Associated Mortality Risk of Proton Pump Inhibitor Therapy for the Prevention of Stress Ulceration in Intensive Care Unit Patients: A Systematic Review and Meta-analysis.

Abstract

The aim was to systematically evaluate risks and benefits of proton pump inhibitor (PPI) use for stress ulcer prophylaxis in the critically ill patient. Whether PPIs increase mortality in the critically ill patient remains controversial. Systematic review and meta-analysis of randomized controlled trials (RCTs) and cohort studies with trial sequential analysis, Bayesian sensitivity analysis, and fragility index analysis. A total of 31 studies in 78,009 critically ill adults receiving PPIs versus any comparator were included. PPI use was associated with an increased mortality risk in all studies [19.6% PPI vs. 17.5% comparator; RR: 1.10; 95% confidence interval (CI): 1.02-1.20; P=0.01], in the subgroup of RCTs (19.4% vs. 18.7%; RR: 1.05; 95% CI: 1.0-1.09, P=0.04), but not cohort studies (19.9% vs. 16.7%; RR: 1.12; 95% CI: 0.98-1.28, P=0.09). Results were maintained with a Bayesian sensitivity analysis (RR: 1.13; 95% credible interval: 1.035-1.227) and a fragility index analysis, but not sequential analysis (P=0.16). RCTs with a higher baseline severity of illness revealed the greatest mortality risk with PPI use (32.1% PPI vs. 29.4% comparator; RR: 1.09; 95% CI: 1.04-1.14; P<0.001). PPI use reduced clinically important bleeding in RCTs (1.4% PPI vs. 2.1% comparator; RR: 0.67; 95% CI: 0.5-0.9; P=0.009) but increased bleeding in cohort studies (2.7% PPI vs. 1.2% comparator; RR: 2.05; 95% CI: 1.2-3.52; P=0.009). PPI use was not associated with a lower incidence of clinically important bleeding when compared with histamine-2 receptor antagonists (1.3% vs. 1.9%; RR: 0.59; 95% CI: 0.28-1.25, P=0.09). This meta-analysis demonstrated an association between PPI use and an increased risk of mortality.