Background: The determination of class G and A serum antigliadin antibodies remains one of the most widely used screening tests for coeliac disease. The results from different laboratories are not always comparable, on account of changes in the technique and the different ways of expressing the results.
Aims: To: a) evaluate the physiological variation of serum antigliadin antibodies expressed in ng/ml, and b) establish the cut-off of quantitative antigliadin antibodies.
Patients: Patients were 127 individuals with active coeliac disease. Controls were 395 non-coeliac subjects (198 females and 197 males) aged 6 months to 45 years (median age: 4.9 years).
Methods: Antigliadin antibody enriched samples were obtained by affinity chromatography. The concentration of the eluted antigliadin antibodies was evaluated by nephelometry and enzyme-linked immunosorbent assay to establish a primary standard. An enzyme-linked immunosorbent assay for antigliadin antibody determination was run according to standard procedures.
Results: In controls, IgG-antigliadin antibody showed high variability in the 50th-90th centile range that peaked during the second year of life while IgA-antigliadin antibodies showed a lower variability and a less pronounced trend to decreasing values with age. A certain degree of overlapping between controls and coeliac patients was seen for both IgA- and IgG-antigliadin antibodies. The receiver operating characteristic analysis showed that the best discrimination was achieved by a cut-off of 8-10 ng/ml for IgA-antigliadin antibodies and 150-200 ng/ml for IgG-antigliadin antibodies.
Conclusions: Antigliadin antibody concentration is not normally distributed and changes with age in non-coeliac subjects. The receiver operating characteristic analysis is a valuable tool for fixing the antigliadin antibody cut-offs between control and diseased individuals. The diagnosis of coeliac disease should always be confirmed by intestinal biopsy.