P077 Toxic Epidermal Necrolysis-Optimal Management (TEN-OM): a retrospective cohort analysis from two centres

TEN-OM (Toxic Epidermal Necrolysis-Optimal Management) is a retrospective dual-centre cohort analysis of adult patients with toxic epidermal necrolysis (TEN) admitted from 2010 to 2023. It is a multidisciplinary project including dermatology, plastic surgery, pharmacy, ophthalmology and gynaecology. The objective of this study is to describe a range of culpable drug classes in the aetiology of TEN, demographic differences, clinical phenotypes and outcomes in patients hospitalized for TEN. Thirty-five cases were identified from clinical records and met the criteria for inclusion in a semiquantitative analysis. The majority of patients were White (n = 18) and female (n = 24). The median age was 53 years [interquartile range (IQR) 30.5–62.5]. The median length of admission was 21 days (IQR 14–32). The median SCORTEN was 2 (n = 23). Antibiotics constituted the largest category of culpable drugs (n = 10), including co-trimoxazole in three cases. All four cases in the antiepileptic category were attributed to lamotrigine. The median time to re-epithelization was 7 days (IQR 7–9.5). The mortality rate was 26% (n = 9). Deceased patients had involved body surface area (BSA) of ≥ 60% and a SCORTEN of 3–5. Most patients received intravenous immunoglobulin (n = 27, 77%), and 33% (n = 9) of those patients followed the Chelsea Protocol (2 g kg−1 on day 1; then 1 g kg−1 on days 2 and 3). Other systemic treatments included ciclosporin (n = 13, 37%), filgrastim (granulocyte colony-stimulating factor; n = 12, 34%), etanercept (n = 4, 11%) and systemic steroids (n = 11, 32%). Most patients had more than one treatment simultaneously. More than half of the female patients had vulval involvement (n = 16, 67%), but only two patients required dilators. The median time for gynaecological opinion was 6 days from admission. Male urethral examination was documented in only one case. Eye involvement tended to lag up to 7 days behind initial skin signs. This was observed in 74% patients (n = 26), and nearly one in four patients (23%, n = 8) required amniotic membrane transplant. Patients requiring amniotic membrane transplant had ≥ 60% BSA involved, but admission SCORTEN ranged between 0 and 4. This suggests that SCORTEN or BSA alone are not reliable predictors of eye involvement, which might be related to drug pharmacokinetics such as half-life. To date, this is the largest retrospective cohort analysis of patients with TEN in the UK. It highlights the need for a multidisciplinary and collaborative approach in management of TEN, with a particular focus on ocular sequalae. Quantitative analysis was not performed due to data paucity, but there was a possible correlation between nadir serum albumin and length of stay. Considering emerging evidence for the role of etanercept in management of TEN, there is an acute need for a prospective, multicentre register to determine treatment efficacy.