Pediatric Severe Cutaneous Adverse Drug Reactions (SCARs) clinical characteristics and comparison of causality assesment tools and severity scoring systems in relation to clinical outcomes from a single center experience

Background: Severe cutaneous adverse reactions (SCARs) such as drug reaction with eosinophilia and systemic symptoms (DRESS), Stevens–Johnson syndrome (SJS), and toxic epidermal necrolysis (TEN) are rare but life-threatening in children. Pediatric data on causative drugs, immunologic features, and outcomes remain limited. Objective: To compare clinical, laboratory, and immunologic findings, causality tools, and severity scoring systems in pediatric DRESS and SJS/TEN. Methods: We retrospectively analyzed 23 children (<18 years) with SCAR at a tertiary center (December 2022–April 2025). DRESS and SJS/TEN were diagnosed using RegiSCAR and consensus criteria, respectively. Causality was evaluated by the World Health Organization–Uppsala Monitoring Centre (WHO–UMC) and Naranjo algorithms for DRESS, and the Algorithm of Drug Causality in Epidermal Necrolysis (ALDEN) for SJS/TEN. Disease severity in DRESS was evaluated using both the Mizukawa severity score and the EAACI 2025 Pediatric Position Paper (PP) severity grading, while SJS/TEN severity was assessed using Severity-of-Illness Score for Toxic Epidermal Necrolysis (SCORTEN). Clinical features, laboratory findings, treatment strategies, and outcomes were compared. Results: DRESS constituted 69.6% of cases. Antibiotics predominated in DRESS (68.7%), and antiepileptics in SJS/TEN (57.1%). Causality assessment yielded predominantly Probable classifications in DRESS (69.2%), while aromatic anticonvulsants scored Very probable in SJS/TEN by ALDEN and severity grading aligned with clinical outcomes. In DRESS patients, severity stratification by the Mizukawa score predominantly classified cases as moderate, guiding systemic corticosteroid use. When reclassified according to the EAACI 2025 PP grading, several patients with limited extra-hepatic involvement were upgraded to higher severity categories, despite favorable clinical response to corticosteroid monotherapy. In SJS/TEN, SCORTEN scores aligned with overall clinical severity but did not correlate with ocular involvement. Intravenous immunoglobulin was administered more frequently in SJS/TEN than in DRESS (85.7% vs 12.5%, p=0.001). No mortality occurred. During follow-up, immune-mediated complications were observed exclusively in the DRESS group, including autoimmune thyroiditis in one patient, autoimmune thyroiditis with concurrent autoimmune neutropenia in another, and chronic urticaria in one patient, whereas no immune-mediated sequelae were detected in the SJS/TEN group. Conclusion: Pediatric DRESS and SJS/TEN display distinct clinical, immunologic, and severity profiles. Early CD19⁺ B-cell lymphopenia appears characteristic of DRESS and may support severity stratification and long-term autoimmune surveillance. Comparative application of the Mizukawa score and EAACI 2025 PP grading highlights potential differences in severity classification that may influence treatment decisions. Integrating standardized causality and severity tools may improve phenotype-specific management of pediatric SCARs.