Onychomycosis is observed less frequently in children than adults. Until recently management of onychomycosis in children included topical formulations, oral griseofulvin, and in some cases deferral of treatment. We attempted to determine the prevalence of onychomycosis in North American children 18 years old or younger attending our dermatology offices (three Canadian, two U.S.) and to report the group's experience using fluconazole, itraconazole, and terbinafine for onychomycosis. We undertook a prospective, multicenter survey in which all children, regardless of presenting complaint, were examined for onychomycosis by a dermatologist. In instances of clinical suspicion appropriate nail samples were obtained for light microscopy and culture. A total of 2500 children under age 18 were examined in the five-center survey (1117 males and 1383 females, mean +/- S.E. age: 11.2 +/- 0.1 years). There was one child with fingernail and ten with mycologically confirmed toenail dermatophyte onychomycosis. The overall prevalence of onychomycosis was 0.44%. Considering those children whose primary or referring diagnosis was not onychomycosis or tinea pedis, the prevalence of onychomycosis was 0.16%. Outside the survey we have seen six other children with dermatophyte onychomycosis; these 17 cases form the basis for the remainder of the report. Of the 17 children, eight (47%) had concomitant tinea pedis infection, and in 11 (65%) a sibling, parent, or grandparent had onychomycosis or tinea pedis. Management included topical terbinafine (two patients: one cured, one failed therapy), topical ketoconazole (one patient: clinical improvement), oral fluconazole (two patients: one cured, one had Down's syndrome and was noncompliant), oral itraconazole (four patients: three cured with subsequent recurrence at follow-up in one patient, one lost to follow-up), oral terbinafine (five patients: four cured with subsequent recurrence at follow-up in one patient, one failed therapy). One child received no therapy following discussion with the parents, one was lost to follow-up and one was found to have asymptomatic hepatic dysfunction with hepatitis C at pretherapy bloodwork. The prevalence of onychomycosis in our sample of North American children 18 years old or younger was 0.44% (n = 2500). In the subset of children whose primary or referring diagnosis was not onychomycosis, the prevalence of onychomycosis was 0.16%. Children with onychomycosis should be carefully examined for concomitant tinea pedis, and their parents and siblings checked for onychomycosis and tinea pedis. The newer oral antifungal agents fluconazole, itraconazole, and terbinafine may be effective and well-tolerated in the treatment of onychomycosis in this age group. These drugs should be carefully evaluated in a larger cohort of children with onychomycosis.