Olive Oil Blended With Thyme Lowered Oxidized LDL: A 12-Person Crossover Trial

Three olive oils, twelve people, and only one arm that moved. A 2017 randomized crossover trial in the European Journal of Nutrition gave adults with high cholesterol 25 mL a day of three different virgin olive oils: an ordinary one, one enriched to 500 mg of olive polyphenols per kg, and one enriched to the same 500 mg per kg using a half and half mixture of olive and thyme phenolics. The oil that lowered oxidized LDL and raised bifidobacteria was the thyme blend. The olive only oil, carrying roughly twice the olive phenol dose, did neither. Sonia Martin-Pelaez and colleagues at the Hospital del Mar Research Institute in Barcelona designed the trial to ask whether olive phenolics change blood lipids through the gut microbiota, and the answer they got is more interesting than a clean yes.

Why This Study Matters

Most olive polyphenol research treats dose as the only variable that matters. Feed people more phenolics, measure more effect. That framing came out of good work: the EUROLIVE trial established that phenolic content changes what olive oil does to blood lipids in a dose dependent way, and a later meta-analysis of 26 randomized trials confirmed the high phenolic versus low phenolic split across a range of endpoints.

This trial asks a different question. It holds total phenolic concentration constant at 500 mg/kg across two enriched oils and changes only where the phenolics came from. One oil is enriched entirely with olive phenolics. The other splits the same total between olive and thyme. If dose were the whole story, the two enriched oils should behave alike. They did not.

The second reason it matters is what got measured alongside the blood work. Very few human trials look at blood lipids, faecal microbial populations, microbial cholesterol metabolites, bile acids, short chain fatty acids and phenolic metabolites in the same participants across the same three weeks. That combination is what lets the authors propose a chain of events rather than a single association, and it is also what makes the null results in this paper worth reading closely.

How It Was Designed

The trial was randomized, controlled, double blind and crossover, registered as ISRCTN77500181 and drawn from the VOHF study, an investigation into virgin olive oil and HDL functionality run in hypercholesterolemic volunteers aged 35 to 80. Each participant took all three oils in sequence: three weeks on each, with two week washout periods in between, so every person served as their own control. Twelve completed the gut analysis reported here.

The oils are the design. All three were matched on fatty acids and on fat soluble micronutrients including alpha tocopherol, lutein, beta cryptoxanthin and beta carotene, in the same types and amounts. Only the phenolic fraction differed. The companion paper from the same trial publishes the full composition, and the numbers reframe the result. Per 25 mL daily dose, the ordinary oil delivered 0.30 mg of hydroxytyrosol derivatives. The olive enriched oil delivered 8.49 mg. The olive and thyme blend delivered 4.30 mg, roughly half the olive enriched oil, because half of its 500 mg/kg came from thyme instead. On top of that the blend carried thyme specific compounds the other two oils had none of: thymol at 0.64 mg, carvacrol at 0.23 mg, rosmarinic acid at 0.41 mg, and thyme flavonoids including thymusin and xanthomicrol.

Compliance was verified rather than assumed. The researchers measured urinary hydroxytyrosol sulfate and thymol sulfate as adherence biomarkers, which confirms both that participants took the oil and which oil they were on. Participants kept three day dietary records at baseline and after each period and were asked to limit other phenol rich foods throughout.

What They Found

Every significant result in this trial landed in one arm or the other, never both, and the split does not follow the dose.

The published abstract reports directions and significance thresholds rather than the underlying concentrations, so this table states what moved, in which arm, and against which comparison. Every row above except the last is a difference the authors report at P less than 0.05. Green marks a favorable direction or evidence of delivery.

Reading the Results

The dose story inverted. The olive enriched oil carried 8.49 mg of hydroxytyrosol derivatives per daily dose against the blend's 4.30 mg. On the logic that has driven most of this literature, it should have been the one to move oxidative markers. It was not. Oxidized LDL fell in the blend arm and nowhere else. Whether that is thyme doing something, or the olive phenol dose being lower, or the two mattering together, this trial cannot separate, because the blend differs from the olive enriched oil on both axes at once.

The olive only oil proved delivery, not effect. Faecal hydroxytyrosol rose against that arm's own baseline and faecal dihydroxyphenylacetic acid rose against the ordinary oil, which means the compounds arrived in the colon and were being worked on there. That is the same signal a later and much larger trial found when olive polyphenols were baked into a biscuit and tracked through urine and blood. It is worth keeping the direction of that fact straight: olive phenols are among the better absorbed dietary polyphenols in the small intestine, and the food matrix they arrive in changes how much gets through. What reaches the colon is a portion of the intake plus its metabolites, not a majority left over from poor absorption.

The bifidobacteria result belongs to the blend. This is the single most misquoted finding in the olive oil and gut literature, and the paper is unambiguous about it. The significant increase in bifidobacteria was the olive and thyme oil measured against the ordinary olive oil. The olive enriched oil produced no quantitative change in any of the microbial populations analysed. Anyone citing this trial as evidence that olive polyphenols raise bifidobacteria is citing the arm that did not.

The proposed chain is a hypothesis. The authors close by suggesting that the drop in oxidized LDL could be mediated by the rise in bifidobacteria together with the rise in phenolic metabolites carrying antioxidant activity. That is a coherent reading of the data and it is explicitly framed as a possibility, not a demonstration. Nothing in a twelve person crossover establishes that the microbial change caused the lipid change. The coprostanone result is a small hint that the two enriched oils left the microbes doing different things with cholesterol, since the olive only oil produced more of it than the blend, but a single microbial metabolite differing between two arms is a thread, not a mechanism.

What Didn't Change

The blood lipid profile. Total cholesterol, LDL, HDL and triglycerides were unmoved by either enriched oil. The only lipid related measure that shifted was oxidized LDL, and only in one arm, against its own starting values rather than against the other two oils.

The olive enriched oil produced no quantitative change in the microbial populations analysed, despite carrying the largest olive phenol dose in the trial. Short chain fatty acids and bile acids were measured in faeces and do not appear among the significant results the authors report.

Limits and What's Still Unknown

Twelve people is the first limit and it shapes everything else. This was a subsample of a larger trial, analysed for outcomes the parent study was not powered to detect, and the crossover design helps with between person variation but does not manufacture statistical power.

The second limit is the comparison the oxidized LDL result rests on. It fell against that arm's own pre-intervention values, not against the other oils. Within arm changes in small samples are the softest kind of positive finding, and they are the ones most likely not to replicate.

The third is that the two enriched oils are not a clean experiment. Going from the olive enriched oil to the blend changes the olive phenol dose and adds an entirely new class of compounds at the same time. Thymol, carvacrol, rosmarinic acid and the thyme flavonoids are all plausible actors, and so is the halved olive phenol load. The trial establishes that the blend behaved differently. It does not establish why.

The same trial also produced a caution in a separate analysis. A companion paper published in Nutrients in 2016 looked at intestinal immune markers in a ten person subsample and found that the high olive phenol oil tended to increase the proportion of IgA coated bacteria and did significantly raise plasma C-reactive protein, while the ordinary oil and the blend did neither. One trend and one significant marker in ten people is not a safety signal. It is a reminder that more olive phenolics is not automatically better, and that this trial found the same asymmetry twice, in two different systems.

Finally, the oils tested here were enriched in a laboratory to specifications no retail bottle carries, and the participants all had raised cholesterol. Three weeks is short. Whether any of this changes a clinical outcome is not a question this trial was built to answer.

Broader Context

The regulatory backdrop makes the dose inversion sharper. The European Food Safety Authority authorized a claim for olive polyphenols protecting blood lipids from oxidative damage at 5 mg of hydroxytyrosol and its derivatives per 20 g of oil. The olive enriched oil in this trial cleared that threshold comfortably. The blend, counting only its olive derived hydroxytyrosol compounds, sat just below it. The arm that lowered oxidized LDL was the one carrying less of the compound the claim is written around.

Against the rest of the literature, that is an outlier worth holding loosely rather than a refutation. Earlier work in this same research tradition showed oils high in phenolics modulating oxidative status in men, and a separate trial found olive oil phenolics reducing LDL oxidation after a meal. Twelve people over three weeks does not overturn that body of work. It complicates the assumption that the relationship is linear all the way up.

On the gut side, this trial sits with a small handful of human studies. The pomace biscuit trial found a bifidobacteria increase that trended across three separate measurement methods and reached significance in none of them, with a fibre confound that would have muddied it anyway. A 2021 review assembled the whole picture of olive oil and the gut microbiome and landed on the same altitude: the human composition data are suggestive, the mechanistic detail is mostly animal and in vitro. Where the intervention is a whole diet rather than a single oil, the signal is stronger and less specific, as in the trial that reshaped the gut microbiome and lowered inflammation markers in 612 older adults. For the plain language version of where this leaves the question, see whether olive oil is good for gut health.

Related Research

Continue exploring olive oil and polyphenol science:

• Olive Polyphenols Reached the Colon: An 8-Week Trial in 62 Adults
• Olive Oil and the Gut Microbiome: What a 2021 Review Found
• EUROLIVE: Olive Polyphenols Raise HDL
• EFSA's 2011 Olive Polyphenols Health Claim: The 5 mg Threshold Explained

Source: View the original study on PubMed. The companion paper on intestinal immune markers from the same trial is also indexed on PubMed.

Olivea's Dosage

The oils in this trial delivered between 0.30 and 8.49 mg of hydroxytyrosol derivatives per 25 mL daily dose, and the two enriched versions were prepared in a laboratory rather than pressed. Olivea works the other side of that range. Our extra virgin olive oil is bottled at its natural phenolic density rather than enriched, and each Olivea capsule delivers over 20 mg of hydroxytyrosol, with our most recent third party certificate of analysis confirming 23.5 mg per capsule.

We share this research for transparency. This is an independent study: we did not fund it, design it, or conduct it, and it did not test an Olivea product. Nothing here tested thyme enriched oil as a product, and nothing here should be read as a reason to expect a specific health outcome.

Editorial Information

Research note. This article summarizes third-party research published in a peer-reviewed journal. Olivea did not conduct or fund the study. Findings reflect the cited paper only and do not establish efficacy of Olivea products.

.