Analysis of discrimination power of prognostic scores in burn patients

Abstract

Introduction: Burns are a serious public health problem. Prognostic indexes estimate prognostic probability and help to quantify disease severity. Objective: To evaluate the power of discrimination of Abbreviated Burn Severity Index (ABSI), Acute Physiology and Chronic Health Evaluation II (APACHE II), Sequential Organ Failure Assessment (SOFA) and Therapeutic Intervention Scoring System 28 (TISS 28) in an Intensive Care Unit (ICU) from a burn center. Patients and Methods: This is a prospective longitudinal study, carried out from May 2011 to May 2013. Patients under 18 years of age and with hospitalization for less than 24 hours were excluded. Clinical and demographic data were collected, and the prognostic scores were calculated. The accuracy of the scores was evaluated by the Receiver Operating Characteristic (ROC) curve, which discriminated against the outcome of the patient (survivor and non-survivor). The area under the curve (AUC) was calculated. Results: A total of 180 patients were included, and 72.8% of them were male (n = 131). The median age was 40 years (IQR: 30 - 52.5). Hospital mortality was 37.2% (n = 67). The highest AUC was from the APACHE II score, with a value of 0.837 (associated criterion = 14, sensitivity = 83.6% and specificity = 72.3%). In the analysis of the scores between survivors and non-survivors statistically significant differences were found in all studied scores. Conclusions: ABSI, APACHE II, SOFA and TISS 28 at ICU resulted in good power of discrimination for survival among burn patients.