A semiquantitative stool antigen test for Helicobacter pylori: shedding new light?

Abstract

The past 2 decades have witnessed a rapid evolution in our understanding the important role played by Helicobacter pylori in human gastrointestinal disease. Current guidelines recommend testing for H. pylori in the initial evaluation of young patients with dyspepsia without “red flag” symptoms. 1 The presence of H. pylori can be confirmed in gastric mucosal biopsy samples by culture, special histology, or rapid urease tests, but these approaches require the expense and inconvenience of endoscopy. As a result, accurate yet noninvasive alternatives have been sought. Helicobacter pylori serology is used widely in primary care, but it is unreliable in low- and high-prevalence populations and cannot be used to monitor eradication therapy. Accordingly and appropriately, its popularity appears to be waning. Until recently, the only available, nonendoscopic means of detecting active H. pylori infection was the urea breath test (UBT). 13C and 14C-urea breath tests are more accurate than is serology and, when available, have quickly become the tests of choice. Indeed, UBT has enjoyed a relative monopoly as a noninvasive test for active infection. However, this status ended recently, when the U.S. Food and Drug Administration approved a novel enzyme immunoassay for H. pylori antigen excreted in the stool. The principal limitation of fecal H. pylori assays is obvious and relates to patients' general reluctance to collect stool samples. Despite this, however, stool H. pylori antigen tests show great promise. Their accuracy for the initial diagnosis of H. pylori infection is high, similar to both UBT and endoscopic biopsy. A recent, comprehensive meta-analysis reported sensitivity, specificity, and positive and negative predictive values in excess of 90% for the detection of primary infection. 2 Importantly, stool antigen tests can be used to monitor eradication therapy, 3–5 although false-positive results may occur in the first 1 to 4 weeks with the shedding of dead organisms. Stool antigen assays also have several distinct and significant advantages over UBT, their principal competitor for noninvasive diagnosis: (1) they are simple to administer; (2) their costs in many jurisdictions are considerably lower 3,4,6; and (3) they require that laboratories acquire little in the way of new equipment. Importantly, stool specimens can be obtained with relative ease from patients who are unlikely to cooperate for endoscopy or breath tests. This feature makes stool H. pylori antigen assays particularly appealing for testing children, a subgroup in whom their diagnostic accuracy has been confirmed in several published series. 6–8 In this issue of the Journal, Ierardi et al. 9 from Italy explore another potential advantage of H. pylori stool antigen tests: their role as a semiquantitative measure of intragastric bacterial load. The authors used known titers of stool-mixed H. pylori to develop a saturation curve and convert stool H. pylori spectrophotometric absorbance in patient samples to estimates of bacterial load. Several conceptual issues arise immediately when considering this approach, including how the results would be affected by variation in stool weight and/or intestinal transit. However, the authors show favorable correlation between their semiquantitative stool assays and the UBT delta value, a marker generally considered to reflect intragastric bacterial load. Of note, these observations have been confirmed recently, with similar results reported by Chang et al. 10 Critics would argue that treatment decisions depend on the presence or absence of H. pylori, not its density. So why would clinicians seek to measure bacterial load? Increased bacterial load has been found to predict failure of eradication therapy and may offer a rationale for a more aggressive or prolonged antimicrobial regimen. 11–13 Higher densities of H. pylori have also been associated with more severe histologic gastritis by some investigators. 10 However, others have failed to confirm this relationship, 14 and it remains unclear whether knowledge of density should influence the clinical decision to treat. Increased H. pylori load has been suggested recently to predict the severity of dyspepsia, 15 but such a relationship would be only of academic interest if the infection were to be eradicated regardless. Other evidence of the importance of H. pylori density in the management of H. pylori-associated disease is surprisingly sparse. In summary, stool antigen tests for H. pylori have arrived on the scene and offer a simple, reliable, and valid alternative to UBT for the diagnosis of active H. pylori infection. These assays will undoubtedly become and remain a part of our diagnostic armamentarium. However, their prominence in primary care practice is limited by their relative acceptability to patients, particularly as point-of-care UBT becomes more widely available. The interpretation of stool antigen tests as semiquantitative measures of gastric H. pylori load offers an intriguing twist to this saga, but more reflection on the relevance of this insight to clinical care is needed.