Objectives: Helicobacter pylori infection is arguably the most common chronic bacterial infection in humans. The high prevalence and the association with peptic ulceration and gastric cancer indicate that simple, noninvasive methods for diagnosis of the infection are needed. In this study, the accuracy of salivary diagnosis for H. pylori infection was assessed.
Methods: Saliva and serum samples of 152 dyspeptic patients were tested for H. pylori IgG and IgA by an in-house ELISA. All patients underwent gastroscopy with biopsy.
Results: One hundred thirty-one patients (86%) were found to be H. pylori positive on histology. Duodenal ulcer was found in 67 patients; 85 had no macroscopic lesion. Salivary and serum H. pylori IgG as well as serum H. pylori IgA titers were significantly higher in H. pylori-positive than in H. pylori-negative patients. The sensitivity and specificity of salivary H. pylori IgG were 82% and 71%, respectively; the positive and negative predictive values were 95% and 40%, respectively; and the accuracy 81%. The corresponding figures for serum H. pylori IgG were 97% and 91%; 98% and 83%; and 96%. Those for serum H. pylori IgA were 80% and 52%; 91% and 30%; and 76%. The sensitivity of salivary H. pylori IgG in detecting duodenal ulcer was 83% (56/67) that of serum H. pylori IgG was 97% (65/67) (odds ratio = 0.15; confidence interval = 0.02-0.8; p = 0.02).
Conclusions: Salivary H. pylori IgG was a fairly sensitive and accurate indicator of gastric H. pylori colonization, with a high positive predictive value in our population. Data, however, suggest that salivary H. pylori IgG measurements do not compare favorably with serology.