Recently a new, noninvasive diagnostic test based on the detection of Helicobacter pylori stool antigen (SA) has been developed. The aim of this study was to systematically review the experience on H. pylori SA test for the diagnosis of H. pylori infection. Bibliographic searches were performed in the PubMed database and abstracts from several congresses. A total of 43 studies fulfilled the inclusion criteria and evaluated H. pylori SA test accuracy for the diagnosis of H. pylori infection in nontreated patients. Overall, 4769 patients were included. The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) (weighted mean) were, respectively: 92.4% (95% CI = 91-93%), 91.9% (91-92%), 92.1% (91-93%), and 90.5% (90-91%). Therefore the SA test can definitively be considered an accurate noninvasive method for the diagnosis of H. pylori infection in untreated patients. A total of 25 studies including 2078 patients evaluated H. pylori SA test for the confirmation of H. pylori eradication > or = 4 wk after completion of therapy. Sensitivity, specificity, PPV and NPV (weighted mean) were: 88.3% (87-90%), 92% (91-93%), 75.1% (73-77%), and 94.8% (94-96%). Although most studies showed that SA test is an accurate method to confirm H. pylori eradication > or = 4 wk after treatment, these favorable results were not confirmed in other studies. Further investigation is necessary to explain these discrepancies, as well as to clarify the precise time for confirmation of eradication after therapy, the appropriate cutoff point for the SA test, and which factors influence it. Proton pump inhibitors seem to affect the accuracy of SA test, but the negative effect disappears 1-2 wk after stopping treatment. The SA test is technically feasible in patients with upper GI bleeding. although the true diagnostic accuracy in this group of patients remains to be more fully assessed. Finally, the SA test seems to be a highly cost-effective method for the diagnosis of H. pylori infection.