Penicillin allergy — considerations
Is the reported allergy even real? Sort intolerance from true allergy and score it with PEN-FAST — then, if it is real, work out which antibiotic is safe by the R1 side-chain.
Assess the reported reaction
Clinical history is the test. Separate a predictable intolerance (de-label it) from a true hypersensitivity (classify it).
A family history of antibiotic allergy is NOT a reason to avoid the drug. Distinguish a predictable adverse reaction from genuine immune-mediated hypersensitivity.
Ask the patient (eTG Fig 2.125)
- Severity & type — what exactly happened? Rash, swelling, breathing difficulty, collapse?
- Management — did it need treatment or hospitalisation (adrenaline, steroids, admission)?
- Timing — how soon after the dose? (minutes–2 h suggests immediate; days suggests delayed)
- How many years ago did it happen?
- Tolerated since — any penicillins or other antibiotics taken without problem since the reaction?
Predictable intolerance — not an allergy
Type A reaction: pharmacologically predictable, no immune features.
- GI upset (nausea, vomiting, diarrhoea)
- Headache, dizziness
- Isolated itch WITHOUT rash, fever or organ involvement
- Family history of antibiotic allergy only
→ De-label: A medical practitioner can directly de-label — remove the allergy on history alone, then document how it was excluded (to avoid relabelling).
Features of true hypersensitivity
Type B reaction: immune-mediated — classify it, don't dismiss it.
- Urticaria, angioedema
- Anaphylaxis / airway compromise / hypotension / collapse
- Maculopapular or other rash
- Severe cutaneous reaction (DRESS, SJS/TEN, AGEP) or organ involvement
→ Next: Classify the reaction (below), then check which antibiotic is safe — and use PEN-FAST to gauge how likely the allergy is real.
Classify the reaction — two axes (eTG Table 2.136)
Combine the two axes (e.g. immediate + severe). A delayed SEVERE reaction contraindicates re-exposure — including desensitisation.
Which antibiotic is safe for each reaction type ↓eTG Antibiotic — Penicillin hypersensitivity, Figs 2.130–2.133 (Dec 2025); reconciled with O&D Figs 13.8–13.11.
eTG Antibiotic — Antimicrobial hypersensitivity (Dec 2025); O&D — Antimicrobial hypersensitivity (Sept 2025).
PEN-FAST — how likely is the allergy real?
A validated score that risk-stratifies a reported penicillin allergy. Adults (and adolescents) — not validated in children.
Three questions → a score out of 5. A score under 3 marks a low-risk allergy that may be suitable for de-labelling / direct oral challenge. PEN-FAST stratifies risk — it does not by itself clear a patient to receive penicillin; that follows a de-labelling pathway or allergy service.
Five years or less scores 2. eTG scores unknown timing as zero.
SCAR = DRESS, SJS/TEN, AGEP. A severe delayed rash with mucosal involvement counts as SCAR even if unconfirmed.
e.g. adrenaline, steroids, antihistamine, or hospitalisation. eTG scores unknown as 1.
Answer all three to see the PEN-FAST score and risk band.
Score under 3 = low risk: in the Australian validation about 1 in 100 had a confirmed penicillin allergy (eTG); the original rule reported NPV 96.3% (Trubiano 2020).
PEN-FAST: Trubiano JA et al. Development and Validation of a Penicillin Allergy Clinical Decision Rule. JAMA Intern Med 2020;180(5):745-52. Reproduced in eTG Antibiotic, Fig 2.127 (Dec 2025).
Calculator rebuilt from the published PEN-FAST decision rule (Trubiano 2020); the eTG/AMA figure itself is not reproduced.
Decide by reaction severity
Classify the penicillin reaction first — its type and severity set what you can and can’t give.
TG O&D 2025 — Penicillin hypersensitivity, Figures 13.8–13.11.
Predictable intolerance (nausea, diarrhoea, headache, itch WITHOUT rash, fever or organ involvement) is not allergy — it can be directly delabelled by a medical practitioner. A family history of antibiotic allergy does not justify avoiding the drug. For a reported cephalosporin allergy, seek expert advice.
Why — the R1 side-chain
Cross-reactivity tracks the R1 side-chain, not the shared β-lactam ring. It is only a real concern when two β-lactams carry the same (or near-identical) R1.
amoxicillin, ampicillin ↔ cefalexin, cefaclor, cefadroxil, cefprozil
Identical R1: ampicillin = cefaclor/cefalexin; amoxicillin = cefadroxil/cefprozil. Almost-identical across the group (they differ only by a hydroxyl). Shared-R1 cross-reactivity ≈ 14–38% — so cefalexin/cefaclor must NOT be given after a SEVERE immediate amoxicillin/ampicillin reaction.
Trubiano 2017 Fig 3 (14–38%); TG O&D 2025 Table 13.3.
cefazolin (IV)
Shares no R1 or R2 with any penicillin or other cephalosporin → the lowest-risk β-lactam. The cephalosporin of choice when a β-lactam is strongly preferred despite severe immediate penicillin allergy (hospital, IV — eg cefazolin + metronidazole for severe spreading odontogenic infection).
TG O&D 2025 (spreading infection, severe); Trubiano 2017; Stevenson 2025.
eg cefuroxime, ceftriaxone, cefotaxime
Different R1 from the aminopenicillins, so cross-reactivity to penicillins is minimal; usable in non-immediate / non-severe penicillin allergy. (They do cross-react amongst themselves by their own shared R1 — eg the ceftriaxone/cefotaxime/cefepime/ceftazidime group.)
Trubiano 2017 Fig 3; TG O&D 2025 Table 13.3.
TG O&D 2025 Table 13.3; Trubiano 2017 Figure 3 (verified from the figure).
The numbers
What the evidence actually shows.
AU validation (Stevenson 2025): R1 side-chain groups explained most skin-test and oral-challenge outcomes, and all 7 unrelated-cephalosporin oral challenges were tolerated. Caveat: 5/49 (10.2%) cefazolin-allergic patients without a prior penicillin label had a positive penicillin skin test — so a confirmed cefazolin allergy still warrants a penicillin-allergy assessment.
Common misconceptions
Scaffold for review only — clinical content is not yet verified. PEN-FAST risk-stratifies a reported allergy to guide de-labelling / referral; it does NOT by itself clear a patient to receive penicillin. De-labelling, oral challenge and prescribing decisions rest with the treating clinician or allergy service. Confirm every threshold against current Therapeutic Guidelines before any clinical use.
Drafted from Therapeutic Guidelines: Antibiotic — Antimicrobial hypersensitivity (Dec 2025) and Oral & Dental — Antimicrobial hypersensitivity (Sept 2025); PEN-FAST from Trubiano JA et al, JAMA Intern Med 2020. Captured in wiki/sources/tg-abx-2025. NOT yet reconciled / signed off.
Therapeutic Guidelines: Oral and Dental — Antimicrobial hypersensitivity in dental practice (Table 13.3; Figures 13.8–13.11), published Sept 2025. Mechanism: Trubiano JA et al. 2017, JACI: In Practice 5(6):1532–1542, Figure 3. AU validation: Stevenson/Trubiano et al. 2025, JACI: Global 5(1):100583.