Sensitivity and specificity of abbreviated burn severity index and revised Baux score in mortality prediction in burn patients admitted to a tertiary care center in South India

Abstract

Background: Mortality still remains the most significant adverse outcome encountered ever since the inception of burn management despite significant advances. Predicting mortality in the initial stages by mortality prediction models can help in many ways. Abbreviated Burn Severity Index (ABSI) and Revised Baux Score (RBS) are two commonly used models. However, these mortality prediction models have varied levels of sensitivity and specificity when applied in different clinical settings. Hence, validation in a particular clinical setting becomes important before using it. Our purpose was to understand the sensitivity and specificity of ABSI and RBS in mortality prediction in our tertiary care center. Methods: Data of all burn patients admitted into our tertiary care center from January 2018 to December 2020 were retrospectively collected. The ABSI and RBS scores were calculated for each patient. The scores between survivors and nonsurvivors were compared using statistical methods. Sensitivity and specificity of ABSI and RBS was calculated. The Receiver Operating Characteristic curve was plotted, and the area under the curve (AUC) was calculated to assess the accuracy of each model. Results: Three hundred and fifty patients were included in the study sample (Mean age: 27.3 years. Avg TBSA: 17.63%). Forty-seven patients had inhalational injury and 131 patients had full thickness burn. The mortality rate was 12.3% (43 patients). Sensitivity, specificity, and AUC for ABSI were 88%, 93%, and 0.96 and that for RBS were 74%, 96%, and 0.93, respectively. The optimal cutoff for ABSI and RBS was 7 and 85, respectively. Conclusion: Both ABSI and RBS mortality prediction models have acceptable levels of sensitivity and specificity in mortality prediction.