Celery Allergy and Photosensitivity

Table of Contents

1. Two Different Problems, Often Confused
2. Furanocoumarins: The Plant’s Chemical Defence
3. Phytophotodermatitis: What It Looks Like
4. Pink Rot and the Grocery-Worker Outbreaks
5. Who Is Actually at Risk for the Skin Reaction
6. Celery Allergy: Serious and Under-Recognised
7. Api g 1, Api g 2 and Why Cooking Does Not Reliably Help
8. Celery–Mugwort–Spice Syndrome
9. Why Europe Labels Celery and North America Does Not
10. Where Celery Hides
11. What to Do
12. Key Research Papers
13. Connections
14. Featured Videos

Two Different Problems, Often Confused

People frequently lump these together as “celery sensitivity”, and separating them matters because the mechanisms, the people affected and the responses are completely different.

• Phytophotodermatitis is a chemical burn, not an allergy. It happens when furanocoumarins from the plant get onto skin and that skin is then exposed to ultraviolet light. It has nothing to do with the immune system. Anyone will get it given enough plant chemical and enough sun — there is no such thing as being immune to it, and there is no such thing as being allergic to it in the immunological sense. It is dose-dependent, it appears a day or two after exposure, and it affects the skin that was in contact, not the whole body.
• Celery allergy is a true IgE-mediated food allergy. It affects a minority of people whose immune systems have made antibodies against specific celery proteins. It can be triggered by a very small amount, it appears within minutes, and it can be systemic and life-threatening.

A useful shorthand: the skin reaction is about how much celery touched you and how much sun followed; the allergy is about who you are.

Back to Table of Contents

Furanocoumarins: The Plant’s Chemical Defence

Furanocoumarins — also called psoralens — are the carrot family’s chemical weapon. Celery, parsnip, wild parsnip, giant hogweed, angelica, figs, limes, bergamot and rue all make them, and they exist to poison insects and fungi that attack the plant. The main ones in celery are psoralen, bergapten (5-methoxypsoralen) and xanthotoxin (8-methoxypsoralen).

What makes them unusual is that they are not toxic on their own. They are phototoxic: they need ultraviolet light to do damage. A psoralen molecule slips between the base pairs of DNA, and when it absorbs UVA energy it forms covalent bonds to the DNA strands, cross-linking them. The cell cannot replicate through a cross-link. Enough cross-linked cells die, the tissue reacts, and the result looks and feels like a burn — because in every meaningful sense it is one.

Medicine uses exactly this reaction on purpose. PUVA therapy — psoralen plus UVA — has treated psoriasis and vitiligo for decades by giving a psoralen drug and then controlled UVA exposure. The compound the dermatologist prescribes is chemically the same family the celery is making in the field.

Three things raise celery’s furanocoumarin content, and all three matter practically:

• Fungal infection. By far the biggest factor. Furanocoumarins are phytoalexins — defence compounds the plant ramps up when attacked. Celery infected with pink rot (Sclerotinia sclerotiorum) or grey mould (Botrytis cinerea) can carry many times the normal amount.
• Storage and age. Levels rise in stored celery even without visible spoilage — documented by Chaudhary and colleagues as far back as 1985 and confirmed since.
• Plant stress. Insect damage and other stressors push the same defensive chemistry up.

Which gives the single most useful practical rule on this page: fresh, sound, crisp celery is low-risk; old, soft, blemished or mouldy celery is the risky kind. The same rule that makes celery taste better also makes it safer to handle.

Back to Table of Contents

Phytophotodermatitis: What It Looks Like

The reaction is distinctive once you know it, and frequently misdiagnosed when you do not.

1. Contact. Plant juice gets on the skin — typically hands, forearms, wrists, the front of the thighs where a box was carried.
2. Sun. That skin is exposed to UVA. Ordinary daylight is enough; cloud cover does not prevent it, and UVA passes through window glass.
3. Delay. Nothing happens for roughly a day. This is the reason it gets misdiagnosed — by the time the skin reacts, nobody connects it to yesterday’s vegetables.
4. Burn. Redness, stinging and often blistering appear about 24 to 48 hours after exposure and peak around 48 to 72 hours. It can be severe enough to need burn care.
5. Pigmentation. As the burn settles, it leaves brown hyperpigmentation in exactly the shape of the contact — drips, streaks, handprints, the outline of where a stalk was held. This can last weeks to months and occasionally more than a year. It is harmless but conspicuous.

The bizarre shapes are the diagnostic clue: streaks, splashes and fingerprints do not look like any rash, because they are a map of where liquid ran. Emergency departments sometimes mistake the pattern for a burn injury or for signs of harm, which is a good reason for the diagnosis to be better known.

Treatment is straightforward: wash the skin with soap and water as soon as contact is noticed and before sun exposure, keep the area covered or out of the sun, and treat the burn as a burn — cool compresses, emollients, topical steroid for inflammation, and medical care for extensive blistering. The pigmentation needs nothing except sun protection and patience.

Back to Table of Contents

Pink Rot and the Grocery-Worker Outbreaks

The definitive documentation of this problem came from American supermarkets in the mid-1980s, and it is worth recounting because it shows the mechanism in full.

Grocery workers — produce clerks handling celery all day — began presenting with an outbreak of blistering, burning dermatitis on the hands, wrists and forearms. Seth Berkley and colleagues investigated and published the results in the Annals of Internal Medicine in 1986, and Paul Seligman and colleagues reported a parallel investigation in the Archives of Dermatology the following year. The findings lined up precisely: the affected workers had handled celery from a crop with pink rot infection, and the infected celery was carrying far higher furanocoumarin concentrations than sound celery — enough to produce clinical phototoxicity from ordinary occupational contact plus ordinary daylight.

This is a satisfying piece of investigative work because it closes the loop. The plant pathology (fungal infection), the plant chemistry (phytoalexin induction), the physics (UVA activation), the biochemistry (DNA cross-linking) and the clinical picture (delayed blistering dermatitis with residual pigmentation) all fit together and were confirmed by measurement rather than inference. Later work by Aharoni and colleagues confirmed the same effect for Botrytis infection and for storage age.

It also explains why this is largely an occupational story. The people affected were handling crates of celery for hours a day. Somebody chopping two stalks for a soup is in a completely different exposure category.

Back to Table of Contents

Who Is Actually at Risk for the Skin Reaction

Genuinely at risk:

• Celery harvesters and field workers — heavy skin contact, outdoors, all day, in strong sun. The highest-exposure group by a wide margin.
• Produce and grocery workers handling large volumes, particularly from a spoiled or infected batch.
• Food-service workers prepping celery in quantity, especially if they then go outside.
• Home gardeners and foragers harvesting celery, parsnip or wild parsnip on a sunny day. Wild parsnip and giant hogweed are considerably worse offenders than cultivated celery.
• Anyone juicing large amounts whose hands are wet with juice and who then goes outside.

Not meaningfully at risk: people who eat celery. Dietary furanocoumarin intake from normal amounts of celery is well below the level that causes clinical photosensitivity. Case reports of severe systemic photosensitivity after eating celery exist, but they involve very large quantities — large amounts of celery root or juice in one sitting — usually followed by deliberate strong UV exposure such as a tanning bed. That is a real published phenomenon and it is also a description of an unusual set of circumstances, not of a normal diet.

One genuine interaction is worth flagging: if you already take a photosensitising medication — some antibiotics, some diuretics, retinoids, certain antifungals and others — or you are having phototherapy, the effects stack. Heavy celery handling plus a photosensitising drug plus sun is the combination that turns a theoretical risk into a real one.

The practical version: if you handle a lot of celery, wear gloves or wash your hands and forearms afterwards, do it before you go out into the sun, and throw away celery that is soft, brown, mouldy or showing pink-tinged rot rather than trimming around it.

Back to Table of Contents

Celery Allergy: Serious and Under-Recognised

Now the immunological half, which is the more dangerous of the two.

Celery allergy is a true IgE-mediated food allergy. Its presentations run the full range:

• Oral allergy syndrome — itching, tingling or mild swelling of the lips, mouth and throat within minutes of eating raw celery. The commonest and mildest form.
• Skin reactions — hives, swelling, or contact urticaria in people who handle it.
• Gastrointestinal and respiratory symptoms — vomiting, abdominal pain, wheezing, and occupational asthma or rhinitis in food workers exposed to celery dust or celery spice.
• Anaphylaxis — a systemic reaction with airway compromise, circulatory collapse or both. Celery is among the plant foods recorded as triggering severe reactions in European anaphylaxis series, particularly in adults.

The evidence base here is unusually solid for a food allergy, because a Swiss and German research group put celery through the gold-standard test. Barbara Ballmer-Weber, Stefan Vieths and colleagues published a study in 2000 in which 32 people with a history of adverse reactions to celery root underwent double-blind, placebo-controlled food challenge — the definitive method, in which neither patient nor clinician knows whether a given dose contains the food. A companion paper by Dirk Lüttkopf and colleagues characterised the allergens in the confirmed responders. Celery allergy is not a self-reported phenomenon; it is a challenge-confirmed one.

A specific and important sub-case is celeriac. Celery root has been the form implicated in a large share of the severe European cases, and component-resolved diagnostic work by Kerstin Bauermeister and colleagues was developed specifically for celeriac-allergic patients. Somebody who tolerates celery stalks should not assume celeriac is equivalent.

Back to Table of Contents

Api g 1, Api g 2 and Why Cooking Does Not Reliably Help

This is the part that matters most, because it overturns advice people are commonly given.

Celery’s allergens are named after the genus: Api g 1 is the major one, a member of the PR-10 protein family and a structural relative of Bet v 1, the major birch pollen allergen. Its crystal structure was solved in 2005 and the cross-reactivity with birch pollen is a direct consequence of that shape. Api g 2 is a non-specific lipid transfer protein found in the stalk, characterised by Gabriele Gadermaier and colleagues and associated with more severe reactions. Api g 4 is a profilin, Api g 5 a glycoprotein that cross-reacts with mugwort, Api g 6 another lipid transfer protein in celeriac, and Api g 7 a more recently characterised defensin-like allergen that has improved diagnosis in celeriac-allergic patients.

Here is the rule most people have been taught: Bet v 1-like allergens are heat-labile. That is why someone with birch-pollen-related apple allergy can usually eat apple pie without trouble — cooking unfolds the protein and the IgE no longer recognises it. It is standard, correct, useful advice for a lot of foods.

Celery is the well-documented exception. Ballmer-Weber and colleagues tested this directly in 2002, running blinded challenges with cooked celery and with celery spice in patients with confirmed celery allergy. Cooking reduced reactivity but did not abolish it: a substantial proportion of the patients still reacted to cooked celery, and reacted to the spice as well. The practical conclusion the authors drew, and the one that matters for anyone with this allergy, is that celery-allergic patients cannot rely on cooking to make celery safe.

Two additional factors make it worse. The lipid transfer proteins, Api g 2 and Api g 6, are inherently more heat- and digestion-stable than the PR-10 allergens — that is what lipid transfer proteins are like — and they are the ones associated with the more serious reactions. And celery is often concentrated in the foods it appears in: stock cubes, celery salt and spice blends deliver far more allergen per mouthful than a stick of celery does.

Back to Table of Contents

Celery–Mugwort–Spice Syndrome

Celery allergy usually does not arrive alone. It belongs to a recognised cross-reactivity cluster, described in the literature as celery–mugwort–birch–spice syndrome, and understanding it explains a great deal about who develops the allergy and what else they need to watch.

The pattern was documented by Brunello Wüthrich and colleagues in 1990, who described celery allergy occurring together with birch and mugwort pollen sensitisation. Later immunoblot work by Bauer, Ebner, Hirschwehr and colleagues showed that the IgE cross-reactivity between birch pollen, mugwort pollen and celery is carried by at least three distinct cross-reacting allergens rather than one — which is why the syndrome is so consistent.

The spice component follows from the same botany. Mugwort is an Asteraceae; celery is an Apiaceae; and the carrot family is full of spices — carrot, parsley, coriander, cumin, caraway, fennel, dill, aniseed. Jürgen Stäger, Wüthrich and Johansson documented spice allergy in celery-sensitive patients in 1991, and case reports since have extended the pattern to star anise and other related spices. Somebody with celery allergy may therefore react to a spice blend without any celery in it.

Two practical consequences:

• Autumn hay fever is the clue. Mugwort pollinates in late summer and autumn. Someone whose hay fever peaks then, and who gets an itchy mouth from raw celery or carrot, fits the syndrome. See Allergic Rhinitis.
• Exercise and other cofactors can convert a mild reaction into a severe one. Food-dependent exercise-induced anaphylaxis has been reported specifically within this syndrome — a food tolerated at rest producing anaphylaxis when eaten before exercise. Alcohol, non-steroidal anti-inflammatory drugs and acute infection act as similar cofactors. This is one reason a previously mild celery reaction should not be assumed to define the ceiling.

Back to Table of Contents

Why Europe Labels Celery and North America Does Not

Here is a fact that catches travellers out. Celery is one of the allergens that must be declared on food labels throughout the European Union, under Regulation (EU) No 1169/2011, which names fourteen allergens including “celery and products thereof”. The United Kingdom retains the same fourteen. In the United States, the Food Allergen Labeling and Consumer Protection Act names nine major allergens — milk, egg, fish, crustacean shellfish, tree nuts, peanuts, wheat, soybeans and, since 2023, sesame. Celery is not among them. Canada’s priority allergen list also omits it.

The difference is not a regulatory oversight. It reflects a real epidemiological difference: celery allergy is genuinely more common and more clinically significant in central and western Europe than in North America, most likely because mugwort and birch pollen sensitisation — the entry route into the syndrome — is more prevalent there, and because celeriac is a far more commonly eaten vegetable in German, Swiss, Dutch, Polish and French cooking than it is in American.

The practical consequence runs in both directions and is worth knowing:

• A celery-allergic person travelling from Europe to North America loses their labelling protection. Celery can be present in an American packaged food, restaurant stock or spice blend without appearing as a highlighted allergen. Ask explicitly rather than relying on the label.
• Someone who reacts to food in Europe and cannot work out why may find celery declared in bold on a European label where the equivalent product at home said nothing.

Back to Table of Contents

Where Celery Hides

Celery is a background flavour ingredient in an enormous amount of savoury food, which is exactly what makes the allergy difficult to manage. Places it commonly appears without being obvious:

• Stock cubes, bouillon powder and liquid stock — the single most common hidden source. Vegetable stock is largely onion, carrot and celery by design.
• Soups, gravies, sauces and ready meals built on that stock.
• Celery salt and spice blends — celery salt is ground celery seed and salt, and it is standard in seasoning mixes, seafood boils and barbecue rubs.
• Bloody Mary mix and savoury tomato drinks, which very often contain celery salt or celery juice.
• Sausages, patés, stuffings and prepared meats. And note the additional route: celery powder is used as a curing agent in “uncured” bacon, hot dogs and deli meat, so celery protein can be present in a product nobody would think of as containing vegetables — see the Blood Pressure and Dietary Nitrate page for how that works.
• Coleslaw, potato salad, tuna and chicken salads, and dressings.
• “Natural flavouring”, “vegetable powder” and “spices” on an ingredients list — in jurisdictions that do not require celery to be declared, these can conceal it.
• Herbal supplements, particularly celery seed extract sold for blood pressure or gout, which is a concentrated allergen source.

Back to Table of Contents

What to Do

If you get an itchy mouth from raw celery: that is most likely oral allergy syndrome, usually in the context of birch or mugwort pollen sensitisation. It is worth getting properly assessed rather than self-managing, because celery is one of the pollen-related foods that can escalate, and because the assessment also tells you what else in the carrot family and the spice cupboard to watch. Do not assume cooking makes it safe; for celery specifically, that assumption does not hold.

If you have had a reaction beyond the mouth: hives, swelling, wheezing, vomiting or faintness after celery warrants a referral to an allergy specialist. Modern component-resolved testing can distinguish between the PR-10 pattern, which usually means milder pollen-related reactivity, and lipid transfer protein sensitisation, which is associated with the more severe reactions — and that distinction genuinely changes the advice you are given. If you are prescribed adrenaline auto-injectors, carry them, including when travelling somewhere celery is not a labelled allergen.

If you handle celery in quantity at work: gloves, or wash hands and forearms after handling and before going into daylight. Discard rotting celery rather than trimming it, and treat a batch with visible pink or brown rot as a hazard rather than as a quality problem. If you develop a delayed blistering rash in odd streaky shapes a day or two after a shift, mention celery to whoever sees it — the diagnosis is easy once someone thinks of it and baffling until they do. See Contact Dermatitis.

If you have no celery problem at all: which is almost everyone. Buy it fresh, keep it crisp, use it up, and enjoy it. None of this page is a reason to avoid a good vegetable — it is a reason to buy sound celery and to take a reported celery allergy seriously when someone tells you they have one.

Back to Table of Contents

Key Research Papers

Author names, titles and journals are plain text; only the DOI is a link, and each opens in a new tab. The 1986 and 1987 outbreak investigations are old but remain the primary documentation of the celery phototoxicity problem and have not been superseded.

1. Berkley SF, Hightower AW, Beier RC, et al. Dermatitis in grocery workers associated with high natural concentrations of furanocoumarins in celery. Annals of Internal Medicine. 1986;105(3):351-355. — doi:10.7326/0003-4819-105-3-351 — the outbreak investigation that linked pink-rot infection, furanocoumarin concentration and occupational dermatitis.
2. Seligman PJ, Mathias CGT, O’Malley MA, et al. Phytophotodermatitis from celery among grocery store workers. Archives of Dermatology. 1987;123(11):1478-1482. — doi:10.1001/archderm.1987.01660350078017
3. Chaudhary SK, Ceska O, Warrington PJ, Ashwood-Smith MJ. Increased furocoumarin content of celery during storage. Journal of Agricultural and Food Chemistry. 1985;33(6):1153-1157. — doi:10.1021/jf00066a032
4. Aharoni N, Afek U, Finkelstein E, et al. Impact of celery age and infection by Botrytis cinerea on linear furanocoumarin (psoralens) content in stored celery. Phytoparasitica. 1996;24(3):195-197. — doi:10.1007/bf02981418
5. Dercks W, Trumble J, Winter C. Impact of atmospheric pollution on linear furanocoumarin content in celery. Journal of Chemical Ecology. 1990;16(2):443-454. — doi:10.1007/bf01021776
6. Knudsen EA, Kroon S. In vitro and in vivo phototoxicity of furocoumarin-containing plants. Clinical and Experimental Dermatology. 1988;13(2):92-96. — doi:10.1111/j.1365-2230.1988.tb00666.x
7. Ballmer-Weber BK, Vieths S, Lüttkopf D, et al. Celery allergy confirmed by double-blind, placebo-controlled food challenge: a clinical study in 32 subjects with a history of adverse reactions to celery root. Journal of Allergy and Clinical Immunology. 2000;106(2):373-378. — doi:10.1067/mai.2000.107196
8. Lüttkopf D, Ballmer-Weber BK, Wüthrich B, Vieths S. Celery allergens in patients with positive double-blind placebo-controlled food challenge. Journal of Allergy and Clinical Immunology. 2000;106(2):390-399. — doi:10.1067/mai.2000.108711
9. Ballmer-Weber BK, Hoffmann A, Wüthrich B, et al. Influence of food processing on the allergenicity of celery: DBPCFC with celery spice and cooked celery in patients with celery allergy. Allergy. 2002;57(3):228-235. — doi:10.1034/j.1398-9995.2002.1o3319.x — the study behind the “cooking does not reliably help” conclusion.
10. Schirmer T, Hoffimann-Sommergruber K, Susani M, et al. Crystal structure of the major celery allergen Api g 1: molecular analysis of cross-reactivity. Journal of Molecular Biology. 2005;351(5):1101-1109. — doi:10.1016/j.jmb.2005.06.054
11. Gadermaier G, Hauser M, Egger M, et al. Sensitization prevalence, antibody cross-reactivity and immunogenic peptide profile of Api g 2, the non-specific lipid transfer protein 1 of celery. PLoS ONE. 2011;6(8):e24150. — doi:10.1371/journal.pone.0024150
12. Bauermeister K, Ballmer-Weber BK, Bublin M, et al. Assessment of component-resolved in vitro diagnosis of celeriac allergy. Journal of Allergy and Clinical Immunology. 2009;124(6):1273-1281. — doi:10.1016/j.jaci.2009.07.033
13. Kosztulska B, Bartuzi Z, Ukleja-Sokołowska N. Current state of celery allergy: is discovering Api g 7 a milestone in diagnosing celeriac-allergic patients? International Journal of Molecular Sciences. 2025;26(12):5840. — doi:10.3390/ijms26125840
14. Wüthrich B, Stäger J, Johansson SGO. Celery allergy associated with birch and mugwort pollinosis. Allergy. 1990;45(8):566-571. — doi:10.1111/j.1398-9995.1990.tb00941.x
15. Bauer L, Ebner C, Hirschwehr R, et al. IgE cross-reactivity between birch pollen, mugwort pollen and celery is due to at least three distinct cross-reacting allergens: immunoblot investigation of the birch–mugwort–celery syndrome. Clinical & Experimental Allergy. 1996;26(10):1161-1170. — doi:10.1111/j.1365-2222.1996.tb00503.x
16. Stäger J, Wüthrich B, Johansson SGO. Spice allergy in celery-sensitive patients. Allergy. 1991;46(6):475-478. — doi:10.1111/j.1398-9995.1991.tb04228.x
17. Baek CH, Bae YJ, Cho YS, et al. Food-dependent exercise-induced anaphylaxis in the celery–mugwort–birch–spice syndrome. Allergy. 2010;65(6):792-793. — doi:10.1111/j.1398-9995.2009.02233.x
18. Katelaris CH. Food allergy and oral allergy or pollen-food syndrome. Current Opinion in Allergy & Clinical Immunology. 2010;10(3):246-251. — doi:10.1097/aci.0b013e32833973fb

External Resources

• EUR-Lex — Regulation (EU) No 1169/2011, Annex II: the fourteen allergens that must be declared, including celery
• US FDA — Food Allergen Labeling and Consumer Protection Act: the nine US major allergens, which do not include celery
• PubMed: celery anaphylaxis and Apium graveolens allergy
• PubMed: phytophotodermatitis, celery and occupational exposure
• PubMed: Api g 1, Api g 2 and component-resolved celery diagnosis

Back to Table of Contents

Connections

• All Food
• Celery
• Celery Benefits Deep Dive
• Celery: History and Origins
• Blood Pressure and Dietary Nitrate
• The Celery Juice Claims
• Anaphylaxis
• Food Allergy
• Allergies
• Allergic Rhinitis
• Contact Dermatitis
• Mugwort — the pollen at the centre of the cross-reactivity cluster
• Parsley
• Fennel
• Carrots
• Celery Juice

Back to Table of Contents

Featured Videos