乌鸦传媒

Despite substantial observational and experimental evidence linking sodium intake and blood pressure (BP), a full understanding of the dose-response relationship has been elusive because of the limitations in study designs and analytic methods. Additionally, the effect of sodium intake on BP in people without hypertension has been a matter of vigorous scientific discussion.

What鈥檚 new here?

To better characterize BP across a wide range of sodium exposure in people with and without hypertension, Filippini and colleagues conducted a dose-response meta-analysis using a novel statistical approach.

The analysis included 85 randomized controlled trials with 4 weeks to 36 months of follow-up. All trials manipulated sodium intake through diet and/or supplementation, measured systolic and diastolic BP, and assessed sodium intake with 24-hour urinary sodium excretion. The trials enrolled participants with hypertension (n=65), without hypertension (n=11), or with and without hypertension (n=9). The researchers used a 1-stage cubic spline mixed effect meta-analytic model for aggregated data, which renders a smoother curve than that yielded by previously used methods, even when working with trials that had only 2 intervention levels.

The median difference in sodium excretion between intervention and control groups was 80 mmol/day (1.8 grams/day), ranging from 5 to 309 mmol/day (0.1 to 7.1 grams/day). In the dose-response analysis, the reference value was set as 87 mmol/d (2 grams/day), corresponding to the range of maximum safe intake levels endorsed by various international and US health organizations.

The analysis demonstrated a positive relationship between achieved sodium intake and mean systolic and diastolic BP across a wide range of sodium excretion. The relationship was approximately linear. There was little change in the relationship at the highest or lowest levels of sodium exposure, even below 2 grams/day.

The dose-response slope for systolic BP was steeper than for diastolic BP. A reduction in mean systolic BP of 2.42 mmHg (95% CI, -1.97 to -2.87) and diastolic BP of 1.01 mmHg (95% CI, -0.72 to -1.31) was observed for every 1 gram/day decrease in sodium excretion. Because of the novel statistical approach, these estimates were more precise than prior meta-analyses.

The effect of sodium exposure was fairly consistent across population subgroups, but was greater in participants with a diagnosis of hypertension at baseline compared to those without hypertension. Among participants with hypertension, systolic BP increased by 10.31 mm Hg (95% CI, +7.86 to +12.75) and diastolic BP by 5.13 mm Hg (95% CI, +3.52 to +6.74) when sodium intake rose from 2 grams/day to as high as 6 grams/day. In participants without hypertension, systolic BP increased by 3.99 mm Hg (95% CI, +0.80 to +7.18) and diastolic BP by 1.66 mm Hg (95% CI, 鈭�0.58 to +3.91).

Participants with higher baseline sodium intake had greater decreases in BP after sodium restriction. The dose-response relationship for both systolic and diastolic BP was stronger in trials with interventions based on dietary modification compared to those based on sodium supplementation. The effect of sodium reduction on BP was little influenced by intervention duration, use of antihypertensive medication, and participants鈥� sex.

The authors concluded that the dose-response relationship between sodium intake and systolic and diastolic BP was largely linear across the entire range of sodium exposure included in the individual trials.

How low can we go?

In an accompanying editorial, Neal and Wu argue that because a beneficial effect of sodium intake reduction on blood pressure was found below a threshold of 2 grams/day, the findings suggest that the global disease burden currently attributed to sodium is likely underestimated. The editorialists also call for renewed public health strategies to reduce sodium intake levels to 2 grams/day and even lower in order to prevent cardiovascular morbidity and mortality.

The Filippini meta-analysis did not directly address the effect of sodium intake on hard cardiovascular endpoints. Prior studies have observed a U-shaped curve, with both low and high sodium intake linked to cardiovascular risk. However, Neal and Wu argue that U-shaped association is likely driven by reverse causation as well as methodologic limitations inherent in large studies which used spot urine samples (instead of 24 hour samples) to estimate sodium intake.

Published: June 29, 2021