Favourable outcomes in episodes of Pseudomonas bacteraemia when associated with tunnelled cuffed catheters (TCCs) in chronic haemodialysis patients

Background. Pseudomonas is regarded as a particularly lethal bacterial isolate. High mortality rates have been reported in episodes of Pseudomonas sepsis when associated with visceral infections as seen in immunosuppressed, hospitalized patients. In comparison, lower mortality rates have been reported with catheter-associated Pseudomonas bacteraemia in non-dialysis patients. The purpose of this study was to determine the risk factors and the outcomes for episodes of Pseudomonas bacteraemia associated with the use of tunnelled cuffed haemodialysis catheters (TCCs) in a chronic out-patient population. Methods. We performed a prospective observational study in seven chronic haemodialysis units over a 2.5 year period. Patients who were diagnosed as having their initial TCC-associated bacteraemia within the study period were followed for 3 months. All episodes of Pseudomonas TCC bacteraemia were identified, and univariate analyses were performed to compare Pseudomonas bacteraemia with non- Pseudomonas bacteraemia. Results. During the study period, 219 episodes of TCC bacteraemia were identified; 18 had a Pseudomonas isolate (8%). Pseudomonas bacteraemia episodes were associated with a significantly higher risk of not receiving appropriate initial antibiotics (odds ratio = 3.6, P = 0.02). There were no deaths in the Pseudomonas bacteraemia group, whereas 19% died in the non- Pseudomonas group. The TCC was removed in 89% of Pseudomonas bacteraemias. There were no significant risk factors for acquiring a Pseudomonas isolate, and no difference in recurrent bacteraemia or infectious complication rates between the groups. Conclusions. In haemodialysis patients with a TCC-associated Pseudomonas bacteraemia, outcomes are remarkably good. This may be because the source of Pseudomonas infection was removed in most cases. Initial antibiotic coverage lacking anti- Pseudomonas activity was not associated with increased mortality.