Accuracy of Preoperative Staging of Rectal Cancer: Comparative Study of Transrectal Ultrasonography and Computerized Tomography.
- Author:
Seung Hui CHEON
1
;
Suk Hwan LEE
;
Kwang Ho KIM
;
Eung Bum PARK
Author Information
1. Department of Surgery, Ewha Womans University College of Medicine, Seoul, Korea. leeshdr@mm.ewha.ac.kr
- Publication Type:Comparative Study ; Original Article
- Keywords:
Rectal neoplasms;
Transrectal ultrasonography;
Computerized tomography;
Staging
- MeSH:
Diagnosis;
Diagnostic Tests, Routine;
Humans;
Lymph Nodes;
Neoplasm Metastasis;
Prospective Studies;
Rectal Neoplasms*;
Sensitivity and Specificity;
Ultrasonography*
- From:Journal of the Korean Society of Coloproctology
2003;19(5):327-333
- CountryRepublic of Korea
- Language:Korean
-
Abstract:
PURPOSE: Preoperative assessment of the depth of invasion in the rectal wall and of lymph node metastases is very important in determining the treatment modality in rectal cancer. The purposes of study were to evaluate the accuracy of transrectal ultrasonography (TRUS) in preoperative staging of rectal cancer and to compare that accuracy with the accuracy for computed tomography (CT). METHODS: We reviewed 59 patients who were diagnosed as having rectal cancer and who had been staged by using TRUS and CT preoperatively. Ultrasonographic tumor (uT) and nodal (uN) stage and computerized tomographic tumor (cT) and nodal (cN) stage were entered into the database prospectively. The accuracy of each staging was compared with the pathologic staging. The accuracy, the sensitivity, the specificity, the positive predictive value, and the negative predictive value of each diagnostic test were calculated. Chi- square tests were conducted to identify the factors influencing the accuracy. RESULTS: The accuracies of TRUS and CT in assessing the depth of invasion were 66.1% and 62.5%, respectively. The accuracies of TRUS and CT in assessing the nodal involvement in patients treated with radical surgery were 70.4% and 63.6%, respectively. For detection of fat infiltration, the sensitivities were 97.4% for TRUS and 76.3% for CT. The specificities were 45.0% for TRUS and 55.6% for CT. The sensitivities for detection of lymph node involvement were 59.3% for TRUS and 42.9% for CT. The specificities were 81.5% for TRUS and 85.2% for CT. The gross appearance of the tumor had a significant influence on the assessment of the depth of invasion (P=0.015). In 9 out of 77 patients (11.7%) could not be performed the TRUS examination due to obstruction or the location of the tumor. CONCLUSIONS: In spite of some limitations, TRUS is considered a very useful tool in the preoperative assessment of the depth of invasion and of the lymph node involvement in rectal cancer. However, CT examination is mandatory to overcome the limitations of TRUS in the preoperative diagnosis of rectal cancers.
