Randomized trials of sodium reduction: an overview

We updated a previously published overview of randomized clinical trials testing the effects of reducing sodium intake. We excluded trials that had confounded designs, enrolled preadolescent study populations, tested intakes outside the usual range for the US population, or reported neither systolic nor diastolic blood pressure. Thirty-two trials with outcome data for 2635 subjects were included. Two reviewers abstracted information independently and differences were reconciled. Pooled blood pressure differences between treated and control groups were highly significant for all trials combined and for trials in hypertensive and normotensive subjects pooled separately. The effects on blood pressure of lowering sodium in hypertensive and normotensive subjects, respectively (each trial weighted according to sample size), were -4.8/-2.5 and -1.9/-1.1 mm Hg (systolic/diastolic). Median differences in sodium excretion between sodium-reduction and control groups in these subgroups were -77 and -76 mmol/24 h, respectively. Weighted linear-regression analyses across the trials showed dose responses, which were more consistent for trials in normotensive subjects. These associations were, per 100 mmol Na/24 h, -5.8/-2.5 and -2.3/-1.4 mm Hg in hypertensive and normotensive subjects, respectively. There is no evidence that sodium reduction as achieved in these trials presents any safety hazards. The blood pressure reduction that would result from a substantial lowering of dietary sodium in the US population could reduce cardiovascular morbidity and mortality.