Archief Roodbont

Computed tomography (CT) colonography is a less-invasive and better-tolerated alternative to colonoscopy as a colorectal cancer (CRC) screening tool in average-risk individuals. However, we know less about its accuracy in detecting advanced colorectal neoplasia in individuals at excess risk for CRC.

Now, researchers have assessed the accuracy of CT colonography in this setting, using unblinded colonoscopy as the reference standard. The European multicenter cross-sectional study involved 937 participants who underwent same-day CT colonography and colonoscopy. All participants had excess risk for CRC, defined by the study as having first-degree relatives with diagnoses of advanced colorectal neoplasia (the family-history group; age range, 40–65), having undergone endoscopic removal of colorectal adenomas (the postpolypectomy group; age range, 18–70), or having positive results on fecal occult blood tests (FOBT; the FOBT-positive group; age range, 59–69). The primary and secondary endpoints were sensitivity, specificity, and positive predictive value (PPV) and negative predictive value (NPV) of CT colonography for advanced neoplasia 6 mm and 10 mm, respectively.

A per-patient analysis showed that CT colonography’s sensitivity and specificity for all patients with advanced neoplasia 6 mm were 85.3% and 87.8%, respectively, and its PPV and NPV were 61.9% and 96.3%. For all patients with advanced neoplasia 10 mm, CT colonography’s sensitivity and specificity were 90.8% and 84.5%, respectively, and its PPV and NPV were 48.8% and 98.3%. Of the three patient subgroups, CT colonography’s specificity and NPV were lowest in the FOBT-positive group (76.4% and 84.9%, respectively). Prevalence of neoplasia 6 mm was 7.5% in the family-history group, 11.1% in the postpolypectomy group, and 50.2% in the FOBT-positive group.

Comment: I find these results disappointing for several reasons. First, they are barely as good as those of the American College of Radiology Imaging Network trial, despite a population prone to neoplasia. Second, a high-prevalence population is bound to elicit higher sensitivity. Third, the idea of performing CT colonography in a population such as the FOBT-positive group is untenable from a cost-effectiveness or efficacy standpoint. Also, both the authors and an editorialist suggest that the lesser performance of CT colonography is a reasonable trade for a test that would improve patient acceptance of CRC screening. This suggestion would be reasonable, except that no evidence exists to show that CT colonography would improve acceptance. This study, despite receiving some attention in the lay press, does not advance our understanding of the best role for CT colonography, which I believe is screening patients with a low preprocedure probability of advanced neoplasia.

— Douglas K. Rex, MD

Published in Journal Watch Gastroenterology June 26, 2009