Detection of Clostridium difficile infection: a suggested laboratory diagnostic algorithm.

Currently, the diagnosis of Clostridium difficile infection (CDI) relies on the detection of toxins A and B in faeces but the sensitivity of these tests has been questioned, particularly in advanced disease. In this context, additional methods to enhance the diagnosis of C. difficile have been investigated. In this study, 1007 faecal samples are tested using toxigenic culture, an immunoassay for toxins AB and the C. difficile-specific glutamate dehydrogenase (GDH) test. Samples positive by any of the above tests are evaluated for the presence of faecal lactoferrin as an indicator of intestinal inflammation. Patients with evidence of inflammation but with negative toxin AB tests are followed up to assess clinical outcome. The toxin AB test was positive in 35 samples (3.4%), while 121 (12%) samples were culture-positive, 87 (8.6%) of which were toxigenic. Glutamate dehydrogenase proved to be a sensitive and specific marker of C. difficile with a negative predictive value of 99.3% (95% CI: 0.98-1.00). Faecal lactoferrin was positive in 52/129 (40.3%) samples tested. A cohort of 15 patients with a negative faecal toxin AB and a positive lactoferrin test was C. difficile culture-positive with a toxigenic isolate; clinically, all had advanced CDI. All demonstrated faecal toxin between five and 41 days later on repeat testing. It is suggested that a two-step algorithm be used to include screening faecal samples for GDH, with positive samples tested for faecal toxin AB and lactoferrin. Patients who present with a negative faecal toxin AB test and a positive lactoferrin test were serially tested for faecal toxin AB every five to seven days until a diagnosis was established. More sensitive tests than enzyme-linked immunosorbent assay (ELISA) for the detection of faecal toxin, or the use of a rapid specific test for the presence of a toxigenic strain, must be considered in such patients.