June 19, 2014
1 min read
Save

Simplified celiac disease serology screening predicted disease before biopsy

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

Positive results from a simplified serological screening panel correlated strongly with positive celiac disease histology, according to new research data.

Researchers at the Rotherham NHS Foundation Trust in the United Kingdom conducted a retrospective analysis of 752 patients (66% women), including 88 with celiac disease (CD), from 2007 through 2008. Duodenal biopsy and blood tests were used to measure immunoglobulin A (IgA) anti-endomysial antibodies (EMA) and IgA anti-tissue transglutaminase (tTG).

CD patients were compared with non-CD patients — based on clinical category (high-risk, low-risk, nutrient deficient and screening), serology profiles and biopsy results — in an effort to develop an algorithm that increases accurate diagnosis at initial presentation of CD symptoms.

CD was diagnosed in 11% of high-risk patients (n=565), 9% of low-risk patients (n=156; high vs. low, P=.47), 25% of nutrient-deficient patients (n=28) and all three screening patients. Of the 71 patients who tested positive for IgA EMA and IgA tTG, the positive predictive value for CD was 97%. Similarly, of the 646 patients with negative blood tests, the negative predictive value of tTG alone for CD was 98%. Nine percent of patients with normal biopsy results (n=708) had CD, as did 59% of patients with abnormal results (n=44).

The researchers recommend that their algorithm for testing for tTG at initial presentation of CD symptoms is “a pragmatic method to detect most CD while minimizing the risk of missing the diagnosis in a few [cases].”

“Biopsying everyone irrespective of serology will potentially capture all with CD at presentation but at a cost of checking histology in 8.6 patients to identify one with CD (88 of 752),” the researchers wrote. “In contrast, our suggested algorithm will miss 8% of CD (7 of 88) but requires biopsying only 1.4 patients to identify one with CD (81 of 113).”

Disclosure: See the study for a full list of relevant financial disclosures.