Units

PubMed

---

Citation

Type

Journal

Publication Date

Issn Print

Pages

Brief description/objective

BACKGROUND: MR-Colonography (MRC) based on MR-imaging is a relatively new diagnostic modality for diagnosing colorectal polyps. The aim of this experimental study was to evaluate its performance in detecting and staging colorectal cancer. METHOD: 23 patients with proven colorectal cancer underwent MR-Colonography one day prior to operation after standard bowel preparation. The colon was filled with a diluted gadolinium enema. Coronal sections were acquired in prone and supine. Virtual colonoscopy (VC) was processed from the acquired 3D data sets and MRC was interactively analysed together with VC. The findings were correlated with colonoscopic and pathology results. RESULTS: A complete MRC was achieved in 21 patients (92%), two patients (8%) could not be conclusively evaluated due to insufficient bowel preparation or technical problems. No complications were observed. Compared to colonoscopy all 23 carcinomas were detected. No lesion > or = 8 mm was missed. In one patient a synchronous carcinoma was newly diagnosed. This lesion was missed by colonoscopy since the distal tumour was endoscopically unpassable. In total eight (33%) colonoscopies were incomplete. CONCLUSION: MRC offers a new and promising diagnostic tool for colorectal cancer. It is particularly valuable when colonoscopy is incomplete. It improves preoperative planning and it holds the potential as an all-in-one investigation including local and liver staging in combination with conventional MRI.