PURPOSE: Empirical antimicrobial treatment is indicated before bacteriological results are available for young children with febrile UTI to minimize renal scarring. To ensure appropriate therapy, knowledge of the prevalence of causative organisms and their susceptibility patterns to antimicrobials is mandatory. We performed a retrospective analysis investigating the local prevalence and resistance patterns of uropathogens, primarily E. coli, isolated from community-acquired UTIs. METHODS: A total of 103 positive urine cultures from children with febrile UTI collected at Bundang CHA General Hospital from February 2004 to February 2005 were analyzed. Inclusion criteria were fever higher than 37.5 degrees C, significant bacteriuria with single strain growth of at least 105 colony forming units/mL urine, and leukocyturia >5/HPF. RESULTS: E. coli(89.3%) was the leading uropathogen followed by Enterococcus spp.(3.9%) Klebsiella spp.(2.9%), Citrobacter spp.(1.9%) and Enterobacter spp.(1.9%). E. coli strains revealed a low proportion of antimicrobial susceptibility to ampicillin(AMP; 27.2%) ampicillin-sulbactam(AMS; 34.8%) and trimethoprim-sulfamethoxazole(SXT; 65.2%). Susceptibility patterns to cephalosporins were as follows; cefazolin(1st generation; 91.3%), cefoxitin(2nd; 100%), ceftriaxone(3rd; 97.8%) and cefepime(4th; 97.8%). Three E. coli isolates produced extended-spectrum beta-lactamase(ESBL). CONCLUSION: Empirical treatment with AMP, AMS and SXT, which are commonly used in pediatric clinics, is not recommended for childhood UTI due to high incidence of resistance. The high level of susceptibility to cephalosporins makes these drugs reasonable alternatives. However the emergence of ESBL-producers, even though they are quite few, may have an impact on cephalosporin treatment in the future.