Cerebral Oximetry in Preterm Infants-To Use or Not to Use, That Is the Question

Abstract

The Safeguarding the Brains of our smallest Children (SafeBoosC) project was initially established to test the patient-relevant benefits and harms of cerebral oximetry in extremely preterm infants in the setting of a randomized clinical trial. Extremely preterm infants constitute a small group of patients with a high risk of death or survival with brain injury and subsequent neurodevelopmental disability. Several cerebral oximeters are approved for clinical use, but the use of additional equipment may disturb and thereby possibly harm these vulnerable, immature patients. Thus, the mission statement of the consortium is "do not disturb-unless necessary." There may also be more tangible risks such as skin breakdown, displacement of tubes and catheters due to more complicated nursing care, and mismanagement of cerebral oxygenation as a physiological variable. Other monitoring modalities have relevance for reducing the risk of hypoxic-ischemic brain injury occurring during acute illness and have found their place in routine clinical care without evidence from randomized clinical trials. In this manuscript, we discuss cerebral oximetry, pulse oximetry, non-invasive electric cardiometry, and invasive monitoring of blood pressure. We discuss the reliability of the measurements, the pathophysiological rationale behind the clinical use, the evidence of benefit and harms, and the costs. By examining similarities and differences, we aim to provide our perspective on the use or non-use of cerebral oximetry in newborn infants during intensive care.

Keywords: brain; clinical benefit; monitoring; near-infrared spectroscopy; oximetry; randomized trial.

Figure 1

Simple mechanistic model of drivers of instant local tissue pO2 (colored in blue) i.e., the final common path to oxidative phosphorylation, oxygen consumption, local aerobic metabolism and organ function and health. The key concept in the model is that systemic vascular conductance (the inverse of resistance) is the sum of all local vascular conductances (in the figure it is labelled “adding up to systemic vascular resistance”). The figure is not an acyclical graph, since there is feedback at several levels. This makes questions of causation as well as the prediction of the effects of manipulation of variables more complex. Here we developed the model merely as an illustration of our view of the four physiological variables (labelled in yellow) for which the monitoring modalities discussed in this essay are more or less close surrogates. The figure was drawn in the Dagitty software for drawing causal graphs. The code can be found at http://dagitty.net/mQNlp4r and modified as needed for your purpose.