Clostridium difficile infections in a tertiary hospital: value of surveillance.

Clostridium difficile is a Gram-positive, spore-forming anaerobic bacillus associated with antibiotic-associated diarrhoea and antibiotic-associated colitis. It is widely distributed in the environment and colonises up to 3e5% of adult humans and 84% of healthy infants without causing symptoms.1 The source of C. difficile may be endogenous or environmental.2 Transmission from patient to patient is via contaminated hands of healthcare workers or contaminated patient environment or equipment.3 We report our recent experiences with C. difficile infections at the Steve Biko Academic Hospital. During the latter half of 2008, we observed an increase in stool specimens being submitted for routine analysis, including requests for C. difficile toxin. It was then decided that the data for the preceding months should be reviewed to inform us whether we were about to encounter an outbreak. In addition, we did not have an accurate knowledge of C. difficile infections at our hospital and the hospital did not have a written policy for the control and prevention of C. difficile infections. A review of laboratory records was initially done for the period January to June 2008 becausewe noticed an increase in the number of stool specimens being submitted in June for the diagnosis of C. difficile infection. When an increase in the number of toxin-positive stool specimens was observed in August, we contacted the infection control nurses and planned our intervention. We drew up guidelines for the prevention and control of C. difficile infections, communicated these to the clinicians and hospital management staff and held training sessions with ward staff members. The diagnosis of C. difficile infection was by commercial C. difficile toxin A test assay (Oxoid, Basingstoke, UK) which detects the presence of toxin A in stool specimens.4 Data were analysed to compare the number of requests and the results from month to month, distribution of positive results per ward and to see which hospital areas were affected. A total of 266 stool specimens were received for the whole year and 46 specimens (17.2%) tested positive. The 46 positive specimens were from 26 inpatients. An increase in the number of toxin positives was first seen in May, and in June the number of specimens submitted also increased (Figure 1). The months of July and August showed an increase both in specimens submitted and toxin-positive specimens (Figure 1). Therefore guidelines were drawn up and were implemented from the month of August. The subsequent months showed a plateau with some decrease in the number of toxin positives whereas December showed a very high number of requests but no real increase in the toxin-positive specimens. Infection control and prevention activities were increased in the later months of the year (August