Diagnosis of Helicobacter pylori in gastric brush and biopsy specimens stained by Romanowsky and immunocytochemical methods: comparison with the CLOtest.

Helicobacter pylori has been implicated in the pathogenesis of chronic gastritis, gastric and duodenal ulcer, and possibly gastric carcinoma. The organism may be detected by invasive or non-invasive methods with variable sensitivity. Paired gastric biopsy and gastric brush specimens were collected from 83 patients presenting with non-ulcer dyspepsia. One biopsy was tested for urease using the CLOtest, the other was processed to paraffin and consecutive sections were stained with haematoxylin and eosin, modified Giemsa and anti-H. pylori antisera. The brush specimens were stained with a rapid Romanowsky stain (Hema-Gurr) and anti-H. pylori. The CLOtest was positive in 31 cases, the Giemsa biopsy in 25, the anti-H. pylori biopsy in 27, the Hema-Gurr smear in 27 and the anti-H. pylori smear in 19. The sensitivities of the methods after omitting one inadequate biopsy were 96%, 93%, 100%, 96% and 78%, respectively. The specificities were 93% for the CLOtest and 100% for the other methods. While immunocytochemical staining of gastric biopsies may be the most sensitive method for H. pylori identification, the cost and turn around time of the technique may preclude its routine use. Gastric brush cytology is a highly sensitive and specific method for H. pylori detection that is quick and simple to perform. Its application is recommended for the routine diagnosis of H. pylori infection.