BACKGROUND. Since 1973, 210 patients with pancreatic carcinoma have un
dergone surgery in our clinic, including 144 with carcinoma of the hea
d of the pancreas. Of these 144 patients, macroscopic curative resecti
ons were performed on 53 (36.8%). Five patients (9.4%) died within 30
postoperative days, and an additional 3 (5.7%) died within 60 days. Th
e overall median survival was 13 months. Eight of the patients who und
erwent macroscopic curative resection survived 5 years, giving a 5-yea
r survival rate of 27.4% using the Kaplan-Meier method. The 5-year sur
vival rate was 39.7% after a microscopically curative resection and 0%
after a microscopically noncurative resection. METHODS. Outcome was c
ompared based on the extent of pancreatic cancer by constructing survi
val curves according to the general rules published by the Japan Pancr
eas Society. RESULTS. There was no statistically significant differenc
e in survival based on tumor size or stage. However, there was a signi
ficant difference in the survival of patients with the absence (so) or
presence (se) of invasion to the anterior capsule of the pancreas, th
e absence (rpo) or presence (rpe) of invasion of the retroperitoneal t
issue, the absence (ew(0)) or presence (ew(2)) of invasion at the surg
ical margin of resection, and the extent (n(0) to n(2)) of lymph node
metastasis. CONCLUSIONS. The results of this study suggest that extend
ed radical pancreatectomy may be indicated for patients with pancreati
c carcinoma because standard dissection may fail when the tumor has sp
read to the retroperitoneum or extrapancreatic nerve plexus. (C) 1996
American Cancer Society.