Application of risk nomogram model in the prevention and control of carbapenem-resistant Klebsiella pneumoniae infection
OBJECTIVE To explore the risk factors of carbapenem resistance in Klebsiella pneumoniae(KP)pa-tients,and to construct the nomogram of prediction model to provide reference for the prevention and control of resistant-carbapenem in KP patients.METHODS Patients hospitalized in a class A tertiary hospital in Guangxi from Jun 2020 to Dec 2022 were recruited,and randomly divided into the training set(n=1 484)and validation set(n=636)at a ratio of 7∶3.According to the results of drug susceptibility test,the training set was divided in-to the drug-resistant group(n=260)and non-resistant group(n=1 224).Clinical data of patients were collected,and the differences between the two groups were compared by univariate analysis.Multivariate logistic regression was used to analyze the risk factors of drug resistance to carbapenem in KP patients.The nomogram for risk pre-diction was constructed,and the predictive efficacy of the prediction model for the occurrence of CR-KP infection in inpatients was evaluated by receiver operating curve(ROC),calibration curve,and decision curve analysis.RESULTS Sputum and urine samples were the main sources of Klebsiella pneumoniae.Multivariate logistic re-gression analysis showed that days of hospitalization prior to the detection of pathogens,respiratory diseases,kid-ney diseases,nervous system diseases,hematological diseases,use of antibiotics in combination,use of ventila-tor,and central venous cannula were risk factors for CR-KP infection.The area under curve(AUC)value of the prediction model in the training set was 0.800(95％CI:0.769-0.831),the average absolute error of calibration curve was 0.018,and the probability range of decision curve of analysis(DCA)threshold was 5％-55％.The AUC value of the model in the verification set was 0.827(95％CI:0.783-0.871),and the average absolute error of the calibration curve was 0.017,with the probability of the DCA threshold ranging from 5％to 57％.CONCLUSION The risk of CR-KP infection is higher in hospitalized patients with long hospitalization,underlying diseases and combined use of antibiotics.The risk prediction model established by nomogram can better predict the risk of CR-KP infection.
Carbapenem resistant Klebsiella pneumoniaeRisk factorInfection prevention and controlNomo-gram modelRisk predictionInpatient
NETL
NSTL
万方数据
