In Reply: We agree with Dr Matuchansky that flat lesions are important to consider regarding colorectal cancer development and prevention, particularly in persons with a positive family history. In our study, we excluded patients with the highest family risk of cancer: those with criteria for hereditary nonpolyposis colorectal cancer syndrome, who have the highest likelihood of developing aggressive flat lesions. Our article did not consider the diagnostic value for flat lesions, but these data are under analysis. Preliminary results1 for 738 individuals with polyps found 39 flat lesions (5.3%) according to the definition used by Matuchansky; this prevalence is consistent with what would be expected having excluded individuals with the highest risk of flat lesions. The study by Soetikno et al2 found a 9.4% average prevalence of flat/depressed lesions in a mixed population, ranging from 5.8% to 15.4% depending on the risk category. Sensitivity of CT colonography for flat lesions in the 738 patients from our series was 61% (24/39); it was 90% (35/39) when combining stand-alone computer-assisted diagnosis with radiologist reading. Because CT colonography is not considered an appropriate test for surveillance of patients at risk of colorectal cancer because of long-standing ulcerative colitis, these patients were excluded from our study.
Galatola G, Laudi C, Regge D. Computed Tomographic Colonography for Detecting Advanced Neoplasia—Reply. JAMA. 2009;302(14):1527–1529. doi:10.1001/jama.2009.1436