ANTIMICROBIAL TREATMENT OF OCCULT BACTEREMIA - A MULTICENTER COOPERATIVE STUDY

This prospective multicenter study was conducted to define more clearly clinical and laboratory criteria that predict a strong probability of occult bacteremia and to evaluate the effect of empiric broad spectrum antimicrobial treatment of these children. Children 3 to 36 months old with fever greater-than-or-equal-to 40-degrees-C (104-degrees-F) or, greater-than-or-equal-to 39.5-degrees-C (103-degrees-F) with white blood cells (WBC) greater-than-or-equal-to 15 x 10(9)/liter, and no focus of infection had blood cultures obtained and were randomized to treatment with oral amoxicillin/potassium clavulanate or intramuscular ceftriaxone. Sixty of 519 (11.6%) study patients had positive blood cultures: Streptococcus pneumoniae, 51; Haemophilus influenzae b, 6; Neisseria meningitidis, 2; and Group B Streptococcus, 1. Subgroups of high risk were identified as fever greater-than-or-equal-to 39.5-degrees-C and WBC greater-than-or-equal-to 15 x 10(9)/liter, 55 of 331 or 16.6% positive with increasing incidence of positive culture with increasing increments of degrees of leukocytosis to WBC greater-than-or-equal-to 30 x 10(9)/liter where 9 of 21 or 42.9% were positive. Subgroups of significantly lower risk were identified as fever greater-than-or-equal-to 39.5-degrees-C and WBC <15 x 10(9)/liter, 5 of 182 or 2.7% positive and those with WBC <10 x 10(9)/liter, 0 of 99 or 0.0% positive. Children with positive cultures who received ceftriaxone were nearly all afebrile after 24 hours whereas a significant number who received amoxicillin/potassium clavulanate remained febrile. In the 459 culture-negative children more amoxicillin/potassium clavulanate-treated children developed diarrhea and had less improvement in clinical scores after 24 hours than ceftriaxone-treated children. Children 3 to 36 months old with fever greater-than-or-equal-to 39.5-degrees-C and WBC greater-than-or-equal-to 15 x 10(9)/liter and no focus of infection are at high risk (>16%) for having occult bacteremia. Antimicrobial treatment of febrile children with these high risk criteria appears prudent whereas routine treatment of those with low risk criteria does not. Both treatment regimens evaluated are rational and all patients did well.