Bo-Moon Shin1, Soo-Jin Yoo, Hye Jun Oh. 1. Department of Laboratory Medicine, Sanggye Paik Hospital, Inje University, Nowon-gu, Seoul, Korea. bmshin@unitel.co.kr
Abstract
BACKGROUND: Enzyme immunoassay (EIA) capable of detecting both toxin A and toxin B is strongly recommended for the diagnosis of Clostridium difficile associated disease. Therefore, we evaluated two different EIAs for the detection of C. difficile toxin A/B. METHODS: For a total of 228 stool specimens we performed bacteriologic cultures for C. difficile and examined for toxin A and toxin B using enzyme linked fluorescent immunoassay (ELFA; VIDAS CDAB, Bio-Merieux sa, France) and ELISA (C.DIFFICILE TOX A/B II, TECHLAB, USA). We also performed PCR assays for toxin A and B genes in 117 C. difficile isolates that grew from the stool cultures and compared the results with those obtained with the two different EIAs. RESULTS: The concordance rate between ELFA and ELISA was 85.5% (195/228). Using the culture and PCR results as the standard, the sensitivity/specificity of the ELFA and ELISA were 65.0%/72.1% and 71.8%/70.3%, and for positive/negative predictive values were 78.4%/69.6% and 71.8%/70.3%, respectively (P value >0.05). No differences were observed between the results of ELFA and ELISA with toxin A(-) toxin B(+) strains of C. difficile. CONCLUSIONS: The sensitivity of the ELISA was slightly higher than that of ELFA for toxin A and toxin B, but the specificity and positive predictive value of the ELFA were rather higher than those of the ELISA, although no statistical differences were observed. A bacteriologic culture and PCR assay for toxin genes are recommended in case the both EIAs are negative.
BACKGROUND: Enzyme immunoassay (EIA) capable of detecting both toxin A and toxin B is strongly recommended for the diagnosis of Clostridium difficile associated disease. Therefore, we evaluated two different EIAs for the detection of C. difficile toxin A/B. METHODS: For a total of 228 stool specimens we performed bacteriologic cultures for C. difficile and examined for toxin A and toxin B using enzyme linked fluorescent immunoassay (ELFA; VIDAS CDAB, Bio-Merieux sa, France) and ELISA (C.DIFFICILE TOX A/B II, TECHLAB, USA). We also performed PCR assays for toxin A and B genes in 117 C. difficile isolates that grew from the stool cultures and compared the results with those obtained with the two different EIAs. RESULTS: The concordance rate between ELFA and ELISA was 85.5% (195/228). Using the culture and PCR results as the standard, the sensitivity/specificity of the ELFA and ELISA were 65.0%/72.1% and 71.8%/70.3%, and for positive/negative predictive values were 78.4%/69.6% and 71.8%/70.3%, respectively (P value >0.05). No differences were observed between the results of ELFA and ELISA with toxin A(-) toxin B(+) strains of C. difficile. CONCLUSIONS: The sensitivity of the ELISA was slightly higher than that of ELFA for toxin A and toxin B, but the specificity and positive predictive value of the ELFA were rather higher than those of the ELISA, although no statistical differences were observed. A bacteriologic culture and PCR assay for toxin genes are recommended in case the both EIAs are negative.
Authors: Susan M Novak-Weekley; Elizabeth M Marlowe; John M Miller; Joven Cumpio; Jim H Nomura; Paula H Vance; Alice Weissfeld Journal: J Clin Microbiol Date: 2010-01-13 Impact factor: 5.948