摘要
Purpose:To investigate which scoring system is the most accurate tool in predicting mortality among the infected patients who present to the emergency department in a middle-income country,and to validate a new scoring system to predict bacterial infections.Methods:This was a retrospective,single-center study among patients who were admitted via the emergency department of a public hospital.All patients who were started on antibiotics were included in the study,while patients aged<18 years were excluded.Data collected includeding patients'demographics,vital signs and basic laboratory parameters like white blood cell count and creatinine.The sensitivity and specificity of different scoring systems were calculated as well as their negative and positive predictive values.Logistic regression was used to derive a novel early warning system for bacterial infections.The area under the receiver operating characteristic(AUROC)was computed for each scoring model.Results:In total,109 patients were included in this study.The quick sequential organ failure assessment(qSOFA),search out severity and rapid acute physiology score had the highest AUROC(≥0.89)for predicting mortality,while qSOFA and universal vital assessment were the simplest scoring systems with an AUROC>0.85;however,these scoring systems failed to predict whether patients were truly infected.The INFECTIONS(short for impaired mental status,not conscious,fast heart rate,elevated creatinine,high temperature,on inotrope,low oxygen,high neutrophils and high sugar)model reached an AUROC of 0.88 to more accurately predict the infectious state of a patient.Conclusions:Middle-income countries should use the qSOFA or universal vital assessment score to identify the sickest patients in emergency department.The INFECTIONS score may help recognize patients with bacterial infections,but it should be further validated in multiple countries prior to widely use.
