STUDY OBJECTIVE: Comparison of two commercially avail-able tests for the detection of Clostridium difficile Glutamate Dehydrogenase (GDH) and toxins A and B for their sensitivity and specificity. MATERIAL AND METHODS: Eighty-six stool samples from patients hospitalised in the Motol University Hospital were analysed. GDH and toxins A and B were assayed in parallel by two tests: C. difficile Quik Chek Complete® (Techlab, USA) and Liaison® C. difficile GDH and Toxins AαB (DiaSorin, USA). From the stool samples, nucleic acids were also isolated using the UltraClean® Fecal DNA kit (MoBio Laboratories, USA). The commercially available C. difficile Elite MGB® kit (Nanogen, Italy) was used for the polymerase chain reaction (PCR). Anaerobic culture on C. difficile selective medium (Oxoid) was performed for all positive samples at least in one test. Pure isolates were characterized by PCR ribotyping. RESULTS: Thirty-six (42%) samples were GDH negative and toxin A/B negative by both tests. Twenty (23%) samples were GDH positive and toxin A/B positive by both tests. Nine (10%) samples were GDH positive and toxin negative by both tests, but were positive by PCR. Eleven (13%) samples that were GDH positive and toxin negative by both tests remained negative by PCR. Six (7%) samples only were GDH positive and toxin positive by the Liaison® test alone. Four (5%) samples were GDH-positive by theLiaison® test alone. Culture failure was observed in 11 (13%) samples, of which seven were positive by PCR. PCR was inhibited in five (6%) samples. The following toxigenic ribotypes: AI-3, 001, 002, 012,014, 017, 020, 049, 054, 078, 176, 203, and 413 and non-toxigenic ribotypes: AI-34, AI-61, 010, 485, 495, and 596 were identified. CONCLUSION: The Liaison® test had seven percent higher sensitivity for the detection of toxins A/B. The two-step protocol of the tests is also cost-saving. The savings can be used e.g. for incorporating the PCR techniques into the diagnostic algorithm of the laboratory.