Respiratory failure & ventilation
Risk Factors for Patient-Important Upper Gastrointestinal Bleeding
Authors' conclusion
Several factors are associated with the risk of patient-important upper gastrointestinal bleeding during invasive ventilation.
Study context
- Population
- Invasively ventilated adults in the ICU
- Participants
- 4,821 participants
- Comparison
- Baseline and time-varying risk factors (preceding 3 days) for patient-important upper GI bleeding
- Primary endpoint
- Patient-important upper gastrointestinal bleeding in the ICU
Methodological detailspreplanned regression analysis, competing risk of death, time-varying risk factors
- Preplanned regression analysis
- Competing risk of death
- Time-varying risk factors
Primary result
As reported in the abstract
Patient-important upper GI bleeding: incidence and risk factors
- Incidence of patient-important upper GI bleeding in the ICU
- 131 of 4,821 (2.7%)
- Baseline APACHE II score, per 5-point increase: HR
- 1.24 (1.12, 1.37) (95% CI)
- Inotropes or vasopressors: HR
- 2.05 (1.35, 3.12) (95% CI)
- Severe thrombocytopenia (platelet count <50 × 109/L): HR
- 2.21 (1.24, 3.94) (95% CI)
- Platelet inhibitor drugs: HR
- 1.69 (1.11, 2.56) (95% CI)
- Pantoprazole: HR
- 0.36 (0.25, 0.54) (95% CI)
- Enteral nutrition, per 500 ml/d increase: HR
- 0.81 (0.68, 0.97) (95% CI)
Further findings
Interaction between enteral nutrition and pantoprazole
- No interaction between enteral nutrition and pantoprazole
- Interaction P = 0.94
Pantoprazole by volume of enteral nutrition (subgroup)
- Pantoprazole, enteral nutrition 500 ml/d: HR
- 0.36 (0.22, 0.58) (95% CI)
- Pantoprazole, no enteral nutrition: HR
- 0.36 (0.18, 0.72) (95% CI)
Enteral nutrition by pantoprazole use (subgroup)
- Enteral nutrition with pantoprazole: HR
- 0.82 (0.63, 1.07) (95% CI)
- Enteral nutrition without pantoprazole: HR
- 0.81 (0.66, 1.00) (95% CI)
Original publication
Risk Factors for Patient-Important Upper Gastrointestinal Bleeding. Am J Respir Crit Care Med. Published 1 September 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →