Current issues of ACP Journal Club are published in Annals of Internal Medicine


Meta-analysis: Reduced dietary sodium decreases only systolic blood pressure in persons who are hypertensive or normotensive

ACP J Club. 1996 Nov-Dec;125:74. doi:10.7326/ACPJC-1996-125-3-074

Source Citation

Midgley JP, Matthew AG, Greenwood CM, Logan AG. Effect of reduced dietary sodium on blood pressure. A meta-analysis of randomized controlled trials. JAMA. 1996 May 22/29;275:1590-7.



To determine whether restriction of dietary sodium lowers blood pressure (BP) in patients who are hypertensive or normotensive.

Data sources

English-language studies were identified by searching MEDLINE (January 1966 to February 1994) and Current Contents using the keywords hypertension, blood pressure, vascular resistance, sodium and dietary, diet and sodium restricted, sodium chloride, clinical trial, randomized controlled trial, and prospective studies. Bibliographies of review articles and personal files were also searched.

Study selection

Studies were selected if they were randomized controlled trials that evaluated a dietary sodium intervention, the intervention was monitored by timed sodium excretion, and both systolic and diastolic BP were included as outcome measures. Studies that reported adjuvant use of antihypertensive medication were excluded.

Data extraction

2 observers independently extracted data on trial design, patient characteristics, intervention, outcomes, and study quality.

Main results

56 trials met the selection criteria. These trials showed statistical heterogeneity. The pooled mean reduction in daily urinary sodium excretion was 95 mmol/d (95% CI 71 to 119 mmol/d) in the 28 trials that involved patients with hypertension and 125 mmol/d (CI 95 to 156 mmol/d) in the 28 trials that involved persons who were normotensive. In trials that involved patients with hypertension and after adjusting for measurement error of urinary sodium excretion, the decrease in BP for a reduction in daily sodium intake of 100 mmol/d was 3.7 mm Hg (CI 2.4 to 5.1 mm Hg) for systolic (P < 0.001) and 0.9 mm Hg (CI -0.1 to 1.9 mm Hg) for diastolic (P = 0.9). Corresponding results in trials that involved persons who were normotensive were 1.0 mm Hg (CI 0.5 to 1.6 mm Hg) for systolic (P < 0.001) and 0.1 mm Hg (CI -0.32 to 0.51 mm Hg) for diastolic (P = 0.64). The BP response to dietary sodium intervention was greater in trials with patients who had hypertension and were > 45 years of age (6.3 mm Hg per 100 mmol/d reduction of sodium intake for systolic BP, CI 4.1 to 8.4 mm Hg, P < 0.001; and 2.2 mm Hg for diastolic, CI 0.6 to 3.9 mm Hg, P = 0.01).


Dietary sodium restriction leads to a reduction in systolic blood pressure in both hypertensive and normotensive persons. The decrease in diastolic blood pressure is small and nonsignificant. This dietary intervention effect is greater for both systolic and diastolic blood pressure in older patients with hypertension.

Sources of funding: Campbell's Institute for Research & Technology, New Jersey, and Medical Research Council of Canada.

For article reprint: Dr. A.G. Logan, Samuel Lunenfeld Research Institute, Mount Sinai Hospital, Suite 850, 600 University Avenue, Toronto, Ontario M5G 1X5, Canada. FAX 416-586-8434.


The popularity of sodium reduction to lower BP has fluctuated for many years. Most patients believe that their diet is low in sodium if no salt is added during cooking or at the table. Unfortunately, 80% of sodium intake comes from salt already added in processed food.

This meta-analysis by Midgley and colleagues includes more randomized controlled trials than did 2 previously published overviews of sodium reduction and its effect on BP (1, 2), and the reductions in BP noted in this meta-analysis were lower. The heterogeneity of results means that they should be interpreted with caution. Systematic differences may exist among the underlying trials in such important factors as age, weight, alcohol intake, and salt sensitivity of the patients studied and in the duration of follow-up. The results of the reviews consistently show that older patients who are hypertensive are more likely to be salt sensitive than are persons who are normotensive. For persons who are salt-sensitive, greater reduction in sodium intake may correlate with greater reduction in BP (3).

The mean systolic BP reduction in older patients who are hypertensive (6.3 mm Hg per 100 mmol/d of reduced sodium) may compel some clinicians to widely recommend dietary salt reduction. Some debate, however, is sparked by observational studies that a low-salt diet may have adverse metabolic and clinical effects (4).

Because the sodium intake of persons who live in developed countries is about 150 to 200 mmol/d, reducing sodium by 100 mmol/d will work only in highly motivated patients and many would require assistance from a dietitian. Therefore, it cannot be recommended for normotensive persons but may be useful for older, hypertensive patients who wish to avoid medication or who cannot tolerate antihypertensive agents.

Bruce Arroll, MD
The University of AucklandAuckland, New Zealand


1. Law MR, Frost CD, Wald NJ. BMJ. 1991;302:819-24.

2. Cutler JA, Follmann D, Elliott P, Suh I. Hypertension. 1991;17(Suppl 1):I27-33.

3. MacGregor GA, Markandu ND, Sagnella GA, Singer DR, Cappucciao FP. Lancet. 1989;2:1244-7.

4. Alderman MH, Madhavan S, Cohen H, Sealy JE, Laragh JH. Hypertension. 1995;25:1144-52.