In order to find an effective way to detect bacteria responsible for respiratory tract infections in children, we first examined as pharyngeal swabs, epi-pharyngeal swabs and nasal aspirates obtained from children hospitalized at our pediatric service during these five months from December 1997 to April 1998. In the rate of bacterial infection, it was found that nasal aspirates came out on top with 92.6%(25/27), followed by epipharyngeal swabs with 71.6%(53/74) and pharyngeal swabs with 26.2%(38/145). Single-species bacteria were found in 78.9%(30/38) of pharyngeal swabs, where as 45.3% of epi-pharyngeal swabs (24/53) and 52.0% of nasal aspirates (13/25) proved mixed infections with two-or three-defferent species. Thus it was suggested that nasal aspirates and epi-pharyngeal swabs would be far more adequate than pharyngeal swabs to detect bacteria with accuracy.
Next, based on the efficiency of bacterial detection, we compared culture media for the specimen obtained from in-patients and out-patients at our pediatric service. The rate of isolation of gram-negative rods was as low as 0.3%(1 of 314 strains) even when BTB agar plate, a selective medium for these bacteria, was employed. The sensitivety was not much different from those observed with nonselective blood agar plate. These results suggest that the conventional blood agar media can substitute for the more expensive type of BTB agar medium for the diagnosis of infections diseases of the airwaysin children.