Go to chapter: 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | 32 | 33 | 34 | 35 | 36 | 37 | 38 | 39
Chapter 34 (page 174)
In a historical review of malignant lymphoid tumors of the gastrointestinal tract, Dawson
et al (l96l) found that these tumors had been reported more frequently in the stomach than
in the remainder of the digestive tract. Up to that time at least 293 reports of lymphoid
tumors of the stomach had appeared in the literature, but no clear distinction had been
made between lymphomas originating primarily in the stomach and those affecting the
stomach as a manifestation of generalized, disseminated lymphomatous disease.
According to these authors the tumors could be classified as primary if the following
conditions were met: there should be no palpable superficial lymphadenopathy and no
enlargement of the mediastinal lymph nodes; the total and differential white blood count
should be within normal limits; at laparotomy the bowel (or gastric) lesion should
predominate, the only lymph nodes obviously affected being those in its immediate
neighbourhood; and finally the liver and spleen should be free of tumor.
Rappaport (l966) proposed a histological classification of the lymphomas which also
included Hodgkin's disease. For the past two or three decades his classification, as well
as five or six others, have been widely used.
Naqvi et al. (l969) found that approximately 1,200 cases of gastric lymphomas had been
recorded, and reviewed 100 cases of their own. In their experience the stomach was the
part of the gastrointestinal tract most frequently involved, and the prepyloric region the
site most commonly affected.
In discussing the pathology Ming (l973) stated that, in the stomach, malignant lymphoma
develops in the mucosa, from where the tumor cells infiltrate the submucosa and
muscularis. Seeing that this mode of growth is similar to that of carcinoma, it often
assumes the gross characteristics of carcinoma and macroscopic differentiation of the two
conditions may be difficult. Grossly the tumors may be ulcerated with rolled borders,
while others are polypoid. There may be extensive infiltration and thickening of the
gastric wall and/or mucosal folds. Histologically there is dense infiltration, with varying
degrees of admixture of mature and immature lymphoid cells and histiocytic (reticulum)
cells; there is a lack of fibrous tissue proliferation. While mucosal glands may be absent,
the remaining gastric glands usually retain their normal architecture. The presence of
reticulum fibres and absence of stainable mucin are features in favour of lymphoma (as
opposed to carcinoma).
Katz et al. (l973) described 15 cases of disseminated lymphoma with gastric involvement.
The following abnormal findings were noted at gastroscopy: enlarged, nondistensible
rugae in 13 cases, multiple superficial ulcerations in 11, nodular ulcerations in 8 and large
polypoid masses in 4. Biopsy furnished a positive diagnosis in not more than 2 of the 15
cases. The poor yield was ascribed to the infiltrative submucosal nature of secondary
lymphoma and the superficial capacity of endoscopic biopsies. The gastroscopic
appearance was found to be of greater help in the diagnosis of gastric lymphoma than
biopsies, directed brush cytology or cytology washings.
Meyers et al. (l975) pointed out that the radiographic features of primary gastric
lymphoma were not specific and included polypoid masses, ulcerations, thickening of the
walls and generalized enlargement of the rugae. These appearances could simulate
ulcerated carcinoma, gastric ulceration, granulomatous disease or hypertrophic rugae due
to other causes, e.g. Menetrier's disease. While lymphoma may occur anywhere in the
stomach, they reiterated that it commonly involved the pyloric and prepyloric area. In 7
proved cases they noted a definite tendency for the condition to spread submucosally
from the distal stomach into the duodenum, resulting in radiographically discernible
contour deformities, filling defects or ulcerations in the duodenal bulb. These features
seldom occur in gastric adenocarcinoma (Chap. 33), and were regarded as characteristic
of lymphoma, permitting a specific diagnosis to be made. In one of their cases, which
was described in detail, duodenal spread occurred both superficial and deep to the lamina
propria. This surrounded, but apparently did not invade, the glands of Brunner.
Koehler et al. (l977) analyzed the records of l9 patients who underwent gastric resection
for lymphoma. Three of these had microscopic evidence of invasion of the duodenum,
and in one of these the duodenal involvement was evident radiographically and grossly.
Lim et al. (l977) defined primary gastric lymphoma as a tumor apparently originating in
the stomach, with no clinical, laboratory or radiographic evidence of systemic
involvement at the time of initial evaluation. They analyzed 50 consecutive cases and
found the most common localities to be the "antrum", the pyloric area and lesser
curvature. According to Rappaport's (1966) classification, the diffuse histiocytic type
occurred in 23 of their 50 patients, the well differentiated lymphocytic type in 12, the
mixed lymphocytic histiocytic in 5, the poorly differentiated lymphocytic in 5, Hodgkin's
disease in 3, and unclassified lymphoma in 2 cases. The prognosis for survival was much
better than in gastric carcinoma since the latter condition presented with a far higher
incidence of serosal penetration and nodal and distant metastases. The treatment of
choice was gastric resection (44 of their cases being resectable), while radiotherapy and
chemotherapy constituted additional therapeutic options.
Seeing that the therapy and prognosis of gastric lymphoma differ significantly from those
of adenocarcinoma, Hricak et al. (l980) reiterated that correct diagnosis was essential. By
reviewing the radiographic and pathological features in 81 cases, they investigated the
incidence of spread from the stomach into the oesophagus, and from the stomach into the
duodenum. In 60 of the cases the tumor appeared to have originated primarily in the
stomach, and only 5 of these showed involvement of adjacent nodes, the mesentery or
small bowel. In 21 patients the involvement was mainly extrinsic, the stomach being
secondarily involved. Radiographically transpyloric extension was seen in 27 of the 81
cases. In 7 the "antrum" was the only site of tumor involvement, 6 of these showing
transpyloric extension. Histologically the tumor was found to have a submucosal origin
(according to Ming it originated in the mucosa). Submucosal extension across the
pylorus was seen in l9 of the 27 cases.
Previous Page | Table of Contents | Next Page
© Copyright PLiG 1998