LAUSR

BACKGROUND Recurrence of community-associated (CA) Clostridiodes difficile infection (CDI) approaches 30%. Studies on risk factors and treatment of choice for pediatric CA-CDI are scarce with variable recommendations. METHODS This was a retrospective cohort study of the electronic health records of children 1-17 years with stool specimens sent for C. difficile at Kaiser Permanente Northern California from January 01, 2012 to December 31, 2016. Children with (1) CA disease, (2) confirmatory C. difficile laboratory testing with no other identified causes of diarrhea and (3) clinical symptoms consistent with CDI were defined as cases. Recurrent CA-CDI was defined using the above-described case criteria and onset of diarrhea within 8 weeks of primary CA-CDI. RESULTS Of the 7350 children with stool samples sent for C. difficile testing, 408 had primary CA-CDI. Forty-five (11%) experienced a recurrence. Using multivariable logistic regression, inflammatory bowel disease [odds ratio (OR) 7.5; 95% confidence interval (CI): 2.6-21.1] and cancer (OR 6.3; 95% CI: 1.6-24.1) diagnoses were risk factors for recurrent disease. Compared with children of Caucasian race, those with multi/other/unknown race had an OR of 3.03 (95% CI: 1.04-8.82) of recurrence. There was no statistically significant difference in the type or duration of therapy as a predictor for recurrent CA CDI. Six percent of children who received metronidazole were switched to vancomycin due to subjective metronidazole allergy or intolerance or metronidazole treatment failure. CONCLUSIONS Recurrent CA-CDI in children in our population is less common than previously reported. This study supports first-line treatment with the standard, short course metronidazole in most cases of primary CA-CDI.