Gastric mucosal density of Helicobacter pylori estimated by real-time PCR compared with results of urea breath test and histological grading.

The accuracy of the urea breath test (UBT) and histological grading for estimation of the density of Helicobacter pylori in gastric mucosa is not known. Real-time (TaqMan) PCR was used to estimate the total number of H. pylori genomes in biopsy samples. These values were compared with those obtained by the UBT and the histological grade obtained by the Sydney system. The UBT and endoscopy with antral and corporal biopsies were performed in 88 consecutive untreated patients with dyspepsia. Bacterial culture and the rapid urease test were done with fresh biopsy materials. TaqMan PCR and histological examination were done on serial paraffin sections of the biopsy samples. Of the five methods tested, TaqMan PCR had the highest sensitivity and specificity (both 100%) in the diagnosis of H. pylori infection. The mean density of H. pylori genomes for pairs of biopsy samples from individual patients was compared with the individual values obtained by the UBT; correlation between the results was significant. The density of H. pylori genomes was higher in histological grades 1, 2 and 3 than in grade 0, without significant differences between adjacent grades from 1 to 3. These results suggest that the severity of H. pylori infection of the stomach can be estimated by the UBT and that histopathologists might state whether the organism is present or absent, rather than making a quantitative statement as recommended in the Sydney system.