Category Archives: Allergy

Prevention of allergy and autoimmune disease

See also asthma prevention.

Immune system starts development in the first trimester, in the womb. Some parts (eg IFN gamma) are constrained, probably to protect the placenta. Immunoglobulin (particular A) still developing through early years – airway mucosal dendritic cells still expanding up until puberty.

The major driving force behind postnatal immune development in all mammalian species is microflora of the gastrointestinal tract – “healthy” development requires not just the right balance of microflora, but also at a crucial window in time.

The hygiene hypothesis (David Strachan) has now been modified to include “old friends” (Graham Rook, talking about microbes that have co-evolved with humans) and finally the biodiversity hypothesis.

So lifestyle becomes important – time spent outdoors, time spent in contact with natural habitats, having diverse, unprocessed diets. Fresh fruit and vegetables have their own microbiome, for instance. And these things are heavily linked to socioeconomic status, particularly housing and green space. Role of artificial sweeteners, emulsifiers, plasticisers and other waste organic compounds?

Mode of birth/delivery affects neonatal colonisation (elective vs emergency section or vaginal) – although so far maternal-infant microbial seeding studies have failed to demonstrate health benefits.

More than 20 studies of probiotics and prevention of allergy.  No safety concerns.  15 found supplementation during pregnancy and lactation may reduce eczema, but no effect in asthma, food allergy, rhinitis.

Fish oil (omega-3) supplementation during pregnancy may reduce risk of allergic sensitisation to egg.

Cochrane 2013 review of prebiotics concluded “promising results in high risk populations”, WAO find in favour (probiotics too), but others institutions have not!

Other dietary exposures, including maternal allergenic food avoidance, vitamin, mineral, fruit and vegetable intake did not appear to influence risk of allergic or autoimmune disease.

There is limited evidence to suggest that supporting mothers to breastfeed for longer reduces risk of eczema in the first year. Longer exclusive breastfeeding duration reduces risk of type 1 diabetes.

EAACI have recommended hypoallergenic formula (“one with documented preventive effect”) for 4/12  (Grade I, evidence A-B).  2024 Chinese systematic review found low quality evidence for extremely hydrolysed formula (EHF) reducing risk of cow’s milk allergy [relative risk (RR): 0.62]. Moderate quality evidence for partially hydrolysed (PHF) and EHF reducing risk of eczema in children (even after 2 y – RR: 0.71-0.79). Moderate quality evidence that PHF only reduced risk of wheeze at age 0-2 y compared with CMF (RR: 0.50), but cf with breast milk, both PHF and EHF increased the risk (RR: 1.61 and 64).

[PLoS Med metanalysis 2018]

Sun exposure around time of birth protects against allergy/atopy, according to French study – for one change in interquartile range, sun exposure prenatally had adjusted odds ratio of 0.47 for sensitisation at age 8-9yrs, and 0.32 for asthma. Postnatal effects less strong. Maternal vitamin D supplementation seemed to increase the risk. All the children were from Paris – sun exposure was calculated from meterological data around time of birth, not time spent outdoors. Ethnicity was not reported.  The suggestion is that the mechanism is vitamin D mediated. Previous studies have had mixed results.

Skin

Since eczema commonly precedes food allergy, it is assumed that disruption of the epithelial barrier is the first step in sensitisation (even when a baby is not eating any food) – the “dual allergen exposure” hypothesis. Peanut protein levels in household correlate with peanut sensitisation in high risk babies.

A couple of different skin “endotypes” – protein/ceramide expression, staph aureus abundance – that seem to predict food allergy. Filaggrin mutations are the strongest genetic risk factor for eczema, and the associated disruption in keratinocytes allows for allergens and irritants to penetrate the skin and trigger inflammation. Detergents/pollutants can further disrupt the skin barrier. Low humidity/temperature also affects (autumn/winter births more at risk).

S aureus colonisation is more likely with eczema, and is associated with severity and persistence – it is also associated with reduced tolerance to egg/peanut, and higher IgE milk/egg/peanut.

Cochrane review of emollients as preventive strategy found no evidence of benefit and possibly more infection. Frequent bathing and oil based bath additives increased the risk of eczema. STOP-AD study (ceramide based emollient, first 2 months of life) found reduced eczema incidence at 1yr in high risk infants. Other studies are looking at trilipid rather than petrolatum based emollients.

Japanese study of “proactive topical steroids” at 7-13 weeks of age (including non-affected skin) found 25% reduction in egg allergy but also reduced weight/length! Calcineurin inhibitors may get around the issue of steroid side effects, if safety data in infants becomes available (so far, so good). [Review, 2024]

House dust mite allergy

People assume they are allergic to dust – more likely is that they are allergic to dust mites (HDM).  These are microscopic, and tend to reside in fabrics in the home esp bedding, carpets, sofas.  So your house can be spotless but still be full of mites!

There are a number of different house dust mite species – most are Dermatophagoides (“skin eating”…), D. farinae is the American type, D. pteronyssinus is the European one.

HDM allergy is common in atopic individuals, contributing to eczema, rhinoconjunctivitis and asthma. Eczema usually comes first, but sensitisation (usually develops in young children) seems to predict development of asthma persisting into school age!

Although common, it doesn’t tend to cause immediate reactions, the way cat/dog does, probably because mites don’t fly – so exposure only caused when surfaces disturbed or if skin/nose/mouth/eyes in direct contact. But inflammation may be chronic, from brief, low quantity, frequent exposure. So even with avoidance/ reduction measures, it can take months to see the benefit…

There is cross-sensitivity with shellfish allergy but not fish allergy.

In some parts of the world, mouse and cockroach allergies more of an issue.

Lots of different allergens identified.  der p 1 and 2 are major allergens in mite faeces, in kids with eczema various other allergens found in mite bodies also important. Mite faeces are the most important cause of sensitisation. Composition of skin prick solutions seems to vary…

Avoidance/reduction

Allergy UK has an advice leaflet with different measures for reducing levels of HDM in the home.  Evidence however is that this is difficult to do and rarely seems to have any significant clinical effect!  So no point in doing piecemeal, certainly.

Dust mites need high humidity to get water, so high altitude and air conditioning suppresses them.  They also like temperatures over 20deg. They are photophobic so hide deep in fabrics.

Get rid of carpets, curtains, excess cushions/pillow/throws, move soft toys off bed.

Wash bedding at 60, or tumble dry! Soft toys go in the freezer overnight once a month (in a bag, obviously, to avoid collecting smells). Protective covers for mattresses and hypoallergenic bedding materials.

Some vacuum cleaners are better than others (in theory) eg with HEPMA filter.  Allergy UK approves products (but only if company pay, of course). But the mechanical action of cleaning (plus moving furniture, changing beds etc) can whip particles up into the air (as can emptying the contents of the cleaner) so best done when kids are out, and wait 20 minutes before letting them back in…  Cleaning and replacing filters important too esp if bagless design.

Vacuuming alone resulted in a significant reduction in carpet house dust mite allergen concentration and load. Levels approached pretreatment values by 4 weeks posttreatment in the intensive vacuuming group, whereas combination of dry steam cleaning plus vacuuming lasted up to 8 weeks. 

You can also use a steam cleaner on a mattress!

Evidence for air purifiers is mixed.  If placed on carpet, they can actually disturb more allergen than they remove! You need to close doors and windows of course. Unfortunately the best ones tend to be big, expensive and noisy, and price does not mean good quality. Plus you have to remember to replace the filters. The cheapest Which? recommended one (the Electriq EAP500HC) costs over £200.

In PAXAMA study in the NW England (kids with a history of wheeze requiring a visit to hospital), impermeable covers (blinded!) halved risk of having an emergency hospital attendance over the next year. [Murray 2017] Maximum benefit was seen in under 11yrs, mono-sensitized to mite, living in nonsmoking households, and requiring more ICS.  

[JACI 2018]

Immunotherapy

Systematic review found good evidence of benefit for eczema with immunotherapy against house dust mite, with subcut immunotherapy superior to sublingual. [Journal of Allergy & Clinical Immunology. 151(1):147-158, 2023 PMID 36191689]

Evidence for immunotherapy for allergic rhinitis, at least in adults… Treatment effect from 14 weeks, so not rapid…

No evidence for asthma but given as indication… although research into whether HDM immunotherapy might prevent.

NICE approved Acarizax for moderate to severe HDM rhinitis from age 12 in the NHS in England in January 2025. Acarizax is sublingual therapy (derived from D pteronyssinus and D farinae), indicated in adolescents (12-17 years) as well as adults diagnosed by clinical history and a positive test of house dust mite sensitisation (skin prick test and/or specific IgE) with persistent moderate to severe house dust mite allergic rhinitis despite use of symptom-relieving medication.

NICE appraisal also comments that avoidance/reduction of HDM burden in home is almost impossible and if problems persist, it can induce strong feelings of guilt in parents/carers.

Also, many patients would prefer not to be using regular, large doses of steroids, even if effective.

£80 a month on NHS, has not been submitted to Scottish Medicines Consortium so currently not recommended.

International treatment guidelines refer to a treatment period of 3 years for allergy immunotherapy to achieve disease modification. Efficacy data is available for 18 months of treatment with ACARIZAX in adults; no data is available for 3 years of treatment. If no benefit in first year, no indication to continue.

First dose should be observed and the patient monitored for at least 30 minutes.

Place under the tongue, avoid swallowing for 1 minute, avoid food/drink for 5 minutes.

Contraindications are FEV1 <70% of predicted value despite treatment, or severe asthma exacerbation within the last 3 months.

If oral ulceration, postpone treatment.

Eosinophilic oesophagitis has been reported.

Accidental Adrenaline self-injection

eg with Epipen or Emerade.

Causes vasoconstriction with potential for gangrene.

Try:

  • warm water immersion
  • local nitroglycerin paste
  • subcut infiltration with a mixture of 1.5mg of phentolamine, 1mL of 2% lidocaine (at site and along course of digital arteries)

[advised by National Poisons Information Service]

n=365 adrenaline injections to hand, 213 to digit.  No cases with clinically apparent systemic effects, only a few patients had ischemia. No patient was admitted or had surgery. [Annals of Emergency Medicine. 56(3):270-4, 2010 Sep. PMID: 20346537]

Peanut allergy

See also Prevention, pregnancy, EAT/LEAP studies and immunotherapy.

One of the most common food allergies, and a frequent cause of anaphylaxis.  Peanut is technically a legume, although there is cross-reactivity with tree nuts so often included when people talk about “nuts”.  Always consider if there are potentially other allergies to lentils, sesame, tree nuts eg cashew etc.

Also known as groundnut or monkey nut.

[Gary Stiefel, Leicester Royal, BSACI guideline peanut allergy]

Diagnosis

Wide range of potential peanut proteins.  In US study, vast majority were Ara h1/2 positive, but European more diverse. h1, h2, h3 most common for systemic reactions.  You would think testing with whole peanut would be more sensitive but component testing probably more sensitive and specific than total IgE – but not better than skin prick testing.

peanut allergy diagnosis BSACI 2017With a decent history, SPT >3mm or IgE >0.35 sufficient.

Before proceeding to a hospital challenge, footnote suggests either 2 negatives, or else both IgE and SPT negative.

Distinguish Pollen Food Syndrome – ie older, rhinitis, oral allergy symptoms with nuts/fruit. These kids will have a milder allergy.  Hazelnut mostly (Cor a1) but almond, walnut too. But can coexist with more severe allergy! Doing grass/birch pollen would support diagnosis, doing components might help assess prognosis. If history unclear, but positive IgE/Skin prick test then do components h2 and h8 (list of different cut offs for different commercial products given, with related specificity/sensitivity, just says positive/negative in flow chart). Footnote suggests adding Ara h 1, 3 and 9 as also suggest primary peanut allergy even if Ara h 2 neg.

Sibling risk 5-9%. Too low to justify routinely screening.  If family likely to just avoid forever, living in fear, then consider SPT to encourage home challenge!?

Prognosis

Up to 20% will outgrow, usually before age 8. Review may not be necessary if PFS only.  Follow up is essentially about education.  Testing can be done periodically, depending on resources.

In a study of adults coming to allergy clinics, 10% of peanut allergic turned out not to be.  Partly this would have been because the diagnosis was wrong (many had never actually reacted to peanut in the past).  Having eczema meant you were more likely to still be positive on testing.  Having asthma and being male made it half as likely you would not be allergic anymore.  But many of the cases described were not formally challenged so these results are of limited value.  [Poster at AAI 2021, Rima Rachid]

Avoidance

Difficult, as often used in biscuits, chocolate, ethnic foods eg satay.

As for any food allergy:

Peanut needs to be specified on food labels under UK/EU law.

Precautionary labels – impossible to eliminate risk. Often these “may contain” warnings and similar just say “nuts” without specifying whether the risk is from peanut or a tree nut (the company may be able to give further information if you enquire directly). 

Snack foods with precautionary labels are higher risk eg biscuits, cakes [Helen Brough].

Balance between convenience and risk (probably a very small risk, as many families ignore these warnings to some degree).  Stratify risk according to type of food, previous reactions, threshold, asthma, illness, time of day, location etc. Crossing the road metaphor – choose a safe place!

Should you avoid all nuts?  Some kids will be allergic to other (tree) nuts, but not all.  Andrew Clark reports v low rate of accidental reactions, 3%, with avoidance of all nuts. But increases quality of life to be allowed other nuts!  Risk assessment for each individual person!

Peanut oil

Probability of any reaction to refined peanut oil is remote (Blom et al, 2017). Little evidence that anyone has ever reacted to refined peanut oil. Code of practice is that presence of UNREFINED peanut oil should be declared on bottles of oil (UK and Europe).

But peanut oil, even if refined, still has to be declared on food labels.  Beware unrefined oil in ethnic foods.  Peanut oil also found in some medicines eg vitamins, antibiotic creams.

Some suspicion that peanut oil in cosmetics and pharmaceuticals might lead to sensitization and subsequent peanut allergy, even if not enough peanut protein to cause a reaction in an allergic person. So advice is avoid if you have a strong family history of allergy.

Immunotherapy

See Peanut immunotherapy.

Central serous chorioretinopathy

Central serous chorioretinopathy (CSCR) = accumulation of subretinal fluid at the posterior pole of the fundus, ultimately leading to retinal detachment. Typically affects one eye only.  Vision becomes blurry and distorted, with objects often appearing smaller in the affected eye. May also cause difficulty with bright lights and contrast sensitivity.

Mechanism unknown, but associated with use of systemic corticosteroids, pregnancy, and Cushing’s syndrome.  Recently also been described after local corticosteroids including inhaled, intranasal, topical and periocular (!). Rare though.

Although blurred vision is a symptom of CSCR, it can be a side effect of periocular steroid treatment, as well as a symptom of whatever underlying eye condition is present (if any).

MHRA therefore says you should inform patients they should report any vision problems or disturbances.

[MHRA]

Anaesthetic allergy

Reactions to local anaesthetics are often reported, but given how often they are used, most turn out not to be allergic but rather toxic (eg to parabens or sulphite preservative) or autonomic. Where allergy is confirmed, it is often of a delayed hypersensitivity type eg 24-72 hours after exposure. Beware latex allergy and C1 esterase inhibitor deficiency too. Local anaesthetics come in 2 main groups, the esters (procaine, benzocaine) and the amides (lidocaine, bupivocaine). Cross-reactivity is common among the esters but not among the amides or between the 2 groups. Neomycin sensitivity may contribute to a reaction.

Exercise induced anaphylaxis

Rare but well recognized allergic condition, about half related to a food trigger.

Anaphylaxis can occur with any degree of exercise, does not need to be extreme, but typically follows submaximal exercise pretty quickly.  Deaths are fortunately rare.

Possible mechanisms:

  • Exercise induced increase in gastric permeability
  • increased tissue transglutaminase activity in gut
  • Exercise induced blood flow redistribution
  • Mast cell heterogeneity
  • Basophil trigger by increased plasma osmolality
  • Mast cell trigger by acidosis

All sound plausible but little evidence!

Co-factors play a role in many cases eg alcohol (but unpredictable), NSAIDs (85% will have reduced threshold, and severity increases too), infection, heat/cold – some evidence too for menstruation, anti-acids, stress, sleep deprivation (as in anaphylaxis).  Co-factors also multiply risk of reaction.

Food triggers

Mostly wheat, including hidden sources eg soap, shampoo, cream.

Diagnosis

SPT useful, also IgE omega-5-gliadin.  But gold standard is provocation test (and still only 70% sensitive).  No uniform protocol – use same as for exercise induced bronchoconstriction? Do 1 hour after meal with suspected trigger.

Management

Prescribe adrenaline auto-injector.  Avoid exercise 4-6 hours after food intake.  Discuss unplanned exercise!

Hydrolysed Formulas

Alternatives and variations on cow’s milk based formulas:

  • Extensively hydrolysed – protein is broken down so good for cow’s milk protein allergy.  Not very nice tasting!  Can be whey-based eg Pepti (which includes lactose, so more palatable but no good for lactose intolerance), else casein-based eg Nutramigen (lactose free).
  • Partially hydrolysed – better tasting but symptoms may persist if true allergy
  • Anti-reflux “stay down”
  • Soya – good for cow’s milk protein allergy but cross reactivity can occur, plus theoretical phyto-oestrogen effect so avoid if under 6 months.  But the only one you can use if you are vegan or have galactosaemia.

Aptamil Pepti is made by Milupa (which is where GP’s will find it on electronic prescribing system). Not suitable for vegetarians and not Halal!

Some are lactose free, others not. Some have medium chain triglycerides as main fat source, eg Pepti Junior, Pregestimil, Peptisorb.

Nutramigen contains prebiotics – should therefore not be given to preterm babies (theoretical risk of gut translocation), and should be made up at room temperature (so not suitable for prep machines).

For those who require a vegetarian or halal diet, the only suitable extensively hydrolysed infant milk is SMA Althéra. Of the amino acid formulas, all are halal, and Neocate/Alfamino are vegetarian. None are vegan friendly.

SMA Alfamino does not have coconut oil, unlike some of the others.  No evidence that there is sufficient coconut protein in formula to cause an allergic reaction but it often gets accused of suspected reactions.

The GINI study

German study from 1998.

Some potential benefit from using hydrolyzed formula in terms of preventing allergy.  The relative risk for the cumulative incidence of any allergic disease in the intention-to-treat analysis (n = 2252) was:

  • 0.87 (95% CI, 0.77-0.99) for partially hydrolysed whey-based formula (pHF-W),
  • 0.94 (95% CI, 0.83-1.07) for extensively hydrolysed whey-based formula (eHF-W) eg Pepti, and
  • 0.83 (95% CI, 0.72-0.95) for extensively hydrolysed casein-based formula (eHF-C) eg Nutramigen compared with standard cow’s milk formula.

The corresponding figures for atopic eczema/dermatits (AD) were 0.82 (95% CI, 0.68-1.00), 0.91 (95% CI, 0.76-1.10), and 0.72 (95% CI, 0.58-0.88), respectively.

In the per-protocol analysis (ie where patients stuck to protocol) effects were stronger (0.49 for eczema at 1yr). The period prevalence of AD at 7 to 10 years was significantly reduced with eHF-C in this analysis, but there was no preventive effect on asthma or allergic rhinitis.

[J Allergy Clin Immunol. 2013 Jun;131(6):1565-73. doi: 10.1016/j.jaci.2013.01.006. ]

Cochrane review 2009 biased towards GINI data.  Since then big Melbourne study (MACS) not in favour; per protocol analysis for eczema at age 1 yr did not show any benefit (0.55-1.93).

Even with GINI, NNT could be as high as 80!

[http://onlinelibrary.wiley.com/doi/10.1111/pai.12138/full]

15 yr follow up of GINI study – between 11 and 15 years,

  • prevalence of asthma was reduced in the eHF‐C group compared to CMF (OR 0.49, 95% CI 0.26–0.89)
  • cumulative incidence of atopic rhinitis was lower in eHF‐C (risk ratio (RR) 0.77, 95% CI 0.59–0.99]) and the AR prevalence lower in pHF‐W (OR 0.67, 95% CI 0.47–0.95) and eHF‐C (OR 0.59, 95% CI 0.41–0.84).
  • cumulative incidence of eczema was reduced in pHF‐W (RR 0.75, 95% CI 0.59–0.96) and eHF‐C (RR 0.60, 95% CI 0.46–0.77), as was the eczema prevalence between 11 and 15 years in eHF‐C (OR 0.42, 95% CI0.23–0.79).
  • No significant effects were found in the eHF‐W group on any manifestation,nor was there an effect on sensitization with any formula.

[Allergy 2016; 71: 210–219. http://onlinelibrary.wiley.com/doi/10.1111/all.12790/abstract]

The EAT study

2016 study of early introduction of six common food allergens into the diet of 1303 breastfed 3 month old infants recruited from a general (not high risk) population.

Randomized.  Breast feeding needed to be maintained until at least 5 months, at least 5 weeks of at least 75% of recommended dose (ie 3g of protein per week) between 3 and 6 months.

2g protein twice weekly was recommended – 2g is roughly:

  • 2 teaspoons peanut butter or 21 Bamba pieces
  • 1 small pot yogurt
  • 1/2 small egg
  • 10g fish
  • 1tsp tahini

In an intention to treat analysis, 7.1% of the standard introduction group (at parental discretion) and 5.6% of the early introduction group developed food allergy to one or more of the six intervention foods (peanuts, egg, cow’s milk, sesame, white fish and wheat) up to 3 years of age (p=0.32, ie no difference).

However, when the analysis was adjusted for adherence to early introduction, there was a statistically significant 67% lower rate of food allergy in the early introduction group (2.4% vs 6.4%; p=0.03), with no cases of peanut allergy (rate was 2.5% in control group) and 75% less egg allergy (1.4% vs 5.5%).  Rate of skin prick test positivity significantly lower for peanut, egg, milk, sesame.

Cooked egg works! Increasing dose, increasing effect.  Modelling suggests 2g protein weekly effective.

Safe!

However, poor adherence to the study protocol (only 32% managed to follow early introduction fully) highlights the challenges around introducing solids.

[Michael Perkins, DOI: 10.1056/NEJMoa1514210]