Pancreatic Lipase—Determination in Canine Pancreatic Secretion and Human Blood Serum

Abstract

A previously reported method for lipase determination was modified to increase the sensitivity so as to permit application of the method without change to both pancreatic secretion and human blood serum. The errors determined in serial dilution experiments with pancreatic secretion were found to be 9.2% for deviations from linearity, and 9.1% for duplicate error (both expressed as coefficient of variation). The method was applied to the reinvestigation of the question of the lipase/protein nitrogen relationship in canine pancreatic secretion obtained a) under as identical conditions as possible, and b) following different test meals and dietary regimens. Two statistical approaches were utilized: a) analysis of covariance, to evaluate variations in the lipase activity/protein nitrogen relationship, and b) analysis of variance, to evaluate variations in the lipase concentration/protein nitrogen relationship. Highly significant variations in both ratios from specimen to specimen were found in the vast majority of our studies. These variations did not conform to the concept of adaption of the individual enzymes to the type of food ingested. Studies with human blood serum revealed the method to be specific for pancreatic-type lipase. No lipolytic activity was detected in normal serum, while in many cases of pancreatic disease, not only was lipolytic activity detected, but the lipase concentrations observed closely followed the clinical course of the disease. The reliability, specificity and rapidity of the method make it suitable for clinical studies.