AETIOLOGY

The exact cause of CPB related neurologic injury remains unclear, the main debate being: is it due to CPB- related hypoperfusion or microvasculature embolisation? If hypoperfusion is responsible, increasing flow rates during CPB will help. If embolisation is responsible, increasing flow rates will deliver a greater embolic load to the brain. Clearly there is a need to identify the cause and develop interventions to minimize the clinical effect, the challenge being that the brain is not homogenous and there is no single test that can quantify cerebral injury.

CEREBRAL PHYSIOLOGY OVERVIEW

Of note: Cerebral Haemodynamics

1. The awake, normothermic brain receives 14% of cardiac output.
2. During alpha- stat controlled, hypothermic CPB, the brain only receives 5- 7% of cardiac output.
3. Hypercapnia causes cerebral vasodilatation, while hypocapnia causes cerebral vasoconstriction.
4. Cerebral blood flow (CBF) increases at the onset of CPB, due to the decreased blood viscosity.

The management of PaCO ₂ during hypothermic CPB is divided into pH- stat management where a temperature corrected PaCO ₂ is maintained at 40mmHg by addition of CO ₂ , and alpha- stat measurement, where the non- temperature corrected PaCO ₂ is maintained at 40mmHg.

• At 28 ^o C, the PaCO ₂ difference between the two methods is 15mmHg.
• Cerebral perfusion pressure autoregulation occurs over the range of a PaCO ₂ of 20- 100mmHg during alpha- stat management, but during pH- stat management, autoregulation is lost. Therefore changes in cerebral perfusion pressure will result in direct changes in cerebral blood flow.
• During deep hypothermia, perfusion pressure autoregulation doesn’t occur, regardless of the method of PaCO ₂ management.

Cerebral Metabolism

1. The most important factor determining CBF is cerebral metabolic demand (assessed by cerebral oxygen consumption, i.e. CMRO ₂ ), a concept known as “flow- metabolism coupling”. A regional increase in metabolic demand and therefore blood flow occurs where a particular area of brain is active. This “flow- metabolism coupling” is lost with pH- stat management of PaCO ₂ .
2. Hypothermia homogenizes cerebral metabolism and blood flow, so that there is little difference between central areas, or between white and grey matter. Cerebral metabolism decreases by 5 –7% for each 1 ^o C drop in temperature.
3. Changes in metabolic rate exceed changes in CBF during hypothermia, therefore the CBF to CMRO ₂ ratio is increased –a phenomenon termed “luxury perfusion”. However, on rewarming, the cerebral oxygen supply may be insufficient to meet the increased oxygen demands due to a sudden increase in cerebral metabolism.
4. Warm blood cardioplegia and normothermic CPB are thought to confer myocardial protection, however, this conflicts with views about neuroprotection from hypothermia. Is it more important to protect the patient’s brain or heart?

PATHOPHYSIOLOGICAL CHARACTERISTICS OF CPB

• All artificial blood pumps are associated with trauma to blood components, which can be reduced by reducing flow rates. Normal flow rates are 3.2l/m2, but currently accepted flow rates are 2.0- 2.4l/m2. Hypothermia helps reduce the mismatch between metabolism and flow.
• Most CPB pumps produce nonpulsatile perfusion. It is uncertain whether the pulse of pulse- generating systems is transmitted to the cerebral arterioles. Nonpulsatile flow causes the release of vasoconstrictors and increased vascular resistance. The effect on the cerebral circulation is unclear.
• Blood contact with synthetic materials initiates activation of coagulation, fibrinolytic and complement pathways, causing a systemic inflammatory response, which may lead to microemboli of cellular aggregates. Emboli may consist of air, debris from disrupted atherosclerotic plaques, calcium, tube fragments, glove powder, silicone antifoam, chylomicrons, remnants of damaged cells or lipid from the surgical field. The number of emboli detected intraoperatively has been associated with postoperative NP defects.

MONITORING AND OUTCOME MEASURES

The ideal neurophysiological monitor should:

• provide noninvasive, continuous, objective, rapid assessment of cerebral perfusion, oxygenation and activity
• be portable, compact, reliable and easy to use
• produce accurate and reproducible results

Early Methods

• Kelly and Schmidt first assessed CBF and CMRO ₂ using 15% inhaled nitrous oxide as a diffusible tracer, the arterial and jugular concentrations of which were measured. Later, labeled krypton was used. However, these methods need a sample time of >10 minutes.
• 133Xe use decreased sample times and allowed measurement of regional and total CBF. This technique was refined and used to define cerebrovascular responses to changes in physiological conditions, e.g. PaCO ₂ . Drawbacks of this method are: inconvenience, expense, and use of radioactive tracers and bulky equipment. This method is confined to research.

Retinal Fluorescein Angiography (RFA)

• Allows visualisation of retinal microcirculation and detection of microembolic events during CPB.
• Fluorescein dye injections enhance the retinal microvasculature, which is then photographed and analysed by digital computer. One study using this technique showed microembolic perfusion defects in 100% of the patients on a bubble oxygenator during CPB, compared with 44% of patients on a membrane oxygenator. Another study related the number of occlusions with NP deficit. Yet another study, in dogs, identified the emboli as platelet- fibrin aggregates.
• Drawbacks include: skill is required to operate the camera, interruption of surgery with camera repositioning and suboptimal image analysis prone to subjective bias.

INTRAOPERATIVE PATIENT MONITORING

Transcranial Doppler (TCD)

• Measures velocity of CBF and can detect, but not identify emboli.
• Traditionally a probe is placed over temporal bone and the sound waves focussed on the middle cerebral artery by process called “insonation”. Placing the probe over a common carotid artery reduces the problems of insonation and studies a greater proportion of the CBF.
• Successfully used to detect emboli, but can only detect relative CBF changes, and not absolute CBF.
• Provides real-time feedback about procedures which result in embolisation, thereby allowing for refinement of surgical technique and a decreased embolic rate.
• Continuous, noninvasive, inexpensive, portable technique without radioactive material.
• Drawbacks include: flow measurements being affected by many factors e.g. bone thickness, probe susceptibility to movement, brain swelling during CPB affecting insonation, background “noise”, requirement of skill.

Aortic Scanning

• Related to neurological outcome by detection of atherosclerosis before instrumentation of the aorta (60% of emboli occur during manipulation of aorta and heart).
• Ultrasonic detection of atherosclerotic plaque in the aorta by handheld probe or tranoesophageal echocardiography(TEE) probe.
• Handheld probe is noninvasive, portable and results are reproducible.
• Drawbacks include: invasiveness of TEE, poor visualisation of distal ascending aorta which is the favoured area for cross clamping AND the area most affected by atherosclerosis!
• Awaiting prospective, randomised trials to test efficacy in minimizing emboli.

Jugular Venous Bulb Oxygen Saturation

• Percutaneous cannulation of the internal jugular vein, so that the tip of the catheter sits in the jugular bulb.
• Intermittent, or continuous, fibreoptic measurement of the cerebral venous oxyhaemoglobin saturation (SjVO ₂ ) allows monitoring for an imbalance between the CBF and CMRO ₂ , i.e. flow- metabolism uncoupling.
• Croughwell et al, used this technique with 133Xe clearance to show desaturation of SjVO ₂ during rewarming on CPB, and defined the critical level of SjVO ₂ desaturation as <50% or a jugular bulb venous oxygen tension of 25mmHg. A later study correlated this desaturation with NP deficits post surgery. Simple and continuous method of global monitoring. Drawbacks include: invasiveness, contamination of sample by blood from extracerebral sources entering the jugular vein, suboptimal fibreoptic oxygenation analysis.

Electroencephalogram (EEG)

• A sensitive and specific indicator of cerebral hypoperfusion and subsequent cerebral ischaemia during carotid endarterectomy.
• Unfortunately changes indicating impending cerebral ischaemia may be caused by hypothermia, anaesthetic agents, haemodilution and changes in PaCO ₂ .
• Hypothermia is the main determinant of the EEG during CPB; the EEG is isoelectric at 18 ^o C.
• Complicated by presence of electrical “noise”.
• Use in cardiac surgery restricted to determining electrical silence before circulatory arrest and detection of major cerebral perfusion defects.

Near Infrared Spectroscopy (NIRS)

• Near infrared light is capable of passing through up to 8cm of tissue, including skin, soft tissue and bone. It is absorbed by oxyhaemoglobin, deoxyhaemoblobin and cytochrome oxidase, and the reflected light intensity gives an indication of the concentration of these substances. Thus intravascular and intracellular oxygenation changes can be measured.
• NIRS assesses oxyhaemoglobin levels in arterial, venous and capillary circulation. The cerebral circulation is 70% venous; therefore NIRS estimates venous cerebral oxygenation predominantly. -NIRS has been used extensively in carotid endarterectomy, but has not been well documented during CPB, various studies have produced differing results in terms of accuracy of NIRS in assessing cerebral venous oxygenation, but usually it follows the trend of the SjVO ₂ . In some cases the cytochrome oxidase oxygenation readings may be decreased despite normal SjVO ₂ levels, which has led to the hypothesis that arteriovenous shunting may occur in the cerebral circulation under certain conditions. This means that there may be cerebral ischaemia despite normal jugular venous saturation levels.
• The significance of extracerebral haemoglobin and the optimal positional relationship of transmitter to probe needs to be established.

Biochemical Markers

• Ideal marker should be organ specific, quantitative, predictable and its kinetics understood.
• Problems include the following:
• the brain is not homogenous in cellular make-up
• site and not size of injury determines the degree of functional impairment
• the blood brain barrier also complicates matters.
• Current research is focussing on S- 100 proteins and NSE. High levels of both are found in all patients after CPB.

Genetic Markers

Apolipoprotein E epsilon-4 allele is associated with cognitive decline in elderly patients with Alzheimer’s disease. Patients with this marker are predisposed to post CPB cognitive impairment.

POSTOPERATIVE NEUROLOGICAL EXAMINATION

• Traditionally this identifies and locates the site of a neurological lesion without giving a quantitative assessment of the injury, e.g. diffuse frontal lobe ischaemia may go unnoticed, while a discrete brainstem lesion will be picked up.
• Problems encountered with examining patients post CPB include patients being too ill or limited by symptoms to co-operate, post operative depression and mood / attention altering drugs.
• The examination should include assessment of the mental state, cranial nerves, motor/ sensory/cerebellar examination, gait, deep tendon and primitive reflexes.
• A thorough visual examination is mandatory and visual abnormalities may influence performance in other tests.
• “Stroke scores” exist to assess changes in neurological status.

NEUROPHYSIOLOGIC EXAMINATION

• This involves the identification and quantification of cognitive impairment, indicative of brain dysfunction.
• An exhaustive battery of tests post cardiac surgery is not possible; therefore a limited number of tests is used to compare pre- and post- surgical cognitive function.
• There is a high incidence of short and long term cognitive impairment after CPB.
• Guidelines for these tests have been drawn up.
• Examples of these tests are the Trail Making, the Digit Symbol and Grooved Peg Board tests.
• Testing should be delayed until 5 days after surgery, to minimize the effects of anaesthesia, fatigue and pain.
• Stump et al, defined a decline of at least 20% from baseline in at least 20% of measures as significant.