Clostridioides difficile Toxin B PCR Cycle Threshold as a Predictor of Toxin Testing in Stool Specimens from Hospitalized Adults

Rapid, accurate detection of Clostridioides difficile toxin may potentially be predicted by toxin B PCR cycle threshold (tcdB C-t). We investigated the validity of this approach in an inpatient adult population. Patients who tested positive by C. difficile PCR (Cepheid GeneXpert) from December 2016 to October 2020 (n = 368) at a tertiary medical center were included. All stool samples were further tested by rapid glutamate dehydrogenase (GDH)/toxin B EIA and cell cytotoxin neutralization assay (CCNA). Receiver operating characteristic curves were analyzed. The area under the curve for tcdB C-t predicting toxin result by EIA was 0.795 (95% confidence interval (CI) 0.747-0.843) and by CCNA was 0.771 (95% CI 0.720-0.822). The Youden C-t cutoff for CCNA was <= 27.8 cycles (sensitivity 65.0%, specificity 77.2%). For specimens with C-t <= 25.0 cycles (n = 115), CCNA toxin was positive in >90%. The negative predictive value of tcdB C-t for CCNA was no greater than 80% regardless of cutoff chosen. In summary, very low C-t values (<= 25.0) could have limited value as a rapid indicator of positive toxin status by CCNA in our patient population. A broad distribution of C-t values for toxin-negative and toxin-positive specimens precluded more robust prediction. Additional data are needed before broader application of C-t values from qualitatively designed assays to clinical laboratory reporting.

Automated summary

This study investigated the validity of using toxin B PCR cycle threshold (tcdB C-t) as a rapid and accurate method for detecting Clostridioides difficile toxin. The findings suggest that very low C-t values may have limited value in predicting positive toxin status using the cell cytotoxin neutralization assay (CCNA). Additional data are needed to support the use of C-t values in clinical laboratory reporting.