Drug allergy and delabelling in children — branching viva

Opening question

Examiner: Take me through this child. Is her penicillin-allergy label likely to be real, and what is your frame for managing it? [3]

Candidate: The label is most likely false. Around 5–10% of children carry a penicillin-allergy label, but more than 90% cannot be confirmed when formally tested — and her history is the classic false label: a maculopapular rash on day 5 of amoxicillin during a viral illness, with no systemic features. My frame is not "what antibiotic can I use instead?" but "is this allergy even real?" I will take a structured history, risk-stratify her with a validated rule, and if she is low-risk — which she appears to be — delabel her with a supervised oral amoxicillin challenge, restoring first-line therapy. [3] [10]

Examiner: Why are most childhood labels false? [10]

Candidate: Three reasons. First, the rash attributed to the antibiotic was usually caused by the intercurrent viral illness — the amoxicillin-EBV rash is the classic, but any childhood exanthem can be misattributed to the drug. Second, many labels are based on non-allergic symptoms such as gastrointestinal upset, or on second-hand family history. Third, where a true IgE allergy once existed, it wanes — most genuinely allergic children tolerate penicillin again within about ten years. So the label outlives the biology. [10] [9]

Branch 1 — classification and mechanism

Examiner: How do you classify drug reactions, and where does hers sit? [4]

Candidate: I classify by Gell-Coombs and by timing, using the one-hour threshold. Immediate reactions, within one hour, are Type I IgE-mediated — urticaria, angioedema, anaphylaxis, bronchospasm. Delayed reactions, after one hour to days, are Type IV T-cell-mediated — the maculopapular exanthem that dominates paediatrics, plus fixed eruptions and the severe blistering syndromes. This child's rash on day 5 is a delayed, non-immediate pattern, mild and non-toxic. [4]

Examiner: What is the immunological basis of these reactions? [4]

Candidate: Beta-lactams are low-molecular-weight haptens that covalently bind host proteins to become immunogenic. In the immediate pathway, drug-specific IgE on mast cells is cross-linked on re-exposure, causing degranulation and the rapid Type I reaction. In the delayed pathway, drug-specific T-cells infiltrate skin hours to days after exposure, producing the exanthem — and, in severe forms, Stevens-Johnson syndrome, DRESS and AGEP. But the key point for this child is that her rash most likely had no immune basis at all — it was the virus. [4]

Branch 2 — risk stratification and PEN-FAST

Examiner: How do you decide who is safe to challenge directly? [2]

Candidate: I take the structured history and apply the PEN-FAST rule, developed and validated by Trubiano and colleagues. PEN-FAST stands for Penicillin allergy (scored by how long ago the reaction occurred), five minutes or less to treat, Anaphylaxis, Severe skin reaction, and Treatment received. A score under 3 identifies a low-risk patient suitable for direct oral amoxicillin challenge without prior skin testing. I also actively seek the history red flags — anaphylaxis, blistering, mucosal involvement, multi-organ disease — that would instead mandate specialist referral and never-challenge. [2] [10]

Examiner: This child's history is low-risk. What would change your mind? [10]

Candidate: Any history red flag. If the original reaction had been anaphylaxis within an hour, blistering or skin pain, mucosal involvement, organ dysfunction with rash, or recurrence on re-exposure, I would not challenge — I would refer to specialist allergy. The severe delayed syndromes — Stevens-Johnson syndrome and toxic epidermal necrolysis, DRESS, AGEP — are absolute, permanent contraindications to re-challenge because re-exposure can be fatal. The whole pathway depends on reading the history accurately. [10] [4]

Branch 3 — the challenge and delabelling

Examiner: Walk me through the direct oral challenge and what you do if she tolerates it. [1]

Candidate: I give a single or graded dose of oral amoxicillin under medical observation, with rescue medication and anaphylaxis capacity available, for a defined observation period. The PALACE randomised trial by Copaescu and colleagues showed that a PEN-FAST-guided direct oral challenge was non-inferior to skin testing for low-risk penicillin allergy and more efficient, and the Srisuwatchari meta-analysis confirmed direct drug provocation is safe in selected children. If she tolerates it, the label is false. [1] [6]

Examiner: She tolerates it. What now? [10]

Candidate: Delabelling is incomplete until it is documented everywhere. I update the electronic medical record, the pharmacy record, MedicAlert, the family-held record and the medication list; counsel the family verbally and in writing that the allergy was not real; and communicate with the general practitioner. A label that survives in one record will resurface at the next encounter and the harm will return. The systems-level gain is antibiotic stewardship — restoring first-line amoxicillin avoids broader-spectrum, more toxic alternatives, shortens admissions and lowers resistance. [10] [9]

Branch 4 — the severe reactor and the contrast

Examiner: Contrast this child with one who had Stevens-Johnson syndrome. Why is that child never challenged? [4]

Candidate: A child with Stevens-Johnson syndrome or toxic epidermal necrolysis, DRESS or AGEP has had a severe, T-cell-mediated reaction in which re-exposure to the culprit can be fatal. These are absolute, permanent contraindications to re-challenge. That child is never part of any delabelling pathway and never challenged directly; the culprit and structurally related drugs are avoided for life, the family is counselled, and the child is referred to specialist allergy for definitive documentation. The contrast is the whole point of the topic: the low-risk, unverified label is challenged and delabelled; the severe, verified reaction is documented and avoided for life. [4]

Examiner: And what about an immediate anaphylactic reaction — how is that managed acutely? [4]

Candidate: Anaphylaxis is treated first with intramuscular adrenaline at 0.01 mg/kg into the anterolateral thigh, repeated every 5 minutes as needed, with airway support, oxygen, supine positioning and fluids. The drug is stopped immediately. Antihistamine and corticosteroid are adjuncts for urticaria, never a substitute for adrenaline when the airway or circulation is involved. After the acute episode, the child has specialist allergy evaluation — but a confirmed anaphylactic reaction is a never-re-challenge label. [4]

Wrap

Examiner: Summarise the drug-allergy and delabelling stance in one sentence. [10]

Candidate: Most paediatric penicillin-allergy labels are false, so take a structured history, risk-stratify with PEN-FAST, delabel the low-risk majority by direct oral amoxicillin challenge, document the removal in every system, never re-challenge the severe reactors (anaphylaxis, Stevens-Johnson syndrome, DRESS, AGEP), and treat drug-induced anaphylaxis first with intramuscular adrenaline — because the goal is to verify, not to assume, and to restore first-line beta-lactam therapy wherever it is safe. [1] [10]