RESUMEN
OBJECTIVE: Abdominal ultrasonography after transrectal filling with contrast agent (AU-TFCA) was retrospectively evaluated with respect to determination of T stage and lesion length in patients with colorectal cancer (CRC) who had previously failed colonoscopy because of severe intestinal stenosis. METHODS: The population comprised 83 patients with CRC with intestinal stenosis and previously failed colonoscopy who underwent AU-TFCA, and in addition contrast-enhanced computed tomography (CECT) and/or magnetic resonance imaging (MRI), 2 wk before surgery. The diagnostic performance of AU-TFCA and CECT/MRI was evaluated relative to the post-operative pathological results (PPRs) by paired sample t-test, receiver operator characteristic (ROC) curve, Pearson's χ2-test and κ and intraclass correlation coefficients. RESULTS: The T staging identified via AU-TFCA, but not CECT/MRI, was relatively consistent with that of the PPRs (linearly weighted κ coefficient: 0.558, p < 0.001, and linearly weighted κ coefficient: 0.237, p < 0.001, respectively). The overall diagnostic accuracy of T staging based on AU-TFCA (83.1%) was significantly higher than that based on CECT/MRI (50.6%). Regarding lesion length, the results of AU-TFCA and PPRs were comparable (t = 1.852, p = 0.068), but those of CECT/MRI and PPRs were significantly different (t = 8.450, p < 0.001). CONCLUSION: AU-TFCA is effective in evaluation of lesion length and T stage in patients with severely stenotic CRC lesions who previously failed colonoscopy. The diagnostic accuracy of AU-TFCA is significantly better compared with that of CECT/MRI.
