Marriott D 1 , Sorrell T 2 , Slavin M 3 , Chen S 2 , Ellis D 4

Author address: 



A recent comprehensive American study demonstrated that Candida spp. are now the fourth most common agent responsible for nosocomial bloodstream infection. Of those patients who develop nosocomial candidiasis, 35% will die from the infection and 30% will die from the underlying disease. Risk factors for the development of candidiasis and appropriate antifungal therapy depend on the infecting species. A number of studies have shown that predominant species vary between geographic regions and even centres within one city. To date there are no Australian data on the incidence, geographic distribution and infecting species in nosocomial candidiasis. We initiated a study over a three-year period to define the epidemiology, clinical features and outcomes associated with candidaemia in public and private health care institutions. Clinical data, including clinical presentation, treatment and outcome, were collected five days and thirty days (or at hospital discharge) after the initial diagnosis was made. All organisms isolated were referred to a reference laboratory for speciation and antifungal susceptibility testing. In addition, whole blood was obtained in some instances to allow the development and assessment of antigen and antibody assays to detect invasive fungal infections. In the first eighteen months of the study, 331 cases of candidaemia were identified nationally from 39 of the 61 participating centres. All States and Territories of Australia were represented by at least one case. The majority of cases were healthcare associated-inpatient infections (n=244, 82.0%). Vascular access devices were the main source of the infection (n=244, 50.4%), however there were also a high proportion of cases originating from an unknown source (26.6%). Other risks factors identified in previous studies such as admission to Intensive Care Units, the use of broad spectrum or multiple antimicrobial agents and total parenteral nutrition are supported in this study. For discharged patients, the crude mortality rate was 40.0% (n=170) with 40.0% of deaths attributable to candidaemia. Of particular interest was the small percentage of immunocompromised patients and HIV/AIDS patients with candidaemia. These two groups were considered at risk but their low representation may be attributable to the widespread use of prophylactic antifungal agents. Patients with complicated medical or surgical gastrointestinal disorders had the highest risk for developing candidaemia. This has important consequences for prophylaxis and pre-emptive therapy. C.albicans (n=169, 57.4%) was the predominant species isolated from the blood with C.parasilosis (17.8%) and C.glabrata (10.1%) the next two most frequent species. If this trend continues it will have significant implications for the choice of initial antifungal therapy. Further analysis on antifungal susceptibility will be conducted on all isolates obtained. This study is the first attempt to define candidaemia in Australia and the first in any country to attempt to collect nation-wide data. Significant differences have already been noted between Australia and published international studies half way through the course of the study. At it’s conclusion, it is anticipated that a better understanding of the risk factors, management strategies and outcomes will result in improved patient diagnosis and treatment.

abstract No: 


Full conference title: 

The 15 th Congress of the International Society for Human and Animal Mycology
    • ISHAM 15th (2003)