Fatores de risco para infecção da corrente sanguínea laboratorialmente confirmada e infecção associada a cateter central em população neonatal

Abstract

Introduction: The use of central venous catheter (CVC) is associated with an increased risk of developing bloodstream infection in newborns. Knowing the risk factors associated with the infection and what may influence the increase of these rates during the insertion and maintenance of these devices can contribute to the implementation of strategies that guarantee the safety in the care of patients hospitalized in a neonatal unit. Objective: To investigate the risk factors for the occurrence of laboratory confirmed bloodstream infection in a neonatal unit Methods: This is a prospective cohort study with patients hospitalized in a neonatal unit of a university hospital, submitted to peripheral central venous catheterization, or not, from May 2014 to December 2016. Results: A total of 384 newborns, 192 with and 192 without central peripheral insertion catheters (PICC) were included. There was no difference in the incidence of ICSLC among the neonates of both groups (p = 0.99) in the multivariate analysis. Those who underwent surgical procedures had a 2.8-fold higher risk of ICSLC (95% CI 1.40-5.9, p = 0.005), and those receiving total parenteral nutrition (TPN) were 7 times 0.75 times more likely develop ICSLC (95% CI: 2.8-20.10, p <0.001). In addition, a clinical score of zero (without surgery or parenteral nutrition) had a negative predictive value of 97%, which may allow more informed decision-making regarding antibiotic therapy in newborns. When evaluating only 213 patients with PICC, it was observed that 230 devices were installed percutaneously with 3,123 day catheters, and 16 (7.5%) had more than one PICC. Logistic regression analysis found that patients who had PICC in the peripheral position on the radiograph of the segment where they were inserted were 2.38 times more likely to have ICSAC. With each increase of one day of the PICC residence time, the infection rate increases by 3 percent. The Hosmer and Lemeshow model fitting test was 0.63. The log rank test for comparison of the survival curves of patients who withdraw PICC or not had a p-value <0.001, indicating that there is an association between catheter withdrawal and infection occurrence, as well as in the adjusted Cox model (Hazard Ratio 18.26 [95% CI 7.98 to 41.76], p <0.001). Conclusion: Despite the care of the care team in the PICC insertion process, the choice of the site of access puncture and improvement of the technique so that the positioning is central to the radiography and the reinforcement of the daily verification of the need of the CVC could contribute to the reduction of ICSLC rates. Preventive measures, such as early enteral feeding, are recommended in order to reduce the patient's stay in the neonatal unit. In addition, adequated care during insertion and maintenance of PICC used for surgical procedures and NPT may decrease the risk of infection.