Severity of Ocular Involvement as an Independent Clinical Marker for Prognosis, Complications, and Short-Term Mortality in Patients With Toxic Epidermal Necrolysis

Abstract

Purpose:: Toxic epidermal necrolysis is a life-threatening illness that involves sloughing of 30% or more of the total body surface area. In this spectrum of disease, the inflammation affects not only skin but also causes erosion of the mucous membranes of the eyes, mouth and genitalia. A severity of the illness rating called SCORTEN is considered the gold standard for predicting mortality for the illness. This score uses multiple laboratory tests and total body surface skin involvement to calculate risk of mortality, but the rating does not include mucosal involvement. Since mucosal involvement, specifically ocular involvement, frequently occurs, the purpose of this study is to determine if the severity of ocular involvement could also be used as a simple independent clinical marker for prognosis, complications, and short-term mortality of the disease.

Methods:: This is a retrospective chart review of patients with biopsy- proven TEN who were treated in the burn unit at Loyola Medical Center between 1997-2003. Ocular and systemic clinical information were recorded. Severity of ocular involvement was rated as none, mild (1+ injection or chemosis, no corneal involvement, eyelid skin sloughing), moderate (2+ injection or chemosis, presence of pseudomembranes, corneal staining, or corneal abrasion), or severe (3+injection, presence of membranes, symblepharon, or corneal ulcer) within 24 hours of presentation to burn unit.

Results:: Preliminary data on 30 patients met the inclusion criteria and data was collected. Seven patients died during hospitalization (23%). Most common ocular finding was eyelid skin involvement (19/30, 63%). Five patients (17%) had severe ocular involvement within 24 hours of admittance to the burn unit. When comparing the SCORTEN value to the severity of ocular involvement, the probability that the outcomes are not significantly different between SCORTEN value and the severity of ocular involvement value was 0.794 (using a one-tailed F-test). For systemic complications, there was weak correlation between conjunctival involvement and respiratory failure (0.21), as well as hospital mortality (0.27). There was no correlation between ocular involvement and development of sepsis (conjunctiva 0.04, cornea 0.03, respectively). Data collection will continue with the goal of 60 patients with ocular involvement in TENS.

Conclusions:: The severity of ocular involvement of patients with toxic epidermal necrolysis may be used as a simple clincial prognostic marker for hospital mortality.

Keywords: anterior segment