Abstract
OBJECTIVES: Clostridium difficile infection (CDI) is a toxin-mediated disease. Oncology patients are at increased risk for developing CDI. Diagnosis of CDI by PCR has led to misclassification of some C. difficile carriers as CDI cases. We determined if an optimized C. difficile PCR cycle threshold value (C(T)) could reliably predict presence of free toxin, and in turn improve the utility of PCR in detecting clinically relevant CDI in oncology patients. METHODS: 183 consecutive patients positive for C. difficile by the Xpert C. difficile were additionally tested using the cell culture cytotoxicity neutralization assay (CYT) and enzyme immunoassays (EIA). C(T) values at diagnosis and relevant clinical information were recorded. Receiver operating characteristic (ROC) curve was used to assess predictive validity and to find optimal C(T) for CYT positive cases. Severity of CDI was assessed by blinded charts review. RESULTS: Using CYT as the reference, ROC-derived Youden cut-off C(T) of 28.0 predicted 77% cytotoxin positive cases, and 91% and 100% of severe and complicated CDI episodes respectively. The median C(T) values for non-severe, severe, and complicated CDI episodes were 28.0, 24.5 and 22.5 respectively (p = 0.005). CONCLUSIONS: Lower C(T) value of the Xpert C. difficile PCR was associated with the presence of toxin and increased CDI severity. C(T) values may be beneficial in interpreting positive C. difficile PCR results.