Objective: The H(2)-sorbitol breath test (H(2)-SBT) has previously been suggested as a screening tool for coeliac disease. We developed an alternative (13)C-sorbitol breath test ((13)C-SBT). The aim of the study was to compare the diagnostic properties of the H(2)-SBT and the (13)C-SBT in a clinical setting.
Material and methods: Thirty-nine coeliac patients, 40 patient controls (mainly patients with irritable bowel syndrome) and 26 healthy volunteers underwent the breath tests. The patients were given an oral load of 5 g sorbitol and 100 mg (13)C-sorbitol dissolved in 250 ml tap-water. H(2), CH(4) and (13)CO(2) concentrations were measured in end-expiratory breath samples every 30 min for 4 h. Increased H(2) concentration > or =20 ppm from basal values was used as the cut-off for the H(2)-SBT.
Results: The H(2)-SBT had a sensitivity of 71%, a specificity of 46% versus healthy controls, and a specificity of 25% versus patient controls. Individuals with methane-producing intestinal flora had significantly lower peak H(2) concentrations than non-methane producers. The (13)C-SBT reached maximal combined sensitivity/specificity (74%/85%) for both control groups after 1 h. A diagnostic algorithm which stratified patients into high-, moderate- and low risk for coeliac disease was proposed. Following the algorithm, 62% of coeliac patients were detected with 100% specificity. The (13)C-SBT, but not the H(2)-SBT, correlated with age and serum IgA tissue-transglutaminase antibody levels in coeliac patients.
Conclusions: The novel (13)C-SBT has superior diagnostic properties compared to the H(2)-SBT, which has unsatisfactory specificity in clinical practice. The 1-h (13)C-SBT may be a useful supplemental test when investigating for coeliac disease.