Acute otitis media caused by Streptococcus pneumoniae in children's hospitals between 1994 and 1997.

OBJECTIVE: To determine patterns of resistance for isolates of Streptococcus pneumoniae recovered from middle ear fluids of children from eight children's hospitals between September, 1994, and August, 1997.
METHODS: Data were extracted retrospectively from the medical records of eight children's hospitals. A standardized data form was completed for each episode of pneumococcal infection. Systemic isolates (blood and pleural, synovial and spinal fluids) of S. pneumoniae were collected during the same period. All isolates of S. pneumoniae from each center were sent to a central laboratory. Susceptibility to penicillin and ceftriaxone was determined by microbroth dilution. Organisms were considered nonsusceptible to penicillin if the minimum inhibitory concentration was > or = 0.1 microg/ml and nonsusceptible to ceftriaxone if the minimum inhibitory concentration was > or = 1.0 microg/ml.
RESULTS: S. pneumoniae was recovered from the middle ear fluids of 707 children from all centers during the study period. Thirty-nine (5.5%) were infections recorded at 4 centers which evaluated middle ear fluid only sporadically and were not included in this analysis. The remaining 668 infections reported by the 4 remaining participating hospitals reflect the experience of 608 children. There were 54% boys; 440 (73%) were Caucasian, 111 (18%) were African-American, 38 (6%) were Hispanic and for 19 (3%) the race was not recorded. The children ranged in age from 16 days to 13.8 years with a mean (+/-sD) of 26.0 (+/- 26.1) months. Children who received antibiotics in the 30 days before the middle ear isolate was recovered were more likely to harbor a resistant strain of S. pneumoniae than children who had not recently received an antibiotic (P < 0.001). Isolates recovered from children with spontaneous otorrhea were more likely to be susceptible to penicillin than isolates recovered during myringotomy, with or without the insertion of tympanostomy tubes (P < 0.01). There was wide variation in the susceptibility of middle ear isolates to penicillin and ceftriaxone according to geographic location; however, in every locale the middle ear isolates were less likely to be susceptible to penicillin and ceftriaxone than systemic isolates of S. pneumoniae.
CONCLUSION: The prevalence of penicillin-resistant and cephalosporin-resistant S. pneumoniae in middle ear isolates derived from children cared for at four different children's hospitals was quite variable. In some locations the prevalence of resistance is still increasing, whereas in other areas the rate of resistance was at a plateau during the period of surveillance. The prevalence of isolates of S. pneumoniae susceptible to penicillin and ceftriaxone was always less common among middle ear isolates than among systemic isolates. Previous antibiotic use remains the most predictive factor for the recovery of isolates resistant to penicillin and ceftriaxone.