Clinicopathologic Analysis of Gastrointestinal Stromal Tumors of the Colon and Rectum

PURPOSE: This study's aim is to investigate the clinicopathologic characteristics of colorectal gastrointestinal stromal tumors (GISTs) and to evaluate the result of those tumors. METHODS: We retrospectively reviewed 22 patients who had been diagnosed with primary colorectal GISTs and who had undergone a surgical resection between October 1996 and July 2008. RESULTS: Colorectal GISTs accounted for 0.28% of all colorectal malignancies and 7.7% of all GISTs. Rectal GISTs (19, 86.4%) were more common than colonic GISTs (3, 13.6%). According to the National Institute of Health's (NIH) grading system, there were 1 (4.5%) very low, 5 (22.7%) low, 4 (18.2%) intermediate, and 12 (54.6%) high-risk tumors. The disease recurred in 7 patients (1 with intermediate risk and 6 with high risk). Recurrence sites were the liver (42.9%), the peritoneum (71.5%), and the lymph nodes (14.3%). Adjuvant imatinib therapy and/or radiation therapy were done for patients with microscopically positive margins of resection and high risk, of which one experienced a recurrence at 95 months after surgery. The five-year recurrence rates were 0% in the very-low-grade and low-grade groups, 33.3% in the intermediate-grade group, and 37.5% in the high-grade group. The five-year overall survival rates were 100% in the very-low-grade and low-grade groups, 66.7% in the intermediate-grade group, and 62.5% in the high-grade group. CONCLUSION: Poor prognosis of colorectal GISTs was closely related to the tumor's histologic grade and size. Integrating surgery, molecular therapy, and radiation therapy might improve outcomes, but further study with more cases is needed.

Risk Factors for Anastomotic Leakage after a Low Anterior Resection for Rectal Cancer

PURPOSE: Clinical anastomotic leakage remains a major problem after a low anterior resection for rectal cancer, so indentifing risk factors influencing anastomotic leakage is important. The aim of this study was to assess the association between risk factors and anastomotic leakage. RESULTS: One thousand two hundred eight patients underwent a primary resection for rectal cancer from June 1993 to March 2007. We used hospital records and the colorectal cancer registry to analyze retrospectively the case histories of those patients. The operations were performed using a low anterior resection with the double stapling method. All patients underwent a tumor-specific mesorectal excision. Of the total, thirty-eight patients showed anastomotic leakage. Univariate and multivariate analyses were performed to assess the risk factors affecting to the anastomotic leakage. RESULTS: The rate of anastomotic leakage was 3.2% (38 of 1,208 patients) with a mortality rate of 7.9% (3 of 38 patients). The overall mortality rate was 0.3% (3 of 1,208 patients). Males accounted for 28 of the 38 patients with leakage, and female accounted for the the account for the remnant 10. The mean age was 53.7 years (33~74 years). The mean leakage day was 11.8th day (3~37th day) after the operation, and the mean hospital day was 39.2 days (7~131 days). The mean body mass index (BMI) was 22.7 kg/m2 (15.7~30.8 kg/m2). The mean operation time was 230.5 minutes (90~425 minutes), and the mean bleeding loss was 519.5 cc (0~3,500 cc). CONCLUSIONS: Significant risk factors for anastomotic leakage after primary resection for rectal cancer were the transfusion amount during surgery, a preliminary colostomy, and nodal stage.

Accuracy of Magnetic Resonance Imaging in Predicting TNM Staging and Circumferential Resection Margin Compared with Pathologic Assessment on Whole-mount Section in Rectal Cancer

PURPOSE: Recently, magnetic resonance imaging (MRI) has become the preferred diagnostic tool for preoperative assessment of TNM staging and circumferential resection margin (CRM) in patients with rectal cancer. The aim of this study is to evaluate the accuracy of preoperative MR imaging in the prediction of T, N stage and CRM compared with pathologic results on whole- mount sections. METHODS: Thirty-five consecutive patients with rectal cancer were enrolled between Dec. 2005 and Apr. 2006. 1.5-T MR imaging, was performed, and pathologic results were investigated on whole-mount sections. The agreement between MR imaging and pathologic examination for the assessment of T, N stage and status of CRM were analyzed using kappa statistics. RESULTS: The accuracy of MR imaging compared with pathologic assessment of T stage was 82.9% (kappa=0.56), and that of N stage was 74.3% (kappa= 0.31). Of the MR imaging planes, the oblique axial plane showed the most accurate prediction of CRM, regardless of tumor position within the circumference of the rectum. The accuracy of MR imaging in the oblique axial plane for predicting the CRM was 81.0% (kappa=0.62) in anterior and posterior rectal tumors and 71.4% (kappa=0.43) in laterally located rectal tumors. With a different CRM criteria for the measured distance in MR imaging, the accuracy of the 2-mm CRM criterion was 77.1% (kappa=0.53). CONCLUSIONS: MR imaging in predicting T stage showed fair agreement according to kappa statistics. Of the MR imaging planes, the oblique axial plane provided the most accurate CRM information compared with pathologic examination. The actual measured distance of the CRM in MR imaging can be applied to the pathologic CRM.

Radiofrequency Ablation of Hepatic Metastasis from Colorectal Cancer; Early Experience

PURPOSE: The aim of this study is to describe the safety and efficacy of radiofrequency ablation (RFA) in the treatment of unresectable liver metastasis from colorectal cancer. METHODS: From January 1999 to December 2000, we undertook 48 RFA procedures in 36 patients with unresectable metastatic liver tumor from colorectal cancer. RFA was performed either via celiotomy (n=23) or using a percutaneous approach (n=25) under ultrasound guidance using a LeVeen needle electrode and a RF 2000 generator. Patients were followed with spiral computed tomographic scans at 1 week after RFA. RESULTS: RFA was performed in 15 patients with synchronous hepatic metastasis and 21 with metachronous hepatic metastasis. The mean tumor size was 3.1 cm (range; 0.5~7.2 cm). In 2 patients, complete ablation failed due to anatomic tumor location. Nine patients (25%) exhibited post RFA complications (9 cases of fever, 8 of abdominal pain, and 1 of hematoma) which showed spontaneous resolution and there was no treatment-related death. At a mean follow up of 10.1 months, 12 patients (33.3%) had recurred in the liver and 3 (8.6%) at the RFA site, while 16 (44.4%) remained clinically free of disease. CONCLUSION: Although RFA is a relatively safe procedure in patients with unresectable hepatic metastasis from colorectal cancer, the possible development of new metastatic disease after RFA is currently a limitation in this form of treatment. Further study on the efficacy of RFA versus other treatment modalities is needed.