Sectional Imaging Anatomy Of The Rectum And Preoperative MRI Study Of Local Invasion Of Rectal Cancer

Objective: The purpose of our study was by comparing the sectional anatomy of the surrounding structures of the rectum with MRI, which is closely related to the total mesorectal excision surgery(TME),to enhance the normal imaging knowledge of the surrounding structures of the rectum and evaluate the local invasion of rectal caner and establish criteria for the visualization of the structures of the rectum; by studying the extent of local infiltration of rectal cancer with preoperative MRI, to provide reliable imaging evidence for surgical treatment. Materials and methods: 20 fresh resected specimens of rectal cancer were dissectioned along the posterior aspect of the mesorectum fascia to identify the gross appearance of the related structures of the rectum. By using the thin sectional anatomy database from the first Chinese digitized visible human of the third medical university and the cadaveric pelvis sectional anatomy, we compared the MR imaging of 60 randomly selected patients(20 men, 40 women, ages rang from 18 to 59, with a mean age of 46.5)to study the imaging of the normal rectal structures. 36 consecutive patients (13 women, 23 men, ages rang from 32 to 78, with a mean age of 42.7) with rectal carcinoma proved by biopsy underwent MR imaging with a 1.0T whole body system, and body coil, each patient underwent bowel preparation and filled the rectal lumen with air or 0.9% Nacl. Each specimen was performed MR with 3 mm slices. All MR imagings ware studied to assess the extent of local rectal cancer infiltration, the depth of extramural spread, the lymph node involvement and the relation of tumor with the mesorectal fascia and finally to predict the potential positive circumferential resection margin. Results: The mesorectal fascia was a continuous smooth structure surrounding the rectum and mrsorectum as a complete "sock", whose main component was collagen and formed by the pelvic visceral fascia. In the structures of mesorectum were fat, blood vessels, nerves, and lymphatics. In thin section anatomy, the rectum and its surrounding structures well displayed it could display the structures of the rectum, such as mucosa, submucosa, and muscle. But in MR imaging we could not distinguish such structures for its poor resolution. MR imaging could display most of the related structures of surrounding the rectum. Mesorectal fascia, as a distinct thin layer, encompassed the mesorectum and was well seen on axial MR images. In T1WI or T2WI MR imaging, the mesorectal fascia appears as a low-signal-intensity linear structure surrounding the mesorectum .The mesorectum was shown on axial and sagittal sections as a high-signal-intensity which was similar to the signal of fat surrounding the rectum, containing vessels and lymphatic tissue. MR images could not clearly display the retrorectal space and the pelvic nerve plexuses. The results of pre-operative MRI in diagnosing the invasion degree of the rectal wall was not satisfied, the overall accuracy rate of the T staging was only 66.7%, which was moderate coincidence with the results of histopathology assessment, and was lower than the accuracy of the post-operative MR images. There was reliable consistence between MRI and histopathology in identifying the depth of extramural spread. In diagnosing of the lymph nodes (LN) of the mesorectum, both post-operative and pre-operative MRI had poor results. The overall accuracy of MRI in prediction the nodal status was only 55.6%, because of its limited indifferentiating metastasis from benign reactive lymph nodes. The sensitivity of pre-operative MRI in predictioning the positive nodes was higher by the criteria of the short-axis diameter in 5mm (64.7%)than in 10mm .Its specificity was 47.4% .The distinct or indistinct border of LN showed the accuracy rate of 47.6% and 52.3%, sensitivities of 45.5% and 54.5%, specificities of 50.0% and 50.0%. The sensitivities of heterogeneity of LNs were 63.7% in predicting nodal status, but the sensitivities were 40.0%. The mean extramural invasion depths of measured by postoperative and preoperative MRI were 11.4 mm and 10.7 mm. In prediction of the potential circumferential resection margins, both postoperative and preoperative MRI had a good consist rate with the histopathology.13 patients had positive circumferential resection margins proved by histological pathology, while postoperative and preoperative MRI identified 12 and 11 patients, respectively. Conclusion: (1)The rectum and its surrounding adjacent structures relevant to the total mesorectal excision of rectal cancer could be well displayed by MRI. The accuracy of depiction of the mesorectum and mesorectal fascia and levator ani muscle would helpimaging doctor recognize the normal imaging of the surrounding structures of the rectum and it would assist in assessment of the invasion degree and scope of rectal cancer in the mesorectum and adjacent structures, which would provide the imaging evidence for the surgeon to develop the total mesorectal excision and pelvic autonomic nerve preservation. (2)The MR technique is more important for displaying the relation of rectal cancer with the rectum and its surrounding structures.(3) There is a direct agreement in assessment of the local invasion and the depth of extramural spread of rectal cancer between MRI and histopathology. In prediction the nodal statues of the mesorectum, MRI is not satisfied for its poor accuracy. The short-axis diameter of LNs, the border of LNs with distinct or indistinct and the heterogeneity or homogeneity of LNs was not a specific criterion in prediction the LN metastasis or reactive benign lymph nodes. The short-axis diameter of LN in 5 mm, which can serve as the MR imaging criteria in prediction the positive nodes, because of its high sensitivity.(4)MRI can assess the relation of the rectal cancer and the mesorectal fascia with high accuracy. MRI also can accurate predict the distance between the rectal cancer and the potential circumferential resection margins. This can assist in defining the type of surgical procedure pattern and the plane of surgical excision; it also could provide valuable information for preoperative radiotherapy and serves as a promising method to determine different treatment policy for rectal cancer patients pre-operatively. The accurate assessment of the status of circumferential resection margins of rectal cancer, which can concur the shortcoming of the imaging staging of rectal caner. This would be helpful for staging the rectal cancer based on the distance between the tumor and the mesorectal fascia, the relation of the tumor with the circumferential resection margins, and on the prognostic factors.