Evaluation of clinical outcomes and comparison of prediction models in the burn population hospitalized from the emergency department: Can burn mortality scores be used in a post-conflict area such as northwest Syria?

Abstract

Burns are a global health problem, especially in low- and middle-income countries. The use of models to predict mortality is more common in developed countries. In northern Syria, internal unrest has continued for 10 years. A lack of infrastruc-ture and difficult living conditions increase the incidence of burns. This study in northern Syria contributes to the predictions of health services provided in conflict regions. The first objective of this study specific to northwestern Syria was to assess and identify risk factors in the burn victim population hospitalized as emergencies. The second objective was to validate the three well-known burn mortality prediction scores to predict mortality: the Abbreviated Burn Severity Index (ABSI) score, Belgium Outcome of Burn Injury (BOBI) score, and revised Baux score. This was a retrospective analysis of the database of patients admitted to the burn center in northwestern Syria. Patients who were admitted to the burn center as emergencies were included in the study. Bivariate logistic regression analysis was performed to compare the effectiveness of the three included burn assessment systems in determining the risk of patient death. A total of 300 burn patients were included in the study. Of them, 149 (49.7%) were treated in the ward, and 46 (15.3%) in the intensive care unit; 54 (18.0%) died, and 246 (82.0%) survived. The median revised Baux scores, BOBI scores, and ABSI scores of the deceased patients were significantly higher than those of the surviving patients (p=0.000). The cut-off values for the revised Baux, BOBI, and ABSI scores were set at 105.50, 4.50, and 10.50, respectively. For predicting mortality at these cut-off values, the revised Baux score had a sensitivity of 94.4% and a specificity of 91.9%, and the ABSI score had a sensitivity of 68.8% and a specificity of 99.6%. However, the cut-off value of the BOBI scale, calculated as 4.50, was found to be low (27.8%). The low sensitivity and negative predictive value of the BOBI model suggest that it was a weaker predictor of mortality than the others. The revised Baux score was successful in predicting burn prognosis in northwestern Syria, a post-conflict region. It is reasonable to assume that the use of such scoring systems will be beneficial in similar post-conflict regions where limited opportunities exist.