulatory BP measurements versus clinical measurements, and intervention durations stratified to ≤1 week versus >1 week. The Wald test was used to assess the statistical significance of these moderators. To examine the potential for publication bias, we used funnel plots and Egger’s regression test. The funnel plots displayed the standard error of each study against its effect estimate, while Egger’s test provided a quantitative measure of bias direction and magnitude.
The dose–response meta-analysis was performed using the dosresmeta command within the metafor package in R 4.2.2 software [19, 20].
RESULTS
Study selection
The study selection process is presented in the PRISMA flow chart (Supplementary data, Fig. S1). The electronic database search resulted in 1826 references. Titles and abstracts were screened, leading to 173 references being deemed eligible for full-text review. After reviewing the full texts, 49 systematic reviews and meta-analyses were considered eligible. In addition, a manual search for eligible studies identified eight additional references, which were also reviewed based on their titles and abstracts. This second search allowed us to refine the comprehensiveness of our literature analysis. In the end, a total of 51 systematic reviews and meta-analyses were eligible. After the selection of eligible systematic reviews and meta-analyses, primary studies for the dose–response meta-analysis were retrieved using a mapping table, resulting in 258 post-2000 RCTs being assessed for eligibility through full-text reading. Ultimately, 43 RCTs were included in the analysis [21–63].
Characteristics of selected studies
Forty-three RCTs with 51 distinct study population groups were included, with publication dates ranging from 2000 to 2024 and set across the following countries: Australia (n = 4), Croatia (n = 1), Denmark (n = 4), Finland (n = 1), Italy (n = 1), the Netherlands (n = 8), Norway (n = 1), Pakistan (n = 1), South Korea (n = 1), Sweden (n = 2), Switzerland (n = 5), Thailand (n = 1), Turkey (n = 1), the UK (n = 6) and the USA (n = 6). A total of 1759 participants were included from all studies. Of the 43 included studies, 39 (1384 total participants) had a crossover design, and 4 (377 total participants) had a parallel design. Mean age ranged from 20 to 69 years, and mean BMI from 21 to 43 kg/m². Most study groups had a mixed-sex ratio, with 10 including only male participants and 7 including only female participants. All analyses were divided into three populations: patients with CKD (n = 7), patients with HTN (n = 12) and patients without HTN (and without CKD) (n = 32).
The duration of interventions ranged from 3 days to 12 weeks, with a median duration of 1 week. Forty-three studies used dietary interventions to modify sodium intake, while 14 used supplements. Detailed characteristics of the selected studies are presented in a table (Supplementary data, Table S3).
Description of the interventions
The interventions involved dietary modifications aimed at reducing sodium intake through controlled feeding and regimens. Participants in the high sodium group mostly received generic dietary advice and sodium supplementations. Most studies ensured balanced nutritional content except for the differing sodium content across groups when preparing or selecting high sodium foods. Participants were generally instructed to maintain their usual habits except for changes in food content. Supplements were provided to the high sodium group in the form of salt tablets or slow sodium tablets (special slow-release form). In crossover trials, participants were randomized into groups, with some trials including a washout period between different phases of the dietary supplement intervention. There was a preliminary period of sodium intake reduction in both groups before the administration of supplementation versus placebo. Only one study provided supplements without a placebo in a crossover design.
In all included studies, 24-h urinary sodium excretion, and BP were assessed by a third party. Twenty-three studies measured and reported ambulatory BP, while 20 studies reported only clinic BP. BP was measured at the start and end of the intervention in parallel studies, and at the end of each intervention phase in crossover designs. Repeated BP measurements throughout the duration of the studies were rarely done while some studies monitored urinary sodium excretion throughout the study to ensure compliance with the intervention.
Risk of bias assessment
Twenty-one studies were assessed to have low risk and 22 studies to have some concerns. Eighteen studies had some concerns in their randomization process as there was no report on the sequence method. Three studies raised some concerns about deviation from the intended intervention due to unblinding of participants or administrators, or unblinding due to the intervention method, including dietary counselling. All studies were assessed as having low risk in the reporting of outcomes and in the measurement of outcome. Most studies in which participants were not blinded indicated that the outcome assessor was blinded. None of the studies included was judged to be at high risk of bias. A summary of the risk of bias assessment for individual studies is available in Supplementary data, Fig. S2.
Dose–response meta-analysis
The average differences in sodium excretion ranged from 9 to 131 mmol/day for subjects with CKD, from 34 to 206 mmol/day for subjects with HTN and from 42 to 364 mmol/day for subjects without HTN. In each of these subpopulations. Across all subpopulations, the dose–response relationship between sodium excretion and SBP/DBP was best described by a linear model, which had the lowest AIC compared with quadratic and cubic spline models (Supplementary data, Tables S4 and S5). Although nonlinear approaches theoretically offer a better fit to the data, the gains observed in our analysis were marginal. For the sake of parsimony, and to avoid the risks of extrapolation associated with complex models, we favoured linear modelling applied to restricted, clinically relevant exposure ranges. This approach allows the dose–response relationship to be represented in a robust and interpretable manner within the ranges of values actually observed. In patients with CKD, the linear and cubic spline models had identical AICs; the linear model was also selected in this subgroup based on the same considerations. The modelling of the dose–response relationship with each of the models is shown (Supplementary data, Fig. S3).
The estimates of predictors in the selected models were statistically significant across all subpopulations. The dose–response curve indicated a particularly strong relationship between sodium excretion and SBP/DBP in subjects at high cardiovascular risk, particularly patients with CKD and those with HTN, unlike subjects without HTN. The potential benefit of reducing sodium intake appeared to be even more pronounced in patients with CKD—a population that is predominantly hypertensive—than in hypertensive subjects without CKD (Fig. 1).
Figure 1:
Dose–response curves linking sodium intake changes to SBP and DBP changes in subjects with CKD, with HTN and without HTN.
The predicted changes in SBP and DBP associated with varying levels of sodium intake reduction (40, 80, 120, 160, 200 and 240 mmol/day) are presented in Table 1. A daily reduction of 40 mmol in sodium intake was associated with a predicted decrease in SBP of –4.5 mmHg (95% CI –5.8, –3.1) and in DBP of –2.2 mmHg (95% CI –3.0, –1.3) among individuals with CKD. In individuals with HTN, the corresponding predicted reductions were –2.3 mmHg (95% CI –3.0, –1.6) for SBP and –1.1 mmHg (95% CI –1.5, –0.6) for DBP. In contrast, among individuals without HTN, the reductions were modest: –0.3 mmHg (95% CI –0.5, –0.1) for SBP and –0.1 mmHg (95% CI –0.2, –0.1) for DBP. A reduction of 40 mmol of sodium per day is equivalent to a clinically achievable decrease of approximately 2.4 g of dietary salt.
Table 1:
Prediction of dose–response for subjects with CKD, with or without HTN.
| SBP/DBP (mmHg) | SBP/DBP (mmHg) | SBP/DBP (mmHg) | |
|---|---|---|---|
| Dose (mmol/day) | Patients with CKD | Patients with HTN | Patients without HTN |
| 40 | –4.5 (–5.8, –3.1)/–2.2 (–3.0, –1.3) | –2.3 (–3.0, –1.6)/–1.1 (–1.5, –0.6) | –0.3 (–0.5, –0.1)/–0.1 (–0.2, –0.1) |
| 80 | –8.9 (–11.7, –6.1)/–4.3 (–6.1, –2.5) | –4.6 (–6.0, –3.2)/–2.1 (–3.0, –1.2) | –0.6 (–0.9, –0.2)/–0.1 (–0.4, –0.1) |
| 120 | –13.4 (–17.5, –9.2)/–6.5 (–9.1, –3.8) | –6.9 (–9.1, –4.8)/–3.2 (–4.5, –1.8) | –0.9 (–1.4, –0.4)/–0.2 (–0.6, –0.2) |
| 160 | –9.3 (–12.1, –6.4)/–4.2 (–6.0, –2.4) | –1.2 (–1.9, –0.5)/–0.3 (–0.9, –0.3) | |
| 200 | –11.6 (–15.1, –8.0)/–5.3 (–7.6, –3.1) | –1.5 (–2.3, –0.6)/–0.4 (–1.1, –0.3) | |
| 240 | –1.8 (–2.8, –0.7)/–0.4 (–1.3, –0.4) |
Variation in SBP (mmHg)/DBP (mmHg) with 95% CI per dose represented by the difference in sodium excretion (mmol/day).
Publication bias
A publication bias analysis was conducted for each population included in the dose–response meta-analysis, in relation to SBP and DBP. No signs of publication bias were observed upon visual examination of the funnel plots for the analyses conducted in subjects with or without HTN, which was corroborated by Egger’s test (subjects with HTN: t = –0.49, P = .63 for SBP; t = –1.26, P = .24 for DBP; subjects without HTN: t = –0.23, P = .82 for SBP; t = –0.50, P = .23 for DBP).
However, in individuals with CKD, publication bias cannot be ruled out in the dose–response meta-analysis on BP. Nevertheless, this interpretation should be viewed with caution, as this analysis is based on a limited number of studies (n = 7). Egger’s test, although suggestive, is used here on an exploratory basis and outside its optimal conditions of application (n ≥ 10). It is therefore difficult to formally conclude that publication bias exists, as this may also reflect a simple lack of available data in this subpopulation.
DISCUSSION
Our study aimed to quantify the relationship between changes in sodium intake (assessed by 24-h urinary sodium excretion) and changes in SBP and DBP using a dose–response meta-analysis in subjects with CKD, with HTN and without HTN, using only studies published after 2000. The main finding of our study is that reducing sodium intake has a significant impact on BP in subjects with CKD and those with HTN. In contrast, variations in sodium intake have a rather marginal effect on SBP and DBP in normotensive subjects. Given the abundance of research on sodium and BP interventions, as well as recent advances in diagnosis (particularly in terms of measurement devices and thresholds) and the therapeutic management of HTN (new antihypertensive drugs on the market) [64], it has become essential to produce an up-to-date, rigorous and clinically relevant summary. That is why we have deliberately restricted our selection to studies conducted after 2000, which are both recent and of high methodological quality. This approach not only better reflects contemporary practices, but also contributes to the harmonization of strategies for measuring and treating HTN, particularly in patients with CKD. Furthermore, to our knowledge, there are no meta-analyses specifically evaluating the effect of sodium reduction on BP in patients with CKD—a factor that reinforces the originality of our approach and minimizes the risk of selection bias. In order to avoid any a priori assumptions about the shape of the dose–response relationship between sodium excretion and SBP/DBP, we compared three types of models: linear, quadratic and cubic spline, in subjects with CKD, with HTN and without HTN [65]. In all subgroups, the linear model proved to be the most representative, with AIC values consistently lower than or equal to those of the more complex models. In patients with CKD, the linear and cubic models had identical AIC values, but the linear model was preferred in accordance with the principle of parsimony. Although nonlinear approaches may theoretically better account for certain inflections in the dose–response relationship, the gains observed in our analysis were marginal. For the sake of simplicity and robustness, and to limit the risks of extrapolation associated with complex models, we therefore opted for linear modelling, applied to limited and clinically relevant exposure ranges. This approach ensures a clear and reliable interpretation of the results.
The estimates from the selected models were statistically significant across all subgroups. In the subgroup analyses, we observed that the relationship between sodium reduction and BP lowering was particularly pronounced in individuals with CKD, strong in those with HTN, and much weaker, or even negligible, in normotensive subjects. However, it should be noted that our conclusions are based on short-term trials, which do not allow for an assessment of the long-term effects of sodium reduction. Beyond the immediate BP response, extensive experimental evidence (in vitro and animal studies) has clearly demonstrated the deleterious effects of high sodium intake on arterial structures, independently of BP variations. Moreover, large-scale epidemiological studies [66], following normotensive cohorts over several decades, have established a robust association between high sodium intake and the incidence of HTN over time. Taken together, these findings underscore the value of a low-salt diet not only as a strategy for BP control but also as a means of long-term vascular and cardiovascular prevention.
Our findings are in line with previous work by Filippini et al. [67], who, using the one-stage cubic spline model, found a near-linear relationship between sodium intake modifications and BP, with a weaker effect in individuals without HTN and a stronger effect in individuals with HTN. The modest effect observed of reduced sodium excretion on BP in individuals without-HTN is largely explained by our rigorous selection strategy. Unlike previous meta-analyses, we excluded studies that specifically recruited normotensive individuals who were sensitive to salt. The inclusion of such enriched populations tends to overestimate the effect of the intervention, introducing bias into previously published estimates [68–71]. In our analysis, participants without HTN included both salt-sensitive and non-salt-sensitive individuals, but without prior selection based on this criterion, allowing for a more representative estimate of the average effect in the general population without HTN.
One of the strengths of our dose–response meta-analysis is the comprehensive evaluation of the different models, as no a priori assumptions were made about a specific dose–response curve. Although non-linear models such as the cubic spline were explored, the linear model provided the best fit in all subpopulations. This exploratory approach captures the varying effects of sodium intake modification in different subpopulations, providing a clearer and more reliable picture of the dose–response relationship [72].
Our study methodology enhances the reliability of our findings through the careful selection of high-quality and more recent studies. Only RCTs were included and studies involving salt substitutes or multiple confounding interventions were excluded. This approach minimizes heterogeneity and reduces biases commonly associated with older studies, and ensured that our results reflect the most current understanding of sodium modification interventions, considering the advancements in patient management, dietary guidelines and transparency in reporting [73]. In addition, only studies estimating sodium intake using 24-h urinary sodium excretion were analysed. Studies using spot urine tests, questionnaires and dietary recalls were not included given the reported inaccuracies of these methods of estimating sodium intake [74].
Twenty-four-hour urinary sodium excretion is not a perfect measure of dietary sodium intake due to biological variability. Nonetheless, repeated 24-h urine collections are rarely feasible in real-world clinical or research settings. Despite these limitations, it remains the most reliable available method for estimating dietary sodium intake.
However, our study has some limitations. Our dose–response meta-analysis, based exclusively on RCTs, reflects the short-term effects of interventions on sodium consumption. By definition, this type of study is not designed to assess the incidence of long-term events, such as the development of HTN. In the absence of long-term follow-up data, it is therefore not possible, in this context, to estimate the impact of a reduction in sodium intake on the onset of HTN over time. Furthermore, the effect of sodium reduction on BP may depend on the level of sodium intake prior to the intervention, regardless of the extent of the reduction achieved. However, this potential interaction between baseline intake and BP response could not be explored in the present dose–response meta-analysis. It should also be acknowledged that the physiological effects of sodium manipulation may vary over time, as changes in sodium intake can influence water–electrolyte balance and activate neurohormonal pathways (e.g. the renin–angiotensin–aldosterone system and the sympathetic nervous system), with potential consequences beyond the short-term BP changes captured by RCTs. These considerations should be kept in mind when interpreting our findings. We also acknowledge the considerable variability in the magnitude of sodium reduction across the included trials, ranging from minimal to very large decreases. While such extremes may not directly reflect usual clinical practice, their inclusion allowed us to provide the most exhaustive synthesis possible and to model the dose–response relationship across the full spectrum of sodium reduction. Heterogeneity based on intervention methods as well as that resulting from clinical characteristics such as age, sex or BMI, antihypertensive treatments, which could provide additional information on the variability of dose–response estimates, were not considered. Future research should consider these factors to enhance the understanding of individual BP responses to sodium dietary modifications. Additionally, the rigorous selection process may have inadvertently excluded some studies in the CKD population, potentially limiting the depth of our analysis for this subgroup. Despite this, the study selection process resulted in the inclusion of many recent, high-quality studies, providing a solid basis for the present meta-analysis. While focusing on recent studies ensures contemporary relevance, it may also mean that some useful older studies were not considered. This trade-off was necessary to balance the need for high-quality evidence with the practicalities of study selection. Nevertheless, our study offers useful insights into the dose–response patterns in different subpopulations, addressing gaps in the literature and contributing to a more nuanced understanding of sodium’s impact on BP. This understanding is crucial for developing targeted dietary recommendations and interventions for different populations, ultimately helping the management and prevention of HTN and related conditions.
CONCLUSION
Our study confirms a strong positive linear dose–response relationship between sodium intake reduction and BP lowering in adults with CKD and/or HTN, with no evidence of a threshold or plateau effect. This indicates that any effort to reduce sodium intake is likely to yield beneficial effects on BP in these high-risk populations. Although the impact appears more limited in individuals without HTN, the overall evidence supports sodium reduction as an essential component of good clinical practice for all patients, in line with international recommendations. Promoting a low-salt diet should therefore be considered a standard intervention, particularly in patients with CKD or HTN, as it contributes to improved BP control and reduces cardiovascular risk.
Supplementary Material
Contributor Information
Omomene Iwelomene, Laboratoire de biométrie et biologie évolutive, équipe EMET Equipe Evaluation et Modélisation des Effets Thérapeutiques, Université Claude Bernard Lyon 1, Villeurbanne, France; Département Universitaire de Santé Publique, Faculté de Médecine, Université de Limoges, France.
Arthur Gougeon, RCTs, Lyon, France.
Michel Burnier, Faculté de Biologie et Médecine, Université de Lausanne, Lausanne, Switzerland.
Léonie Belot, Laboratoire de biométrie et biologie évolutive, équipe EMET Equipe Evaluation et Modélisation des Effets Thérapeutiques, Université Claude Bernard Lyon 1, Villeurbanne, France.
Victoria Sourd, Laboratoire de biométrie et biologie évolutive, équipe EMET Equipe Evaluation et Modélisation des Effets Thérapeutiques, Université Claude Bernard Lyon 1, Villeurbanne, France.
Jean Pierre Fauvel, Hospices Civils de Lyon, Hôpital Edouard Herriot, Service de Néphrologie, Lyon, France.
Maelys Granal, Laboratoire de biométrie et biologie évolutive, équipe EMET Equipe Evaluation et Modélisation des Effets Thérapeutiques, Université Claude Bernard Lyon 1, Villeurbanne, France; Hospices Civils de Lyon, Hôpital Edouard Herriot, Service de Néphrologie, Lyon, France.
FUNDING
We would like to thank the Fondation Philanthopia, the Droux’ family and A. Lefranc for their selfless financial support, which made this research possible.
DATA AVAILABILITY STATEMENT
The data underlying this article will be shared on reasonable request to the corresponding author.
CONFLICT OF INTEREST STATEMENT
None declared.
REFERENCES
- 1. Mancia G, Kreutz R, Brunström M et al. 2023 ESH guidelines for the management of arterial hypertension the task force for the management of arterial hypertension of the European Society of Hypertension: endorsed by the International Society Of Hypertension (ISH) and the European Renal Association (ERA). J Hypertens 2023;41:1874–71. 10.1097/HJH.0000000000003480 [DOI] [PubMed] [Google Scholar]
- 2. McEvoy JW, McCarthy CP, Bruno RM et al. 2024 ESC guidelines for the management of elevated blood pressure and hypertension. Eur Heart J 2024;45:3912–4018. 10.1093/eurheartj/ehae178 [DOI] [PubMed] [Google Scholar]
- 3. Whelton PK. Is the current recommendation by guidelines for hypertension to keep daily sodium intake below 5 G (sodium 2300 MG) realistic? J Hypertens 2023;41:e153. 10.1097/01.hjh.0000914244.67145.8d [DOI] [Google Scholar]
- 4. Jha V, Garcia-Garcia G, Iseki K et al. Chronic kidney disease: global dimension and perspectives. Lancet 2013;382:260–72. 10.1016/S0140-6736(13)60687-X [DOI] [PubMed] [Google Scholar]
- 5. Guideline: Sodium Intake for Adults and Children. Available from: https://www.who.int/publications/i/item/9789241504836 (16 June 2024, date last accessed). [PubMed] [Google Scholar]
- 6. Huang L, Trieu K, Yoshimura S et al. Effect of dose and duration of reduction in dietary sodium on blood pressure levels: systematic review and meta-analysis of randomised trials. BMJ 2020;368:m315. 10.1136/bmj.m315 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Filippini T, Malavolti M, Whelton PK et al. Blood pressure effects of sodium reduction: dose-response meta-analysis of experimental studies. Circulation 2021;143:1542–67. 10.1161/CIRCULATIONAHA.120.050371 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Graudal N, Hubeck-Graudal T, Jürgens G et al. Dose-response relation between dietary sodium and blood pressure: a meta-regression analysis of 133 randomized controlled trials. Am J Clin Nutr 2019;109:1273–8. 10.1093/ajcn/nqy384 [DOI] [PubMed] [Google Scholar]
- 9. Hedges LV. Advances in statistical methods for meta-analysis. New Dir Program Eval 1984;1984:25–42. 10.1002/ev.1376 [DOI] [Google Scholar]
- 10. Crippa A. Novel Methods for Dose–Response Meta-Analysis. Available from: https://bookdown.org/alecri/thesis/4-methods.html (16 June 2024, date last accessed). [Google Scholar]
- 11. Cirillo M, Bilancio G, Cavallo P et al. Sodium intake and kidney function in the general population: an observational, population-based study. Clin Kidney J 2020;14:647–55. 10.1093/ckj/sfaa158 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Healthy Eating for People with Chronic Kidney Disease (CKD). National Kidney Federation, 29 September 2023. Available from: https://www.kidney.org.uk/healthy-eating-for-people-with-chronic-kidney-disease-ckd (24 May 2024, date last accessed). [Google Scholar]
- 13. Pigott TD, Polanin JR. Methodological Guidance paper: high-quality meta-analysis in a systematic review. Rev Educ Res 2020;90:24–46. 10.3102/0034654319877153 [DOI] [Google Scholar]
- 14. Higgins JPT, Thomas J, Chandler J et al. Cochrane Handbook for Systematic Reviews of Interventions. Wiley-Blackwell, 2019. Available from: https://research.monash.edu/en/publications/cochrane-handbook-for-systematic-reviews-of-interventions (18 May 2023, date last accessed) [Google Scholar]
- 15. Page MJ, McKenzie JE, Bossuyt PM et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372:n71. 10.1136/bmj.n71 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Sterne JAC, Savović J, Page MJ et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019;366:l4898. 10.1136/bmj.l4898 [DOI] [PubMed] [Google Scholar]
- 17. Abbas A, Hefnawy MT, Negida A. Meta-analysis accelerator: a comprehensive tool for statistical data conversion in systematic reviews with meta-analysis. BMC Med Res Methodol 2024;24:243. 10.1186/s12874-024-02356-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Harrell FE. Regression Modeling Strategies: With Applications to Linear Models, Logistic Regression, and Survival Analysis. New York: Springer, 2001. 10.1007/978-1-4757-3462-1 [DOI] [Google Scholar]
- 19. R Core Team . R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing, 2022. Available from: https://www.R-project.org/ (n.d.). (16 June 2024, date last accessed). [Google Scholar]
- 20. Balduzzi S, Rücker G, Schwarzer G. How to perform a meta-analysis with R: a practical tutorial. Evid Based Ment Health 2019;22:153–60. 10.1136/ebmental-2019-300117 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Pechère-Bertschi A, Maillard M, Stalder H et al. Blood pressure and renal haemodynamic response to salt during the normal menstrual cycle. Clin Sci 2000;98:697–702. 10.1042/cs0980697 [DOI] [PubMed] [Google Scholar]
- 22. Houlihan CA, Allen TJ, Baxter AL et al. A low-sodium diet potentiates the effects of losartan in type 2 diabetes. Diabetes Care 2002;25:663–71. 10.2337/diacare.25.4.663 [DOI] [PubMed] [Google Scholar]
- 23. Dishy V, Sofowora GG, Imamura H et al. Nitric oxide production decreases after salt loading but is not related to blood pressure changes or nitric oxide-mediated vascular responses. J Hypertens 2003;21:153. 10.1097/00004872-200301000-00025 [DOI] [PubMed] [Google Scholar]
- 24. Perry CG, Palmer T, Cleland SJ et al. Decreased insulin sensitivity during dietary sodium restriction is not mediated by effects of angiotensin II on insulin action. Clin Sci 2003;105:187–94. 10.1042/CS20020320 [DOI] [PubMed] [Google Scholar]
- 25. Pechère-Bertschi A, Maillard M, Stalder H et al. Renal hemodynamic and tubular responses to salt in women using oral contraceptives. Kidney Int 2003;64:1374–80. 10.1046/j.1523-1755.2003.00239.x [DOI] [PubMed] [Google Scholar]
- 26. Beeks E, van der Klauw MM, Kroon AA et al. α-Adducin Gly460Trp polymorphism and renal hemodynamics in essential hypertension. Hypertension 2004;44:419–23. 10.1161/01.HYP.0000141410.72537.fd [DOI] [PubMed] [Google Scholar]
- 27. Zanchi A, Chiolero A, Maillard M et al. Effects of the peroxisomal proliferator-activated receptor-γ agonist pioglitazone on renal and hormonal responses to salt in healthy men. J Clin Endocrinol Metab 2004;89:1140–5. 10.1210/jc.2003-031526 [DOI] [PubMed] [Google Scholar]
- 28. Swift PA, Markandu ND, Sagnella GA et al. Modest salt reduction reduces blood pressure and urine protein excretion in black hypertensives. Hypertension 2005;46:308–12. 10.1161/01.HYP.0000172662.12480.7f [DOI] [PubMed] [Google Scholar]
- 29. Keven K, Yalçin S, Canbakan B et al. The impact of daily sodium intake on posttransplant hypertension in kidney allograft recipients. Transplant Proc 2006;38:1323–6. 10.1016/j.transproceed.2006.02.103 [DOI] [PubMed] [Google Scholar]
- 30. Melander O, von WF, Frandsen E et al. Moderate salt restriction effectively lowers blood pressure and degree of salt sensitivity is related to baseline concentration of renin and N-terminal atrial natriuretic peptide in plasma. J Hypertens 2007;25:619. 10.1097/HJH.0b013e328013cd50 [DOI] [PubMed] [Google Scholar]
- 31. Townsend RR, Kapoor S, McFadden CB. Salt intake and insulin sensitivity in healthy human volunteers. Clin Sci 2007;113:141–8. 10.1042/CS20060361 [DOI] [PubMed] [Google Scholar]
- 32. Vogt L, Waanders F, Boomsma F et al. Effects of dietary sodium and hydrochlorothiazide on the antiproteinuric efficacy of losartan. J Am Soc Nephrol 2008;19:999–1007. 10.1681/ASN.2007060693 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Tzemos N, Lim PO, Wong S et al. Adverse Cardiovascular effects of acute salt loading in young normotensive individuals. Hypertension 2008;51:1525–30. 10.1161/HYPERTENSIONAHA.108.109868 [DOI] [PubMed] [Google Scholar]
- 34. Jessani S, Hatcher J, Chaturvedi N et al. Effect of low vs. high dietary sodium on blood pressure levels in a normotensive indo-asian population. Am J Hypertens 2008;21:1238–44. 10.1038/ajh.2008.256 [DOI] [PubMed] [Google Scholar]
- 35. Meland E, Aamland A. Salt restriction among hypertensive patients: modest blood pressure effect and no adverse effects. Scand J Prim Health Care 2009;27:97–103. 10.1080/02813430802661795 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Paulsen L, Holst LM, Bech JN et al. Glomerular filtration rate and blood pressure are unchanged by increased sodium intake in atorvastatin-treated healthy men. Scand J Clin Lab Invest 2009;69:323–9. 10.1080/00365510802571007 [DOI] [PubMed] [Google Scholar]
- 37. Dickinson KM, Keogh JB, Clifton PM. Effects of a low-salt diet on flow-mediated dilatation in humans. Am J Clin Nutr 2009;89:485–90. 10.3945/ajcn.2008.26856 [DOI] [PubMed] [Google Scholar]
- 38. He FJ, Marciniak M, Visagie E et al. Effect of modest salt reduction on blood pressure, urinary albumin, and pulse wave velocity in white, black, and Asian mild hypertensives. Hypertension 2009;54:482–8. 10.1161/HYPERTENSIONAHA.109.133223 [DOI] [PubMed] [Google Scholar]
- 39. Pimenta E, Gaddam KK, Oparil S et al. Effects of dietary sodium reduction on blood pressure in subjects with resistant hypertension. Hypertension 2009;54:475–81. 10.1161/HYPERTENSIONAHA.109.131235 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Starmans-Kool MJ, Stanton AV, Xu YY et al. High dietary salt intake increases carotid blood pressure and wave reflection in normotensive healthy young men. J Appl Physiol 2011;110:468–71. 10.1152/japplphysiol.00917.2010 [DOI] [PubMed] [Google Scholar]
- 41. Slagman MCJ, Waanders F, Hemmelder MH et al. Moderate dietary sodium restriction added to angiotensin converting enzyme inhibition compared with dual blockade in lowering proteinuria and blood pressure: randomised controlled trial. BMJ 2011;343:d4366. 10.1136/bmj.d4366 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Graffe CC, Bech JN, Pedersen EB. Effect of high and low sodium intake on urinary aquaporin-2 excretion in healthy humans. Am J Physiol-Ren Physiol 2012;302:F264–75. 10.1152/ajprenal.00442.2010 [DOI] [PubMed] [Google Scholar]
- 43. Mallamaci F, Leonardis D, Pizzini P et al. Procalcitonin and the inflammatory response to salt in essential hypertension: a randomized cross-over clinical trial. J Hypertens 2013;31:1424. 10.1097/HJH.0b013e328360ddd5 [DOI] [PubMed] [Google Scholar]
- 44. Bonfils PK, Taskiran M, Damgaard M et al. The influence of high versus low sodium intake on blood pressure and haemodynamics in patients with morbid obesity. J Hypertens 2013;31:2220. 10.1097/HJH.0b013e328363c769 [DOI] [PubMed] [Google Scholar]
- 45. Dickinson KM, Clifton PM, Keogh JB. A reduction of 3 g/day from a usual 9 g/day salt diet improves endothelial function and decreases endothelin-1 in a randomised cross_over study in normotensive overweight and obese subjects. Atherosclerosis 2014;233:32–38. 10.1016/j.atherosclerosis.2013.11.078 [DOI] [PubMed] [Google Scholar]
- 46. Campbell KL, Johnson DW, Bauer JD et al. A randomized trial of sodium-restriction on kidney function, fluid volume and adipokines in CKD patients. BMC Nephrol 2014;15:57. 10.1186/1471-2369-15-57 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Kwakernaak AJ, Krikken JA, Binnenmars SH et al. Effects of sodium restriction and hydrochlorothiazide on RAAS blockade efficacy in diabetic nephropathy: a randomised clinical trial. Lancet Diabetes Endocrinol 2014;2:385–95. 10.1016/S2213-8587(14)70030-0 [DOI] [PubMed] [Google Scholar]
- 48. Cavka A, Cosic A, Jukic I et al. The role of cyclo-oxygenase-1 in high-salt diet-induced microvascular dysfunction in humans. J Physiol 2015;593:5313–24. 10.1113/JP271631 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Gu N, Cho JY, Shin KH et al. The influence of dietary sodium content on the pharmacokinetics and pharmacodynamics of fimasartan. Drug Des Devel Ther 2016;10:1525–31. 10.2147/DDDT.S94694 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50. de Vries LV, Dobrowolski LC, van den Bosch JJON et al. Effects of dietary sodium restriction in kidney transplant recipients treated with renin-angiotensin-aldosterone system blockade: a randomized clinical trial. Am J Kidney Dis 2016;67:936–44. 10.1053/j.ajkd.2015.11.026 [DOI] [PubMed] [Google Scholar]
- 51. van Twist DJ, Houben AJ, de Leeuw PW et al. Acute eprosartan-induced intrarenal vasodilation in hypertensive humans is not influenced by dietary sodium intake or angiotensin II co-infusion. J Hypertens 2016;34:1607. 10.1097/HJH.0000000000000974 [DOI] [PubMed] [Google Scholar]
- 52. Nielsen LH, Ovesen P, Hansen MR et al. Changes in the renin-angiotensin-aldosterone system in response to dietary salt intake in normal and hypertensive pregnancy. A randomized trial. J Am Soc Hypertens 2016;10:881–90.e4. 10.1016/j.jash.2016.10.001 [DOI] [PubMed] [Google Scholar]
- 53. Gefke M, Christensen NJ, Bech P et al. Hemodynamic responses to mental stress during salt loading. Clin Physiol Funct Imaging 2017;37:688–94. 10.1111/cpf.12360 [DOI] [PubMed] [Google Scholar]
- 54. Brian MS, Dalpiaz A, Matthews EL et al. Dietary sodium and nocturnal blood pressure dipping in normotensive men and women. J Hum Hypertens 2017;31:145–50. 10.1038/jhh.2016.53 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55. Selvarajah V, Mäki-Petäjä KM, Pedro L et al. Novel mechanism for buffering dietary salt in humans. Hypertension 2017;70:930–7. 10.1161/HYPERTENSIONAHA.117.10003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56. Rorije NMG, Olde Engberink RHG, Chahid Y et al. Microvascular permeability after an acute and chronic salt load in healthy subjects: a randomized open-label crossover intervention study. Anesthesiology 2018;128:352–60. 10.1097/ALN.0000000000001989 [DOI] [PubMed] [Google Scholar]
- 57. Babcock MC, Robinson AT, Migdal KU et al. Reducing dietary sodium to 1000 mg per day reduces neurovascular transduction without stimulating sympathetic outflow. Hypertension 2019;73:587–93. 10.1161/HYPERTENSIONAHA.118.12074 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58. Rorije NMG, Rademaker E, Schrooten EM et al. High-salt intake affects sublingual microcirculation and is linked to body weight change in healthy volunteers: a randomized cross-over trial. J Hypertens 2019;37:1254. 10.1097/HJH.0000000000002015 [DOI] [PubMed] [Google Scholar]
- 59. Braconnier P, Milani B, Loncle N et al. Short-term changes in dietary sodium intake influence sweat sodium concentration and muscle sodium content in healthy individuals. J Hypertens 2020;38:159. 10.1097/HJH.0000000000002234 [DOI] [PubMed] [Google Scholar]
- 60. Babcock MC, Robinson AT, Watso JC et al. Salt loading blunts central and peripheral postexercise hypotension. Med Sci Sports Exerc 2020;52:935. 10.1249/MSS.0000000000002187 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61. Migdal KU, Babcock MC, Robinson AT et al. The impact of high dietary sodium consumption on blood pressure variability in healthy, young adults. Am J Hypertens 2020;33:422–9. 10.1093/ajh/hpaa014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 62. Burnier M, Monod ML, Chioléro A et al. Renal sodium handling in acute and chronic salt loading/depletion protocols: the confounding influence of acute water loading. J Hypertens 2000;18:1657–64. 10.1097/00004872-200018110-00018 [DOI] [PubMed] [Google Scholar]
- 63. Trakarnvanich T, Chailimpamontree W, Kantachuvesiri S et al. Effect of a low salt diet on the progression of chronic kidney disease: a prospective, open-label, randomized controlled trial. J Prim Care Community Health 2024;15:21501319241297766. 10.1177/21501319241297766 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 64. Carey RM, Whelton PK; 2017 ACC/AHA Hypertension Guideline Writing Committee . Prevention, detection, evaluation, and management of high blood pressure in adults: synopsis of the 2017 American College of Cardiology/American Heart Association hypertension guideline. Ann Intern Med 2018;168:351–8. 10.7326/M17-3203 [DOI] [PubMed] [Google Scholar]
- 65. Crippa A, Discacciati A, Bottai M et al. One-stage dose-response meta-analysis for aggregated data. Stat Methods Med Res 2019;28:1579–96. 10.1177/0962280218773122 [DOI] [PubMed] [Google Scholar]
- 66. Hisamatsu T, Lloyd-Jones DM, Colangelo LA et al. Urinary sodium and potassium excretions in young adulthood and blood pressure by middle age: the Coronary Artery Risk Development in Young Adults (CARDIA) study. J Hypertens 2021;39:1586–93. 10.1097/HJH.0000000000002802 [DOI] [PubMed] [Google Scholar]
- 67. Filippini T, Malavolti M, Whelton PK et al. Blood pressure effects of sodium reduction: dose–response meta-analysis of experimental studies. Circulation 2021;143:1542–67. 10.1161/CIRCULATIONAHA.120.050371 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 68. Shenouda N, Ramick MG, Lennon SL et al. High dietary sodium augments vascular tone and attenuates low-flow mediated constriction in salt-resistant adults. Eur J Appl Physiol 2020;120:1383–9. 10.1007/s00421-020-04370-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 69. Visser FW, Boonstra AH, Lely AT et al. Renal response to angiotensin II is blunted in sodium-sensitive normotensive men. Am J Hypertens 2008;21:323–8. 10.1038/ajh.2007.63 [DOI] [PubMed] [Google Scholar]
- 70. Matthews EL, Brian MS, Ramick MG et al. High dietary sodium reduces brachial artery flow-mediated dilation in humans with salt-sensitive and salt-resistant blood pressure. J Appl Physiol 2015;118:1510–5. 10.1152/japplphysiol.00023.2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 71. Ramick MG, Brian MS, Matthews EL et al. Apocynin and tempol ameliorate dietary sodium-induced declines in cutaneous microvascular function in salt-resistant humans. Am J Physiol Heart Circ Physiol 2019;317:H97–103. 10.1152/ajpheart.00786.2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 72. Orsini N, Bellocco R, Greenland S. Generalized least squares for trend estimation of summarized dose–response data. Stata J 2006;6:40–57. [Google Scholar]
- 73. Bradley SH, DeVito NJ, Lloyd KE et al. Reducing bias and improving transparency in medical research: a critical overview of the problems, progress and suggested next steps. J R Soc Med 2020;113:433–43. 10.1177/0141076820956799 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 74. Ginos BNR, Engberink RHGO. Estimation of sodium and potassium intake: current limitations and future perspectives. Nutrients 2020;12:3275. 10.3390/nu12113275 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying this article will be shared on reasonable request to the corresponding author.


