What are the sensitivity and specificity of serologic tests for celiac disease? Do sensitivity and specificity vary in different populations?

Do we need to measure total serum IgA to exclude IgA deficiency in coeliac disease?

BACKGROUND

Screening for IgA deficiency in patients with coeliac disease is essential because of the increased incidence of IgA deficiency associated with the disease, which usually relies on the estimation of IgA levels in each case.

AIM

To devise a method of excluding IgA deficiency without measuring total serum IgA in each case.

MATERIALS AND METHODS

The optical density readings on enzyme-linked immunosorbent assay (ELISA) of 608 routine samples received for tissue transglutaminase (TTG) antibody testing for coeliac disease were compared with their total IgA concentrations. Dilution experiments were also carried out to ensure linear relationships between optical density on ELISA and IgA concentrations and to compare the sensitivities for TTG and endomysium antibodies in TTG-positive samples.

RESULTS AND DISCUSSION

A clear relationship was shown between total IgA concentration and TTG optical density readings by ELISA. To ensure a positive TTG result if antibodies are present, it was possible to recommend an optical density level above which all samples have sufficient IgA. Samples with optical density <0.05 should be investigated further by estimating total IgA and, if low, samples should be subjected to immunofluorescence microscopy testing for IgA and IgG endomysium antibodies.

CONCLUSIONS

An easier, more cost-effective and practical way of excluding IgA deficiency in the investigation on coeliac disease is reported.

Gluten sensitive enteropathy and other causes of small intestinal lymphocytosis

Recent studies on gluten sensitive enteropathy (GSE) have heightened the pathologist's awareness of intraepithelial lymphocytes in duodenal biopsies. Mild histologic forms of GSE are now recognized with increased IELs as the only histologic change. In this review, both the classic morphology and "minimal deviation" GSE are discussed in detail, with special emphasis on the differential diagnosis and how to avoid overdiagnosing this condition. The Marsh-Oberhuber classification is described and contrasted with a more practical diagnostic terminology that is recommended for use in pathology reports. Other important conditions associated with duodenal lymphocytosis, such as refractory sprue, autoimmune enteropathy, and Helicobacter pylori-associated duodenitis, are also presented.

Anti-tissue transglutaminase antibodies in patients with abnormal liver tests: is it always coeliac disease?

BACKGROUND

Coeliac disease (CD) is found in 5-10% of patients with chronically abnormal liver tests and no obvious cause of liver disease. In this population the efficacy of screening for CD by anti-tissue transglutaminase (anti-tTG) may be impaired by the high rate of positive anti-tTG found in chronic liver disease.

AIMS

To evaluate the prevalence of coeliac disease and the role of anti-tTG in patients with non-viral, non-autoimmune chronic and no obvious cause of liver damage.

METHODS

Out of 2,512 consecutive patients with abnormal liver tests, 168 (118 men, 50 women; mean age 40.7 +/- 12.6 years) were defined, on the basis of clinical data and liver biopsy, as NAFLD or cryptogenic chronic hepatitis. All were tested by recombinant IgA and IgG anti-tissue transglutaminase. Patients with a positive serology underwent endoscopy with duodenal biopsies.

RESULTS

NAFLD was diagnosed in 121 patients, in 6 associated with cirrhosis, while 47 patients were considered as cryptogenic hepatitis in the absence of steatosis. Anti-tTG were positive in 20/168 patients (3 IgA alone; 11 IgG alone; 6 both IgA and IgG). Coeliac disease was found at endoscopy and confirmed by histopathology only in the 6 patients (3.6%) with both IgA and IgG anti-tTG positivity. Four of the patients with CD had NAFLD (3.3%), in 2 of them associated with cirrhosis; while 2 of those with cryptogenic hepatitis (4.2%) had CD.

CONCLUSIONS

The prevalence of CD in patients with chronically abnormal liver tests of unexplained etiology is 4%, with no relation with the degree of liver steatosis. Screening should be done by testing for IgA and IgG antibodies and then evaluating by endoscopy and biopsy only patients positive for both.