Risk factors for upper gastrointestinal bleeding in intensive care unit patients: role of helicobacter pylori. Federal Hyperimmune Immunoglobulin Therapy Study Group

Abstract

OBJECTIVE: To determine the role of preexisting Helicobacter pylori infection in the development of acute upper gastrointestinal (GI) hemorrhage in intensive care unit (ICU) patients in relation to other potential predisposing risk factors. DESIGN: Prospective, multicenter, cohort study. SETTING: Medical and surgical ICUs in six tertiary care Department of Veterans Affairs Medical Centers. PATIENTS: Eight-hundred seventy-four patients without previous GI bleeding or peptic ulcer disease who were enrolled in a multicenter, randomized, controlled trial of prophylactic intravenous immunoglobulin to prevent ICU-associated infections. INTERVENTIONS: This substudy of the larger intravenous immunoglobulin study only involved data analysis and had no intervention. All patients were enrolled in the larger study where they received intravenous immunoglobulin or placebo as intervention. MEASUREMENTS AND MAIN RESULTS: Patients were prospectively evaluated for the development of acute upper GI hemorrhage while in an ICU. Anti-H. pylori immunoglobulin G and immnoglobulin A concentrations were determined by enzyme immunoassay on preintervention serum samples. Seventy-six (9%) patients had over upper GI bleeding and a mortality rate of 49%, as compared with a 15% mortality rate in patients who did not bleed (p < .001). By logistic regression analysis, the following factors were associated with an increased risk of bleeding: acute hepatic failure, prolonged duration of nasogastric tube placement, alcoholism, and an increased serum concentration of anti-H. pylori immunoglobulin A. CONCLUSIONS: Increased anti-H. pylori immunoglobulin A concentrations, prolonged nasogastric intubation, alcoholism, and acute hepatic failure were found to be independently correlated with the development of acute GI bleeding in an ICU setting. These observations should be prospectively confirmed in an independent population before being used for treatment guidelines.