RESUMO
BACKGROUND: Epidermal necrolysis (EN), comprising Stevens Johnson syndrome and toxic epidermal necrolysis is a rare and severe blistering reaction, mainly induced by drugs. Differences have been discussed between pediatric and adult patients regarding incidence, causes, and outcomes, but only based on a limited number of patients, in small case-series. OBJECTIVES: To directly compare incidence, causes and prognosis between adult and pediatric EN patients. METHODS: We used the French Health System Database from January 1st 2013 to December 31st 2022. We included adult and pediatric patients hospitalized for EN using the international classification of diseases, 10th revision codes combined with validated algorithms. The outcomes were incidence of EN; presence of a suspected drug before EN onset, defined by a new drug dispensation from 5 to 56â days before hospitalization; and in-hospital mortality. To estimate the association between pediatric EN and the presence of a suspected drug, we computed a multivariable logistic regression with odd ratios (OR). To estimate the association with mortality, we computed a multivariable Cox proportional hazard ratio (HR) model. RESULTS: A total of 1440 EN patients were included, with 799 (55.5%) females, comprising 219 children and 1221 adults. Among children, EN incidence was 1.5 cases (95%CI: 1.3-1.7) per million people year compared with 2.6 cases (95%CI: 2.5-2.7) in adults. Moreover, children had less chances to have a culprit drug before EN onset (93/219 (42.5%) versus 829/1221 (67.9%)), with an adjusted OR of 0.43 (95%CI: 0.32-0.59), p < 0.0001, together with a better prognosis, with death rates of 1.4% (95%CI: 0.4%-3.7%) in pediatric patients compared with 19.4% (95%CI: 17.3%-21.7%) in adult patients, and an adjusted HR of 0.12 (95%CI: 0.04-0.38), p = 0.0003 for in-hospital mortality. CONCLUSION: Pediatric EN seems to be rarer, with less chances to be caused by drugs, together with a better prognosis than adult EN. These results suggest the existence of different underlying pathophysiological mechanisms and clinical particularities between adult and pediatric patients.
RESUMO
BACKGROUND: Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) are rare life-threatening mucocutaneous reactions most often induced by drugs. To date, no large pharmacovigilance study has been conducted in the paediatric population. OBJECTIVES: To describe the spectrum of drugs associated with SJS-TEN in children through the analysis of cases reported in the WHO pharmacovigilance database (VigiBase). METHODS: Disproportionality study using data from VigiBase. All paediatric (age under 18 years) cases reported between January 1, 1967, and July 6, 2022, were included. For each molecule, a case-non-case study was performed to assess a potential pharmacovigilance signal by computing the lower end of the 95% credibility interval for the information component (IC025). We performed sensitivity analyses, (i) taking into account only cases reported by physicians and (ii) taking into account only cases reported in the last 10 years. RESULTS: Among 31,376,783 adverse drug reactions reported in VigiBase, 2,248,727 were paediatric cases and 7342 were encoded as paediatric SJS-TEN. Significant statistical pharmacovigilance signals were observed for 165 drugs. The two most represented drug classes were antiepileptics and anti-infectious drugs. The five drugs with the highest IC025 were lamotrigine (IC025 4.99), carbamazepine (IC025 4.88), phenobarbital (IC025 4.67), phenytoin (IC025 4.52) and nimesulide (IC025 4.23). Acetaminophen was significantly associated with paediatric SJS-TEN (IC025 2.85) and we also described various new suspected drugs. Vaccines had no significant pharmacovigilance signal. These results were confirmed with the sensitivity analyses. CONCLUSIONS: This study updates the spectrum of drugs potentially associated with paediatric SJS-TEN.