Servet Tatli MD, Presenter: Nothing to Disclose

Koenraad Julien Mortele MD, Abstract Co-Author: Nothing to Disclose

Elizabeth Lisa Breen, Abstract Co-Author: Nothing to Disclose

Ronald Bleday, Abstract Co-Author: Nothing to Disclose

Stuart George Silverman MD, Abstract Co-Author: Nothing to Disclose

To assess the accuracy of MRI using combined pelvic phased-array and endorectal coil for preoperative staging of rectal cancer.

Between August 2001 and December 2003, 91 consecutive patients with rectal adenocarcinoma underwent preoperative MR imaging using both a pelvic phased-array and an endorectal coil with axial/coronal T2-weighted fast spin echo, axial T1-weighted spin echo, and pre-and-post gadolinium axial spoiled gradient refocused acquisition in steady state sequences. Fifty-one patients (26 men, 25 women) had surgical tumor resection and served as the study population. An abdominal radiologist experienced in the interpretation of anorectal MRI staged tumors retrospectively using TNM classification. Original preoperative MRI staging, performed one of seven abdominal radiologists, was also recorded from the radiology reports. Both MRI staging were compared to postoperative pathologic staging. In addition, patients were stratified according to surgical treatment groups using following criteria: Stage I= T1-2/N0, Stage II=T3/N0, and Stage III=T1-4/N1-3. Kappa statistic was calculated to determine the level of interobserver agreement in the identification of T3 tumor and nodal metastases between two readings.

At pathology, 15 (32%) of 51 tumors were classified as T3 and 34 (68%) as T0-T2. Overall, sensitivity and specificity of MRI for T3 staging was 93% and 86%, respectively (PPV: 74%, NPV: 97%, accuracy: 88%). MRI correctly predicted lymph node metastases in 11 out of 13 patients with a sensitivity of 85% and, specificity of 69% (PPV: 58%, NPV: 90%, accuracy: 74%). MRI correctly identified surgical treatment groups in 33 out of 39 (85%) patients. Five patients were overstaged and one patient was understaged. Interobserver agreement between the retrospective and prospective readings was excellent in prediction of T3 tumor (kappa=0.85) and good (kappa=0.80) in prediction of nodal metastasis.

Pelvic phased-array and endorectal coil MRI is an accurate technique with low interobserver variability in local staging of rectal carcinoma. Combination of the coils especially increases the ability of MRI to predict the degree of mesorectal tumor invasion.

Tatli, S, Mortele, K, Breen, E, Bleday, R, Silverman, S, Rectal Carcinoma Staging with Pelvic Phased-Array and Endorectal Coil Magnetic Resonance Imaging.  Radiological Society of North America 2004 Scientific Assembly and Annual Meeting, November 28 - December 3, 2004 ,Chicago IL. http://archive.rsna.org/2004/4408633.html