Two-year audit of computed tomographic colonography in a teaching hospital: are we meeting the standard?

We aimed to determine whether adopting the published recommendations has led to successful implementation of computed tomographic colonography (CTC) in a teaching hospital setting outside the context of a clinical trial.

An audit of all the CTC examinations between April 2005 and June 2007 was conducted to determine the following: adequacy of bowel preparation, CTC indications and findings (compared with available colonoscopy), complications and experience of reporting radiologist.

The most common indications for the 111 CTC patients reviewed included exclusion of synchronous colonic tumours, incomplete colonoscopy and altered bowel habit. Only 16% of ascending colon/caecal segments was clear of faecal or fluid contamination. The rectum and sigmoid colon were free of contamination in 78% and 74% of cases respectively. Appropriately trained radiologists reported 91% of studies. Thirty-two percent of studies were normal. The most common positive findings were diverticular disease or a rectal tumour. Sensitivity, specificity and positive predictive value were 89%, 94% and 90% respectively (all polyps) with a sensitivity of 98.5% for lesions > 5 mm in size. Twenty-five percent of patients had extra colonic abnormalities. There were no recorded complications.

Our CTC practice is within accepted published guidelines. Bowel preparation is suboptimal in a significant proportion of cases and faecal tagging is being implemented.