A fundamental principle of specific antimicrobial therapy is accurate diagnosis. Numerous validated methods to diagnose patients with H. pylori infection are in use. These methods can be divided into invasive and noninvasive diagnostic tests.

The invasive tests include endoscopy followed by gastric biopsy and histologic demonstration of organisms, biopsy with direct detection of urease activity in the tissue specimen, and biopsy with culture of the H. pylori organism. Although culturing the organism is traditionally considered the "gold standard" for diagnosis of many infectious agents, it is the least sensitive diagnostic test (approximately 70-80 percent positivity). Both histologic demonstration of the organism by Giemsa or Warthin-Starry stains and urease testing have sensitivities and specificities above 90 percent.

Excellent diagnostic sensitivities and specificities (less than 95 percent) are also obtained with noninvasive tests for the initial diagnosis of H. pylori infection. These include serology for immunoglobulin G antibodies to H. pylori antigens and breath tests of urease activity using orally administered ¹⁴C- or ¹³C-labeled urea. A number of highly accurate serologic kits for diagnosis of H. pylori infection are available. Labeled urea breath tests have had restricted availability as research tools in the past, but commercial assays will be available in the near future.

It is important to note that, with the exception of the serologic assays, all the tests for diagnosis of H. pylori infection may be falsely negative in patients who have taken antibiotics, bismuth compounds, or omeprazole in the recent past.

Presently, there is no readily available, inexpensive, and accurate noninvasive method to monitor eradication of H. pylori. Without such an assay, routine monitoring for relapse, reinfection, or treatment failure cannot be recommended. Even if such a test were available, testing all patients treated for H. pylori infection would probably not be necessary in view of the high efficacy of treatment and low reinfection rate. Important exceptions would be patients with complicated, recurrent, or refractory peptic ulcers who should be evaluated for successful eradication of infection before cessation of antiulcer therapy. Antibody levels decrease slowly following successful eradication of H. pylori infection. If the same well-standardized assay is used, a dramatic fall in antibody titer 6-12 months following antimicrobial treatment indicates successful eradication. However, variability among serology tests applied in commercial laboratories may limit their usefulness in confirming H. pylori eradication. Although breath testing is the best noninvasive assay for evaluating success of eradication, there are unresolved issues of availability, cost, and ease of use in the practical application of this method. Invasive tests can also be used for documenting cure, but these incur the cost and morbidity associated with endoscopy.

Therapy of H. pylori poses several unique challenges. The organism resides under a mucus gel layer in the highly acidic milieu of the stomach, where rapid removal of ingested antimicrobials may occur. These and other factors may contribute to the variable correlation between in vitro and in vivo antimicrobial activity. A problem in selection of a therapeutic regimen has been the lack of a suitable animal model. For these reasons, much of the available information concerning choice of antimicrobial agents is based on small empirical trials in humans. Multiple agents that have been studied in various combinations include metronidazole, tetracycline, amoxicillin, clarithromycin, bismuth compounds, H₂-receptor antagonists, and proton-pump inhibitors. The choice of a particular regimen must be tempered by the rapidly developing data on optimal therapy.

Guidelines for the routine antimicrobial treatment of H. pylori infection