To evaluate outcomes of the diagnostic workup in pediatric patients with a history of mild nonimmediate reactions (NIR) (>6 h) to amoxicillin (AMX) or amoxicillin-clavulanic acid (AMX/CL) and who underwent a Drug Provocation Test (DPT) with the culprit drug.

The study included 354 children with a history of mild NIR to AMX or AMX/CL who underwent DPT with the culprit drug at a single pediatric hospital. Patients with chronic urticaria, uncontrolled asthma, and severe cutaneous adverse reactions were excluded.

The study was a 5-year retrospective review examining children with a history of mild NIR to AMX or AMX/CL and underwent DPT to the culprit drug. NIR was defined as a reaction to the culprit drug occurring >1 h to 48 h. Diagnostic workup included delayed intradermal testing (IDT) to the culprit drug before DPT. Patients underwent a graded DPT (1/10 - 2/10 - 7/10 of the full dose [25 mg/kg] every 30 min) and observed for 2 h after last dose. Negative DPTs received a single full dose on the 2nd day and resumed twice daily dosing for 5 days. Lymphocyte transformation testing (LTT) was performed in DPT positive patients.

Of the 354 patients, 34 (9.6%) had prior history of NIR to AMX and 320 (90.4%) to AMX/CLV. Before the index reaction, 60% previously tolerated the culprit drug. The mean age of index reaction was 4.8 years, whereas the mean latency period between index reaction and DPT was 2.5 years. All patients experienced a skin eruption with the index reaction: delayed urticaria (48.6%), maculopapular exanthem (17.2%), macular exanthema (9.3%), papular exanthema (6.5%), angioedema (4%), undefined rash (4%), and scarlatiniform or morbilliform rash (1.1%). Only 4 children had gastrointestinal symptoms, and no one had respiratory symptoms with the index reaction. Delayed IDT was negative in all but one child who had a positive late reading; however, DPT was continued in this patient. Of the 354 patients, only 23 (6.5%) had a positive DPT (2 AMX; 21 AMX/CL). Eleven out of 23 reacted within 2 h to 8 h of the last DPT dose on either day 1 or day 2. Eleven out of 23 reacted at home 24 h to 48 h after the last dose on day 5. Only one out of 23 reacted within 30 min of the 1/10 DPT dose. All 23 positive DPT reactions were mild and self-resolving or only required antihistamines. Results reinforced the poor utility of delayed IDT in NIR to AMX or AMX/CL. LTT was performed only on 16 of the 23 DPT positive patients and only 6 of 16 (37.5%) were positive.

Given the overlap between immediate and NIR, it is appropriate to perform a graded DPT protocol in children experiencing reactions within 2 h to 6 h, whereas only conducting a single dose DPT for those reacting >6 hours. Collecting a thorough and accurate drug reaction history is crucial in determining the next steps for the DPT. Additionally, IDT and LTT did not predict reactivity. Even though the NPV for a prolonged DPT is almost comparable to the one for a single-day DPT, prolonged DPT is associated with a higher parental confidence about patient’s future intake of the culprit drug, when needed.

DPT remains the gold standard for confirmatory diagnosis of drug reactions. This retrospective study demonstrated that for mild NIR (>6 hours after last drug intake) to AMX or AMX/CL, a DPT with a single, 1-time therapeutic dose is safe to administer. Overall, it is important to personalize DPTs (graded versus single dose) to each patient’s case after risk-benefit evaluation based on a detailed reaction history.