To the Editors: Treatment of AOM in the current era of high prevalence of antibiotic-resistant Streptococcus pneumoniae is a great concern and an issue of controversy. A major step towards reducing nonjudicious antibiotic treatment of this infection has been the recognition that antibiotic treatment can be postponed and even omitted in many cases. Yet antibiotic prescriptions for this indication are currently still very frequent, and nonjudicious treatment of this disease continues to contribute to the increase in resistance rate among pathogens. One frequently measured outcome is nasopharyngeal carriage of resistant pathogens after different antibiotic treatments, because such a carriage constitutes a potential source of spread of resistant pathogens in the community. In the study by Toltzis et al,1 4 antimicrobial regimens frequently used in the treatment of acute otitis media (AOM) (amoxicillin, ceftriaxone, cefprozil and azithromycin) were compared, and the outcomes of nasopharyngeal colonization by penicillin-nonsusceptible S. pneumoniae (PNSSP) and in particular penicillin-resistant S. pneumoniae (PRSP) were measured. Although the relative proportion of PRSP of all S. pneumoniae isolated was higher in the amoxicillin group, no significant difference was observed in the prevalence rate of PRSP among the children treated with all 4 regimens during the follow-up period (up to 30 days), when the 4 treatment groups were compared. Neither was there any difference in PRSP carriage rates between the treatment groups. In fact, the only antibiotic that reduced PNSSP carriage was amoxicillin. Because about 33% of PNSSP and 66% of PRSP in the United States are also macrolide-resistant,2,3 missing information in the report by Toltzis et al that is crucial is the carriage rate of macrolide-resistant and multidrug-resistant S. pneumoniae during the period after treatment. In fact, because one of the treatment arms was azithromycin, it is surprising that macrolide resistance (and multidrug resistance) rates were not reported or at least that the reason why they were not reported was not provided. In communities in which macrolide resistance of S. pneumoniae is low, this may not be significant, but in populations with higher prevalence rates (as has been reported in the United States and many countries in Europe), this information is essential. Toltzis et al4 calculated both the simple odds ratio (ORs) and the conditional odds ratio (ORc). Because there was no change in the absolute number of PRSP isolates in each group and thus of the prevalence rate of PRSP, and even the change in PNSSP rates were not significant, the pairwise comparisons of ORs indeed did not detect any differences between the groups (P > 0.29). A recent review5 emphasized how misleading the use of ORc alone can be in interpreting data on carriage and spread of antibiotic-resistant organisms. Toltzis et al used the ORc to claim that acquisition of PRSP was higher in the amoxicillin-treated group, whereas the data show that in the amoxicillin-treated group absolute numbers of PNSP decreased. However, both PNSP and PSSP decreased, but PRSP as a subgroup did not change; the proportion of PRSP increased. The acquisition rate of PNSP or any other subgroup was not measured. In fact, without the use of molecular measures, or at least serotyping, one cannot really calculate acquisition or eradication rates of S. pneumoniae given that both spontaneous and antibiotic-induced eradication and acquisition of new strains are common. Therefore the conclusions drawn by the authors (that amoxicillin administration, compared with the other 3 regimens, resulted in a larger shift toward PRSP colonization and might promote the dissemination of the nonsusceptible phenotype) do not correspond to their results. To date, the published data still support the use of high dose amoxicillin given at high doses (90–100 mg/kg/d, with or without clavulanate)6,7 as the primary treatment of AOM. This regimen is still regarded as optimal when considering both the efficacy of the antimicrobial regimen and the population dynamics associated with carriage and spread of antibiotic-resistant S. pneumoniae. Ron Dagan, MD Soroka University Medical Center The Faculty of Health Science Ben-Gurion University of the Negev Beer-Sheva, Israel Gili Regev-Yochay, MD Ethan Rubinstein, MD Sheba Medical Center Tel-Hashomer, Israel
Read full abstract