Drug eruptions: recognising and managing cADRs in general practice

Cutaneous adverse drug reactions in Australian general practice

Drug rashes are the most common manifestation of adverse drug reactions encountered in general practice, occurring in 2–3% of all inpatient drug courses and accounting for a large proportion of dermatology referrals in patients taking four or more medicines. The spectrum runs from mild morbilliform rashes that resolve with drug cessation to life-threatening severe cutaneous adverse reactions (SCAR) with multiorgan involvement and significant mortality if not recognised early.

The correct clinical approach turns on pattern recognition: timing, morphology, culprit drug identification, and the presence or absence of systemic features. Drug history is the diagnosis — every cutaneous adverse drug reaction (cADR) requires a thorough review of all agents started in the relevant latency window, including over-the-counter medicines, supplements, and recently ceased drugs. eTG Drug-induced skin reactions and the Australasian College of Dermatologists’ drug eruption resource are the primary Australian clinical references.

A. Core clinical — the AU general-practice framework

Classification by phenotype

Phenotype	Typical latency (first exposure)	Rechallenge latency	AU frequency
Morbilliform (maculopapular exanthem)	7–14 days	Hours–2 days	~90% of drug rashes
Fixed drug eruption (FDE)	1–2 weeks	<24 hours at same site	2–5%
Drug-induced photosensitivity (phototoxic or photoallergic)	Hours (phototoxic) – days (photoallergic)	Hours	Common with doxycycline
AGEP (acute generalised exanthematous pustulosis)	1–3 days, usually <48 h	Hours	Rare (~1–5 per million/year)
DRESS / DiHS (drug reaction with eosinophilia and systemic symptoms)	2–8 weeks	Days	Rare (~1–10 per 10,000 anticonvulsant exposures)
SJS/TEN	4–28 days	Days	Rare — see linked SJS/TEN article

Severe cutaneous adverse reactions (SCAR) = DRESS + AGEP + SJS/TEN + generalised bullous FDE. All require drug cessation, hospital admission, and mandatory TGA DAEN reporting.

Taking a focused history

Drug history is the diagnosis. For every patient with a drug rash, systematically review all prescribed medicines, OTC products, herbal and complementary preparations, and drugs ceased in the past 8 weeks — DRESS can present up to 8 weeks after drug initiation and the culprit may already have been stopped.

Key questions:

• Exact start dates for all agents
• Timing: when did the rash start relative to each drug?
• Prior reactions: previous rash to this or a chemically related agent?
• Asian or Aboriginal and Torres Strait Islander ancestry: pharmacogenomic HLA risk (see Section C)
• HIV status, immunosuppression, or autoimmune disease: all amplify SCAR risk 100-fold (especially sulfonamides and nevirapine in HIV)
• Family history of SCAR: HLA-linked predisposition in first-degree relatives

Physical examination

Morbilliform: symmetrical erythematous macules and papules on the trunk spreading centrifugally; blanching; may be pruritic; no fever; no mucosal involvement; no skin pain; no Nikolsky sign.

Fixed drug eruption: single or few well-demarcated round/oval dusky-violaceous to brown plaques, recurring at exactly the same site with each drug exposure; common sites — genitals, lips, acral areas.

Photosensitivity: confined to sun-exposed areas (face, V-neck, dorsal hands, forearms, shins) with sharp cut-off at clothing lines; submental triangle, retroauricular skin, and eyelids are spared.

AGEP: dozens to hundreds of pinhead-sized sterile non-follicular pustules on erythematous skin, starting in flexures (axillae, groin, neck) and spreading rapidly; fever; resolution with desquamation after drug cessation.

DRESS: morbilliform to erythrodermic rash (>50% body surface area in 75% of cases) with prominent facial and periorbital oedema (a cardinal sign); lymphadenopathy; systemic organ involvement. Mucosal involvement is less prominent than in SJS/TEN.

Red flags demanding immediate escalation: painful skin; positive Nikolsky sign; blisters; erosions or ulcers on lips, eyes, or genitalia; temperature >38.5°C; facial oedema; jaundice — these indicate possible SJS/TEN or DRESS; move immediately to hospital workup.

Investigations

First-line blood tests for any non-trivial drug eruption (febrile, widespread, facial oedema, systemic symptoms):

• FBC with differential — eosinophilia >700/µL and atypical lymphocytes are RegiSCAR DRESS criteria (Kardaun et al. Br J Dermatol 2013)
• Liver function tests — hepatitis occurs in 50% of DRESS (hepatocellular or cholestatic)
• U&E and creatinine — interstitial nephritis / acute kidney injury in ~30% of DRESS
• CRP — non-specific; usually elevated in SCAR
• TSH — baseline before DRESS diagnosis; repeat at 2, 4, and 6 months (late autoimmune thyroiditis is a recognised DRESS sequela)
• Troponin and ECG — if any chest symptoms (DRESS myocarditis is rare but lethal)
• Viral PCR (HHV-6, EBV, CMV) — useful in confirming DiHS subtype and explaining relapsing course

Severity scoring: use the RegiSCAR DRESS validation score and EuroSCAR AGEP score to guide triage and admission threshold.

Management — guideline-aligned

eTG Drug-induced skin reactions, ACD A-Z drug eruptions, and ASCIA drug allergy 2024 all agree on the first step:

Step 1 — Stop the suspect drug. This is the single highest-yield intervention for every cADR. If polypharmacy makes the culprit unclear, stop all non-essential drugs introduced in the relevant latency window. Do not attempt to treat through a morbilliform rash caused by an aromatic anticonvulsant or allopurinol without specialist input — escalation to SCAR is a real risk.

Uncomplicated morbilliform (afebrile, no organ involvement, no mucosae, no Nikolsky): outpatient management. Non-sedating oral antihistamine — cetirizine 10 mg daily, fexofenadine 180 mg daily, or loratadine 10 mg daily. Medium-potency topical corticosteroid (mometasone 0.1% ointment or betamethasone valerate 0.05% ointment) twice daily until clear. Liberal emollient. Safety-net: return immediately for fever, facial swelling, painful skin, mouth or eye ulcers, or blistering.

Fixed drug eruption: stop drug; topical corticosteroid for active lesion; counsel about post-inflammatory hyperpigmentation persisting for months; document on My Health Record; avoid the culprit drug class lifelong.

Photosensitivity: cease the drug where possible; if continuation is clinically necessary (e.g., amiodarone), counsel about rigorous broad-spectrum SPF 50+ sun protection and protective clothing; pigmentary changes may persist months after cessation.

AGEP: admit if extensive or febrile; stop drug; supportive care — IV fluids, paracetamol for fever (confirm paracetamol is not the culprit), topical corticosteroid, careful skin care; resolution typically 1–2 weeks on drug cessation alone.

DRESS: admit immediately; dermatology and clinical immunology involvement; organ-specific consultations (gastroenterology if hepatitis, nephrology if acute kidney injury, cardiology if troponin elevated). Commence systemic prednisolone 1 mg/kg/day (maximum 60–80 mg) once the clinical and biochemical diagnosis is confirmed and infection excluded. Long taper over 6–9 months to cover the viral reactivation phase and prevent rebound — weekly to fortnightly biochemical review (Shiohara Allergol Int 2006; Cacoub Am J Med 2011).

B. Evidence appraisal

DRESS steroid taper duration — long versus short

Shiohara and Kano (Allergol Int 2006) and Cacoub et al. (Am J Med 2011) advocate a prolonged taper over 6–9 months to cover the HHV-6/EBV reactivation phase. European ESPI 2019 and Husain JAAD 2013 data support shorter tapers (6–12 weeks) with comparable mortality but higher relapse rates. The comparison is based on observational cohort data only — no head-to-head RCT exists. Australian practice follows the longer taper approach for moderate-to-severe DRESS with organ involvement; shorter tapers are acceptable in mild DRESS without organ failure, under specialist guidance.

HLA-B*57:01 screening before abacavir — landmark evidence

Mallal et al. PREDICT-1 (NEJM 2008) — a randomised controlled trial of 1,956 HIV-positive patients — demonstrated that universal HLA-B57:01 pre-screening reduced abacavir hypersensitivity reaction to near zero, with 100% negative predictive value. Under the Australian PBS, HLA-B57:01 testing is mandatory before prescribing abacavir. No exceptions.

HLA-B*15:02 screening before aromatic anticonvulsants

Chen et al. (NEJM 2011) — a population-based study in Han Chinese patients — demonstrated that universal HLA-B*15:02 prescreening before carbamazepine reduced carbamazepine-induced SJS/TEN to near-zero. The TGA issued a black-triangle advisory in 2009 endorsing pre-testing in patients of Han Chinese, Thai, Vietnamese, Filipino, or Malay ancestry. A positive result contraindicates the entire aromatic anticonvulsant class (carbamazepine, phenytoin, lamotrigine, oxcarbazepine, phenobarbital all cross-react). The test is not routinely PBS-funded for this indication in Australia — out-of-pocket cost is approximately $50–150 — but the investment is warranted given the catastrophic alternative.

Cease versus treat-through in mild morbilliform

Guideline default is to stop the suspect drug. An exception exists in HIV and tuberculosis management where an antiretroviral or antituberculous drug may be continued under specialist supervision if the reaction is mild, afebrile, and no SCAR features are present. This exception does not apply to allopurinol or aromatic anticonvulsants — these must be stopped at first sign of rash, per ASCIA drug allergy 2024 and eTG.

C. HLA pharmacogenomics — AU-specific risks

Three pharmacogenomic HLA loci carry disproportionate SCAR risk in populations well-represented in Australia:

HLA allele	Drug / reaction	High-risk ancestry in AU	PBS-funded testing?
HLA-B*57:01	Abacavir → hypersensitivity syndrome	All ancestry	Yes — mandatory pre-abacavir
HLA-B*15:02	Aromatic anticonvulsants → SJS/TEN/DRESS	Han Chinese, Thai, Vietnamese, Filipino, Malay	No — private ($50–150)
HLA-B*58:01	Allopurinol → SCAR	Han Chinese, Thai, Vietnamese, Korean, ATSI	Yes — in defined high-risk groups

Counsel patients with a confirmed HLA-driven SCAR that first-degree relatives share the risk and should undergo HLA typing before being prescribed the relevant drug class. Document results in the patient’s My Health Record allergy section.

D. Australian operations

Reporting to TGA DAEN

Every GP who manages a confirmed SCAR — or any hospitalisation due to a drug rash — is required to submit a report to the TGA Database of Adverse Event Notifications. Online reporting is straightforward. Reporting for moderate cADRs (not requiring hospitalisation) is recommended but not mandatory. Post-market surveillance depends on this reporting, particularly for TGA black-triangle monitored drugs.

MBS pathology items

FBC 65070; liver function tests 66512; U&E and creatinine 66500; CRP 66503; TSH 66719; troponin 66509; ECG 11700; skin biopsy by dermatologist item 30071. Viral PCR for HHV-6/EBV/CMV is rebatable in defined clinical contexts.

PBS treatments for drug rashes

No PBS therapy specifically targets drug eruptions — standard antihistamines (cetirizine, loratadine, fexofenadine) and topical corticosteroids (mometasone, betamethasone valerate) are PBS general. Prednisolone is PBS general for DRESS management. Ciclosporin for steroid-refractory DRESS requires PBS Authority at hospital initiation. IVIG for DRESS requires National Blood Authority (BloodSTAR) criteria approval.

Documentation and patient safety pack

For every SCAR or confirmed drug allergy, document in the clinical record: drug name and brand, dose, indication, start date, reaction date, phenotype, severity. Issue the patient a written drug-allergy card listing the culprit drug, the cross-reactive class to avoid, severity, and treating clinician contact. Update My Health Record allergy/adverse reaction section. Advise on a MedicAlert Foundation Australia bracelet.

E. Special populations

Older adults. Polypharmacy amplifies cADR risk. The 8-week history window is more complex with multiple agents. Sedating antihistamines (promethazine, chlorpheniramine) carry anticholinergic and falls risk in older patients — use non-sedating alternatives (cetirizine, fexofenadine, loratadine). DRESS in older adults carries higher mortality from hepatic and renal failure.

Aboriginal and Torres Strait Islander patients. HLA-B58:01 carrier frequency is elevated in some communities. Allopurinol is one of the most-prescribed medicines in Indigenous Australians for gout (high prevalence). Consider HLA-B58:01 testing before initiating allopurinol. Coordinate pharmacogenomic counselling with the local Aboriginal Medical Service or ACCHO where available.

HIV-positive patients. SCAR risk is 100-fold higher, particularly with sulfonamides, nevirapine, and abacavir. Mandatory HLA-B*57:01 pre-screening before abacavir is non-negotiable. Liaise closely with the HIV physician for any drug change.

Patients on multiple anticonvulsants. Aromatic anticonvulsants extensively cross-react. A patient who has had carbamazepine-induced SCAR cannot safely receive phenytoin, lamotrigine, oxcarbazepine, or phenobarbital. Alternative anticonvulsant classes (valproate, levetiracetam, gabapentin) may be substituted under neurology guidance.

When to escalate

Refer to emergency department or specialist services when:

• Any suspected SCAR (DRESS, AGEP with systemic features, generalised bullous FDE, SJS/TEN features — mucosal involvement, Nikolsky sign, blisters, skin pain) — same-day hospital transfer
• Fever above 38.5°C with a drug rash — at minimum urgent blood tests; low threshold for admission pending results
• DRESS confirmed or probable — dermatology + clinical immunology + organ-specific consultation
• Diagnostic uncertainty about phenotype — dermatology referral for biopsy or specialist assessment
• Post-SCAR drug allergy clarification — clinical immunology for patch testing (6 weeks to 6 months after recovery), intradermal testing, or graded challenge

Refer to clinical immunology routinely for all SCAR survivors (after recovery), ambiguous cADR, or where a definitive drug-allergy clarification will affect future treatment options.

What this article is and is not

This is general health information drawn from current Australian clinical guidelines — eTG Drug-induced skin reactions, Australasian College of Dermatologists, ASCIA drug allergy 2024, and major pharmacogenomics evidence. It is not personal medical advice and does not substitute for a consultation with a qualified practitioner, particularly for diagnosis of severe drug reactions. Any person who develops a drug rash with fever, facial swelling, painful skin, or mucosal involvement should seek urgent medical assessment.

For further information: HealthDirect — Drug allergies, Better Health Channel, ASCIA patient information, ACD — A-Z drug eruptions.

---

Sources cited

1. eTG complete — Drug-induced skin reactions
2. Australasian College of Dermatologists — A-Z drug eruptions
3. ASCIA — Drug allergy clinical position 2024
4. AMH — Antihistamine, corticosteroid, prednisolone, ciclosporin monographs
5. TGA — Database of Adverse Event Notifications (DAEN) + carbamazepine HLA-B*15:02 advisory 2009
6. Kardaun SH et al. — RegiSCAR DRESS validation (Br J Dermatol 2013)
7. Sidoroff A et al. — EuroSCAR AGEP scoring (Br J Dermatol 2001)
8. Chen P et al. — HLA-B*15:02 screening before carbamazepine (NEJM 2011)
9. Mallal S et al. — PREDICT-1 HLA-B*57:01 screening for abacavir (NEJM 2008)
10. Hung S-I et al. — HLA-B*58:01 and allopurinol SCAR (PNAS 2005)
11. Cacoub P et al. — DRESS syndrome review (Am J Med 2011)
12. Shiohara T, Kano Y — DiHS/DRESS and HHV-6 reactivation (Allergol Int 2006)
13. Barbaud A et al. — Skin tests in cADR investigation (Contact Dermatitis 2013)
14. HealthDirect — Drug allergies
15. Better Health Channel — Medicines and rashes
16. ASCIA — Drug allergy patient information
17. MedicAlert Foundation Australia
18. My Health Record