Performance of CT Colonography in Diagnosis of Synchronous Colonic Lesions in Patients With Occlusive Colorectal Cancer

OBJECTIVE. The purpose of this study was to evaluate the accuracy of CT colonography (CTC) in the diagnosis of synchronous colonic lesions in a cohort of patients with an occlusive colorectal cancer (CRC) causing incomplete colonoscopy. SUBJECTS AND METHODS. Among 109 patients with CRC causing incomplete colonoscopy who underwent CTC with IV contrast enhancement after cathartic purgation, fecal tagging, and colon distention, 70 (mean age, 70 years) for whom reference standards (surgical reports, first surveillance colonoscopy) were available were evaluated. Per-patient and per-lesion sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV) of CTC in the diagnosis of synchronous colonic lesions measuring 6 mm or larger were assessed. RESULTS. Twenty-seven of the 70 patients (39%) had at least one 6-mm or larger synchronous lesion, and four patients (6%) had a total of five synchronous CRCs. Per-patient sensitivity in diagnosing synchronous CRC was 1.00 (4/4). There were 59 lesions: 20 with a diameter of 10 mm or greater; 30, 6-9 mm; and nine, 5 mm or less. The overall per-patient CTC sensitivity in detecting synchronous lesions 6 mm or larger was 0.93 (25/27); specificity, 0.98 (42/43); PPV, 0.96; and NPV, 0.95. Per-patient sensitivity for the diagnosis of synchronous advanced neoplasia (advanced adenoma and colorectal cancers) was 0.94 (15/16). Per-lesion CTC sensitivity for detecting synchronous lesions 6 mm or larger was 0.88 (37/42); all adenomatous lesions, 0.89 (55/62); and advanced neoplasia, 0.92 (22/24). CONCLUSION. CTC is a highly accurate test for detecting synchronous colonic lesions in patients with occlusive CRC. The prevalence of advanced neoplasia is high (23%).