Pseudomonas aeruginosa blood stream infections: clinical and molecular epidemiology, virulence and outcome

Abstract

IntroductionPseudomonas aeruginosa is a virulent pathogen. It can exist both in the environment and the human host. Recent work by Turner et al suggests that it is the environment that P. aeruginosa exists that determines the virulence factors that are expressed [1]. P. aeruginosa bloodstream infections (BSI) are predominantly seen in the hospital or healthcare-associated host [2-5]. This BSI is no longer the rare entity of the original case series in the 1950’s, with increasing prevalence over the subsequent decades [6, 7]. The associated morbidity and mortality with this infection is considerable [2, 8-10]. This has not improved over time [7]. The antibiotic treatment options available, in the current age of global antibiotic resistance, has not kept pace. A mortality benefit of combination antibiotic therapy in the setting of P. aeruginosa BSI has not been consistently shown [11-17]. In the settings of drug resistance or the immunocompromised host, more effective ways to utilise current antibiotic therapy must be considered. It is therefore timely that we further characterise the clinical and molecular epidemiology, virulence genotype and outcome of P. aeruginosa BSI. MethodsBSI episode data was collected retrospectively over a three year time period from the first of January 2008 to the first of January 2011. This involved seven tertiary care institutions, servicing an urban population of 2.24 million. Extensive epidemiological, clinical, laboratory, treatment and outcome data was collected as per a pre-formulated data collection sheet (Appendix A). For purposes of the study on BSI in the setting of febrile neutropenia, data was collected from a single institution over an extended period of time from the 1st of January 2006 to the 1st of April 2014. Ethics approval was obtained from each of the study sites.A P. aeruginosa BSI isolate collection was also collated for study. This involved five public and private laboratories. Isolates were collected from the time period of the first of January 2008 to the first of January 2013. The laboratories service eight tertiary care institutions and one secondary care institution. Permission was obtained from each of the laboratories.Results595 P. aeruginosa BSI episodes were characterised from the study centres. 942 BSI isolates were collated and analysed as part of the BSI isolate collection. A retrospective cohort study of monomicrobial P. aeruginosa BSI described the recent Australian epidemiology of this BSI. The longitudinal mortality of this infection was found to increase with time and at one year was greater than would be expected for the comparative death rate predicted by the median Charlson’s co-morbidity index (CCI) of the cohort [18]. Detailed description of a retrospective cohort of community acquired (CAI) P. aeruginosa BSI gave new insight into the epidemiology of this group of patients. Multivariate analysis comparing CAI and health-care associated infection cohorts (HCAI) found that CAI is not associated with a shorter length of hospital stay, lower mortality rates or less antimicrobial resistance [19]. Review of patients who developed recurrent P. aeruginosa BSI gave insight into a previously uncharacterised cohort of patients. A recurrent BSI was found to place an individual at an increased risk for a further recurrence and had a higher mortality rate than the primary BSI episode [20]. Further work into the molecular epidemiology of this cohort revealed that the closer in time BSI relapse occurred, the more likely it was from the same clonal type. Molecular analysis of P. aeruginosa BSI isolates from a single institution in a non-outbreak setting demonstrated unexpectedly a polyclonal population structure [21]. This work points to ongoing hospital environmental exposure of the vulnerable inpatient to P. aeruginosa. Thus raising important infection control questions.Finally the virulence genes of five BSI isolates associated with rapid death in a non-neutropenic host were characterised. This allowed interpretation of the results in a clinical context. Almost complete conservation of virulence genes was seen. ConclusionWithin this thesis unique insights into P. aeruginosa BSI have been obtained. Looking forward, further work on the pathogenesis of the BSI in the human host and how this may potentially manifest as longer term mortality is required. Furthermore research looking at the metabolic pathways that drive the expression of genotypic virulence is also needed. Both these strategies may potentially lead to the development of non-antibiotic treatment options in the patient with a P. aeruginosa BSI. Looking more closely at the transmission pathways of this organism in the hospital setting that result in patient colonisation, may allow the implementation of preventative infection control strategies. Finally how we utilise the antibiotics we have available, depending on the host we are treating, is an area of research that requires ongoing attention.