Modeling Of Early Infection Prediction In Patients With Acute Pancreatitis SIRS

Background and Purpose:Patients with acute pancreatitis(AP)combined with systemic infammactery response syndrome(SIRS)have difficulty distinguishing between inflammation and infection.Prophylactic antibiotic therapy in patients with acute pancreatitis leads not only to antibiotic misuse but also to multidrug resistance and fungal infections in non-infected patients.However,appropriate antibiotic therapy will significantly improve outcomes in patients with acute pancreatitis infection at the right time.This study aimed to explore the risk factors for infection in patients with acute pancreatitis and systemic inflammatory response syndrome,and to accurately predict the probability of infection to improve the prognosis of patients with acute pancreatitis.Method:By screening all inpatients in the Department of Gastroenterology from January2020 to January 2022 in the case collection system of the First Affiliated Hospital of Nanchang University,all patients with acute pancreatitis were searched and SIRS patients in AP who met the criteria were screened.According to the presence or absence of infection,they were divided into the non-infected SIRS group and the infected SIRS group.The clinical data between the two groups were statistically described,compared between groups,and binary logistics regression analysis was performed to establish a prognostic model to predict the occurrence of SIRS infection in AP patients.The area under the curve(AUC)of the receiver operating characteristic(ROC)is used to evaluate the identification ability of the predictive model,and the accuracy and clinical application value of the model are evaluated by calibration curve and decision curve analysis(DCA).Internal validation was assessed using Bootstrap self-sampling.Outcome:A total of 500 AP patients with SIRS were included in the study without infection,with 209 male patients accounting for 77.7% and 60 female patients accounting for 22.3%.There were 159 male patients in the SIRS group,accounting for 68.8%,and 72 female patients,accounting for 31.2%.A total of 269 patients in the SIRS group without infection and 231 patients in the SIRS group with infection were analyzed,including neutrophil count,lymphocyte count,white blood cell count,neutrophil percentage and neutrophil and lymphocyte percentage(NLR)in both groups.There were significant differences in serum C-responsive eggs(CRP),procalcitonin(PCT),partial pressure of oxygen,and lactate dehydrogenase(LDH)(P<0.05).Reverse stepwise binary logistic multivariate analysis of factors affecting SIRS infection in acute pancreatitis include etiology,biliary origin,severe acute pancreatitis,APACEH-Ⅱ score,neutrophil-lymphocyte ratio(NLR),procalcitonin(PCT),and lactate dehydrogenase(LDH).Furthermore,DCA demonstrated the clinical utility of this predictive model.Conclusion:The new risk prediction model established in this study can accurately predict the incidence of infection in SIRS patients in acute pancreatitis,thereby helping clinicians decide anti-infection treatment.Improving the prognosis of patients with acute pancreatitis is conducive to the rational allocation of medical resources and is of great significance for tertiary prevention.Moreover,the predictive model is calibrated well and can predict the presence of infection in SIRS patients with acute pancreatitis.