MORTALITY PREDICTION MODEL IN SEPSIS EMERGENCIES: COMBINATION OF BIOMARKERS AND CLINICAL PARAMETERS

Abstract

Sepsis is a medical emergency characterized by a systemic inflammatory response due to infection and can cause organ dysfunction leading to death. Identification of patients with high risk of mortality is a challenge in clinical practice. Predictive models based on biomarkers and clinical parameters can help in early detection and patient management. This study aims to develop a mortality prediction model in sepsis patients by analyzing a combination of inflammatory biomarkers such as procalcitonin (PCT), C-reactive protein (CRP), and interleukin-6 (IL-6) and clinical parameters such as Sequential Organ Failure Assessment (SOFA) score, serum lactate levels, and blood pressure. Data were obtained from sepsis patients at a referral hospital and analyzed using multivariate logistic regression methods to evaluate the relationship between independent variables and mortality. The results showed that the combination of biomarkers and clinical parameters provided better predictive value than single-based models. The developed prediction model had an accuracy of 87% with an area under the curve (AUC) of 0.91, indicating a high level of reliability in detecting patients at risk of death. Implementation of this model in clinical practice is expected to assist physicians in decision making and improve the prognosis of sepsis patients. Further research is needed for external validation and application of this model in various patient populations.