HOME
Return to Table of Contents
233 PHYSICIAN MANAGEMENT OF POSITIVE CELIAC DISEASE SCREENS KE McGowan1,3 , SD Loken2,3, ME Lyon2,3, JD Butzner1 We evaluated test accuracy as well as physician interpretation and action on positive IgA-endomysial antibody (EMA) screening test results in the Calgary Health Region.
1Department of Pediatrics, 2Department of Pathology & Laboratory Medicine, 3Calgary Laboratory Services (CLS), University of Calgary, Calgary, Alberta
METHODS: Consecutive EMA results were obtained from the CLS laboratory database between March 2003 and July 2004. Seropositive EMA results were cross-referenced with the CLS pathology database to identify patients with corresponding intestinal and skin biopsies. Questionnaires were sent to ordering physicians regarding the management of seropositive patients with no record of biopsy.
RESULTS: Over 17 months 11,716 EMA tests were ordered for 9,533 patients. 349 positive EMA tests were observed in 313 (3%) patients (233 adults, 80 children) and intestinal biopsies were performed in 218 (70%) patients. One skin and 193 intestinal biopsies were diagnostic of celiac disease yielding a PPV of 89% and a disease prevalence of 2%. Eight biopsies demonstrated other diagnoses, 15 were normal (four on gluten-free diet [GFD]) and one specimen was inadequate. An additional ten patients with a negative EMA displayed celiac disease on biopsy yielding a maximum sensitivity of 95%. Of the 109 positive screens performed in 95 patients with no biopsy, 28 had appropriate indications for no biopsy, including follow-up of a previously biopsy-diagnosed patient (21), the patient was unable to undergo biopsy due to a medical condition (6) and repeated screening was negative one year later in an asymptomatic patient. Physicians inappropriately managed 33 tests. In 21 of these tests the physician recommended a GFD without a biopsy, six received no further investigation and six were unnecessary repeats less than three months after the intestinal biopsy. Administrative issues prevented evaluation in 19 positive screens (test or patient lost). Eleven patients refused further evaluation (five began GFD) and 18 surveys were not returned.
CONCLUSIONS: EMA test accuracy is approximately 90%, similar to published values, in a population with 2% prevalence. Most positive EMA screening tests were appropriately managed by physicians. Beginning a GFD without biopsy or no follow-up of a positive screen remain common errors of management.