Aim: To determine efficacy and safety of intravenous micafungin vs. intravenous fluconazole in the treatment of oesophageal candidiasis. Methods: A total of 523 patients ≥16 years with documented oesophageal candidiasis were randomized (1:1) in this controlled, non-inferiority study to receive either micafungin (150 mg/day) or fluconazole (200 mg/day). Response was evaluated clinically and endoscopically. Post-treatment assessments were performed at 2 and 4 weeks after discontinuation of therapy. Results: Median duration of therapy was 14 days. For the primary end-point of endoscopic cure, treatment difference was -0.3% (micafungin 87.7%, fluconazole 88.0%). Documented persistent invasive disease at the end of therapy was reported in 2.7% and 3.9% of patients, respectively. Both 84.8% of micafungin and 88.7% of fluconazole patients remained recurrence free at 4-weeks post-treatment. The overall therapeutic response rate was 87.3% for micafungin and 87.2% for fluconazole. The incidence of drug-related adverse events was 27.7% for micafungin and 21.3% for fluconazole. Six (2.3%) micafungin- and two (0.8%) fluconazole-treated patients discontinued therapy; rash was the most common event leading to discontinuation. Conclusion: Intravenous micafungin (150 mg daily) is well tolerated and as efficacious as intravenous fluconazole (200 mg daily) in the primary treatment of oesophageal candidiasis. achieving high rates of clinical and endoscopic cure. © 2005 Blackwell Publishing Ltd.