Continuous chemoprophylaxis is effective in the prevention of new episodes of acute otitis media (AOM) in otitis-prone children, but compliance can be a problem and thus efficacy can be decreased. Intermittent chemoprophylaxis has so far shown conflicting results. Azithromycin, which has a peculiar pharmacokinetics, resulting, even after a single dose, in persistently elevated concentrations in respiratory tissues, could permit a periodic administration with higher compliance. We compared a 6-month course of once-weekly azithromycin (5 or 10 mg/kg of body weight) with that of once-daily amoxicillin (20 mg/kg) in a single-blind, randomized study of prophylaxis for recurrent AOM in 159 children aged 6 months to 5 years with at least three episodes of AOM in the preceding 6 months. In the amoxicillin group, 23 (31.1%) of 74 children developed 29 episodes of AOM, while in the 10-mg/kg azithromycin group, 11 (14.9%) of 74 children experienced 15 episodes. The 5-mg/kg/week azithromycin trial was prematurely interrupted after nine cases, due to the high occurrence rate of AOM (55.5%). During the 6-month prophylaxis period, the proportion of children with middle ear effusion declined similarly in both groups. No substantial modification of the nasopharyngeal flora was noted at the end of prophylaxis in both antimicrobial groups. In the 6-month-postprophylaxis follow-up period, about 40% of children in both groups again developed AOM. Azithromycin at 10 mg/kg once weekly can be regarded as a valid alternative to once-daily low-dose amoxicillin for the prophylaxis of AOM. Although in the present study no microbiological drawback was noted, accurate selection of children eligible for prophylaxis is mandatory to avoid the risk of emergence of resistant strains.