Research Desk · Immunology

Your Histamine Test Probably Didn’t Tell You Anything

Stephen Duncan FDN-P BSc Hons MSc  ·  9 min read

You paid for a blood test. It came back low, and someone told you that explained the flushing, the headaches and the reaction to red wine. Then you were given a list of foods to avoid, and the list was long.

I want to take that apart, because most of it doesn’t survive contact with the evidence. Not the symptoms — those are real, and I’ll come back to them. The test and the list.

The test

The blood test almost everyone gets is serum or plasma diamine oxidase. DAO is the enzyme that breaks down the histamine you eat, and it works mainly in the lining of the small intestine. The logic is clean: low enzyme, poor clearance, symptoms.

The largest study to evaluate it properly was published in 2023. It measured DAO in 249 patients with suspected histamine intolerance against 50 healthy controls. The group with the highest clinical probability did have lower DAO — a median of 8 U/mL against 18 in controls. So there is a real difference at group level.

Now look at what happens at the level of the individual person, which is the only level that matters when you are the person. At the strict cut-off, the test’s sensitivity was 2 per cent. At the looser cut-off, 71 per cent. In plain terms: a very low result means something, and a normal result means almost nothing, because the test misses most of the people it is meant to find.

The authors’ own conclusion is that diagnosis “should not solely rely on DAO measurements”.

They are not alone. A prospective multicentre study of 207 adults found no correlation between serum DAO and clinical status in any group, and said so in its title: not a useful marker. Some smaller studies have found the opposite. The literature genuinely conflicts, and where it conflicts I weight the largest and most rigorous work.

There is also a mechanical problem underneath the statistical one. Circulating DAO is used in gastroenterology as a marker of how intact your intestinal lining is — it leaks into the blood when that lining is damaged. It is not a direct readout of how much histamine-degrading enzyme is present at the gut surface doing the actual work. The thing being measured is adjacent to the thing you want to know.

And there is no agreed cut-off. Different laboratories run different assays reporting in different units, with thresholds set by manufacturers rather than by consensus.

The finding that should give everyone pause

In 2023, 59 people with suspected histamine intolerance were given a blinded, placebo-controlled histamine challenge.

62.7 per cent of them had symptoms in response to the placebo.

Nearly two-thirds reacted to nothing.

Read that carefully, because it does not mean the symptoms were imagined. Placebo responses are real physiological events. What it means is that if you eat something, feel unwell, and conclude the food caused it, you will be wrong most of the time — because you would have felt unwell anyway.

The attribution is the unreliable step. Not the sensation.

This is also why a food diary feels so convincing and proves so little. It records what you ate and how you felt. It cannot separate the two.

The food list

There are several in circulation and they do not agree with each other.

A group at the University of Barcelona did the obvious thing and checked. They compared the foods excluded by low-histamine diets against the measured histamine content of those foods. Only around 32 per cent of the exclusions could be explained by the food actually containing much histamine. Fermented foods were the only category every list agreed on.

A separate review found three different classifications — high, moderate and low — for the same foods across different published tables. Strawberries. Blue cheese. Cooked ham. Yoghurt. Green tea.

There is a reason for this, and it isn’t that the list-writers are careless. Histamine is not a fixed property of a food. It is produced by bacteria acting on the amino acid histidine, and it accumulates over time — faster when warm.

One controlled study measured exactly this in tuna. Held at 4°C, histamine rose from 12.8 to 68.2 mg/kg over six days. The same fish held at 12°C reached 2,721 mg/kg. A forty-fold difference from storage temperature alone.

So “is tuna high in histamine?” has no answer. Fresh tuna handled cold is low. The same fish left warm is very high. The question a food list is trying to answer cannot be answered by a list.

Which means the single most useful rule in this whole area isn’t a food list at all. It is: cook it fresh, cool it fast, don’t keep it. That has a real mechanism behind it. Most individual food classifications do not.

Does the diet work?

Sometimes. Less often than the enthusiasm suggests, and the best study on it is uncomfortable reading.

157 people with moderate to severe chronic urticaria followed a low-histamine diet for three weeks, then underwent a double-blind placebo-controlled histamine challenge. 46 per cent improved on the diet. Only 17 per cent reacted on blinded challenge.

Most of the improvement was not about histamine.

A systematic review of diet in chronic spontaneous urticaria arrived at the same place from a different direction: complete remission in about 12 per cent, partial in 44 per cent, and a conclusion that the evidence supports benefit “only in individual patients” because controlled trials are lacking.

The German allergy societies — the only bodies to have produced a formal guideline on ingested histamine — state that the evidence for a causal link between dietary histamine and reproducible symptoms is “limited and contradictory”, that there is no reliable laboratory test, and that randomised trials are absent. There is no UK equivalent, which is part of why your GP has no framework to offer you.

What that guideline actually recommends

Not a permanent diet. This is the part that gets lost.

1. A short low-histamine phase — around two weeks. Long enough to see whether anything changes. Not a way of life.
2. Systematic reintroduction — one food at a time, to find out what you personally react to and at what dose.
3. A personalised long-term diet — the widest one you tolerate.

The guideline is explicit that the whole point of the structure is to prevent people ending up on unnecessarily restrictive diets. That is its stated purpose.

The failure mode I see constantly is phase one, indefinitely. Someone removes thirty foods, feels somewhat better, and never reintroduces anything. They now avoid thirty foods, of which perhaps two were ever a problem, and they have no way of knowing which two. Every food removed and never formally rechallenged sits in permanent limbo.

That is not a cautious approach. It is an unfinished one, and the cost is real — narrowed nutrition, lost fibre and polyphenol diversity, a shrinking social world, and for some people the beginnings of a genuinely difficult relationship with eating.

Why the diet gets sold and the reintroduction doesn’t

I want to be careful here, because I don’t think this is about anyone acting in bad faith.

Look at where the commercial gradient runs. A downloadable low-histamine plan is a product. It scales, it’s the same for everybody, and it can be sold to someone you’ve never met. Reintroduction is none of those things — it is slow, individual, unglamorous, and it ends with the client needing you less.

So the two-week phase that the guideline treats as a diagnostic step becomes the thing that gets packaged, and the reintroduction phase that produces the actual answer quietly doesn’t. Nobody has to intend that for it to happen. Incentives shape what gets built.

If you have bought a low-histamine plan and it contained no reintroduction protocol, you bought half a process.

Three more things you may have been told

DAO supplements. A handful of small trials. In chronic urticaria, 22 people, with benefit confined to the subgroup who already had low DAO. In migraine, 100 people with diagnosed DAO deficiency, showing reduced hours of headache. Several come from research groups with commercial links to DAO products. Not nothing — but thin, small in effect, and inconsistently replicated.

“B6 and vitamin C are DAO cofactors.” This one is simply wrong, and it is everywhere. Human DAO is a copper enzyme — copper plus a built-in cofactor called topaquinone, formed from one of its own amino acids. Vitamin B6 is the cofactor for histidine decarboxylase, which is the enzyme that produces histamine. The claim has the biochemistry inverted. And no trial has tested whether B6, vitamin C or copper supplementation improves histamine intolerance symptoms at all.

Probiotic strain lists. The lists sorting strains into histamine-producing and histamine-degrading come from what bacteria do in a dish, and from searching their genomes for the relevant gene. Carrying that gene is strain-specific rather than species-wide, so “avoid this species” is too blunt to be right. And no human trial has shown that choosing probiotics by histamine profile changes anyone’s symptoms.

There is a wrinkle worth knowing, because it shows how loose the reasoning gets. Certain Lactobacillus reuteri strains do produce histamine — and that histamine acts on a receptor that suppresses inflammation. Producing histamine is not automatically a bad thing to do.

So what is going on?

Something. The symptoms are consistent enough across enough people that dismissing them is as unserious as over-diagnosing them.

There is real mechanism here. Corticotropin-releasing hormone, released under stress, acts directly on mast cells and triggers them to release histamine with no allergy involved at all. That is a well-established pathway, and it means stress-triggered flushing or gut symptoms are physiological rather than psychological. Not in your head. In your mast cells.

Gut bacteria produce histamine too. The best work on this colonised germ-free mice with stool from people with IBS and produced visceral pain — elegant, mechanistically convincing, and done in mice. The human end of it is association, not proof.

What I take from all of this is not that histamine intolerance is fictional. It is that we have a plausible mechanism, an unreliable test, contradictory food lists and a treatment that mostly hasn’t been trialled properly — and a marketplace selling all four with a confidence none of them has earned.

What I’d actually do

Get the testable things tested first. Coeliac serology while still eating gluten. Thyroid function. Ferritin. A tryptase level if there is flushing or anything resembling anaphylaxis — that test is validated, unlike DAO.

Don’t buy the DAO test. If you already have the result, it does not mean what you were told, in either direction.

Try the diet if you want to — for two weeks, not forever. Start with the exclusions that have real mechanism behind them: fermented and aged foods, alcohol, cured and smoked meats, leftovers. Add the freshness rule, which is doing more work than most of the list.

Then reintroduce, one food at a time. This is the part that produces the answer. Skipping it means you learn nothing and lose thirty foods.

Do one thing at a time. Low-histamine on top of low-FODMAP on top of low-oxalate on top of gluten-free is not a careful diet. It is a diet nobody can meet their nutritional needs on, and it makes every finding uninterpretable.

Notice if it is becoming something else. Weight loss, growing fear of food, an ever-shrinking list of safe things. That is a signal to stop and get help, not to restrict further.

Every numerical claim on this page was checked against primary literature before publication. Where the evidence conflicts, I have said so. Where a widely repeated claim has no traceable source — the accuracy figures that circulate for DAO testing, the vitamin C cofactor claim — I have said that too.

Stephen Duncan is a Functional Diagnostic Nutrition Practitioner (FDN-P, BSc Hons, MSc) based in Edinburgh with thirty-seven years in clinical practice.

Educational content, not medical advice. If you have experienced any immediate reaction to food involving swelling or breathing difficulty, seek medical assessment — that requires an allergist, not a nutrition practitioner.