CT Virtual Colonoscopy to Screen for Colorectal Neoplasia in Asymptomatic Adults

Abstract & Commentary

Synopsis: When analyzed using 3-dimensional methods, virtual colonoscopy achieves comparable accuracy in screening asymptomatic adults for colonic polyps as optical colonoscopy.

Source: Pickhardt PJ, et al. N Engl J Med. 2003; 349:2191-2200.

Pickhardt and colleagues compared the per-formance of computed tomographic (CT), "virtual" colonoscopy with standard optical colonoscopy for the detection of colorectal neoplasia in an asymptomatic population. A total of 1233 adults aged 50-79 years underwent same-day virtual colonoscopy followed immediately by optical colonoscopy. The sensitivity of virtual colonoscopy for adenomatous polyps was 93.9% for polyps > 8 mm in diameter and 88.7% for those > 6 mm. The sensitivity of optical colonoscopy was 92.2% and 79.6%, respectively. Also, CT identified clinically important, extra-colonic findings in 56 patients, 5 of which turned out to be cancerous. Two abdominal aortic aneurysms were found and repaired. The mean time spent by patients was 14 minutes for CT and 31 minutes for optical colonoscopy. Most patients preferred the CT colonoscopy, even though they rated it as equally uncomfortable (because the patient must introduce sufficient air to achieve pneumocolon for the CT to be readable). The only apparent drawback to virtual colonoscopy is that in practice, polyps or other significant lesions identified on screening virtual colonoscopy would require optical colonoscopy for biopsy afterward. In contrast, with optical colonoscopy, the biopsy can be done at the time of the procedure. The other limitations to the use of virtual colonoscopy are the need for dedicated training of radiologists and technologists and the lack of availability of the software systems that permit 3-dimensional analysis.

Comment by Sarah L. Berga, MD

This article grabbed my attention because I had just seen a glaring example of direct-to-consumer advertising in the form of a huge billboard along the interstate advocating virtual colonoscopy. I had not read too much about this technique in the medical literature, as most of the published debate has focused on the pros and cons of occult heme testing from stool samples vs standard sigmoidoscopy vs optical colonoscopy. Although I was not necessarily the intended consumer, the billboard worked well in that it garnered my professional attention. Interestingly, most professional organizations that make screening recommendations do not even endorse optical colonoscopy because of the cost and risk. Rather, they generally advocate sigmoidoscopy or occult heme testing. However, optical colonoscopy does have a higher sensitivity, if only because more of the colon is examined and colonic cancers are evenly distributed throughout the length of the colon. (I like to think of a sigmoidoscopy as similar to doing a screening mammogram of only 1 breast.) Since colorectal cancer is the second leading cause of cancer-related death, it is prudent to recommend some form of screening. In average-risk patients, the guidance has been to start screening at age 50 years.

Most colorectal cancers are believed to arise within benign adenomatous polyps, the removal of which markedly decreases the incidence of colorectal cancer. In the discussion of the present article, Pickhardt et al note that the number of patients who would require subsequent optical colonoscopy for removal of polyps identified on virtual colonoscopy depends on the recommendation regarding the size of the polyp that must be removed. If the cut-off is 6 mm, then 30% of patients in the present study would have required follow-up optical colonoscopy. Pickhardt et al recommend removal for polyps > 8 mm, in which case only about 15% of the study population would have required optical colonoscopy. Of note, only about half of the eligible population has not undergone screening of any type. Thus, a technique that works and has high patient acceptance with low medical risk would represent a true medical advance.

The erstwhile OB/GYN who has time to do appropriate well-care counseling clearly should include a discussion of screening for colon cancer. However, to do this, one has to have both an opinion about and the time to make a recommendation. The present article is intended to help with forming an opinion. I have still not conquered the time barrier. Assuming comparable cost (a topic not covered in the article), it would appear that virtual colonoscopy has higher patient acceptance and comparable sensitivity and clearly can be recommended to interested patients older than 50. If the first optical colonoscopy is negative in an average-risk, asymptomatic individual, current recommendations suggest that the next one be done in 10 years. Given the comparable sensitivity, one assumes that a similar recommendation would hold for a negative virtual colonoscopy.

Dr. Berga is James Robert McCord Professor and Chair, Department of Gynecology and Obstetrics, Emory University School of Medicine, Atlanta, Ga.