Aortic arch repair surgery is a technically complex and challenging procedure to treat aortic pathologies. Despite advancements in perioperative care, detrimental neurological complications occur during or after surgery. The neurological complications increase the economic burden of healthcare, morbidity and quality of life for the patients, even if they survive. Stroke, for example, leads to an increase in healthcare and social care costs, requiring a subsequent lengthy rehabilitation. Milder neurological impairments include transient ischaemic attacks, confusion and delirium, necessitating longer intensive care and hospital stay.

Currently applied cerebral monitoring modalities are electroencephalogram and cerebral oximetry. However, they are not specific enough to timely detect early cerebral ischaemia to prevent neurological complications. S100B protein and neuron-specific enolase are serum markers that reflect cerebral damage, however, their applicability in the hyperacute setting is limited. However, rapid measurements of glial fibrillary protein have paved new pathways to detect cerebral injury. Recent studies reveal more sensitive biomarkers of glucose, lactate, pyruvate, glutamate and glycerol. These biomarkers could potentially detect cerebral ischaemia on a near real-time basis using the microdialysis method. The aim of the project is to develop a bedside system for early detection of cerebral ischaemia on a near real-time basis during aortic arch surgery.

Early detection of cerebral ischaemia could mandate more aggressive cerebral protection strategies by further optimisation of hypothermia and antegrade selective cerebral perfusion during surgery, and optimisation of blood pressure and oxygenation in the intensive care unit. Ultimately, early detection of cerebral ischemia during surgery will prevent disabling and costly neurological complications following surgery.

Aortic arch surgery is used to treat life-threatening aortic dissections, aortic aneurysms involving the arch and other aortic pathologies. The procedure is performed under deep hypothermic circulatory arrest. Neurological complications are detrimental and disabling for the patients, even if they survive. In the 1990s, the risk of stroke after aortic arch surgery was 48% with deep hypothermic circulatory arrest. The risk of stroke has largely decreased to about 3% after elective surgery and 11% after emergency aortic arch repair in recent years, largely attributed to the development of current cerebral protection strategies. Despite this, the risk of stroke is considerably high and it accompanies devastating disabilities for the patients leading to longer hospital stays and lower quality of life for surviving patients.

Neurological complications are not without an economic burden on the NHS healthcare system. After analysing 84,184 patients admitted to hospital with a stroke, it is reported that the mean total healthcare and social care cost of a stroke patient in the UK is £22,429 at 1 year, and this amount increases to £46,039 in 5-year. The costs also increase with age and severity of the stroke (Figure 1).

Electroencephalogram (EEG) and cerebral oximetry are currently used modalities for cerebral ischaemia monitoring, yet, they fail to detect early cerebral ischaemia. EEG only processes data from superficial cortex areas only. It is also affected by anaesthetic agents, neuromuscular blocking agents and other intraoperative factors. Cerebral oximetry uses near-infrared spectroscopy (NIRS) to monitor cerebral regional oxygen saturation in the frontal cortex. It has a weak predictive value of neurological events, and no data exists to support the optimum threshold of oximetry. Moreover, there is no association between spectroscopy measurements and stroke. The latest emerging imaging modality for monitoring is transcranial doppler ultrasound. It is operator-dependent and images are usually acquired through a small transtemporal window. It measures velocity in cranial arteries such as middle cerebral arteries, and it is still yet to be validated with larger studies. Limitations in these monitoring modalities could often bring the adequacy of cerebral protection into question.

The brain is a very sensitive organ to hypoxemia, and mixed venous oxygen saturation can be used to estimate the oxygen delivery and usage in the body by measuring the blood returning to the right side of the heart. Mixed venous saturation is influenced by many factors, such as cardiac output, respiration and oxygenation, and tissue metabolism in the perioperative period. Intraoperatively, it could serve as a more specific marker of global oxygen delivery during the cardiopulmonary bypass since the pump flow and metabolic rates are fairly constant.

Alternatively, certain proteins, such as the S100B protein, neuron-specific enolase and glial fibrillary protein (GFAP) could be monitored. They are released into the cerebrospinal fluid (CSF) and blood after cerebral damage. Their concentrations correlate with the extent of cerebral injury and predict the adverse clinical outcomes. However, S100B and neuron-specific enolase cannot accurately diagnose stroke, even after 6 hours from the onset of clinical symptoms. Moreover, their measurement results could take at least 2 hours, and also reliability of the measurements decrease with the use of cardiopulmonary bypass. Hence, their applicability in the hyperacute setting of aortic surgery is very limited.

On the other hand, GFAP and ubiquitin carboxy-terminal hydrolase L1 (UCH-L1) have been described as emerging biomarkers for cerebral injury in an acute setting. The conventional measurement of GFAP using ELISA assays cannot reliably detect low levels of GFAP, especially in the blood. However, since the introduction of the portable point-of-care device to measure GFAP and UCH-L1 levels in the blood, these could be measured in as little as 10-15 minutes. This has paved the way for further exploring the role of GFAP as a biomarker for the early identification of cerebral ischaemia in aortic arch surgery.

Recent studies have evaluated the role of metabolic biomarkers; glucose, pyruvate, lactate, glutamate, and glycerol, in cardiac surgery, using the microdialysis method. Although conventional analysis of biomarkers is difficult on a near real-time basis, it is feasible using the microdialysis method. It has been validated and in clinical use for more than 30 years. The microdialysis system consists of a small catheter with a semipermeable membrane placed in the interstitial space and physiologic salt solution is constantly perfused in the catheter allowing some of the solutes to diffuse from the interstitial space into the microdialysis catheter along a concentration gradient. The dialysate can then be analysed for concentrations of biomarkers.

Choice of biomarkers (glucose, pyruvate, lactate, glutamate, and glycerol) is largely dependent on the commercially available analysis capacity of microdialysis method and previous studies on biomarkers in cardiac surgery. Glucose is the sole fuel of the brain. It is broken down in the glycolytic pathway to generate pyruvate, which is then used in mitochondria by oxidative metabolism via the tricarboxylic acid cycle. In the absence of oxygen, pyruvate is converted to lactate by the lactate dehydrogenase reaction for anaerobic glycolysis. An increase in the lactate-to-pyruvate ratio is shown to be a very sensitive biomarker for impending cerebral damage. Glutamate, which acts as the primary excitatory neurotransmitter in the brain, is released in excessive amounts following trauma, various brain pathologies and ischaemic events, such as stroke, leading to excitotoxic injury and neural cell death. Glycerol is known to be a sensitive marker for phospholipid cellular membrane degradation after ischaemic cell injury as well. Detection of biomarkers using the near real-time microdialysis method is a promising way of developing a better cerebral monitoring system during aortic surgery to prevent neurological complications.

Inclusion Criteria:

• Patients undergoing aortic arch surgery
• Patients aged 18 and over
• Patients who are able to provide informed consent

Exclusion Criteria:

• Patients with intraoperative death
• Patients with preoperative neurological dysfunction
• Patients who are not able to provide informed consent (language barrier, unconscious, unable to understand, retain and process information)

The study focuses on patients undergoing aortic arch surgery secondary to aortic pathologies such as aortic aneurysm and aortic dissection.