Noncathartic CT colonography to screen for colorectal neoplasia in subjects with a family history of colorectal cancer.

ABSTRACT

PURPOSE: To prospectively assess the diagnostic performance of noncathartic computed tomographic (CT) colonography in the detection of clinically relevant colorectal lesions (≥6 mm polyps or masses) in a well-defined cohort of first-degree relatives of patients with colorectal cancer (CRC), using colonoscopy and histologic review as the standard of reference. MATERIALS AND METHODS: Institutional review board approval was obtained, and all subjects provided written informed consent. Consecutive patients admitted with CRC (index cases) were prospectively evaluated, and those who agreed to contact their first-degree relatives who were at least 40 years old were included. Available first-degree relatives were invited to undergo noncathartic CT colonography (200 mL of diatrizoate meglumine and diatrizoate sodium). Colonoscopy was performed the following day, and findings from CT colonography were disclosed for each segment. Sensitivity, specificity, and positive and negative predictive values of CT colonography were assessed for detecting subjects with any lesion at least 6 mm, any lesion at least 10 mm, and advanced neoplasia at least 6 mm. Colonoscopy with segmental unblinding and histologic diagnosis were used as the standard of reference. Matching between findings from CT colonography and colonoscopy was allowed when lesions were located in the same or adjacent colon segments and when the size difference was 50% or less. RESULTS: Three hundred four first-degree relatives (median age, 47 years; age range, 40-79 years; 46.7% women) identified from 221 index cases were included. Overall, CT colonography helped identify 17 of 22 subjects with polyps measuring at least 6 mm (sensitivity, 0.77; 95% confidence interval [CI] 0.59, 0.95) and helped correctly classify as negative 278 of 282 subjects without lesions measuring at least 6 mm (specificity, 0.99; 95% CI 0.97, 1.00). CT colonography helped detect eight of nine subjects with polyps measuring at least 10 mm as well as eight of nine subjects with advanced neoplasia measuring at least 6 mm (sensitivity, 0.89 for both). Per-subject positive and negative predictive values for lesions measuring at least 6 mm were 0.81 (17 of 21 subjects; 95% CI 0.65, 0.97) and 0.98 (282 of 287 subjects; 95% CI 0.96, 0.99), respectively. CONCLUSION: Noncathartic CT colonography is an effective screening method in first-degree relatives of patients with CRC.