Population-based incidence of severe pneumonia in children in Kiel, Germany

Background: Elaborated data on the descriptive epidemiology of community-acquired pneumonia (CAP) are a prerequisite to estimate the impact of new vaccines. Patients and Methods: From July 1996 to June 2000, all children (0-16 years) admitted to one of the two pediatric hospitals in Kiel and being resident in the municipal area of Kiel were investigated by cross-sectional studies and prospective testing using a 9-valent in-house m-RT-PCR method. Results: In the 4-year period, 514 children were included (mean age 46, median 40 months): 279 were diagnosed with bronchopneumonia (BPN, median age 26 months), 235 with pneumonia (PN) (47 months); within the latter 69 cases had lobar PN (55 months), 41 atypical PN (51 months) and 28 parapneumonic effusions (74 months). An underlying chronic condition was present in 22.8% and 10.1% were born prematurely. The population-based incidence rates (per 100,000 per year) were on average 300 for children 0-16 years, 163 for BPN, 136 for PN, 53 for lobar PN, 24 for atypical PN and 16 for parapneumonic effusions. The rate was stable or slightly declined over the observation period. 61 % of infants and 45 % of children under 5 years of age have to be hospitalized having contracted CAP. The highest fraction of 34 and 25%, respectively, was attributable to RSV. Viruses were not diagnosed significantly more often in BPN than in PN, if stratified by age. Conclusion: The incidence and the admission rate of severe CAP is lower than in the USA. The high rate of empyema warrants enhanced surveillance as an indicator for antibiotic resistance or changing impact of pneumococcal serotypes. Misclassification, also with ICD codes, is a major issue. Well analyzed epidemiological recruitment areas are a valid tool to generate precise data in Germany.