Effects of Conservative Versus Liberal Oxygen Therapy on the Outcomes of Critically Ill Patients: A Systematic Review and Meta-Analysis

Abstract

Background: The liberal use of supplemental oxygen therapy is common in intensive care units (ICUs) for critically ill patients, but this practice still lacks high-quality evidence, especially for patients receiving mechanical ventilation (MV). We systematically reviewed the effects of conservative versus liberal oxygen therapy in critically ill adults. Methods: We searched the following electronic databases, including dates from Jan 1st, 2001 until Nov 31st, 2018: PubMed/MEDLINE, Embase, the Cochrane (Central) database and Ovid. Randomized controlled clinical trials (RCT), before-and-after trials or cohort studies comparing conservative with liberal oxygen therapy in critically ill adults (aged ≥18 years) were retrieved. The main outcomes were mortality and ventilator-free days, length of stay, and adverse effects assessed by random-effects meta-analyses. We used the GRADEpro Guideline Development Tool and Trial sequential analysis (TSA) to evaluate the quality of evidence and information size, respectively. Meta-regression was conducted to explore heterogeneity among studies. Findings: Five RCTs and 2 before-and-after trials enrolled 16188 ICU patients. Compared with liberal oxygen therapy, a conservative oxygen strategy (maintaining peripheral oxygen [SpO2] ≤ 98% or arterial oxygen partial pressure [PaO2] ≤ 100 mmHg at the lowest possible FiO2) decreased hospital mortality (RR = 0.88; 95% CI, 0.81-0.95; p = 0.002, moderate quality), especially for patients receiving MV (N = 12327; RR = 0.91; 95% CI, 0.84-1.00; p = 0.05, moderate quality). Moreover, it reduced hospital length of stay (Std. MD = -0.04; 95% CI, -0.06- -0.01; p = 0.01, moderate quality) and increased ventilator-free days either at day 28 (Std. MD = 0.04; 95% CI, 0.01-0.08; p = 0.005, low quality) or during ICU stay (Std. MD = 0.49; 95% CI, 0.30-0.68; p < 0.00001, low quality). The incidence of new bacteremia was much lower in the conservative oxygen therapy group (RR = 0.53; 95% CI, 0.30-0.92; p = 0.02, very low quality). TSA indicated that no more clinical studies should be conducted to prove a beneficial effect of conservative oxygen therapy on hospital length of stay rather than ICU length of stay for critically ill patients. Interpretation: In critically ill adults, conservative oxygen therapy improves hospital mortality, decreases hospital length of stay and increases ventilator-free days, without impacting overall adverse effects. More clinical trials with better designs are needed to optimize the strategy of conservative oxygen therapy in ICU. Funding Statement: This work was supported by the National Natural Science Foundations of China (No. 81571874, 81670074, 81870066), the Key Research and Development Plan of Jiangsu Province (BE2018743), the Postgraduate Research & Practice Innovation Program of Jiangsu Province (KYCX17_0168), Natural Science Foundation of Jiangsu Province (BK20171271), the Jiangsu Provincial Medical Youth Talent (QNRC2016807) and Third Level Talents of the “333 High Level Talents Training Project” in the fifth phase in Jiangsu (LGY2016051). Declaration of Interests: None declared.