I M M H Oen1, C H van der Vlies1,2, Y W F Roeleveld1, J Dokter1, M J Hop3, M E van Baar3. 1. Burn Centre, Maasstad Hospital, Rotterdam, the Netherlands. 2. Department of Surgery, Maasstad Hospital, Rotterdam, the Netherlands. 3. Association of Dutch Burn Centres, Burn Centre, Maasstad Hospital, Rotterdam, the Netherlands.
Abstract
BACKGROUND: Recently, the importance of reporting the results and principles of management in Toxic epidermal necrolysis (TEN) patients was underscored. Treatment of TEN focuses on supportive care, often provided in a burn centre setting. Mortality in TEN patients can be high; the SCORTEN score is a scoring system that predicts mortality in patients with TEN. The predictive value of the SCORTEN score in our setting is unclear, as are the treatment costs of TEN patients. OBJECTIVE: To describe patient characteristics, treatment, outcome and direct medical costs of patients with TEN treated in one Dutch burn centre in a 27-year period. In addition, determinants of mortality and the predictive value of the SCORTEN score were assessed. METHODS: A retrospective study was conducted in all patients with TEN (including Stevens-Johnson syndrome (SJS) and overlap SJS-TEN) admitted to the burn centre Rotterdam between January 1987 and December 2013. The discriminative value of the SCORTEN score was assessed by receiver-operator characteristics curve analysis. RESULTS: A total of 63 patients were admitted in 27-year period. Overall mortality was 39.7%, mortality in TEN patients (>30%TBSA) was 37.1%. A higher age (OR = 1.04, 95%CI: 1.02-1.07) and comorbidity (OR = 4.25, 95%CI: 1.2-14.7) were associated with mortality. The discriminative value of the SCORTEN prediction model in our population was limited (AUC=0.72, 95%CI: 0.57-0.86). The mean direct medical hospital-based costs was €41.361. CONCLUSION: Toxic epidermal necrolysis is a severe adverse drug reaction, with a high mortality. Elderly patients and patients with comorbidity, especially circulatory comorbidity, have a relatively high risk of decease. The SCORTEN score, a frequently used prediction model in patients with TEN, underestimated the mortality in our study, mainly due to limited availability in patients with a good prognosis. The treatment of patient with TEN is associated with high direct medical hospital-based costs, also compared to burn patients in general.
BACKGROUND: Recently, the importance of reporting the results and principles of management in Toxic epidermal necrolysis (TEN) patients was underscored. Treatment of TEN focuses on supportive care, often provided in a burn centre setting. Mortality in TEN patients can be high; the SCORTEN score is a scoring system that predicts mortality in patients with TEN. The predictive value of the SCORTEN score in our setting is unclear, as are the treatment costs of TEN patients. OBJECTIVE: To describe patient characteristics, treatment, outcome and direct medical costs of patients with TEN treated in one Dutch burn centre in a 27-year period. In addition, determinants of mortality and the predictive value of the SCORTEN score were assessed. METHODS: A retrospective study was conducted in all patients with TEN (including Stevens-Johnson syndrome (SJS) and overlap SJS-TEN) admitted to the burn centre Rotterdam between January 1987 and December 2013. The discriminative value of the SCORTEN score was assessed by receiver-operator characteristics curve analysis. RESULTS: A total of 63 patients were admitted in 27-year period. Overall mortality was 39.7%, mortality in TEN patients (>30%TBSA) was 37.1%. A higher age (OR = 1.04, 95%CI: 1.02-1.07) and comorbidity (OR = 4.25, 95%CI: 1.2-14.7) were associated with mortality. The discriminative value of the SCORTEN prediction model in our population was limited (AUC=0.72, 95%CI: 0.57-0.86). The mean direct medical hospital-based costs was €41.361. CONCLUSION: Toxic epidermal necrolysis is a severe adverse drug reaction, with a high mortality. Elderly patients and patients with comorbidity, especially circulatory comorbidity, have a relatively high risk of decease. The SCORTEN score, a frequently used prediction model in patients with TEN, underestimated the mortality in our study, mainly due to limited availability in patients with a good prognosis. The treatment of patient with TEN is associated with high direct medical hospital-based costs, also compared to burn patients in general.