Celiac Disease Diagnosis in Patients With Weakly Positive Serum Anti-Transglutaminase: Duodenal Anti-Endomysium Assay.

Celiac disease (CD) is a chronic immune-mediated disorder that occurs in genetically predisposed populations. Patients affected by the disease may be asymptomatic or manifest classic malabsorption symptoms of diarrhea, steatorrhea, abdominal pain, and weight loss after gluten ingestion (and related derivatives found in other grains). The astute clinician must be aware of a more subtle GI picture, as well as non-GI signs and symptoms (e.g., iron-deficiency anemia, abnormal liver function tests, and type 1 diabetes mellitus). Diagnosis and screening begin with the use of serologic tests, i.e. immunoglobulin A (IgA) anti-tissue transglutaminase (tTG) and IgA anti-endomysial antibodies (EmA). The EmA test is performed by indirect immunofluorescence against reagent monkey esophagus or human umbilical cord. The EmA bind to the smooth muscle antigen tTG, and the test result is reported as titers. EmA are both moderately sensitive and highly specific (sensitivity 85%-98%, specificity 97%-100%) for CD, but quickly become negative when the patient starts a gluten-free diet. Anti-tTG is a highly sensitive and specific test for CD and is widely available as an automated enzyme-linked immunosorbent assay from multiple manufacturers. Although the sensitivity and specificity of this test is high, tTG can be present in as many as 5% of control patients. In addition, false-positive tTG has been reported in patients with Crohn's disease. Duodenal biopsy, still considered by many as the criterion necessary for diagnosis, demonstrates the pathologic findings of small intestinal villous atrophy, crypt hyperplasia, and intraepithelial lymphocytosis that occur on exposure to dietary gluten. These changes exhibit improvement after withdrawal of gluten from the diet. Genetic tests, revealing permissive haplotypes, may be helpful in identifying susceptible individuals.

CD diagnosis is still anchored to the criteria established by the European Society of Pediatric Gastroenterology Hepatology and Nutrition in 1990. These require the mandatory presence of (a) villous atrophy with crypt hyperplasia and increased intraepithelial lymphocytes (IELs) count when the patient is eating gluten, and (b) a full clinical remission after elimination of gluten from the diet. However, 20 years after those diagnostic criteria were established, there is now growing evidence that at least two other factors must be considered. First, symptomatic gluten sensitivity and malabsorption may coexist with a normal-looking mucosa. Second, the finding of circulating IgA antibodies to tissue transglutaminase (anti-tTG) or endomysium (EmA) at diagnosis is an extremely accurate diagnostic instrument. As a consequence, CD diagnosis is nowadays also being done in symptomatic patients presenting with a normal-looking mucosa but with a high titer of serum EMA and/or anti-tTG antibodies. However, patients with minimal or no intestinal histology lesions pose a considerable problem, as serum anti-tTG and EmA are known to be often negative, or weakly positive, in patients with CD with mild intestinal damage.

Picarelli et al., in 1996, tried to establish whether the small intestine of CD patients is the site of EmA production and whether gliadin challenge could induce their release. The authors evaluated small intestine biopsy samples from treated and untreated CD patients and controls, cultured in vitro for 24-48 h in the presence of gliadin. EmA were detected in the supernatants of these organ culture biopsy samples by immunofluorescence technique and enzyme-linked immunosorbent assay (ELISA), respectively. No EmA were found in the culture supernatants of biopsy samples of controls, whereas they were detected in the culture supernatants of all untreated CD patients, irrespective of gliadin challenge. Conversely, EmA were not detected in supernatants of biopsy samples cultured in medium only from treated CD patients, but were detected in the majority of the biopsy samples challenged with gliadin.

The investigators, in 2002, measured anti-tTG antibody in the culture medium of intestinal biopsy specimens from patients with suspected CD and evaluated the relationship between antibody production and severity of intestinal mucosal damage, and demonstrated that anti-tTG assay of the culture medium of biopsy specimens can improve the accuracy of CD diagnosis in patients negative for serum antibodies.

The same investigators, in 2011, evaluated the diagnostic accuracy of EmA assay in the culture medium of intestinal biopsies for CD diagnosis and demonstrated that EmA assay in the culture medium had a higher sensitivity and specificity than serum EmA and anti-tTG assay. All patients with CD who were tested as false-negatives for serum EmA and/or anti-tTG carried the human leukocyte antigen alleles associated to CD. Furthermore, during the follow-up, a subgroup of patients with negative-serum EmA/anti-tTG, normal villi architecture, and positive-EmA in the culture medium, developed villous atrophy and underwent gluten-free diet, with consequent resolution of the symptoms and complete intestinal histology recovery. The investigators concluded that EmA assay in the culture medium should be included in the diagnostic criteria for CD diagnosis in "seronegative" patients.

The present study is performed to investigate the clinical usefulness of the in vitro production of EmA in CD diagnosis in a large number of consecutive adult patients with suspected CD and weakly positive [e.g. normal value multiplied for 2-3 times, 2-3xN] serum anti-tTG.