OPTIMIZATION OF THE C-14 UREA BREATH TEST FOR THE DETECTION OF HELICOBACTER-PYLORI IN DYSPEPTIC PATIENTS

Helicobacter py(ori (Hp) has been associated to gastritis, peptic ulce r, gastric cancer, and other gastrointestinal disorders. The C-14-urea breath test (UBT) has been proposed as a simple and noninvasive metho d for its detection, and has recently been implemented in our institut ion. To optimize our resources, we performed a sensitivity analysis to determine the minimal sampling and duration of testing required, and to establish objective criteria for its interpretation. With this purp ose, endoscopy, antral biopsy and UBT were performed in 104 dyspeptic patients. For the UBT, a basal breath sample was taken before the admi nistration of 10 mu Ci of C-14-urea and followed by sequential breath sampling at 5, 15, 30 and 60 minutes. Considering histologic findings as the gold standard, receiver operating characteristic (ROC) curves w ere constructed for the following three (CO2)-C-14 excretion strategie s: excretion by sample, maximum excretion, and cumulative excretion. H p was detected in 74 (71%) of the patients, and its presence coincided with significantly higher (CO2)-C-14 excretion than in the Hp negativ e (p < 0.001). The three excretion strategies were comparable in terms of diagnostic accuracy, but the most efficient results were given by the 15 minute sample. With a cut-off point of greater than or equal to 1.7%, the sensitivity and specificity of this sample was greater than or equal to 83%, the positive and negative predictive values were 93% and 68%, and the accuracy was 84%. We conclude that UBT can be comple ted with a single breath sampling at 15 min, and its results objective ly interpreted as positive if the (CO2)-C-14 excretion is greater than or equal to 1.7%.