Evaluation of a fecal pancreatic elastase-1 enzyme-linked immunosorbent assay: Assessment versus an established assay and implication in classifying pancreatic function

Abstract

Background Disagreement continues regarding 2 fecal pancreatic elastase-1 (PE-1) ELISAs and their respective capabilities to assess pancreatic function. Methods The BioServ Diagnostics polyclonal PE-1 ELISA was validated and its performance characteristics compared to the previously validated ScheBo ® Biotech monoclonal PE-1 ELISA. Split sample study results were analyzed by Deming regression and Bland-Altman plot analysis. Data mining was utilized to explore PE-1 distribution and evaluate PE-1 and fecal fat correlation. Results Analysis demonstrates limited quantitative agreement; slope = 0.9640, intercept = 10.787, R 2 = 0.633. Means were 228.8 and 226.2 µg PE-1/g stool for the polyclonal and monoclonal assays respectively. Bland-Altman analysis showed 91% of paired values within 2 SD of their means. There was good qualitative agreement when interpreted against established intervals with 91% of results equivalent in pancreatic function classification. The remaining 9% varied by one classification level with no bias evident. The distribution of PE-1 concentrations ( n = 400, 0–25 years) classified 78% of subjects with normal pancreatic function, 7% with moderate pancreatic insufficiency and 15% with severe insufficiency. There was little agreement between PE-1 and fecal fat results. Conclusions The polyclonal PE-1 ELISA is an acceptable alternative to the monoclonal PE-1 ELISA. PE-1 is a potential substitute for fecal fat for evaluating pancreatic function.