If a study named eight additives you almost certainly ate this week, and linked them to high blood pressure and heart disease, the obvious reaction is to start avoiding them. That reaction is half right, and the other half is where most reporting goes wrong.
The research, published in the European Heart Journal, followed 112,395 adults across France for close to eight years and found that the people eating the most preservatives were meaningfully more likely to develop high blood pressure and heart disease. As far as the researchers know, it is the first study of its kind to look at a wide range of preservatives and heart health in humans at this scale.
But the study did not show that these eight ingredients cause heart attacks. It found patterns in a large group of people, but a pattern in a group is not the same as a proven cause. That difference is the single most important thing to understand before you change anything about your diet. This article explains what was actually found, where each additive shows up on a shop shelf, and what a reasonable reader should do with the information.
What the largest preservative study actually found
The analysis comes from NutriNet-Santé, a long-running French nutrition research project that began in 2009 and has followed the same volunteers for years. What sets it apart from most diet research is how detailed the food records were. Participants wrote down everything they ate and drank over three-day periods, repeated as often as 96 times across the study, covering home cooking and packaged brands alike. Researchers then matched those records against ingredient databases and laboratory tests of actual food samples to work out which additives each person was consuming.
Across the group, 78.7% of volunteers were women and the average age was 42.8 years, and the study followed them for a median of 7.9 years. The first result added up all the preservatives together rather than looking at them one by one. People who ate the most non-antioxidant preservatives had a 29% higher chance of developing high blood pressure, and a 16% higher chance of heart disease, stroke or angina. People who ate the most antioxidant preservatives had a 22% higher chance of high blood pressure. The researchers report these as hazard ratios of 1.29, 1.16 and 1.22, and also give the likely range around each figure, which is explained below.
Only one of those three findings is solid across its whole range. The cardiovascular figure has a lower bound of 1.04, which means the real increase could be as small as 4% rather than 16%. The two high blood pressure figures sit comfortably clear of no effect, which makes them the more reliable results. The key point is that these are group-level averages, not a prediction about any one person’s health.
Almost everyone ate them. During the first two years of taking part, 99.5% of volunteers had eaten at least one food preservative. Across everyone in the study, the researchers counted 58 different preservatives.
The eight additives and the strength of each finding
The team then looked at the 17 preservatives that at least one in ten participants ate. Eight came back linked to higher rates of high blood pressure: potassium sorbate (E202), potassium metabisulphite (E224), sodium nitrite (E250), ascorbic acid (E300), sodium ascorbate (E301), sodium erythorbate (E316), citric acid (E330) and rosemary extracts (E392). Ascorbic acid (E300) was also linked to heart disease on its own, though that was a single result rather than a pattern across the group.
The ascorbic acid result is the one to watch. Vitamin C also occurs naturally in fruit and vegetables in far larger amounts than any food fortification provides, and a commentary published in the same journal in August 2026 argues the association is difficult to reconcile with randomized trial evidence. That argument is set out in the next section.
That distinction is where most write-ups on this study go wrong, so it is worth stating plainly. The NutriNet-Santé authors never said these substances are dangerous in themselves. What they measured was the pattern of additives in someone’s diet. If the people eating the most additives were also eating the most processed food overall, the additives may simply be standing in for the wider diet rather than causing harm themselves. Telling those two possibilities apart is exactly what the next round of research needs to do, and the team is already working on it.
| Additive | E-number | What it does | Linked to high blood pressure | Linked to heart disease |
|---|---|---|---|---|
| Sodium nitrite | E250 | Non-antioxidant | Yes | No |
| Ascorbic acid (vitamin C) | E300 | Antioxidant | Yes | Yes |
| Potassium sorbate | E202 | Non-antioxidant | Yes | No |
| Citric acid | E330 | Antioxidant | Yes | No |
| Potassium metabisulphite | E224 | Non-antioxidant | Yes | No |
| Sodium ascorbate | E301 | Antioxidant | Yes | No |
| Sodium erythorbate | E316 | Antioxidant | Yes | No |
| Rosemary extracts | E392 | Antioxidant | Yes | No |
The European Society of Cardiology, which published the findings alongside the journal paper, quoted lead researcher Dr. Mathilde Touvier, saying the results are a reason for regulators to take another look at the risks and benefits of these additives. Her framing was notably measured: the results support existing advice to favor less processed food and avoid unnecessary additives, rather than replacing that advice with something new.
[interactive_interactive_chart]{“type”:”bar”,”themeMode”:”light”,”background”:”#f8fafc”,”title”:”How Much Higher Was the Risk?”,”subtitle”:”Percentage increase in risk for people who ate the most preservatives compared with the least.”,”source”:”Hasenböhler et al., European Heart Journal”,”sourceUrl”:”https://academic.oup.com/eurheartj/advance-article/doi/10.1093/eurheartj/ehag308/8679203″,”unitSuffix”:”%”,”dataLabels”:true,”categories”:[“Non-antioxidant preservatives and high blood pressure”,”Non-antioxidant preservatives and heart disease”,”Antioxidant preservatives and high blood pressure”],”series”:[{“name”:”Increase in risk”,”data”:[29,16,22]}],”colors”:[“#0f766e”],”xAxis”:{“title”:”Risk increase (%)”},”yAxis”:{“title”:”Increase in risk”},”allowDownload”:true}[/interactive_interactive_chart]
The chart shows the three overall figures. The possible range around each one matters just as much as the headline number, though, and the bars alone do not show it. A 29% increase where the real figure sits somewhere between 20% and 39% is unlikely to be chance. The 16% figure for heart disease could be as low as 4%, which is much closer to no effect at all.
Antioxidant and non-antioxidant preservatives are not interchangeable
One of the confusing parts of the press coverage is that both kinds of preservative were linked to higher rates of high blood pressure. Understanding why requires knowing what each type does.
Non-antioxidant preservatives stop harmful organisms. Bacteria and mold are the targets. Antioxidant preservatives slow down the process that makes food go brown, change color, or taste rancid. In practice they often appear alongside each other, because stopping spoilage and stopping browning are separate jobs needing separate ingredients.
| Non-antioxidant | Antioxidant | |
|---|---|---|
| What they do | Stop bacteria and mold growing | Stop food browning or going rancid |
| Of the eight flagged | 3 additives | 5 additives |
| High blood pressure | 29% higher | 22% higher |
| Possible range | 20% to 39% | 13% to 31% |
| Heart disease | 16% higher, but could be as little as 4% | No link found for the group overall |
| Additives in each group | Sodium nitrite, potassium sorbate, potassium metabisulphite | Ascorbic acid, sodium ascorbate, citric acid, sodium erythorbate, rosemary extracts |
| Where you meet them | Cured meats, wine and beer, cheese, dried fruit | Bread, soft drinks, crisps and snacks, cooking oil, jams |
Note: the study reported overall results for the two groups separately, then listed which individual additives were linked. The paper does not spell out which of the eight falls into which group, so the group assignments here follow each additive’s standard function.
Two details in that table are worth pausing on. The non-antioxidant group has the higher high blood pressure figure at 29%, but its heart disease result is the weaker of the two, with a range that drops to just 4% at the bottom. And the single additive linked to heart disease on its own, ascorbic acid, sits in the antioxidant group rather than the non-antioxidant one. Most write-ups on this study do not mention either point.
There is a reason not to read the two figures as separate results. People who eat a lot of processed food tend to run into both kinds of preservatives at once, which is why finding higher risk in both is harder to interpret than it first looks.
Where these eight additives actually show up
The most useful thing a reader can do with this study is identify where these ingredients appear. The list below covers common uses rather than exhaustive ones, and no individual product will necessarily contain any given additive. The ingredient list on the specific package is the only reliable source.
| Additive | Commonly found in |
|---|---|
| Potassium sorbate (E202) | Cheese, yogurt, cakes and pastries, margarine, jams, sauces, dried fruit, fruit drinks |
| Potassium metabisulphite (E224) | Wine, cider and beer, dried fruit, bottled lemon and lime juice, prawns, pickles, sauces |
| Sodium nitrite (E250) | Bacon, ham, sausages, salami, hot dogs, and cured and processed meats |
| Ascorbic acid (E300) | Bread and baked goods, breakfast cereals, fruit drinks, soft drinks, processed meat, frozen fruit and vegetables |
| Sodium ascorbate (E301) | Ham, bacon, sausages, other processed meats, some bread, drinks, and canned foods |
| Sodium erythorbate (E316) | Sausages, ham, bacon, hot dogs, deli meats, other cured and processed meats |
| Citric acid (E330) | Soft drinks, sweets, jams, desserts, sauces, tinned fruit and vegetables, flavored dairy, ready meals |
| Rosemary extracts (E392) | Cooking oils, margarine, meat products, sauces, dressings, crisps and snacks, bakery products |
Read across that table and a pattern emerges. The E-numbers do not cluster in obscure or industrial corners of the supermarket. Rosemary extract, a natural herb extract, is a preservative, and it appears in crisps and cooking oil. Ascorbic acid, which is vitamin C, appears in bread. Sodium nitrite, the additive most readers will have heard of, is confined almost entirely to cured meats.
This is also why the “eat less processed food” advice feels vague to many people. It is not a single category. It ranges from a bag of crisps with rosemary extract to a bottle of wine with sulphites to a pack of bacon with nitrite, and the nutritional profiles of those foods differ substantially. If you are deciding what to cut, the evidence on ultra-processed foods and cardiovascular outcomes offers a broader framework, including the finding that cardiovascular mortality is the strongest signal in that literature.
What foods have the most preservatives?
There is no single answer, because it depends on whether you are counting the number of different additives on the label, their weight, or how far down the ingredient list they appear. A plain loaf of white bread can carry four or five additives while ranking none of them in the top three. A sausage roll can carry one.
By the measure most people mean, which is the number of distinct preservative additives, these categories tend to sit at the top:

- Cured and processed meats. Ham, bacon, sausages, salami, and hot dogs commonly combine sodium nitrite with sodium ascorbate or sodium erythorbate and often add potassium sorbate as well. This is the most additive-dense shelf in most supermarkets.
- Soft drinks and powdered drinks. Citric acid, ascorbic acid, and benzoates are common together, and a single fruit-flavored drink can list several.
- Fruit spreads, jams, and baked goods. Potassium sorbate appears alongside citric acid and ascorbic acid in most commercial versions.
- Salad dressings, sauces, and pickles. Potassium sorbate, citric acid, and sulfites are all common, and many products use two or three.
- Breakfast cereals and snack bars. Frequently carry ascorbic acid, citric acid, and sometimes rosemary or mixed tocopherols alongside several flavorings.
- Wines and dried fruit. Sulphites are the defining preservative here, sometimes alongside ascorbic acid.
How much of each additive is in a product matters more than how many. Two preservatives near the top of the ingredient list expose you more than six listed at the end, which is why the NutriNet-Santé team added up everything the volunteers ate across their whole diet rather than counting products. The study turned up 58 different preservatives across the group, a truer picture of what people actually eat than any single label can give.
It is worth repeating the point from earlier. People who eat the most of these ingredients are also, statistically, the people eating the most processed food overall. If the additives are simply standing in for that pattern rather than causing harm themselves, then cutting back on processed food does more good than hunting down individual E-numbers.
Five additives readers most often ask about
Some ingredients come up in almost every conversation about food labeling, whether or not this study looked at them. For context on the wider landscape:
- Sodium benzoate (E211), a common preservative in soft drinks, sauces, and pickles.
- Sodium nitrite and sodium nitrate (E250 and E251), the cured-meat pair, where nitrate converts slowly into nitrite.
- Sulphur dioxide and sulphites (E220 to E228), which must be declared on wine labels above a threshold in many countries.
- Calcium propionate (E282), the usual anti-mould additive in sliced bread and baked goods, which is why it appears alongside ascorbic acid.
- Artificial sweeteners, which are not preservatives and were not part of this study, though the broader ultra-processed food research does examine their effects separately.
Is any preservative genuinely unsafe?
Which of the eight is most worrying is not something this study can settle, because it did not test any one additive in isolation. But two things are worth separating here.
The first is the safety record that already exists. Before an additive can be used, regulators set a limit on how much a person can safely eat each day, even over a lifetime. For sodium nitrite, the FAO/WHO Joint Expert Committee on Food Additives has set that limit at 0.07 milligrams per kilogram of body weight. The European Food Safety Authority is partway through a review of 315 additives approved before 2009, with 137 already finished and 70 still to go.
The second point concerns sodium nitrite specifically. The International Agency for Research on Cancer classifies nitrite as probably carcinogenic to humans, though the wording matters: the classification covers nitrite consumed under conditions that lead to nitrosation inside the body, which happens when it reacts with other compounds in the digestive tract. The same additive is used deliberately to stop Clostridium botulinum, the bacterium behind botulism, and it gives cured meat its color and flavor. A reasonable position acknowledges both halves of that, rather than presenting nitrite as just another line on a list of eight.
Notably, ascorbic acid and erythorbic acid are added to cured meat specifically to inhibit nitrosamine formation, which is one reason cured meat formulations frequently include several of the flagged additives at once.

Is the vitamin C finding disputed?
Yes. In August 2026, the same journal published a commentary by Harri Hemilä arguing that the study’s vitamin C result is difficult to reconcile with what clinical trials have already found.
The argument runs as follows. The study found that vitamin C, which the researchers refer to as total ascorbates, was associated with a 13% higher risk of high blood pressure, and the median amount people ate as a preservative was 44 milligrams a day. Dr. Hemilä’s point is that randomized controlled trials, where researchers assign treatments at random, are not vulnerable to the kinds of hidden variables that can distort an observational study. He notes that three relevant trial reports were not cited in the original paper, which he says may leave readers with an unbalanced picture.
The trial evidence he cites comes from a meta-analysis of 29 randomized trials, which found that vitamin C supplements lowered systolic blood pressure by an average of 3.8 millimeters of mercury and diastolic pressure by 1.5. Both were statistically significant. In other words, the trial evidence points in the opposite direction to the NutriNet-Santé finding.
There is an important wrinkle, and the commentary raises it himself. Those trials used a median dose of 500 milligrams a day, which is more than ten times the 44 milligrams that people in the study were getting from preservatives. So the two figures are not measuring the same thing, and a supplement taken in a large pill is not the same as a trace amount used to keep bread from going grey. That gap does not settle the question either way, but it means neither study is a direct answer to the other.
It is worth being clear about what this does and does not change. A commentary is an argument, not a replication, and not a retraction. The NutriNet-Santé paper has not been withdrawn, and the authors acknowledged in it that hidden factors cannot be entirely ruled out.
What we already know cuts in the other direction
That said, the framing is misleading in a different way. The NutriNet-Santé team spent more than a decade collecting unusually detailed dietary records, and the published analysis examined 58 individual preservatives. Salt, the most established dietary factor in high blood pressure, was not its subject. This is not a criticism, because answering one question well is not the same as ranking every dietary factor. But it does mean the 29% figure compares high consumers against low consumers who also ate salt. The number is real, and it is not a measure of what a preservative costs you compared with what salt does.
The DASH-Sodium trial and the Salt Substitute and Stroke Study both point in the other direction, and much more firmly than any additive result.
The DASH-Sodium trial, published in the New England Journal of Medicine in 2001, randomized adults to different combinations of sodium level and eating pattern. Among the 153 participants whose blood pressure was measured directly, those following a DASH diet with low sodium had systolic pressure lower by 7.1 millimeters of mercury compared with a high-sodium control diet if they did not already have hypertension, and 11.5 lower if they did.
The Salt Substitute and Stroke Study went further than blood pressure readings. It randomized 20,995 people across 600 villages in rural China, all of whom had either a previous stroke or uncontrolled high blood pressure, to use either a potassium-enriched salt substitute or ordinary salt. Over roughly five years, those using the substitute had 13% fewer major heart attacks and strokes. That is a hard clinical outcome rather than a survey finding, and it remains one of the clearest examples in nutrition of an everyday change producing a measurable health benefit.
Compare those magnitudes with the study at the center of this article. Every additive result here sits in the teens in percentage terms. Salt reduction and DASH-style eating in randomized trials produce effects several times larger. If you have high blood pressure and limited time, the evidence says start with salt and overall eating pattern rather than reading ingredient labels for E-numbers.
There is a real caveat. Trials of dietary change are notoriously hard to run, which is why good cohort evidence matters too, and none of this settles the preservative question either way. But it does mean preservatives are not the only item on the list, and a reader deciding what to do with this information should know that.
What this study cannot tell you
Healthy skepticism is warranted here, and there are four specific limitations rather than a generic disclaimer.
Association is not the same as cause. Studies that watch people over time can spot patterns, but they cannot prove that one thing produced another. They show that people with more of something were more likely to experience something else. The researchers were upfront about this, noting the limits built into the design of their research.
The volunteers were not a random cross-section of the public. With 78.7% women and an average age of 42.8, the participants looked quite different from the general adult population, and people who sign up for a health study tend to be more health-conscious than average. Their results may not transfer cleanly to everyone else.
They ruled out chance, but the additives are still hard to pin down. The researchers accounted for other factors that raise or lower heart risk, and they used a method that corrects for testing many additives at once, which is how they narrowed 17 down to eight fairly. That is a real strength. It does not solve a harder problem: people who eat a lot of these additives also tend to eat more processed food overall, with more salt, sugar, and calories. Separating the additive from the whole pattern is what remains unresolved.
The one figure that is softer than the headline suggests is the 16% increase in heart disease. Its lower bound is 4%, which is close enough to no change to matter. The honest reading is that heart disease risk probably did rise, but probably by less than 16%.
Set against that, the study’s real strength is its data quality. The repeated three-day diaries, the inclusion of branded products, and the laboratory confirmation of what was actually in the food make these exposure estimates far more accurate than those in most diet research. That is why Dr. Touvier felt the findings were worth putting in front of regulators despite the design.
What happens next?
The team is now looking at how these additives and processed foods affect inflammation, oxidative stress, blood markers like cholesterol and glucose, and the mix of bacteria in the gut. Finding a biological mechanism would be the next real step, because that is what would turn a statistical pattern into a proven cause.
On the regulatory side, Dr. Touvier called for the European Food Safety Authority in Europe and the Food and Drug Administration in the United States to take another look at the risks and benefits of these additives. She framed the immediate takeaway as supporting existing advice to eat less processed food, rather than issuing new guidance.
For now, no regulator has changed how much of an additive people can safely eat because of this study, and it would be misleading to imply otherwise. Regulators also look at an additive’s whole history of use, not just one study, so nothing changes on the strength of a single study like this one.
Practical ways to reduce exposure
The practical message from the research, in the researchers’ own words, is really just a reminder of advice people already have. That can feel unsatisfying, so it is worth making it concrete.
- Check the ingredient list, not the front of the pack. Additives are listed by name or E-number. Ingredients appear in descending order by weight, so an additive near the top is present in a larger quantity.
- Cured meat is the most concentrated category. Four of the eight flagged additives are used primarily or extensively in processed meat, and E250 is used almost nowhere else. If you eat bacon, ham, or sausages regularly, that is the highest-leverage change available.
- Bread and breakfast cereal carry ascorbic acid. This is one of the easier swaps, since the flour improver function in bread can be replicated by buying bread without it or choosing products that use it minimally.
- Wine, dried fruit, and bottled juices carry sulfites. The same applies to prawns and pickles.
- Read the wine label for sulphite declarations. In many jurisdictions sulphites above a threshold must be declared, and a “contains sulphites” line is a reliable signal.
- Do not attempt to eliminate all preservatives. They exist for food safety reasons. Avoiding every additive risks more than it solves, and the researchers did not recommend that.
For readers already working on high blood pressure, the evidence on potassium-enriched salt substitutes looks at a part of the diet with a much more established link to high blood pressure than any preservative, and the role of dietary patterns in inflammation is relevant to one of the mechanisms the researchers are now investigating.
Frequently asked questions
Should I stop eating bread, bacon, and soft drinks because of this study?
No single food needs to be cut out based on one study like this. A more sensible approach is to notice which of these categories you eat most of, since cured meat carries four of the eight flagged additives, and ease back on those. The researchers themselves pointed to existing advice about eating less processed food rather than to any new restriction.
Is vitamin C bad for me, since E300 is on the list?
No conclusion of that kind can be drawn, and the point is actively disputed. Ascorbic acid was flagged as part of a study of what additives people were eating, and it also occurs naturally in fruit and vegetables in far larger amounts than any food fortification provides. Nobody involved in the study suggested vitamin C is harmful, and a commentary published in the same journal in August 2026 argues the opposite, citing randomized trials in which vitamin C supplements lowered blood pressure rather than raising it. That argument is set out in our section on the disputed vitamin C finding.
Are natural or organic preservatives safer than synthetic ones?
The distinction this study draws is between antioxidant and non-antioxidant function, not natural and artificial. Rosemary extract is a natural herb product, and it was among the flagged additives. “Natural” on a label is not a reliable proxy for a lower risk profile.
Do any two foods never expire?
Foods that do not require preservatives are the ones most likely to be exempt from best-before dates, and these tend to be whole or minimally processed items such as fresh fruit and vegetables, plain dried pulses, and honey. This is a separate question from additive safety, but it points in the same direction: the items least likely to carry E-numbers are usually the ones needing the least protection from spoilage.
Has any regulator banned these preservatives because of this study?
No. Dr. Touvier asked regulators to take another look at the risks and benefits of these additives, which is a request, not a decision. The existing daily limits for additives like sodium nitrite are all still in place, and a study like this is one piece of information among many that regulators weigh.
What if I already have high blood pressure?
Dietary changes are one part of managing blood pressure, alongside medication where prescribed, physical activity, weight, and other factors. Preservatives are not a recognized treatment target. If you are concerned, the more established dietary levers, such as sodium reduction and the potassium-rich approaches, have stronger evidence behind them. Our guide to potassium salt substitutes for resistant hypertension covers that approach in detail. Please discuss any dietary change with your doctor.
The takeaway
A large, carefully conducted study found that people eating the most food preservatives were more likely to develop high blood pressure and heart disease over roughly eight years. Eight specific additives were linked to high blood pressure, and one of those was also linked to heart disease. Those findings are real, and they are worth taking seriously as a warning sign.
What they are not is a verdict. The study could not show cause and effect, most of the volunteers were middle-aged women, and three of the eight flagged substances are things you also eat in fruit and vegetables. The most useful response is not panic and not dismissal, but a return to the ingredient list and a sense of proportion about which foods carry these additives in the largest amounts.
For the wider evidence on how processed food affects heart health across a much larger body of research, see our guide to the health risks of ultra-processed foods, which covers all 32 outcomes studied.
