Clostridium difficile is a particularly troublesome 'superbug' that commonly affects people in healthcare facilities who have received antimicrobial medication. With the normal gut flora destroyed by antibiotic treatment, C. difficile quickly overruns the colon and releases toxins that cause bloating, diarrhea and abdominal pain. Left unchecked, C. difficile infection (CDI) can cause pseudomembranous colitis (a severe inflammation of the colon) and in rare cases will progress to a life-threatening form of colonic swelling called toxic megacolon.

C. difficile is extremely transmissible and can quickly spread. Standard diagnostic tests for CDI are slow to produce results (cytotoxin assays requiring cell cultures) or are very expensive and require special equipment and expertise (nucleic amplification tests). The resultant delays in diagnosis allow ample time for CDI to become a ward-wide outbreak.

CDI-associated diarrhea has a characteristic scent. So distinct is the odor that nursing staff have been known to detect potential cases by smell. It was with this fact in mind that researchers at the VU University Medical Center (Netherlands) enlisted the help of a beagle named Cliff.

In a proof of principle study, Marije Bomers and her colleagues examined whether a dog's superior sense of smell could be used to detect C. difficile in stool samples and hospital patients (BMJ published online 13 December 2012; doi:10.1136/bmj.e7396).

After two months of training, the dog was presented with two empirical tasks. First, Cliff was exposed to 100 stool samples (50 positive for C. difficile, 50 negative) and correctly identified all 50 positive samples and 47 of 50 negative samples (100% sensitivity and 94% specificity). Cliff was then taken to two hospital wards to test his ability to detect C. difficile in patients. Here he correctly identified 25 of 30 cases (83% sensitivity) and 265 of 270 controls (98% specificity).

Credit: iStockphoto.com/GlobalP

The authors note that Cliff was quick and efficient, spending less than ten minutes with each patient. They also point out some limitations to using an animal as a diagnostic tool, including the potential to spread infection and the possible unpredictability in behavioral responses to other stimuli.

This study was the first to demonstrate that a dog can be used to detect C. difficile in stool samples and in patients. Cliff's diagnostic accuracy with the stool samples suggests that immediate CDI identification is achievable. To a slightly lesser degree, using dogs to identify CDI on the clinical ward may be possible too.