Introduction C.difficile is a Gram-positive anaerobe that can cause life-threatening diarrhoea. There is no current method with which to identify pathogenic C.difficile colonisation (PCDC) quickly. In order to be pathogenic, C.difficile must have the ability to make toxin B. We have designed and validated an rtPCR assay targeting the toxin B with which to diagnose PCDC quickly using a perianal swab.
Methods Perianal swabs were taken prospectively from 99 patients with proven stool culture-positive C.difficile infection (CDI) within 24 hours of diagnosis. DNA from swab tips was extracted using the automated QIAsymphony platform. Half of the DNA extracts from patients with CDI along with control C.difficile DNA were used to optimise an rtPCR assay using primers targeting the C.difficile toxin B gene. PCRs were run as 10μL reactions on 96 well plates in duplicate and the volume of extract per reaction was varied (1–3μL). PCR positive was defined as: amplification curve crossing the threshold at <40 cycles, <0.5 CT difference between duplicates, melting points to be ± 1.25 OC of that of C.difficile DNA and no amplification in negative controls. PCR positivity was compared to culture result as the gold standard. Once optimised, the assay was validated using the remaining 48 DNA extracts from CDI patients and extracts from swabs taken from 11 control patients with C.difficile-negative diarrhoea.
Results In optimisation, assay sensitivity increased as more DNA extract was added (40%, 62% and 78% for 1μL, 2μL and 3μL respectively, p = 0.0042). The 3μL assay showed 92% efficiency and linearity of 0.997 was selected for the validation study. For validation using perianal swabs from 48 patients with proven CDI and 11 C.difficile stool culture-negative patients, assay sensitivity was 69%, specificity 93%, positive predictive value (PPV) 97% and negative predictive value (NPV) 48%.
Conclusion We designed an rtPCR assay to identify C.difficile toxin B in DNA extracted from perianal swabs taken from patients with diarrhoea. Although the sensitivity in the validation experiments was low (69%), the high specificity and PPV mean that the assay may have clinical applicability. As almost all patients testing positive with the PCR swab test will be colonised with pathogenic C.difficile, they could be isolated on hospital admission to reduce the chance of nosocomial spread. Patients testing positive could also be prospectively followed to see if colonisation with pathogenic C.difficile on admission is a risk factor for subsequent C.difficile-induced diarrhoea (CDI), poorer outcome if diagnosed with CDI, or relapsing disease.
Disclosure of Interest None Declared