Conference: 2008 International PHA Conference and Scientific Sessions
Release Date: 06.20.2008
Presentation Type: Abstracts
Ivy D.1, Doran A.1, Calderbank M.1, Dolan S.3, Nyquist A.2,3
1. Section of Pediatric Cardiology, University of Colorado Denver School of Medicine
2. Pediatric Infectious Disease, University of Colorado Denver School of Medicine
3. Department of Epidemiology, The Children’s Hospital, Aurora, CO, USA
BACKGROUND: Intravenous epoprostenol and treprostinil have proven effective therapies for many patients with pulmonary arterial hypertension; however, these continuous infusion therapies carry the risk of catheter-related bloodstream infections (CR-BSIs). The catheter hub is the most common port of bacterial contamination. Following a significant increase in gram negative CR-BSIs in patients receiving treprostinil, a closed hub catheter connection as well as water-proof protection of all connections of the infusion system during showering was implemented and evaluated at one center.
METHODS: Intravenous epoprostenol was used exclusively until July 2004 and subsequently a combination of intravenous epoprostenol and treprostinil was used. Implementation of the closed hub system with the split-septum BD Q-Syte ® and sealing connections during showering for all patients was initiated in July 2007. Incidence and type of CR-BSIs for all patients at The Children’s Hospital on intravenous epoprostenol and intravenous treprostinil was collected during two time periods: from January 2003 until July 2007, at which time the change to a closed hub system and water-proof connection was made; and then until February 2008. We evaluated overall CR-BSI rate before and after the intervention as well as the type of bacterial infection.
RESULTS: 48 patients were treated with intravenous prostanoid therapy during the study period for a total for 37,179 treatment days (epoprostenol 23,506, treprostinil 13,673). From January 2003 to February 2008, there were 0.94 cases of CR-BSI per 1000 treatment days. During the entire study period, the infection rate was 0.55/1000 days for epoprostenol and was 1.61/1000 days for treprostinil (p< 0.01). Patients treated with treprostinil following the implemented changes had a significant decrease in CR-BSI rate (1.95 vs 0.00 per 1000 treatment days, p = 0.03). Patients treated with intravenous treprostinil had a greater likelihood of gram negative BSI than those treated with epoprostenol (1.02 vs 0.09 per 1000 treatment days, p < 0.01). Since closed hub implementation, there have been no cases of gram negative CR-BSIs and only one case of gram positive infection in a patient treated with intravenous epoprostenol.
COMMENTS AND CONCLUSIONS: The closed hub system and maintaining dry connections significantly reduced the incidence of BSI in patients on treprostinil, including a significant reduction in gram negative infections. We speculate that protection of the catheter hub connection may be more critical with treprostinil versus epoprostenol due the neutral pH (7.4 vs 10.4). Patients receiving intravenous prostanoids suspected of a BSI should receive antibiotic coverage for gram positive and gram negative bacteria until identified. Continued follow-up and further investigation is warranted regarding incidence of BSI in all patients on prostanoid therapy, use of the closed hub system, and improvements in the delivery system to prevent contamination.