Radioactive oxygen-15 in the study of cerebral blood flow, blood volume, and oxygen metabolism

Radioactive Oxygen-15 in the Study of Cerebral Blood Flow, Blood Volume, and Oxygen Metabolism Michel M. Ter-Pogossian and Peter Herscovitch The short half-life of 150 led early observers t o believe that it was unsuitable for use as a biological tracer. However, initial studies with this nuclide demonstrated its potential usefulness for in vivo, regional physiologic measurements. Subsequently, techniques w e r e developed to measure cerebral blood f l o w (CBF), blood volume, and oxygen metabolism using intracarotid injection of 1sO-labeled radiopharmaceuticals and highly collimated scintillation probes to record the time course of radioactivity in the brain, The development of positron emission tomography (PET) made possible the in vivo, noninvasiva measurement of the absolute concentration of positron-emitting nuclides, A v a r i e t y of tracer kinetic models w e r e formulated to obtain physiologic m e a s u r e m e n t s from tomographic images of the distribution of 1sO-labeled radiopharmaceuticals in the brain. 1sO-labeled carbon monoxide, administered by

XYGEN-15 is a radionuclide which decays

O with a half-life of approximately 122.5 seconds by the emission of 1.74 meV (maximum)

energy positrons.l This radionuclide is most often generated through the 14N(d, n) reaction by means of deuterons accelerated to an energy of 7 to 8 meV using a cyclotron. A number of other nuclear reactions such as 14N(p, -g), 160(3H, or), ~2C(a, n), 160(% n), and 13N(p, n) can also be used for the generation of 150. Because of its short half-life, 150 may appear to be an unlikely candidate for in vivo tracer nuclear medicine studies. In fact, highly perceptive early investigators of tracer methodology unequivocally dismissed the possible use of 150 as a physiologic tracer. W.E. Siri, in Isotopic Tracers and Nuclear Radiations with Applications to Biology and Medicine, 2 states in a paragraph on isotopic tracing of oxygen, "The unstable species of longest half-life is 015 (126 sec); this has not been employed for tracer work and does not offer much promise." M.D. Kamen, in Isotopic Tracers in Biology: Introduction to Tracer Methodology, was equally pessimistic about 150. He states 3 "No radioactive isotope of oxygen is sufficiently long lived to be useful in tracer work. One, 014, is a positron emitter with 7"1/2 = 76.5 sec and the other, 015, is a positron emitter with 7"1/2 = 118 sec; the third, 019, is a negatron emitter with

Seminars in Nuclear Medicine, Vol XV, No 4 (October), 1985

inhalation, binds to hemoglobin in RBCs, and therefore can be used as an intravascular tracer to measure regional cerebral blood volume (rCBV). Several strategies have been developed to measure regional CBF using 150-labeled w a t e r as an inert, diffusible f l o w tracer. Regional cerebral oxygen metabolism is measured using scan data obtained following the inhalation of 150-labeled oxygen; independent determinations of local blood flow and blood volume are also required for this measurement. The tracer kinetic models used to measure rCBV, blood flow, and oxygen metabolism will be described and their relative advantages and limitations discussed. Several examples of the use of 1SO tracer methods will be reviewed to demonstrate their widespread applicability to the study of cerebral physiology and pathophysiology. 9 198.6 b y G r u n e & S t r a t t o n . Inc. = 29 sec. Only the isotope with mass 18 is useful as a tracer." It is ironic that today 150 is a radioactive tracer which has proven itself to be of major importance for the in vivo, regional measurement of blood flow, blood volume, and oxygen metabolism in the brain and other organs. With regard to these measurements, 150 offers the only method for the in vivo, regional determination of oxygen metabolism; at this time, it is probably the best tool for the measurement of blood flow and one of the best for the determination of blood volume. Probably the first application of 150 as a regional biologic tracer took place in the mid 1950's, when a group of investigators at Washington University in St Louis used this radionuclide for the assessment of oxygen distribution in

7"1/2

From the Division o f Radiation Sciences, Edward Mallinkrodt Institute o f Radiology, and the Department o f Neurology and Neurological Surgery, the Washington University School o f Medicine, St Louis. The P E T tracer and imaging sections were contributed by Dr. Herscovitch with Dr. Ter-Pogossian contributing the balance. Address reprint requests to Michel M. Ter-Pogossian, PhD, Washington University School o f Medicine, 510 S Kingshighway Blvd, St Louis, MO 63110. 9 1985 by Grune & Stratton, Inc. 0001-2998/85/1504-0009505.00/0

377

378

TER-POGOSSIAN ANO HERSCOVITCH

B

lem. malignant neoplasms. 4 The technique consisted of exposing tumor-bearing mice to air labeled with cyclotron-produced 1502 and dropping them into liquid nitrogen. The tumor was then sectioned and radioautographs were obtained (Fig 1) showing the irregular distribution of the tracer. This work with ~sO demonstrated its potential usefulness in biomedical investigations, and injected into the literature a considerably more optimistic opinion of the worth of this radionuclide.* The usefulness of 150 in physiologic studies was reemphasized in 1958 at the Hammersmith Hospital (London) by one of the members of the Washington University group (M.T-P) while on sabbatical leave in England) Beginning in 1958, a team led by P. Hugh-Jones with J. West and C. Dollery carried out a number of innovative studies on regional pulmonary function. 5-9 Concurrently, studies of the usefulness of 150 in biologic applications were also carried out at Washington University. 1~ It was quickly realized by both groups that the oxygen of respiration is rapidly converted into water of metabolism and that ~sO can indeed be used as a tracer of regional metabolism and blood flow. Methods for the labeling and quality control of 150*"It is the opinion of the authors that oxygen-15 can be applied to the study of considerable numbers of biological problems, particularly in cases where a steady influx of irradiated air can be used. For example, the irradiated air can be fed to a subject by means of a mask and its distribution can be followed by means of an external scintillation counter."

Fig 1. Photograph (AI of a section of a spontaneous m a m m a r y mouse tumor obtained, and radioautograph (BI of the same section by means of radioactive 1sO administered t o the mouse by inhalation before sectioning,

radiopharmaceuticals were steadily improved between 1957 and 1966. 4'6'13-16

Cerebral Studies With tSO The above studies, which showed that 150 could be effectively used as a tracer for in vivo, regional measurements by the systemic administration of 1SO-labeled oxygen, were carried out mostly in muscle and lung. By that time, it was recognized that " . . . metabolic oxidation is a process which in most of its stages is comparable in time scale to the half-life of oxygen (lSO). Therefore, 150 can indeed be used to study many phases of metabolism. Thus, the combination of the need for a radioactive tracer of oxygen and the fact that most phenomena studied that way are short in duration has made 150 a useful and particularly attractive tracer in biological and medical studies. ''~7 Following the recognition of the usefulness of 150 as a biologic tracer, it was a natural extrapolation to extend its use to studies of the brain because of the importance of oxidative metabolism to the functional integrity of that organ. A method for the assessment of regional cerebral oxygen supply and use by means of ~50 was described in 1967 by Ter-Pogossian et al. ~8 This method consisted of the administration by inhalation of 150-labeled oxygen in the form of air. The distribution of the radioactive label was followed regionally in the brain by means of highly collimated scintillation probes (Fig 2). The model on which the validity of this approach

CEREBRAL STUDIES WITH OXYGEN-15

Fig 2. (A) Sketch of t w o scintillation collimated probes used in early studies of regional cerebral o x y g e n metabolism and blood f l o w by means of ~"0. (B) Isocount curves for these probes for the annihilation radiation.

379

380

TER-POGOSSIAN AND HERSCOVITCH 10

I

9-

I

~k/ ~r

8-

;

I

z

I

EXTERNAL RECORDING OF ACTIVITY IN BRAIN HUMAN SUBJECT (N.M,)

~7:D 6>-

5-

~

4-

<

3-

~2 ./.~ ii!~

-003t~ OI5WATER

~

~c (]-e

9

)

2-

A

10

U <

O

1oo '

0

B-

I

200 '

300

I

I

'

7REPARTITION OF ACTIVITY IN ARTERIAL BLOOD x

5-

~4E ~" 3 s

-015

OXYHEMOGLOBIN . . ~ O

15 WATER ( x S )

2-

B

I0 0

200

400

TIME AFTER 015 I N H A L A T I O N

was as follows (Fig 3). Oxygen of respiration is pulse-labeled by adding a small volume of radioactive oxygen to the inspired air. The radioactivity observed in a volume of tissue, including blood vessels, progresses to its ultimate conversion to water of metabolism, by combination with hydrogen ions supplied by the cytochrome system, through the following compartments: (1) hemoglobin-bound oxygen; (2) oxygen dissolved in blood; (3) oxygen dissolved in tissues; and (4) intrace|lular water of metabolism. In turn, the intracellular water of metabolism exchanges with the total water pool. This methodology was tested in experimental animals and applied in a series of volunteers and patients with cerebral pathology (Fig 4). It should be noted that this method did not distinguish between oxygen supplied to the tissues and oxygen use. In 1969, a method was described for the determination of regional cerebral blood flow (CBF) by means of water labeled with radioactive ]50.19 The approach used was a modification of earlier methods using saline solution of

600

(SECONDS)

~

Fig 3. Distribution of 1SO radioactivity in a human subject subsequent to inhalation of air labeled with 150. (A) (1) External recording of radioactivity with a scintillation probe positioned over parietal area; (2) clearance of hemoglobinbound radioactivity from carotid blood: (3) difference of curve 1 and curve 2 representing contribution of waterbound activity to the external recording: (4) plot of equation C(l-e ~ which fits curve 3. C is a constant and t is time expressed in seconds. (B) Repartition of activity in carotid blood between oxyhemoglobin and water of metabolism. (Reproduced with permission, TM)

radioactive inert gases such as krypton-85 and xenon-133 injected into the internal carotid artery. The proposed approach offered the following advantages over the use of radioactive inert gases: (1) tissue:blood partition coefficient values required for flow calculation are easier to determine for water than for inert gases; (2) water is less soluble in lipids than inert gases commonly used for blood flow studies; and (3) the higher energy of the annihilation radiation permitted more reliable measurements in the brain in depth. The fundamental concepts embodied in the above two techniques were shortly after incorporated into methods yielding regional measurements in the brain of (1) blood flow; (2) oxygen extraction fraction; (3) oxygen metabolism; and (4) blood volume. 2~ These measurements rely upon the exclusive use of ~50 as a radioactive label. The approach used for performing these measurements was as follows: "The internal carotid artery of a subject is injected first with a small volume of his own blood with the

CEREBRAL STUDIES WITH OXYGEN-15

381

oxyhemoglobin labeled with radioactive oxygen-15. Then a second injection is performed under identical conditions but with blood labeled with water-l~O. After each injection, the distribution of the radioactive label in the brain is measured, as a function of time, by means of a series of narrowly collimated scintillation probes placed over the subject's head. The recording obtained after the first injection [Fig 5] reflects: (a) the arrival of the oxygen label into the tissues, (b) its partial conversion into water of metabolism, and (e) the washout of that water from the brain. The ratio of the amount of labeled water formed to the amount of oxygen presented to the tissues provides a measure of fractional oxygen utilization. The measure of the rate of washout of the water subsequent to the second injection permits the determination of cerebral blood flow. The product, fractional oxygen utilization x blood flow x arterial oxygen content, gives a measure of regional oxygen utilization rate. ''2o

fusible intravascular indicator, carboxyhemoglobin labeled with 150, which provided a measure of regional blood volume. 2' This approach was validated in experimental animals (miniature Pitman-Moore pigs) and applied in a series of patients with a variety of neurologic complaints. 23 Later, these methods were applied in studies of cerebrovascular disease, 24'25 dementia, 26 pseudotumor cerebri, 27 and cerebral functional activation. 2gThese investigations were carried out by means of numerous scintillation probes located strategically over the brain (Fig

6). The above approaches provided the foundations, albeit with important modifications, for several methods which are presently used widely to measure CBF, oxygen metabolism, oxygen extraction fraction, and blood volume in the brain.

Some aspects of the validity of this method were tested by the intracarotid injection of a nondif4

L.H. 7-11-67 front.

Lf.

2

t

p r e op

b I Fig 4. External recordings of radioactivity detected with scintillation probes positioned over contralateral and frontal areas of a patient with cerebral pathology. (A) Prior to corrective surgery; (B) after surgery. The patient was a 55-year-old male with transient ischemic attack (TIA) involving the right arm, leg, and speech. Arteriography showed severe stenosis of the left internal carotid artery that only partially carried the left middle cerebral artery system. Right to left intracranial collateral flow was also demonstrated. A carotid endarterectomy was performed between the t w o 1SO studies, (Reproduced with permission. TM)

I

~4

I

A

L.H. 7-11-67 front.

RL and U 3 ~b

post op.

I

TIME

I

100 200 A F T E R 0 ~5 I N H A L A T I O N ( s e c o n d s ) ~

,B 300

382

TER-POGOSSIAN AND HERSCOVITCH

N.M.

3-7-69

parietal

150- Hgb inj. H g b sol. 9 6 % 3000 rcbf: rcrr:

51 i n t . / r a i n . / t O 0 g. 3.1 m l . / m i n . / l O O g .

t/J ~J 2000

I.-

>Iu

I000

I I[ I0

20 TIME

30 AFTER

40 INJECTION

The development of positron emission tomography (PET) approximately two decades after the early studies with 150 considerably enlarged the scope of cerebral investigations with that radionuclide. PET made it possible to obtain tomographic images of the distribution of 150 with a higher contrast and spatial resolution than achievable with the previously used collimated scintillation probes. However, the gains yielded by PET were achieved at the cost of a considerable loss in temporal resolution as compared to scintillation probes. Typically, the early cerebral studies carried out with probes permitted temporal sampling with a high statistical accuracy at the rate of five samples per second, whereas even

Ii Fig 6. Multiprobe probe system designed for the measure of CBF, blood volume, and oxygen metabolism by means of VSO. The device includes 26 highly collimated scintillation probes distributed over a patient's brain in the attempt to minimize overlap of their fields of view.

50 SEC.

60

Fig 5. Recording of radioactivity obtained over the parietal lobe of a patient subsequent to the injection into the internal carotid artery of blood with hemoglobin labeled w i t h 1SO demonstrating the method used to determine the oxygen extraction fraction in the brain, here equal to 32%.

the fastest PET devices do not allow rates faster than approximately one image every 10 seconds. The addition of PET to the use of 150 in cerebral studies has made possible the application of biologic models unsuitable to regional probe measurements, but PET has also forced the development of strategies compatible with the relatively low temporal resolution of some of these devices. These factors will be discussed further below.

150 Tracer Methods for Cerebral Studies with PET The development of PET permitted the in vivo absolute quantitation of radiotracers in a noninvasive fashion. However, appropriate tracer strategies were required to obtain physiologically meaningful measurements from the tomographic measurements of regional radioactivity. As noted above, these strategies had to be compatible with the limited temporal resolution of PET. Thus, sufficiently long data collection times are required to obtain an image with adequate statistical accuracy. In addition, in keeping with the noninvasive nature of PET, these newer tracer strategies involved more acceptable methods of tracer administration, inhalation, or intravenous (IV) injection, rather than the technique of intracarotid injection used in earlier physiologic studies with 150. A variety of methods have been developed to measure regional CBF, blood volume, and oxygen metabolism using 150-labeled radiotracers and PET. These methods and their

CEREBRAL STUDIES WITH OXYGEN-15

383

relative advantages and disadvantages will be discussed further. CEREBRAL BLOOD VOLUME

The determination of regional cerebral blood volume (rCBV) was one of the earliest and, conceptually one of the simplest, measurements made with PET. Trace amounts of carbon monoxide, labeled with either ~C or 150 are administered by inhalation.29-3aThe labeled carbon monoxide binds to hemoglobin in RBCs, and is thus confined to the intravascular space. Following equilibration of the labeled carboxyhemoglobin throughout the body's blood pool, which requires approximately two minutes, the local radioactivity recorded from the brain is directly proportional to the local RBC volume. Conceptually, rCBV can be calculated from the ratio of radioactivity concentration in brain to that in peripheral blood. However, as the tracer labels only the RBC volume, one must correct for the difference between peripheral, large vessel hematocrit, and the hematocrit in the cerebral vasculature.32 Thus, rCBV (mL/100g of brain) (Fig 7) is calculated from the local tissue radiotracer concentration, C t (cps/g), and peripheral blood radiotracer concentration, Cbl (cps/mL),

Fig 7. Quantitative PET image of rCBV in a normal subject, obtained with the PETT VI tomograph following the brief inhalation of C~sO. The superior saggital sinus is seen anteriorly and posteriorly end the differences in vascular density between gray and white matter are clearly delineated.

as

rCBV

Ct

Cbt • R

• 100

[1]

where R is the ratio of the cerebral hematocrit to peripheral, large vessel hematocrit. A value of 0.85 has been used for R, based on an average of values obtained in both animal and human studies. 29"3~ However, recent tomographic studies in which cerebral hematocrit was calculated from measurements of both plasma volume and RBC volume indicate that the use of this single, standard value may be incorrect. Not only may this ratio be less than 0.85, 33'34but it may also vary in different physiologic conditions, such as during CO 2 inhalation)4 Although the same tracer model is applied when either HC or 1sO-labeled carbon monoxide is used, there are certain practical advantages to using 150. First, the use of the shorter-lived nuclide results in a rapid decay of the radioactive background following the rCBV measurement, so that other PET determinations, some of which

may require the use of longer-lived nuclides such as ~'C or ~SF,can be made. Second, the use of ~50 results in a lower radiation exposure to the subject. 35 Finally, as the rCBV determination is often made in conjunction with measurements of CBF and oxygen metabolism that require the use of ~50, the use of 150 for the rCBV measurement as well, avoids changing cyclotron targetry during the patient study. However, 11CO may be preferable when single-slice tomographs are used, as multiple tomographic slices can be obtained sequentially following a single administration of the longer-lived tracer. The measurement of rCBV is of physiologic importance because it may reflect local vasodilation of the cerebral vasculature in response to decreased perfusion pressure. 24'36"37 In addition, determination of rCBV is often an important component of other PET radiotracer methods. This is because rCBV data are often required to correct measurements of local radiotracer concentration for radiotracer located in the intravascular space of the brain, so that the amount of

384

TER-POGOSSIAN AND HERSCOVlTCH

radiotracer actually entering brain tissue can be accurately determined. 31'38'39 CEREBRAL BLOOD FLOW

Most methods for measuring regional CBF (rCBF) with PET use ~50-labeled water as the flow t r a c e r . H2150 has several desirable characteristics for this application. Water is a biologically inert, naturally occurring, body constituent and H2150 has no undesirable physiologic or pharmacologic side effects. It is easily synthesized in large millicurie quantities 16and is chemically stable in the body. Because of the short half-life of 150, relatively large amounts of radioactivity can be administered to obtain statistically satisfactory images over brief time periods while keeping the radiation exposure to the subject within acceptable limits. In addition, the short half-life results in rapid decay of the radioactive background once the rCBF determination is completed, so that other PET measurements can be easily performed in conjunction with the flow study. Methods for measuring rCBF with PET are based on the approach developed by S. Kety and his colleagues to describe the in vivo behavior of inert, diffusible flow tracers. 4~44 Once such a tracer is introduced into the circulation, the rate of change of tracer concentration in a tissue region is equal to the difference in the rate at which the tracer is transported to the tissue in the arterial blood, and the rate at which it is washed out from the tissue in the venous blood. This concept, known as the Fick principle, is expressed mathematically as dC t

dt - fCa - fC~

[2]

where f is the tissue blood flow (mL/min 9 g), Ct is the tissue radiotracer concentration (cps/g), and Ca and Cv are the tracer concentrations (cps/mL) in the arterial input and venous drainage, respectively, of the tissue. Because Cv cannot be measured in practice on a regional basis, Kety4~ introduced the following substitution: C~ ~ Ct/X. Here, X is the brain:blood partition coefficient for the tracer, defined as the ratio at equilibrium between the tissue and blood radiotracer concentrations. X Can be determined from independent experiments or can be calculated as

the ratio of the solubilities of the tracer in brain and blood. 4~ In the absence of limitation to diffusion of the tracer across the blood-brain barrier, the local venous radiotracer concentration remains in equilibrium with that of the tissue, so that Cv in equation 2 can be replaced by C,/X: dC t

dt

= f ( C a - Ct/~k ).

[3]

This equation, originally developed to measure rCBF in laboratory animals with tissue autoradiography, forms the basis for tracer models used to measure rCBF with PET and H2150. The steady-state inhalation technique was the earliest described method for measuring rCBF with 150 and PET. 46~8 rCBF is measured during the continuous inhalation of trace amounts of 1SO-labeled CO2, delivered at a constant rate. Under the catalytic action of carbonic anhydrase in the RBCs in the pulmonary vasculature, the ~50 label is transferred to water: 9 As a result, H2~50 is constantly generated in the lungs and circulates throughout the body. After approximately ten minutes of inhalation, a steady state is reached in which the amount of radiotracer delivered to a brain region equals that leaving the region by radioactive decay and by washout into the venous circulation. As a result, the distribution of regional radioactivity in the brain remains constant. In this case, equation 3 may be reformulated as dC--2= f ( C a - C t / ~ ) - K d C t = 0 dt

[4]

where Kd is the decay constant of ~50. Thus flow (f) can be expressed in terms of the measured arterial and tissue radiotracer concentrations: f

Kd C a / C t - 1/~k"

[5]

The steady-state approach is particularly suited to tomographs that can operate accurately only at relatively low count rates, as the count data required to construct the tomographic image can be accumulated over several minutes of C~502 inhalation. It is also convenient for use with single-ring tomographs, because multiple tomographic slices can be obtained by repositioning the patient during a prolonged period of

CEREBRAL STUDIES WITH OXYGEN-15

inhalation. In addition, rapid sampling of arterial blood is not required and the specific times when the samples are obtained are not critical. Although this method has not been validated by direct comparison with another accepted method for measuring rCBF, it has been shown to provide rCBF measurements that vary appropriately in response to variations in arterial pCO2 in experimental animals. 5~ In addition, rCBF measurements obtained in normal human subjects are comparable to those that have been obtained by other techniquesfl The steady-state rCBF method has been extensively analyzed in the literature) 2-59 Some of its limitations arise from the nonlinear relationship between rCBF and measured tissue radiotracer concentration (equation 4), such that at higher flow levels, a large change in rCBF produces a relatively smaller change in brain 150 concentration.47's2"56Thus, at high levels of flow, errors in measurement of tissue radioactivity produce proportionately larger errors in flow measurement. The relationships between the statistical accuracy of tomographic radioactivity quantitation, the total amount of radioactivity administered, and error in rCBF determination have been analyzed in detail: s'56 Similarly, calculated flow values are sensitive to errors in the measurement of Ca53and to any difference in the value of ~, used in equation 4 and the actual value of X, which may vary in pathologic conditions. 45'54's6'5~ Another limitation resulting from the model's nonlinearity relates to its behavior in heterogeneous tissue regions. 53'57'5s This technique, as well as all other PET-CBF methods, assumes that the region of interest in which the measurement is made is uniform in tissue composition and physiologic properties. However, due to the limited spatial resolution of PET, a region will receive tissue count contributions from both gray and white matter. 6~ Because of the nonlinear nature of the steady-state operational equation, the flow measured in such a heterogeneous region underestimates the true weighted regional flow by up to 20%. Deviation from the steadystate requirement of constant arterial radiotracer concentration, due either to fluctuation in cyclotron delivery of C~502, or to variation in the patient's respiratory pattern, has been a matter of concern) 3 However, Meyer et al have demonstrated that if the average concentration from

385

multiple arterial samples is used, this problem is easily avoided:9 Jones et al have implemented a modification of the steady-state inhalation method by using continuous IV administration of H2~sO.61The stated advantages of this approach include lowered radiation dose to the lungs, reduced radioactivity in the nasopharynx, which decreases scattered and random coincidence counts, and ease of tracer administration in patients unable to tolerate inhalation. In addition, they have modified the operational equation of the steady-state technique so that equilibrium conditions are not required during the scanning period: 2 Thus, tomographic data collection can begin at the onset of radiotracer administration, resulting in more counts collected for the same radiation dose to the subject, and also in a more linear relationship between local brain radiotracer concentration and flow. An alternative approach to measuring rCBF with H2150 is the adaptation to PET 63~7of Kety's tissue autoradiographic method for measuring rCBF in laboratory animalsf1-44 With Kety's technique, a biologically inert, freely diffusible radiotracer is infused for a brief time period (T), followed by decapitation of the animal. The behavior of the radiotracer is described by equation 3, which can be rearranged 42-44to give C,(T) = f C,(t) * exp(- ft/X).

[6]

One solves this equation for flow (f) from determinations of the local brain radiotracer concentration at the end of the infusion (Ct(T)), as measured by quantitative tissue autoradiography, and of the arterial time-activity curve (Ca(t)) obtained by frequent blood sampling; * denotes the mathematical operation of convolution. To adapt this method to PET, one must take into account that, because of limited temporal resolution, current tomographs cannot measure the instantaneous radiotracer concentration, Ct(T ). A scan must be performed over many seconds, essentially summing the decay events occurring during the scan. Thus, the operational equation (equation 6) was modified 63'6a by an integration of the instantaneous count rate (Ct(T)) over the scan time (T~ to T 0 to correspond to the summing process of tomographic data collection:

386

TER-POGOSSIAN AND HERSCOVITCH

C = rLT~ Ct(T)dT "tl

[7]

= fja. T2r C~(t) *exp(-ft/),)dt. Here, C is the local number of counts per unit weight of tissue recorded by the tomograph. To implement this approach, 6s Hzl50 is administered by bolus IV injection, and a 40-second scan is obtained following arrival of radioactivity in the head (Fig 8). Blood is sampled every 4 to 5 seconds via an in-dwelling radial artery catheter. Once a value for ), is specified, 45 equation 7 can be solved numerically for flow.65 rCBF measurements with the PET/autoradiographic technique have been validated in the baboon by comparison with flow measured by intracarotid injection of radiotracer and external residue detection.65 In order to obtain statistically adequate images during the brief tomographic data collection time, a relatively large amount of radioactivity must be administered, typically, 50 to 75 mCi. Thus, the scanner must be able to operate accurately at the resulting high count rates. 68 In addition, a multislice scanner is preferable69 so that the entire brain can be imaged following a single radioisotope injection.

Fig 8. Quantitative image of rCBF in a normal subject, obtained following the bolus IV injection of H21"0.

The relationship between tissue counts and rCBF for the PET/autoradiographic approach is almost linear. This has several advantageous consequences. 6. Errors in measurement of tissue radioactivity result in approximately equivalent errors in rCBF; there is no amplification of error. The method works well in the unavoidable situation of tissue heterogeneity, with the measured flow approximately equal to the true weighted flow in a mixed gray matter-white matter brain region. In addition, any inaccuracy in the value of )` used in the operational equation results in only minimal error in flow. A relative disadvantage of the technique is the requirement for frequent, accurately timed, arterial blood sampies, and the assumption that the measured peripheral arterial time-activity curve equals the input to the brain. In fact, the bolus of radioactivity arrives at the radial artery sampling site several seconds later than in the brain. However, this time difference can be measured, and the peripheral arterial curve appropriately shifted in time.65 Huang et al 7~ have described a technique for measuring both rCBF and the local value of), for water using Hz150. Following the bolus IV administration of H2150, scan data are collected over a ten-minute period and arterial sampling is performed. Two image sets are reconstructed, using decay-corrected and nondecay-corrected scan data, respectively. The operational equations, derived from the basic Kety formulation (equation 3), permit the estimation of both rCBF and ), for H2150. The values for ), obtained were about 15% lower than would be expected based on the known water content of brain. A potential explanation for this result is that not all water in tissue is freely exchangeable with water in blood. However, further studies are required to clarify this issue. Simulation studies of this approach have shown that the error in rCBF measurement in heterogenous tissue is quite small, and that propagation of noise in tissue count measurement is modest. All of these methods for measuring rCBF assume that the tracer is freely diffusible across the blood-brain barrier, so that the amount of tracer entering tissue depends only upon the local flow. However, H2150 does not have this ideal behavior. At higher flow levels, there is a progressive decline in the extraction of H2~50 from

CEREBRAL STUDIES WITH OXYGEN-15

blood. 72'7JThis results in less radiotracer entering tissue than would be predicted at higher flows, leading to an underestimation of rCBF. This has been shown to occur with the PET/autoradiographic method in baboons, 65 and would be expected to occur with the other methods using H2~50 as well.54'57'71 A recent study74 using the PET/autoradiographic method has compared flow measurements obtained in the same subjects using both H2150 and nC-butanol, a flow tracer which is not diffusion limited. In comparison to HC-butanol, H2tSO was found to underestimate rCBF by approximately l 5% because of its diffusion limitation. CEREBRAL OXYGEN METABOLISM

Two methods have been described and implemented to measure regional cerebral oxygen metabolism (rCMRO2) with PET. One, which uses the continuous inhalation of ~502, was developed in conjunction with the steady-state technique for measuring rCBF, 47'4s while the other, which uses a brief inhalation of ~502, is a companion method to the PET/autoradiographic approach. 3~ The principles underlying both methods are the same and are based on the concepts used in the measurement of cerebral oxygen metabolism with intracarotid administration of ~50-radiotracers described above. 2~ A certain fraction of the oxygen delivered to the brain is extracted and used in the oxidative metabolism of glucose. As there are no stores of oxygen in brain, the rate of oxygen use can be determined from the product of the regional oxygen extraction fraction (rOEF) and the rate of oxygen delivery, which equals rCBF x arterial oxygen content. Strategies using 1502 as a tracer for measuring rOEF and rCMRO2 must adequately describe the fate of the ~50 label following zsO2 inhalation. The ~sO2 that is extracted from arterial blood by brain tissue is converted to t50-labeled water of metabolism and washed out of the brain. The labeled recirculating water of metabolism, produced by both the brain and the rest of the body, will subsequently be delivered to brain tissue via its arterial input. Thus, the tracer kinetic model must take into account the various sources from which the measured 150 activity arises: ~502 in the incoming arterial blood; extracted ~502 that is converted to 150 water of metabolism and washed out of the brain; unex-

387

tracted 1502in the brain's venous circulation; and the circulating ~50 water of metabolism that washes into and out of brain tissue. With the steady-state method, rCBF is first determined using C~502 inhalation and then scanning is performed during the continuous inhalation of ~50z. By using a two-compartment model,47:~ rOEF is calculated from the ratio of tissue counts during the ~O2 and C~50 inhalations and from measurements of blood radioactivity. As originally formulated, this model did not include a term accounting for intravascular ~sO2, thus leading to an overestimation of both rOEF and rCMRO2. 53'54The amount of overestimation depends on the local blood volume and is greater at low levels of rCBF and rOEF. A technique for correcting for intravascular 1502 has been developed 75 and implemented using data from a separate determination of rCBV, and the importance of this correction has been clearly demonstrated.76'7~ The steady-state measurement of rCMRO2 has the same practical features as the C~502 flow technique. It is particularly suited to tomographs requiring low count rates and to single-slice tomographs. Of note, tissue heterogeneity does not affect the accuracy of the OEF calculation. 75'78The accuracy of rOEF and rCMRO2 measurements in relation to tomographic counting statistics has been discussed by Lammertsma et al. 55'75 rOEF measurements with the steady-state method have been compared to direct measurements of the cerebral arterialvenous oxygen difference in baboons, s~ There was a consistant 13% overestimation of rOEF, most likely due to the lack of a rCBV correction in these experiments. Mintun et al 3~ have described an alternative method for measuring rOEF and rCMRO2 (Figs 9 and I 0) that uses scan data obtained following the brief inhalation of =50~. The method also involves measurement of rCBF with H2~50 and the PET/autoradiographic approach, and of rCBV with C~O. The two-compartment model used to analyze the scan data accounts for the production and egress of water of metabolism in the tissue, recirculating water of metabolism, and the arterial, venous, and capillary contents of 1502 in the brain. To implement this technique, a 40-second emission scan is obtained following the brief inhalation of approximately 80 to 100 mCi

388

"TER-POGOSSIAN AND HERSCOVITCH

study so that registration of these images will be obtained. PET STUDIES USING 1sO-LABELED RADIOTRACERS

Fig 9. Tomographic image of rCMRO=. This quantitative image was obtained using scan data obtained following the brief inhalation of 1"02, as well as measurements of

Most PET studies to date have used either the 150 techniques described above, or have used lSF-labeled deoxyglucose (ISFDG) to measure regional cerebral glucose metabolism,s~ A detailed discussion of the investigations using tsO-labeled radiotracers is beyond the scope of this report, and the reader is directed to recent reviews in the literature, s2~6 However, we will provide a selective overview of the use of ~50 methods to demonstrate their widespread applicability to the study of cerebral physiology and pathophysiology. 150 techniques have thus far found their widest application in the study of cerebrovascular disease, s5'8~The utility of these methods lies in their ability to provide measurements not only of rCBF, but also of rCBV, rCMRO2, and rOEF, the latter indicating the balance between local oxygen supply and demand. A variety of issues

rCBV (Fig 7) and rCBF (Fig 8).

of ~502, and frequent arterial blood samples are obtained. As with the PET/autoradiographic flow technique, this method requires a tomograph capable of operating at high count rates. Although the equation for the calculation of rOEF from the measured PET and arterial blood curve data is mathematically complex, an accurate simplification of this equation has been described. 79The accuracy of the brief inhalation technique has been demonstrated in baboons by the comparison of rOEF measured with PET to OEF measured by intracarotid injection of 150:. Simulation studies of this method have demonstrated that measurement errors in rCBV or rCBF cause approximately equivalent errors in rOEF and rCMRO2 determinations at high or normal levels of rCMRO2, although errors are amplified at low metabolic rates. Finally, it should be noted that both methods for measuring cerebral oxygen metabolism with PET, the steady-state technique, and the brief inhalation method, require the combination of tomographic data obtained from three separate emission scans. Thus, it is important to constantly maintain the subject's head position throughout the

Fig 10. Tomographic image of rOEF. Note that rOEF is relatively uniform throughout the brain because areas of high oxygen metabolism have matching high blood flows, as Can be seen in Figs 8 and 9.

CEREBRAL STUDIES WITH OXYGEN-15

relating to the pathophysiology, and potentially to the treatment of ischemic cerebrovascular disease, are under investigation. In acute cerebral infarction, the nature and time course of the alterations in rCBF and oxygen metabolism have been studied.87-~9 Areas of decreased rCBF and rCMRO2 have been found in brain regions distant to the site of cerebral infarction, probably resulting from functional depression of local neuronal activity due to interruption of afferent or efferent pathways associated with these regions. The response of ischemic, noninfarcted brain to decreased perfusion pressure caused by occlusive vascular disease has been studied.9~ Increased oxygen extraction and dilatation of intraparenchymal blood vessels, both serving to maintain oxygen metabolism, have been observed. Thresholds of rCBF and rCMRO2 for normal neuronal function and for irreversible cerebral infarction have been examined,92 suggesting that it may be possible to differentiate reversible ischemia from irreversible infarction. The hemodynamic and metabolic response of the brain to therapeutic intervention, such as extracranial/intracranial bypass surgery, is a subject of ongoing investigation.93-95 The measurement of rCBF alone in cerebrovascular disease is insufficient to characterize tissue integrity. For example, blood flow may be normal in acutely infarcted tissue because of luxury perfusion although metabolism is markedly decreased. Similarly, rCBF may be decreased in noninfarcted tissue distal to an occluded internal carotid artery although metabolism may be relatively maintained (Fig 11). Thus, measurements of metabolism are required in addition to those of flow. Although ~SFDG is widely used to measure regional cerebral glucose metabolism, the accuracy of this method in ischemic or infarcted tissue is open to question.96'97Thus, zSO tracer techniques are preferable for measuring cerebral metabolism in cerebrovascular disease. Measurements using ~50-labeled radiotracers have been performed in a variety of neurologic diseases which are not caused by disturbances of CBF or metabolism. However, such measurements are used as indirect indicators of the level of local neuronal activity.2s,gs Therefore, they may demonstrate the site and degree of local abnormalities of neuronal function and the

389

response of these abnormalities to therapeutic interventions. For example, local blood flow abnormalities have been demonstrated in basal ganglia and mesocortical regions in hemiparkinsonism, 99 and the effects of L-dopa on rCBF and rCMRO2 have been investigated.99'1~176 In senile dementia of the Alzheimer type, declines in rCBF and rCMRO2 have been observed, 1~ presumably paralleling the loss of functioning neurons. rCBF has been measured in patients with panic disorder, 1~ which is characterized by recurrent anxiety attacks in the absence of a frightening stimulus. Analysis of rCBF in brain regions thought to mediate symptoms of panic and anxiety demonstrated an abnormal asymmetry of flow in a region of the parahippocampal gyrus, rCBF and rCMROz have been measured in a variety of other diseases to seek areas of neuronal dysfunction, including schizophrenia, focal epilepsy, and multiple sclerosis) ~176 PET has been widely applied to study the response of the brain to a variety of sensory, motor, and cognitive tasks. 82 Most of these studies have used ~SFDG to measure local changes in the use of glucose. However, this approach has certain limitations. ~~ The tracer strategy used requires a period of approximately 40 minutes between ~SFDG administration and onset of scanning. Thus, it may be difficult to maintain the required physiologic steady-state during specific neurobehavioral tasks over the period of radiotracer uptake by the brain. Because of the relatively long half-life of 18F (110 minutes), studies must be performed on consecutive days, and only one or two repeat studies can be obtained. These factors introduce the difficulties involved in repositioning subjects for follow-up studies and limit the range of experimental conditions that can be studied in a given subject. Some of these difficulties may be partially overcome by the use of the shorter-lived radiotracer, HC-deoxyglucose.l~ An alternative approach to functional activation studies involves the measurement of local CBF responses using bolus IV Hz~50 and the PET/autoradiographic method. 64"65'1~ Because of the short half-life of ~50, up to eight measurements of rCBF can be performed in the same subject in a two-hour period, allowing great flexibility in experimental design. Each flow determination is performed in less than one minute. The steady-state C~502 inhala-

390

TER-POGOSSIAN AND HERSCOVITCH

Fig 11. PET study of a 69-year-old male with occlusion of the right internal carotid artery and a single right hemisphere transient ischemic attack. Although rCBF was markedly decreased in the distribution of the right middle cerebral a r t e r y (A), rCMRO 2 was only slightly decreased in this region (B). There was increase in both rCBV (C) and rOEF (D) in the right hemisphere, demonstrating compensatory mechanisms of the brain to decreased perfusion pressure.

CEREBRAL STUDIES WITH OXYGEN-15

391

tion method can also be used to measure rCBF in neurobehavioral studies) ~ However, this approach requires maintaining a physiologic steady-state for several minutes, and may provide inaccurate estimates of local flow changes because of nonlinearity in the flow model. 5s Another advantage of the PET/autoradiographic approach using H2~50 is that the image of tissue counts reflects the relative difference in flow in different brain regions. Thus, in functional mapping studies, where only relative changes in rCBF are sought, arterial sampling to provide absolute flow measurements is not necessarily required. 64'1~ These examples have served to demonstrate

the widespread applicability of ~50-labeled radiotracers to the study of cerebral physiology and pathophysiology)50, initially felt to be unsuitable for use as a biologic radiotracer, 2'3 has in fact several advantageous characteristics for this use, as described above. These characteristics lead to its application in cerebral studies, initially using intracarotid injection of radiotracer and, more recently, using PET. Although a variety of other PET radiotracer strategies have been developed or are under investigation,s3 these are complementary to the 150 techniques that remain the mainstays for the measurement of cerebral hemodynamics and oxygen metabolism.

REFERENCES

1. Lederer CM, Hollander JM, Perlman I (eds): Table of Isotopes, (ed 6). New York, Wiley, 1967 2. Siri WE: Major organic metabolites, in Isotopic Tracers and Nuclear Radiations With Applications to Biology and Medicine. New York, McGraw-Hill, 1949, p 517 3. Kamen MD:Theisotopesofoxygen, nitrogen, phosphorus, and sulfer, in Isotopic Tracers in Biology: An Introduction to Tracer Methodology. Orlando, Fla, Academic, 1957, p 339 4. Ter-Pogossian MM, Powers WE: The use of radioactive oxygen-15 in the determination of oxygen content in malignant neoplasms. Radioisotopes in scientific research, Vol 3, in Proceedings of the First UNESCO International Conference (Paris, 1957). New York, Pergamon, 1958, pp 1-11 5. Jones T: Radioisotopic measurements of regional tissue function, in Post RJ (ed): Medical Research Council Cyclotron Unit Silver Jubilee Book. London, Hammersmith Hospital, 1980, p 92 6. Dyson NA, Hugh-Jones P, Newbery GR, et al: Preparation and use of 150 with particular reference to its value in 9 study of pulmonary malfunction. Proceedings of the Second UN International Conference on the Peaceful Uses of Atomic Energy (UN, Geneva, Sept 1958). New York, Pergamon, 1959 7. Dyson NA, Hugh-Jones P, Newbery GR, et al: Studies of regional lung function using radioactive oxygen. Br Med J 1:231-238, 1960 8. Dollery CT, West JB: Regional uptake of radioactive oxygen, carbon monoxide and carbon dioxide in lungs of patients with mitral stenosis. Circ Res 8:765-771, 1960 9. West JB, Dollery CT: Absorption of inhaled radioactive water vapour. Nature 189:588, 1961 10. Ter-Pogossian MM, Spratt JS Jr, Rudman S, et al: Radioactive oxygen 15 in study of kinetics of oxygen of respiration. Am J Physio1201:582-586, 1961 11. Spratt JS, Ter-Pogossian MM, Rudman S, et al: The measurement of the pulmonary venous cross-circulation through the conjugated heart of thoracopagous twins with radioactive O ~5. Surgery 50:941-946, 1961 12. Spratt JS, Ter-Pogossian MM, Rudman S, et al:

Radioactive oxygen-15 in the tracer study of oxygen transport. Surg Forum 12:7-9, 1961 13. Ter-Pogossian MM: Short-lived isotopes and positron emitters in medicine, in Lawrence JH (ed): Progress in Atomic Medicine, Vol 1. Orlando, Fla, Grune & Stratton, 1965, pp 107-126 14. Ter-Pogossian MM: Cyclotron produced short-lived radioactive isotopes. Am J Roentgenol Radium Ther Nucl Med 96:737-743, 1966 15. Welch M J, Ter-Pogossian MM: Preparation of short half-lived radioactive gases for medical studies. Radiat Res 36:580-587, 1968 16. Welch M J, Lifton JF, Ter-Pogossian MM: Preparation of millicurie quantities of oxygen-15 labeled water. J Labeled Compounds 5:168-172, 1969 17. Ter-Pogossian MM, Wagner HN Jr: A new look at the cyclotron for making short-lived isotopes. Nucleonics 24:5057, 1966 18. Ter-Pogossian MM, Taveras JM, Davis DO, et al: A study of regional cerebral oxygen supply and utilization by means of radioactive oxygen-15, in Taveras J, Fischgold H, Dilenge D (eds): Recent Advances in the Study of Circulation. Springfield, I11, Charles C. Thomas, 1969, pp 156-174 19. Ter-Pogossian MM, Eichling JO, Davis DO, et al: The determination of regional cerebral blood flow by means of water labeled with radioactive oxygen 15. Radiology 93:3140, 1969 20. Ter-Pogossian MM, Eichling JO, Davis DO, et al: The measurement in vivo of regional cerebral oxygen utilization by means of oxyhemoglobin labeled with radioactive oxygen15. J Clin Invest 49:381-391, 1970 21. Eiehling JO, Raichle ME, Grubb RL Jr, et al: tn vivo determination of cerebral blood volume with radioactive oxygen-15 in the monkey. Circ Res 37:707-714, 1975 22. Raichle ME, Grubb RL Jr, Eichling JO, et al: Measurement of brain oxygen utilization with radiotracer oxygen-15: Experimental verification. J Appl Physiol 40:638640, 1976 23. Carter CC, Eichling JO, Davis DO, et al: Correlation

392

of regional cerebral blood ftow with regional oxygen uptake using 150 method. Neurology 22:755-762, 1972 24. Grubb RL Jr, Raichle ME, Eichling JO, et al: Effects of subarachnoid hemorrhage on cerebral blood volume, blood flow, and oxygen utilization in humans. J Neurosurg 46:446453, 1977 25. Grubb RL Jr, Ratcheson RA, Raichle ME, et al: Regional cerebral blood flow and oxygen utilization in superficial temporal-middle cerebral artery anastomosis patients. J Neurosurg 50:733-741, 1979 26. Grubb RL Jr, Raichle ME, Gado MH, et al: Cerebral blood flow, oxygen utilization, and blood volume in dementia. Neurology 27:905-909, 1977 27. Raichle ME, Grubb RL Jr, Phelps ME, et al: Cerebral hemodynamics and metabolism in pseudotumor cerebri. Ann Neurol 4:104-111, 1978 28. Raichle ME, Grubb RL Jr, Gado MH, et al: In vivo correlation between regional cerebral blood flow and oxidative metabolism in man. Arch Neurol 33:523-526, 1976 29. Grubb RL Jr, Raichle ME, Higgins CS, et al: Measurement of regional cerebral blood volume by emission tomography. Ann Neurol 4:322-328, 1978 30. Phelps ME, Huang SC, Hoffman E J, et al: Validation of tomographic measurement of cerebral blood volume with C-1 l-labeled carboxyhemoglobin. J Nucl Med 20:328-334, 1979 31. Mintun MA, Raichle ME, Martin WRW, et al: Brain oxygen utilization measured with O-15 radiotracers and positron emission tomography. J Nucl Med 25:177-187, 1984 32. Grubb RL Jr, Phelps ME, Ter-Pogossian MM: Regional cerebral blood volume in humans. X-Ray fluorescence studies. Arch Neurol 28:38-44, 1973 33. Lammertsma AA, Brooks D J, Beaney RP, et al: In vivo measurement of regional cerebral haematocrit using positron emission tomography. J Cereb Blood Flow Metab 4:317-322, 1984 34. Sakai F, Nakazawa K, Tazaki Y, et al: Regional cerebral blood volume and hematocrit measured in normal human volunteers by single-photon emission computed tomography. J Cereb Blood Flow Metab 5:207-213, 1985 35. Kearfott K J: Absorbed dose estimates for positron emission tomography (PET): clsO, lifo, and COlSO. J Nucl Med 23:1031-1037, 1982 36. Grubb RL Jr, Raichle ME, Phelps ME, et al: Effects of increased intracranial pressure on cerebral blood volume, blood flow and oxygen utilization in monkeys. J Neurosurg 43:385-398, 1975 37. Powers W, Martin W, Herscovitch P, et al: The value of regional blood volume measurements in the diagnosis of cerebral ischemia. J Cereb Blood Flow Metab 3:598-599, 1983 (suppl 1) 38. Mintun MA, Raichle ME, Kilbourn MR, et al: A quantitative model for the in vivo assessment of drug binding sites with positron emission tomography. Ann Neurol 15:217-227, 1984 39. Koeppe RA, Holden JE, Hutchins GD: Importance of correction for intravascular radioactivity in dynamic PCT studies using glucose analogs. J N u d Med 26:P90, 1985 40. Kety SS: The theory and applications of the exchange

TER-POGOSSIAN AND HERSCOVITCH

of inert gas at the lungs and tissues. Pharmacol Rev 3:1~,1, 1951 41. Landau WM, Freygang WH Jr, Rowland LP, et al: The local circulation of the living brain; values in the unanesthetized and anesthetized cat. Trans Am Neurol Assoc 80:125-129, 1955 42. Freygang WH Jr, Sokoloff L: Quantitative measurement of regional circulation in the central nervous system by the use of radioactive inert gas. Adv Biol Med Phys 6:263279, 1958 43. Kety SS: Measurement of local blood flow by the exchange of an inert diffusible substance. Methods Med Res 8:228-236, 1960 44. Sakurada O, Kennedy C, Jehle J, et al: Measurement of local cerebral blood flow with iodo[14C]antipyrine. Am J Physiol 234:H59-H66, 1978 45. Herscovitch P, Raichle ME: What is the correct value for the brain-blood partition coefficient for water? J Cereb Blood Flow Metab 5:65-69, 1985 46. Jones T, Chester DA, Ter-Pogossian MM: The continuous inhalation of oxygen-15 for assessing regional oxygen extraction in the brain of man. Br J Radiol 49:339-343, 1976 47. Subramanyam R, Alpert NM, Hoop B Jr, et al: A model for regional cerebral oxygen distribution during continuous inhalation of 1502, C150, and ClsO2. J Nucl Med 19:48-53, 1978 48. Frackowiak RSJ, Lenzi G-L, Jones T, et al: Quantitative measurement of regional cerebral blood flow and oxygen metabolism in man using 1sO and positron emission tomography: Theory, procedure and normal values. J Comput Assist Tomogr 4:727-736, 1980 49. West JB, Dollery CT: Uptake of oxygen-15-1abeled CO2 compared with carbon-1 l-labeled CO2 in the lung. J Appl Physiol 17:9-13, 1962 50. Baron JC, Steinling M, Tanaka T, et al: Quantitative measurement of CBF, oxygen extraction fraction (OEF) and CMRO2 with the ~50 continuous inhalation technique and positron emission tomography (PET): Experimental evidence and normal values in man. J Cereb Blood Flow Metab 1:5-6, 1981 (suppl 1) 51. Rhodes CG, Lenzi GL, Frackowiak RSJ, et al: Measurement of CBF and CMRO2 using continuous inhalation of C1502 and ~502: Experimental validation using CO2 reactivity in the anaesthetised dog. J Neurol Sci 50:381-389, 1981 52. Huang S-C, Phelps ME, Hoffman E J, et al: A theoretical study of quantitative flow measurements with constant infusion of short-lived isotopes. Phys Med Biol 24:11511161, 1979 53. Lammertsma AA, Frackowiak RSJ, Lenzi GL, et al: Accuracy of the oxygen-15 steady state technique for measuring rCBF and rCMRO~: Tracer modelling, statistics and spatial sampling. J Cereb Blood Flow Metab 1:$3~$4, 1981 (suppl 1) 54. Lammertsma AA, Jones T, Frackowiak RSJ, et al: A theoretical study of the steady-state model for measuring regional cerebral blood flow and oxygen utilization using oxygen- 15. J Comput Assist Tomogr 5:544-550, 1981 55. Lammertsma AA, Heather JD, Jones T, et al: A statistical study of the steady state technique for measuring

CEREBRAL STUDIES WITH OXYGEN-15

regional cerebral blood flow and oxygen utilization using ~50. J Comput Assist Tomogr 6:566-573, 1982 56. Jones SC, Greenberg JH, Reivich M: Error analysis for the determination of cerebral blood flow with the continuous inhalation of ~50-labeled carbon dioxide and positron emission tomography. J Comput Assist Tomogr 6:116-124, 1982 57. Steinling M, Baron JC: Mesure du d6bit sanguin c6r6bral local par inhalation continue de C1502 et tomographie d'6mission: l~tude des limites du mod61e. J Biophys Med Nucl 6:89-95, 1982 58. Herscovitch P, Raichle ME: Effect of tissue heterogeneity on the measurement of cerebral blood flow with the equilibrium C1502 inhalation technique. J Cereb Blood Flow Metab 3:407-415, 1983 59. Meyer E, Yamamoto YL: The requirement for constant arterial radioactivity in the ClsO2 steady-state bloodflow model J Nucl Med 25:455-460, 1984 60. Mazziotta JC, Phelps ME, Plummer D, et al: quantitation in positron computed tomography: 5. Physicalanatomical effects. J Comput Assist Tomogr 5:734-743, 198I 6l. Jones SC, Reivich M, Robinson GD Jr, et al: The measurement of cerebral blood flow with positron emission tomography using the continuous infusion of oxygen-15 labeled water. Proceedings of the Third World Congress of Nuclear Medicine and Biology, Paris. Oxford, Pergamon, 1982, pp 1744-1747 62. Greenberg JH, Jones SC, Reivich M: A non-equilibrium technique for the measurement of regional cerebral blood flow by positron tomography. J Nucl Med 25:P62, 1984 63. Raichle ME, Markham J, Larson K, et al: Measurement of local cerebral blood flow in man with positron emission tomography. J Cereb Blood Flow Metab 1:S19$20, 1981 (suppl 1) 64. Herscovitch P, Markham J, Raichle ME: Brain blood flow measured with intravenous H2~50. I. Theory and error analysis. J Nuel Med: 24:782-789, 1983 65. Raichle ME, Martin WRW, Herseovitch P, et al: Brain blood flow measured with intravenous H2~50. II. Implementation and validation. J Nuel Med 24:790-798, 1983 66. Howard BE, Ginsberg MD, Hassel WR, et al: On the uniqueness of cerebral blood flow measured by the in vivo autoradiographic strategy and positron emission tomography. J Cereb Blood Flow Metab 3:432-441, 1983 67. Kanno I, Lammertsma AA, Heather JD, et al: Measurement of cerebral blood flow using bolus inhalation of Ct~O 2 and positron emission tomography: Description of the method and its comparison with the C~502 continuous inhalation method. J Cereb Blood Flow Metab 4:224--234, 1984 68. Ter-Pogossian MM: Special characteristics and potential for dynamic function studies with PET. Sem Nucl Med 11:13-23, 1981 69. Ter-Pogossian MM, Fieke DC, Hood JT Sr, et al: PETT VI: A positron emission tomograph utilizing cesium fluoride scintillation detectors. J Comput Assist Tomogr 6:125-133, 1982 70. Huang S-C, Carson RE, Phelps ME: Measurement of

393

local blood flow and distribution volume with short-lived isotopes: A general input technique. J Cereb Blood Flow Metab 2:99-108, 1982 71. Huang S-C, Carson RE, Hoffman EJ, et al: Quantitative measurement of local cerebral blood flow in humans by positron computed tomography and ~50-water. J Cereb Blood Flow Metab 3:141-153, 1983 72. Eichling JO, Raiehle ME, Grubb RJ Jr, et al: Evidence of the limitations of water as a freely diffusible tracer in the brain of the rhesus monkey. Cire Res 35:358-364, 1974 73. Raichle ME, Eichling JO, Straatmann MG, et al: Blood-brain barrier permeability of HC-labeled alcohols and =50-labeled water. Am J Physiol 230:543-552, 1976 74. Herscovitch P, Raichle ME, Kilbourn MR, et al: Measurement of cerebral blood flow and water permeability with positron emission tomography using O-15 water and C-11 butanol. J Cereb Blood Flow Metab (in press) (suppl) 75. Lammertsma AA, Jones T: Correction for the presence of intravascular oxygen- 15 in the steady-state technique for measuring regional oxygen extraction ratio in the brain: 1. Description of the method. J Cereb Blood Flow Metab 13:416-424, 1983 76. Lammertsma AA, Wise RJS, Heather JD, et al: Correction for the presence of intravaseular oxygen- 15 in the steady-state technique for measuring regional oxygen extraction ratio in the brain: 2. Results in normal subjects and brain tumour and stroke patients. J Cereb Blood Flow Metab 3:425-431, 1983 77. Pantano P, Baron J-C, Crouzel C, et al: The 150 continuous-inhalation method: Correction for intravascular signal using C150. Eur J Nucl Med 10:387-391, 1985 78. Herscovitch P, Raichle ME: Effect of tissue heterogeneity on the measurement of regional cerebral oxygen extraction and metabolic rate with positron emission tomography. J Nucl Med 26:P62, 1985 79. Herscovitch P, Mintun MA, Raichle ME: Brain oxygen utilization measured with oxygen-15 radiotracers and positron emission tomography: Generation of metabolic images. J Nncl Med 26:416--417, 1985 80. Reivich M, Kuhl D, Wolf A, et al: The (tSF)fluorodeoxyglucose method for the measurement of local cerebral glucose utilization in man. Circ Res 44:127-137, 1979 81. Huang SC, Phelps ME, Hoffman EJ, et al: Noninvasive determination of local cerebral metabolic rate of glucose in man. Am J Physiol 238:E69-E82, 1980 82. Phelps ME, Mazziotta JC, Huang S-C: Study of cerebral function with positron computed tomography. J Cereb Blood Flow Metab 2:113-162, 1982 83. Leenders KL, Gibbs JM, Frackowiak RSJ, et al: Positron emission tomography of the brain: New possibilities for the investigation of human cerebral pathophysiology. Prog Neurobiol 23:1-38, 1984 84. Beaney RP: Positron emission tomography in the study of human tumors. Sem Nucl Med 14:324-341, 1984 85. Powers W J, Raichle ME: Positron emission tomography and its application to the study of cerebrovascular disease in man. Stroke 16:361-376, 1985 86. Ackerman RH, Correia JA, Alpert NM, et al: Posi-

394

tron imaging in ischemic and hemorrhagic stroke disease, in Greitz T, Ingvar DH, Wid6n L (eds): The Metabolism of the Human Brain Studied with Positron Emission Tomography. New York, Raven, 1985, pp 387-397 87. Lenzi GL, Frackowiak RSJ, Jones T: Cerebral oxygen metabolism and blood flow in human cerebral ischemic infarction. ,l Cereb Blood Flow Metab 2:321-335, 1982 88. Wise R,IS, Bernardi S, Frackowiak RSJ, et al: Serial observations on the pathophysiology of acute stroke. Brain 106:197-222, 1983 89. Baron 3C, Rougemont D, Soussaline F, et al: Local interrelationships of cerebral oxygen consumption and glucose utilization in normal subjects and ischemic stroke patients: A positron tomography study. J Cereb Blood Flow Metab 4:140-149, 1984 90. Gibbs JM, Wise RJS, Leenders KL, et al: Evaluation of cerebral perfusion reserve in patients with carotid-artery occlusion. Lancet 1:310-314, 1984 91. Powers W J, Grubb RL Jr, Raichle ME: Physiologic responses to focal cerebral ischemia in humans. Ann Neurol 16:546-552, 1984 92. Powers W J, Darriet D, Grubb RL Jr, et al: CBF and CMRO2 requirements for cerebral function and viability in man. J Cereb Blood Flow Metab (in press) 93. Powers W J, Martin WRW, Herscovitch P, et al: Extracranial-intracranial bypass surgery: Hemodynamic and metabolic effects. Neurology 34:1168-1174, 1984 94. Samson Y, Baron JC, Bousser MG, et al: Extraintracranial arterial by-pass (EIAB) increases CMRO2 in both cerebral hemispheres. J Cereb Blood Flow Metab (in press) (suppl) 95. Gibbs JM, Wise R,IS, Ross Russell RW, et al: Cerebral circulatory reserve before and after surgery for occlusive carotid artery disease. J Cereb Blood Flow Metab (in press) (suppl) 96. Choki J, Greenberg J, Reivich M: Regional cerebral glucose metabolism during and after bilateral cerebral ischemia in the gerbil. Stroke 14:568-574, 1983 97. Wise RJS, Rhodes CG, Gibbs JM, et al: Disturbance of oxidative metabolism of glucose in recent human cerebral infarcts. Ann Neurol 14:627--637, 1983 98. Yarowsky PJ, Ingvar DH: Neuronal activity and energy metabolism. Fed Proc 40:2353-2362, 1981 99. Perlmutter JS, Raichle ME: Regional blood flow in hemiparkinsonism. Neurology 35:1127-1134, 1985

TER-POGOSSIAN AND HERSCOVITCH

100. Leenders KL, Wolfson L, Gibbs JM, et al: The effects of L-dopa on regional cerebral blood flow and oxygen metabolism in patients with Parkinson's disease. Brain 108:171-191, 1985 101. Frackowiak RSJ, Pozzilli C, Legg N J, et al: Regional cerebral oxygen supply and utilization in dementia: A clinical and physiological study with oxygen-15 and positron tomography. Brain 104:753-778, 1981 102. Reiman EM, Raichle ME, Butler FK, et al: A focal brain abnormality in panic disorder, a severe form of anxiety. Nature 310:683-685, 1984 103. Sheppard G, Gruzelier J, Manchandon R, et al: 150 positron emission tomographic scanning in predominantly never-treated acute schizophrenic patients. Lancet 2:14481452, 1983 104. Brooks DJ, Leenders KL, Head G, et al: Studies on regional cerebral oxygen utilization and cognitive function in multiple sclerosis. J Neurol Neurosurg Psychiatry 47:11821191, 1984 105. Bernardi S, Trimble MR, Frackowiak RSJ, et al: An interictal study of partial epilepsy using positron emission tomography and the oxygen-15 inhalation technique. J Neurol Neurosurg Psychiatry 46:473~,77, 1983 106. Mazziotta JC, Huang S-C, Phelps ME, et al: A noninvasive positron computed tomography technique using oxygen- 15-labeled water for the evaluation of neurobehavioral task batteries. J Cereb Blood Flow Metab 5:70-78, 1985 107. Reivich M, Alavi A, Wolf A, et al: Use of 2deoxy-D[1-HC]glucose for the determination of local cerebral glucose metabolism in humans: Variation within and between subjects. J Cereb Blood Flow Metab 2:307-319, 1982 108. Fox PT, Raichle ME: Stimulus rate dependence of regional cerebral blood flow in human striate cortex, demonstrated by positron emission tomography. ,l Neurophysiol 51:1109-1120, 1984 109. Kearfott K J, Rottenberg DA, Volpe BT: Design of steady-state positron emission tomography protocols for neurobehavioral studies: COtSO and 19Ne. J Comput Assist Tomogr 7:51-58, 1983 110. Fox PT, Mintun MA, Raichle ME, et al: A noninvasire approach to quantitative functional brain mapping with H2~SO and positron emission tomography. J Cereb Blood Flow Metab 4:329-333, 1984