Discrepancy between Helicobacter pylori stool antigen assay and urea breath test in the detection of Helicobacter pylori infection

BACKGROUND

The reference diagnostic methods available for detection of Helicobacter pylori infection are either invasive (histology) or expensive and highly sophisticated (Urea Breath Test). A new enzyme immunoassay, which can be easily performed in any laboratory, has been developed to detect Helicobacter pylori in stool specimens (HpSA-Meridian Diagnostics, Cincinnati, USA). Aim of the study was to compare HpSA to Urea Breath Test.

PATIENTS AND METHODS

A total of 125 patients (52 never treated for Helicobacter pylori infection and 73 after Helicobacter pylori eradication therapy) referring to our Department, underwent both tests within two weeks.

RESULTS

Contrasting results between the two tests were found in 30% of cases: in 19% of the untreated patients and in 37% of the treated patients (p<0.001). The main discrepancy consisted in positive HpSA associated with negative Urea Breath Test. Mean HpSA value in such conditions was 0.273 optical density, while in patients with both positive tests, it was 1.192 optical density. In untreated, but not in treated patients, raising the HpSA cut off value significantly decreased the percentage of conflicting results.

CONCLUSIONS

Some disagreement was detected between HpSA and Urea Breath Test results, especially in treated patients. Possible explanations for our findings are a low HpSA cut off value together with the identification of Helicobacter pylori coccoid forms by the immunoassay but not by the urease based Urea Breath Test. The higher percentage of discrepancy detected in treated patients might support this hypothesis.

Candida overgrowth in gastric juice of peptic ulcer subjects on short- and long-term treatment with H2-receptor antagonists

Candida overgrowth in gastric juice of peptic ulcer subjects under therapy with H2-receptor (H2-R) antagonists has been detected in 21.4 and 53.8% of cases after short- and long-term treatment respectively, and in 8% of controls. Both types of H2-R antagonists, ranitidine and cimetidine, were equally associated with production of yeasts. The location of ulcers, whether gastric or duodenal, seems to have no influence on fungal growth. Females were more susceptible than males to develop Candida in gastric juice. In the short-term course with H2-R blockers fungal colonization of gastric juice was associated with delay of the rate of ulcer healing. Fungal detection in gastric juice was not associated with mucosal invasion by Candida since in none of the patients who had a biopsy for gastric ulcer was Candida detected by histology.