Tests for the diagnosis of Clostridium difficile infection: the next generation.

Clostridium difficile (C. difficile) causes 25-30% of cases of antibiotic associated diarrhea and most cases of pseudomembranous colitis. Patients presenting with diarrhea after hospitalization for 3 or more days should be tested for C. difficile. There are many options available for testing, each of which has inherent advantages and disadvantages. Most laboratories perform toxin testing using an enzyme immunoassay method. In general these tests have sensitivities ranging from 60 to 70% and specificities of 98%. When using these methods, symptomatic patients with negative tests should be tested by another more sensitive method. Until recently, cell culture cytotoxicity neutralization assays (CCNAs) were considered the gold standard in the U.S. A two-step algorithm using an EIA for glutamate dehydrogenase detection followed by testing positives using CCNA, offered an improved alternative until the availability of molecular assays. Although early studies that compared the GDH assay to CCNA demonstrated high sensitivity and negative predictive values, more recent comparisons to toxigenic culture and PCR have shown the sensitivity to be in the mid to high 80's. When testing using a sensitive assay, repeat testing is not cost-effective. Outbreaks caused by a toxin variant epidemic strain have renewed interest in bacterial culture. Toxigenic culture has emerged as the new gold standard against which newer assays should be compared. However, there is no agreed upon standard method for culture performance. At least 4 FDA cleared nucleic acid amplification assays are available to clinical laboratories and several of these have been well evaluated in the literature. Because these assays detect a gene that encodes toxin and not the toxin itself it is important that laboratories test only patients with diarrhea. These molecular assays have been shown to be superior to toxin EIAs, CCNA and 2-step algorithms, but not to toxigenic culture. More studies are needed to assess the impact of molecular tests on treatment and nosocomial spread of Clostridium difficile infections.