Clostridium difficile: value of the GDH, toxin and tcdB testing algorithm

Clostridium difficile has been identified as the primary cause of nosocomial diarrhoea worldwide. Symptoms can vary from life-threatening pseudomembranous colitis (PMC) and toxic megacolon to mild diarrhoea and asymptomatic carriage.¹ C. difficile infection (CDI) is associated with high rates of morbidity and mortality, which in recent years has increased clinical and public awareness. This has subsequently highlighted the importance of appropriate patient management and infection control measures aided by the rapid and accurate diagnosis of CDI.

The clinical manifestations of CDI are caused by the effects of the C. difficile exotoxins commonly known as toxins A and B. Numerous studies over the past two decades have concluded that the use of a toxin A/B enzyme immunoassay (EIA) as a single test was inadequate due to poor specificity.^2–4 Eastwood et al.⁵ found that a nucleic acid amplification test (NAAT) for the toxin B gene (tcdB) gave a negative predictive value (NPV) of 99% and was the optimal single test method for excluding disease from the nine assays tested. However, NAATs only confirm the presence of a specific gene, not the toxin itself, and therefore only indicate an organism’s potential to produce toxin. The detection of free toxin is still regarded by many as critical for diagnosis of CDI, as this correlates best with severity and higher complication rates.^6,7

In 2009, as a result of mounting evidence, the Advisory Committee on Antimicrobial Resistance and Healthcare Associated infection and the Department of Health (ARHAI/DH) advised that laboratories should not rely solely on a single test for the accurate diagnosis of CDI. A two-step algorithm was recommended and this approach is supported by the European Society of Clinical Microbiology and Infectious Disease (ESCMID).⁸