While it is accepted that lowering sodium intake from a high level is beneficial for health by lowing blood pressure, there is a lack of consensus on how far sodium intake should be lowered and whether it is even possible to achieve the reductions recommended in current guidelines, say experts.
Consequently, clinicians should perhaps move away from giving strict numeric targets to their patients for salt intake and counsel more "practical" approaches such as avoiding food with a high salt content.
That was the suggestion during a debate on 29th June as part of Food for Thought 2020: The science and politics of nutrition, hosted by the Swiss Re Institute in partnership with The BMJ, and held virtually due to the coronavirus pandemic.
How Much Is Too Much?
Professor Nancy Cook, Division of Preventive Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, US, opened the discussion by summarising the accompanying paper she led for The BMJ, noting that one of the main questions that arose was "how much" sodium effects health.
However, there are, she said, "a number of different thoughts on that", resulting in debate over the optimum goal of sodium intake, with many guidelines from across the world differing on what to recommend.
Guidelines from the UK suggest a consumption limit of 2400 mg a day, while the American Heart Association goes further, suggesting a limit of 1500 mg per day in most adults.
"So the question is: how low to go?"
Prof Cook points out that small amounts of salt are necessary for ongoing health but large amounts can have adverse effects, including fluid retention, and lead to higher blood pressure through volume expansion, "but where in between is the optimal range?".
This issue is compounded by difficulties in measuring sodium intake, not least due to being unable to determine from dietary questionnaires the exact sodium content of the foods consumed. Urine measurement is indicative, as 90% of consumed sodium is secreted via the urine. However, repeated 24-hour measurements, which is the best assessment of sodium intake, is impractical.
Despite these issues, Prof Cook said that, for their paper, the authors were able to identify some areas of agreement, primarily that lowering sodium levels lowers blood pressure in individuals with hypertension in a dose–response relationship. For individuals with high-normal blood pressure levels, or stage 1 hypertension, the effect is "small", however.
The impact of sodium reduction in individuals with lower than average blood pressure is "unclear", Prof Cook added, which may reflect genetic variations, including around sodium sensitivity.
She also said that there is general consensus that very high levels of sodium, such as over 5000 mg per day, are associated with cardiovascular disease risk, "especially in those with hypertension".
Blood Pressure Impact
However, the consensus breaks down when it comes to the effect of sodium lowering on hypertension, with the many studies having been conducted so far leading to conflicting results.
Prof Cook therefore asked: "What can we do?" She said that many people have advocated for a long-term trial, but that is "very difficult, because you're taking a nutrient away".
This is not the case with calcium, vitamin D, or fish oil, where individuals are given a supplement to increase levels - but it would not be ethical to give a sodium supplement to artificially raise levels.
Another approach is salt substitution, which is being tested in a number of ongoing trials, including a study in China involving almost 21,000 individuals that should report during 2020.
However, Prof Cook asked the question whether, with all the data that has been amassed already, including from observational studies: "Do we really need to wait for a long-term trial for another 5 or 10 years before we can implement population recommendations?"
Taking up the baton, Martin O'Donnell, professor of translational medicine, National University of Ireland Galway, who said that he agreed with "most" of what Prof Cook said, adding that he would like to "amplify" some points.
Regarding the impact of increased sodium intake on blood pressure, he said that, "as you transition from high to moderate", compared to moderate to low, blood pressure, "you have a larger magnitude of blood pressure reduction per gram of sodium reduction...and when you go from moderate to low [blood pressure] you get activation of neurohormonal systems".
"These are necessary systems in the body because of the essential nature of sodium, so you're invoking almost salvage systems to promote retention of sodium”, Prof O'Donnell said, some of which have been associated with increased cardiovascular risk.
This has been demonstrated in epidemiological studies, with a J-shaped association between sodium levels and risk, with the lowest risk seen around 2700 mg per day.
This, Prof O'Donnell said, "is the expected relationship between an essential nutrient and health", and is seen with other electrolytes, such as magnesium, calcium and potassium, with an increased risk with lower consumption.
Trial Data
He continued that one of the main reasons that outcome studies on sodium intake lowering have reported differing results is due to many being underpowered to draw definitive conclusions.
"But of critical importance is there's no large trial that's achieved sustained low sodium recommended by the current guidelines. So in truth we don't know the safety or the effect of sustained intakes at this level in general populations," Prof O'Donnell said.
He suggested that clinicians should "remove the numeric targets when counselling patients and emphasise more practical considerations," such as "avoiding food with high salt intake and behaviours that may be associated with high salt intake".
While supporting recommendations to bring average intakes below 5000 mg per day, Prof O'Donnell said: "What we don't know is whether we should reduce populations from moderate levels down to low intakes recommended by guidelines.
"We don't have clinical trials and we don't have sustained reductions to these levels to confirm feasibility. We don't have effective interventions at maintaining that level, and critically we don't have evidence associated with cardiovascular disease."
He asked: "What evidentiary basis should we demand to inform public health policy and nutrition, particularly when that nutrient is an essential electrolyte?
"Our contention would be that before you manipulate an essential electrolyte to levels that we have very little experience [of], it needs to be informed by definitive randomised controlled trials."
Confounding Studies
Graham MacGregor, professor of cardiovascular medicine, Wolfson Institute of Preventive Medicine, and Honorary Consultant, Barts & The London, spoke next, saying that humans "come from a salt-poor environment", and consequently the current high salt intake is a consequence of the foods we now eat.
He went on to say that "10 million people die of raised blood pressure a year", making it the "biggest cause of death in the world by far", of which "three million are due to eating too much salt".
Confounding studies on sodium intake, however, is that the Kawasaki equation is "contaminated" by including age and weight, "both of which are potent risk factors for death".
This, he believes, is the cause of the J-shaped curve, which he says is an "artefact of using a very peculiar equation" to calculate sodium intake.
Nevertheless, countries that have successfully reduced salt intake on a population level, specifically Japan, Finland, and the UK, have shown a fall in intake, a fall in blood pressure, and a fall in cardiovascular mortality, Prof MacGregor said.
"Not all of that fall is due to the reduction in salt," he added, noting that factors such as smoking have contributed to the reduction in mortality, "but if we look at how many deaths were due to that fall in population blood pressure, it's 9000 fatal events...prevented in one year from cardiovascular disease", with an estimated reduction in healthcare costs of £1.5 billion.
"So I feel the evidence, like for all dietary things, is controversial because we can't do the studies we want, but I would not advocate stopping reducing salt intake," Prof MacGregor said, adding that he endorsed the World Health Organisation's recommendation to reduce salt intake from 10,000 mg to 5000 mg per day.
Finally, Franz Messerli, professor of medicine, Swiss Cardiovascular Center, University of Bern, Switzerland, put his case forward, saying that some papers have demonstrated no difference in blood pressure between individuals on a low sodium versus a high sodium diet, despite some differences in hormonal levels.
He highlighted that there was, however, a difference in weight between the two groups. This, he said, implies that individuals on a high salt diet were "volume expanded", and "obviously this creates a powerful protection against the risk of volume depletion, which may be very common in the summer and in hot and humid tropical countries".
Prof Messerli also said that studies have shown that there is a positive correlation between sodium intake and outcomes such as mortality only "up to a certain point", with women in Hong Kong having the longest lifespan in the world despite having an average daily intake of sodium of 8000 mg.
He believes, therefore, that sodium recommendations should be weight adjusted, and that the data "argue against salt being the cardiac nemesis, or a devastating plague of mankind, which has been argued time and again".
Controlled Environments
In the subsequent open discussion, Fiona Godlee, editor-in-chief of The BMJ, who chaired the debate, asked what prospects there are for "better, more reliable" research in the field.
Prof Cook replied that the answer is to develop higher quality measures of sodium, and look at other types of studies, such as those in cosmonauts, to understand day-to-day variations in physiology in carefully controlled environments.
However, a trial per se would be "very difficult", as, leaving aside the question as to whether sodium reduction can even be achieved, it would require a highly intensive lifestyle intervention, rather than simply "giving a pill".
In addition to which, it would be "very expensive, much more so than the other types of studies we do...in giving supplement pills".
Prof O'Donnell added that there is the "obvious contradiction that we believe that randomised controlled trials can't achieve low sodium intake, but we're still recommending it in guidelines", which is a "definite challenge".
He believes that observational studies have "a large contribution to make, but I would also think that if we're going to invest this level of energy in recommending an intervention to reduce sodium intake, we're obliged to determine the feasibility and effectiveness of it".
Prof MacGregor pointed out that those kinds of outcome studies are also lacking for reducing sugar intake, for increasing fruit and vegetable consumption, and for reducing weight.
He also took issue with Prof Messerli's assertions over the relationship between salt consumption and life expectancy in Hong Kong, as Prof MacGregor has been advising the government there on how to reduce salt intake on a population level and ascribes their longevity to factors such as large fruit and vegetable consumption, and low dietary fat and sugar.
Goldilocks
The panel then discussed how clinicians could individualise salt intake recommendations to the patient they have in front of them, which Navjoyt Ladher, head of scholarly comment at The BMJ, who was putting together questions from comments from the audience, characterised as finding the "Goldilocks figure...of not too much and not too little but just right".
Prof O'Donnell replied by saying that there is "the very occasional person who adheres to the advice that they're given, sometimes with very low sodium intake," who will then report symptoms of orthostatic hypotension.
"Then the guideline recommendations are to increase salt intake, so...it goes back to the old maxim: everything in moderation."
In other words, "excessive restriction may cause harm, and there's certainly no evidence of benefit," and he believes that "we're setting up our patients to fail by giving them a target recommendation they can't interpret, they can't measure, and there's no intervention within which they can achieve it".
Therefore, for Prof O'Donnell, pitching the salt intake recommendations, where "we have consensus, where we know there is increased disease, makes sense", which for him is between 3000 and 5000 mg of sodium per day, while recognising that there may be benefits in some populations of reducing the target.
Prof MacGregor said, however, that the emphasis should not be on the individual but rather on the food industry to "take out the salt, sugar and fat" from processed foods, and "encourage them to reformulate with fruit and vegetable residues" so that intake can be changed without "people really knowing".
Following the debate, John Schoonbee, global chief medical officer at Swiss Re, commented that the debate raises a "really difficult" issue, which is that, despite sodium being "a single element that's essential to life...questions remain and consensus eludes us", which is a "real reminder of how challenging the intersection is between nutrition and health".
Open access fees for the paper were funded by Swiss Re.
Prof Cook has received past research funding from the National Institutes of Health and the American Heart Association, and consultancy fees from Jazz Pharmaceuticals and OptumCare. She was a member of the Committee to Review the Dietary Reference Intakes for Sodium and Potassium, National Academies of Sciences, Engineering, and Medicine, 2019. is chairman of Blood Pressure UK, chairman of Consensus Action on Salt and Health and chairman of World Action on Salt and Health.
No other conflicts of interest declared.
Food for Thought 2020: Should current salt guidelines be taken with a "pinch of salt"? Presented June 29th 2020.
BMJ 2020;369:m2440 doi: 10.1136/bmj.m2440
