Résumé : Objectives: We evaluated demographic characteristics in HIV-positive patients receiving highly active antiretroviral therapy (HAART) who had upper gastrointestinal (UGI) symptoms requiring UGI endoscopy and compared the findings in patients with and without H. Pylori coinfection. Methods: We prospectively observed all HIV-infected patients treated with antiretroviral therapy who underwent UGI endoscopy for the first time and were tested for H. pylori from January 2004 to December 2008. Data collected included the following: demographics (age, gender, ethnicity, body mass index [BMI], tobacco use, alcohol intake, and HIV risk behavior); comorbidity (viral hepatitis B or C, any organ dysfunction, or opportunistic disease); medication, including antibiotics, H2 blockers, proton pump inhibitors, and NSAIDs; CD4 cell counts, viral load; symptoms; and endoscopic and histologic diagnoses (H. pylori determined by Giemsa staining). Patients were compared according to H. pylori status (presence vs absence). Results: One hundred and forty-five patients were evaluated. Compared to patients without H. pylori infection (n=97), those with H. pylori infection (n=48) had a significantly higher CD4 cell count (p=.008), were more likely to be heterosexual (p=.047), had a higher BMI (p=.027), had a greater incidence of duodenal ulcers (p=.005), had lower viral loads (p<.01), were less likely to have received macrolide antibiotics in the last 3months (p=.00), and had less comorbidity (p=.03). They were also more frequently of Black African than Caucasians. In multivariate analysis, being heterosexual and having a low viral load were independently associated with an increased risk of having H. Pylori coinfection. Conclusion: In the antiretroviral therapy era, HIV-H. pylori coinfection is associated with a greater incidence of duodenal ulcers and higher CD4 counts, higher BMI, less comorbidity, and less frequent use of macrolides. © 2012 Blackwell Publishing Ltd.