Results:

The American Heart Association (AHA) currently recommends a sodium intake <1500 mg/d for the US population. The US Department of Agriculture and US Department of Health and Human Services joint 2010 Dietary Guidelines for Americans call for no more than 1500 mg/d in African-Americans, in people >51 years of age, and in people with hypertension, diabetes mellitus, or chronic kidney disease, and no more than 2300 mg/d in all others. Data from recent observational studies and a meta-analysis have suggested that current recommendations for dietary sodium intake are not appropriated in the prevention of cardiovascular disease (CVD) events. This 2012 AHA Presidential Advisory statement addresses this controversy. The authors note that the AHA recommendations published in 2011 were based on strong evidence that excess sodium intake is associated with high blood pressure, CVD, and stroke, and that reduction in sodium intake can assist in both the prevention and treatment of hypertension, and the reduction of CVD and stroke events.

Since 2011, two studies have observed a positive relationship, two a curved (J-shaped) relationship, and two an inverse relationship between sodium intake and CVD risk. Studies that observed a positive relationship included one from Japan where salt intake was identified as an independent risk factor for stroke mortality. The second study was a prospective study of US adults, which also observed a strong association between sodium intake and stroke incidence. Participants who consumed ≥4000 mg/d sodium had a stroke hazard ratio of 2.59 (95% confidence interval, 1.27-5.28) compared with participants who consumed ≤1500 mg/d sodium. In the two studies that observed a J-shape relationship, one was in 2,807 Finnish adults with type 1 diabetes, and the other was a secondary analysis of two randomized controlled trials (ONTARGET and TRANSSEND). In the Finnish study, the association between sodium and risk for all-cause mortality was positive for the majority of subjects. The AHA statement authors hypothesize that sicker participants may have reduced caloric intake, hence lower sodium intake, or may follow a lower sodium diet due to significant health concerns. Thus, the association of very low sodium intake and elevated risk for mortality may be the result of reverse causality. No results were reported for CVD and stroke mortality or nonfatal CVD events. The data that demonstrated a J-curve were a secondary analysis of two randomized clinical trials. Concerns were raised by the authors of this 2012 AHA statement (in addition to editorials and other commentary) that measurement of spot urine sodium, which was used in these studies, is subject to systematic error.

Last, the two studies that reported inverse relationships between sodium intake and risk were both conducted in high-risk populations, which thus may lack generalizability and be subject to reverse causality. One study was in 638 adults with type 2 diabetes; urinary sodium assessed by 24-hour collection was inversely associated with all-cause mortality and CVD. The cohort appears to be fairly sick, with 27.4% of subjects dying during follow-up. The AHA authors note that illness was particularly prevalent among subjects in the lowest tertile of sodium excretion; thus, reverse causality may be a concern. In the second study of 3,681 European adults, a similar finding was noted (i.e., sodium intake was inversely associated with CVD mortality). Several methodological concerns have been raised regarding this study including missing data, lack of adjustment for multiple comparisons, and reverse causality.

The 2012 AHA statement states that “clinical trials provided confirmation of previous meta-analyses indicating that a reduction in sodium lowers blood pressure in both patients with hypertension and individuals who are normotensive. The statistical power to identify the effect of sodium reduction on CVD and stroke events was limited, but a consistent trend suggested the benefit of a lower sodium intake in trials in patients with hypertension and people who are normotensive. Meta-analysis of these trials identified a statistically significant 20% reduction in CVD and stroke events.”