Objective: The management of hypoxemia in critically ill patients is challenging. Whilst the harms of tissue hypoxia are well recognized, the possibility of harm from excess oxygen administration, or other interventions targeted at mitigating hypoxemia, may be inadequately appreciated. The benefits of attempting to fully reverse arterial hypoxemia may be outweighed by the harms associated with high concentrations of supplemental oxygen and invasive mechanical ventilation strategies. We propose two novel related strategies for the management of hypoxemia in critically ill patients. First, we describe precise control of arterial oxygenation involving the specific targeting of arterial partial pressure of oxygen or arterial hemoglobin oxygen saturation to individualized target values, with the avoidance of significant variation from these levels. The aim of precise control of arterial oxygenation is to avoid the harms associated with inadvertent hyperoxia or hypoxia through careful and precise control of arterial oxygen levels. Secondly, we describe permissive hypoxemia: the acceptance of levels of arterial oxygenation lower than is conventionally tolerated in patients. The aim of permissive hypoxemia is to minimize the possible harms caused by restoration of normoxemia while avoiding tissue hypoxia. This review sets out to discuss the strengths and limitations of precise control of arterial oxygenation and permissive hypoxemia as candidate management strategies in hypoxemic critically ill patients.
Design: We searched PubMed for references to “permissive hypoxemia/hypoxaemia” and “precise control of arterial oxygenation” as well as reference to “profound hypoxemia/hypoxaemia/hypoxia,” “severe hypoxemia/hypoxaemia/hypoxia.” We searched personal reference libraries in the areas of critical illness and high altitude physiology and medicine. We also identified large clinical studies in patients with critical illness characterized by hypoxemia such as acute respiratory distress syndrome.
Subjects: Studies were selected that explored the physiology of hypoxemia in healthy volunteers or critically ill patients.
Setting: The data were subjectively assessed and combined to generate the narrative.
Results: Inadequate tissue oxygenation and excessive oxygen administration can be detrimental to outcome but safety thresholds lack definition in critically ill patients. Precise control of arterial oxygenation provides a rational approach to the management of arterial oxygenation that reflects recent clinical developments in other settings. Permissive hypoxemia is a concept that is untested clinically and requires robust investigation prior to consideration of implementation. Both strategies will require accurate monitoring of oxygen administration and arterial oxygenation. Effective, reliable measurement of tissue oxygenation along with the use of selected biomarkers to identify suitable candidates and monitor harm will aid the development of permissive hypoxemia as viable clinical strategy.
Conclusions: Implementation of precise control of arterial oxygenation may avoid the harms associated with excessive and inadequate oxygenation. However, at present there is no direct evidence to support the immediate implementation of permissive hypoxemia and a comprehensive evaluation of its value in critically ill patients should be a high research priority.
1 UCL Centre for Altitude, Space and Extreme Environment Medicine, Portex Unit, Institute of Child Health, London, United Kingdom.
2 Intensive Care Unit and University College London Division of Surgery and Interventional Science, Royal Free Hospital, London, United Kingdom.
3 Integrative Physiology and Critical Illness Group, Clinical and Experimental Sciences, Sir Henry Wellcome Laboratories, University of Southampton, University Hospital Southampton NHS Foundation Trust, Southampton, United Kingdom.
4 Anaesthesia and Critical Care Research Unit, University Hospital Southampton NHS Foundation Trust, Southampton, United Kingdom.
*See also p. 664.
The authors have not disclosed any potential conflicts of interest.
For information regarding this article, E-mail: firstname.lastname@example.org