Analysis of risk factors and construction of predictive model for carbapenem-resistant Klebsiella pneumoniae infection

Objective To explore the risk factors of carbapenem-resistant Klebsiella pneumoniae (CRKP) infection, construct and validate a risk predictive model.

Methods A total of 2 377 patients with Klebsiella pneumoniae infection were collected, with 2 117 cases excluded.Finally, 260 eligible patients were included and randomly divided into the training set of 195 cases and the validation set of 65 cases in a 3∶1 ratio.The training group was divided into a CRKP group with 72 cases and a carbapenem-sensitive Klebsiella pneumoniae (CSKP) group with 123 cases based on whether carbapenem was resistant.The model was constructed using R language and validated.

Results ALB decrease, indwelling urethral catheter, ventilator use for ≥7 days, ICU stay for ≥7 days, and combined use of antibiotics were risk factors for CRKP infection (P < 0.05).A risk prediction model was constructed and a nomogram was drawn using these 5 variables.The area under the ROC curve (AUC) of the training set was 0.864(95%CI: 0.814-0.915), and the AUC of the validation set was 0.855(95%CI: 0.763-0.947).The consistency of the calibration curve test results between the training set and validation set was good, and the Hosmer-Lemeshow fit was good.The clinical decision curve showed that the model had good prediction accuracy.

Conclusions The risk predictive model constructed in this study can provide reference for the clinical assessment of CRKP infection risk in patients, which has certain guiding significance for early diagnosis and treatment.