Risk factors and prediction model construction of bacterial infection after gastroscopy in patients with cirrhosis and gastroesophageal varices
10.3760/cma.j.cn311365-20250507-00137
- VernacularTitle:肝硬化食管胃静脉曲张患者胃镜治疗后细菌感染的危险因素及预测模型构建
- Author:
Xiangyu FANG
1
;
Chuan SHEN
1
;
Luyuan MA
1
;
Yuexia LIU
1
;
Chun LIU
1
;
Caiyan ZHAO
1
Author Information
1. 河北医科大学第三医院感染科，石家庄 050051
- Publication Type:Journal Article
- Keywords:
Liver cirrhosis;
Esophageal and gastric varices;
Bacterial infections;
Risk factors;
Gastroscopy treatment;
Prediction model
- From:
Chinese Journal of Infectious Diseases
2025;43(8):457-464
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To analyze the risk factors of bacterial infection in patients with liver cirrhosis complicated with gastroesophageal varices after gastroscopy, and to construct a prediction model.Methods:Patients with gastroesophageal varices due to cirrhosis who underwent gastroscopy in the Third Hospital of Hebei Medical University from January 2021 to May 2023 were enrolled. All patients were divided into infection group and non-infection group according to whether bacterial infection occurred after gastroscopy. The detection of pathogens in the infection group and the source of specimens were analyzed. Multivariate binary logistic regression was used to analyze the risk factors of postoperative bacterial infection in patients with cirrhosis and gastroesophageal varices. The nomogram was drawn by R language to construct a risk prediction model. Receiver operator characteristic curve (ROC curve), calibration curve, Hosmer-Lemeshow test and decision curve were used to evaluate the model.Results:Among the 480 patients, 57 had postoperative bacterial infection and 423 had no infection. The incidence of infection was 11.88%(57/480). Seventy bacterial culture positive samples were obtained, mainly from blood and respiratory tract (30 samples (42.86%) and 25 samples (35.71%), respectively). A total of 82 strains of pathogenic bacteria were isolated, including 16 strains of Escherichia coli and 14 strains of Staphylococcus aureus. Multivariate binary regression analysis showed that length of hospital stay (odds ratio ( OR)=1.13, 95% confidence interval ( CI) 1.06 to 1.20, P<0.001), age ( OR=1.06, 95% CI 1.02 to 1.10, P=0.006), model for end-stage liver disease combined with sodium (MELD-Na) score ( OR=1.10, 95% CI 1.02 to 1.18, P=0.014), diabetes ( OR=1.25, 95% CI 1.07 to 1.96, P=0.043) and emergency gastroscopy ( OR=2.95, 95% CI 1.20 to 7.25, P=0.019) were independent risk factors for bacterial infection after gastroscopy in patients with cirrhosis and gastroesophageal varices. Based on the above risk factors, a nomogram prediction model was constructed. The results of ROC curve analysis showed that the area under the curve of the nomogram model for predicting bacterial infection after gastroscopy of gastroesophageal varices in cirrhosis was 0.82 (95% CI 0.73 to 0.90). The slope of the calibration curve was 0.98(95% CI 0.92 to 1.04), indicating that the predicted probability of the model was in good agreement with the actual probability. The results of Hosmer-Lemeshow test showed that the nomogram model fitted well ( χ2=6.35, P=0.415). The decision curve analysis showed that the clinical net benefit rate of the nomogram model was >0 when the threshold probability was 0.039 to 0.410. Conclusions:Older age, length of hospital stay, MELD-Na score, history of diabetes, and emergency gastroscopy are independent risk factors for bacterial infection after gastroscopy in patients with cirrhosis and gastroesophageal varices. The prediction model constructed in this study has a good predictive value for bacterial infection in such patients.
