Effect of endorectal coils on staging of rectal cancers by magnetic resonance imaging.

OBJECTIVE To compare the use of endorectal plus phased-array coils with use of phased-array coil alone with respect to the accuracy of magnetic resonance imaging for detecting mesorectal involvement of rectal cancer. DESIGN Retrospective study. SETTING A tertiary referral centre in Hong Kong. PATIENTS Ethnic Chinese patients with rectal adenocarcinoma who underwent staging magnetic resonance imaging during the years 2003 to 2008 in our centre were selected; those who received preoperative neoadjuvant therapy were excluded. Unless otherwise contra-indicated, endorectal coils have been used since 2006. MAIN OUTCOME MEASURES Magnetic resonance images were retrieved and reviewed by two radiologists blinded to the pathological results. The radiological findings were then correlated with the pathological reports to determine diagnostic accuracy. RESULTS A total of 50 patients were studied; 13 of the examinations were in patients having an endorectal coil. The overall accuracy of magnetic resonance imaging in detecting mesorectal tumour involvement was 80%. Subgroup analysis showed higher accuracy in the group with endorectal coils than in those with phased-array coils alone. Over-detection of mesorectal involvement was noted in 12% of the cases, with lower rate being observed in patients with endorectal coils. Underdetection of mesorectal tumour involvement was only noted in the group without endorectal coils. With the use of endorectal coils, the sensitivity reached 100% and the specificity increased to 86%. CONCLUSION Use of endorectal coil in staging magnetic resonance imaging of the rectum improves diagnostic accuracy. Whenever feasible, endorectal coil use is therefore recommendable to enhance diagnostic accuracy. The study results substantiate the understanding of staging by magnetic resonance imaging of rectal cancer in the local Chinese population.