This European multicentre trial randomised ICU patients to receive 40 mg of IV pantoprazole or placebo daily in 3298 patients following admission for an acute event. The authors reported no differences in at least one clinically important event (a composite of clinically important GI bleeding, pneumonia, Clostridium difficile infection, or myocardial ischaemia) in the treated group compared to placebo (21.9% vs. 22.6%, respectively). However, 2.5% of patients had clinically important GI bleeding compared with 4.2% in the placebo group, but at 90 days there was no difference in mortality (31.1% vs. 30.4% of the patients), the primary outcome of the study.
The absence of differences in outcomes between pantoprazole and placebo except for GI bleeding is not surprising. Although fewer patients in the pantoprazole group had clinically important GI bleeding, the authors did not correct for multiple comparisons of secondary outcomes potentially limiting the findings. This report is similar to a prior meta-analysis, but different from observational studies suggesting an increased risk of infectious complications and myocardial ischaemia in association with proton-pump inhibitors.
Fortunately, the New England Journal of Medicine embraces large, well-conducted clinical studies such as the one thoughtfully performed by the investigators, despite prior reported multiple smaller studies or observational data that may have shown equipoise or adverse effects of certain therapies due to their higher use in critically ill patients.
– Jerrold H. Levy