Guideline review: BSACI guideline for the set-up of penicillin allergy delabelling services by non-allergists working in a hospital setting

Background

The prevalence of penicillin allergy labels in the general population is 5.9%, with 95% of these labels incorrect on testing.1 2 True allergic reactions to penicillin are less common in children than in adults, with less than 7% of children reacting on re-exposure.3 Erroneous labels of ‘penicillin allergy’ are associated with increased antibiotic resistance, higher healthcare costs and poorer clinical outcomes.4

Despite this, penicillin allergy testing is limited in the National Health Service, conducted only by specialist allergists, leaving many patients unable to access testing. The authors feel that widespread delabelling is therefore only possible through the engagement of clinicians not specially trained in allergy and immunology, referred to as ‘non-allergists’.

Information about the current guideline

The guideline was published in June 2022 by the Standards of Care Committee of the British Society for Allergy and Clinical Immunology (BSACI). It provides a framework for non-allergists working in secondary care to set up a penicillin delabelling service for patients. It is a joint guideline covering adults and paediatrics, with separate sections for both patient groups.

Allergy testing is currently performed in specialist clinics by trained paediatric allergists and immunologists in accordance with previous National Institute for Health and Care Excellence (NICE) guidance (NICE CG183), limiting this service to select patient groups. This guideline gives recommendations for identifying patients at low risk of allergy and a framework for the conduct of drug provocation testing (DPT) by non-allergists. It should be used in conjunction with the BSACI 2015 guideline ‘Management of allergy to penicillins and other beta-lactams’ (box 1).

Recommendations for children of all ages are the same as those for adults in terms of risk stratification and the conduct of DPT (see boxes 2–4). Doses of medication should be age and weight adjusted for the paediatric population.

Box 2

Low-risk symptoms in adults and children5

Minor gastrointestinal symptoms (nausea, abdominal pain, diarrhoea).^a
Candidiasis (thrush).^a
Minor symptoms unrelated to any form of allergic reaction, for example, headache, arthralgia, strange taste in mouth.^a
Family history of penicillin allergy but without personal history of allergy.^a
Patient has taken and tolerated the same penicillin subsequent to the index reaction.^a
Patient reports ‘benign’^b rash which developed more than 1 hour after the first dose of a course of penicillin.
Patient reports a childhood rash with no other history available.^c
Patient cannot remember what happened during index reaction but was told it was not serious and did not require hospital treatment.

^aThese patients do not require allergy testing in the form of either skin tests or drug provocation testing (DPT). However, some patients may continue avoiding penicillin if they do not have the reassurance of a negative allergy test. In these circumstances, a DPT should be considered.
^bBenign defined as non-urticarial, non-itchy, non-blistering, short lived (less than 24 hours) and did not require treatment.
^c‘No other history available’ assumes that such information has been sought from patients, carers, relatives and healthcare records where possible.

Box 3

Exclusion criteria for drug provocation testing (DPT) in adults and children5

Rash occurring within 1 hour of the first dose of penicillin.
Rash lasting more than 24 hours and/or affecting more than 10% of body surface.
Rash associated with blisters, skin peeling, mucosal inflammation (eyes, mouth, genitals) and purpura.
Patients reporting any symptoms suggestive of a type 1 immediate hypersensitivity reaction to penicillin, including swelling, urticaria, angioedema, shortness of breath, wheeze, loss of consciousness or collapse.
Patients who required hospital treatment due to their reaction.
Patients who required treatment with epinephrine for their reaction.
Patients who cannot remember what happened during the index reaction but were told it was serious and/or required medical intervention.
Unable to give informed consent.
Severe or uncontrolled asthma.
Severe chronic obstructive airway disease.
Severe aortic stenosis.
Patients who, at the time they are being considered for DPT, are acutely unwell or clinically unstable. This includes patients with respiratory and/or cardiac compromise.
Pregnancy.
Previous penicillin allergy testing which concluded that the patient was allergic to penicillin.

Box 4

Principles for the conduct of a drug provocation testing (DPT)5

Written informed consent is required.
If the index penicillin is not known, amoxicillin should be used.
Blood pressure, heart rate and oxygen saturations should be checked prior to commencing the DPT and after each dose given.
Single or divided dose DPT may be used according to local preference.
- Single-dose DPT
- Graded DPT
  - Administer 10% of a full dose of amoxicillin.
  - Observe for 30 min.
  - Administer 50% of a full dose of amoxicillin.
  - Observe for 30 min.
  - Administer remainder of a full dose of amoxicillin.
If an alternative penicillin is used for the DPT, follow the same percentage dosing schedules as above.
Peak flow should be included in the observations measured, depending on local guidelines in place during the COVID-19 pandemic.
Should symptoms consistent with anaphylaxis develop during the test, treat the patient in accordance with the Resuscitation Council Guidelines for management of anaphylaxis.
The patient should be observed for 1 hour after the last dose (if an inpatient, ensure this observation is conducted by a member of the delabelling team).
The patient should be provided with clear written instructions about what to do if symptoms develop after leaving the hospital.
If a prolonged course is required (see ‘Use of prolonged DPT’ below), this should be provided to the patient. They must be contacted at the end of the course to check for delayed reactions.
A system should be in place to inform the general practitioner (GP) and other relevant healthcare professionals about the result of the DPT. The patient should receive clear written information about the test result and its implications.
Note: Training for those delivering the service should be developed in collaboration with the allergy/immunology service providing support for the delabelling service.

Key issues addressed in this guideline

Risk stratification

The BSACI guideline outlines how clinicians can stratify a patient’s risk of allergy, and thus their suitability for direct DPT:

‘Low-risk’ symptoms, consistent with non-immunological side effects and probable non-allergic phenomena (box 2).
Symptoms consistent with type 1 (immunoglobulin E-mediated) or type 4 (typically T cell-mediated) hypersensitivity reactions (box 3). These patients are not suitable for direct DPT by non-allergists and require evaluation by an allergy specialist.
Symptoms entirely consistent with side effects and other non-allergic phenomena in whom allergy testing is not indicated (box 2).

Suitability for DPT

Patients with symptoms listed in box 2 but none of the exclusion criteria listed in box 3 are suitable for direct DPT performed by a non-allergist outside an allergy clinic setting.

Patients in whom there is a clinical imperative for penicillin treatment but DPT is not appropriate should be referred to an allergy specialist for evaluation.

Conduct of DPT

Box 4 outlines the safe conduct of a DPT. This can be a single or graded DPT, dependent on the preference of the allergy team supporting the set-up of the service. Amoxicillin is the drug of choice, unless the index penicillin is known, in which case the index penicillin should be used. Doses should be age adjusted as per the British National Formulary for Children (BNFC) for the paediatric population.

All personnel administering a DPT should have up-to-date training in advanced paediatric life support and management of anaphylaxis, and immediate access to on-site resuscitation facilities including provision of critical care.

Use of prolonged DPT

Evidence supporting the use of prolonged DPT is weak, and consideration should be given to the risk of antimicrobial resistance. It is not known whether a single dose of antibiotic is sufficient to trigger a delayed hypersensitivity reaction or whether this is a problem which develops only after accumulation of the drug. The local/regional allergy service overseeing the delabelling service should define which patients require prolonged DPT and how long this should be.

Retention of delabelled status

Failure to retain ‘de-labelled’ status is associated with persistent and unnecessary penicillin avoidance by the patient. The patient, their carer(s) and their general practitioner should be provided with clear written information about the test result and its implications for future prescribing.

What do I need to know

What should I start doing?

Approach the regional paediatric allergy/immunology service to support the set-up of a penicillin delabelling service by non-allergists.
Take a thorough drug allergy history from all children and young people labelled ‘penicillin allergic’.
Identify patients at low risk for a true penicillin allergy and offer these patients DPT, provided no exclusion criteria are met.

What should I stop doing?

What can I continue to do as before?

Maintain up-to-date training in advanced paediatric life support and the management of anaphylaxis.
Consider referring patients with type 1 or type 4 hypersensitivity reactions to penicillin to a paediatric allergy specialist for evaluation, especially if it is of medical importance to the patient due to an underlying condition requiring frequent antibiotics.
Prescribe epinephrine autoinjectors to patients with a history of anaphylaxis and refer to paediatric allergy services.

What should I do differently?

Critical review

The delabelling of children with ‘penicillin allergy’ is challenging due to the frequency with which children develop skin rashes. These are most often secondary to viral infections, not drug allergy,3 though the aetiology of a rash can be difficult to distinguish both acutely and retrospectively. The guideline authors describe a benign rash as ‘non-urticarial’ (box 2) but also acknowledge that urticarial rashes in children are often viral. Furthermore, the exclusion criteria (box 3) described by the authors include a ‘rash lasting more than 24 h and/or affecting more than 10% of body surface’, which is a common characteristic of many viral exanthemata. Children who might otherwise be considered to have a low risk of penicillin allergy may be deemed unsuitable for DPT by these criteria. Investigation of patients with penicillin allergy is currently limited to specific indications (box 5), meaning many children may continue to be erroneously labelled as penicillin allergic without the widespread provision of a delabelling service. We feel that further research is required to refine these criteria to better suit the unique challenges the paediatric population presents, and therefore increase the number of children who could be successfully delabelled by non-allergist paediatricians using this guideline.

Box 5

Indications for the investigation of patients with immediate or non-immediate* reaction/s to penicillin/s and cephalosporin/s6

Patients with a label of ‘multiple antibiotic allergy’.
Patients with a history of immediate or non-immediate* reaction to penicillin/s and/or cephalosporin/s with a requirement for frequent antibiotics, for example, patients with bronchiectasis, cystic fibrosis (CF), diabetes, primary and secondary immunodeficiencies or with asplenia/hyposplenism.
Patients with a history of immediate or non-immediate* reaction to penicillin/s and/or cephalosporin/s requiring specific treatment with a beta-lactam.
Anaphylaxis during general anaesthesia when penicillin was administered alongside multiple other agents.

The authors of the guideline identify that further research (box 6) is required to determine those children who require a single or prolonged DPT. The optimal length of a prolonged DPT is also unknown, with the authors deferring to specialist allergy services to define this. The BSACI guideline ‘Management of allergy to penicillins and other beta-lactams’ suggests a 5-day course should be offered to children with a history of non-immediate urticaria or maculopapular rashes, with the first dose to be given in hospital. Support and guidance from specialist paediatric allergists is therefore needed for non-allergist paediatricians to establish a safe and effective delabelling service. Take home messages are outlined in box 7.

Box 6

Areas identified for future research5

Optimal length of prolonged drug provocation testing (DPT) in adults and children.
Feasibility of penicillin allergy delabelling in primary care settings.
Optimising retention of delabelled status.
Development of educational packages for non-allergists delivering delabelling services.

Box 7

Take home messages

True penicillin allergy is uncommon in the paediatric population and reactions are rarely reproduced with challenge testing.
All children and young people labelled ‘penicillin allergic’ presenting to hospital should be risk stratified and considered for penicillin allergy investigation.
Non-allergists providing delabelling services should have support and training provided by specialist paediatric allergy services.