• Although the brain comprises only 2% of the body's mass, it accounts for ~20% of the body's metabolic requirements and receives 14% of the cardiac output.
• Regional differences of cerebral blood flow (CBF) exist and reflect the metabolic requirements (cerebral metabolic rate of oxygen, CMRO₂). CBF in:
  • Gray matter is 80 mL/100 g/min (CMRO₂ of 6 mL/100 g/min).
  • White matter is 20 mL/100 g/min (CMRO₂ of 2 mL/100 g/min).
  • Global average is 50 mL/100 g/min (CMRO₂ of 3.2 mL/100 g/min).
• Demographics and CBF. In addition, CBF is higher in females than males and also decreases with age.

• The brain requires a constant supply of a substrate and oxidizer in order to meet the high metabolic requirements.
  • Glucose is the main metabolic precursor with a consumption of ~5 mg/100 g/min; ~90% is metabolized aerobically. When unavailable, secondary metabolic precursors include acetoacetate and beta-hydroxybutyrate.
  • There is no adenosine triphosphate (ATP) storage mechanism in the brain. Thus, in the event that the oxygen supply is lost, ATP levels will fall to zero within 7 minutes.
  • 40% of the brain's energy is used for basal function (membrane potential, "housekeeping" activities) and the remainder 60% is used for functional activity (signal transmission).
• Cerebral oxygen extraction is high compared to the systemic circulation. Hemoglobin saturations measured at the jugular vein (S_jO₂) range from 45–60%, compared to normal mixed venous saturations (S_vO₂) of 55–70%.
• Regional flow. CBF lacks global control. Instead, areas of the brain act in an independent fashion to control flow via heterogeneous and homeostatic control mechanisms. Homeostatic mechanisms of CBF control include metabolism, temperature, metabolic products, PaCO₂, PaO₂, blood viscosity, and autoregulation.
  • Metabolism. Blood flow-metabolism coupling is the most important mediator of the cerebral circulation and is maintained during sleep and general anesthesia. The CMRO₂ varies regionally in the brain. Under normal circumstances, its main determinants are sleep or wakefulness, temperature, age, and degree of neuronal activity (e.g., seizures).
  • Hypothermia reduces CMRO₂ (and therefore CBF) by 5–7% for every degree Celsius drop in body temperature (functional and basal energy requirements). The Q₁₀ for CBF in humans is 2 (represents the factor by which the reaction rate decreases when the temperature is decreased by ten degrees; it is a unitless quantity). Complete suppression of metabolic activity occurs at 18–20°C.
  • Metabolic products. Mediators of cerebral vasodilation and flow-metabolism coupling include glutamate (indirectly by causing release of vasodilators, a feed-forward mechanism), arachidonate derivatives, nitric oxide (NO), prostaglandin E₂ (PGE₂) adenosine, potassium, vasoactive intestinal peptide, lactate, and carbon dioxide (CO₂).
  • PaCO₂. CO₂ crosses the blood–brain barrier (BBB) and changes the pH of the CSF. CBF increases linearly by 2–4% (1–2 mL/100 g/min) per 1 mm Hg increase in PaCO₂ (when the PaCO₂ is in the range of 25–75 mm Hg). The change begins within seconds, and fully equilibrates within 2 minutes. Conditions that cause a reduction in CO₂ reactivity include severe carotid artery stenosis, head injury, subarachnoid hemorrhage, hypotension, cardiac failure, female gender, and age (>40 years). Acids and protons themselves cannot cross the BBB (1,2) [A].
  • PaO₂. A PaO₂ <60 mm Hg will increase CBF and takes 6 minutes to equilibrate (2) [A].
  • Viscosity. Blood viscosity is primarily determined by hematocrit and temperature. As viscosity decreases, the CBF increases due to changes in blood rheology. The effect that this has on the microvascular circulation is not consistent and depends on local factors. Therefore, the generalization that a lower hematocrit improves microcirculatory flow cannot be made. Additionally, a lower viscosity may increase intracranial pressure (ICP).
  • CBF autoregulation. Under normal conditions, there is minimal change in the CBF with mean arterial pressure (MAP) changes between 60 and 160 mm Hg. Both arterioles (smooth muscle in vessel wall) and capillaries (pericytes that encircle the capillary) dilate or constrict in response to perfusion pressure, in order to maintain a constant blood flow. However, larger arteries and the venous system do not play a role in dynamic flow changes. It should be noted that there is significant individual variation in the range of autoregulation, based on a person's baseline BP.
• Global and regional blood flow can be measured by nitrous oxide washout and Xenon CT scanning/Xe clearance. Neither is done routinely.

• Generally, branches of the common carotid arteries supply the cerebrum and branches of the vertebral arteries supply the cerebellum.
• All areas supplied by the carotid arteries have good collateral circulation except for the middle cerebral artery. This area is most prone to ischemia.
• The classical Circle of Willis for collateral circulation is present in only 50% of people.
• The superior sympathetic ganglion provides sympathetic stimulation to the cerebral vasculature, and plays a role in cerebral vasospasm following injury and stroke.

• The fixed volume of the intracranial vault consists of brain tissue (80%), blood (12%), and CSF (8%). ICPs above 30 mm Hg generally produce pathological changes in CBF (with 15 mm Hg being defined as an elevated ICP). CBF at this point becomes strictly flow dependent (entirely reliant on the MAP being greater than the ICP) (3) [A].
• Cushing's response: A physiologic phenomenon resulting from increased ICP and subsequent cerebral ischemia. It consists of
  • Hypertension (to increase the cerebral perfusion pressure, or CPP)
  • Bradycardia (an inappropriate reflex to the hypertension)
  • Irregular respirations (from pressure on the brain stem); not always observable
• Cerebral autoregulation is impaired by
  • Trauma or traumatic brain injury
  • Hypoxemia
  • Hypercapnia (PaCO₂ >60 mm Hg)
  • High-dose volatile anesthetics
  • Subarachnoid hemorrhage
  • Ischemic cerebrovascular disease
• Ischemia. Because of regional differences, CBF in one area does not correlate with blood flow in another ischemia-prone area.
  • Normal CPP is 80–100 mm Hg in adults.
  • CPP <50 mm Hg manifests as EEG slowing.
  • CPP <25–40 mm Hg will produce a flat EEG.
  • CBF <15–20 mL/100 g/min shows an isoelectric EEG signal.
  • CBF <5 mL/100 g/min produces irreversible damage (1) [A].
• Moyamoya: Creation of a web of collateral vessels to deeper brain structures as a result of longstanding stenosis/malformation of the internal carotid arteries. Pathology of the blood vessels shows intimal hyperplasia and elastic fiber formation with sparing of the media and adventitia. This is likely from an elevation of basic fibroblast growth factor. The Moya-Moya regions have decreased responsiveness to CO₂ and are dependent on CPP alone to avoid underperfusion (ischemia) and overperfusion (rupture). Brain areas around the stenosed vessels show a decreased CMRO₂ and regional blood flow with an increased oxygen extraction ratio (signs of inadequate blood supply). There is a bimodal age of presentation in the 1st and 4th decades of life. Pediatric patients commonly present with signs of ischemia (~80%), whereas adults predominantly present with a rupture (40–65%, usually basal ganglia). Symptoms may be elicited by maneuvers that lower the PaCO₂, causing vasoconstriction and focal ischemia (coughing, crying, straining, hyperventilation). During anesthesia, it is vitally important to preserve blood flow to areas with a tenuous supply. This is achieved with preoperative volume restitution (preadmittance with IV fluids), avoidance of cerebral vasoconstriction (avoid crying and hyperventilation before induction), intraoperative EEG with or without cerebral oximetry, frequent blood gases to avoid both hypocapnea (watershed ischemia), and hypercapnea (steal of blood from the Moya-Moya regions by dilated cerebral vessels). Efforts should be made to minimize intraoperative CMRO₂ (avoid pain intraoperatively with adequate opiates/anesthesia and avoid hyperthermia). Steal may also occur with the return to normocarbia after a period of hypocarbia and the EEG may take time to normalize (3,4)[A and C].
• Carotid stenosis/carotid endarterectomy (CEA). Stenosis can impair CBF and perfusion pressures, resulting in syncopal episodes, ischemia, and stroke; repair is undertaken in order to restore flow. Intraoperatively, the MAP (and therefore the CPP) can change rapidly with increases and decreases in pressures, catecholamine release from manipulation of the carotid sinus, and bradycardia from vagus nerve clamping. Postoperatively, hyperperfusion syndrome may result; its etiology is not clear, but is related to postoperative hypertension to previously underperfused areas and usually does not occur until several hours after surgery.
• Ruptured cerebral aneurysms/subarachnoid hemorrhage. Inadequate regional blood flow or ischemia is experienced in the area of the aneurysm. Extravascular hemoglobin can cause cerebral vasospasm, exacerbating the ischemia.
• Vasospasm can be treated with nimodipine or "Triple H Therapy": Hemodilution, hypervolemia, and hypertension.

• Anesthetic agents decrease the neuronal functional energy expenditure, but do not change the basal energy expenditure. Thus, all volatile agents, propofol, and barbiturates can reduce the CMRO₂ to a flat EEG reflecting electrical silence, but can act no further (unlike hypothermia).
  • Regional differences. Propofol and barbiturates produce a larger overall reduction in CMRO₂ than inhalational anesthetics; this is of unknown benefit. Additionally, this decrease is not uniform across the brain; propofol and barbiturates produce a larger reduction in the cerebellum and deep structures, whereas volatile agents produce a greater reduction in the neocortex.
  • Direct effects. Propofol and barbiturates have minimal effect on cerebrovascular tone but may decrease the CPP by lowering the MAP. Propofol has slight cerebral vasoconstrictive properties. Opiates have the least direct effect on CBF (if the MAP is maintained); exceptions include the ability of alfentanil and normeperidine to induce seizure foci, resulting in a focal increase in CMRO₂ in the seizure area. Ketamine increases CMRO₂ in focal areas and therefore CBF. It also dilates cerebral vasculature, increasing the ICP, and can induce seizure activity.
  • Autoregulation. Volatile agents dilate the cerebral vessels but do not abolish autoregulation. The effect is transitory and, within 5 hours, blood flow returns to normal.
• CBF in chronic hypertensives. The autoregulation region is right-shifted and vessels are narrower, thus patients can experience ischemia at moderately low blood pressures compared to normotensives. Perioperatively, patients should have their BP maintained at a higher pressure to compensate and avoid ischemia. Generally, a change of less than 30% from baseline is recommended.
• Perioperative stroke. The rate of postoperative stroke is small (0.08–0.4%) but is slightly higher in patients with known cerebrovascular disease (0.4–3.3%). It is highest in patients having open cardiac surgery (4%). Documenting preoperative neurological deficits is imperative in identifying brain regions susceptible to ischemia, to document changes postoperatively, and to identify patients at risk of difficult extubations (cranial nerve involvement or cognitive problems). The mortality after postoperative stroke is high (26%).
• EEG. The absolute EEG frequencies are less important than sudden changes or trends during intraoperative occurrences or surgical events (e.g., carotid clamping during a CEA in order to decide whether to create an arterial shunt).
• Cerebral vessel anatomy. for relevant surgeries, the patency of the Circle of Willis and presence of collateral blood flow can be tested preoperatively using cerebral angiography. This may aid with estimating the chances of cerebral ischemia.

CPP = MAP – (CVP or ICP); the greater of the two values is utilized. CPP is cerebral perfusion pressure, MAP is mean arterial pressure, CVP is central venous pressure, and ICP is intracranial pressure.

• Cerebral Metabolic Rate
• Carotid Endarterectomy
• Electroencephalogram (EEG)
• Craniotomy, Cerebral Aneurysm Clipping
• Circle of Willis
• Postoperative Stroke

Jayson T. Maynes , MD, PhD

Ivan M. Kangrga , MD, PhD