Format

Send to

Choose Destination
J Hosp Infect. 2010 Jan;74(1):48-54. doi: 10.1016/j.jhin.2009.08.014. Epub 2009 Nov 8.

Two-step glutamate dehydrogenase antigen real-time polymerase chain reaction assay for detection of toxigenic Clostridium difficile.

Author information

1
Guy's & St Thomas' NHS Foundation Trust, Westminster Bridge Road, London SE1 7EH, UK. simon.goldenberg@gstt.nhs.uk

Abstract

Current diagnosis of Clostridium difficile infection (CDI) relies upon detection of toxins A/B in stool by enzyme immunoassay [EIA(A/B)]. This strategy is unsatisfactory because it has a low sensitivity resulting in significant false negatives. We investigated the performance of a two-step algorithm for diagnosis of CDI using detection of glutamate dehydrogenase (GDH). GDH-positive samples were tested for C. difficile toxin B gene (tcdB) by polymerase chain reaction (PCR). The performance of the two-step protocol was compared with toxin detection by the Meridian Premier EIA kit in 500 consecutive stool samples from patients with suspected CDI. The reference standard among samples that were positive by either EIA(A/B) or GDH testing was culture cytotoxin neutralisation (culture/CTN). Thirty-six (7%) of 500 samples were identified as true positives by culture/CTN. EIA(A/B) identified 14 of the positive specimens with 22 false negatives and two false positives. The two-step protocol identified 34 of the positive samples with two false positives and two false negatives. EIA(A/B) had a sensitivity of 39%, specificity of 99%, positive predictive value of 88% and negative predictive value of 95%. The two-step algorithm performed better, with corresponding values of 94%, 99%, 94% and 99% respectively. Screening for GDH before confirmation of positives by PCR is cheaper than screening all specimens by PCR and is an effective method for routine use. Current EIA(A/B) tests for CDI are of inadequate sensitivity and should be replaced; however, this may result in apparent changes in CDI rates that would need to be explained in national surveillance statistics.

PMID:
19900734
DOI:
10.1016/j.jhin.2009.08.014
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center