Faulting Salt? New PURE Analysis Argues Against Low Sodium Intake
There seems to be growing agreement, however, that ultimately a randomized trial will be needed to resolve the so-called salt wars.
Another salvo in the ongoing battle between proponents of restricting dietary sodium to low levels and those advocating for more moderate intakes has been fired in the form of a new community-level analysis from the PURE study. The analysis supports the idea that consuming about 3 to 5 grams of sodium per day is the “sweet spot” in terms of cardiovascular risk.
Though risks of cardiovascular events and stroke increased along with sodium intake, the relationships were observed only in communities consuming more than 5.08 grams per day on average. At lower levels of average intake, risks of MI and mortality declined with greater sodium consumption and the association between sodium and stroke was no longer significant, researchers led by Andrew Mente, PhD (Population Health Research Institute, Hamilton, Canada), report in a study published in the August 11, 2018, issue of the Lancet.
Additionally, greater potassium intake was strongly associated with improved outcomes, independent of sodium consumption.
“What our findings do is they reaffirm our previous findings . . . that about 3 to 5 grams [of sodium] per day appears to be the optimal level associated with lowest risk,” Mente told TCTMD.
Communities with that level of average intake “don’t need to worry about sodium,” he added, “and it’s best that we focus on improving the overall quality of the diet by eating more foods that are higher in potassium. That’s the bottom line.”
That advice regarding sodium, however, is at odds with recommendations from the World Health Organization, which advises consuming fewer than 2 grams per day, and from the American Heart Association, which recommends eating no more than 2.3 grams per day, with an ideal limit of 1.5 grams.
According to Franz Messerli, MD (University Hospital, Bern, Switzerland), who co-authored an editorial accompanying the study, both organizations should reconsider their recommendations because it’s not reasonable to expect people to maintain those levels of intake over the long term.
“The 1.5 grams of sodium is just absolutely not realistic,” he told TCTMD. “I have yet to see a patient who is willing to do so and can do so over weeks, months, and years. It just does not happen.”
The approach advocated by Mente et al in their paper—to focus sodium restriction efforts on communities with high levels of average intake (over 5 grams per day)—is “a reasonable way of going,” Messerli said.
“What we have taught for years—that everybody should restrict salt intake—is wrong,” he said. “We should agree that once you have high blood pressure or once you are at risk of stroke . . . it’s probably a good idea to restrict salt intake some. But otherwise I think we can be pretty liberal in terms of the amount of salt we consume.”
Not surprisingly, the AHA disagrees. In an emailed statement from the organization, AHA volunteer Frank Sacks, MD (Harvard School of Public Health, Boston, MA), said: “Most Americans are consuming more sodium than their body needs. Most Americans also underestimate how much sodium they eat. A vast body of diverse research tells us that lowering sodium intake lowers blood pressure, which can save lives.”
Sacks cited some limitations of studies like this one—including a lack of multiple 24-hour urine measurements to assess sodium intake and the problem of reverse causation—that can impact the accuracy of the results.
“Let’s not let studies like this distract us from the truth, which is based on strong scientific evidence on the harms of high sodium intake,” Sacks said. “Lowering sodium can improve health and even save lives. The majority (more than 70%) of the sodium Americans eat comes from processed, prepackaged, and restaurant foods. We need to make a change in the foods offered to us, and that means that the community, healthcare providers, governments, and professional organizations must work together.”
High Sodium Intake Clustered in Chinese Communities
For this new analysis, Mente and colleagues looked at data from 18 of the 21 countries participating in the PURE study on adults ages 35 to 70 without cardiovascular disease at baseline. They calculated community-level sodium and potassium intake using morning fasting urine and examined relationships with blood pressure (in 95,767 people living in 369 communities) and cardiovascular outcomes and mortality (in 82,544 people living in 255 communities).
High levels of sodium intake were more likely to be seen in China than in other countries. In China, 80% of communities had an average intake exceeding 5 grams per day, whereas the majority of communities elsewhere (84%) had an average intake between 3 and 5 grams per day.
An increase in community-level sodium intake by 1 gram per day was associated with an average increase in systolic blood pressure of 2.86 mm Hg, but the relationship was seen only in areas with the highest levels of sodium intake.
We like smoked salmon, we like Big Macs and all the salty food, and it’s extremely unlikely that this will happen in the general population. Franz Messerli
Similarly, as mean sodium intake increased, so did risks of overall cardiovascular events and stroke. But again, the relationships were only observed in the highest tertile of sodium intake (< 5.08 grams per day).
For MI, there was an inverse relationship between sodium intake and risk in the lowest tertile (< 4.43 grams per day), and for all-cause mortality, there was an inverse relationship with risk in the bottom two tertiles.
“Rather than a population-wide approach, our data suggest that a targeted approach of intervening in communities and countries with high mean sodium intakes (> 5 g/day) might improve reduction of cardiovascular disease (and strokes),” Mente et al write.
“Such an approach would avoid diversion of resources to communities with lower sodium intakes, where no correlation with increased rates of clinical events are seen, and those in which associations with BP are small,” they explain. “If the inverse associations between low sodium intake and increased rates of myocardial infarction and death are real, such a targeted strategy would minimize the potential for harm by sodium reductions in populations with average sodium intake.”
Going to Prison to Settle the Issue
With both sides of the sodium debate seemingly entrenched, the question becomes: how will this be resolved?
According to Mente, there’s seems to be growing agreement that a randomized controlled trial comparing low levels of sodium intake to more moderate levels is needed. He pointed to a paper published last month in Hypertension that suggested that prisons might an ideal setting for conducting just such a study.
“It’s very encouraging that there is increasing consensus that a large randomized controlled trial is needed, and ultimately that’s what’s needed to settle the debate,” Mente said, who added that “the fact that a prison population is proposed as a way to make the study feasible just really strongly speaks to how off the charts the current recommendations are.”
Both Mente and Messerli agreed that trying to perform a trial that carefully controls how much sodium a person consumes would not be feasible in the general population.
“It’s very simply not going to happen,” Messerli said. “We like smoked salmon, we like Big Macs and all the salty food, and it’s extremely unlikely that this will happen in the general population.”
While awaiting results from a randomized trial, Mente said it’s best to proceed with caution. Instead of focusing on sodium, we should “instead focus on improving the overall quality of the diet and get people to eat more potassium-rich foods like fruits and vegetables, dairy, potatoes, beans, nuts—that stuff.”
In communities with very high average sodium intakes, like many in China, it would likely be beneficial to reduce sodium consumption, he added. “But to reduce intake in communities with moderate levels down to low levels, there’s no evidence it will result in benefit and there’s a lot of evidence that it may even be harmful.”
Mente A, O’Donnell M, Rangarajan S, et al. Urinary sodium excretion, blood pressure, cardiovascular disease, and mortality: a community-level prospective epidemiological cohort study. Lancet. 2018;392:496-506.
Messerli FH, Hofstetter L, Bangalore S. Salt and heart disease: a second round of “bad science”? Lancet. 2018;392:456-458.
- Mente reports no relevant conflicts of interest.
- Messerli reports having received grants and speaking honoraria from and serving on advisory boards for Menarini, Servier, Pfizer, Novartis, Medscape, and Medtronic.