Helicobacter gastritis induces changes in the oxyntic mucosa indistinguishable from the effects of proton pump inhibitors.

A causal relationship between oxyntic glands dilatation with protruding parietal cells, referred to as proton pump inhibitor (PPI) effects, and PPI use has been suspected but not established. We designed this study to evaluate the association between these changes and the use of PPIs and histamine2-receptor blockers (H2-blockers). We obtained five Sydney System-compliant biopsy specimens from patients recruited into a therapeutic trial for H. pylori. Medication history with details on PPI and H2-blockers use was collected. Two blinded pathologists graded gastritis and the intensity of putative PPI effects using a 0 to 3 scale. PPI and H2-blocker use was then disclosed and the accuracy of pathologists' assessment was analyzed. There were 138 H. pylori-negative and 104 positive patients. In H. pylori-negative patients the histologic assessment for PPI use had 77.5% sensitivity and 51.8% specificity, with a positive predictive value of 86.9% and a negative predictive value of 35.9%. In H. pylori-positive patients, sensitivity was 74.1% and specificity 26.1%. Positive and negative predictive values were 55.8% and 44.4%, respectively. Neither glandular dilatations nor parietal cell protrusions related to H2-blocker use. We conclude that these changes are associated with PPI use only in H. pylori-negative subjects. In H. pylori gastritis, so-called PPI-effects were equally prevalent in PPI-users and non-users, indicating that other factors are involved in the induction of oxyntic cell hyperplasia. We suggest that comments regarding the supposed evidence of PPI use are too often wrong to be useful and should be avoided in the diagnosis of gastric biopsy specimens.