Accuracy of magnetic resonance imaging for preoperative staging of rectal cancer.

Abstract

New Zealand has one of the highest rates of rectal adenocarcinoma in the world. Magnetic resonance imaging (MRI) is widely used for preoperative staging of rectal cancer. The accuracy of MRI varies, which may affect treatment decisions. The accuracy of MRI for pretreatment staging of rectal adenocarcinoma in a provincial centre in New Zealand has not been investigated. We aimed to assess the accuracy of MRI for pretreatment staging of early rectal adenocarcinoma in patients managed via the MidCentral Regional Cancer Service Multidisciplinary Team. A retrospective review of the MidCentral Regional Cancer Service Multidisciplinary Team database identified 54 patients with rectal adenocarcinoma who proceeded to surgery without preoperative long course chemo-radiotherapy. The pretreatment MRI stage was compared with the histological stage for each of these patients. MRI correctly staged the tumour invasion (T stage) in 24 patients (44% of cases), and lymph node stage in 38 patients (70% of cases). There was moderate agreement between MRI and histological staging for tumour invasion (κ=0.46) and for lymph node involvement (κ=0.41). Twenty-one cases were under-staged and five cases were over-staged with regards to invasion of the muscularis propria. Fourteen cases were under-staged, and two cases over-staged in regards to lymph node involvement. Although MRI provides important pretreatment staging information for rectal adenocarcinoma, in our experience MRI is not as accurate as in other reports. Multidisciplinary teams managing patients with rectal adenocarcinoma should be aware of the limitations of MRI for pretreatment staging.