Risk factors for resistance to "first-line" antimicrobials among urinary tract isolates of Escherichia coli in children

Background: There are increasing concerns regarding antimicrobial resistanc e in Canada. Data are limited on the prevalence, patterns of resistance and risk factors associated with resistant organisms, including coliforms, in children. This study was done to address these issues as they relate to uri nary tract isolates of Escherichia coil in a tertiary care pediatric centre in Ottawa. Methods: A surveillance study was conducted from December 1992 to December 1994. Susceptibility testing of urinary tract isolates of E. coil was perfo rmed using a panel of antimicrobial agents. A case-control study was also c onducted for subjects with isolates resistant to trimethoprim-sulfamethoxaz ole (T-S), this drug being used a representative "first-line" agent. Results: A total of 1636 consecutive isolates were obtained from 967 subjec ts. Of the 1636 isolates, 736 (45.0%) were resistant to ampicillin, 514 (31 .4%) were resistant to T-S, 363 (22.2%) were resistant to both ampicillin a nd T-S, and 27 (1.7%) were resistant to both ampicillin and gentamicin. In the case-control study 274 children with isolates resistant to T-S were mat ched with 274 children who had T-S-sensitive isolates obtained during the s tudy period or the preceding or subsequent 6 months. Multivariate analyses indicated that subjects who had received antimicrobials for more than 4 wee ks in the previous 6 months were about 23 times more likely to have isolate s resistant to T-S than were subjects without this risk factor (odds ratio [OR] 23.4, 95% confidence interval [CI] 12.0-47.6). Children with genitouri nary tract abnormalities were 2.4 times more likely to have resistant isola tes than those without such abnormalities (95% CI 1.2-4.5). Compared with c hildren who had no hospital admissions in the previous year, those with 1 a dmission in that period were more likely to have resistant isolates (OR 2.3 , 95% CI 1.4-7.5), as were those with 2 or more admissions in that period ( OR 3.2, 95% CI 1.1-4.8). Compared with children aged 2-6 years, children un der 2 years of age were less likely to have resistant isolates (OR 0.3, 95% CI 0.2-0.8). Interpretation: Selective antimicrobial pressure and multiple admissions to hospital were among the risk factors associated with antimicrobial resista nce. The finding of a low but definite level of resistance to both ampicill in and gentamicin is important for the selection of empiric therapy for sep sis in neonates. The role of inexpensive first-line agents in the outpatien t treatment and prevention of urinary tract infections requires re-examinat ion, particularly in children who have recently received antimicrobial ther apy.