A total of 590 exocrine pancreatic cancers of the body or tail of the
pancreas, operated on between January 1982 and December 1988, were ana
lysed. There were 128 pancreatic resections (group 1), 164 palliative
bypasses (group 2) and 293 exploratory laparotomies which included 74
splanchnicectomies (group 3). The mortality rate was lower in group 1
(9 per cent) than in group 2 (19 per cent) (P = 0.012). The mortality
rate exceeded 40 per cent in groups 1 and 2 for patients aged more tha
n 70 years with pre-existing organ failure. The morbidity rate was 32
per cent in group 1 and 29 per cent in group 2. Patients with metastas
es had a median survival of 3.4 months, whatever the operative treatme
nt. In the presence of lymph node involvement there was no significant
difference in survival between groups 1 and 2. Patients with no metas
tases and no lymph node involvement had 1- and 3-year survival rates o
f 38 and 12 per cent respectively after pancreatic resection. Only pat
ients with a small tumour (less than or equal to 4 cm), no lymph node
involvement and no metastases achieved a significantly better survival
after resection (P=0.049). Curative resection should be reserved for
a small tumour confined to the pancreas. Fewer than 10 per cent of pat
ients will be suitable for surgery, For the other cases, resection mus
t be considered as a palliative procedure without a significant improv
ement in survival. It seems justified to limit palliative surgery to c
andidates for digestive bypass and to use non-surgical palliation for
the remainder.