Differentiation of Benign and Malignant Ulcers of the Stomach on Computed Tomography

PURPOSE: We wanted to determine the multidetector computed tomography (MDCT) findings for differentiating benign ulcers from malignant ones. MATERIALS AND METHODS: 18 clinicopathologically proven benign ulcers that had been detected by both endoscopy and MDCT were the focus of this study. 26 ulcerative advanced gastric cancers and 26 early gastric cancers with ulceration, all of which had been surgically proven, were selected as a control group. Five of the 26 early gastric cancers that were confined to the mucosa and that were not detected on CT were excluded in this study. The following CT findings were reviewed by two radiologists; ulcer size, the degree of enhancement and the thickness of inner enhancing layer in the ulcer base, the total thickness and the enhancing inner layer thickness in the largest part of the thickened ulcer mound, the presence of ulcer that projected beyond the healthy lumen, and the presence of perigastric fat infiltration and perigastric lymphadenopathy. RESULTS: An indiscernible thin-walled ulcer base (less than 1.5 mm) and suboptimal enhancement of the ulcer base for the discrimination of benign gastric ulcers from the malignant gastric ulcers showed sensitivities of 100% (18/18) and 78% (14/18), respectively, with specificities of 98% (46/47) and 92% (43/47), respectively. Ulcer projection was more significantly present in benign ulcer (13/18, 72%) than in the malignant gastric ulcers (7/47, 15%). The enhancing inner layer thickness in the ulcer mound was significantly greater in the AGC (mean: 7.4 mm) than in the benign gastric ulcers (mean, 2.2 mm). There were insignificant differences for ulcer size, total thickness of the ulcer mound, the perigastric fat infiltration and perigastric lymphadenopathy between the benign and malignant gastric ulcers. CONCLUSION: MDCT is an additional helpful diagnostic tool when benign gastric ulcers are histologically difficult to distinguish from malignant gastric ones.

Recurrent Pediatric Mesenteroaxial Gastric Volvulus: Case Report Focusing on Ultrasonographic and CT Findings

Gastric volvulus is a rare condition, and it is classified as the organoaxial or mesentericaxial type according to the axis of rotation. We experienced 1 case of pediatric recurrent mesenteroaxial gastric volvulus and we report here the ultrasonographic and CT findings.

Gastric Inflammatory Myofibroblastic Tumor: Case Report

We report a case of gastric inflammatory myofibroblastic tumor in a 25-month-old boy. The condition is very rare and is of unknown pathogenesis. Because it is large and invades adjacent organs, imaging procedure do not provide easy differentiation between an inflammatory myofibroblastic tumor and malignacy, and it should thus be included in the differential diagnosis of an infiltrative upper abdominal mass occurring in children.

Radiologic Findings of Multiple Myeloma with Gastric Involvement: A Case Report

We report a case of multiple myeloma with gastric involvement occurring in a patient who underwent an upper gastrointestinal series (UGIS), CT and MRI. UGIS depicted a luminal protruding mass, while contrast-enhanced CT demonstrated marked thickening of the gastric wall, with subtle contrast enhancement. At T1- and T2-weighted MR imaging, the mass showed iso- and intermediate signal intensity, respectively. After the administration of contrast material, subtle homogeneous enhancement was apparent.

Remote Metastasis to the Gastric Cardia in Esophageal Carcinoma: CT Findings

PURPOSE: To assess the CT findings of remote metastasis to the gastric cardia in patients with esophageal carcinoma. MATERIALS AND METHODS: Among patients with esophageal carcinomas treated between June 1994 and May 1999, five males aged 65-75 (mean, 67.4) years with histologically proven remote metastasis to the gastric cardia from esophageal squamous cell carcinoma, detected at surgery (n=2) or endoscopic biopsy (n=3), underwent CT scanning. We retrospectively evaluated the findings in terms of the location, size and appearance of each lesion and the presence or absence of associated lymphadenopathy. RESULTS: The primary esophageal carcinomas were located in the middle third (n=4) and lower third (n=1) of the esophagus. All five gastric metastases were solitary and occurred in the gastric cardia, and were separated from the primary tumors. CT showed that the metastases ranged in size from 4.2 to 8.0 (mean, 6.7) cm, and all were larger than the primary tumors. All were ulcerated, and in four cases there was associated abdominal lymphadenopathy. They were all well defined, poorly enhanced, submucosal masses that were endogastric in three cases and exogastric in two. The latter were difficult to differentiate from extrinsic masses compressing the gastric cardia. CONCLUSION: Our results suggest that when a submucosal gastric cardial mass with associated lymphadenopathy is detected by CT during the initial staging or follow-up evaluation of esophageal carcinoma, remote gastric metastasis should be considered.

Gastric Neuroendocrine Carcinoma Presenting as a Wandering Exophytic Mass: A Case Report

We describe a case of poorly differentiated gastric neuroendocrine carcinoma presenting as a wandering exophytic mass. CT imaging revealed a 14.5x10.0x8.0 cm, lobulated, solid mass with a multifocal necrotic portion at the right of the peritoneal cavity. It was attached to the antrum of the stomach by a broad stalk, and shown by MR imaging to be well-defined, lobulated and solid, with a multifocal necrotic portion, and at the left of the peritoneal cavity. Isointensity was apparent at T1-weighted imaging, and slightly hyperintensity at T2-weighted imaging, and after gadolinium injection, enhancement was stronger than at precontrast imaging.

Usefulness of Follow-up Computed Tomography after Surgery for Early Gastric Cancer

PURPOSE: To analyze the recurrent rate, time of recurrence, type of recurrence and the relationship between recurrence and histopathologic findings after radical gastrectomy for early gastric cancer and evaluate the usefulness of follow up abdominal computed tomography after surgery. MATERIALS AND METHODS: We retrospectively evaluated 617 abdominal computed tomographic examinations of 144 patients (101 male, 43 female, mean age, 53 years) who underwent radical subtotal gastrectomy for early gastric cancer between July 1994 and July 1997. Follow-up abdominal CT scans were reviewed by three abdominal radiologists for detection of recurrence of early gastric cancer, and endoscopic and pathologic findings were correlated. We also reviewed the surgical pathologic reports for location, size, cell type and depth of invasion of early gastric cancer and lymph node invasion. We analyzed the recurrent rate, time and type of recurrence, and relationship between recurrence rate and pathologic characteristics of early gastric cancer. RESULTS: The recurrent rate was 4.2% (6/144) during 5-7 years after radical subtotal gastrectomy for early gastric cancer. The recurrence was detected on 2-5 years after operation. The types of recurrence were lymph node metastasis (n=5), liver metastasis (n=4), recurrence in the residual stomach or anastomotic site (n=3), adrenal metastasis (n=1), and lung metastasis (n=1). Relationship between recurrence and location, size, depth of invasion and cell type of early gastric cancer and lymph node metastasis was not significant statistically (p>0.4). CONCLUSION: The recurrence rate of early gastric cancer after radical subtotal gastrectomy is very low and occurs after two years. The follow up-CT scans can detect all recurrence of early gastric cancer, so regular follow-up abdominal CT examination is useful.

Intramuscular Bronchogenic Cyst of Gastric Body: A Case Report

Developmental foregut cysts, whether bronchogenic, esophageal, gastroenteric or pericardial, are frequently encountered in the mediastinum, and are also occasionally found in the upper abdomen, where they can mimic adrenal, pancreatic, renal or gastric masses. We present the computed tomographic (CT) and histologic findings of an intramuscular bronchogenic cyst of the gastric body, mimicking a retroperitoneal cystic mass. CT scanning demonstrated the presence of a relatively hyperattenuating cystic mass without enhancement. Histologic examination revealed a bronchogenic cyst secreting mucoid materials.

Ductal Adenocarcinoma Arising from Heterotopic Pancreas in the Stomach: A Case Report

Malignant transformation of heterotopic pancreas is extremely rare. We report a case of ductal adenocarcinoma arising from heterotopic pancreas in the stomach of a 64-year-old man. Preoperative CT scans showed the lesion as a submucosal mass along the greater curvature of the pyloric antrum and protruding into the pyloric canal. After gastric surgery, the resected tumor was histopathologically diagnosed as a ductal adenocarcinoma arising from heterotopic pancreas with cystic dilatation of aberrant pancreatic duct.

Right Intrathoracic Stomach Secondary to Congenital Hiatal Hernia with Organoaxial Torsion: A Report of Two Cases

Right intrathoracic stomach associated with organoaxial torsion is a rare form of congenital hiatal hernia. We report the radiologic findings in two cases of complete or partial right intrathoracic stomach secondary to congenital hiatal hernia. The barium meal test demonstrated the presence of complete or partial right intrathoracic stomach and non-obstructive organoaxial torsion with the greater curvature lying against the right chest wall. The esophagogastric junction was located above the diaphragm. CT revealed a cystic mass in the right posterior mediastinum. This cystic lesion should be differentiated from other congenital mediastinal cysts.

Sarcomatoid Carcinoma of the Stomach: Report of Three Cases

Sarcomatoid carcinoma is a rare neoplasm of epithelial origin but consists partly of variable differentiated tumor cells of mesenchymal origin. Accurate diagnosis, including differentiation from adenocarcinoma or gastrointestinal stromal tumor, is difficult. We experienced three cases of sarcomatoid carcinoma of the stomach, and describe the radiological and pathologic findings. One case involved a polypoid mass in the antrum, another a mass with a large ulcer mimicking a Bormann type-II adenocarcinoma in the body, while in the third case, an intraluminal bulky mass arising from the cardia of the stomach was present. This was not differentiated from cancer or stromal tumor.

Helical CT Findings of Gastric Wall Thickening by Peptic Ulcer: Compared with Gastric Adenocarcinoma with Ulcer

PURPOSE: To compare on the basis of helical CT findings gastric wall thickening of peptic gastric ulcer with that of gastric adenocarcinoma with ulcer. MATERIALS AND METHODS: Thirty-eight patients with athologically proven gastric lesion [17 cases of peptic ulcer and 21 cases of ulcerative or ulceroinfiltrative gastric cancer (Borrman type II, III)] underwent helical CT, and the findings were retrospectively reviewed in terms of maximum abnormal wall thickness, preservation of the inner enhancing layer, the presence of three discriminate layers of gastric wall, and enhancment pattern. The enhancment pattern of abnormally thick wall was compared with that of the portal phase of back muscle, and was defined as low, iso, or high. The Chi-square test and Student t test were used for statistical analysis. RESULTS: In cases of peptic ulcer and gastric cancer with ulceration, maximum abnormal wall thickness was 7-30 (mean, 16.1)mm, and 11-33 (mean, 21.8)mm, respectively. The inner enhancing layer was preserved in 15 of 17 patients (88.2%) and one of 21 (4.8%); three discriminate layers of gastric wall were observed in 8 of 17 patients (47.0%), and one of 21 (4.8%). The enhancement pattern was low in 12 of 17 patients (70.5%), and 3 of 21 (14.3%); iso in 4 of 17 (23.5%), and 4 of 21 (19.0%), and high in one of 17 (5.9%), and 14 of 21 (66.7 % ). All figures refer, respectively, to the two distinct onditioins. In terms of preservation of the inner enhancing layer, three discriminate layers of gastric wall, and a low enhancement pattern, there were statistically significant differences between peptic ulcer and gastric adenocarcinoma with ulcer. Where the enhancement was high, however, the statistically significant difference between the two conditions was even greater. There was no statistically significant difference in terms of gastric wall thickness or iso-attenuation of thickened gastric. CONCLUSION: Helical CT findings of gastric wall thickening, preservation of the inner enhancing layer, and three discriminate layers of gastric wall, as well as the nature of the observed enhancement pattern, may help differentiate between peptic ulcer and gastric adenocarcinoma with ulcer.

Gastric Wall Thickening on Spiral CT after Subtotal Gastrectomy for Gastric Cancer: Comparision between Recurrent Cancer and Benign Thickening

PURPOSE: To determine the features revealed by two-phase spiral CT scanning useful for differential diagnosis between recurrent cancer and benign wall thickening in patients who have undergone subtotal gastrectomy for stomach cancer. MATERIALS AND METHODS: We retrospectively reviewed 25 cases in which wall thickening of more than 1 cm in the remnant stomach after subtotal gastrectomy was revealed by two-phase spiral CT scanning. All cases were confirmed: 11 were recurrent cancer, and in 14, benign wall thickening was demonstrated. We analyzed the CT findings including maximal thickness of the gastric wall, patterns of wall thickening, degree of contrast enhancement seen during the arterial and portal phases, and the presence of perigastric strands. Maximal wall thickness was classified as either more or less than 15 mm, and as either focal or diffuse. We also determined whether lymphadenopathy was present. RESULTS: Mean maximal gastric wall thickness was 18.4 mm in the recurrent cancer group ("group A") and 12.6 mm in the benign group ("group B") . The gastric wall was thicker than 15 mm in 10 of 11 group A cases and in 3 of 14 in group B; wall thickening was focal (n=3) or diffuse (n=8) in group A, and focal (n=13) or diffuse (n=1) in group B, while the enhancement patterns seen during the arterial and portal phase, respectively, were high/high (n=8), low/high (n=1) and low/low (n=2) in group A, and low/low (n=7), low/high (n=4), high/low (n=1) and high/high (n=2) in group B. Perigastric strands were observed in nine cases in group A, but in none in group B, while lymphadenopathy was combined with wall thickening in seven group A cases but in none of those in group B. CONCLUSION: In patients who have undergone subtotal gastrectomy for gastric cancer, two-phase spiral CT findings including maximal thickness of the gastric wall, patterns of wall thickening, degree of contrast enhancement seen during the arterial and portal phase, the presence of perigastric strands, and lymphadenopathy are useful for differential diagnosis between recurrent cancer and benign wall thickening.

Gastric Stromal Tumor: Two-Phase Dynamic CT Findings with Water as Oral Contrast Agents

PURPOSE: To evaluate two-phase dynamic CT with water as oral contrast agents in the CT diagnosis of gastric stromal tumors. MATERIALS AND METHODS: We retrospectively reviewed the CT findings in 21 patients with pathologically proven gastric stromal tumors. Six were found to be benign, twelve were malignant, and there were three cases of STUMP (stromal tumor uncertain malignant potential). Two-phase dynamic CT scans with water as oral contrast agents were obtained 60-70secs (portal phase) and 3 mins (equilibrium phase) after the start of IV contrast administration. We determined the size, growth pattern, and enhancement pattern of the tumors and overlying mucosa, the presence or absence of ulceration and necrosis, tumor extent, and lymph node and distant metastasis. The CT and pathologic findings were correlated. RESULTS: All six benign tumors and three STUMP were less than 5.5 cm in size, and during the portal phase showed round endogastric masses with highly enhanced, intact overlying mucosa. Twelve malignant tumors were 4.5-15.5 cm in size (mean, 11.5cm); an endogastric mass was seen in three cases, an exogastric mass in one, and a mixed pattern in eight. On portal phase images the tumors were not significantly enhanced, but highly enhanced feeding vessels were noted in five larger tumors (> 10 cm). All 12 malignant tumors showed ulceration and necrosis, and interruption of overlying mucosa was clearly seen during the portal phase. We were readily able to evaluate tumor extent during this phase, and in ten malignant tumors there was no invasion of adjacent organs. Seven malignant tumors showed air density within their necrotic portion (p<0.05). On equilibrium phase images, all malignant tumors showed heterogeneous enhancement due to necrosis, and poorly enhanced overlying mucosa. CONCLUSION: Dynamic CT during the portal phase with water as oral contrast agents was useful for depicting the submucosal origin of gastric stromal tumors and for evaluating the extent of malignant stromal tumors. Our results suggest that these CT findings may be helpful for differentiating between benign and malignant stromal tumors, and in distinguishing them from other gastric tumors.

Imaging Findings of Intragastric Gallstone and Bouveret's Syndrome

Gallstone ileus is a well-known complication of cholelithiasis, but is relatively rare. Most ectopic gallstones are located in the small bowel; they are rarely found in the stomach and duodenum. We describe the imaging findings of a case of intragastric gallstone, as well as a case in which duodenal obstruction was caused by a large gallstone (Bouveret's syndrome).

Cystic Lesions in the Stomach

Cystic lesions of the stomach are rare and usually detected incidentally during surgery or autopsy. Amongseven cases of cystic masses, duplication cysts accounted for four, retension cysts of ectopic pancreas for two,and cystic lymphangioma remaining one. In the upper gastrointestinal series, all were submucosally, whileendoscopic ultrasonography showed that the location of cystic masses was also submucosal. Except for two cases ofduplication cyst and cystic lymphangioma which were thin-walled, lesions were well-defined and showed lowattenuation. In addition, abdominal CT scanning showed two cases of retension cyst of ectopic pancreas.

Radiologic Finding of Gastric Antral Vascular Ectasia: A Case Report

Gastric antral vascular ectasia ("watermelon stomach") is characterized by a prominent longitudinalerythematous fold of gastric antrum. Because it has usually been diagnosed by gastroscopy and biopsy, itsradiologic findings have not been well described. We report a case of gastric antral vascular ectasia, anddescribe its findings, as seen on UGIS, ultrasonography, and computed tomography.

Borrmann Type IV Adenocarcinoma versus Gastric Lymphoma: Spiral CT Evaluation

PURPOSE: To distinguish the spiral CT findings of Borrmann type IV adenocarcinoma from those of gastric lymphoma with diffuse gastric wall thickening. MATERIALS AND METHODS: We retrospectively reviewed the spiral CT scans of 30 patients with Borrmann type IV adenocarcinoma and nine with gastric lymphoma with diffuse gastric wall thickening. In all patients the respective condition was pathologically confirmed by gastrectomy. CT scanning was performed after peroral administration of 500-700ml of water. A total of 120-140ml bolus of nonionic contrast material was administered intravenously at a flow rate of 3ml/sec and two-phase images were obtained at 35-45 sec(early phase) and 180 sec(delayed phase) after the start of bolus injection. Spiral CT was performed with 10mm collimation, 10mm/sec table feed and 10mm reconstruction. We evaluated the degree and homogeneity of enhancement of thickened entire gastric wall, and the enhancement pattern of gastric inner layer, as seen on earlyphase CT scans. On early and delayed views, the thickness of gastric wall and the presence of perigastric fat infiltration were determined. The enhancement patterns of gastric inner layer were classified as either continuous or discontinuous thick enhancement, thin enhancement, or nonenhancement. RESULTS: The thickness of gastric wall was 1.2-3.5cm(mean 2.2cm) in cases of adenocarcinoma and 1.2-7.6c m (mean 4cm) in lymphoma. Perigastric fat infiltration was seen in 24 patients with adenocarcinoma(80 %) and four with lymphoma(44%). In those with adenocarcinoma, the degree of enhancement of entire gastric wall was hyperdense in fifteen patients(50%) and isointense in eleven (37 %). Seven patients with lymphoma(78 % ) showed hypodensity. In those with adenocarcinoma, continuous thick enhancement of gastric inner layer was seen in 18 patients(60 %) and discontinuous thick enhancement in nine(30%). In lymphoma cases, no thick enhancement was observed. Thin enhancement of gastric inner layer was demonstrated in three patients with adenocarcinoma( 10 %) and two with lymphoma(22 %). In seven patients with lymphoma(78 %), there was no enhancement. CONCLUSION: The following early-phase findings are highly suggestive of gastric lymphoma: a gastric wall thickness of more than 3 cm; no or minimal perigastric fat infiltration, hypodense enhancement of thickened entire gastric wall; and no or thin enhancement of gastric inner layer.

Preoperative Lymph Node Staging of Advanced Gastric Cancer with Helical CT: Assessment of New TNMClassification

PURPOSE: To evaluate preoperative N staging of advanced gastric cancer(AGC) using helical CT according to thenew TNM classification. MATERIALS AND METHODS: Helical CT findings of AGCs in N staging were prospectivelyevaluated and correlated with pathologic staging in 60 patients with AGCs who underwent surgery. In all patients,contrast-enhanced helical CT with 5 -7 mm silce thickness and 5 -7 mm reconstruction was performed after ingestionof 600 - 800ml of water. A total of 150ml of contrast medium was administered intravenously at a rate of 4mL/secand CT scans were obtained 60 seconds after the initiation of intravenous administration of contrast medium. CTnodal status was assessed according to the 1997 UICC/AJCC N staging system as: N0, no lymph node metasta-sis; N1,1 -6 metastatic regional lymph nodes; N2, 7 -15 metastatic regional lymph nodes; N3, more than 15 metastaticregional lymph nodes. Lymph nodes at least 5mm in short-axis diameter or more than three lymph nodes in a focalarea (clustered appearance) regardless of size were interpretated as positive for metastasis. RESULTS: Of thetotal 1,334 lymph nodes dissected, 352(26%) were positive for metastasis. The sensitivity of helical CT scans in Nstaging of AGCs was 61%, specificity was 36 %, and overall accuracy was 55% (33 of 60cases), Nine(15 %) cases wereoverstaged and 18(30 %) were understaged. CONCLUSION: Our results indicate that the results of helical CT inpreoperative N staging of AGCs according to the new TNM classification showed no improvement despite theapplication of favorable criteria for lymph node metastasis. Further evaluation using various analytic approachesis necessary.

The Usefulness of 0.5 %-Methylcellulous as Oral Contrast Agent: In Subtotal Gastrectomy Patients

PURPOSE: To assess the usefulness of 0.5 %-methylcellulose as oral contrast agent in spiral CT examinationsfor the evaluation of anastomotic site and remnant stomach in patients who have undergone subtotal gastrectomy dueto stomach cancer. MATERIALS AND METHODS: Twenty-seven patients who underwent subtotal gastrectomy for stomachcancer and were referred for the evaluation of anastomosis recurrence and lymph node metastasis wereprosepectively analyzed by spiral CT. They were divided into two groups: before scanning, group A patients drank0.5 %-methylcellulous 500ml as oral contrast agent, while those in group B drank diluted gastrografin 500ml. Threepatients were examined twice. Anatomic delineation of the anastomosis site was graded by two radiologists asexcellent (3), good (2), fair (1), or poor (0). To evaluate the degree of distension, maximal transverse andanterior-posterior diameter of remnant stomach and anastomotic sites were measured. RESULTS: In Group A, anatomicdelineation of the anastomotic site was very much better than in group B (mean score: 2.93 vs 1.80, p<0.05). Inaddition, the maximum diameters of remnant stomach and anastomotic site were significantly larger in group A thanin group B (transverse A-P remnant stomach and anastomosis site: 87.5 +/-14.7mm, 103.3 +/-20.1mm, 17.6 +/-2.9mm vs57.6 +/-20.1mm, 69.9 +/-3 5 . 0 m m , 10.7 +/-7 . 2 m m ) CONCLUSION: In patients who had undergone subtotalgastrectomy, the use of 0.5 %-methylcellulose as oral contrast agent for spiral CT showed excellent anatomicdelineation of the anastomotic site and distension of remnant stomach.