In a randomized, placebo-controlled, double-blind, multicentre study, 130 patients with recurrence of group A beta-hemolytic streptococci (GAS) and clinical signs of pharyngotonsillitis were enrolled. The patients received antibiotic treatment for 10 days, followed by 10 days of alpha-streptococci-inhibitory to GAS-or placebo spray treatment. Patients taking antibiotic treatment for at least 9 days and using the spray for at least 5 days were included in the efficacy analysis. In addition, recurrence within the first 5 days of spray treatment was classified as 'early treatment failure'. The clinical recurrences (bacteriologically verified) in the alpha- (n = 51) and placebo-treated (n = 61) patient groups were 2% (n = 1) and 23% (n = 14) respectively, in patients given spray for at least 5 days (p = 0.004). The inclusion of 'early treatment failures' reduces this difference (p = 0.064). Both treatments were equally well tolerated. Thus, alpha-streptococci given as a spray and used for at least 5 days significantly prevented recurrence of GAS pharyngotonsillitis.