160 children with an average age of 9 years (range 6-15) affected by acute bacterial tonsillitis, were selected and assigned, following an open, parallel group design to: a) brodimoprim at the dose of 10 mg/kg on the first day, in single administration, and of 5 mg/kg on the following days; b) cotrimoxazole suspension, at the dosage of 6 mg of trimethoprim/kg/day, in two daily administrations; c) amoxicillin with clavulanic acid suspension (amoxi-clavulanate) 50 mg/kg every 12 hours. Quantity of pharynx and tonsillar exudate, pharynx pain, dysphonia and dysphagia were checked at the basal time, 3rd, 7th and at the last day of therapy. These symptoms were evaluated using a four-step rating scale. The evolution of body temperature was measured at two different times (1 and 5 o'clock p.m.), until the end of treatment, foreseen five days after disappearance of fever. Microbiological evaluation through a pharynx swab was performed at the beginning and at the end of therapy. Side-effects were registered during all the observation period. Lab-tests were carried out at the enrollment and at the end of treatment. The frequency and intensity of symptoms decreased significantly in all treatment groups. In comparison with amoxi-clavulanate, the brodimoprim group showed an earlier improvement (3rd day) of the clinical situation and a significantly better regression of pharynx exudate (p < 0.01), pharynx pain (p < 0.05) and dysphonia (p < 0.01).(ABSTRACT TRUNCATED AT 250 WORDS)