Turkey is a country with a very high prevalence of Helicobacter pylori infection and diagnosis of the infection via invasive or noninvasive tests is a major issue increasing the costs. The aim of this study was to assess the reliability of anti-H. pylori IgG ELISA serology, a simple and inexpensive method, for the primary diagnosis of H. pylori infection in dyspeptic patients. Serum samples of dyspeptic patients, who were referred to the Endoscopy Unit, Cerrahpaşa Medical School, İstanbul, between June 2003 and March 2004, were tested with a commercially available anti-H. pylori ELISA kit. Helicobacter pylori status of each patient was determined by histology, rapid urease test (RUT) and culture. According to our gold standard, a patient was classified as being H. pylori-positive if the culture and/or both histology, RUT were positive and as H. pylori-negative only if all of these tests remained negative. Patients with discordant results were excluded. Standard methods were used to calculate sensitivity, specificity, predictive values of positive and negative results and 95% confidence intervals of these values. Totally 152 patients were studied. According to our defined gold standard, 131 patients (86%) were H. pylori-positive and 21 patients (14%) were H. pylori-negative. The sensitivity, specificity and diagnostic accuracy of the IgG-ELISA were 97%, 67% and 93%, respectively. Due to its lack for specificity, it is concluded that anti- H. pylori ELISA test may not be a suitable alternative for the primary diagnosis of H. pylori infection in dyspepsia patients.