Brook I, Gruenwald L D
South Med J. 1984 Oct;77(10):1240-2. doi: 10.1097/00007611-198410000-00008.
Blood cultures were obtained during a one-year period from 147 febrile children seen in the hospital outpatient clinic (group 1), and 42 seen in a private practice office (group 2). Although the two clinics were located in the same geographic area, they served different socioeconomic groups. Bacteremia was detected in 12 (8.2%) of the children in group 1; the organisms recovered were Streptococcus pneumoniae in nine and Haemophilus influenzae in three. Bacteremia was observed in two (4.8%) of the children seen in group 2, and both cases were due to S pneumoniae. The difference in prevalence rates for bacteremia between groups was not significant by chi-square testing (P = .46). The age of all the 13 patients with bacteremia was less than 24 months, their temperature was above 104 F (40 C), their polymorphonuclear leukocyte count was above 15,000/cu mm, and a localized site of infection was present in ten of them. These data suggest that the presence of bacteremia in febrile children depends more on the patients' clinical and laboratory findings than on the medical facility in which they are seen.