Treatment outcomes in patients receiving conventional amphotericin B therapy: a prospective multicentre study in Taiwan

OBJECTIVES: To evaluate treatment outcomes and healthcare resource use with conventional amphotericin B therapy for invasive fungal infections (IFIs). PATIENTS AND METHODS: A prospective observational study in hospitalized adult patients receiving amphotericin B treatment was undertaken at four hospitals in Taiwan. Patients were observed from the start of therapy to hospital discharge. RESULTS: A total of 108 patients (October 2000 to April 2002) were included in the study. Proven or probable IFIs as defined by the EORTC/MSG criteria were the reasons for the initiation of amphotericin B in 35.2% of the sample. A total of 24.1% patients developed nephrotoxicity (NT) (defined as a 50% increase in the baseline serum creatinine and achieving a peak of at least 2.0 mg/dL). Treatment of proven/probable IFIs [odds ratio (OR) = 4.16, 95% confidence interval (CI) = 1.61-10.75] was a significant predictor of the development of NT. The in-hospital mortality rate was 38.0%. Proven/probable IFIs (OR = 6.93, 95% CI = 2.62-18.29) and the development of NT (OR = 3.68, 95% CI = 1.22-11.04) were independent predictors of in-hospital mortality. For patients alive at discharge, those with NT had a trend of longer hospital stay compared with patients who had not developed NT (mean, 49.3 +/- 18.2 versus 29.3 +/- 22.3 days, P = 0.069). For patients who died, those who had developed NT died sooner (15.5 +/- 16.7 versus 33. 8 +/- 26.9 days, P = 0.0004). CONCLUSIONS: NT was associated with accelerated mortality and increased hospital stay for patients who survived. Using amphotericin B carefully or the use of antifungal agents with less potential for NT might improve patient outcomes.