Comparison of Complication Rates of Central Venous Catheters Versus Peripherally Inserted Central Venous Catheters in Pediatric Patients.

OBJECTIVES: The purpose of our study is to compare the rate of central line-associated blood stream infections and venous thromboembolism in central venous catheters versus peripherally inserted central catheters in hospitalized children. There is a growing body of literature in adults describing an increased rate of venous thromboembolisms and similar rates of central line-associated blood stream infection associated with peripherally inserted central catheters versus central venous catheters. It is not known if the rate of central line-associated blood stream infection and venous thromboembolism differs between peripherally inserted central catheters and central venous catheters in children. Based on current adult literature, we hypothesize that central line-associated blood stream infection rates for peripherally inserted central catheters and central venous catheters will be similar, and the rate of venous thromboembolism will be higher for peripherally inserted central catheters versus central venous catheters.

DESIGN: This is a cohort study using retrospective review of medical records and prospectively collected hospital quality improvement databases.

SETTING: Quaternary-care pediatric hospital from October 2012 to March 2016.

PATIENTS: All patients age 1 day to 18 years old with central venous catheters and peripherally inserted central catheters placed during hospital admission over the study dates were included. Central venous catheters that were present upon hospital admission were excluded. The primary outcomes were rate of central line-associated blood stream infection and rate of venous thromboembolism.

INTERVENTIONS: None.

MEASUREMENTS AND MAIN RESULTS: Of 2,709 catheters included in the study, 1,126 were peripherally inserted central catheters and 1,583 were central venous catheters. Peripherally inserted central catheters demonstrated a higher rate of both infection and venous thromboembolism than central venous catheters in all reported measures. In multivariable analysis, peripherally inserted central catheters had increased association with central line-associated blood stream infection (odds ratio of 3.15; 95% CI, 1.74-5.71; p = 0.0002) and increased association with venous thromboembolism (odds ratio of 2.71; 95% CI, 1.65-4.45; p < 0.0001) compared with central venous catheters.

CONCLUSIONS: Rates of central line-associated blood stream infection and venous thromboembolism were higher in hospitalized pediatric patients with peripherally inserted central catheters as compared to central venous catheters. Our study confirms the need for further investigation into the safety of central access devices to assist in proper catheter selection.