Mortality and morbidity in acutely ill adults treated with liberal versus conservative oxygen therapy (IOTA): a systematic review and meta-analysis. (Lancet 4/2018) (By Contributor Dr. Daniel J. Green)
Background: Most patients that are in the hospital will have a nasal cannula on whether they need it or not. How do we know who needs supplemental oxygen therapy?
METHODS: Systematic review and meta-analysis of randomized control trials in a variety of different databases comparing liberal versus conservative oxygen therapy in acutely ill adult inpatients (pts with chronic respiratory disease and a few other things like ECMO were excluded). The outcome of interest was mortality (in hospital, 30-day, longest follow-up reported) and morbidity (hospital-acquired pneumonia, any infection, length of hospital stay, others).
Findings: Twenty-five RCTs including ~16,000 patients with diseases such as sepsis, critical illness, stroke, trauma, myocardial infarction, cardiac arrest, and emergency surgery. They compared conservative oxygen strategy (‘less is more’) verses liberal oxygen strategy (O2 sat median greater than 96%). More oxygen had a relative risk for in-hospital death of RR 1.2, 30-day mortality relative risk RR 1.14, all were statistically significant, and the I-square test for heterogeneity was low (favorable, meaning the studies were similar and easy to compare).
Critical Appraisal: Broadly speaking, internally valid with high quality of evidence, low risk bias with low heterogeneity among numerous large randomized control trials. Reasonably well generalized as the patient's included had a variety of different medical conditions. Our veteran population has a high prevalence of chronic respiratory disease, I could not easily find what this term includes. If they excluded stable COPD, asthma patients etc., it would limit the generalization to our VA population.
Take Home: There is small but persistent risk of death, both short and long-term, from supplemental oxygen therapy titration to a saturation greater than ~96%. Consider removing the nasal cannula to prevent death, prevent your patient from sitting in bed all day unnecessarily, and prevent dry nasal passages and discomfort. But mostly death.
Link to Paper Here