摘要
Background: The diagnostic accuracy of IgA tissue transglutaminase antibodies (TGA) for coeliac disease (CD) has been assessed following the introduction of the test into routine practice in 2002. Methods: Specificity was assessed in serum samples received from 1554 adults for routine coeliac serology. The population for assessing sensitivity was 75 consecutive new adult diagnoses of CD. TGA was measured by enzyme-linked immunoassay using human tissue transglutaminase as antigen. Concordance between TGA and endomysial antibody (EMA)was also assessed. Results: The prevalence of new diagnoses of CD in the population tested was 2.8%with similar proportions of new diagnoses in males and females. The positive predictive values at a cut-off of 3 units/mL were 0.77 for samples from Primary Care and 0.92 for samples from Hospital sources, with a sensitivity of 92%. At TGA < 3 units/mL, EMA was usually negative; when TGA was >4.9 units/mL, EMA was rarely negative. Conclusions: We have assessed the role of TGA in routine clinical practice and confirmed high diagnostic accuracy. Sensitivity (92%) is identical to the sensitivity for EMA. IgA deficiency should be excluded in samples showing low absorbance readings in the TGA assay and interference from monoclonal and polyclonal IgA should be excluded in samples with slightly raised TGA levels and negative EMA. TGA is recommended as the first-line serological test for coeliac disease.
Background: The diagnostic accuracy of IgA tissue transglutaminase antibodies (TGA) for coeliac disease (CD) has been assessed following the introduction of the test into routine practice in 2002. Methods: Specificity was assessed in serum samples received from 1554 adults for routine coeliac serology. The population for assessing sensitivity was 75 consecutive new adult diagnoses of CD. TGA was measured by enzyme-linked immunoassay using human tissue transglutaminase as antigen. Concordance between TGA and endomysial antibody (EMA)was also assessed. Results: The prevalence of new diagnoses of CD in the population tested was 2.8%with similar proportions of new diagnoses in males and females. The positive predictive values at a cut-off of 3 units/mL were 0.77 for samples from Primary Care and 0.92 for samples from Hospital sources, with a sensitivity of 92%. At TGA < 3 units/mL, EMA was usually negative; when TGA was >4.9 units/mL, EMA was rarely negative. Conclusions: We have assessed the role of TGA in routine clinical practice and confirmed high diagnostic accuracy. Sensitivity (92%) is identical to the sensitivity for EMA. IgA deficiency should be excluded in samples showing low absorbance readings in the TGA assay and interference from monoclonal and polyclonal IgA should be excluded in samples with slightly raised TGA levels and negative EMA. TGA is recommended as the first-line serological test for coeliac disease.
