Adherence to biopsy guidelines increases celiac disease diagnosis

Background: Celiac disease (CD) is common but underdiagnosed in the United States. A proposed quality guideline recommends that >= 4 specimens be submitted during duodenal biopsy. The degree of adherence to this recommendation in clinical practice is unknown. Objective: To measure the number of specimens submitted during duodenal biopsy among patients throughout the United States and to determine the incremental diagnostic yield of adherence to the recommended number of specimens. Design: Retrospective cohort study. Patients: This study involved 132,352 patients without known CD who underwent duodenal biopsy. Intervention: Duodenal biopsy. Main Outcome Measurements: Duodenal biopsy specimens were submitted to a pathology laboratory operating in 43 states in the United States. We used multivariate logistic regression to identify factors associated with submitting >= 4 specimens. We also compared the prevalence of newly diagnosed CD in biopsies with >= 4 specimens with that in biopsies with <4 specimens. Results: Of the 132,352 patients who underwent biopsy (67% women, mean age 52.9 years), >= 4 specimens were submitted in 45,995 cases (35%). A modest increase in the proportion of biopsies with >= 4 specimens occurred after this guideline was proposed in 2006 (odds ratio for 2009 vs 2006, 1.51; 95% confidence interval, 1.22-1.88), but the rate of adherence in 2009 remained low at 37%. Among patients in whom the indication was malabsorption/suspected CD (n = 3261), adherence to this standard was only 39.5%. The probability of a new diagnosis of CD was increased when >= 4 specimens were submitted (1.8% vs 0.7%; P < .0001). Limitations: Retrospective analysis lacking clinical follow-up. The guideline publication occurred during the study period, possibly influencing clinical practice and confounding results. Conclusion: Although this proposed standard remains a subject of debate, adherence to submitting >= 4 specimens is low in the United States. Adherence yields a diagnosis rate of 1.8%, a small absolute increase but a doubling of the diagnosis rate of CD. Efforts to increase adherence are warranted. (Gastrointest Endosc 2011;74:103-9.)