Comparison of the TOX A/B test to a cell culture cytotoxicity assay for the detection of Clostridium difficile in stools

The TOX A/B Test (Techlab, Blacksburg, VA, USA) was compared to cell cultur e cytotoxicity assay on 1109 consecutive diarrheal stool samples collected from patients with the presumptive diagnosis of Clostridium difficile disea se. The TOX A/B Test is an enzyme immunoassay in a microtiter format that d etects both toxins A and B. The procedure used for this study takes approxi mately 1.5 h to perform. Cell culture cytoxicity was performed by using a f ibroblast cell line in a microtiter format read at 4 h, 24 h, and 48 h. One hundred ninety-four of the 1109 samples were positive by the "gold stan dard" cytoxicity assay, whereas 189 were positive by EIA. There was a 98.5% agreement between the two assays. When compared to the cytoxicity assay, t he EIA had an initial sensitivity of 94.3% and a specificity of 99.3%. Howe ver, after resolution of six discrepants using another ELISA for toxin A de tection the sensitivity, specificity, positive and negative predictive valu es for the TOX A/B test are as follows: 94.5%; 100%; 100%; 98.8%. The corre sponding values for the cytotoxicity assay are: 97%; 100%; 100%; and 99.3%. This test seems to have excellent sensitivity and specificity as compared to an in-house cell culture cytotoxicity assay. It is sensitive enough to u se as a stand-alone test for the detection of C. difficile toxin in laborat ories that do not have cell culture cytotoxicity testing capability. (C) 20 00 Elsevier Science Inc. All rights reserved.