RESUMO
OBJECTIVES: We aimed to identify the occurrence and risk factors for unplanned catheter removal due to catheter-associated complications and the effects on catheter survival probability in a PICU. DESIGN: Retrospective, single-center, observational study of cases involving conventional central venous catheters or peripherally inserted central venous catheters. SETTING: The PICU of a tertiary children's hospital. PATIENTS: Consecutive PICU patients with central venous catheters between April 2016 and February 2019. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: We identified unplanned catheter removals that were related to central line-associated bloodstream infection, thrombosis, and mechanical complications. During the study period, 582 central venous catheters and 474 peripherally inserted central venous catheters were identified. The median durations of catheter placement were 4.0 days for central venous catheters and 13.0 days for peripherally inserted central venous catheters (p < 0.001), and unplanned catheter removals due to catheter-associated complications were in 52 (8.9%) central venous catheter cases and 132 (27.8%) peripherally inserted central venous catheter cases (p < 0.001) (15.0 and 16.0 per 1,000 catheter-days, respectively [p = 0.75]). Unplanned catheter removal was associated with a peripheral catheter tip position among both central venous catheters and peripherally inserted central venous catheters (p < 0.001 and p = 0.001), and it was associated with surgical patient status among peripherally inserted central venous catheters (p = 0.009). In contrast, the use of ultrasound-guided insertion was associated with a lower occurrence of unplanned catheter removal among peripherally inserted central venous catheters (p = 0.01). With regard to catheter survival probability, there was no significant difference between central venous catheters and peripherally inserted central venous catheters (p = 0.23). However, peripherally inserted central venous catheters had a lower occurrence of central line-associated bloodstream infection than central venous catheters (p = 0.03), whereas there was no significant difference in the rates of thrombosis (p = 0.29) and mechanical complications (p = 0.84) between central venous catheters and peripherally inserted central venous catheters. CONCLUSIONS: In a PICU, peripherally inserted central venous catheters had lower occurrence of central line-associated bloodstream infection than central venous catheters; however, similar catheter survival probabilities were observed between both catheters. A central catheter tip position for both catheters and ultrasound-guided insertion for peripherally inserted central venous catheters may help limit unplanned catheter removal due to catheter-associated complications.