Helicobacter pylori detection: a quality and cost analysis.

Histopathologic interpretation of hematoxylin and eosin (H&E)-stained endoscopic biopsies is a common method for identifying Helicobacter pylori. Few studies report the accuracy of this method, and none have compared costs of other diagnostic methods. In the clinical setting of a community hospital using standard diagnostic techniques, the purpose of this study were to determine 1) the comparative sensitivities and specificities of the H&E stain, the Warthin-Starry silver stain, the Giemsa stain, and the CLOtest; 2) the sensitivity and specificity of an "experienced" pathologist in identifying H. pylori by H&E stains, compared with a rotating pathology faculty; and 3) the time to diagnosis (turnaround time) and current patient charges for each diagnostic method. Bacterial identification by the silver stain (or a combination of other tests which were likely to compensate for false-positive and false-negative silver stains) were used as the diagnostic standard in evaluating 94 consecutive cases with the following results: The H&E stain interpreted by the rotating pathology staff was the least sensitive method and one of the least specific tests that were studied. The silver and Giemsa stains were equally sensitive in identifying H. pylori; the silver stain was more specific. The CLOtest was less sensitive than the silver and Giemsa stains, but was equally specific. CLOtest was similar in sensitivity to the H&E stain examined by the "experienced" pathologist, but was more specific. An experienced pathologist was significantly more sensitive than the rotating pathologists in evaluating H&E-stained slides. Therefore, if H&E stains are used to identify H. pylori, which is a common practice, it may be advantageous to use an experienced pathologist. The CLOtest was a simple, rapid, and cost effective substitute for H&E stains in the identification of H. pylori.