Induction of telithromycin resistance in Streptococcus pneumoniae.

efficacy; whereas in the patients treated with penicillin V, no differences would be found with respect to bacteriological (P = 1.0) and clinical efficacy (P = 1.0) between patients with a clarithromycin-susceptible strain and those with a clarithromycin-resistant strain. In the PP analysis of the group treated with clarithromycin, clinical cure at visit three was obtained in 95.2% (120 of 126) of patients with susceptible strains, and in 71.4% (10 of 14) of patients with resistant strains (P < 0.01), with an OR of 8 (95% CI 1.38–39.76). As in the ITT analysis, the principal criterion for efficacy analysis (bac-teriological eradication rate) was achieved at a much higher rate in the group with susceptible strains than in the group with resistant but not for clinical efficacy (P = 0.1) between patients with clarithromycin-susceptible and those with clarithromycin-resistant strains. As with the ITT analysis, when patients classified as indeterminate are excluded, in the clarithromycin group for bacteriological (94. efficacy the differences between patients with clarithromycin-susceptible and-resistant strains persist. Nevertheless, in the penicillin V group, no differences are observed in either bacteriological (P = 0.5) or clinical efficacy (P = 1.0). These data support the effect of macrolide resistance on bacterio-logical and clinical failure in GABHS pharyngitis, and indicate that failure in bacteriological eradication during clarithromycin therapy is more likely to occur in patients infected with clarithromycin-resistant GABHS. This finding may be of utmost importance in areas with a high prevalence of macrolide resistance, such as Spain. 2 The difference observed in bacteriological efficacy in the penicillin V group between patients with clarithromycin-susceptible and patients with clarithromycin-resistant strains may have two explanations: (i) biologically , there is an association between macrolide resistance and cell invasiveness in GABHS (strains combining macrolide resistance and ability to enter human respiratory tract cells may be able to escape both β-lactams, by virtue of intracellular location, and macrolides, by virtue of resistance); 3 and (ii) methodologically, the proportion of patients classified as indeterminate was more unbalanced in patients with clarithromycin-resistant strains (33.3%) than in those with clari-thromycin-susceptible strains (3.6% and 0.8% in the ITT and PP analysis, respectively; P < 0.001 in both cases), which could have biased negatively the bacteriological results of penicillin V. The major contribution of this study is the demonstration that macrolide resistance is clinically relevant in GABHS pharyngitis. Further studies with a larger number of resistant cases are needed to confirm these …