In 1942, a 33-year-old woman walked into a hospital with a blood pressure so high, it was destroying her eyes. Her kidneys were failing. At the time, there were no effective treatments. So in a last-ditch effort, her doctor — a German refugee named Walter Kempner — put her on a radical experiment: nothing but rice, fruit, and juice. Almost zero salt. He told her to come back in two weeks [1].
But Kempner had a thick German accent. She misunderstood — and came back in two months [1].
When she returned, instead of being dead or malnourished, her blood pressure had dropped from 190/120 to 124/84. Her eye damage had resolved. Her previously swollen heart had shrunk [1]. By accident, Kempner had stumbled onto something extraordinary.
That was over 80 years ago. Yet the debate about salt continues. And depending on which side of the argument one lands on, either cutting sodium is essential for heart health — or the conventional advice could actually make things worse.
There are viral videos claiming that cutting sodium intake is unnecessary. At the same time, the American Heart Association recommends staying below 1,500 mg of sodium a day to support blood pressure and heart health [2].
A major new study — a comprehensive digest of all the research conducted to date on sodium and cardiovascular health — provides the clearest picture yet of what the evidence actually shows.
Table of Contents
- The Scientific Consensus on Salt
- Dahl and the Genetic Factor
- The Contrarian View
- What the New Study Found
- Practical Takeaways
- References
The Scientific Consensus on Salt
Kempner's accidental discovery was followed by decades of research. The consensus that formed was clear and nearly unanimous. The American Heart Association is not an outlier.
A quick note: sodium and salt are not the same thing — salt is about 40% sodium, which is why the WHO's 2,000 mg sodium limit works out to roughly a teaspoon of salt [3].
A landmark study in 1988 — INTERSALT — tracked sodium in the urine of 10,000 adults across dozens of countries and found a clear, consistent association: more sodium, higher blood pressure. The relationship held across populations with very different dietary patterns and baseline blood pressure levels [4].
Then a randomised controlled trial — the DASH-Sodium trial — went further. It tested three sodium intake levels across two different diets (the DASH diet and a typical Western diet), and in both, reducing sodium led to significant drops in blood pressure — especially for those who started with the highest levels. The combined effects of the low-sodium DASH diet versus the high-sodium control diet on systolic blood pressure were -5.3, -7.5, -9.7, and -20.8 mmHg across baseline blood pressure strata, with the greatest benefit seen in those with pre-existing hypertension [5].
And when researchers followed a group that had participated in a sodium reduction trial for 24 years — one of the longest sodium follow-ups ever conducted — those with an intake under 2,300 mg had a 25% lower risk of death than those consuming between 3,600 and 4,800 mg. This was not just a statistical association; it was tracked in the same individuals over more than two decades [6].
Lower salt, lower blood pressure, longer life. Case closed — or so it seemed.
Dahl and the Genetic Factor
But even in the early days, there were hints that the story was not so simple.
In the early 1960s, a physician named Lewis Dahl at Brookhaven National Laboratory started feeding rats high-salt diets. Some rats developed hypertension. Others were completely unaffected. So he selectively bred them. Within just three generations, he had two distinct lines: salt-sensitive rats that died of hypertension on a high-salt diet, and salt-resistant rats that were totally immune to it [7].
It was the first clear proof that salt sensitivity has a genetic component — and that not everyone responds to salt the same way. This finding becomes important later.
Dahl spent the next 15 years proving salt's dangers — he even fed commercial baby food to his salt-sensitive rats and watched them all develop hypertension, which triggered a US Senate investigation into sodium in infant food. In 1975, he won the Ciba Award — the highest prize in hypertension research [7].
The Contrarian View
But then a widely-discussed study raised a serious objection to the consensus.
In 2014, cardiologist Salim Yusuf and epidemiologist Andrew Mente at McMaster University published a study that would ignite a scientific controversy. Their team analysed urine samples from over 100,000 people across 17 countries, looking for indicators of sodium and potassium levels. Over a follow-up period of almost four years, they examined how heart attacks, strokes, and deaths related to those levels [8].
Partly, what they found matched the established picture. When sodium intake was high, there was an increased risk of death, heart attacks, and strokes [8].
But they found something else that was surprising. Low sodium intake was also associated with elevated risks of heart attack and stroke. As compared with the reference range, an estimated sodium excretion below 3,000 mg per day was also associated with an increased risk of death and cardiovascular events [8].
In other words, the data revealed a J-shaped curve. It was not a case of "the less sodium, the better." Instead, there appeared to be an optimal range the research placed between 3,000 and 6,000 mg a day [8].
As Mente put it in an interview: "Sodium is an essential nutrient. It's not tobacco. Without sodium, you die. The optimal level of tobacco is zero, but with sodium, it's not zero" [9].
The response from the medical establishment was immediate — and personal.
Yusuf, the study's lead, later described it bluntly: "There has been a smear campaign by a group of people against anybody who questions salt. The moment you stand up and say, 'Well, it might not be as bad as we think,' you get attacked personally" [10].
Daniel Jones, speaking on behalf of the American Heart Association, fired back: "This is a flawed study, and no health policy should be based on this study" [11].
Critics of the prevailing view were quick to point out a glaring potential problem. Both the WHO's recommendation of staying under 2,000 mg/day and the American Heart Association's ideal target of 1,500 mg/day are significantly below this apparent sweet spot. The worry, according to Mente, is that pushing sodium intake that low might actually raise cardiovascular risks — the opposite of the intended effect.
The critics' conclusion: most people should not necessarily be trying to lower sodium intake, particularly if potassium intake is adequate. Because there was another important finding of this study and others. Higher potassium intake lowers the risks of heart attacks and strokes [8]. Potassium helps lower blood pressure, especially when it is elevated [12].
There are also certain medical conditions — such as postural orthostatic tachycardia syndrome (POTS), where the body struggles to maintain blood flow to the brain upon standing — where increasing sodium intake is clinically indicated.
All of this has generated considerable confusion, leaving many people unsure whether they should be cutting sodium or not.
What the New Study Found
So who is right — the guidelines that recommend cutting sodium as low as possible, or the PURE researchers who suggest that could be harmful? A major new study provides the clearest answer yet.
The study is not just another trial adding one more data point. And it is not even a standard meta-analysis pooling results from several trials. Instead, it is essentially a meta-analysis of meta-analyses of randomised controlled trials and observational studies — a comprehensive digest of all the research conducted to date on sodium intake and cardiovascular health [13].
The headline findings reinforce the consensus. Low sodium intake is associated with reduced risk of heart-related and all-cause mortality. It reduced the risk of death from strokes by 26%. High sodium intake raised stroke mortality by 40%. Each extra 1,000 mg/day of sodium raised heart disease and stroke risks by 4% and 6%, respectively [13].
So that settles it, right? Not so fast. What about Mente's J-curve — the finding that going too low also increases risk?
This is where it gets interesting. The umbrella review found no signal of elevated risk at low intakes. No J-curve [13].
But if low sodium is safe, why did the PURE study find the opposite? The PURE researchers had proposed a mechanism: when sodium drops very low, the body activates a compensatory system — the renin-angiotensin-aldosterone system (RAAS) — which causes hormones to constrict blood vessels and force the kidneys to retain sodium. Think of it like a thermostat that overreacts to a small temperature drop by cranking the heat to maximum.
This system — called the renin-angiotensin-aldosterone system — can contribute to cardiovascular problems when it is chronically overactive [13].
So is this backup system a real danger? The umbrella review examined the evidence. Researchers found evidence of partial RAAS activation — but the response was mild. And with sustained lower intake, the body adapts. Evidence suggests that with prolonged sodium reduction exceeding one year, compensatory RAAS activation diminishes over time [13].
But the researchers found something that adds important nuance — and it brings the analysis back to Lewis Dahl's rats from the 1960s.
The response to low salt intake was not uniform across populations. It significantly lowered blood pressure in the Western Pacific, Europe, and Southeast Asia — but not in the Americas [13].
Similarly, high salt intake was not significantly associated with cardiovascular deaths in US populations, but was strongly linked to them in Japanese populations. This could reflect different dietary patterns, or differences in sodium sensitivity across populations [13].
So was Mente wrong? Not entirely. The umbrella review found no J-curve for mortality — but his observation about population differences turned out to be real. Dahl proved that salt sensitivity is genetically determined — some rats were profoundly affected, others barely at all. Sixty years later, the same pattern shows up across human populations. One size does not necessarily fit all. Critics of existing salt guidelines will likely find that part vindicating.
It is also theoretically possible for sodium intake to be too low. The body is estimated to need about 500 mg a day to function normally [2]. But in practice, it is almost unheard of to reach such a low level through diet alone. While low blood sodium is relatively common in hospitalised patients, it generally results from serious underlying health problems — not from eating too little salt [14].
Walter Kempner's rice diet — under 230 mg of sodium per day, far below anything currently recommended — was used at Duke University for nearly 60 years, treating over 17,000 patients. When researchers digitised those records, they found even that extreme restriction appeared safe, with a five-year survival probability of 95.6% [15].
For most people, the problem lies at the other end of the spectrum. The global average sodium intake is nearly 4,000 mg/day [13]. Concern about recommendations being too low is analogous to pushing back against exercise guidelines because it is possible to over-train — the typical situation is excessive intake, not insufficient.
Practical Takeaways
What does this evidence mean in practice?
For most people, the guidelines from the WHO and the American Heart Association are supported by the best evidence available. Keeping sodium intake low is associated with lower blood pressure, which is in turn linked to lower risks for a range of cardiovascular outcomes.
But should the target be 2,300 mg? 2,000 mg? Or even 1,500 mg of sodium per day?
Mente has pointed out how challenging the lowest targets are to achieve. Authors of one journal article proposed using prison populations to study the effects of lower intakes, because adults find it very difficult to reduce sodium consumption to the 1,800–2,300 mg/day range [16]. Mente commented: "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" [17].
The honest answer: for most people, the lower the sodium intake, the better.
There is substantial evidence that blood pressure benefits continue to accumulate as sodium intake decreases. An earlier meta-analysis found the relationship holds at both the high and low ends of the sodium consumption spectrum [18].
When it comes to cutting back on sodium, it is useful to recognise that most of the salt people consume does not come from the salt shaker — it comes from packaged and processed foods, which often contain very high levels [19]. Reducing packaged food consumption is an effective way to lower sodium intake while simultaneously shifting toward healthier whole foods.
Even modest changes can produce meaningful results. A large study in China had participants switch from regular salt to a salt substitute. Instead of standard table salt (100% sodium chloride), the substitute was 75% sodium chloride and 25% potassium chloride [20].
That single dietary swap cut stroke risk by 14%, reduced a combined measure of major cardiovascular events and strokes by 13%, and lowered the risk of death by 12% over nearly five years of follow-up. Average blood pressure among those using the salt substitute dropped by 3.34 mmHg [20].
Lead investigator Bruce Neal put the potential scale in perspective: approximately 10 million major cardiovascular events occur each year in China. Around 1 million of those could be avoided by this single change — one swap in the salt shaker [21].
A salt substitute is a particularly effective intervention because it accomplishes two things at once: it cuts sodium intake while simultaneously increasing potassium intake. A follow-up analysis found that the majority of the blood pressure reduction in the China study came from the addition of potassium, not the subtraction of sodium [22].
The evidence supports both strategies in parallel: keeping sodium intake low and increasing potassium intake. Since most people do not consume enough potassium through diet, increasing it represents an achievable and high-impact change. Because the two nutrients interact — potassium helps the kidneys excrete sodium and directly relaxes blood vessel walls — addressing both simultaneously produces greater benefit than focusing on either alone.
A meta-analysis on potassium and blood pressure found that when daily potassium intake reached 3,500 to 4,700 mg, the blood pressure reduction was 7.16 mmHg [23]. To put that in perspective, a 7 mmHg systolic reduction is roughly comparable to a moderate dose of a common blood pressure medication. Natural food sources — leafy green vegetables, beans, lentils, and bananas — are the preferred way to achieve this. Important caveat: people with chronic kidney disease should consult a healthcare professional before substantially increasing potassium intake, as impaired kidney function reduces the body's ability to excrete excess potassium safely.
Taken together, the evidence from 80 years of research reinforces what Kempner stumbled onto by accident in 1942: for most people, less sodium means lower blood pressure and better long-term cardiovascular outcomes. The critics are right that not everyone responds equally — genetic and regional differences in sodium sensitivity are real. But the global average intake is nearly double what the evidence supports as beneficial. The practical steps are clear: reduce packaged food consumption, consider a potassium-enriched salt substitute, and prioritise natural potassium-rich whole foods like leafy greens, beans, lentils, and bananas.
References
1. https://www.drmcdougall.com/education/information-all/walter-kempner-mdfounder-of-the-rice-diet/
3. https://www.who.int/news-room/fact-sheets/detail/sodium-reduction
4. https://www.bmj.com/content/297/6644/319
5. https://pmc.ncbi.nlm.nih.gov/articles/PMC5742671/
6. https://pmc.ncbi.nlm.nih.gov/articles/PMC5098805/
7. https://pmc.ncbi.nlm.nih.gov/articles/PMC4393342/
8. https://www.nejm.org/doi/full/10.1056/NEJMoa1311889
10. https://www.medscape.com/viewarticle/824749
12. https://academic.oup.com/ckj/article/18/7/sfaf173/8177122
13. https://pmc.ncbi.nlm.nih.gov/articles/PMC12624901/
14. https://www.ncbi.nlm.nih.gov/books/NBK470386/
15. https://pmc.ncbi.nlm.nih.gov/articles/PMC11773661
16. https://www.ahajournals.org/doi/10.1161/HYPERTENSIONAHA.118.11103
17. https://www.tctmd.com/news/faulting-salt-new-pure-analysis-argues-against-low-sodium-intake
18. https://pmc.ncbi.nlm.nih.gov/articles/PMC8055199/
19. https://pmc.ncbi.nlm.nih.gov/articles/PMC9962803/
20. https://www.nejm.org/doi/full/10.1056/NEJMoa2105675
21. https://www.tctmd.com/news/massive-ssass-study-shows-switch-salt-substitute-cuts-stroke-cvd
