Antibody to Tissue Transglutaminase May Fall Short for Accurate Triage of Celiac Disease

Peggy Peck

November 03, 2004

Nov. 3, 2004 (Orlando) — A study evaluating the utility of antibody to tissue transglutaminase (tTG) to aid in diagnosis of celiac disease suggests that relying on tTG is likely to result in "underestimating the prevalence of celiac disease," said Julian Abrams, MD, from Columbia University Medical Center in New York City.

Dr. Abrams presented his analysis during the presidential plenary at the 69th annual scientific meeting of the American College of Gastroenterology (ACG).

He noted that "people like tTG because of its convenience and low cost, but because there is no standard assay used by commercial labs, both the sensitivity and selectivity of tTG vary from lab to lab." Dr. Abrams noted that studies promoting tTG "used research laboratories, which are not subject to the wide variation seen in commercial labs."

The favorable tTG studies in the literature have suggested that tTG had a sensitivity and specificity similar to endomyosial antibody (EMA), Dr. Abrams said. "This finding resulted in many labs using tTG rather than EMA, and a recent [National Institutes of Health] consensus conference advocated the use of tTG. It is currently the most widely used antibody test for celiac disease."

In the study, Dr. Abrams and colleagues attempted to evaluate the efficacy of tTG in a general referral practice setting. They reviewed patient records from January 2000 through December 2003 and identified 137 patients who had a duodenal biopsy for celiac disease and tTG performed at diagnosis. Biopsy-confirmed celiac disease was confirmed in 117 patients. Serum was sent to four commercial laboratories for analysis, and the results from those labs were compared.

Biopsies were reported as normal, partial villous atrophy (PVA), subtotal villous atrophy, or total villous atrophy (TVA). Overall, the average tTG sensitivity was 71%, and specificity was 67%. For patients with "TVA, sensitivity was as high as 92%, and for patients with PVA, it was as low as 38%," Dr. Abrams said.

One of the four laboratories received samples from 48 patients, he said. "At that lab, tTG sensitivity was 51% and specificity was 100%," Dr. Abrams said.

The commercial laboratories were not named, and Douglas K. Rex, MD, a professor of medicine at the Indiana University School of Medicine in Indianapolis, told Medscape, "that will be very interesting information. The study makes the point that the accuracy of the results depend upon the lab that you chose, so I think we would all like to know more about these labs so that we can choose the right one."

Dr. Rex, who is the immediate past president of ACG, chaired the presidential plenary session where the paper was presented. He was not involved in the study.

The findings are thought-provoking, Dr. Rex said, because "we've all been carrying around the impression that tTG had sufficiently high sensitivity and specificity that the results are almost diagnostic and could possibly be used by itself. But now we see that when you take it out of the setting of clinical study, this is not the case."

ACG 69th Annual Scientific Meeting: Abstract 4. Presented Nov. 1, 2004.

Reviewed by Gary D. Vogin, MD