My Healthcare News & Research — September 28, 2026 · Eight Common Food Preservatives Linked to High Blood Pressure: What the 112,000-Person French Study Actually Found

On September 20, 2026, the European Society of Cardiology issued a press release, carried by ScienceDaily under the headline “8 common food additives linked to high blood pressure and heart disease.” The study behind it, published online in the European Heart Journal on May 20, 2026, comes from the French NutriNet-Santé cohort — the group whose earlier papers linked emulsifiers and artificial sweeteners to cardiovascular disease, emulsifiers to type 2 diabetes, and additive nitrites to hypertension and type 2 diabetes. It followed 112,395 adults for a median of almost eight years, logging what they ate down to the brand, and asked whether the people who took in the most preservatives developed high blood pressure or cardiovascular disease more often than those who took in the least.

They did. The third of participants with the highest intake of non-antioxidant preservatives (the class that stops mould and bacteria: sorbates, sulphites, nitrites) had a 29% higher rate of new hypertension and a 16% higher rate of cardiovascular disease than the lowest third; the highest consumers of antioxidant preservatives (the class that stops rancidity and browning: ascorbates, erythorbates, citric acid, rosemary extract) had a 22% higher rate of hypertension. Of the 17 preservatives that at least one in ten participants consumed, eight stayed associated with hypertension after correcting for the number of tests run, and one — ascorbic acid, which is vitamin C used as an additive — was also associated with cardiovascular disease itself.

These are not exotic chemicals: citric acid was in the diet of 91% of participants, ascorbic acid 83%, sodium nitrite 73%, potassium sorbate 65%, and, as the paper notes, the preservatives studied are authorised across the European Union and widely used elsewhere, including North America. But the study is observational, so it cannot prove the preservatives caused anything, and people who eat more preservatives eat more ultra-processed food generally. This article walks through what was measured, what each number means, where the headline overstates the paper, how a preservative might plausibly raise blood pressure, what regulators have already said, and what the authors conclude.

Table of Contents

  1. The Study at a Glance
  2. What the Researchers Actually Measured
  3. The Eight Preservatives and Their Numbers
  4. Where the Exposure Comes From
  5. Where the Headline Overreaches
  6. How a Preservative Could Raise Blood Pressure
  7. The Limits: What a Cohort Study Cannot Tell You
  8. How It Fits With Earlier NutriNet-Santé Findings
  9. What the Regulators Say
  10. What the Findings Show
  11. Sources and Primary Documents
  12. Key Research Papers
  13. Connections
  14. Featured Videos

The Study at a Glance

A hazard ratio is the rate of an event in one group divided by the rate in a comparison group: 1.29 means new hypertension diagnoses accumulated 29% faster among the highest consumers. The 95% confidence interval is the range compatible with the data; when it excludes 1.00 the result is conventionally called statistically significant.

Back to Table of Contents

What the Researchers Actually Measured

The reason this cohort can ask about additives at all, when almost no other can, is how it records food: every six months, three 24-hour dietary records on randomly assigned days, through a validated web interface that asks for the brand of every industrial product (median 18 records per person; maximum 96). Two brands of the same sliced ham can contain quite different preservatives, and a food-frequency questionnaire cannot see the difference.

Those records were merged with three composition databases (among them the crowd-sourced Open Food Facts), matching each consumption to the closest composition record in time to allow for reformulation. Doses came from laboratory assays the team ran on real food matrices and from other sources, such as dose data obtained from the European Food Safety Authority (EFSA) under a public-access-to-documents request, giving an estimated daily intake of each of 58 preservatives per participant, updated every two years.

Two definitions matter. “Preservative” here covers both preservatives proper (the E200s, which stop micro-organisms) and antioxidants (the E300s, which stop rancidity and browning). And several of these substances occur naturally — citric acid in lemons, ascorbic acid in fruit, nitrates in leafy greens — so natural intake was quantified separately and adjusted for. On average the additive source accounted for 16% of citric acid intake, 29% of ascorbate intake and 63% of sulphite intake.

Diagnoses were validated by a physician committee from medical records, and the cohort is linked to the national health-insurance database and the national mortality registry. A hypertension case means a diagnosis made between enrolment and December 31, 2024, in someone who did not have hypertension at entry.

The statistics were Cox proportional-hazards models with age as the time scale, comparing sex-specific thirds of intake (for an additive eaten by fewer than two-thirds of participants: non-consumers against lower and higher consumers split at the median), adjusted for sex, height, body mass index, physical activity, smoking, education, family history, number of dietary records, and daily intakes of energy, alcohol, saturated fat, sodium, fibre, sugars, fruit and vegetables, dairy, and red and processed meat. A sensitivity analysis added the ultra-processed share of the diet (NOVA classification); the authors report the associations held.

Back to Table of Contents

The Eight Preservatives and Their Numbers

Hazard ratios for new hypertension, higher versus lower consumers (or versus non-consumers, for additives eaten by fewer than two-thirds of participants). All survived false-discovery-rate correction, the standard adjustment for having tested 17 substances at once.

Non-antioxidant preservatives (stop microbes)

Antioxidant preservatives (stop oxidation)

What was not associated, and a sense of scale

Nitrates, tocopherols (vitamin E), lecithins and the other substances tested were not among the associations that remained significant after correction.

The paper reports hazard ratios, not absolute rates by tertile. For scale, an illustrative calculation from the crude figures (not reported in the paper): 5,544 diagnoses among 103,386 people is about 5.4 new hypertension cases per 100 participants over a median 7.6 years, so a 29% relative increase would mean roughly 1.5 extra diagnoses per 100 people over that period; for cardiovascular disease (2,450 among 110,356, about 2.2 per 100), 16% more is roughly a third of one extra event per 100. These are illustrations, and this is a young cohort: the same relative increase in an older population would translate into more absolute cases.

Back to Table of Contents

Where the Exposure Comes From

In this cohort alcoholic drinks supplied 84% of sulphites; processed meat supplied 54% of nitrites, 77% of nitrates and 42% of erythorbates; processed fruit and vegetable products supplied 51% of ascorbates and 26% of citrates; and sorbates and citrates had no single dominant source: the paper’s Figure 1 shows sorbates coming in shares of under a quarter each from processed fruit and vegetable products, refined grain products, dairy products, and fats and sauces, and citrates, beyond processed fruit and vegetable products, from confectionery, cakes and biscuits, and sweetened and unsweetened drinks. The authors describe most preservatives as ubiquitous rather than tied to one food.

The highest third of preservative consumers were younger (mean 39 versus 46), less physically active, and had less family history of hypertension; they ate more sugar (104 g a day versus 83 g), more sodium (2,851 mg versus 2,590 mg), more red and processed meat, and a much larger share of ultra-processed food (21% of food weight versus 14%). All of those were adjusted for in the main model except the ultra-processed share, which was added in a sensitivity analysis; whether adjustment fully removes their influence is taken up under limits.

Against Europe’s acceptable daily intakes (ADIs) the cohort mostly stayed within bounds: nobody exceeded the ADI for sorbates, erythorbates or nitrates, though 96 participants exceeded the sulphite ADI and 54 the nitrite ADI. For almost all participants, then, the associations were observed at intakes below these ADIs. (The sulphite ADI the paper uses, 0.7 mg sulphur dioxide equivalent per kg, is the temporary value EFSA withdrew in 2022; see the regulators section.)

Back to Table of Contents

Where the Headline Overreaches

The headline — “8 common food additives linked to high blood pressure and heart disease” — is accurate about the count and honest about “linked,” but it compresses three distinctions the paper keeps separate.

First, the study is about preservatives, not food additives in general. Colours, sweeteners and thickeners were not examined, and emulsifiers only where they also act as preservatives or antioxidants (lecithins, for example); this paper says nothing about xanthan gum or aspartame.

Second, the eight were linked to high blood pressure; only one was linked to heart disease. Ascorbic acid alone also reached significance for cardiovascular disease, with a corrected P-value exactly at 0.05. The heart-disease signal is carried by the total of non-antioxidant preservatives, a summed exposure.

Third, “linked” is doing all the work it always does. The release’s own summary says so: the observational study “cannot prove they were directly responsible.” The authors estimate that 16% of the association between non-antioxidant preservatives and cardiovascular disease ran through raised blood pressure and about 5% through type 2 diabetes, a pattern consistent with a causal pathway and also with a confounder shared by all three.

The authors describe the work as the first prospective study of a wide range of preservatives and cardiovascular health, built on brand-level exposure data, and write that, if confirmed, the results “call for the re-evaluation of regulations governing the use of these additives to improve consumer protection.” The study does not show that the preservative in any single product raises an individual’s blood pressure.

Back to Table of Contents

How a Preservative Could Raise Blood Pressure

The authors write that experimental research is needed to establish mechanisms. The evidence they cite comes from cell studies and animal models, some at pharmacological doses, whereas authorised additive doses are, in their words, “physiological and not pharmacological.” What they assemble:

Sorbates. The authors describe potassium sorbate as a trans fatty acid, note that the harm of trans fats to cardiovascular health, including raised blood pressure, is well documented, and suggest the trans configuration may play a role. The in vitro studies they cite on trans fats concern eighteen-carbon trans fatty acids; sorbic acid has six carbons. In laboratory studies potassium sorbate showed cytotoxicity in human cell models and activated advanced glycation end-products, sugar-protein compounds implicated in arterial stiffening.

Nitrites. The authors write that nitrites can promote oxidative damage and that N-nitroso compounds, formed essentially during meat processing, increase the risk of insulin resistance, itself a contributor to hypertension; nitrites increased insulin resistance in rats. Additive nitrates (E251, E252), which came mostly from processed meat in this cohort, showed no association. Nitrate naturally present in vegetables, which the body can convert to nitric oxide, a vasodilator, has shown cardiovascular benefits in some intervention studies, the authors note, but their results do not support cardiovascular benefits from additive nitrites or nitrates.

Ascorbic acid and erythorbates — the vitamin C paradox. In systematic reviews of prospective studies, higher intake and blood levels of vitamin C from foods tended to be associated with lower cardiovascular risk, yet added ascorbic acid, chemically identical, was associated with more hypertension and cardiovascular disease here, while added tocopherols were not. The authors note that identical molecules can act differently depending on the food matrix (vitamin C in an orange arrives with fibre, potassium and polyphenols; as E300 it arrives in a processed product), the dose, and interactions that affect bioavailability, and that an umbrella review found no overall effect of vitamin C supplementation on arterial stiffness and other cardiovascular biomarkers, with some studies even suggesting increased risk. A further reading, which an observational design cannot exclude, is that added ascorbic acid marks the consumption of processed fruit products and cured meats. Sodium erythorbate, an isomer with little vitamin C activity, gave an almost identical hypertension hazard ratio (1.14), which is consistent with either reading.

The pancreas and the microbiome. Several of these compounds disturb insulin secretion or action in experimental systems, and the same group reported in Nature Communications in January 2026 that twelve of the 17 commonly consumed preservatives were associated with new type 2 diabetes in this cohort. A second hypothesis concerns the gut: preservatives are by definition substances that inhibit micro-organisms, and a co-author of this paper, Benoit Chassaing (now at the Institut Pasteur), reported in 2015 that two dietary emulsifiers, carboxymethylcellulose and polysorbate-80, altered the mouse gut microbiota and induced low-grade inflammation and metabolic syndrome. Whether sorbates or sulphites do the same at food doses in humans is not known.

Back to Table of Contents

The Limits: What a Cohort Study Cannot Tell You

It is observational. The paper’s first stated limitation is that the design “precludes causal inference based on this study alone, and residual confounding cannot be entirely ruled out.” Long-term randomised trials feeding people additives such as sodium nitrite have not been run, so cohort studies are the main human evidence.

Confounding by ultra-processed food. Preservatives travel almost exclusively in industrially processed products, and the same cohort showed in 2019 that each 10-percentage-point increase in the ultra-processed share of the diet predicts a 12% higher rate of cardiovascular disease. Everything else in those products — sugar, salt, refined starch, packaging chemicals, displaced whole foods, the habits of the person eating them — is a candidate explanation. The authors’ response: the model adjusts for the major nutrients and food groups, adding the ultra-processed share did not remove the associations, and preservatives are used across products with diverse processing levels. The results also differed within classes (additive nitrites associated, additive nitrates not; ascorbates associated, tocopherols not). Adjustment for a variable that is measured with error removes only part of its influence, which is why residual confounding remains possible.

The healthy-volunteer problem. Participants were mostly women, better educated and healthier than the French population, and only 5% developed hypertension during follow-up against roughly 30% prevalence in France, largely because the cohort is young. Undiagnosed hypertension could not be excluded, though the authors argue that such misclassification would more likely dilute associations than create them, and the expected associations for sodium (HR 1.13) and smoking did appear.

Exposure measurement. Doses came partly from laboratory assays and partly from other sources such as EFSA use data, rather than from measurement of each item eaten, and there are no biomarkers for most preservatives, so no biological validation was possible.

Multiple testing and small effects. The authors applied false-discovery-rate correction, ran a negative-control outcome (hip fracture, an outcome with no expected causal link to preservatives; none was associated), and computed E-values, all 1.5 or above except rosemary extract. An E-value of 1.5 means an unmeasured confounder associated with both exposure and outcome by a factor of 1.5 could fully explain the result; the authors conclude that it is “unlikely for detected associations to be entirely ruled out by residual confounding.” Most individual hazard ratios lie between 1.10 and 1.16, a range in which modest biases can account for an association; the sorbate (1.39) and citric acid (1.25) associations are larger.

One cohort, one group. Every paper in this series comes from one cohort, one food database and one analytic tradition; a systematic bias in how it measures diet or selects volunteers would reproduce itself in every paper. The paper notes that, apart from nitrites and nitrates in processed meat, no other cohort study has examined these preservatives and cardiovascular outcomes, probably because brand-level data are lacking, and that the results need confirmation in other epidemiological studies.

Back to Table of Contents

How It Fits With Earlier NutriNet-Santé Findings

Since 2019 the group has examined one class of additive after another in the same cohort:

Across the series, several additive classes have been associated with cardiometabolic outcomes, and within each class some members were associated and others were not. Because every paper comes from the same cohort and methods, the single-cohort limitation described above applies to all of them.

Back to Table of Contents

What the Regulators Say

Europe: EFSA’s rolling re-evaluations

Food additives authorised in the EU before January 2009 are subject to a re-evaluation programme carried out by EFSA’s panels; according to the paper, 25 EFSA re-evaluations over the past ten years concerned preservatives, and 16 of them set an acceptable daily intake (ADI). Where the eight stand:

According to the paper, ADIs were derived from experimental toxicology endpoints — behavioural, carcinogenic, developmental, haematological, reproductive and thyroid toxicity, growth retardation, raised blood methaemoglobin and mortality; blood pressure and cardiovascular outcomes are not among those listed. The authors frame any re-evaluation as a benefit/risk balance between food preservation and cardiovascular health; nitrite, for example, inhibits Clostridium botulinum in cured meat.

United States: the GRAS proposal

In the United States most of these substances are listed or affirmed as “generally recognized as safe” (GRAS) in FDA regulations — potassium sorbate, potassium metabisulfite, ascorbic acid, sodium ascorbate and erythorbic acid in 21 CFR Part 182, citric acid in Part 184 — while sodium nitrite is an approved food additive (21 CFR 172.175). Under the GRAS provision a company may also conclude on its own that an ingredient is GRAS and notify the FDA only voluntarily. On August 11, 2026 the FDA published a proposed rule (91 FR 51834, Docket FDA-2025-N-3262) that would make GRAS notification mandatory; comments close December 9, 2026. Substances already listed or affirmed as GRAS in Parts 182, 184 or 186 would be excepted from the notice requirement, and uses already on the market on the basis of a company’s own GRAS conclusion could instead be reported in a one-year streamlined submission, which the FDA says would help it prioritise post-market review. The proposal is the subject of the August 10 News article. It does not by itself open a new safety review of the listed preservatives of the kind the French authors call for.

Back to Table of Contents

What the Findings Show

The authors write that their results “do not call into question food-based dietary guidelines or results based on scores for compliance to these diets (DASH, Mediterranean diet) but rather complement them.” Alongside sufficient fruit and vegetables, fish, legumes and fibre and limited processed meat and salt, they write that “it would also seem preferable to favour ‘fresh and minimally processed’ versions of these products as officially recommended for consumers and patients in the French nutritional health programme.” In the press release, the senior author called for a re-evaluation of the risks and benefits of these additives by the authorities in charge, such as EFSA in Europe and the FDA in the United States.

How the preservatives appear on labels

In the EU and UK, preservatives and antioxidants are declared by E-number or by name; in the United States, by name. The groups in this study, with where the cohort’s intake came from:

The preservatives often travel together. The paper identified three mixtures that participants tended to consume together; none was associated with cardiovascular disease, but two were associated with higher hypertension incidence: one mainly acetic and tartaric acids, citric acid esters of mono- and diglycerides, potassium metabisulphite and potassium sorbate; the other mainly citric and phosphoric acids, sodium nitrite, sodium erythorbate and potassium sorbate.

Which foods carried which preservatives

Foods in their fresh or minimally processed form — freshly cooked meat, poultry or fish, eggs, whole fruit and vegetables, plain whole-milk yogurt, water, and intact whole grains such as brown rice and oats — do not contain these substances as additives; this is the contrast with industrial products that the authors draw.

The vitamin C result concerns ascorbic acid added to processed products: the systematic reviews the authors cite found that higher intake and blood levels of vitamin C from foods tended to be associated with lower risk of cardiovascular disease and hypertension. The study did not examine treatment; people who already had hypertension at entry were excluded from the hypertension analysis, and decisions about blood-pressure treatment belong with a clinician.

Back to Table of Contents

Sources and Primary Documents

  1. Hasenböhler A, Javaux G, Payen de la Garanderie M, Szabo de Edelenyi F, Yvroud-Hoyos P, Agaësse C, De Sa A, Huybrechts I, Pierre F, Coumoul X, Fezeu LK, Galan P, Blacher J, Julia C, Allès B, Hercberg S, Chassaing B, Deschasaux-Tanguy M, Srour B, Touvier M (2026). Preservative food additives, hypertension, and cardiovascular diseases: the NutriNet-Santé study. European Heart Journal. — doi:10.1093/eurheartj/ehag308
  2. The same paper on PubMed (online ahead of print, May 20, 2026). — Search PubMed
  3. Open-access author manuscript, HAL open archive (INRAE), record hal-05630247. hal.inrae.fr/hal-05630247
  4. European Society of Cardiology, via ScienceDaily. 8 common food additives linked to high blood pressure and heart disease. September 20, 2026. sciencedaily.com
  5. NutriNet-Santé cohort registration, ClinicalTrials.gov NCT03335644. clinicaltrials.gov · Study website: etude-nutrinet-sante.fr
  6. EFSA Panel on Food Additives and Flavourings; Younes M, Aquilina G, et al. (2022). Follow-up of the re-evaluation of sulfur dioxide (E 220), sodium sulfite (E 221), sodium bisulfite (E 222), sodium metabisulfite (E 223), potassium metabisulfite (E 224), calcium sulfite (E 226), calcium bisulfite (E 227) and potassium bisulfite (E 228). EFSA Journal. — doi:10.2903/j.efsa.2022.7594
  7. EFSA Panel on Food Additives and Nutrient Sources added to Food; Mortensen A, Aguilar F, et al. (2017). Re-evaluation of potassium nitrite (E 249) and sodium nitrite (E 250) as food additives. EFSA Journal. — doi:10.2903/j.efsa.2017.4786
  8. EFSA Panel on Food Additives and Flavourings (2019). Opinion on the follow-up of the re-evaluation of sorbic acid (E200) and potassium sorbate (E202) as food additives. EFSA Journal. — doi:10.2903/j.efsa.2019.5625
  9. EFSA Panel on Food Additives and Nutrient Sources added to Food; Younes M, Aggett P, et al. (2018). Refined exposure assessment of extracts of rosemary (E 392) from its use as food additive. EFSA Journal. — doi:10.2903/j.efsa.2018.5373
  10. EFSA Panel on Food Additives and Nutrient Sources added to Food (2015). Scientific Opinion on the re-evaluation of ascorbic acid (E 300), sodium ascorbate (E 301) and calcium ascorbate (E 302) as food additives. EFSA Journal 13(5):4087. — doi:10.2903/j.efsa.2015.4087
  11. EFSA Panel on Food Additives and Nutrient Sources added to Food (2016). Scientific Opinion on the re-evaluation of erythorbic acid (E 315) and sodium erythorbate (E 316) as food additives. EFSA Journal 14(1):4360. — doi:10.2903/j.efsa.2016.4360
  12. EFSA Panel on Food Additives and Nutrient Sources added to Food (2016). Scientific Opinion on the re-evaluation of sulfur dioxide (E 220), sodium sulfite (E 221), sodium bisulfite (E 222), sodium metabisulfite (E 223), potassium metabisulfite (E 224), calcium sulfite (E 226), calcium bisulfite (E 227) and potassium bisulfite (E 228) as food additives. EFSA Journal 14(4):4438. — doi:10.2903/j.efsa.2016.4438
  13. U.S. Food and Drug Administration. Substances Generally Recognized as Safe. Proposed rule, 91 FR 51834, Docket No. FDA-2025-N-3262, published August 11, 2026; comments close December 9, 2026. federalregister.gov/d/2026-16296

Back to Table of Contents

Key Research Papers

The background literature: the earlier NutriNet-Santé additive papers this study builds on, its companion diabetes paper, and the experimental work behind the microbiome hypothesis, followed by the group’s 2022 nitrite paper and the study itself.

  1. Srour B, Fezeu LK, Kesse-Guyot E, et al. (2019). Ultra-processed food intake and risk of cardiovascular disease: prospective cohort study (NutriNet-Santé). BMJ 365:l1451. — doi:10.1136/bmj.l1451 · PubMed PMID: 31142457
  2. Debras C, Chazelas E, Sellem L, et al. (2022). Artificial sweeteners and risk of cardiovascular diseases: results from the prospective NutriNet-Santé cohort. BMJ 378:e071204. — doi:10.1136/bmj-2022-071204 · PubMed PMID: 36638072
  3. Sellem L, Srour B, Javaux G, et al. (2023). Food additive emulsifiers and risk of cardiovascular disease in the NutriNet-Santé cohort: prospective cohort study. BMJ 382:e076058. — doi:10.1136/bmj-2023-076058 · PubMed PMID: 37673430
  4. Srour B, Chazelas E, Druesne-Pecollo N, et al. (2023). Dietary exposure to nitrites and nitrates in association with type 2 diabetes risk: Results from the NutriNet-Santé population-based cohort study. PLOS Medicine 20(1):e1004149. — doi:10.1371/journal.pmed.1004149 · PubMed PMID: 36649248
  5. Salame C, Javaux G, Sellem L, et al. (2024). Food additive emulsifiers and the risk of type 2 diabetes: analysis of data from the NutriNet-Santé prospective cohort study. The Lancet Diabetes & Endocrinology 12(5):339-349. — doi:10.1016/S2213-8587(24)00086-X · PubMed PMID: 38663950
  6. Payen de la Garanderie M, Hasenbohler A, Dechamp N, et al. (2025). Food additive mixtures and type 2 diabetes incidence: Results from the NutriNet-Santé prospective cohort. PLOS Medicine 22(4):e1004570. — doi:10.1371/journal.pmed.1004570 · PubMed PMID: 40198579
  7. Hasenböhler A, Javaux G, Payen de la Garanderie M, et al. (2026). Associations between preservative food additives and type 2 diabetes incidence in the NutriNet-Santé prospective cohort. Nature Communications 16:11199. — doi:10.1038/s41467-025-67360-w · PubMed PMID: 41501013
  8. Chassaing B, Koren O, Goodrich JK, et al. (2015). Dietary emulsifiers impact the mouse gut microbiota promoting colitis and metabolic syndrome. Nature 519(7541):92-96. — doi:10.1038/nature14232 · PubMed PMID: 25731162
  9. Srour B, Chazelas E, Fezeu LK, et al. (2022). Nitrites, nitrates, and cardiovascular outcomes: are we living “la vie en rose” with pink processed meats? Journal of the American Heart Association 11(24):e027627. — doi:10.1161/JAHA.122.027627 · PubMed PMID: 36533633
  10. Hasenböhler A, Javaux G, Payen de la Garanderie M, et al. (2026). Preservative food additives, hypertension, and cardiovascular diseases: the NutriNet-Santé study. European Heart Journal, online May 20, 2026: ehag308. — doi:10.1093/eurheartj/ehag308 · Search PubMed

PubMed Topic Searches

  1. PubMed: NutriNet-Santé food additives

Back to Table of Contents

Connections

Back to Table of Contents