ADEQUACY OF DUODENAL BIOPSIES FOR THE DIAGNOSIS AND EXCLUSION OF COELIAC DISEASE- A MISSED OPPORTUNITY?
Introduction: Published guidelines recommend at least 5 biopsies of the small bowel be performed to make an accurate diagnosis of, or confidently exclude coeliac disease, with at least two from the first part (D1) and at least four from the second part (D2). We aimed to retrospectively audit the site and number of biopsies being taken from the proximal small bowel in patients undergoing endoscopy at a regional hospital network to evaluate the degree of compliance with published guidelines. Secondary aims were to determine the correlation between the number and site of biopsies with proceduralist specialty and clinical suspicion of coeliac disease.
Methods: Histopathology reports from adult patients undergoing endoscopy with small bowel biopsies between 2011-2012 were retrospectively examined and cross-referenced with endoscopy reports and patient files. For patients undergoing multiple endoscopies only the first endoscopy was included. The indication for endoscopy and proceduralist designation was also recorded. Where available, data from other biopsy sites and coeliac antibodies were examined to identify a diagnosis CD.
Results: Of 1091 histopathology reports examined, 111 (10.17%) had at least one biopsy from D1 and 82 (7.52%) had 2 or more biopsies obtained. Of the 1078 procedures where D2 biopsies were obtained 448 (41.56%) had 4 or more biopsies taken. Gastroenterologists were significantly more likely to obtain an appropriate number of biopsies compared to surgeons (465/967 (48.09%) vs. 21/99 (21.21%) respectively, (OR 3.4 (CI 2.09-5.66)). Of all endoscopies, 70 had known or strongly suspected CD (positive coeliac antibodies (n=39) or pre-existing CD (n=31)). Of these procedures, 43 (61.42%) had any D1 biopsies and 36 (51.43%) had 2 or more biopsies. Sixty-five (92.86%) had D2 biopsies, and 54 (77.14%) had 4 or more biopsies. Known or suspected coeliac disease was associated with an increased likelihood of an appropriate number of biopsies being taken (OR 4.75 (95% CI 2.68-8.40)).
Conclusions: Despite existing guidelines relevant to the number of duodenal biopsies for the diagnosis or exclusion of coeliac disease, less than half of procedures meet the minimum requirements. Even where coeliac disease was strongly suspected or known of the number of biopsies was deficient in a third of procedures and only 60% had biopsies obtained from D1. Gastroenterologists demonstrated better compliance with the guidelines compared to surgeons although numbers in the surgeon group were small. There is a risk that coeliac disease is being missed given the known patchy nature of this common condition and awareness of current guidelines needs to improve.