Abstract
Aim: Clostridium difficile is the major pathogen associated with nosocomial diarrhoea. We evaluated the performances of a commercially available toxin A enzyme immunoassay (EIA; Clearview C. difficile Toxin A Test), culture and tissue culture cytotoxin assay in the diagnosis of C. difficile-associated diarrhoea. Methods: Comparative test performance was determined from data obtained from 166 faecal samples. The initial analysis compared the performance of toxin A EIA and culture with that of cytotoxin assay, this being defined as a 'laboratory gold standard'. A second analysis compared the individual performance of the toxin A EIA, culture and cytotoxin assay using a combined clinical and laboratory diagnostic assessment as a 'clinical gold standard'. In a parallel, study three selective culture media were compared. Results: From the initial analysis, the sensitivity and specificity of the methods were, respectively, 84.6 and 65.4% for the toxin A EIA, and 38.5 and 93.5% for culture. From the second analysis, the sensitivity and specificity of the methods were, respectively, 100 and 67.5% for the toxin A EIA, 63.6 and 96.7% for culture and 72.7 and 98.0% for cytotoxin assay. Media containing d-cycloserine 250 mg/L and cefoxitin 8 mg/L performed best, growing 88.2% of the isolates. Conclusion: The toxin A EIA we evaluated had poor specificity in the diagnosis of C. difficile-associated diarrhoea. We conclude that in our laboratory the combination of culture and cytotoxin assay is a preferred approach to the diagnosis of C. difficile-associated diarrhoea.
