In this prospective study of AOM characteristics in Germany, H. Influenzae was the most frequently identified pathogen in children 12 months of age and older, and was also frequently identified in 3–11 month olds, consistent with a growing body of evidence implicating H. Influenzae as a key pathogen in AOM [2, 4], and of a possibly increasing role due to PCV vaccination. S. pyogenes was the second most frequently identified pathogen. These results are consistent with preliminary results from another study conducted in Germany , but contrast with other studies suggesting a possibly diminishing role for S. pyogenes in AOM etiology . In a large evaluation of AOM etiology conducted in Israel, S. pyogenes AOM was observed more frequently in older children than in younger children, and was often associated with acute TM perforation . In our study S. pyogenes was only isolated from otorrhea samples (not from tympanocentesis samples), supporting the notion that S. pyogenes infection rapidly leads to TM perforation . The high rate of S.pyogenes isolation in the present study likely reflects the severity of disease in the enrolled population (children with TM perforation or with clinically indicated tympanocentesis according to German clinical guidelines).
In the present cohort at least 74% of children had received at least one PCV7 dose. Although the low number of subjects precludes firm conclusions, the results are generally consistent with larger observational studies showing an increase in AOM due to non-PCV7 S. pneumoniae serotypes in countries where PCV7 uptake is high . New pneumococcal conjugate vaccines targeting H. influenzae by use of Protein D as carrier protein (PHiD-CV) , may provide additional benefits in Germany.
In contrast with other countries in Europe, antibiotic susceptibility rates were high in German children. This may be due in part to reduced penicillin consumption in Germany, which is lower than in many other countries in Europe .
The potential limitations of the study include the low number of culture-positive samples for which a cause was not able to be identified, although the percentage of positive cultures increased (from 43% to 61%) after changes were made to ensure that sample delivery times and temperature constraints were met. It is also possible that the initial delay between sample collection and processing might have introduced a bias toward a differential recovery of more resilient organisms. For example, S. aureus and S. pyogenes are known to overgrow other bacteria on swabs when processing delay increases, and both S. pneumoniae and NTHi are fastidious organisms more likely to perish after 6 hours and in conditions of freezing.
Because tympanocentesis is not routinely performed for AOM management in Germany, only children with spontaneous perforations in which a specimen was obtained using a procedure avoiding contamination or those in whom tympanocentesis was clinically indicated were eligible. Thus, our population had more severe AOM and the results may not necessarily be able to be extrapolated to all cases of AOM.