Preoperative Staging of Rectal Cancer by CT & MRI.
- Author:
Jong Keun SHIN
1
;
Sang Mock LEE
;
Ho Chul PARK
;
Kee Hyung LEE
;
Soo Myung OH
;
Choong YOON
;
Dong Ho LEE
Author Information
1. Department of Surgery, Kyung Hee University Hospital, Korea.
- Publication Type:Original Article
- Keywords:
Rectal cancer;
Preoperative staging;
CT;
MRI;
TNM classification
- MeSH:
Classification;
Humans;
Magnetic Resonance Imaging*;
Rectal Neoplasms*;
Ultrasonography
- From:Journal of the Korean Society of Coloproctology
2000;16(3):177-185
- CountryRepublic of Korea
- Language:Korean
-
Abstract:
Accurate staging of rectal cancer preoperatively is important to plan a proper treatment and to predict treatment results. For the preoperative staging of rectal cancer, computed tomography (CT), transrectal ultrasonography (TRUS), and magnetic resonance imaging (MRI) have been used, but the role of them remains controversial. This research was intended to compare and analyze the accuracy of CT and MRI in the preoperative staging of rectal cancer. METHODS: From January 1998 to June 1999, sixty patients were studied by CT and MRI before their operations for rectal cancer in our institution, but two patients with local excision were excluded in N-staging as objects. The patients who had preoperative irradiation were also excluded in this study. Preoperative staging with CT and MRI were conducted by one radiologist according to 1997's TNM classification based on AJCC. On the results of pathological findings after operation, preoperative staging with CT and MRI were classified into T-staging and N-staging. Accuracy and agreement rate between pathological staging and preoperative staging by CT and MRI were compared and analyzed by Kappa value. RESULTS: The accuracy of CT was 68 percent in T-staging, and 58 percent in N-staging, MRI showed accuracy of 82 percent in T-staging and 64 percent in N-staging. In the T-staging, the agreement rate between pathological staging and CT staging was 0.54 (95% confidence interval), while the agreement rate was 0.70 in MRI staging, resulting in a higher agreement rate with MRI than with CT. In the N-staging, the agreement rate between pathological staging and CT staging was 0.38, with a relatively lower agreement rate, while the agreement rate was 0.56 in MRI staging. In our study, MRI showed a higher agreement rate than CT. CONCLUSIONS: In the future, more research should be conducted, but it can be conclued that in preoperative staging for rectal cancer, MRI using body arrayed coil has a better accuracy than CT. Subsequently MRI staging should be considered as a more useful investigation method before operation than CT.
