Even 'charitable bodies' like Consensus Action on Salt and Health (CASH) are exhorting us all to lay off the tasty white crystals, telling us in a recent press release: "Salt . . . is a direct toxin that puts up our blood pressure, which is the biggest cause of strokes, heart attacks and heart failure, and is the commonest cause of death and disability in the UK."
The notion that salt is toxic can be traced back to the 1970s-incidentally, the decade that also saw the rise of scares about cholesterol, alcohol and high-fat foods. For years, doctors had noted a connection between high blood pressure and high salt intakes. However, while salt in high quantities was considered "noxious", the actual evidence that it caused high blood pressure was "conflicting".
Nevertheless, such was the finger-wagging medical zeitgeist at the time that, in 1977, the US medical authorities decided the public should be "educated" about salt and high blood pressure, or hypertension. According to the Dietary Goals for the United States, 2nd edition, prepared in 1977 by the Select Committee on Nutrition and Human Needs,3 the maximum intake decreed was set at about 5 g/day, or roughly a teaspoonful. Other developed countries, as well as the World Health Organization (WHO), dutifully fell in line.
And yet, within a few years, the whole initiative was being questioned-and from within the medical profession itself.
In 2000, the American Journal of Clinical Nutrition published a scathing report on the official salt guidelines, which were roundly condemned as "a rush to judgment" that lacked "controlled trials" and "carefully executed observational studies", including any evidence that salt was responsible for deaths from heart disease.
One of the key "unproved postulates" was that populations with high salt intakes (for example, the Japanese) had higher blood pressure than those with low salt intakes (such as Eskimos)-and yet this was "the cornerstone of the salt/blood pressure myth" (see page 38 for official WHO figures).
The response was predictable. Questioning the science behind the salt guidelines was an "aggressive campaign"; the by-then recommended US limit of 2.4 g/day was reasserted as "critically important" for reducing high blood pressure in the general population-and indeed might not be "low enough".
This tussle has continued to this day, with salt hawks confidently pointing out that increasing salt increases blood pressure. But that's not a universal finding, reply the doves: in large-scale trials, some people have an increase in blood pressure, but not others, and a few show reduced blood pressure.
The hawks cite trials showing that, at a population level, reducing salt intake has apparent health benefits. In a comparative study of more than 3,000 people with prehypertension, those who reduced their salt intake to about 6 g/day had a 25 per cent reduction in risk of "cardiovascular events", and a corresponding reduction in actual deaths.
The trouble is that, despite literally hundreds of studies on the salt/health connection, the outcomes don't all agree. That's why the many media stories we've read over the years have been so confusing. One British newspaper ran a 2011 headline announcing "Now salt is safe to eat", only to rail against "KILLER levels of salt" being served in restaurant foods three years later.
The standard way to decide among disparate study outcomes is to pool the data in a 'meta-analysis' to assess the evidence as a whole. Usually, it takes only a few such analyses to arrive at a fairly clear conclusion about any health issue.
But not about salt. In the last 17 years, a staggering 26 meta-analyses have arrived at radically different-indeed, polarizing-conclusions based on the overall evidence so far (see timeline, right). While these surveys all acknowledge that very high salt intakes are potentially bad for your health, there's considerable disagreement on whether everyday intakes are worth worrying about.
So, while the US Institute of Medicine acknowledges that the average 3.4 g/day of salt most Americans consume is well over the recommended 2.3 g/day or less, there was no evidence of benefit with such low levels of salt.10 Yet, the expert cardiologists of the American Heart Association (AHA), along with most international governments, suggest setting limits at less than 1.5 g/day.
But according to other experts, such low salt intakes are not only unwarranted but downright dangerous. In a recent eight-year study of nearly 4,000 Europeans, heart experts found that most cardiovascular-related deaths were of people with the least salty diets-in fact, the low-salt eaters had a staggering five times higher death rate than those who consumed the most salt.
This shouldn't have come as a surprise, though. Twenty years ago, US experts came up with similar findings in people with hypertension. According to the conventional theory, high salt intakes should have resulted in more heart problems. But the researchers found the reverse: those who consumed the least salt had the most heart attacks.
Last September, an international team pooled the findings of 25 different studies-totalling over a quarter of a million people-and clearly demonstrated that those with the lowest salt intakes had the highest risk of both heart problems and premature death due to any cause.
This was astonishing enough, but the fine print of the study contained another bombshell. The researchers had split people into three categories of salt intake, and the results showed that death rates were lowest with the "usual" salt intake (from 9.5-12.5 g/day), slightly higher in the "high" salt intake group (above 12.5 g/day) and highest in the "low" group (below 6.8 g/day).
So, if people followed the official guidelines and cut their salt intakes to 6 g/day or less, they had shorter lives and more heart disease than those who simply ignored the rules.
Not surprisingly, these findings were fiercely attacked. "The study relied on flawed data, and should not change the way anyone looks at salt," declared the AHA in a press release on 1 April 2014, insisting that "excess salt intake is a major health problem".
The physiology of salt
All sides of the debate agree that salt is an essential nutrient: it's an electrolyte without which our bodily systems would collapse. Its two constituents-sodium and chloride-are the principal ions in the fluids surrounding all body cells, including blood cells. Together with potassium, sodium regulates cell electrochemical functions-a task which takes up to 40 per cent of the body's resting energy.
But the body can't make its own salt and has to get it from food, and because dietary intakes can be uncertain, the body uses sophisticated control mechanisms to conserve salt when supplies are low and to jettison any excess via the kidneys when there's an overload.
But how much salt does the body actually need? Recommended minimums range from 1 to 4 g/day, the latter to allow for salt loss in people with sweaty lifestyles. Incidentally, that 4 g/day minimum comes from the AHA, the very organization that recommends we all reduce our intakes to 1.5/day or less.11 Here we have a professional body specifying the same intake as both the safe minimum and maximum, a unique restriction for public-health advice.
Despite the body's natural salt control mechanisms, in practice they can only help to prevent a physiological calamity-they cannot maintain optimal health. Older people, who commonly suffer from low salt levels (or 'hyponatraemia'), are particularly affected, and low salt levels may be the cause of many of the so-called symptoms of old age. A recent review reported that "even mild, chronic hyponatremia can lead to cognitive impairment, falls and fractures", the latter due in part to bone demineralization and poor bone quality.
One case-control study even describes some hyponatraemia-related symptoms as "comparable to those observed after alcohol intake", with "falls, unsteadiness, and cognitive impairment". Yet, despite such clear evidence of harm, official guidelines steadfastly maintain that old people should restrict their salt intakes to very low levels.
As for the rest of us, a number of papers published over the last 10 years have firmly established that low salt levels have major effects on key hormones, enzymes and blood profiles. A 2012 review summarizing the findings of 167 different papers reported that people who reduced their salt intakes had significant increases in blood levels of renin, aldosterone, noradrenaline, adrenaline, cholesterol and triglycerides, all of which contribute to heart disease. This might explain the counterintuitive findings that low salt intakes can lead to more heart disease.
A recent multinational study specifically assessed the effect of salt on heart disease. Experts from nine academic institutions across the globe monitored the intakes of almost 30,000 people and tracked their health for about five years.18 In that time, roughly 4,000 of them died or were hospitalized because of heart disease. Contrary to the salt/heart-disease theory, most of these heart problems were in people with "low" salt intakes, defined as less than 7.5 g/day. And taking less than 5 g/day was even more dangerous, resulting in the most heart attacks, strokes and heart failure.
Low salt has also been implicated in diabetes. In one trial, healthy people who followed a low-salt diet for just seven days ended up with increased insulin resistance, which can lead to type 2 diabetes.
Ironically, diabetics are routinely exhorted to reduce their salt intakes to very low levels-advice that's 100 per cent wrong. When actual death rates are checked, diabetics with low salt intakes have the poorest outcomes. A recent audit of Australian diabetics reported that "contrary to established dogma . . . the highest mortality risks were observed in individuals with the lowest salt intake and vice versa".20 Diabetics who ignored the medical advice and piled on the salt were the longest-lived.
What's the evidence?
Questions over the evidence linking low salt and poorer/better health are now ricocheting through the medical community, but are largely kept under wraps for fear of undermining the official 'salt is toxic' message. "Debates among scientists are common, but debates create doubt and should not derail sound policy," argues Dr Lawrence Appel, an AHA spokesman. "To prevent cardiovascular disease and improve cardiovascular health, we have to improve blood pressure control. Lowering salt intake is key to achieving this goal."
But the cat's got out of the bag. Last August, one of America's premier medical journals invited Professor Suzanne Oparil to write an editorial on the thorny issue. As past president of both the AHA and the American Society of Hypertension (ASH), Dr Oparil is an 'eminence grise in the field. Her editorial pointed to a "lack of high-quality data" on all sides, and asserted that the evidence so far "argues against reduction of dietary sodium to currently recommended levels as an isolated public health recommendation". Her conclusion: "Let's ditch the salt guidelines until we know what we're talking about." Given the lack of high-quality data, why were salt guidelines proposed in the first place? The answer may be that they were the result of theory trumping evidence.
As Dr Appel reminded us, the prevailing view is that high blood pressure is a major contributor to heart disease. So when it was clear that extra salt can increase blood pressure, the logic was simple: salt must cause heart disease.
But now, it seems there are major problems with this reasoning. First, there is growing professional uncertainty as to the details of the basic theory, as admitted by experts at the 22nd Scientific Meeting of the European Society of Hypertension in London in 2012.
"Despite decades of trials, we still are uncertain as to what level we should lower blood pressure to achieve maximal cardiovascular protection," said Dr Giuseppe Mancia of Milan University. This was echoed by UK expert Professor Bryan Williams, who said: "We have no idea what to lower blood pressure to."
There is also the so-called 'J-curve' for blood pressure and heart disease, which means that reducing blood pressure beyond a certain point can actually increase heart disease.22 At the other end of the scale, blood pressure targets in the elderly are being revised upwards as they're found to be perfectly safe.
In addition, although in trial settings dosing people with salt does indeed make their blood pressure rise-while reducing salt lowers it-in the real world, this doesn't always happen, as one major review revealed.
The analysis found two major holes in the evidence. First, it blew the gaff on the huge international Intersalt study (1982-1985), which claimed a "significant association" between salt intake and blood pressure, and predicted substantial decreases in national heart disease rates if populations reduced their salt intakes. But a later reanalysis revealed that, in fact, the study had actually found "a minimal impact of salt intake on blood pressure".
Indeed, 11 other long-term controlled trials had arrived at a similar conclusion: reducing salt intakes resulted in a clinically trivial "small decrease in average systolic but not diastolic blood pressure". However, the reanalysis did confirm that people with high blood pressure could reduce it by cutting down on salt, but this has never been in doubt.
Re-move the goalposts?
While professional bodies and anti-salt lobbyists doggedly stick to their guns and insist that governments continue to tell their citizens not to exceed salt intakes of 6 g/day, academics are becoming more circumspect and have revived the notion of 'salt resistance' and 'salt sensitivity', first proposed in the 1980s.26 In essence, this suggests that people differ genetically in how they handle salt, and only 'salt-sensitives' react with changes in blood pressure.
How many salt-sensitives are there in the general population? One US survey estimated that 15 per cent of whites and 27 per cent of blacks are affected (with figures roughly doubling in hypertensives).28 But although salt-sensitives do show rises in blood pressure with salt consumption, the extra danger to health is only a doubling of cardiovascular risk, even in people with hypertension.
As the vast majority of people don't have a problem with salt and as low salt is harmful to health, some academics are now urging that we scrap the official guidelines. Bolstering their view is a new idea-that the human body instinctively knows how much salt it needs for optimal health. A recent analysis of international population statistics reveals that salt intakes lie within "a narrow range . . . determined by physiologic needs"-between 6.5 and 12 g/day of salt-largely irrespective of eating habits.
Controlled experiments have also shown that restricting salt intake doesn't work: people simply eat more salt-laden foods to compensate.31 And it's not just the body, but the brain that regulates salt intake, claims Dr Joel Geerling, professor of neurology at Harvard Medical School. "These findings are consistent with decades of efforts by the neuroscience community directed at understanding the brain's role in the regulation of salt appetite," he says.
"The current guidelines need to be rethought," says McMaster University epidemiology professor Salim Yusuf. "We are making recommendations that 99 per cent of people cannot follow. From a practical point of view it makes no sense, and from a scientific point of view it makes no sense."
Even more outspoken is Professor Michael Alderman of Albert Einstein College of Medicine in New York, an elder statesman in the field of hypertension: "There is no justification for a population-wide, public health recommendation for radical reduction . . . in salt intake", nor is there "evidence that this reduction is either safe, in terms of ultimate health impact, or that it will produce cardioprotection."
Yet, the response of professional bodies has been both unyielding and vicious. Alderman has been accused not only of scientific incompetence, but of being in the pockets of the 'salt industry'-attacks that he refutes. "The low-salt dogma is often preached with a fervour usually associated with religious zealotry," adds Alderman.
As Professor Suzanne Oparil put it: "The thinking was that we should advise people to reduce their salt because it's something that's relatively easy to do and it can do no harm. What's new now is that there are studies suggesting that severe salt restriction over time can be harmful."
Tony Edwards is a freelance author, journalist and broadcaster who focuses on cutting-edge science and medicine.
The meta-analysis muddle In the last 17 years, 26 meta-analyses have arrived at radically different conclusions about salt and health:
"These results do not support a general recommendation to reduce salt intake", although it "may be used as a supplementary treatment in hypertension"
JAMA, 1998; 279: 1383-91
"[A] modest reduction in salt intake... does have a significant and... important effect on blood pressure"
J Hum Hypertens, 2002; 16: 761-70
"A further reduction [from 6] to 3 g/day... should now become the long-term target"
Hypertension, 2003; 42: 1093-9
"Intensive interventions [to reduce salt]... provide only minimal reductions in blood pressure"
Cochrane Database Syst Rev, 2003; 3: CD003656
A "reduction in salt intake... to 5 g/day or less is not scientifically justified"
Tidsskr Nor Laegeforen, 2004; 124: 3191-3
"The magnitude of the effect in Caucasians with normal blood pressure does not warrant a general recommendation to reduce salt intake"
Cochrane Database Syst Rev, 2004; 1: CD004022
A "modest reduction in salt intake... has a significant and, from a population viewpoint, important effect on blood pressure"
Cochrane Database Syst Rev, 2004; 3: CD004937
"Patients with elevated blood pressure should... restrict... salt intake"
J Hypertens, 2006; 24: 215-33
The "ability of dietary sodium restriction to reduce... cardiovascular events is... controversial"
Interact Cardiovasc Thorac Surg, 2007; 6: 793-8
The "long-term impact of reduced salt intake on blood pressure, mortality, and morbidity remains to be defined"
J Ren Nutr, 2009; 19: 57-60
"High salt intake is associated with significantly increased risk of stroke and total cardiovascular disease"
BMJ, 2009; 339: b4567
"[T]his meta-analysis shows a large fall in [blood pressure] with salt restriction, similar to that of single drug therapy"
Cochrane Database Syst Rev, 2010; 12: CD006763
Salt reduction "showed no strong evidence" of benefit for heart disease, but "increased the risk of all-cause mortality in those with heart failure"
Am J Hypertens, 2011; 24: 843-53
"Salt reduction lowers cardiovascular risk"
Lancet, 2011; 378: 380-2
The "benefits of salt reduction are clear and consistent, and reinforces the recommendations... for a population reduction in salt intake to prevent strokes, heart attacks and other cardiovascular events"
Nutr Metab Cardiovasc Dis, 2011; 21: 617-9
"Sodium reduction resulted in a 1% decrease in blood pressure in normotensives [and] a 3.5% decrease in hypertensives"
Cochrane Database Syst Rev, 2011; 11: CD004022
"We consider salt restriction to be an important tool for improving population health"
Ugeskr Laeger, 2012; 174: 637-9
"High salt intake is associated with significantly increased risk of stroke"
CNS Neurosci Ther, 2012; 18: 691-701
The "evidence base supporting recommendations for reduced sodium intake in the general population remains robust and persuasive"
Circulation, 2012; 126: 2880-9
"The current recommendations to reduce salt intake... to 5-6 g/day will have a major effect on blood pressure, but a further reduction to 3 g/day will have a greater effect"
BMJ, 2013; 346: f1325
The "evidence... is inconsistent and insufficient to conclude that lowering sodium intakes below 2,300 [2.3] mg/day either increases or decreases risk of [cardiovascular] outcomes... or all-cause mortality"
Institute of Medicine. Sodium Intake in Populations: Assessment of Evidence (2013)
"Both low sodium intakes and high sodium intakes are associated with increased mortality"
Am J Hypertens, 2014; 27: 1129-37
"There is insufficient power to confirm clinically important effects of dietary advice and salt substitution on cardiovascular mortality in normotensive or hypertensive populations"
Cochrane Database Syst Rev, 2014; 12: CD009217
This review "found overwhelming evidence that elevated sodium consumption could cause hypertension"
J Integr Med, 2015; 13: 1-8
"Taken together, available evidence does not support the current recommendations of a... reduction of salt intake at the population level" Adv Chronic Kidney Dis, 2015; 22: 108-15
This meta-analysis of 167 controlled trials could find "no relation between [sodium reduction]... and [blood pressure] outcome" whereas "prehypertensive and hypertensive populations showed a significant dose-response relation"
Adv Nutr, 2015; 6: 169-77
Salt and disease
These figures from the World Health Organization (WHO) show the average salt consumption in grams per day, per person, for different countries (dark blue numbers) and the annual death rates per 100,000 people by cause.
The death rates for heart disease are represented by a heart, the rates for stroke by a brain and the rates for hypertension (high blood pressure) by a cardiogram.
There appears to be no correlation between national salt intakes and heart-related deaths. Take Japan, for example. It has the lowest death rates for hypertension and the second lowest for heart disease, yet has the highest salt intake-which is more than double the WHO's recommended daily intake of
5 g per day per person.
Table salt or sea salt?
Table salt is the so-called 'refined' salt mostly from salt mines-huge underground deposits of dried-up ancient oceans. The refining process removes all salts except sodium chloride, and adds anti-caking agents for smooth flow from the salt shaker. In the US, iodine is mandatorily added to support thyroid function-an optional extra in the UK.
Sea salt is made by evaporating seawater in large shallow ponds. It's mostly sodium chloride but, depending on its origin, it may also contain minerals like potassium, magnesium, copper, iron, manganese, aluminium and iodine. Although the amounts are tiny, they may have nutritional value.
Most dietitians believe there's no difference between the two types, but nutritionists recommend sea salt, as it more closely corresponds to the natural salt composition of our blood.
2 CRC Crit Rev Toxicol, 1978; 5: 355-76
4 Am J Clin Nutr, 2000; 71: 1013-9
5 Am J Clin Nutr, 2000; 71: 1020-6
6 Hypertension, 1993; 22: 331-8
7 BMJ, 2007; 334: 885-8
10 Institute of Medicine. Sodium Intake in Populations: Assessment of Evidence (2013)
11 AHA. 'Shaking the Salt Habit'; www.heart.org/HEARTORG/Conditions/HighBloodPressure/PreventionTreatmentofHighBloodPressure/Shaking-the-Salt-Habit_UCM_303241_Article.jsp
12 JAMA, 2011; 305: 1777-85
13 Hypertension, 1995; 25: 1144-52
14 Am J Hypertens, 2014; 27: 1129-37
15 J Clin Med, 2014; 3: 944-58
16 Am J Med, 2006; 119: 71.e1-8
17 Am J Hypertens, 2012; 25: 1-15
18 JAMA, 2011; 306: 2229-38
19 Metabolism, 2011; 60: 965-8
20 Diabetes Care, 2011; 34: 703-9
21 N Engl J Med, 2014; 371: 677-9
22 Eur Heart J, 2010; 31: 2897-908
23 JAMA Intern Med, 2013; 173: 325
24 J Ren Nutr, 2009; 19: 57-60
25 Clin Exp Hypertens A, 1989; 11: 1025-34
26 Kidney Int, 1982; 21: 371-8
27 Curr Hypertens Rep, 2011; 13: 55-66
28 Hypertension, 1991; 17 [1 Suppl]: I61-8
29 Lancet, 1997; 350: 1734-7
30 Am J Hypertens, 2013; 26: 1218-23
31 Am J Hypertens, 1997; 10: 68-76
32 Int J Epidemiol, 2002; 31: 311-6