Pantoprazole in ICU patients at risk for gastrointestinal bleeding - one-year mortality in the SUP-ICU trial.

Marker, Søren; Krag, Mette; Perner, Anders; Wetterslev, Jørn; Lange, Theis; Wise, Matt P; Borthwick, Mark; Bendel, Stepani; Keus, Frederik; Guttormsen, Anne Berit; Schefold, Joerg C; Rasmussen, Bodil S; Elkmann, Thomas; Bestle, Morten; Arenkiel, Bjørn; Laake, Jon H; Kamper, Maj K; Lång, Maarit; Pawlowicz-Dworzanska, Malgorzata B; Karlsson, Sari; Liisanantti, Janne; Dey, Nilanjan; Knudsen, Heidi; Granholm, Anders; Møller, Morten Hylander

BACKGROUND The long-term effects of stress ulcer prophylaxis with pantoprazole are unknown in ICU patients. We report one-year mortality outcome in the Stress Ulcer Prophylaxis in the Intensive Care Unit (SUP-ICU) trial. METHODS In the SUP-ICU trial, acutely admitted adult ICU patients at risk of gastrointestinal bleeding were randomised to intravenous pantoprazole 40 mg vs. placebo (saline) once daily during their ICU stay. We assessed mortality at one year and did sensitivity analyses according to the trial protocol and statistical analysis plan. RESULTS A total of 3261 of the 3291 patients with available data (99.1%) were followed up at one year after randomisation; 1635 were allocated to pantoprazole and 1626 to placebo. At one year after randomisation, 610 of 1635 patients (37.3%) had died in the pantoprazole group as compared with 601 of 1626 (37.0%) in the placebo group (relative risk, 1.01; 95% confidence interval 0.92-1.10). The results were consistent in the sensitivity analysis adjusted for baseline risk factors and in those of the per-protocol population. We did not observe heterogeneity in the effect of pantoprazole versus placebo on one-year mortality in the predefined subgroups, i.e. patients with and without shock, mechanical ventilation, liver disease, coagulopathy, high disease severity (SAPS II > 53) or in medical vs. surgical ICU patients. CONCLUSION We did not observe a difference in one-year mortality among acutely admitted adult ICU patients with risk factors for gastrointestinal bleeding allocated to stress ulcer prophylaxis with pantoprazole or placebo during the ICU stay. This article is protected by copyright. All rights reserved.