A French cohort of 112,395 adults who logged their diets for a median of 7.9 years developed hypertension and cardiovascular disease at higher rates when their food contained more preservatives, according to a peer-reviewed analysis published in the European Heart Journal. The findings single out non-antioxidant preservatives as the category most strongly tied to elevated risk, raising pointed questions about whether current regulatory limits account for the cumulative effect of eating these additives day after day.
Why a nearly eight-year French diet study matters right now
Hypertension is the single largest modifiable driver of heart attack and stroke worldwide, and processed foods remain a dominant source of daily calories across Europe and North America. The NutriNet-Sante study, which enrolled participants across France, tracked not just broad dietary patterns but the specific preservative additives people consumed. Over a median follow-up of 7.9 years, the researchers found that non-antioxidant preservatives were associated with higher hypertension incidence and with cardiovascular disease events. Antioxidant preservatives showed a link to hypertension as well, though the association with broader cardiovascular outcomes was less pronounced.
The distinction between preservative subtypes matters because it points toward a possible biological mechanism. Non-antioxidant preservatives, a group that includes compounds such as nitrites, benzoates, and sorbates, have been studied in smaller trials for their effects on gut bacteria and vascular function. One working hypothesis holds that repeated low-dose exposure to these chemicals alters gut microbiota composition in ways that promote low-grade inflammation and endothelial dysfunction, the cellular damage that precedes clinical hypertension. Shorter feeding studies tracking inflammatory biomarkers could test that chain of events, but the NutriNet-Sante data set is the first large-scale, long-duration cohort to document the downstream clinical outcomes at population scale.
What 112,395 participants and 7.9 years of data actually show
The study, formally titled “Preservative food additives, hypertension, and cardiovascular diseases: the NutriNet-Sante study,” drew on repeated 24-hour dietary records that captured brand-level product information, allowing researchers to estimate individual preservative intake with unusual precision. The cohort of 112,395 adults in France was large enough to stratify results by preservative type while adjusting for overall diet quality, sodium intake, body mass index, and other established cardiovascular risk factors.
The analysis separated preservatives into two families. Non-antioxidant preservatives, used primarily to inhibit microbial growth in processed meats, baked goods, and beverages, showed the clearest dose-response relationship with both new-onset hypertension and cardiovascular disease. Antioxidant preservatives, added to prevent fat oxidation in oils and snack foods, were associated with hypertension but did not show the same strength of association with cardiovascular events. That split suggests the health signal is not simply a proxy for eating processed food in general. If it were, both preservative categories would track similarly, since they appear in overlapping product types.
The study is registered as a clinical trial, which means its methods and endpoints were declared before the data were analyzed. That registration reduces the risk of selective reporting, a common concern in observational nutrition research. The European Heart Journal, where the results were published, is among the highest-impact cardiology journals, and the paper underwent formal peer review.
Regulatory oversight of these additives in Europe falls to the European Food Safety Authority, which evaluates each compound individually and assigns acceptable daily intake levels. The European Commission maintains a database of authorized food additives and their permitted uses. Those safety evaluations, however, are designed around single-additive toxicity testing. They do not typically model what happens when a person consumes multiple preservatives across several meals every day for years, which is exactly the exposure pattern the NutriNet-Sante cohort captured.
Gaps in the evidence and what to watch next
The study is observational, not experimental, so it cannot prove that preservatives directly cause hypertension or heart disease. The researchers adjusted for many confounders, but residual confounding from ultra-processed food intake, socioeconomic factors, or unmeasured lifestyle variables is always possible in epidemiological work. The individual-level dietary records and exact preservative quantities consumed are not publicly released in the cohort dataset, which limits independent replication.
No direct statements from the study investigators address whether the statistical adjustments fully separate the effect of preservatives from the broader health consequences of eating ultra-processed food. That distinction is the sharpest unresolved question in the data. If the preservative signal is really just a marker for a diet heavy in packaged food, then reformulating products to remove specific additives would not reduce cardiovascular risk. If the signal is specific to certain chemicals, targeted reformulation or tighter intake limits could have real public health value.
Shorter-term feeding studies that measure inflammatory biomarkers, gut microbiota shifts, and early vascular changes in response to controlled preservative doses would help clarify causality. Randomized trials could, for example, compare two versions of the same processed foods-identical in salt, sugar, and fat but differing in preservative content-to see whether blood pressure or endothelial function diverges over weeks or months. Parallel mechanistic work in cell and animal models could probe how nitrites, benzoates, and sorbates interact with vascular tissue and immune signaling pathways.
Another gap is the lack of data on cumulative and mixture effects. Current safety assessments typically test one preservative at a time, yet consumers encounter cocktails of additives in single meals and across the day. The NutriNet-Sante findings hint that the total burden of non-antioxidant preservatives may matter more than any single compound. Future cohort analyses could explicitly model mixture exposures, while laboratory studies could examine whether combinations of preservatives have additive or synergistic effects on blood vessels and the gut microbiome.
Researchers also note that the cohort is French and skewed toward individuals who volunteer for detailed nutrition tracking, which may limit how broadly the findings apply. Diets in other regions feature different processed foods, culinary traditions, and regulatory limits on additives. Replicating similar analyses in other large cohorts, ideally with comparable brand-level dietary data, will be important for confirming whether the observed associations hold in more diverse populations.
What this means for consumers and regulators
For individual consumers, the study reinforces a familiar but still challenging message: relying heavily on packaged and processed foods may carry cardiovascular risks that go beyond salt, sugar, and fat. Because non-antioxidant preservatives are most common in processed meats, ready-made baked goods, and sweetened beverages, cutting back on those products and cooking more meals from basic ingredients is a pragmatic way to reduce exposure. Reading ingredient lists for nitrites, benzoates, sorbates, and other preservatives can help people identify where these additives cluster in their diets.
For regulators, the findings raise uncomfortable questions about whether current acceptable daily intakes adequately protect against long-term cardiovascular outcomes. Even if each preservative meets safety thresholds in isolation, the NutriNet-Sante data suggest that real-world, chronic exposure to multiple additives may still be linked to hypertension and heart disease. Authorities could respond by commissioning updated risk assessments that incorporate mixture modeling, funding mechanistic studies focused on vascular endpoints, and encouraging or mandating reformulation where feasible.
Food manufacturers, meanwhile, face a strategic choice. Waiting for definitive causal proof could take years, but consumer concern about additives is already shaping purchasing decisions. Companies that proactively reduce non-essential preservatives, especially in categories like processed meats and snack foods, may be better positioned if regulators tighten rules or if further research strengthens the case for harm. Reformulation is not trivial-preservatives play real roles in food safety and shelf life-but the French cohort’s findings suggest that innovation in this area could yield both health and market benefits.
The NutriNet-Sante study does not settle the debate over food additives, but it shifts the burden of proof. With more than 100,000 adults followed for nearly eight years, the signal linking higher preservative intake to hypertension and cardiovascular disease is difficult to dismiss as mere statistical noise. The next phase of research will determine whether that signal points to specific chemicals that can be regulated or reformulated away, or whether it is simply another warning light on the broader dashboard of ultra-processed diets. Either way, the message is clear: what keeps food on the shelf longer may not be neutral for the heart over the long term.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
