Prospective identification of Helicobacter pylori in routine gastric biopsies without reflex ancillary stains is cost-efficient for our health care system.

Despite the recommendation of expert gastrointestinal pathologists, private and academic centers (including our own) have continued to use ancillary stains for identification of Helicobacter pylori. For a 1-month period, gastric biopsies were prospectively evaluated for H pylori using routine hematoxylin and eosin (H&E) and a reflex Diff-Quik stain. During this time, 379 gastric biopsies were collected on 326 patients. H pylori organisms were prospectively identified in 23 (7%) patients, all of whom had superficial dense lymphoplasmacytic inflammation expanding the lamina propria. An additional 2 patients with neutrophilic inflammation were found to have H pylori by immunohistochemical staining. One patient diagnosed as having normal gastric mucosa was retrospectively found to have inflammation with rare H pylori organisms originally overlooked on both H&E and Diff-Quik but later identified on immunostain (0.5%). No patients with chemical gastritis (16%) or chronic inflammation (27%) were found to have H pylori. During the study month, 9 immunostains for H pylori were performed in addition to the 379 Diff-Quik. After discontinuation of reflex Diff-Quik, approximately 20 immunostains are performed for H pylori each month, which decreases technical time spent for processing gastric biopsies and reduces cost to the health care system. In our population with a low prevalence of H pylori, reflex staining for organisms is not cost-effective. The organisms can be seen on routine H&E; when suspicious superficial or active inflammation is present without visible organisms, immunohistochemical stains will confirm presence or absence within a day. Discontinuation of up-front ancillary studies is cost-effective without compromising patient care.