Can We Diagnose C. difficile Diarrhea with One Sample and One Test?

Abstract & Commentary

By Ellen Jo Baron, Ph.D., D(ABMM), Professor Emerita, Stanford University School of Medicine Director of Medical Affairs, Cepheid, Sunnyvale, CA Dr. Baron is a consultant for MorphDesign, is a speaker for bioMerieux, is a stockholder for Cepheid, Key Scientific Products, and Immunogenetics, is Director of Medical Affairs for Cepheid, and is on the Scientific Advisory Board for OpGen, Key Scientific Products, Immunogenetics, and NanoMR. This article originally appeared in the December 2010 issue of Infectious Disease Alert. It was edited by Stan Deresinski, MD, FACP, and peer reviewed by Timothy Jenkins, MD. Dr. Deresinski is Clinical Professor of Medicine, Stanford; Associate Chief of Infectious Diseases, Santa Clara Valley Medical Center, and Dr. Jenkins is Assistant Professor of Medicine, University of Colorado, Denver Health Medical Center. Dr. Deresinski serves on the speaker's bureau for Merck, Pfizer, Wyeth, Ortho-McNeil (J&J), Schering-Plough, and Cubist, does research for the National Institute of Health, and is an advisory board member for Schering-Plough, Ortho-McNeil (J&J), and Cepheid, and Dr. Jenkins reports no financial relationships relevant to this field of study.

Synopsis: Clinicians who send samples to the 95% of U.S. laboratories that test for Clostridium difficile infection using an enzyme-immunoassay (EIA) for toxins A and B are often frustrated by a negative laboratory result that doesn't fit their clinical impression or their olfactory suspicions. A recent spate of publications has shown that the EIA tests are shamefully insensitive vs. the new gold standard, recovery of the organism on anaerobic culture and subsequent detection of toxin production by the C. difficile isolate recovered using a cell-culture cytotoxin neutralization method. A new study shows that if a reliable test is used, in this case nucleic acid amplification of the toxin B gene sequence by PCR, an initial negative result for toxigenic C. difficile can be trusted, and a repeat test is useful only for a subset of patients with new onset of symptoms ≥ 7 days later.

Source: Luo RF, Banaei N. Is repeat PCR needed for diagnosis of Clostridium difficile infection? J Clin Microbiol. 2010;48:3738.

The authors tested samples sent to Stanford Hospital microbiology laboratory for diagnosis of C. difficile infection (CDI). That laboratory had been using the cell culture cytotoxin neutralization test as a first assay, followed by a bioMerieux Vidas™ EIA method for toxins A and B if the cell culture exhibited non-specific toxicity. Although the laboratory had decreed no repeat testing within 7 days of a positive test, there were no rules for negative test results. Apparently, physicians were sending multiple stool samples, as shown by the data presented in the publication.

Toxin tests are now known to lack sensitivity for diagnosis of CDI.1,2,3 Even use of a glutamate dehydrogenase initial test (thought to be highly sensitive but known to be non-specific) required either a molecular or cell culture-based confirmatory test to yield reliable results, which took more than a day to complete.1,4 A straight-talking paper by Peterson and Robicsek explained, with an example, how the predictive value of a positive result fell rapidly with repeated tests on the same patient when the sensitivity of the test was 73%, specificity was 97%, and prevalence was 10%.5 And based on numerous studies, we now know that the sensitivity of EIA tests is more akin to 50%.1,2,3 A previous publication from Mayo Clinic also found that repeat testing (using another method) was not clinically helpful.6

The authors' laboratory-developed PCR assay had a sensitivity of 87.2% and a specificity of 98.6%, compared with toxigenic culture, the gold standard. After implementation of the new PCR test, previous ordering patterns continued. The samples were received over 6 months in 2009, during which time both toxigenic cultures and PCR were performed on all samples. A total of 1,287 patients were initially evaluated, mostly adults. Almost one-fourth of all patients (293) had at least one repeat specimen sent for testing. There were a total of 405 repeat samples received, or an average of 1.5 samples per patient tested for CDI. Only 10 samples yielded a new positive result after an initial negative result, one of which was considered to be a false-positive. Among the new positive results, seven occurred at 7 or more days later. Chart reviews of those patients showed medical reasons, relapse, or new onset of diarrhea in most cases. For all negative test results that had another sample submitted for the same test within the first 7 days, only 1% yielded a different (positive) result. In fact, overall, 97.5% of all tests remained negative on repeat(s). There may be a small subset of patients who exhibit new disease symptoms after 7 days for whom another molecular test for CDI could be valuable.

Finally, a very recent paper evaluating results of a seven-site study using a commercially available real-time PCR assay compared with various other test methods showed that the performance of EIA and GDH assays varied widely depending on the ribotype of C. difficile tested.7 For the 027 epidemic strain, known to produce higher toxin levels and many more spores, the sensitivity of several EIA assays was relatively higher, but not to the level of toxigenic culture or PCR.7 But even GDH no longer appears to be equally robust for all strains of C. difficile, with sensitivity dropping to 72% overall for stools containing non-027 strains vs. 91% for stools from patients harboring 027 strains.3,7 This startling new fact may help explain published differences in sensitivity and specificity of the same test method in different settings. Fortunately for patients and physicians, the new molecular tests, similar to the old toxigenic culture test methods, seem to perform well regardless of the strain type of the pathogen. This is comforting to those whose laboratories have made the switch to molecular diagnostics, as recommended by several authorities.5,8


  1. Eastwood K, et al. Comparison of nine commercially available Clostridium difficile toxin detection assays, a real-time PCR assay for C. difficile tcdB, and a glutamate dehydrogenase detection assay to cytotoxin testing and cytotoxigenic culture methods. J Clin Microbiol. 2009;47:3211-3217.
  2. Doing KM, et al. Reevaluation of the premier Clostridium difficile toxin A and B immunoassay with comparison to glutamate dehydrogenase common antigen testing evaluating Bartels cytotoxin and Prodesse ProGastro polymerase chain reaction as confirmatory procedures. Diagn Microbiol Infect Dis. 2010;66:129-134.
  3. Novak-Weekley SM, et al. Clostridium difficile testing in the clinical laboratory using multiple testing algorithms. J Clin Microbiol. 2010;48:889-893.
  4. Gilligan PH. Is a two-step glutamate dehyrogenase antigen-cytotoxicity neutralization assay algorithm superior to the Premier toxin A and B enzyme immunoassay for laboratory detection of Clostridium difficile? J Clin Microbiol. 2008;46:1523-1525.
  5. Peterson LR, Robicsek A. Does my patient have Clostridium difficile infection? Ann Intern Med. 2009;151:176-179.
  6. Aichinger E, et al. Nonutility of repeat laboratory testing for detection of Clostridium difficile by use of PCR or enzyme immunoassay. J Clin Microbiol. 2008;46:3795-3797.
  7. Tenover FC, et al. Impact of strain type on detection of toxigenic Clostridium difficile: Comparison of molecular diagnostic and enzyme immunoassay approaches. J Clin Microbiol. 2010;48:3719-3724.
  8. Schmidt ML, Gilligan PH. 2009. Clostridium difficile testing algorithms: What is practical and feasible? Anaerobe. 2009;15:270-273.