Efficacy of high resolution magnetic resonance imaging in preoperative local staging of rectal cancer

OBJECTIVE: To assess the efficacy of high-resolution magnetic resonance imaging (HRMRI) for preoperative local staging in patients with rectal cancer who did not receive preoperative radiochemotherapy.
METHODS: In this retrospective study, 30 patients with biopsy proved primary rectal cancer were evaluated by HRMRI. Two observers independently scored the tumour and lymph node stages, and circumferential resection margin (CRM) involvement. The sensitivity, specificity, the negative predictive value and the positive predictive value of HRMRI findings were calculated within the 95% confidence interval. The area under the curve was measured for each result. Agreement between two observers was assessed by means of the Kappa test.
RESULTS: : In T staging the accuracy rate of HRMRI was 47-67%, overstaging was 10-21%, and understaging was 13-43%. In the prediction of extramural invasion with HRMRI, the sensitivity was 79-89%, the specificity was 72-100%, the PPV was 85-100%, the NPV was 73-86%, and the area under the curve was 0.81-0.89. In the prediction of lymph node metastasis, the sensitivity was 58-58%, the specificity was 50-55%, the PPV was 43-46%, and the NPV was 64-66%. The area under the curve was 0.54-0.57. When the cut off value was selected as 1 mm, the sensitivity of HRMRI was 38-42%, the specificity was 73-82%, the PPV was 33-42%, and NPV was 79-81% in the prediction of the CRM involvement. The correlation between the two observers was moderate for tumour staging, substantial for lymph node staging and predicting of CRM involvement.
CONCLUSION: Preoperative HRMRI provides good predictive data for extramural invasion but poor prediction of lymph node status and CRM involvement.