Contains supporting figures & tables.

<div><p>Background</p><p>Though 60–80% of hospitalized patients have an intravascular device placed during hospitalization, there is a substantial risk of complication related to the placement, maintenance and removal of these devices. The objectives of this study were to describe vascular access device use, device complications and lumen dysfunction.</p><p>Methods</p><p>An observational cohort study was conducted over a 4.5 years period, in two quaternary, university-affiliated paediatric intensive care units (ICU). Eligible patients were admitted to either the paediatric or cardiac ICU and had one or more vascular access devices in-situ at the time of enrolment. Vascular access devices were defined as any external connection directly into the circulation including peripheral and central veins, arteries or bone marrow. Consistent with practice in the studied ICUs removal of peripheral catheters was regarded as due to dysfunction or complication.</p><p>Results</p><p>220 patients, 683 devices and 924 lumens were observed. The device complication rate was 21% and the lumen complication rate was 25%. The median duration without complication was 18 days for arterial catheters, 14 days for central venous catheters; 13 days for peripherally inserted central catheters and 4 days for peripheral intravenous catheters. On one third of all patient days, the volume of fluid administered to maintain VADs was equal to or greater than 20% of the total hourly total fluid intake.</p><p>Conclusion</p><p>Approximately 1 in 5 vascular devices had one or more complications in ICU, most devices continued to be used without removal, and some complications resolved. The implications of the fluids infused to maintain device function warrants further study, as do strategies to resolve or limit the their complications in children.</p></div>