Pancreatic surgery, not pancreatitis, is the primary cause of diabetes after acute fulminant pancreatitis.

Acute fulminant pancreatitis is associated with significant morbidity and mortality. To examine the outcome of conservative and surgical treatment of this disorder, 36 patients who survived an initial episode were restudied after a mean of six years. Fifty three per cent had developed diabetes mellitus, half of whom required insulin therapy. Pancreatic resection was associated with a 100% frequency of diabetes, while only 26% of those treated with peritoneal lavage developed this (p less than 0.001). Insulin secretion and sensitivity were assessed using the hyperglycaemic glucose clamp technique. First phase insulin secretion was impaired in surgically treated patients (mean (SEM) 14 (5) microU/ml x 10 minutes) compared with conservatively treated patients and control subjects (144 (66) and 87 (12) microU/ml x 10 minutes, respectively; p less than 0.05). Second phase and 'maximal' insulin secretion were also impaired among the surgically treated patients compared with the conservatively treated patients and the controls. Insulin sensitivity was reduced among the surgically treated patients (2.88 (58) mg/kg.minute) when compared with conservatively treated patients and healthy control subjects (5.87 (1.02) and 6.45 (0.66) mg/kg.minute; p less than 0.05). Pancreatic resection is associated with a very high frequency of diabetes compared with peritoneal lavage, and these results favour conservative treatment of active fulminant pancreatitis whenever possible.