Comparison of three enzyme immunoassays for detection of Clostridium difficile toxins A and B

AIM OF THE STUDY: Diagnosis of Clostridium difficile-associated diarrhea is classically based on detection of toxin A and/or toxins A+B by using several techniques. However, these techniques show important differences in terms of sensitivity and specificity. In this work, we compared three commercial immunoenzymatic tests for detecting C. difficile toxins. METHODS: We used three immunoenzymatic techniques: ELISA Premiertrade mark Toxins A and B (Meridian), ImmunoCard (IC) Toxins A and B (Meridian), and Triage (TRI) Antigen GDH and Toxin A (Biosite). Culture on the specific media C. difficile (bioMérieux) was performed in parallel. RESULTS: During two years, 898 stool specimens have been tested by using the three different techniques. According to the manufacturer's recommendations (sample positive if optical density [OD] value greater or equal to 0.15), 205 (22.8%) were positive for ELISA. Among them, 65 (31.7%) were negative for all other tests, and they have been considered as false-positive results. This discrepancy led us to choose other OD values (greater than 0.75: positive result; 0.15-0.75: limit result). For limit results, IC, TRI, antigen GDH, and culture methods were positive in 30, 2, 41, and 29% of cases, respectively, whereas for positive results (>0.75), they were positive in 82, 54, 84, and 76% of cases, respectively. CONCLUSION: ELISA and IC tests are the most powerful and are concordant together if interpretation is performed with OD values redefined by the microbiology laboratory. The choice of the technique to use depends on the number of samples to analyze, the rapidity of the result, and the cost.