Background: Many laboratories use polymerase chain reaction (PCR)-based assays to detect the Clostridium difficile toxin B gene (tcdB) in stool. However, PCR testing experience in pediatric patients is limited. We compared the detection of C. difficile by PCR in hospitalized children with and without diarrhea.
Methods: Stool samples from patients aged 1-18 years with diarrhea (symptomatic) and from patients without diarrhea (asymptomatic) were tested for C. difficile tcdB using a commercial PCR assay. Samples were cultured for C. difficile using standard techniques with tcdB PCR and cytotoxicity assays performed on C. difficile isolates. Demographic, clinical and laboratory data were abstracted. Categorical and continuous variables were compared between the 2 groups using Fisher Exact test and the Mann-Whitney test, respectively.
Results: Thirty-five of 188 (19%) stool samples from symptomatic patients and 18 of 74 (24%) samples from asymptomatic patients were positive by PCR (P=0.31). Among PCR-positive patients, symptomatic patients had a significantly higher proportion of subjects who received antimicrobials in the preceding 30 days (P=0.04) and a greater number of preceding antimicrobial days than did asymptomatic patients (P=0.02) but were comparable with respect to the other variables analyzed.
Conclusions: C. difficile PCR assays are frequently positive in hospitalized children both with and without diarrhea. As we observed a high level of toxigenic C. difficile colonization in children, our findings suggest that a positive C. difficile PCR result in a child with diarrhea should be interpreted with caution.