Laboratory testing considerations for C. difficile disease

During the brief history of Clostridium difficile as a pathogen, there have been a remarkable number of diagnostic laboratory testing strategies for disease related to this organism. Tests for diagnosis of C. difficile disease have been based on detection of the organism, its toxins, other cellular antigens, and most recently toxin-specific genes. Upon recognition of C. difficile as the major cause of antibiotic-associated diarrhea in the late 1970s, emphasis was initially placed on recovery of the organism. Effective culture techniques were developed based on selective isolation, allowing recovery of C.difficile in the presence of other fecal flora. Cycloserine-cefoxitin-fructose agar was one medium used in the early techniques, and it is still in use today. Due to the existence of toxigenic and nontoxigenic strains, it became apparent that the laboratory needed to assay directly for the toxin, which could be done by cytotoxicity testing using a variety of cell lines, with antitoxin neutralization. Eventually immunodiagnostic assays became commercially available; thus, the testing was facilitated for the majority of clinical laboratories. The status of C. difficile testing in 2005 is that strategies vary widely and laboratories continue to employ one or more of the above methods. Although there is limited standardization of diagnostic strategies, there is consensus concerning such matters as timing and frequency of testing. Basic guidelines have been suggested to promote clinically relevant, cost-effective testing for C. difficile by hospital laboratories (Table).

Table.

Testing recommendations for Clostridium difficile

Do	Don't
—Submit stool samples for testing from symptomatic patients.	—Submit stool samples for testing from asymptomatic patients.
—Avoid repeat testing. Submit one, or a maximum of two, specimens per patient.	—Perform C. difficile testing on patients <6 months old or “test-of-cure” toxin assay on any patients.
—Consider C. difficile instead of routine microbiologic studies for inpatients over 6 months of age.	—Test stool samples of patients with diarrhea after the third hospital day for Salmonella, Shigella, Campylobacter, or parasites culture without first considering C. difficile.

In his presentation (1), Dr. Fordtran highlights two features of the organism that have had a direct impact on the direction that laboratories will need to take to improve diagnostic assays. current immunoassays offered by most laboratories are woefully First, the increasing incidence of C. difficile disease justifies avail-inadequate in terms of diagnostic sensitivity. This has been espeability of on-site, rapid, and accurate testing in most hospital cially borne out by recent studies, which have begun evaluation laboratories. A testing protocol that includes immunoassays for of polymerase chain reaction (PCR) as a means for rapid and both toxins A and B is feasible, and that is the current standard of practice. Numerous rapid tests are commercially available, and many have been clinically evaluated (2, 3). The clinical significance of toxin A-negative/toxin B-positive variants (4) warrants use of immunoassays for both toxins, while other screening tests that rely on only one of the toxins, or on antigens other than toxins (5), should be avoided. A European survey found that 58% of laboratories continued to perform an immunoassay for only toxin A, even after commercial immunoassays for toxins A and B were introduced (6). Second, the emergence of strains with enhanced virulence demands attention from laboratorians to enhance clinicians' ability to diagnose C. difficile disease. The organism's pathogenic properties have shifted more than once with respect to toxin production, altering the clinical features and diagnostic requirements for C. difficile disease. It is now known that some strains produce a third toxin (binary toxin) and that certain strains may produce all three toxins (7) or an unregulated quantity of toxin.

Another point that Dr. Fordtran candidly makes is that the current immunoassays offered by most laboratories are woefully inadequate in terms of diagnostic sensitivity. This has been especially borne out by recent studies, which have begun evaluation of polymerase chain reaction (PCR) as a means for rapid and sensitive detection of C. difficile (8–10). The sensitivity of the direct fecal toxin immunoassay was below 60% in one study when compared with PCR detection of toxin A and B genes (10). The latter, combined with culture, is the approach referred to as the toxigenic culture. PCR is preferred over immunoassay for the toxins in the toxigenic culture because the immunoassays tend to be less specific when testing isolates, resulting in overdiagnosis of toxin-mediated C. difficile disease. Laboratories with advanced diagnostic capabilities are now offering this approach, and it will be increasingly advocated as the laboratory service of choice for diagnosing C. difficile disease. A practical algorithm would seem to be for laboratories to continue performing immunoassays for toxins, since their specificity is high with direct fecal assays. If the initial test result is negative, the toxigenic culture and PCR could either be reflexively performed or clinicians could consult the laboratory for a toxigenic culture when suspicion of C. difficile is high.