Chapter 33 (page 162)

The Pyloric Sphincteric Cylinder in Health and Disease

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 33 (page 162)

Chapter 33

Pyloric Carcinoma

As pointed out by Castleman (l936), there was a time when it was widely believed that gastric carcinoma did not invade the duodenum. In most of the earlier text-books of surgery and surgical pathology it was stated that, while carcinoma of the stomach spread extensively in many directions by means of lymphatic and haematogenous metastases and through direct extension, it stopped abruptly at the pyloric ring. However, on reviewing the literature, Castleman (l936) collected 38 case reports of gastric carcinoma in which duodenal spread of the tumor had occurred. In the majority the extent of duodenal invasion was not more than 1.0 to 2.0 cm, but in isolated cases it could be as much as 5.0 cm. In view of these findings it was necessary to remove most of the first part of the duodenum in resections for gastric carcinoma.

In his own investigation Castleman (l936) examined microscopic preparations of 134 surgical and 65 autopsy specimens of pyloric carcinoma which had accumulated in his department during the previous 34 years. There was microscopic invasion of the duodenum in 6 (9 percent) of the autopsy, and in 15 (11 percent) of the surgical specimens. The extent of duodenal spread in the 21 cases varied from 4.0 mm to 2.3 cm. (In most of the surgical specimens carcinoma cells were found at the distal cut edge, i.e. the tumor had been transected). Spread usually occurred along the submucosa, often in the lymphatics, and rarely involved the mucosa. It was stated that spread had taken place deep to Brunner's glands (i.e. between Brunner's glands and the muscularis externa) and in the accompanying illustrations it appeared as if these glands were not involved. Usually duodenal infiltration was so slight that the surgeon had difficulty in palpating the thickened duodenal wall and it could only be detected microscopically. It was recommended that at least 3.0 cm of the duodenum be removed during gastric resection for carcinoma at the pylorus.

Dixon and Stevens (l936) drew attention to 6 cases of linitis plastica of the stomach in which histologically similar but separate, discrete lesions were found elsewhere in the gastrointestinal tract, namely in the oesophagus, jejunum, colon and rectum. No mention was made of direct extension into the duodenum in these cases. The authors had encountered 37 similar case reports in the literature.

Coller et al. (l941) studied the routes and extent of spread in 53 cases of gastric carcinoma. Duodenal involvement was seen in 14 cases (26.4 percent). In some it had occurred through direct extension of the tumor, and in others via submucosal and intermuscular lymphatic channels. The exact extent of duodenal spread was not mentioned.

Zinninger and Collins (l949) studied microscopic sections of gastric resection specimens in 36 cases of carcinoma of the stomach without any gross evidence of invasion of the duodenum. Microscopically 9 (25 percent) showed duodenal invasion by carcinoma. If only those cases were considered in which the lesion was situated within 5.0 cm of the pylorus, there were 9 cases out of 30 (30 percent) in which duodenal invasion had occurred. The extent of duodenal spread was as follows: in 2 cases it was 3.0 mm, in 4 cases 7.0 to 15.0 mm, in 2 cases 20.0 mm and in one case 60.0 mm. (The latter was an unusual case. It appeared to be a benign gastric ulcer, but was diagnosed microscopically as a mucoid scirrhous carcinoma with widespread extension. The cell types in the other 8 cases were not mentioned). The invasion of the duodenum seemed to be mostly by direct infiltration of the muscle or extension through the subserosal lymphatics. In 3 of the 9 cases spread had taken place in the submucosa. In some cases an isolated lymph node adjacent to the duodenum beyond the pylorus contained metastases without any carcinoma being found in the duodenal wall. No mention was made of carcinomatous infiltration in relation to Brunner's glands.

Eker (l951) studied the lymphatic spread of gastric carcinoma in 70 total and 100 partial gastrectomy specimens. Generally speaking extension into the oesophagus was much more frequent than spread into the duodenum. In the 70 specimens of total resection, 32.8 percent showed involvement of the oesophagus, while in the entire material of 170 cases only 2 tumors were found to have extended into the duodenum. It appeared as if the pyloric "sphincter" formed a barrier which prevented distal spread of carcinoma; similar circumstances did not pertain at the gastro-oesophageal junction.

Subsequently Eker and Efskind (l952) analyzed the frequency and extent of spread of gastric carcinoma within the different layers of the wall. For this purpose 42 total and 38 partial gastrectomy specimens were examined. Duodenal extension of the tumor was apparent in 14 of the 80 specimens. In all 14 cases the distance of duodenal infiltration beyond Brunner's glands was said to be short. Microscopically the cases were divided into adenocarcinomas, mucinous scirrhus carcinomas and colloid carcinomas.

It was found that gastric adenocarcinoma spread mainly in the mucosa and submucosa of the stomach, with spread decreasing gradually in the muscle layers toward the subserous layer. Spread in the subserous layer was slight, a fact which was especially obvious in the highly differentiated adenocarcinomas. In the 6 cases of gastric adenocarcinoma with duodenal extension, the spread in the mucous membrane stopped at the point where Brunner's glands commenced, i.e. at the line of demarcation between the stomach and duodenum. In other words, the duodenal mucosa was not involved by spread of gastric adenocarcinoma across the pylorus, while the other duodenal layers were.

In gastric mucinous scirrhus carcinomas conditions were found to be different. In most of these cases there was little spread in the gastric mucosa, most of the spread occurring in the submucosa and deeper layers of the gastric wall. In 6 of these cases duodenal extension of the tumor had occurred; in 4 of the 6 cases mucosal involvement stopped at the gastroduodenal junction, the duodenal mucosa being spared. In the other 2 cases the duodenal mucosa was involved for a distance of 2.0 mm in the first and 7.0 mm in the second case. All the other duodenal layers were involved in the spread.