Epidemiology of nosocomial Candidaemia in a Finnish tertiary care hospital, 1987–2004

Abstract number: P1794

Poikonen E., Lyytikäinen O., Anttila V., Kuusela P., Koukila-Kähkölä P., Ruutu P.

Objectives: Nosocomial bloodstream infections (BSI) caused by Candida sp. (i.e. candidaemia) have been shown to be a growing problem in the US. In addition to immunocompromised patients, intensive care and surgical patients are also at risk, mainly because of wide spectrum antibacterial treatment, invasive monitoring and gastrointestinal surgery. We studied the changes in epidemiology of candidaemia by assessing the incidence and outcome of candidaemia as well as causative Candida sp. in the Helsinki University Central Hospital (HUCH) during 1987–2004.

Methods: HUCH with 1600 beds serves a population of 1.66 million in Southern Finland. All patients with at least one blood culture positive for Candida sp. during 1987–1998 were retrospectively identified from the laboratory logbooks, and the patient charts were reviewed. During 1999–2004, candidaemia episodes were identified through prospective laboratory-based surveillance of nosocomial BSIs, which was carried out as a part of the Finnish Hospital Infection Program. Centers for Disease Control and Prevention definitions for nosocomial BSIs were used.

Overall incidence of candidaemia per 10,000 patient-days as well as incidences by Candida sp., Helsinki University Central Hospital, 1987–2004.

Results: A total of 373 episodes of nosocomial candidaemias were observed (average annual incidence, 0.40 per 10,000 patient-days; range by year, 0.27–0.59); no increase in annual incidence was detected (Figure 1). Median age of candidaemia patients was 52 years (range, 0–89 years) and 59% were male. Most common causative species was C. albicans (65%), followed by C. parapsilosis (13%), C. glabrata (9%), C. krusei (5%), C. tropicalis (3%) and others (13%); no increase in proportion of non-albicans sp. was observed. During 1987–1998, the proportion of intensive care patients with candidaemia varied between 24–27%, but by the period of years 1999–2004 it increased to 44%. The increase was observed in neonatal (15–21%) and surgical (46–55%) patients but not in patients with haematologic malignancy, whose proportion decreased (21–11%). Most (68%) patients with candidaemia had central-lines. The one-month case fatality was 31%; no changes over the six-year periods were observed (range, 30–32%).

Conclusions: During the 18-year study period, no increasing trend in the incidence of nosocomial candidaemia was detected and no shift towards non-albicans sp. occurred. Crude mortality remained high. The changes in the patients groups at risk may reflect differences in prevention strategies which need to be analysed for further improvement of prevention.