Reply to “Comparison of Detection Methods for Clostridium difficile”

REPLY

We thank Monica Ianosi-Irimie et al. for their interest in our publication. In their letter, the authors questioned if the enhanced sensitivity of the Illumigene Clostridium difficile loop-mediated isothermal amplification (LAMP; Meridian Bioscience, Inc.) assay in the frozen (retrospective) arm of the study could have been influenced by the freeze-thaw cycle. The authors tested fresh and frozen stool specimens and noted no difference in sensitivity of the Illumigene C. difficile LAMP assay. They questioned if we had an explanation for the enhanced detection of C. difficile by the Illumigene assay in the retrospective cell culture cytotoxin assay (CCNA)-positive frozen specimens used in our study.

In our studies, the Xpert C. difficile PCR assay (Cepheid, CA) showed a sensitivity of 100%, while the Illumigene C. difficile LAMP assay showed sensitivities of 81% and 94% in the prospective and retrospective arms of the study, respectively. As stated in our publication (1), we favored the idea that the difference in sensitivity with frozen and fresh stool specimens was most likely due to the preselection of specimens. While we tested sequential specimens in the prospective arm of the study, previously frozen CCNA-positive stool samples were tested for the retrospective arm of the study. The latter specimens are most likely toxin-positive samples with high concentrations of C. difficile DNA, so we should expect a higher sensitivity based on the fact that the CCNA is a less sensitive method than nucleic acid amplification tests (NAATs). This would explain the difference in sensitivity with the two groups of specimens. Similar results have been published in two other studies with sensitivities similar to ours for the Illumigene C. difficile and Xpert C. difficile assays (2, 3).

We also cultured C. difficile from the discrepant specimens that were detected by the Xpert C. difficile but not by the Illumigene C. difficile LAMP assay and sent these isolates to Meridian for further investigation to determine if the C. difficile strains could be detected by the primers used in the Illumigene C. difficile LAMP assay. Meridian communicated that the C. difficile isolates could be detected by the Illumigene C. difficile LAMP assay, suggesting to us that the discrepant results may have been due to a difference in the analytical sensitivities of these two NAATs (Illumigene C. difficile and Xpert C. difficile) and not due to a lack of amplification because of DNA sequence polymorphisms. In fact, after our publication, C. E. Gyorke et al. showed that the Illumigene C. difficile LAMP and Xpert C. difficile assays performed similarly for toxin-positive samples but that the Illumigene C. difficile LAMP assay was less sensitive with toxin-negative samples (58% for the Illumigene C. difficile LAMP assay versus 100% for the Xpert C. difficile assay) (2). The authors investigated this difference and showed that it correlated with the limit of detection (LOD) of each assay, showing that all clinical specimens with discrepant results had low C. difficile DNA concentrations. They concluded that the Xpert LOD was 1.2 log₁₀ DNA copies/ml higher than that of the Illumigene C. difficile LAMP assay, in agreement with the LOD data provided in the manufacturer's package insert for each assay (4, 5). Furthermore, the authors also suggested that sample dilution steps in the preanalytical processing with the Illumigene C. difficile LAMP assay could also be responsible for decreasing the sensitivity of the assay. While we are aware that the sensitivity of the Illumigene C. difficile LAMP assay has been reported to be higher, ranging from 92 to 98% (2, 3, 6–11), these results may be influenced by the method used for comparison and the inclusion of high numbers of specimens with elevated levels of C. difficile target DNA.