Methods: 262 children were enrolled in the MAL-ED Tanzania site. We randomly selected 100 children who had E. coli specimens archived every 6 months through 60 months. Up to 5 lactose-fermenting colonies were selected from growth on MacConkey agar. Drug susceptibility testing of 18 antibiotics was performed by disk diffusion. CLSI interpretive criteria were used for determination of resistance. Generalized estimating equations were used to create a multivariate Poisson regression model for drug resistance risk factors.
Results: 823 E. coli specimens were available for testing. The highest rates of resistance were to ampicillin, cefazolin, and cotrimoxazole, respectively. No carbapenem resistance was found. 1.8% met criteria for extended-spectrum beta-lactamase production based on combination disk testing. 696 (84.6%) specimens met criteria for multi-drug resistance (non-susceptible to at least 1 drug in at least 3 drug categories). In terms of dynamic risk factors, there was no association between antibiotic use or episodes of diarrhea and antibiotic resistance. For static risk factors, there was an association between higher income and increased antibiotic resistance.
Conclusion: Antibiotic resistance carriage is an under recognized problem in this setting. Resistance rates at 6 months of age are higher than expected, with surprisingly little variance explained by individual level risk factors for resistance in this community.
T. Walongo, None
J. Gratz, None
J. Platts-Mills, None
E. Mduma, None
E. Houpt, None
See more of: Poster Abstract Session