BACKGROUND Onychomycosis or nail fungal infection is the most common nail disease. Despite the wide range of studies on this condition, it remains difficult to establish the correct diagnosis and effective treatment. OBJECTIVES To evaluate the efficacy of classical laboratory methods for the diagnosis of onychomycosis, and the in vitro susceptibility of the its main etiological agent to antifungals used in routine. METHODS Nail samples of 100 patients with clinically suspected feet onychomycosis were collected to confirm the diagnosis by direct mycological examination and fungal culture. In vitro antifungal susceptibility testing was performed against strains of the main dermatophyte isolated by microdilution, according to the standardized protocol (M38-A2 - CLSI) RESULTS Clinical diagnosis of onychomycosis was confirmed by laboratory analysis in 59% of patients. Of these, 54.2% were positive only in direct mycological examination, 44.1% in direct mycological examination and culture, and one case (1.7%) was positive only in culture, resulting in weak agreement between these tests (Kappa = 0.385; p <0.001) High minimum inhibitory concentration values of fluconazole and itraconazole were observed in 66.7% and 25.0% of isolates of T. rubrum tested. Additionally, high MIC values of terbinafine and ciclopirox was detected in only one isolate, and this was one of the strains in which in vitro activity of itraconazole and fluconazole has not been proven. CONCLUSIONS Poor agreement was observed between direct mycological examination and culture for the diagnosis of onychomycosis, with direct mycological examination being significantly more sensitive. Except for fluconazole, the other three antifungals tested showed good in vitro activity against clinical isolates of T. rubrum.