MORE CONVENIENT C-13-UREA BREATH TEST MODIFICATIONS STILL MEET THE CRITERIA FOR VALID DIAGNOSIS OF HELICOBACTER-PYLORI INFECTION

Aims and methods C-13-Urea breath tests (UBT) were performed in 217 pa tients with epigastric symptoms to assess (a) the validity of shorter intervals of breath sample collection and (b) of the non-fasting state on diagnosing H. pylori infection and to evaluate (c) a more simple s ampling method. (a) In 186 patients, breath samples were taken before and at 10, 20, 30, 40, 50, 60, 90, and 120 minutes after ingestion of 75 mg C-13-urea. C-13-enrichment was measured by isotope ratio mass sp ectrometry [delta parts per thousand] and the cumulative recovery rate (CRR) was calculated. (b) 31 patients performed C-13-UBTs on two occa sions, i.e. fasted and in the postprandial state collecting breath bef ore and 30 min after substrate ingestion. (c) 61 breath samples were o btained in duplicate, i.e. both with an aluminium foil breath bag usin g evacuated glass tubes, and by exhalation into open vials via straw. Results (a) DOB-responses at 10, 20,30, and 40 minutes after dose show ed a linear correlation with the 120 min-CRR. Referring to a CRR > 3% at 120 min (criterion of H. pylori-infection), DOB-responses at 20 min with a cut off-value at 5 parts per thousand (ROC-analysis) were show n to be valid for diagnosing H. pylori status in man (sensitivity 99.0 %, specificity 100%). (b) Postprandially, mean DOB-responses of H.pylo ri positive patients were lower than in the fasting state (17.46 parts per thousand vs. 27.63 parts per thousand; p < 0.01). With the cut-of f at 5 parts per thousand, however, sensitivity still was 94.4 % and s pecificity was 1 00 %. (c) The straw and breath bag method gave equiva lent results (r = 0.98, p < 0.0001). Conclusions (a) A two sample meas urement using the DOB at 20 min and a cutoff value of 5 parts per thou sand can replace the CRR-calculation over 120 minutes. (b) Although qu antitative DOB-responses are significantly affected, C-13-UBT can be p erformed in the postprandial state with no significant effect on the d etection of the H. pylori infection. (c) The sample collection by a si mple straw method meets precision requirements < 1.5 delta parts per t housand. Thus, the C-13-UBT can be performed cheaper, faster and more conveniently.