Gastric Granuloma Research Articles

Circumferential Gastric Wall Thickening With Lymphadenopathy

Question: A 78-year-old Japanese man with a history of hypertension and abdominal aortic aneurysm visited our hospital for epigastric discomfort lasting more than a week. Blood examination revealed mildly elevated white blood cell counts (9900/μL) and normal red blood cell and platelet counts. Biochemical analyses indicated no specific abnormalities except mildly elevated C-reactive protein levels (1.89 mg/dL, normal range <0.14 mg/dL). Serum tumor marker levels, including carcinoembryonic antigen and carbohydrate antigen 19-9, were within the normal ranges. Although Helicobacter pylori infection was detected via serum antibody test, he had not received eradication therapy previously. Contrast-enhanced computed tomography (CT) showed circumferential gastric wall thickening with hyperenhancement on the mucosal surface, accompanied by infrapyloric lymphadenopathies (Figure A). Minimal ascites was found around the liver and in the pelvis. On fluorodeoxyglucose (FDG) positron emission tomography–CT, the thickened gastric wall and swollen lymph nodes showed intense accumulation of FDG with a maximum standardized uptake value of 5.23 and 4.33, respectively (Figure B). Moreover, minor mucosal erosions and ulcerations on the gastric body and the antrum were observed during esophagogastroduodenoscopy (EGD) (Figure C). The distensibility of the gastric wall by CO2 insufflation was not impaired on EGD. Endoscopic mucosal biopsies were performed on the thickened gastric wall with ulcerations. What is the most likely diagnosis here, and how should these lesions be treated? Look on page 339 for the answer and see the Gastroenterology website (www.gastrojournal.org) for more information on submitting to Gastro Curbside Consult. Gastric linitis plastica (GLP), a distinct subtype of gastric cancer characterized by thickening and rigidity of the gastric wall,1Vivier-Chicoteau J. Lambert J. Coriat R. et al.Development and internal validation of a diagnostic score for gastric linitis plastica.Gastric Cancer. 2020; 23: 639-647Crossref PubMed Scopus (10) Google Scholar was primarily suspected from the imaging findings. Perigastric lymphadenopathies and ascites suggested lymph node metastases and peritoneal dissemination, reflecting the aggressive behavior of GLP.1Vivier-Chicoteau J. Lambert J. Coriat R. et al.Development and internal validation of a diagnostic score for gastric linitis plastica.Gastric Cancer. 2020; 23: 639-647Crossref PubMed Scopus (10) Google Scholar However, the biopsy obtained from the margins of the ulceration did not reveal any malignant cells suggestive of GLP. Hematoxylin and eosin (HE) staining showed infiltration of numerous histiocytes and noncaseating granuloma formation (Figure D). Because the superficial biopsy often results in inconclusive diagnosis of GLP, laparoscopic resection of the swollen infrapyloric lymph nodes was performed to obtain a more substantial amount of tissue. HE staining of the surgically resected lymph nodes showed the formation of multiple small-size epithelioid granulomas (Figure E). To determine the etiology of granuloma formation, various kinds of immunohistochemical studies were performed. Accumulation of H pylori was observed inside the granuloma of both gastric mucosa (Figure F) and lymph nodes (Figure G) with the use of H pylori–specific antibody (B0471; Dako, Denmark) whereas Epstein-Barr virus or tuberculosis was not detected by in situ hybridization or Ziehl-Neelsen staining. In addition, the possibility of gastric sarcoidosis was low because the serum angiotensin-converting enzyme level was not elevated and gastric granuloma was negative for immunostaining of monoclonal antibody against Propionibacterium acnes (D371-3; MBL, Japan).2Yamaguchi T. Costabel U. McDowell A. et al.Immunohistochemical detection of potential microbial antigens in granulomas in the diagnosis of sarcoidosis.J Clin Med. 2021; 10: 983Crossref PubMed Scopus (17) Google Scholar,3Inomata M. Ikushima S. Awano N. et al.Upper gastrointestinal sarcoidosis: report of three cases.Intern Med. 2012; 51: 1689-1694Crossref PubMed Scopus (16) Google Scholar Therefore, the diagnosis of granulomatous gastritis associated with H pylori infection was made. Successful H pylori eradication therapy with vonoprazan, amoxicillin, and clarithromycin improved his symptoms. Furthermore, gastric wall thickening on CT as well as gastric mucosal ulcerations on EGD disappeared entirely 6 months after the H pylori eradication (Figures H and I), which further verified the diagnosis. This case highlights that granulomatous gastritis associated with H pylori infection needs to be considered in the differential diagnosis when encountering H pylori–positive cases exhibiting circumferential gastric wall thickening with lymphadenopathy.

Granulomatous Gastritis in Crohn’s Disease - is it Common or Rare: Review Article

Background: Granulomas in gastric biopsy specimens are extremely rare; more than half are associated with Crohn’s disease. In Crohn’s disease, clinical involvement is rare; microscopic disease is more common. Gastric granulomas are found in 9-15%, and a high proportion of Crohn’s patients show upper GI involvement, usually antral. Focal acute inflammation in a background of non-inflamed H. pylori negative mucosa is suggestive of Crohn’s disease and active inflammation in 56% (some inflammatory changes in up to 70%), but only 10% are H. pylori positive. Objective: The aim of the study was to highlight granulomatous gastritis in Crohn’s disease. Conclusion: The incidence of gastric Crohn’s is 10% of the cases. The histological diagnosis of Crohn’s disease on gastric samples is suggested by the presence of 1) a focal distribution; 2) chronic active inflammatory lesions; 3) the absence of the Helicobacter Pylori (HP) organism; and 4) granuloma. Focal stromal abnormalities could be a new diagnostic morphologic pattern in gastric Crohn’s disease, sharing the same value as focal cryptal architectural abnormalities in intestinal involvement. Since the majority of patients with Crohn’s disease lack significant symptoms, the true incidence of the disease depends on whether such pictures are sought for and identified. Keywords: Granulomatous Gastritis, Crohn’s disease, Chronic inflammatory condition

Anisakis pegreffii Larvae in Sphyraena viridensis and Description of Granulomatous Lesions.

Simple SummaryFish-borne zoonoses are caused by bacteria and parasites, while no viral fish-borne zoonoses have been reported, to date. Regarding zoonoses caused by parasites, Anisakiasis is one of the most important, with Anisakis simplex and Anisakis pegreffii agents in the central Mediterranean Sea. Humans can be infected by accidental ingestion of third-stage larvae in raw, undercooked or improperly processed fish or cephalopods. After ingestion, the larvae migrate from the gastrointestinal tract to gastrointestinal tissue, causing pain and, subsequently, inflammatory reaction leading to eosinophilic granuloma. This kind of reaction has not been described to date, in fish, and the aim of this study is to describe gastric wall lesions caused by A. pegreffii in Sphyraena viridensis and to compare them to those reported in humans, which appear macroscopically identical, albeit showing significant microscopic differences.The aim of the present study was to describe gastric granuloma caused by Anisakis pegreffii in Sphyraena viridensis caught in the central Mediterranean Sea. Sixty-eight S. viridensis specimens were collected from different fish markets on the east coast of Sicily. Coelomic organs were observed both macroscopically and with the aid of stereomicroscope. Parasite specimens and lesioned tissues were collected for identification, histological and molecular analyses. Twelve specimens (p = 17.6%) were positive for the presence of nematode larvae, morphologically identified as larvae of Anisakis sp., with values of mean abundance and mean intensity of 0.9 and 4.8, respectively. One large female specimen showed massive parasite infection associated with nodular lesions of the gastric wall. By histology, several nematode larvae encysted through the gastric wall were found. The parasite bodies were surrounded by a granulomatous reaction made up of macrophages, epithelioid cells, some lymphocytes and an external connective sheet. Molecular analysis of 18S rRNA and cox2 genes from Anisakis sp. collected larvae, identified them as A. pegreffii. The lesions here described, though macroscopically superimposable on human eosinophilic granuloma, microscopically showed significant differences in the inflammatory cells involved and in the type of immune reaction.

Clinicopathological characteristics and aetiological factors of granulomatous gastritis.

To examine the clinicopathological characteristics of granulomatous gastritis (GG) among different aetiologies, particularly Crohn disease (CD), and determine the contribution of Helicobacter pylori and the clinical significance of isolated GG. We identified 269 GG cases overall (0.19% prevalence): 220 had an underlying granulomatous disease (CD, sarcoidosis, tuberculosis) and only eight of these (3.6%) had H. pylori, fewer than the 10.3% rate among non-GG biopsies (P < 0.001). Conversely, among 49 GG cases without known cause (foreign body, undetermined, idiopathic), 13 (26.5%) had H. pylori more than background (P = 0.001). Most patients (n = 185/68.8%) had CD and these were more probably male (P < 0.001), younger (P < 0.001), white (P < 0.001) and had single (P = 0.010), smaller (P = 0.005) and antral (P = 0.027) granulomas amid inflammation (P = 0.005) compared to non-CD GG cases; younger age was independently associated with CD [P = 0.003; odds ratio (OR) = 1.13, 95% confidence interval (CI) = 1.04-1.22]. Among CD patients, younger age (P = 0.003; OR = 1.04, 95% CI = 1.01-1.07) and upper gastrointestinal (GI) symptoms (P = 0.017; OR = 2.53, 95% CI = 1.18-5.43) were associated with new (versus established) diagnosis, whereas multiple gastric granulomas (P = 0.003; OR = 4.67, 95% CI = 1.67-13.04) and lack of upper GI symptoms (P < 0.001; OR = 6.75, 95% CI = 2.94-15.49) were associated with lower GI granulomas. Of 86 isolated GG cases (i.e. no prior diagnosis or lower GI granulomas), 51 (59.3%) were eventually diagnosed with CD, and this was independently associated with younger age (P = 0.014; OR = 1.11, 95% CI = 1.02-1.21) and upper GI symptoms (P = 0.033; OR = 19.27, 95% CI = 1.27-293.31). The positive predictive value of finding isolated GG towards a CD diagnosis in patients aged <30 years was 91%, increasing in males (93%), with single (94%), antral (97%) granulomas or upper GI symptoms (94%). GG does not correlate with H. pylori in patients with granulomatous disease, but may be associated with the organism when such diagnosis is lacking. In CD patients with GG, younger age and upper GI symptoms are associated with a new CD diagnosis, whereas multiple gastric granulomas and lack of upper GI symptoms correlate with lower GI granulomas. GG, including in isolated cases with no prior clinical history or granuloma, probably signifies CD, particularly in younger, male patients or those with single, antral granulomas or upper GI symptoms.

Cover Image

The cover image is based on the Original Article Gastric Granulomas and Helicobacter pylori: An incidental relationship by Robert Genta et al., https://doi.org/10.1111/hel.12805.

Gastric granulomas and Helicobacter pylori: An incidental relationship.

Studies investigating the relationship between granulomatous gastritis (GG) and Helicobacter pylori infection have been largely inconclusive. This study was designed to determine whether the analysis of a very large number of patients would provide clearer answers evaluate the association between H.pylori infection and gastric granulomas. We used a large national database of clinicopathological data to extract 1,673,086 patients who underwent esophagogastroduodenoscopy with gastric biopsies between 2008 and 2020. In a case-control study, we evaluated the occurrence of H.pylori infection in patients with and without gastric granulomas. We also explored other clinical and histopathological associations. H.pylori infection was present in 44 of 496 (8.9%) patients with gastric granulomas, compared to 158,949 (9.5%) in the control group (OR=0.93, 95% CI=0.68-1.26). Of the 129 patients with gastric granulomas, 50 had documented inflammatory bowel disease. The results of this study show that the prevalence of H.pylori infection in patients with gastric granulomas is essentially identical to that of controls with no evidence of granulomas or granulomatous disease. When patients with and without a plausible-known association for gastric granulomas were analyzed separately, the prevalence of H.pylori infection remained remarkably similar in GG patients and controls. Considering the very large numbers of patients with gastric biopsies analyzed in this study, we submit that there is no basis for suggesting that H.pylori is etiologically related to GG.

Eosinophil granuloma due to parasite treated by laparoscopic and endoscopic cooperative surgery: a case report

BackgroundGastric eosinophilic granuloma caused by parasitic infection is rare. It is often suspected to be a malignant disease and it is difficult to diagnose. We successfully diagnosed and removed a gastric eosinophilic granuloma using laparoscopic and endoscopic cooperative surgery (LECS).Case presentationA 35-year-old woman visited our hospital because of epigastric pain. Upper gastrointestinal endoscopy revealed a 15 mm submucosal tumor (SMT) with changes in the folds, such as enlargement and convergence, located in the greater curvature of the lower gastric body. Computed tomography (CT) showed a dense, nonenhanced area of 15 mm at the same site. SMT was suspected, but undifferentiated cancer could not be excluded. We performed laparoscopic partial gastrectomy using LECS for resection biopsy.Histopathological examination showed an SMT 8 × 8 × 5 mm in size with an unclear boundary and necrosed insects at the core of the tumor. There was marked eosinophilic infiltration around the area. The diagnosis was gastric granuloma caused by parasitic infection.ConclusionsIt is difficult to differentiate gastric eosinophilic granuloma caused by parasitic infection from malignant disease. In this case, LECS is considered a minimally invasive and useful procedure.

P207 Prevalence and characteristics of Romanian patients with upper GI tract involvement in Crohn's disease

Background: Prevalence of upper GI involvement in CD has been reported to be low. This can be due to difficulties encountered in establishing a clear diagnosis. The aim of the study was to evaluate the rate of upper GI involvement in CD patients based on the data coming from a regional, population-based romanian registry, and to identify their chracteristics on diagnosis and during follow-up. Methods: In order to study the characteristics of patients with upper GI Crohn's disease, we used data from EPIROM- a regional, population-based registry. Data was collected between 2014–2016. Upper GI tract involvement was considered when endoscopy described erythema, edema, linear or aphtoid erosions or deep ulcers, combined with a histologic report describing nonspecific chronic inflammation, focally enhanced gastritis or granuloma. Results: Out of 113 patients with CD, fourteen (12.38%) were considered to have upper GI tract involvement: 10 had gastroduodenal lesions, 3 had only gastric lesions and one patient had only duodenal lesions. None of the patients had esophageal CD. Apthoid erosions and deep ulcers were the most common endoscopic findings (64.28%). Chronic active gastritis was the most common histologic finding (71.42%). Only one patient was found to have gastric granuloma. Two thirds of patients with upper GI tract involvement were considered to have severe disease, based on the endoscopical and histological scores. Upper GI tract disease was most commonly found in association with ileocolonic disease (50%), followed by ileal disease (28.57%) and colonic disease (21.42%) No cases with exclusive involvement of the upper GI tract were registered. Intestinal stenosis was the most common complication of CD in these patients and it was found to be significantly associated with upper GI tract involvement (p=0.021). Intestinal stenosis was found in 9 patients with upper GI tract involvement (64.28%), 7 of these associated ileal disease, and 2 associated ileocolonic disease. Conclusions: The rate of upper GI involvement in CD is low. Most of the patients had gastroduodenal lesions combined with ileocolonic involvement. The disease is more severe in these cases and the development of intestinal stenosis occurs more frequently. These patients usually associate a poor prognosis, thus a prompt and highly efficient therapeutic method is needed early in the course of the disease.

Upper Gastrointestinal Involvement in Crohn Disease: Histopathologic and Endoscopic Findings.

Studies describing the prevalence of upper gastrointestinal (GI) Crohn disease (CD) and its histopathologic changes have been inconsistent as a result of different definitions used for upper GI involvement, diverse populations, and varying indications for endoscopy. We reviewed the literature describing endoscopic findings and histologic lesions in gastric and duodenal mucosa of patients with established CD. PubMed, EMBASE, and the Cochrane Library were searched for gastroduodenal biopsy findings in patients with CD from 1970 to 2014. We included all retrospective and prospective studies in adults. We calculated the prevalence of the most common endoscopic and histopathological findings among patients with overall CD and upper GI CD. Of the 385 articles identified, 20 eligible studies were included. A total of 2511 patients had CD and 815 had upper GI CD. In the CD group, the most common histopathological finding was nonspecific gastric inflammation in 32% of patients, followed by gastric granuloma in 7.9%. Focal gastritis was prevalent in 30.9% of patients. In the upper GI CD group, gastric inflammation was present in 84% of patients, followed by duodenal inflammation in 28.2% and gastric granuloma in 23.2%. The most common gastric endoscopic finding in patients with CD was erythema in 5.9%, followed by erosions in 3.7%. Duodenal endoscopic findings included ulcers and erythema in 5.3% and 3.0% of patients, respectively. We found a prevalence of 34% for CD involving the upper GI tract across these 20 studies. Routine upper endoscopy with biopsies of the upper GI tract in the diagnostic workup of patients with CD can correctly classify the distribution and extent of the disease.

Crohnʼs Disease Misdiagnosed as Ulcerative Colitis: Can We Improve Our Diagnostic Accuracy?

Purpose: Clinicians rely on clinical, endoscopic, and histological findings to differentiate between Crohn's colitis (CD) and ulcerative colitis (UC). However, there are instances where the colitis is deemed to be indeterminate, or an initial diagnosis of UC is converted to CD with time. In this study, we looked at our patient population who had a change in diagnosis to try to identify potential red flags for CD in patients with an initial diagnosis of UC. Methods: This is a retrospective study performed at Georgetown University Hospital, looking at patients who had both ICD9 codes of UC and CD in our EMR between January, 2005 and May, 2013. Patient who underwent a colectomy were excluded. Patients' demographics and clinical symptoms at time of UC diagnosis and the endoscopic or clinical findings that prompted a change of diagnosis from UC to CD were recorded. Results: We identified 23 patients with an initial diagnosis of UC who were later found to have CD. Eleven were males (48%) and 12 were females (52%). There were 15 Caucasians (65%), four African Americans (17%), and four others (17%). The median age at diagnosis of UC was 21 years old (range 8-51). Twenty-two patients (95.6%) were non-smokers. Symptoms at diagnosis of UC: abdominal pain in 19 (83%), diarrhea in 23 (100%), rectal bleeding in 22 (96%). All patients had the appearance of UC on colonoscopy. The median time interval between the initial diagnosis of UC and that of CD was 8 years (range 2 months to 29 years). The diagnosis of CD was made based on the following one or more findings: colonic stricture in four (17%), spontaneous colon perforation in one (4%), anorectal abscess in one (4%), deep linear ulceration of the colon in five (22%), colonic granuloma in eight (35%). Three had anal fissures with either small bowel abscess and perforation, or deep linear ulcerations of the colon. Upper GI involvement was found in five patients (22%), gastric granulomas in three, duodenal granuloma in one, small bowel-associated abscess in one. Conclusion: A portion of patients initially diagnosed with UC are later found to have CD. The diagnosis of CD is made after clinical, endoscopic, or histological findings/complications were seen. A portion of patients were diagnosed with CD after an EGD was done for abdominal pain, and found to have upper GI disease. The lack of a history of smoking and the presence of abdominal pain as a major symptom should alert the clinician of the possibility of Crohn's disease in a patient with an initial endoscopic finding of UC.

Canine spirocercosis: clinical, diagnostic, pathologic, and epidemiologic characteristics

The nematode Spirocerca lupi is a parasite of dogs with beetles of several species serving as intermediate hosts. The medical records of 50 dogs diagnosed with spirocercosis at the Hebrew University Veterinary Teaching Hospital (HUVTH) in Israel during 1991–1999 were retrospectively reviewed and compared to a control group ( n=100). There was a seven-fold increase in the annual number of dogs diagnosed with spirocercosis during these years while the hospital caseload increased by 80%, indicating an emerging outbreak of this infection. Dogs from the greater Tel Aviv area were at the highest risk of being diagnosed with spirocercosis with 74% of the cases originating from this region compared to only 17% of the controls. The disease appeared to have a primarily urban pattern of distribution with a significantly higher percentage ( P=0.025) of dogs from cities versus rural areas, as compared to the control group. Sixty-two percent of the cases were diagnosed during the colder months of December through April. The median age of infected dogs was 5 years, with dogs 1 year old or younger at the lowest risk of being diagnosed with spirocercosis. Large breeds were at a higher risk of infection in comparison to small breeds and the Labrador Retriever was significantly over represented ( P=0.027) in the study group compared to the control population. The most common signs were vomiting or regurgitation (60%), pyrexia (24%), weakness (22%), respiratory abnormalities (20%), anorexia (18%), melena (18%) and paraparesis (14%). A caudal esophageal mass was identified by radiography in 53% of the dogs and spondylitis of the thoracic vertebrae in 33%. Fecal flotation was positive for S. lupi eggs in 80% of the dogs, and endoscopy was found to be the most sensitive diagnostic procedure and allowed diagnosis in 100% of the examined dogs. Fifty-three percent of the dogs were anemic and creatine kinase (CK) activities were elevated in 54%. Necropsy of 14 dogs revealed esophageal or gastric granulomas in 13 dogs, and an esophageal osteosarcoma in a single animal. Aortic aneurysms were found in six (43%) dogs. Out of 24, 15 dogs (63%) for which follow-up information was available died or were euthanized within 1 month of admission. The case-fatality rate decreased toward the end of the study period when improved therapy with avermectins became available.

Emerging gastritides.

The purpose of this review is to highlight two types of gastritis that have recently received much greater attention: lymphocytic gastritis and the gastritis associated with Crohn's disease. Lymphocytic gastritis is a distinctive pattern of inflammation that resembles that seen in celiac disease and lymphocytic colitis. It is associated with a diverse and unusual group of disorders in their own right, as well as having a possible relationship (real or phantom) with H. pylori infection. With respect to Crohn's disease, there is a growing recognition that, much more common than gastric granulomas, is the existence in one third or more of patients of a highly focal non-H. pylori gastritis. This recognition may help secure the diagnosis of Crohn's disease where it is equivocal, especially in children, in whom follow-up radiography and endoscopy cannot be done as readily as in adults.

Endoscopic and bioptic study of the upper gastrointestinal tract in crohn’s disease patients

In the present study, virtually all of 225 patients suffering from Crohn's disease of the lower gastrointestinal tract (small and/or large bowel) were subjected to endoscopic examination of the upper gastrointestinal tract (esophagus, stomach, duodenum); while histologic examination of the upper gastrointestinal tract was performed in a portion of the patients (54 initial esophageal, 221 initial gastric and 210 initial duodenal examinations). Statistical evaluation of the findings from the upper gastrointestinal tract revealed that: Endoscopic lesions were observed in the esophagus of 15%, the stomach of 49%, and the duodenum of 34% of the 225 Crohn's disease patients. Of the 54 patients from which esophageal biopsies were taken, 31 (57%) revealed histopathologic alterations. Of the 221 patients from which gastric biopsies were obtained, 60% revealed histopathologic alterations; the rate was 53% in the 210 patients from which duodenal biopsies were taken. Calculated from the present data, noncaseating granulomas, i.e., Crohn's disease, were present only in the stomach of 29.4% of the patients, only in the duodenum in 3.4% of patients, and in both the stomach and duodenum in 4.9% of patients. Gastric granulomas were confined to the region of the stomach body and fundus in 3.4% of the patients from which gastric biopsies were obtained and to the antrum in 15.6% of the respective patients. Both gastric regions were involved in 8.3% of the respective patients. The incidence of gastric granulomas was significantly increased in young patients, patients with enterocolic manifestations of Crohn's disease, and those with brief duration of disease. Patient sex or previous drug therapy had no effect on the incidence of granulomas. The most frequent endoscopic findings in the stomach of patients with Crohn's disease were mucosal edema, mucosal redness, and acute or chronic erosions. Only chronic erosions were of significant predictive value for the presence of granulomas, i.e., diagnosis of Crohn's disease. The most frequent endoscopic lesion in the duodenum was mucosal redness, followed by mucosal edema and aphthous lesions. Ulcers, stenosis, and mucosal redness had significant predictive values for the presence of granulomas.

Eosinophilic Gastritis

INFLAMMATORY CHANGES with eosinophilic infiltration of the walls of the gastrointestinal tract may occur in two forms. One is a circumscribed polypoid lesion, and the other consists of a diffuse infiltrative thickening of the walls. The former has been called in general eosinophilic granuloma,^1-3and in the specific case of the stomach has been described as inflammatory fibroid polyps⁴and gastric fibroma with eosinophilic infiltration.⁵This form is not usually associated with blood eosinophilia or an allergic history. The diffuse type has been recorded in the literature under such names as eosinophilic granuloma,⁶eosinophilic gastroduodenopathy,⁷eosinophilic gastroenteritis,^2,8and in the specific case of the stomach as eosinophilic gastritis,⁹gastric lesion of Loeffler's syndrome,¹⁰eosinophilic linitis plastica,¹¹and gastric granuloma with eosinophilic infiltration.¹²This type is commonly associated with blood eosinophilia and an allergic history. Both forms have been described as

Gastric Granuloma with Eosinophilic Infiltration

Gastric granuloma with eosinophilic infiltration was first described in 1949 by Vanek (6), who reported 6 cases, all showing polypoid lesions of the antrum of the stomach. The literature has been reviewed in large part in an accompanying article, in this issue of Radiology, by Rigler, Blank, and Hebbel (4). No previous description of the lesion has appeared in the radiologic literature, although 21 cases are recorded, primarily for their pathologic interest. Two additional examples are presented here. It may be added that Polayes and Krieger (3) have described a similar process in the jejunum. Certain cases have been reported (1, 2, 5) in which there was extensive infiltration of the stomach with eosinophils without the features of a gastric granuloma. These cases were seemingly of allergic origin. In the cases to be presented here, as well as in those referred to by Rigler and his associates, there was no convincing history of allergy, either in the patient's medical background or as reflected by peripheral blood studies. Histologically, these granulomas are not related to the eosinophilic granuloma of bone or eosinophilic granuloma elsewhere in the body; in these latter conditions the basic cell is the histiocyte, rather than the fibroblast. It is difficult to distinguish granuloma with eosinophilic infiltration from gastric adenoma or other submucosal lesions such as leiomyoma, heterotopic pancreatic tissue, or even a small carcinoma. Histological examination in these instances is definitive. Case Histories Case I: F. K., a 54-year-old female, complained of epigastric pain, anorexia, and low-grade fever; for two years the pain had been relieved by alkalis. The blood count was normal, and agglutination and gastric acidity studies showed nothing of pathologic significance. A gastrointestinal examination revealed a small submucosal lesion in the prepyloric region. Two months later (Fig. 1) this appeared larger; there was no demonstrable peptic ulceration. A Billroth I gastrectomy was performed and the resected specimen (Figs. 2 and 3) showed a lesion of the mucosa 3 cm. above the pyloric ring, measuring 1.5 × 1.5 cm. and protruding 5 mm. above the mucosal surface. On section, the abnormal tissue, freely movable from the submucosal fat, was of an opaque, milky-gray color. Microscopically (Fig. 4) this was a granuloma lying in both the mucosa and submucosa and interrupting the muscularis mucosae. It was not encapsulated. Basically, it consisted of delicate fibrocytic tissue with many blood vessels. Large numbers of eosinophils were scattered diffusely through the lesion, with fewer numbers of plasma and lymphoid cells. The gastric mucosa elsewhere showed an accentuation of its scalloped outline, packing of the interglandular stroma by lymphoid and plasma cells, and a minimal, widely scattered intestinalization of the lining epithelium and that of the necks of the glands.

Granulomas of stomach. I. Response to injury of muscle and fibrous tissue of wall of human stomach.

Six recently studied chronic granulomas of the stomach have aroused our interest in their etiology and pathogenesis. These 6 granulomas were found during the routine examination of 166 consecutive partially resected stomachs at the Veterans Hospital, Aspinwall, Pa. Our observations suggest that these granulomas are essentially reactions to foreign material which may be ingested or may be derived from the gastric wall through action of the gastric juice. Numerous reports in the literature of unusual gastric granulomas present varying etiologic concepts. Prior to 1940 most of the granulomas of the stomach were reported as examples of atypical neoplastic, tuberculous or syphilitic disease. Later reports have stressed the nonspecific character of these granulomas. Descriptions of the reactions in maii3 of these reports include certain basic similarities to the granulomas in our series. These lesions may complicate pre-existing ulcers,are pyloric or prepyloric, 6 ' 8 and may give rise to symptoms of pyloric obstruction. Weight loss is frequently striking, and carcinoma is often suspected. The gastric wall is always thickened and all layers of the wall may be involved. The lesion is not circumscribed but fades gradually into the surrounding normal structure. Yellow or brown miliary foci are sometimes disseminated through the edematous, but tough, thickened wall. Polypoid masses are frequent in the mucosa and may extend from the submucosa or muscularis through the serosal surface into the peritoneal cavity. The histologic characteristics of these lesions are variable but certain features are constant. The reaction involves all layers of the gastric wall but may be maximum in any one. The cellular exudate consists of macrophages, multinucleated giant cells, lymphocytes, eosinophils and plasma cells. These cells are grouped perivascularly, or around irregular, eosinophilic masses of unknown ori^'u. Vascular and connective-tissue proliferation varies in quantity and eventually leads to healing by scar. Allergic response and foreign-body response' have been suggested in discussions of the pathogenesis of these lesions.

Gastric granuloma with eosinophilic infiltration

1. 1. A case of gastric submucosal granuloma with eosinophilic infiltration is reported as the ninth case in the literature. It is characterized by a history, clinical findings and a gross lesion suggestive of a gastric malignancy. The microscopic appearance is that of submucosal fibroblastic collagenous tissue with an evenly distributed infiltration of eosinophilic leukocytes and collections of lymphocytes. 2. 2. Although an allergic reaction is suspected, the etiologic nature of the lesion is undetermined. 3. 3. This lesion does not appear to be related to the eosinophilic granuloma of bone or soft tissues either locally or microscopically. 4. 4. We believe this lesion is more common than reported to be and that by calling attention to it we may stimulate a search for the causative agent and a method of diagnosis in future cases reported.