Limited utilization of serologic testing in patients undergoing duodenal biopsy for celiac disease

BACKGROUND Clinical algorithms for the workup of celiac disease often recommend the use of serologic assays for initial screening, followed by duodenal biopsy for histologic confirmation. However, the majority of duodenal biopsies submitted to pathology for "rule out celiac" are negative. The objective of this study was to determine the underlying causes for this low diagnostic yield. METHODS We performed a retrospective review of pathology reports from 1432 consecutive duodenal biopsies submitted for pathologic assessment to "rule out celiac" and correlated biopsy results with results for concurrent serologic testing for celiac autoantibodies. RESULTS The majority of patients had no record of serologic testing prior to biopsy, and evidence of positive serology results was found in only 5% of patients. Most duodenal biopsies were submitted as part of a multi-site GI sampling strategy that included biopsies from other locations. In this context, serologic results correlated with the likelihood of significant duodenal and non-duodenal findings, and were also helpful in evaluating patients with indeterminate duodenal histology. CONCLUSIONS The presence of a positive screening test for celiac autoantibodies does not appear to be a major driver in the decision to submit duodenal biopsies for evaluation of celiac disease, which accounts for the low incidence of findings in these samples. In patients where celiac serology testing was performed, the results were a good predictor of the likelihood of findings on biopsy.