OBJECTIVES: To evaluate the efficacy of low dose fluconazole treatment for the prevention of yeast colonization and infection in severely neutropenic patients. METHODS: An open randomized trial, comparing fluconazole (100 mg per day) with nystatin (800,000 IU per day), in a University Hospital setting. RESULTS: Antifungal prophylaxis was given during the period of neutropenia, defined as less than 500 polymorphonuclear cells (PMN)/mm3). Thirty-six patients were randomly assigned to fluconazole and 33 to nystatin treatment groups. New oropharyngeal colonizations were significantly reduced by fluconazole (P=0.005), and oropharyngeal infections occurred less frequently in the fluconazole group (3% versus 16%, P=0.07). Stool colonization was identical between both groups. Systemic fungal infections were rare; one fluconazole patient had pulmonary aspergillosis and one nystatin patient developped Candida pseudotropicalis fungemia. Empiric amphotericin B was given with the same frequency in both groups. No side effects were associated with fluconazole. However, the administration of nystatin became impossible for three patients because of vomiting and lack of compliance. CONCLUSIONS: Fluconazole (100 mg per day) is more effective than nystatin for the prevention of oropharyngeal yeast colonization. Comparison with results in the literature suggests that a 100-mg dose of fluconazole has similar effects to 200 or 400 mg per day.