Carcinoma of the pancreatobiliary system often produces perineural invasion extending to extrabiliary and extrapancreatic sites. A surgical technique has been developed to manage this invasion and has been used since 1974. Serial sections were prepared from 90 resected specimens and examined for perineural involvement of intramural and extramural biliary plexuses as well as the pancreatic nerve plexus. Perineural invasion was seen in 34 of 40 patients with carcinoma of the common bile duct, in ten of 14 with cancer of the gallbladder and in four of 15 with carcinoma of the papilla of Vater. Invasion extended to the extramural biliary or pancreatic nerve plexuses in 24 of 40 patients with carcinoma of the bile duct. Involvement of the intrapancreatic nerves was seen in all 21 patients with carcinoma of the pancreatic head and that of the pancreatic nerve plexus in 17 of 21. Perineural invasion was often found at the most progressive margin of the tumour. The survival rate of patients with perineural invasion was low compared with that of those without such invasion, although survival of patients with perineural invasion tended to be longer after extensive resection. The perineural space should be regarded as an important route for the spread of pancreatic and biliary carcinoma.

Four cases are reported of the intraductal variant of peripheral cholangiocarcinoma among surgical specimens from 32 cases of hepatolithiasis. The cancers arose from the periphery of the stone‐containing bile duct and spread chiefly along the luminal surface. Microscopically, these tumors showed papillary proliferation and therefore were diagnosed as the intraductal spreading type of peripheral cholangiocarcinoma. Mucosal dysplasia also was noticed in the vicinity of the tumors. In six other cases, mucosal dysplasia was observed in the periphery of the stone. Immunohistochemically, anti‐CA 19‐9 staining was observed diffusely in the cytoplasm of dysplastic lesions and carcinomas. Anticarcinoembryonic antigen staining was restricted to the luminal surface and/or the supranuclear region of the cytoplasm in carcinomas. It was not identified in dysplastic cells. These results suggest that the mucosal dysplasia occasionally observed near stones is a precursor of the intraductal spreading type of peripheral cholangiocarcinoma in the presence of hepatolithiasis. The authors hypothesize that the lining epithelium of the large bile duct, when persistently exposed to biochemically altered bile, may undergo a carcinomatous transformation through a stage of mucosal dysplasia. Cancer 68:2217–2223, 1991.

To clarify whether or not the lymphatic routes that have long been generally accepted are indeed correct, we retrospectively examined the clinical records of patients with solitary lymph node metastasis from gastric carcinoma. From 735 patients gastrectomized with lymph node dissection (more than D1), 51 (7%) were histologically proven to have only one lymph node involved. In 44 of these 51 patients, the involved nodes were all in the perigastric region (N1). There were also 7 patients with a jumping metastasis to the N2-N3 nodes. Three of them were found along the left gastric artery (#7 according to Japanese classification) and the other 4 were found along either the common hepatic artery (#8) or the proper hepatic artery (#12). The depth of invasion was submucosal in 2, proper-muscular in 2, subserosal in 1, and serosa-exposed in 2, and the conclusive stage was II in 2, IIIa in 3, and IIIb in 2. However, 1 of these patients died of liver cirrhosis and 2 died of pneumonia, while the other 4 were still alive at the time of this report more than 5 years after surgery. These results suggest that not every sentinel node is located in the perigastric region near the primary tumor and that, if the preoperative examination indicates submucosal invasion, then a systematic regional lymph node dissection should therefore be carried out.

No abstract

Since 1973, 152 patients with pancreatic carcinoma have undergone surgery in our clinic, including 110 with carcinoma of the head of the pancreas. Of these 110 patients, resections were performed on 43 (39.1 per cent), 33 (30 per cent) of whom underwent a curative resection based on macroscopic evidence. Six of the patients who underwent macroscopic curative resection survived for five years, giving a five-year survival rate of 36.5 per cent by the Kaplan-Meier method after excepting 6 operative deaths. We compared the extent of pancreatic cancer by constructing survival curves according to the General Rules published by the Japan Pancreas Society. There was no statistical difference in survival based on tumor size or stage, however, there was a significant difference in the survival curves of so and se, being the absence or presence of the anterior capsule of the pancreas, rpo and rpe, being the absence or presence of invasion of the retroperitoneal tissue; ew(-) and ew(+) being the absence or presence of invasion at the surgical margin of resection, or n0 and n1 being the extent of lymph node metastasis. The results of this comparison suggest that extended radical pancreatectomy may be indicated for the treatment of pancreatic cancer as the standard radical operation for pancreatic cancer may miss tumors which have spread to the retroperitoneum and extrapancreatic nerve plexus.

We present a case of portal-systemic encephalopathy due to a congenital splenorenal shunt. A 69 year old woman was admitted to hospital because of recurrent episodes of disturbed consciousness. The present episode had begun 3 months prior to admission. Although the patient demonstrated mildly slurred speech, the remainder of her neurological examination was unremarkable. She showed no clinical signs of portal hypertension and her liver function, except for a serum hepaplastin test of 58% and an ICG retention rate of 28% at 15 min, was normal. Her serum ammonium level was 210 micrograms/dL. The venous phase of a superior mesenteric arteriogram revealed a splenorenal shunt and narrowing of the portal vein, which was 4 mm in diameter. The histological findings, demonstrated by a needle liver biopsy specimen, were consistent with mild fibrosis and lymphocytic infiltration. Following the diagnosis of a splenorenal shunt in the absence of liver cirrhosis, ligature of the shunt was performed with a splenectomy. The portal vein pressure after ligature of the shunt rose from 12.5 to 18.8 mmHg. This shunt was thought to be of congenital origin. The high preoperative serum ammonia concentration decreased to the normal range postoperatively and the serum hepaplastin test and ICG retention rate similarly improved postoperatively. A follow-up superior mesenteric arteriogram was performed during the venous phase, demonstrating resolution of the shunt and decreased portal vein narrowing. The patient has suffered no further episodes of disturbed consciousness postoperatively.

To clarify whether or not the lymphatic routes that have long been generally accepted are indeed correct, we retrospectively examined the clinical records of patients with solitary lymph node metastasis from gastric carcinoma. From 735 patients gastrectomized with lymph node dissection (more than D1), 51 (7%) were histologically proven to have only one lymph node involved. In 44 of these 51 patients, the involved nodes were all in the perigastric region (N1). There were also 7 patients with a jumping metastasis to the N2-N3 nodes. Three of them were found along the left gastric artery (#7 according to Japanese classification) and the other 4 were found along either the common hepatic artery (#8) or the proper hepatic artery (#12). The depth of invasion was submucosal in 2, proper-muscular in 2, subserosal in 1, and serosa-exposed in 2, and the conclusive stage was II in 2, IIIa in 3, and IIIb in 2. However, 1 of these patients died of liver cirrhosis and 2 died of pneumonia, while the other 4 were still alive at the time of this report more than 5 years after surgery. These results suggest that not every sentinel node is located in the perigastric region near the primary tumor and that, if the preoperative examination indicates submucosal invasion, then a systematic regional lymph node dissection should therefore be carried out.

Between 1978 and 1988, 15 patients with gallbladder cancer and 2 patients with bile duct cancer were seen among 49 patients with anomalous union of the pancreaticobiliary ductal system. Radiographic findings revealed two types of this anomalous condition: one in which the pancreatic duct entered the common bile duct (type 1) and one in which the common bile duct entered the pancreatic duct (type 2). In gallbladder cancer, the common bile duct presented no dilatation, or in some patients, mild dilatation, and type-1 anomalous union was frequently found among these patients. In contrast, the two patients with bile duct cancer had cystic dilatation of the common bile duct and type-2 anomalous union. The bile amylase level, which was determined in seven patients, was extremely high in all the patients. Histopathologically, the tumors in most patients showed papillary to papillo-tubular proliferation in the mucosal layer while atypical epithelial hyperplasia was noted in the vicinity of the tumor area. These findings suggest that this congenital anomaly in both ducts results in a loss of the normal sphincteric mechanism of the duodenal papilla, and that chronic relapsing cholecystitis or cholangitis, caused by the reflux of pancreatic juice into the biliary tract, can induced progressive changes to atypical epithelial hyperplasia which may develop into carcinoma.