Attributable mortality of candidemia at a German tertiary hospital from 1997 to 2001 before the introduction of echinocandins.

OBJECTIVES: The relevance of candidemia has increased over the last decades due to higher incidence rates in an ageing society. Studies on amphotericin B and fluconazole have shown high attributable mortality rates of 38% and 49% in the United States. Incidence rates and locational factors might have an impact on the mortality rates at the University Hospital of Cologne (UHC), Germany. METHODS: We performed a matched case-control study including 57 patients with candidemia, hospitalised at the UHC between 1 July 1997 and 30 June 2001. Controls were matched by age, sex, admission date, treatment on intensive care unit (ICU), number of days at risk, underlying diseases, surgical procedures and the Charlson Comorbidity Index. RESULTS: The incidence of candidemia was 3.5 per 10 000 admissions. For cases and controls, we observed in-hospital-mortality rates of 33.3% and 11.8%, and a 30-day mortality of 23.5% and 7.8% respectively. The attributable mortality rate to candidemia was 21.5%, and at 30 days, it was 15.7%. Underlying conditions were more frequent in cases than in controls, especially central venous catheter (80% vs 33%, P < .001), chronic cardiovascular disease (39.2% vs 25.5%, P = .138), treatment on ICU (31.4% vs 13.7%, P = .033) and chronic liver disease (21.6% vs 0%, P < .001). CONCLUSIONS: The attributable mortality of candidemia at the UHC between 1997 and 2001 was lower compared to studies performed in the United States with a similar design. Contributing factors might be lower incidence rates and less comorbidities in our study.