577 A Potential Association Between Remnant Bowel Anatomy and the Incidence of Catheter-Related Blood-Stream Infections (CRBSIS) in Adult Intestinal Failure (IF) Patients Depending in Home Parenteral Nutrition(HPN)

Abstract

Listen

Translate article

Patients with intestinal failure (IF) depend on parenteral support through a central line for survival, but are challenged by the risk of catheter-related bloodstream infections (CRBSIs). Employing the Copenhagen HPN database, we investigated the association between the remaining bowel anatomy and the incidence and infectious species in CRBSIs with the aim of identifying risk factors. Methods: The CopenhagenHPNdatabase is based on a retrospective annually review of all charts from adult patients, who have received HPN from Rigshospitalet, Denmark. The diagnosis of a CRBSI required clinical signs of a systemic infection and positive blood cultures, with the exclusion of other causes of infections. Results: From 1970 to 2010, 510 IF patients were discharged with HPN, in total contributing to 1745 HPN years. In 256 of the IF patients 873 CRBSI were detected: 595 mono-bacteraemia, 140 polybacteraemia, 85 fungemia and 49 combinations of bacteraemia and Candidemia. The species of bacteraemia and Candidemia were determined in 869 CRBSIs, while 4 positive cultures were recorded without species. The cohort were divided into three groups according to remain bowel anatomy (-colon, +colon and no-surgical). The overall incidence of CRBSIs in HPN were 1.37 per 1000 HPN days. In the group without colon, the incidence of CRBSIs were 1.54 per 1000 HPN days. The median remaining small intestine was 125 cm, and the median age at complication was 57.8 years. The group were subdivided dependent on the remaining small intestine ( 200cm) with minor variations in incidence (1.59, 1.69, 1.42 and 1.60 per 1000 HPN days). The CRBSI incidence in IF patients with remaining colon was 0.92 per 1000 HPN days. The median small intestine was 100 cm, and the patients had a median of 70.5% of remaining colon. The median age at complication in this group was 49.1 years. The colon-group was first split into two: less/ above 50% remaining colon with CBRSI incidences of 0.42 and 1.06 per 1000 HPN days. Staphylococcus spp were detected in 55 % of blood cultures in IF patients with colon, while only seen in 30 % in IF patients without remaining colon. In the no colon group enterobacteriaceae were more frequent (50 %) with the exception of IF patients with < 50 cm small intestine (24 %). Candidemia were detected in 14 % of the cultures from IF patients without colon, while only 6.8 % in IF patients with colon. Conclusion: The CRBSI incidence rate in IF patients without colon is almost two fold higher compared to IF patients with remaining colon. Most frequently IF patients with a colon had bacteraemia with Staphylococci, while CRBSI in patients without a colon as frequently were caused by enterobacteriaceae. Candidemia was most frequently seen in patients without colon. The role of remaining bowel on the incidence of CRBSI needs further investigation. Tabel 1: Incidence of CRBSIs in HPN years and 1000 HPN days.