BMC Medical Informatics and Decision Making (Oct 2022)

Accuracy of SCORTEN in predicting mortality in toxic epidermal necrolysis

  • Jerzy Strużyna,
  • Agnieszka Surowiecka,
  • Tomasz Korzeniowski,
  • Patrycja Korulczyk,
  • Lukasz Drozd,
  • Aldona Stachura,
  • Kamil Torres,
  • Andrzej Krajewski

DOI
https://doi.org/10.1186/s12911-022-02013-2
Journal volume & issue
Vol. 22, no. 1
pp. 1 – 7

Abstract

Read online

Abstract Background Toxic epidermal necrolysis (TEN) patients require multi-directional and multi-disciplinary treatment. In most cases, they are hospitalised at intensive care units and require multi-directional, burn-complication preventive care. Choosing the most appropriate treatment option might be troublesome even when predicting scores are used. SCORTEN is the most renowned prognostic score for TEN patients, however, there are some data indicating that the accuracy of this test may be limited. The credibility of not just the predicted mortality risk, but also componential laboratory results and clinical features subject to debate. The aim of this study was to evaluate the efficacy and credibility of SCORTEN in clinical practice, on proprietary material. Methods A retrospective analysis of 35 patients with diagnosed in histopathology TEN was performed. The inclusion criteria were as follows: day of submission before 5th day from the onset of the symptoms, full protocol of plasmaphereses and IVIGs according to our scheme. Our protocol includes cycle of plasmapheresis with frozen fresh plasma twice daily for the first 2 days following admission, and once daily for the subsequent 5 to 7 days. IVIGs were administered after the first two sessions of plasmapheresis, for 4 to 7 days. The dosage was calculated according to body weight, at 0.4 to 0.5 g/kg per dose. Results The sensitivity of SCORTEN for the analysed cohort was 100%, with a specificity of 24%. The estimated death was 41,9%, while the actual death rates were 12,5%. Our protocol improved the survival, OR = 26,57, RR = 6,34, p = 0,022. Decrease in mortality was caused by a combined treatment protocol we use- plasmaphereses with IVIGs. No independent risk factor was significant in death evaluation. Conclusion Our data suggest that the scoring system for predicting death among TEN patients are reliable when they are high. New prognostic factors should be found to improve the evaluation of patients with low SCORTEN.

Keywords