Every research record we hold on this ingredient, printed in full — the citation, who funded the work, what the authors declared, and the date we read the source. The ingredient page shows the same records in short.
This concerns the EXTRACTION PROCESS, not the fatty acid — a separate question from the linoleic acid record, and one where the evidence gap is real.
Most conventionally produced seed oils are extracted using n-hexane, a petroleum-derived solvent, because it recovers far more oil than mechanical pressing. Nearly all of it is removed during refining.
WHAT IS ESTABLISHED: EPA classifies hexane as a hazardous air pollutant. Chronic INHALATION exposure in humans is associated with polyneuropathy — numbness in the extremities, muscular weakness, blurred vision. In the EU it carries harmonised classification as a reproductive toxicity hazard (Cat. 2) and organ toxicity hazard.
WHAT IS NOT ESTABLISHED, AND THIS IS THE POINT: the ORAL toxicity of hexane has never been properly characterised. EPA has no Reference Dose for chronic oral exposure, and per EPA no studies have evaluated the toxicity of hexane consumption in humans or laboratory animals. Peer-reviewed assessment notes that residue data in refined oils are "scarce and outdated" and that existing toxicology does not reflect long-term, low-level consumer exposure. EFSA is currently conducting a re-assessment.
REGULATORY DIVERGENCE: the United States sets NO maximum residue limit for hexane in edible oils. FDA treats it as a processing aid rather than an ingredient, so it need not appear on labels and residues are not monitored. The EU caps it at 1 mg/kg, Canada at 10 ppm, Australia and New Zealand at 0.3 mg/kg. See this ingredient's regulatory status records for the comparison.
PRACTICAL NOTE: expeller-pressed and cold-pressed oils are mechanically extracted and do not use hexane at all. Where a label states expeller-pressed, this concern does not apply.
This record makes no claim that hexane residue in oil is harmful. It states that in the US it is unregulated, unmonitored, and its oral toxicity is uncharacterised — so the honest answer is that nobody knows, which is itself a legitimate basis for preferring mechanically pressed oils.
Studies raise questions; harm in humans has not been established.
SCOPE: this record addresses the fatty acid (linoleic acid), NOT the extraction process. For extraction solvents, see the separate hexane record on this ingredient — the concerns are different and the evidence quality differs sharply.
On linoleic acid and inflammation: the widely circulated claim that it promotes inflammation is not supported by the current evidence. Randomized controlled trials consistently find linoleic acid does not raise inflammatory markers, and several meta-analyses associate higher intake with LOWER C-reactive protein and interleukin-6. A pooled analysis of roughly 70,000 participants across 30 prospective cohorts (Marklund et al., Circulation 2019) found higher blood linoleic acid associated with about 22% lower cardiovascular mortality. The proposed mechanism behind the concern — linoleic acid converting to arachidonic acid, which then drives inflammation — is specifically contradicted: dietary linoleic acid intake has negligible effect on blood or tissue arachidonic acid levels.
FUNDING CAVEAT, STATED PLAINLY: a frequently cited 2026 review supporting this conclusion (Petersen, Messina & Flickinger, Nutrition Today) is substantially industry-connected. Per its own disclosure, author M.M. is an EMPLOYEE of Soy Nutrition Institute Global, which is funded by the United Soybean Board; author B.F. consults for the same body and provides regulatory affairs services to food industry clients; author K.S.P. received honoraria from SNI Global plus grants from Cotton Incorporated, the National Cattlemen's Beef Association and others. That paper is promoted on SNI Global's own website. It has been REMOVED as a supporting citation for this record. The citation now rests on the Marklund pooled cohort analysis and an independent academic review. Note the funding basis of the Marklund analysis itself has not been separately verified, so this record is marked undetermined rather than independent.
WHAT THIS RECORD DOES NOT SAY: it does not claim seed oils are optimal, nor that avoiding them is unreasonable. Reasonable grounds for avoidance exist that this evidence does not address — extraction solvent residue (see the hexane record), a preference for whole-food fats, or the fact that seed oils concentrate in ultra-processed foods.