LAUSR

Introduction: Central venous catheters (CVC) are often required in critical care settings to provide life-sustaining care. Their usage has improved the quality of care in pediatric patients by providing a secure point of access which minimizes the number of peripheral needle sticks. However, CVC usage is not without risks. Children can experience acute CVC-related complications such as venous thromboembolisms (VTEs) and infections. Over the past 20 years, CVC utilization has increased and its associated complications (VTE, central line associated blood stream infection [CLABSI], catheter dysfunction) result in increased lengths of stay and hospital costs. Clinical studies identify CVCs as the single most important risk for DVT in children. Incidence rates in critically-ill children are widely divergent with historical rates of 2.7-18% and 24.7-32.5 per 1000 catheter days. In addition, other historical rates for another CVC-related complication, CLABSI, are limited but demonstrate a rate of 1.21-6.5% and 14.4 per 1000 catheter days. Prevention of catheter-related thrombosis with prophylactic anticoagulants, such as unfractionated heparin, low molecular weight heparin or warfarin, have been tested in several studies in specific subgroups of patients and the majority of studies in pediatric populations were not able to clearly demonstrate a benefit. Despite this, nearly 50% of surveyed Pediatric Intensive Care Unit (PICUs) utilize various forms and doses of heparin prophylaxis to prevent CVC-associated complications. The Penn State Health Children's Hospital PICU utilizes low dose unfractionated heparin infusions (LDUFHI) at 5-15 units/kg/hr for catheter patency and for the prevention of CVC associated complications. Methods: A retrospective chart review of all CVCs, which include non-tunneled CVC (NT-CVC), tunneled lines (TL) and peripherally inserted central catheters (PICC), utilized in the PICU at The Penn State Health Children's Hospital in 2015 was performed to determine if the usage of LDUFHI in critically ill children to maintain catheter patency results in a lower incidence of VTE and CLABSI without increasing bleeding complications in comparison to historically reported data (Beck et al.,J Pediatr. 1998;133(2):237-241; Faustino et al.,J Pediatr. 2013;162(2):387-391; Krishnaiah et al.,Pediatr Crit Care Med. 2012;13(3):e176-e180.). Results: We identified 178 subjects who had CVC which received LDUFHI in 2015. There were a total of 228 CVCs with the majority being non-tunneled CVCs (54%) followed by tunneled lines (34%) and PICCs (12%). Combined, these catheters resulted in 2086 days of use. There was a total of 5 VTEs events for an overall incidence rate of 2.19% and 2.39 per 1000 catheter days. NT-CVC had the lowest rate (1.63%) followed by TL (2.56%) and PICCs (3.7%). However, PICCs has a slightly lower rate when measuring by catheter days (NT-CVC = 2.2 per 1000 catheter days; TL = 4.01 per 1000 catheter days; PICC = 1.46 per 1000 catheter days). There were 3 CLABSI events for an incidence rate of 1.32% and 1.43 per 1000 catheter days (NT-CVC = 1.63% and 2.2 per 1000 catheter days; TL no events; PICC = 3.7% and 1.46 per 1000 catheter days). There were no episodes of heparin induced thrombocytopenia or major bleeding. Overall, the VTE and CLABSI incidence rates are dramatically lower than historical comparisons (VTE = 2.19% and 2.39 per 1000 catheter days versus 15.8-18% and 24.7-32.5 per 1000 catheter days; CLABSI = 1.32% and 1.43 per 1000 catheter days versus 6.5% and 14.4 per 1000 catheter days). In addition, this study can offer more accurate rates as incidence was also measured as a factor in catheter days. Conclusion: This study used a single tertiary pediatric intensive care site in a non-randomized fashion. These data suggest an overall lower incidence of catheter related VTE and CLABSI in critically-ill children who received LDUFHI when compared to published with comparable pediatric study populations. The incidence of complications such as CLABSI were also lower with LDUFHI usage when compared to published data involving critically ill children. The incidence rates were even lower than recently published data involving non-critically ill children. These findings suggest that a multicenter randomized study should be pursued. No relevant conflicts of interest to declare.