Cerebral Blood Flow and Metabolism: Regulation and Pathophysiology in Cerebrovascular Disease

Introduction

Energy production in the brain relies on the metabolism of exogenous compounds with a high-energy content, primarily the oxidation of glucose. Since the storage of substrates for energy metabolism is minimal, the brain is normally highly dependent on a continuous supply of oxygen and glucose from the blood for its functional and structural integrity.

Normal Values of Cerebral Blood Flow and Cerebral Metabolism

The brain accounts for approximately 2% of body weight and receives 16% of cardiac output. Healthy young adults have an average whole-brain CBF of approximately 46 mL/100 g/min, cerebral metabolic rate of oxygen (CMRO ₂ ) of 3.0 mL/100 g/min (134 μmol/100 g/min), and cerebral metabolic rate of glucose (CMRglc) of 25 μmol/100 g/min using the Kety-Schmidt technique. Normal whole-brain values obtained by positron emission tomography (PET) are comparable: CBF 48 mL/100 g/min, CMRO ₂ 2.8 mL/100 g/min (125 μmol/100 g/min), and CMRglc of 22 μmol/100 g/min. PET measured CBV is 3.7 mL/100 g and vascular MTT is 4.7 seconds (one author’s data). When plasma ketones are elevated by ketogenic diet or fasting, oxidation of ketones partially replaces glucose oxidation.

In newborn infants, mean whole-brain CBF is much lower, 6–35 mL/100 g/min. Preterm infants with subsequently normal development at 6 months of age may have CBF below 10 mL/100 g/min. Whole-brain CMRO ₂ in the normal newborn is also very low, often below 1.3 mL/100g/min. ^, Mean whole-brain CMRglc in the newborn has been reported to be 4–19 μmol/100 g/min.

Beyond the neonatal period, average whole-brain CBF, CMRO ₂ , and CMRglc progressively increase, reaching a maximum at 3–10 years with peak CBF 60–140 mL/100 g/min, peak CMRO ₂ 4.3–6.2 mL/100 g/min, and peak CMRglc 49–65 μmol/100 g/min. ^, By late adolescence, cerebral flow, oxygen, and glucose metabolism decrease to adult levels. ^{, , ,} Many studies report that CBF declines further from the third decade onward, albeit much more slowly than the decrease in adolescence. The changes in CMRO ₂ and CMRglc with age are less clear, with several studies showing a decrease ^{, ,} and others showing no change. Studies that have corrected for brain atrophy show lesser or absent changes in CBF with increasing age. ^, One author’s own data corrected for brain atrophy from 23 normal subjects, ages 23–71 years, show no significant change in CBF or CMRO ₂ , but a small decline in CMRglc.

Control of Cerebral Blood Flow

CBF is regulated by the cerebral perfusion pressure (CPP) and the cerebrovascular resistance (CVR):

CPP is equal to the difference between arterial and cerebral venous pressures. Cerebral venous pressure is negligible unless there is elevated intracranial pressure (ICP) or venous obstruction. Under constant CPP, changes in CBF must occur as a result of changes in CVR. CVR is determined by blood viscosity, vessel length, and vessel radius. Only vessel radius is amenable to rapid physiologic regulation. Thus, the control of CBF under conditions of normal CPP is determined by the caliber of the resistance vessels, primarily arterioles but also larger intracranial and intracranial arteries, which dilate and constrict in response to a variety of stimuli.

Relationship of Cerebral Blood Flow and Metabolism

Although it may seem self-contradictory, the term “resting brain” has proven to be a useful concept for understanding the relationship between CBF and metabolism. The brain is considered to be “resting” in awake humans not specifically engaged in any cognitive or behavioral task, often stipulated with eyes closed. In resting brain, CBF in gray matter (80 mL/100 g/min) is approximately four times higher than in white matter (20 mL/100 g/min). Regional blood flow is closely matched to the resting metabolic rates of oxygen and glucose, both also higher in gray than in white matter (see Fig. 3.3 ). For both oxygen and glucose, the amount delivered to the brain by the arterial blood (CBF × arterial concentration) exceeds the normal metabolic rate. Only approximately one-third of the blood-borne oxygen and one-tenth of the blood-borne glucose is metabolized. ^{, ,} As a consequence of the close regional coupling of resting CBF with both CMRO ₂ and CMRglc, the fractions extracted by the brain of available oxygen (oxygen extraction fraction, OEF) and glucose (glucose extraction fraction, GEF) are strikingly more uniform than either CBF or metabolism (see Fig. 3.3 ). Similarly, resting CBV is also higher in gray than white matter and closely matched to CBF resulting in a uniform vascular MTT (CBV/CBF) throughout the normal brain (see Fig. 3.2 ).

There is substantial interindividual variation in resting whole-brain values of CBF, CMRO ₂ , and CMRglc among normal humans. Normal CBF variation is significantly correlated with normal variations in CMRO ₂ and arterial oxygen content, but not with normal variations in CMRglc or arterial plasma glucose concentration, thus indicating that the metabolic factor controlling CBF in the normal resting brain is CMRO ₂ and not CMRglc. When there is a primary reduction in the metabolic rate of brain cells, there is a secondary decline in CBF to a comparable degree with little or no change in OEF. This close coupling of reduced CBF to reduced CMRO ₂ in the resting brain can be demonstrated experimentally during metabolic depression from hypothermia or barbiturate anesthesia but is also an essential CBF regulatory mechanism that influences CBF in a variety of physiologic and pathologic conditions ( Fig. 3.4 ). Because of the coupling of CBF to CMRO ₂ in the resting state, it isn’t possible to accurately interpret CBF values without knowledge of CMRO ₂ . Low CBF may be due to simply reduced metabolic demand and not an indication of restriction in supply. ^,

When simultaneously measured from arterial-jugular venous differences in the human resting brain, the CMRO ₂ /CMRglc molar ratio is 5.4, rather than 6.0, as would occur with complete oxidation of glucose. A small amount of lactate is produced by glycolysis even under these normal conditions of excess oxygen supply. ^{, ,}

When brain neurons are engaged in a specific cognitive or behavioral task, both regional CBF and CMRglc increase in the area of increased neuronal activity. Increases in CMRO ₂ , if they occur at all, are much lower. ^, As a consequence, the oxygen delivery further outstrips the oxygen demands in these areas causing a decrease in OEF and an increase in the venous oxygen concentration. Metabolism in these areas shifts to nonoxidative glycolysis with increased lactate production, even in the face of excess oxygen supply. Compartmentalization of metabolism by cell type between mainly oxidative neurons and mainly glycolytic astrocytes has been proposed, with the astrocytes acting to produce and distribute lactate to active neurons, but this hypothesis has not been generally accepted.

Blood-oxygen-level-dependent (BOLD) magnetic resonance imaging is sensitive to the changes in blood deoxygenated hemoglobin. Changes in BOLD contrast can be used detect increases in the venous oxygen concentration that occur with physiologic neuronal activity, and the measurement of this signal allows particular brain regions to be implicated in the performance of motor, sensory, language, or cognitive tasks. Examining the intercorrelation of regional BOLD activity over time in the resting brain has revealed multiple networks of brain regions that are tightly coupled. This analysis method, termed resting-state functional connectivity , provides insight into intrinsic brain function. The mechanism by which increases in neuronal activity cause increases in regional CBF remains a field of active investigation.

Response of Cerebral Blood Flow to Changes in Arterial Partial Pressure of Oxygen and Oxygen Content

Because of the sigmoid shape of the hemoglobin–oxygen dissociation curve, a significant reduction in arterial oxygen content (CaO ₂ ) and cerebral oxygen delivery (CBF × CaO ₂ ) does not occur until arterial pO ₂ falls to about 50–60 mm Hg. ^, During arterial hypoxemia, CBF does not increase until arterial pO ₂ is reduced below about 30–50 mm Hg, ^, indicating that it is primarily CaO ₂ and not pO ₂ that drives this vasodilatory response. Reductions in CaO ₂ due to anemia cause compensatory increases in CBF, and increases in CaO ₂ with polycythemia are associated with a decrease in CBF. ^{, ,} In neither of these cases does cerebral oxygen metabolism change. ^, In contrast to the vasodilation with arterial hypoxemia, anemia causes a decrease in pial artery diameter, with the resultant changes in CBF due mostly to decreased viscosity. ^, With chronic changes in CaO ₂ due to differing hemoglobin concentrations, there is a reciprocal inverse relationship between CaO ₂ and CBF throughout the range of oxygen content levels. Acute changes in CaO ₂ due to reduction in hemoglobin or pO ₂ also produce reciprocal increases in CBF, but less so than do chronic changes. Acute hypoxic hypoxemia causes a small increase in CMRO ₂ , whereas acute hemodilutional anemia does not. These changes are best interpreted as an active cerebrovascular mechanism that adjusts oxygen delivery to maintain tissue oxygenation.

Response of Cerebral Blood Flow to Changes in Blood Glucose

In contrast to the clear reciprocal relationship between CBF and CaO ₂ , alterations in arterial plasma glucose concentration have little effect on CBF. Decreasing blood glucose concentration to 2.3–3 mmol/L in normal subjects has generally been reported to cause no statistically significant change in CBF. ^, One study with greater measurement precision reported a slight decrease in CBF of 6%–8% at a blood glucose concentration of 3.0 mmol/L. More severe reductions in blood glucose down to 1.1–2.2 mmol/L produced a modest but significant increase in CBF. This likely does not represent a compensatory mechanism to maintain glucose delivery to the brain. A blood glucose level of 2 mmol/L is well below the level at which brain dysfunction and counterregulatory hormone response occur. ^, Furthermore, increases in CBF do not increase blood-brain glucose transport. ^,