Investigating the prevalence of, and risk factors for, antimicrobial resistance in paediatric urinary bacteria: retrospective analysis of the ‘DUTY’ (Diagnosis of Urinary Tract infection in Young children) cohort study
Bacterial resistance to antibiotics is an internationally recognised health threat, particularly in primary care where 80% of antibiotics are prescribed. Children are high frequency recipients of primary care services, and receive a disproportionately high number of antibiotics compared to middle-age populations. There is currently limited knowledge around the prevalence of resistant bacteria in children, and even less known about risk factors associated with resistance, which could be targeted in future antimicrobial stewardship interventions. This study aims to estimate the prevalence of resistance to common primary care prescribed antibiotics in children’s urinary bacteria, and identify associated risk factors, including the relationship between demographic, clinical factors and previous antibiotic exposure and resistance.
This was a follow-up to the Diagnosis of Urinary Tract Infection in Young Children (DUTY) study. DUTY children were aged 0-5 years presenting to primary care across England and Wales with any acute infectious illnesses. All 7163 children were invited to submit a urine sample, 5017 provided a reference standard urine sample from which Escherichia coli urinary isolates were obtained in a subset. Minimum inhibitory concentrations were used to determine antimicrobial susceptibilities. Risk factors were obtained from case report forms completed by parents at recruitment and included demographics, previous medical history, symptoms, physical examination findings and urinalysis results. Information on previous exposure to antibiotics was collected from primary care records for the 12 months prior to DUTY recruitment. Logistic regression analysis was used to obtain crude and adjusted odds ratios.
Antimicrobial sensitivities were obtained for 824 E. coli urinary isolates, 79 met the laboratory criteria for urinary tract infection (UTI) and 745 did not (non-UTI). Resistance was found against amoxicillin (38.5%), co-amoxiclav (21.0%) and trimethoprim (17.6%). No isolates were nitrofurantoin-resistant. Amoxicillin and trimethoprim resistance was higher in UTI isolates than in non-UTI isolates, both by over 10%. Older children were more likely to carry resistant urinary bacteria (adjusted OR: 1.13, 95% CI: 1.01-1.30). Children with UTI were more than twice as likely to carry a resistant E. coli isolate as those with a non-UTI (aOR: 2.36, 1.26-4.48).There was no association between being prescribed any antibiotic in the 12 months prior to urine sampling and resistance.
Prevalence of resistance to several common primary care prescribed antibiotics in children’s E. coli isolates is high, suggesting some antibiotics should no longer be considered first-line. Unlike some previous studies, we did not find an association between antibiotic exposure in the preceding 12 months and resistance in children’s urinary E. coli. If this reflects a ‘ceiling effect’ for on-going antibiotic use, then the effects of other risk factors (age, susceptibility to UTI, factors associated with deprivation) may require further investigation as possible targets for antimicrobial stewardship interventions.