Journal Article

Searching for the optimal oxygen saturation range in acutely unwell patients

Beasley, Richard, Pilcher, Janine Marie, Kearns, Ciléin

2021

Overview

Correspondence to Professor Richard Beasley, Medical Research Institute of New Zealand, Wellington 6242, New Zealand; [e-mail] Recognition that there are risks associated with hyperoxaemia, as well as hypoxaemia, with the use of oxygen therapy in critically ill patients, has led to a paradigm shift in clinical practice, from the traditional approach of liberal administration to titration within a target oxygen saturation range.1 This paradigm shift, colloquially referred to as ‘swimming between the flags’,2 particularly applies to the use of oxygen in severe exacerbations of chronic obstructive pulmonary disease (COPD), in which guidelines now recommend that oxygen is titrated to a target oxygen saturation range of 88%–92%.2–4 This recommendation is based on the landmark randomised controlled trial in patients with acute exacerbations of COPD, in which oxygen therapy titrated to maintain oxygen saturations between 88% and 92% in the prehospital setting reduced the risk of mortality by 58%, when compared with high concentration oxygen therapy.5 While this randomised controlled trial provides the evidence base for the recommended 88%–92% target range, it raises the important question as to whether a ‘physiologically normal’ target of 93%–96%, that also avoids hyperoxaemia, may result in better outcomes than the 88%–92% target. In support of this higher target range, the nadir within the U-shaped risk of a serious adverse outcome with oxygen saturation in COPD extends from 88% to 96%, with insufficient delineation of risk within this range.6 The U-shaped association between oxygen saturations and risk of mortality in exacerbations of COPD has been examined further by Echevarria et al in a study published in this journal, and which provides data directly relevant to clinical practice.7 The strengths of their observational study are the power with 2645 consecutive admissions, data obtained on whether oxygen was administered and the ability to control for mortality risk through use of validated mortality prediction scores. External validity of this calculation is obtained from the number needed to cause one death of 14 calculated from the landmark randomised controlled trial of liberal versus titrated oxygen therapy in COPD.5 It may be that the period of grace, in which tolerance to excessive administration of high concentration oxygen to patients with COPD is accepted as entrenched behaviour that is understandable when treating a breathless distressed patient, should now be over.

• Publisher: BMJ Publishing Group Ltd and the British Association for Accident & Emergency Medicine,BMJ Publishing Group LTD
• Subject: Chronic obstructive pulmonary disease / Clinical medicine / Commentary / COPD / Emergency medical care / Medical personnel / Mortality / Oxygen saturation / Oxygen therapy / Patients / Physiology