Lowering dietary sodium intake to reduce blood pressure: cum grano salis

Comment on ‘Effect of dietary sodium on blood pressure: a crossover trial’ which was published in JAMA, doi: 10.1001/jama.2023.23651. The Coronary Artery Risk Development in Young Adults—Salt Sensitivity of Blood Pressure (CARDIA-SSBP) is a prospective study with allocated diet order crossover design aiming to compare blood pressure (BP) responses in individuals exposed to high or low dietary sodium intakes.1 Inclusion criterion was age between 50 and 75 years, and key exclusion criteria were systolic or diastolic BP (SBP or DBP) outside the range of 90–160 or 50–100 mmHg at enrolment visit, respectively, and resistant hypertension. Participants were allocated to either 1-week high-sodium diet first (followed by 1-week low-sodium diet) or 1-week low-sodium diet first (followed by 1-week high-sodium diet). Allocation to diet order alternated by weekday at each site. High-sodium diet was achieved pragmatically by adding ∼2200 mg sodium daily to each individual’s usual diet using 2 bullion packets, whereas low-sodium diet was standardized to provide daily averages of ∼500 mg of sodium, ∼4500 mg of potassium, and ∼1000 mg of calcium. Both 24-h ambulatory BP monitoring and 24-h urine collections (for urine sodium and creatinine level assessment) were performed starting the day before baseline, end of first week, and end of second week. Main outcome measures were average 24-h ambulatory SBP and DBP, mean arterial pressure (MAP), and pulse pressure. From a total of 228 individuals enrolled in 2 US cities, 213 participants (mean age, 61 years; 65% female; 64% Black) attended all the planned study visits and were included in the analysis (118 high-sodium diet first and 95 low-sodium diet first). The study included individuals with normal BP, controlled hypertension, uncontrolled hypertension, and untreated hypertension. Median body mass index (BMI) of enrolled individuals was 31 kg/m2, and more than 20% had diabetes. Sodium levels of usual diets were high [median, 4.5 g/d; interquartile range (IQR), 2.6–6.9 g/d], increasing to a median of 5.0 g/d (IQR, 3.3–7.6) after high-sodium diet and decreasing to a median of 1.3 g/d (IQR, 0.5–3.8) after low-sodium diet. The median intra-individual difference in MAP between high- and low-sodium diets was 4 mmHg (IQR, 0–8 mmHg; P < .001), regardless of hypertension status and antihypertensive medication use. Mean arterial pressure declined in 73% of individuals after low-sodium diet, as compared with high-sodium diet, while it remained stable in 1% and increased in 26% of them, with similar trends for SBP, DBP, and pulse pressure. In the overall cohort, 46% of individuals had a 5 mmHg or greater reduction in MAP defining ‘salt sensitivity’ in response to a change in sodium dietary intake, with a median difference in 24-h urine sodium of 3.4 g (IQR, 1.4–4.9 g; P < .001) between high-sodium and low-sodium diets. As compared with individuals allocated to a high-sodium diet, those receiving a low-sodium diet had a 8 mmHg lower mean SBP (95% confidence interval, 4–11 mmHg; P < .001) at the end of the first dietary intervention week, an effect comparable with that of a common first-line antihypertensive medication. A comparable response was observed at the end of the second diet week, without evidence of sequence or carryover effects. This finding was consistent across multiple subgroups of age, sex, race, hypertension status, baseline BP, diabetes, and BMI. Substantial evidence supports a causal relationship between dietary sodium intake and increase in BP, and excessive sodium consumption (defined by the World Health Organization as >5 g sodium per day) has been linked with the onset of hypertension and its cardiovascular (CV) complications.2 For this reason, most guidelines recommend a low dietary salt intake, defined as <2 g of sodium (equivalent to <5 g of sodium chloride) daily, among the lifestyle interventions for patients with hypertension or high-normal BP.3 The putative mechanisms underlying the increase in BP as a consequence of high sodium intake include water retention, volume expansion, increased systemic peripheral resistance (as high sodium triggers vascular remodelling in small resistant arteries as well as in large elastic arteries), endothelial dysfunction, and autonomic dysregulation of the CV system.2,4 While substantial evidence supports an overall increased risk of CV events and mortality with high sodium intake,5 the effect of reducing dietary sodium on CV events remains controversial.6 A number of cohort studies and meta-analyses have shown a J-curve phenomenon in the relationship between dietary sodium intake and prognosis, with low sodium intake (<3 g per day) being paradoxically associated with higher risk of CV events and mortality.7 Based on this conflicting evidence, some argue that sodium intake restriction should be reserved only to subjects with high sodium dietary intake and ‘salt-sensitive’ hypertension. The CARDIA-SSBP study found that dietary sodium restriction lowered MAP (i.e. the primary outcome measure) by a median of 4 mmHg as compared with a high-sodium diet at the end of a 1-week study period, in a small sample of middle-aged to elderly individuals.1 Major strengths of the study are the crossover design enabling multiple within-individual and between parallel-group analyses and the enrolment of individuals across a broad range of clinical conditions (from normotensive subjects to patients with treated and untreated hypertension). Of note, both within-individual and between-group declines in BP from a high-sodium to a low-sodium diet were independent of hypertension status and antihypertensive medication use. Almost three out of four individuals experienced a decline in MAP after a low-sodium diet, whereas 26% of them showed an increase. While the lack of BP response to dietary sodium restriction might be interpreted as a ‘salt-insensitivity’, the potential for measurement error cannot be completely ruled out. In fact, estimating dietary sodium based on 24-h urine samples reflects ∼90% of intake and can also be limited by incomplete collection. Furthermore, urinary sodium excretion varies enormously from day to day and is affected by concomitant clinical conditions, medications, and hormonal fluctuations.6 In addition, BP is highly variable, even when measured using ambulatory BP monitoring and intra-individual BP variability is commonly observed. In fact, ambulatory BP variability is influenced by many factors, including psychological stress, physical activity, and hours of sleep,8 which were not assessed in this study. Another potential reason for the paradoxical response to sodium restriction observed in a sizable proportion of individuals might be dietary non-adherence (the median urine 24-h sodium levels were 1.3 g daily during the low-sodium diet, almost three-fold higher than expected based on the prescribed diet). Moreover, participants received each diet for only 1 week, without a washout period, while previous observations suggest that several weeks are required to achieve a steady-state effect of dietary sodium restriction.9 Of note, the low-sodium diet included high intake of potassium and calcium which might have contributed to lower BP. Interestingly, epidemiological and interventional studies suggest that high potassium intake not only lowers BP but also strikingly reduces ‘salt sensitivity’, through both direct effects on the endothelium and inhibition of a newly discovered kinase signalling pathway controlling a ‘potassium switch’ in the distal nephron.10 Finally, the study had a limited sample size, if one considers the multiple heterogeneous BP profiles included in the study. Also, two-thirds of the participants were Black and the population was substantially overweight. This is particularly relevant as ‘salt sensitivity’ is usually considered a hallmark of hypertension in Black people, while it is significantly less prevalent in White hypertensive individuals.11 The features of the population selected for this study may have resulted in an overestimation of the prevalence of ‘salt sensitivity’, thus limiting the extension of the findings to the general hypertensive population. In conclusion, the CARDIA-SSBP study provides evidence that dietary sodium restriction is able to lower BP in a high proportion of people within a population of middle-aged to elderly, mostly Black American, adults and that the observed BP reduction with low-sodium diet compared with the high-sodium regimen is independent of the hypertension status and the concomitant use of antihypertensive medications. The long investigated and debated issue of the clinical relevance of ‘salt-sensitive’ hypertension still requires further efforts to more easily identify ‘salt-sensitive’ individuals in daily practice, in whom dietary restriction of sodium may translate into clinical benefits. Nonetheless, population-based initiatives should continue to be implemented to discourage excessive consumption of sodium and to promote the public awareness about its potential disadvantages for health. R.V. received consulting or lecturing fees from Abbott Vascular, Abiomed, Daiichi Sankyo, Amgen, Medtronic, and Terumo, outside the submitted work. M.V. reports personal fees for speaker bureau and/or consulting in advisory boards from AstraZeneca, Menarini Int, Novartis Pharma, Novo Nordisk, and Sanofi Pasteur, outside the submitted work.