COMPARISON OF FLUCONAZOLE WITH ORAL POLYENES IN THE PREVENTION OF FUNGAL-INFECTIONS IN NEUTROPENIC PATIENTS - A PROSPECTIVE, RANDOMIZED, SINGLE-CENTER STUDY

The goal of this prospective randomized single-center study was the co mparison of safety and efficacy of high-dose oral/intravenous fluconaz ole (400 mg daily) (group A) with oral nystatin plus miconazole inhala tions (group B) in the prevention of fungal infections on a hemato-onc ological isolation Ward. Of 157 patients admitted to the isolation war d during the study period only 90 (57%) were eligible for randomizatio n 22 (14%) had a fungal infection at admission. Of the 90 randomized p atients, 89 were evaluable, 43 in group A and 46 in group B. The age, sex, diagnosis, planned therapy and risk factors for fungal infections at admission as well as the duration of neutropenia were in the same proportions in both groups. Oral thrush and mucocutaneous candidiasis were prevented in all patients of both groups, and 29 patients (32%: 1 7 in group A, 12 in group B) were discharged after successful prophyla xis (NS). Empiric amphotericin B was given according to predetermined criteria to 45 patients (51%: 23 group A, 22 group B; NS). Fluconazole significantly delayed the time before the start of intravenous amphot ericin B. It was begun after a median of 10 days (0-45 days, range) of neutropenia below 0.5 x10(9) granulocytes/l in group A and 7.5 days ( 0-26, range) in group B (P< 0.05). The duration of successful prophyla xis was significantly longer in group A (26 days median) than in group B (21 days, median) (P<0.05). Systemic fungal infection was documente d in 3 patients (1 group A, 2 group B; NS). Colonisation with Candida persisted for more than 14 days or occurred de novo after admission in 1 patient in group A and in 7 patients in group B (NS). Oral nystatin had to be discontinued because of oral intolerance in 3 patients and fluconazol had to be stopped because of increased liver values in one patient. Compliance was worse (P< 0.01) in group B; 82% of the planned dose was given in group B compared to 99% in group A. Both regimens s uccessfully prevented oral fungal complications. Fluconazole was bette r tolerated and delayed the need for empiric amphotericin B. Neither a pproach cancelled the need for the empiric use of amphotericin B nor p revented fungal infections or colonization. Systemic fungal infections occur probably independently of oral or mucocutaneous candidiasis.