ABSTRACT

The goal of this retrospective study was to evaluate the correlation between pathologically verified tumor stages and clinical stages predicted by Magnetic Resonance Imaging (MRI) before and after combined chemoradiotherapy retrospectively, whose biopsy confirmed as rectal cancer and preoperative MR staging of locally advanced rectal cancer was detected.

14 patients with histologically proven primary rectal cancer were evaluated with 1.5 T MRI before and after chemoradiotherapy. Rectum MRI were evaluated by a specialist radiologist and a radiology resident. Two observers independently scored the tumor stage, lymph node stage and measured the distance to the mesorectal facia named as environmental resection border (ERB). This findings were compared with postoperative histopathologic results using Friedman, Wilcoxon, Cochran tests.

In T staging the accuracy rate of MRI was 57.1%, overstaging was 42.8% for both observers. Both observers did not downstage. In N staging the accuracy rate of MRI was 64.2%, overstaging was 21.4%, downstaging was 14.2% for both observer. In ERB evaluation the accuracy of MRI was 85.7% for the first observer, 92.8% for the second observer. There was not any statistically significant difference between observers and with histopathologic results regarding T, N staging and ERB evaluation. When the MRI imaging of patients who had MRI pre-CRT, compared with histopathological findings, 58.3% regression was detected for T staging for both observer and regressions of 75% for the first, 66.6% for the second observer were detected for N staging.

In patients who received preoperative CRT due to desmoplastic rection in T-staging, reactive lymph nodes in N-staging leads to false positive results and thus the accuracy rate is reduced. Better results were obtained in ERB evaluation.