Background: Recent studies have shown that the prevalence of anti-endomysial antibodies (EMAs) in clinical practice is lower than expected; the aim of our study was therefore to compare the sorbitol H2-breath test (BT) with EMAs in the diagnosis of subclinical/silent coeliac disease and to compare with histologic lesions.
Methods: We studied 123 consecutive patients with subclinical (96) and silent (27) coeliac disease. Expiratory samples were collected before the patients drank the test solution (5 g of sorbitol in 150 ml of tap water) and every 30 min for 4 h. An increase in H2 concentration of at least 20 ppm over fasting baseline was considered positive for sorbitol malabsorption. EMAs were screened by the indirect immunofluorescence method.
Results: EMAs were positive in 77/96 (80.80%) and sorbitol H2-BT in 94/96 (97.91%) patients with subclinical coeliac disease, while EMAs were positive in 17/27 (62.96%) and sorbitol H2-BT in 26/27 (96.29%) patients with silent coeliac disease (P < 0.001 in both forms of coeliac disease). The best cut-off values in ppm and minutes are higher and shorter in the severe form than in the minor form of intestinal damage, respectively (P < 0.001 in both forms).
Conclusions: This study indicates that almost all subclinical/silent coeliac patients show abnormal sorbitol H2-BT and that there is a strict correlation between cut-off value (in ppm and minutes) and histologic lesions. In particular, the maximal cut-off value (in ppm and in minutes) correlates statistically with the more severe the grade of intestinal damage. Finally, the prevalence of EMA in subclinical/silent disease is lower than expected.