Knowledge of the treatment of recurrent group A streptococcal pharyngotonsillitis has, so far, been based on studies of non-recurrent rather than recurrent episodes of the disease. This multicentre, double-blind, randomized trial was designed to compare the efficacy of loracarbef (200 mg twice daily) vs phenoxymethylpenicillin (penicillin V) (800/1000 mg twice daily) each for 10 days in the treatment of recurrent group A streptococcal pharyngotonsillitis. Among the 331 patients enrolled in the study, 265 were evaluable for efficacy. The combined clinical and bacteriological failure rate was 8.2% in the loracarbef group and 21.5% in the penicillin V group (p = 0.008). Bacterial eradication was noted in 90% of loracarbef-treated patients compared to 66% of penicillin V-treated patients (p < 0.0005). The higher bacteriological eradication and clinical efficacy rate among loracarbef patients might be related to higher stability of loracarbef in the presence of beta-lactamases produced by the oral microflora. These results suggest loracarbef to be a strong candidate for treatment of patients with recurrent group A streptococcal pharyngotonsillitis.