Stevens-Johnson syndrome and toxic epidermal necrolysis

Overview

SJS and TEN are a spectrum of severe, life-threatening drug-induced mucocutaneous reactions characterised by widespread epidermal detachment and mucosal erosions. Distinction is by percentage BSA affected.

Classification

SJS vs overlap vs TEN by BSA detachment
FeatureSJSSJS-TEN overlapTEN
BSA detachment<10%10-30%>30%
Mortality~5-10%Intermediate25-40%

Presentation

Prodrome (1-3 days before skin eruption): fever, malaise, sore throat, burning eyes, dysphagia - onset 1-3 weeks after starting causative drug
Skin lesions: dusky-red erythematous macules or atypical (flat) target lesions - trunk, face, proximal limbs, palms and soles
Epidermal detachment: blisters coalesce, epidermis peels in sheets
Nikolsky sign positive: lateral shear on intact erythematous skin causes epidermis to slide off
Mucosal involvement (>90%): oral erosions/haemorrhagic crusting, conjunctivitis, genital/urethral erosions
Systemic: fever, tachycardia, hypotension, confusion in severe cases
🚨
Prodromal fever, sore throat, and burning eyes in a patient recently started on carbamazepine, lamotrigine, or allopurinol = SJS/TEN until proven otherwise. Stop the drug immediately - do not wait for blistering.

Investigations

Clinical diagnosis - history, morphology, BSA assessment; investigations support management
Skin biopsy: full-thickness epidermal necrosis - distinguishes from SSSS (superficial/subcorneal cleavage)
FBC, U&E, LFTs, CRP: electrolyte derangement, organ involvement; lymphopenia = severity marker
Cultures (blood, skin, sputum): infection is leading cause of death
Ophthalmology review: slit-lamp for conjunctival/corneal involvement - essential in all confirmed cases
SCORTEN score: calculate on admission and day 3 to stratify mortality

Differential diagnosis

Key differentials
FeatureSJS/TENErythema multiformeSSSS
Lesion morphologyFlat dusky macules, blistersRaised 3-zone target lesionsSuperficial blistering/peeling
TriggerDrugs (>85%)HSV/MycoplasmaStaphylococcal exotoxin
Mucosal involvement>90%Less severeAbsent
Biopsy cleavageFull-thickness epidermalMinimal detachmentSuperficial (subcorneal)
Relation to SJS spectrum-Distinct - NOT on same spectrumDistinct

Management

Immediate: stop causative drug - single most important intervention
Specialist setting: dermatology ward; extensive BSA involvement warrants burns unit transfer
Supportive care: IV fluid resuscitation, electrolyte replacement, non-adherent wound dressings, nutritional support
Ophthalmology: urgent review for all confirmed cases
Immunomodulation (specialist decision): ciclosporin is considered
⚠️
Prophylactic antibiotics are NOT recommended - empirical use selects for resistant organisms and increases mortality. Use antibiotics only for confirmed infection guided by cultures.

Complications

Sepsis - leading cause of death; Staphylococcus aureus and Pseudomonas aeruginosa common
Ocular: symblepharon, corneal scarring, dry eye, permanent visual impairment
Respiratory failure: tracheobronchial mucosal sloughing, ARDS
Genitourinary: urethral strictures, vaginal adhesions
Long-term: post-inflammatory hyperpigmentation, PTSD
📌
Survivors must never re-take the causative drug or structurally related compounds. Document allergy in all medical records, issue allergy alert card, and inform GP and all future prescribers.

Causative drugs

High-risk drugs (must know): carbamazepine, lamotrigine, allopurinol, sulfonamides, phenobarbital, phenytoin
Other implicated drugs: salicylates, sertraline, imidazole antifungals, nevirapine
Non-drug triggers (especially in children): Mycoplasma pneumoniae, HSV
💡
HLA-B*15:02 strongly associated with carbamazepine-induced SJS/TEN in Han Chinese, Thai, and South-East Asian populations - MHRA recommends HLA screening before prescribing carbamazepine in these groups.

SCORTEN score

Each criterion scores 1 point. Calculate on day 1 and day 3.

SCORTEN criteria (1 point each)
Age >40 years
Malignancy present
BSA detachment >10% at presentation
Heart rate >120 bpm
Serum urea >10 mmol/L
Serum bicarbonate <20 mmol/L
Serum glucose >14 mmol/L
Score 0-1 = ~3% mortality; score ≥5 = >90% mortality