5. Cerebral Oximetry

A reliable non-invasive monitor of cerebral oxygenation would render these less important, on the general assumption that if the brain is being perfused with oxygenated blood then other systems will also function. There are such non-invasive devices available, and it remains somewhat surprising that they have not found their way into mainstream anaesthetic practice.

Near infrared spectroscopy (NIRS) is based on physical principles similar to those which underlie pulse oximetry, and it can be used in the same way to measure the oxygen saturation of peripheral tissues. However, its particular value lies in its ability to provide a non-invasive measure not only of cerebral oxygenation but also of other deeper structures that are usually inaccessible, such as the spinal cord and the splanchnic circulation.

A ‘niroscope’ is essentially an optical spectrophotometer that uses wavelengths in the near infrared range, from 700 to 1,100 nanometres (nm). This utilizes the fact that tissues, including those of the bony cranium, allow penetration of light at these wavelengths to depths as great as 8 centimetres. Visible light, in contrast, with wavelengths that range from 450 to 700 nm, s able to penetrate little more than 1 cm of tissue due to much greater attenuation by absorption, reflectance and scattering.

The instruments use laser diodes to generate infrared (IR) light and employ reflectance-mode spectroscopy. The light is emitted and received by optodes (the term used in this context for optical sensors). The transmitter and receiver are on the same side but are separated by around 5 cm. This is because the pathway of IR light from the emitter, through the tissue, and back to the detector is elliptical; and its maximal penetration is proportional to the separation between the two. This is variously quoted as being one-third to one-half of that distance.

The emitted IR light is absorbed by chromophores, of which oxygenated haemoglobin (HbO2) and deoxygenated haemoglobin (Hb) are two such species with absorption spectra of 800–850 nm and 650–800 nm, respectively. Others include cytochrome P450 enzymes, myoglobin (mainly in the frontalis muscle) and bilirubin (of significance only in jaundiced patients). Each different chromophore has a specific absorption, or extinction, coefficient which is expressed as a function of wavelength. The selected wavelengths generally are between 700 and 850 nm, which is the range within which the absorption spectra of HbO2 and Hb show maximal separation.

Cerebral oxygen saturation is determined by application of the Beer-Lambert law, which put at its simplest states that the concentration of a substance can be measured by the amount of light that it absorbs.

The available devices can determine changes in oxygenated and deoxygenated haemoglobin but are unable to distinguish between arterial and venous concentrations. The distribution of haemoglobin in cortical tissue is approximately 70% in the venous system and 30% in arterial blood ‘normal’ rSO2 is not 96% or greater as would be expected from a peripheral oximeter probe, but is between 55% and 80%, and so trend monitoring is of more value than absolute values below 50%, or a fall of 20% from baseline suggests cortical hypoxaemia

One potential difficulty is that in non-metabolizing or dead brain, HbO2 may be sequestered in veins and capillaries which will lead to misleadingly normal rSO2 values. This is more likely to be a problem when cortical oxygenation is being assessed after brain injury than during surgery when a falling rSO2 will indicate hypoperfusion or hypoxia before cell death has occurred.

Much of the work on cerebral oxygenation has related to cardiac surgery, but a systematic review that was published in 2014 looked at changes in regional (cerebral) oxygen saturation (rSO2) in non-cardiac surgery and found significant falls in onelung ventilation during thoracic surgery, in major abdominal surgery, in carotid endarterectomy with arterial cross-clamping, in hip surgery and in laparoscopic surgery performed in the reverse Trendelenberg (head-up) position. The review also noted ‘pronounced cerebral desaturation during shoulder arthroscopy in the beach chair position’ (H.B. Nielsen. Systematic review of near infrared determined cerebral oxygenation during non-cardiac surgery. March 2014,

The clinical importance of maintaining cerebral oxygenation is self-evident, and the studies so far done would seem to illustrate two important realities. The first is that silent cerebral cortical desaturation can occur without any obvious perturbations of routine systemic vital signs such as heart rate, arterial blood pressure and peripheral oxygen saturation. The second is that such episodes of desaturation do appear to correlate with the development of postoperative cognitive dysfunction

Given that routine monitoring may be falsely reassuring, it is difficult to know exactly how specifically to manage anaesthesia during those higher-risk procedures. But if the beach chair position for shoulder surgery is taken as an example, there should be a high index of suspicion for any factors that might potentially lead to cerebral hypoperfusion. Hence, in addition to maintaining cardiac output, volaemic status, normocapnia and an adequate haemoglobin concentration to ensure oxygen delivery, the anaesthetist should perhaps tolerate only minimal divergence from the patient’s baseline levels. (These can readily be determined in the anaesthetic room in the sitting position prior to the induction of anaesthesia.)