Risk factors and prediction model construction for peripherally inserted central catheter-related infections.

OBJECTIVE: To establish a predictive modeling for the risk of bloodstream infection associated with peripherally inserted central catheter (PICC).

METHODS: Patients receiving PICC treatment in Shenzhen People's Hospital from June 2020 to December 2020 were retrospectively enrolled and divided into the infection group and the non-infection group according to the presence and absence of PICC-related infections. Then, relevant clinical information of patients was collected and the predictors of PICC-related infection were screened by the least absolute shrinkage and selection operator regression (LASSO) model. Besides, multivariate logistic regression was used to analyze the influencing factors of PICC-related infection, A nomogram was constructed based on the results of the multivariate analysis. Ultimately, a receiver operating characteristic (ROC) curve was plotted to analyze the application value of influencing factors to predict PICC-related infections.

RESULTS: A total of 505 patients were included, including 75 patients with PICC-related infections (14.85%). The main pathogen was gram-positive cocci. The predictors screened by LASSO included age >60 years, catheter movement, catheter maintenance cycle, insertion technique, immune function, complications, and body temperature ≥37.2 °C before PICC placement. Multivariate logistic regression analysis showed that independent risk factors of infections related to PICC included age >60 years [odds ratio (OR) = 1.722; 95% confidence interval (CI) = 1.312-3.579; P  = 0.006], catheter movement (OR = 1.313; 95% CI = 1.119-3.240; P  = 0.014), catheter maintenance cycle >7 days (OR = 2.199; 95% CI = 1.677-4.653; P  = 0.000), direct insertion (OR = 1.036; 95% CI = 1.019-2.743; P  = 0.000), poor immune function (OR = 2.322; 95% CI = 2.012-4.579; P  = 0.000), complications (OR = 1.611; 95% CI = 1.133-3.454; P  = 0.019), and body temperature ≥37.2 °C before PICC placement (OR = 1.713; 95% CI = 1.172-3.654; P  = 0.012). Besides, the area under the ROC curve was 0.889.

CONCLUSION: PICC-related infections are associated with factors such as age >60 years, catheter movement, catheter maintenance cycle, insertion technique, immune function, complications, and body temperature ≥37.2 °C before PICC placement. Additionally, the LASSO model is moderately predictive for predicting the occurrence of PICC-related infections.