Yeast isolates from burned patients were analyzed retrospectively for a 7-year period (1984-1991). Topical nystatin was used routinely in the burn wound dressing as antifungal therapy beginning in July 1986. Nystatin used was associated with a significant decrease in overall yeast acquisitions in burn wounds; yeasts were isolated from 15.5% of admitted patients before the use of nystatin vs. 10.5% with use of nystatin (odds ratio [OR] = 0.64; 95% confidence interval [CI], 0.48-0.86). New acquisitions of Candida rugosa in burn wounds increased from 0.36% of admissions during the period July 1984 to June 1986 (before nystatin use) to 5.25% in the period July 1986 to June 1991 (during use of nystatin) (OR = 15.3; 95% CI, 4.1-128). The incidence of fungemia decreased from 3.25% of admissions in the pre-nystatin period to 1.43% in the postnystatin period (OR = 0.43; 95% CI, 0.22-0.87). C. rugosa caused none of 18 fungemias in the former period and 15 of 21 in the latter period (P = .002). Susceptibility testing of recent C. rugosa isolates demonstrated resistance to nystatin and moderate susceptibility to amphotericin B and fluconazole. Topical nystatin use was associated with a decrease in fungemias and acquisition of yeasts in burn wounds but with an increase in colonization and fungemias caused by nystatin-resistant, amphotericin B-susceptible C. rugosa.