RESUMO
OBJECTIVE: To determine the prevalence of Helicobacter pylori (H. pylori) in critically ill patients who develop upper gastrointestinal bleeding after cardiac surgery in relation to other risk factors. DESIGN: Prospective, single center, cohort study. SETTING: Surgical intensive care unit in a university hospital. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Over a 1-yr period, all consecutive patients with upper gastrointestinal hemorrhage from the stomach or duodenum were studied for H. pylori infection by serology. Additionally, the need for mechanical ventilation over 48 hrs, the duration cardiopulmonary bypass, and the aortic cross-clamp time were analyzed. For control, 229 patients with no evidence of gastrointestinal hemorrhage after cardiac surgery were studied. All patients received stress ulcer prophylaxis with ranitidine. Operations were performed on 2,956 patients during the study period. The incidence of upper gastrointestinal bleeding was 0.9%. Twenty (77%) of the 26 patients with upper gastrointestinal bleeding and 145 (63%) patients of the control group had serologic evidence for H. pylori infection (odds ratio, 1.9; 95% confidence interval 0.7-5.0; p = .2). Patients who required prolonged mechanical ventilation had a significantly greater risk for upper gastrointestinal bleeding (odds ratio, 22.1; 95% confidence interval 8.6-56.7; p<.001). Patients with upper gastrointestinal bleeding also had a significantly longer duration of cardiopulmonary bypass and aortic cross-clamp time (p<.001) CONCLUSIONS: H. pylori is not associated with upper gastrointestinal bleeding in critically ill patients who receive stress ulcer prophylaxis, whereas patients who require prolonged mechanical ventilation are at high risk. A prophylactic eradication of H. pylori is not justified.