Abstract:Objective To explore the prognostic influencing factors of patients with multidrug-resistant Klebsiella pneumoniae (MDR-KP) nosocomial infection and to construct a nomogram model for prognosis prediction. Methods A total of 168 cases of MDR-KP nosocomial infection in a tertiary hospital in Beijing from December 2016 to December 2020 were selected and divided into death group (n=41) and survival group (n=127) according prognosis. Lasso logistic regression was performed to obtain the risk factors of higher correlation with prognosis of the patients, which was incorporated into the “rms” package based on R language for analysis and model construction. Results Lasso regression analysis showed that age ≥ 80 years, rescue history, ventilator use and blood transfusion 30 days before infection, surgical history 3 months before infection, chest puncture 30 days before infection and infection in the lungs and lower respiratory tract were the independent risk factors for death outcome in patients with MDR-KP. Based on the above risk factors, the corresponding nomogram model was constructed for prediction. The area under the ROC curve (AUC) in the training set and test set for modelling were 0.834(95%CI： 0.753-0.915) and 0.744(95%CI： 0.590-0.897), respectively. Conclusion The nomogram model for prognosis prediction of MDR-KP nosocomial infection has good test efficiency and goodness of fit, which helps to identify patients with poor prognosis in early clinical stage and take relevant clinical intervention measures as soon as possible.