Distribution analysis of positive and negative pathogenic bacteria in patients with acute pancreatitis and the clinical characteristics and model prediction analysis of positive infection bacteria.

Background: There were bacteria in the early pancreatic juice culture of severe acute pancreatitis (SAP) patients, but during the clinical time, some patients showed more positive bacteria and some patients showed more negative bacteria. Many scholars have different test results, and further clinical research needs to be carried out to clarify this fact. To determine evidence of infection in the early stage of acute pancreatitis (AP) by pancreatic juice bacterial culture and provide a reference for the anti-infective therapy of AP. Methods: Patients with AP who underwent pancreatic juice bacterial culture in the Department of hepatobiliary surgery of the General Hospital of Ningxia Medical University from January 1, 2019 to June 30, 2020were reviewed. Endoscopic retrograde cholangiopancreatography (ERCP) was used to collect pancreatic juice, which was sent to the laboratory for culturing. The clinical data and bacterial culture results of the patients were then recorded and analyzed. According to the results of the pancreatic juice culture, the patients were divided into a positive bacterial culture group (n=64) and a negative bacterial culture group (n=92). It was compared the data results of two groups [age, gender, etiology, acute physiology and chronic health evaluation (APACHE) II score, cultured bacteria, complications, local complications, Balthazar computed tomography (CT) score, inflammatory factors, the use of antibiotics, drug sensitivity analysis results, and the patient's co-infection] and performed multivariate analysis to identify the clinically valuable indicators. Moreover, a receiver operating characteristic (ROC) curve was drawn to predict the model of positive pancreatic juice culture in AP. Results: The patients in the positive bacterial culture group and the negative bacterial culture group had statistically significant differences in gender, age, body mass index (BMI), amylase, white blood cell count and the two groups of patients were comparable. A total of 156 patients were included in the study and pathogenic bacteria were cultured in the pancreatic juice of 64 patients (41.03%) and 94 strains of bacteria were found (Gram-positive bacteria, 38.30%; Gram-negative bacteria, 58.51%; fungi, 3.19%). A history of ERCP and early pancreatic necrosis were independent influencing factors of positive pancreatic juice culture. The incidence of complications, APACHE II, and inflammatory factor levels of patients with positive pancreatic juice bacterial culture were significantly higher than those of negative pancreatic juice bacterial culture (P<0.05). Multivariate regression and the ROC curve of pancreatic infection showed that positive pancreatic and Balthazar CT score >7 on admission were independent risk factors of pancreatic. The area under the ROC curve of patients with later pancreatic infection was 0.863 [95% confidence interval (CI): 0.769-0.957], specificity was 65.30%, sensitivity was 90.50%, and the Youden index was 0.603. Conclusions: Bacterial culturing of pancreatic juice provides evidence of infection in the early stage of AP, which has certain significance for the anti-infective therapy of AP.