Purpose: The choice of the therapeutic strategies in patients affected with rectal cancer is strictly dependent by the tumor stage. So, in order to obtain an improvement in preoperative staging accuracy, new imaging modalities are now under investigation. The aim of this work is the evaluation of endorectal-coil MRI in the local staging of rectal cancer.
Material and methods: Fourty-three patients affected with histologically proven rectal cancer, have been evaluated by an high-field strength magnet (1.5 T). In 14/43 patients neoadjuvant pre-operative chemotherapy had been previously performed. In all cases axial SE T1w and FSE T2w sequences and coronal or sagittal FSE T2w sequences, with and without fat suppression, were performed. Basing upon the TNM staging system and the previously reported MRI signs the local extent of the tumor was evaluated, focusing about the rectal wall infiltration and the perirectal lymph nodes involvement. All the patients underwent surgery and a comparative evaluation of MRI and pathological staging was done.
Results: At MRI the tumor was detected in 38/43 patients. In evaluating wall infiltration the MRI results agreed with pathological results in 89% of patients and showed 92% accuracy in T1-T2 stage and 94% in T3. In evaluating perirectal lymph nodes metastases MRI showed 69% accuracy, 82% sensitivity and 55%specificity.
Discussion and conclusions: The poor accuracy of CT and body-coil MRI in evaluating wall involvement in patients with rectal cancer is mainly related to their inability to demonstrate the single layers of the rectal wall. So transrectal ultrasound is now the first choice modalitiy in local staging of rectal cancer. However transrectal ultrasound showed low sensitivity in detecting perirectal lymph nodes metastases and low accuracy in evaluating the patients previously undergone to neoadjuvant chemotherapy or radiotherapy. On the other hand the improvement of MRI sequences and the availability of the endorectal coils allowed to visualize the single layers of the rectal wall so making the endorectal-coil MRI a reliable imaging technique to stage rectal cancer. The results of our work demonstrate a good diagnostic accuracy of endorectal-coil MRI in local staging of rectal cancer, in particular the degree of rectal wall infiltration was well demonstrated, while the perirectal lymph nodes metastases were demonstrated with less accuracy. The long examination time, the costs and the movement-related artefacts are the main limits of MRI. In particular the movement-related artifacts sometime do not allow the visualization of the wall layers so lowering the diagnostic accuracy in demonstrating the tumor wall infiltration. In conclusion, even though endorectal coil MRI proved to be a reliable imaging technique in local staging of rectal cancer, at present we are not able to state what may be its real role in diagnostic evaluation of the patients with rectal cancer, in particular if compared to endorectal ultrasound. Further, comparative studies, based upon larger patients series are probably needed to draw a definitive conclusion.