A double-blind randomized comparative phase II study of flutrimazole site-release vaginal cream (1, 2 and 4%) with placebo site-release vaginal cream was undertaken in patients with acute vulvovaginal candidosis. Vaginitis was demonstrated by both positive findings on microscopic examination of vaginal smears and positive culture as well as by the presence of clinical signs and symptoms. The vaginal monodose treatment was inserted in the evening at bedtime using a vaginal applicator and, in addition, all four groups of patients received additional topical external cream for application to the vulva twice-daily for 7 days; the placebo group received a placebo cream and the active therapy groups all received a 2% flutrimazole cream. A total of 133 patients who were seen over a 10-month period were screened and randomized: five patients did not take the allocated medication, and four patients whose menstrual period began shortly after study entry were excluded from the study, leaving 124 patients who were randomly allocated to receive a monodose vaginal 1% cream (regimen A, 28 patients), a monodose vaginal 2% cream (regimen B, 32 patients), a monodose vaginal 4% cream (regimen C, 31 patients) or a monodose vaginal placebo cream (regimen D, 33 patients). At the assessment 9 days after the end of therapy the proportion of patients who were cured was 82% in group A, 87.4% in group B, 83.8% in group C and 63.5% in group D. Three patients (10.7%) in group A, four (12.5%) in group B, one (3.2%) in group C and 12 (36.36%) in group D did not respond to the treatment. One patient (3.5%) in group A, and two patients (6.4%) in group C terminated the treatment prematurely due to intolerance. There was a significant association between Candida glabrata and treatment failure (P < 0.04) and C. glabrata and carrier state (P = 0.01) in vagina (chi 2 test, P = 0.01) and vulvovagina (chi 2 test, P = 0.00001). At the assessment 4 weeks after the end of therapy the proportion of cured patients was 60.6% in group A, 78% in group B, 80.6% in group C and 48.4% in group D. Group D (placebo) versus group B (2%) and group C (4%) showed a significant difference (P = 0.01 and P = 0.007, respectively). Although there were no significant differences in clinical and mycological activity between the three active groups, group B (flutrimazole 2% site-release vaginal cream) was chosen for clinical use due to its tolerance profile. Seven patients (25%) in group A, three (9.3%) in group B, two (6.4%) in group C and five (15.1%) in group D relapsed 4 weeks after the end of therapy; the relapse rate was not significantly associated with positive culture results 9 days after treatment. There was a significant association between C. glabrata and the carrier state (P < 0.01). The overall ineffective treatment (includes failures at control 1, relapses at control 2 and premature terminations) was 39% in group A, 21.7% in group B, 16% in group C and 51.3% in group D. There was a significant difference in the overall ineffective treatment when C and D groups were compared with placebo (P = 0.01 and P = 0.003, respectively).