Carbon urea breath test is not as accurate as endoscopy to detect Helicobacter pylori after gastrectomy

BACKGROUND: This study was conducted to determine (1) whether Helicobacter pylori infection decreases in conjunction with time elapsed after gastrectomy and (2) the diagnostic efficacy of (13)C urea breath test (UBT) for H pylori in patients after gastrectomy. METHODS: From January 1997 to June 1998, 86 patients who had undergone gastrectomy and 180 patients with dyspepsia without gastrectomy were enrolled. A UBT for the analysis of excess (13)CO(2)/(12)CO(2) ratio (ECR) was obtained for each patient. Each patient also underwent endoscopy to obtain gastric biopsies for histology and H pylori culture. The presence of H pylori by either histology or culture served as the standard to test the efficacy of UBT. The 86 patients with a prior gastrectomy were categorized into 3 subgroups (I, less than 1 year; II, 1 to 3 years; III, greater than 3 years), according to the interval between surgery and UBT. The initial H pylori status of these 86 patients was determined by histologic evaluation of the resected stomach. RESULTS: At trial initiation, the postgastrectomy group had a lower H pylori infection rate (52.3%) as compared with the dyspeptic control group (80%). The initial H pylori status among subgroups I, II, and III was similar. There was a trend for the presence of H pylori in the stomach to decrease with increasing time elapsed after surgery (I to III: 68.8%, 48.3%, 36%, respectively; p < 0.05). The maximum UBT sensitivity and specificity achieved were 82.2% and 87.8% in the gastrectomy group and 97.2% and 96.3% in the dyspeptic group, with cutoff points of 2.5 and 4.0, respectively. CONCLUSION: The prevalence of H pylori diminishes with time elapsed after gastrectomy. UBT for detection of H pylori is more effective in patients without prior gastrectomy than in patients who have undergone gastrectomy and is less effective than endoscopy for patients who have had a gastrectomy.