Pump infusion pressures in patients on long term parenteral nutrition: a novel approach and monitoring technique to anticipate catheter occlusion.

Abstract

equivalent to 4.46 infections/1000 PN days. Comparison of baseline statistics between patients with and without confirmed PN-aCVC sepsis revealed no significant difference between age, pre-feed BMI or baseline albumin. PN-aCVC sepsis positive patients had a longer feed length (36.67 days vs 13.54 days, p ¼ 0.0007). The table demonstrates multivariate logistic regression for RR of line sepsis given various patient characteristics. Six month mortality in the entire cohort was 27%. Age above 65 (RR 3.17, 95%CI 1.05 e 9.86, p ¼ 0.05) and baseline albumin less than 20 (RR 5.21, 95%CI 1.19-22.91, p ¼ 0.02) was associated with increased risk of mortality, although confirmed PN-aCVC sepsis was not related to increased mortality. In summary, PN-aCVC sepsis is not an infrequent occurrence in patients receiving short-term PN. Feed length > 14 days, feed received via PICC line, and patient location in ITU have been isolated as potential predisposing factors for this outcome. With this in mind, the aim is that high risk patients are identified earlier, and that vigilance for suspected line sepsis is raised. Although overall mortality is high in the PN cohort, this is probably a reflection of the type of patient receiving PN at our centre (tertiary referral centre for complex upper GI cancer and HPB surgery) rather than the rates of line sepsis. 1. Catheter application, insertion vein and length of ICU stay prior to insertion affect the risk of catheter-related bloodstream infection'.van der Kooi et al.