Computed axial colonography, usually called virtual colonoscopy, is a new diagnostic method potentially useful for investigating polyps and tumors of the colon and rectum. It uses spiral axial tomography data to build up images similar to those given by conventional colonoscopy, offering advantages in that it causes no adverse reactions, and is accepted well by patients, who do not need to be sedated. Reports to date show its sensitivity and specificity vary in identifying polyps in the colon, and there are still many problems with the new method, such as its cost, its learning curve, the identification of flat lesions, and detection of extracolonic abnormalities. There is a pressing need to establish the accuracy and cost-effectiveness of virtual colonoscopy as a screening method for colorectal tumors, and for post-polypectomy surveillance.