Clinical evaluation of a serum pancreatic amylase assay
The first objective of this thesis has been to assess the clinical value of measuring serum pancreatic amylase activity in patients suspected of having disease of the exocrine pancreas. The second objective has been to identify factors other than exocrine pancreatic diseases which might influence serum pancreatic amylase activity. The stimulus to the first objective was recognition of the possibility that activity of salivary amylase might be a contributory factor in the deficiencies of total serum amylase assay as a diagnostic test. The methods employed have been the study of normal subjects and groups of patients with well defined pancreatic and non-pancreatic disorders and the estimation of their serum pancreatic amylase activity using an inhibitor method of assay, the Phadebas Isoamylase Test (Pharmacia). This method is technically simple, may be performed rapidly and gives values which approximate closely to true serum pancreatic amylase activity. The potential diagnostic performance of the results obtained have been compared to those of toal serum amylase activity. In patients with acute pancreatitis, serum pancreatic amylase activity was significantly elevated, returning towards normal in parallel with total serum amylase activity. The diagnostic performances of these 2 indices, in terms of their sensitivity, specificity and predictive value for the diagnosis, were not significantly different. Both serum pancreatic amylase activity and total serum amylase activity were significantly subnormal in patients with chronic pancreatitis. A subnormal serum pancreatic amylase activity was found to have high specificity and predictive value for the diagnosis of chronic pancreatitis. Its sensitivity for this diagnosis (80%) was less satisfactory, although significantly greater than that of total serum amylase activity. Patients with pancreatic cancer had significantly lower serum pancreatic amylase activity and total serum amylase activity than normal adults. Neither of these indices were sufficiently sensitive or specific for this disease to recommend their use in diagnosis. Serum pancreatic amylase activity did not differ significantly between male and female subjects and showed no correlation with age. In the absence of exocrine pancreatic disease, serum pancreatic amylase activity was significantly elevated in patients with chronically impaired renal function and those receiving systemic corticosteroid therapy and significantly subnormal in patients with a past history of Polya partial gastrectomy. Serum pancreatic amylase activity tended to be subnormal in association with insulin-dependent diabetes mellitus, an elevated serum bilirubin concentration and therapy with β-adrenergic blocking drugs. The conclusion of this thesis has been that serum pancreatic amylase assay with the Phadebas Isoamylase Test has no definite clinical application. It might have a role of limited value as a preliminary diagnostic test in patients suspected of having chronic pancreatitis.