Pancreatic phlegmon. Clinical features and course.

The clinical course of 19 patients with pancreatic phlegmon, as diagnosed by computed tomography (CT) and clinical criteria, was assessed retrospectively and compared to that of eight patients with pancreatic abscess diagnosed either at surgery or with percutaneous aspiration. Controls consisted of 55 patients with uncomplicated acute pancreatitis without CT scans and 11 patients with acute pancreatitis in whom CT scans were negative or only consistent with acute pancreatitis (no phlegmon). The age, sex, and presumed etiology of the pancreatitis were not significantly different in the four groups. Patients with phlegmon had a higher incidence of severe pancreatitis as defined by Ranson's criteria, presence of an abdominal mass, as well as a longer duration of fever, abdominal pain and leukocytosis than controls without CT scans. With the exception of a palpable abdominal mass and fever lasting over five days, the results were similar when comparing the phlegmon group and controls with CT scans, although the severity of the disease and prolonged abdominal pain tended to be increased in the former patients. There was no statistically significant difference in clinical or laboratory criteria between the phlegmon and abscess groups, although the latter group had longer hospital stays and periods with no oral intake (npo). Management of patients with phlegmon tended to include TPN, longer npo periods, antibiotics, and longer hospital stay than in controls without CT scans. Controls with CT scans were managed similarly to the phlegmon group because of prolonged amylase elevation and abdominal pain. Percutaneous aspiration was successful in differentiating abscess from phlegmon in five of six cases. Major complications were rare in the phlegmon group and spontaneous resolution was the rule. Pancreatic phlegmon is a distinct clinical/radiologic entity which may be very difficult to differentiate clinically from pancreatic abscess. Early percutaneous thin-needle aspiration of the inflammatory mass (under CT guidance) seems to be the diagnostic procedure of choice. Management is nonsurgical unless complications arise. The role of TPN and antibiotics is unknown, and controlled studies of these therapeutic approaches in pancreatic phlegmon are needed.