The child with a delayed triad after an antibiotic — structured encounter

Setting and candidate instructions

You are the paediatric registrar in the emergency department. A four-year-old boy is brought by his parents with a four-day history of a blotchy, moving rash over his trunk and limbs, low-grade fever and stiff, swollen knees. He finished a five-day course of cefaclor for an ear infection eight days ago. He looks miserable but not toxic, with normal observations, no mucosal involvement and no breathing difficulty. The examiner will ask you to build the diagnosis, exclude the dangerous mimics, choose and justify the investigations, construct the stepwise management plan, and make a defensible avoidance decision.

Information released on request

1. On the timeline: the rash, fever and joint stiffness began four days ago; the cefaclor course was completed eight days before onset, given for a middle-ear infection. [8]
2. On the rash: blotchy, polymorphic, some annular lesions, mildly itchy, migratory, no blistering, no skin pain, no mucosal involvement, no residual bruising. [13]
3. On the joints: knees and ankles swollen, warm and tender; the child limps but can weight-bear with encouragement; no morning stiffness history. [8]
4. On the examination: afebrile on arrival, normal observations, well-hydrated, no lymphadenopathy, no abdominal tenderness, no palpable purpura, urinalysis negative, normal observations and perfusion. [2]

Examiner prompts

1. "What is the most likely diagnosis, and what single feature of the history points you there?" [8]
2. "Which conditions must you exclude before settling on that diagnosis, and how?" [12]
3. "Which investigations do and do not help here, and why?" [13]
4. "Outline your stepwise management." [2]
5. "The parents ask whether he is now allergic to penicillin for life. What do you say?" [1]

Model performance

"The most likely diagnosis is a serum-sickness-like reaction. The single feature that points me there is the latency — the triad of rash, fever and arthralgia arriving eight days after completing a cefaclor course, which is the signature six-to-twenty-one-day window, with cefaclor the paradigmatic paediatric cause." [8]

"Before settling on serum sickness I exclude anaphylaxis (the latency of over a week and the absence of airway or circulatory compromise argue against it), Stevens-Johnson syndrome or toxic epidermal necrolysis (no mucosal blistering, skin pain or epidermal detachment), and DRESS (no eosinophilia or transaminitis, onset too early for the two-to-eight-week window). I also consider a viral exanthem with reactive arthralgia, the commonest confounder, and a vasculitis such as Henoch-Schönlein purpura if palpable purpura were present." [12] [13]

"The diagnosis is clinical, built on the latency and the triad. A routine battery of specific IgE or skin-prick tests does not help because these are not IgE-mediated reactions. The complement level is expected to be normal in a cefaclor SSLR and distinguishes it from true serum sickness, so I check it if there is any doubt about a biologic exposure but not routinely. A urinalysis and basic bloods are reasonable to detect organ involvement and grade severity." [13]

"My management follows the four-step ladder. I stop and document the cefaclor precisely. I give a non-sedating second-generation oral antihistamine at the local weight-based dose for the rash and itch, and reserve a short course of oral corticosteroid for severe arthralgia or a distressing rash. I provide analgesia for the joints, and I would escalate to admission and systemic corticosteroid only if organ involvement were present, which it is not." [2]

"On the lifelong penicillin question: a blanket lifelong label is not justified. The modern evidence supports an individualised approach — I avoid cefaclor and closely related drugs for now and document the reaction, and I arrange a specialist drug-allergy review with a graded oral challenge if a beta-lactam is clinically important in the future, because a coincidental viral exanthem is the common confounder and many labelled children tolerate the drug on challenge. I also make clear that this was serum sickness, not anaphylaxis or SJS, so the danger is not acute re-exposure in the same way." [1] [3]

Marking domains

Disposition and safety-net standard for this encounter

The child is managed as an outpatient once symptoms are controlled and the dangerous mimics are excluded. A clear written plan names the suspected drug (cefaclor), the avoidance stance (avoid cefaclor and closely related drugs for now; individualised decision with specialist review), the symptom-control regimen, and the return precautions (worsening rash, new mucosal involvement, high fever, reduced urine output, increasing joint pain or refusal to walk). The family is counselled that this was a serum-sickness-like reaction — not anaphylaxis or Stevens-Johnson syndrome — and that a graded challenge under specialist supervision can clarify the true drug-allergy status before any permanent label is applied. The avoidance decision and the planned specialist review are documented in the chart, the allergy record, and the discharge summary, with a named primary-care follow-up point. [1] [13]