Sage Journals: Discover world-class research

Abstract

The authors tested the hypothesis that the oxygen content of brain tissue is negligible by injecting an intracarotid bolus of ¹⁵O-labeled tracer into rats. Under the hypothesis, the clearance rates of ¹⁵O radioactivity from the brain after injections of both ¹⁵O-labeled water (H₂¹⁵O) and ¹⁵O-labeled oxyhemoglobin (HbO¹⁵O) should be identical. However, the logarithmic slope of the ¹⁵O radioactivity curve after HbO¹⁵O injection (0.494 ± 0.071 min^-1) was steeper than that after H₂¹⁵O injection (0.406 ± 0.038 min⁻¹) (P<0.001, n = 13), where the time range used in the comparison was between 60 and 120 seconds after the injection. A possible interpretation of this result is that nonmetabolized O¹⁵O may dwell in the brain tissue for a finite period of time before it is eventually metabolized or returned to the blood stream unaltered. These findings contradict assumptions made by models currently used to measure cerebral oxygen metabolism.

Keywords

Oxygen metabolism Metabolized water Intracarotid bolus injection Dissolved oxygen

All methods used to measure the regional cerebral metabolic rate of oxygen using ¹⁵O radioactivity assume that oxygen supplied from the capillaries is immediately utilized and converted into water, implying that there is no oxygen reservoir in brain tissue. This idea is supported by many reports showing that tissue oxygen tension is significantly lower than that for venous blood (Erdmann, 1973; Hempel, 1977). Chance's (1965) finding that dissolved oxygen does not diffuse freely across the blood–brain barrier also supports this notion. On the other hand, several recent reports suggest that interstitial oxygen does exist in the brain. For example, the interstitial partial pressure of oxygen during normoxia was reported to be 30 mm Hg using a magnetic resonance imaging spectroscopic technique (Duong et al., 2001), and 32 mm Hg using oxygen microelectrodes (Ances et al., 2001). The significant presence of oxygen in brain tissue was also suggested by a recent quantitative PET CBF activation study using combined visual activation and hypoxia (Mintun et al., 2001).

Motivated by these conflicting reports, we have designed an experiment to directly test the hypothesis that oxygen content in the brain is negligible. The method observes the washout from brain tissue after the injection of ¹⁵O radioactivity. Assuming that the hypothesis is valid, it can be shown that the clearance rate of ¹⁵O radioactivity after an injection of HbO¹⁵O should be identical with that observed after an injection of H₂¹⁵O. Therefore, the kinetics of ¹⁵O-labeled tracers after intracarotid bolus injection of HbO¹⁵O and H₂¹⁵O were measured in rat using a homemade β detector (Yamamoto et al., 1997), and nonextracted intravascular ¹⁵O radioactivity was excluded by comparing the clearance curves after the passage of this intravascular component. The results showed that the clearance of HbO¹⁵O was faster than that of H₂¹⁵O.

THEORY

H₂¹⁵O kinetics

Assuming that H₂¹⁵O is a highly diffusible tracer under physiologic conditions, the relationship between the arterial input function C_a(t) and the brain tissue response function C_b(t) during washout is given by the following equation:

C_{b} (t) = β E_{w} f C_{a} (t) * exp {- (E_{w} f / p_{w}) t}

(1)

where f is the cerebral blood flow, E_w is the extraction fraction of water, β is a geometrical scale factor, p_w is the blood–brain partition coefficient of water (assumed to be 0.95; Herscovitch and Raichle, 1985), and the asterisk denotes a convolution integral. When an intracarotid bolus injection is applied, C_a(t) can be approximated as a unit impulse and Eq.1 becomes a monoexponential function with a clearance rate of E_wf/p_w:

C_{b} (t) = β E_{w} f B_{w} exp {- (E_{w} f / p_{w}) t}

(2)

where B_W is the concentration of radioactivity contained in the bolus injection. Thus, the rate constant E_wf/p_w can be uniquely determined from the logarithmic slope of the curve measured after intracarotid bolus injection of H₂¹⁵O.

HbO¹⁵O kinetics

There are three components to consider when analyzing the tracer kinetics of ¹⁵O radioactivity after HbO¹⁵O bolus injection: the concentrations of nonmetabolized O¹⁵O [C_n(t)] and metabolized H₂¹⁵O [C_m(t)] in the tissue and hemoglobin-bound O¹⁵O confined in the intravascular space [C_v(t)] (Mintun et al., 1984). Hence,

C_{b} (t) = C_{n} (t) + C_{m} (t) + C_{v} (t)

(3)

However, the hypothesis that the brain oxygen content is negligible requires that C_n(t) is zero. Moreover, if it is also assumed that the fraction of activity that recirculates is small, then C_v(t) is only present during the first passage of the hemoglobin-bound tracer in the intravascular space. Its contribution can be omitted by discarding the early part of the kinetics as determined by observing the vascular transit of an intracarotid bolus injection of HbC¹⁵O, a substance that does not diffuse out of the intravascular space. Therefore, for times longer than the vascular transit time, τ, the solitary source of radioactivity in the brain is metabolized oxygen. That is,

C_{b} (t) \approx C_{m} (t), t > τ

(4)

Under these conditions, it is straightforward to show that the metabolized component behaves in the same way as for the water kinetics of Eq. 2, but modified for the rate of oxygen extraction

C_{b} (t) = β O E F f B_{o} exp {- (E_{w} f / p_{w}) t}, t > τ

(5)

where B_O is the concentration of HbO¹⁵O radioactivity injected into the brain tissue and OEF is the oxygen extraction fraction. Thus, after omitting the first passage of hemoglobin-bound tracer, the logarithmic slope of the activity curve after intracarotid bolus injection of HbO¹⁵O should be identical with that of the curve measured after intracarotid bolus injection of H₂¹⁵O.

MATERIALS AND METHODS

Animal preparation

The institute's animal use committee approved the animal experiments. Anesthesia and muscle relaxation were maintained throughout the experiment with intravenous α-chloralose (35 mg/kg/h) and pancuronium bromide (0.8 mg/kg/h). Body temperature was maintained at 37.0°C with a rectal thermometer and a heating pad. Arterial blood pressure was continuously monitored and arterial blood gas measured during each experiment. The left external carotid artery was ligated and polyethylene tube (PE-10, 0.28-mm inner diameter) was introduced through the left external carotid artery to the origin for injection of the tracer to the internal carotid artery. The pterygopalatine artery was ligated to prevent leakage of the tracers into the extracerebral tissues. The rat was fixed with a stereotactic frame and the parietal bone of the left cortex was thinned with a drill to an approximate thickness of 0.5 mm. The “window” covered a 3 × 5-mm² rectangle and a β detector was positioned over it. An LDF (Unique Medicals, Tokyo) set in close proximity to the β detector was used to monitor global CBF change (Fig. 1).

FIG. 1.

Experimental setup. The head of the rat was fixed with a stereotactic frame and a β detector was put on the thinned skull of the left hemisphere. Three catheters were inserted into the left femoral artery for blood sampling for PaCO₂ and blood pressure monitoring, the left femoral vein for administration of anesthesia and muscle relaxant, and the left internal carotid artery for bolus injection of radioactive tracers. The HbO¹⁵O and HbC¹⁵O tracers were prepared by bubbling radioactive O¹⁵O and C¹⁵O into the blood, respectively. An LDF was set in the vicinity of the β detector.

Preparation of the ¹⁵O-labeled tracers

¹⁵O radioactivity was produced by a compact in-house cyclotron (BC168, Japan Steel Works, Muroran, Japan). H₂¹⁵O was synthesized by the in-target method (Hagami, 1986), and vapor of H₂¹⁵O was trapped in a saline vial. HbO¹⁵O and HbC¹⁵O were made by bubbling O¹⁵O and C¹⁵O, respectively, into heparinized blood (2 mL) taken from the donor rat and put in a glass tube (100 mL). Each injectate was placed in a polyethylene tube (PE-50) and introduced into the internal carotid artery cannula with a 2- to 3-second injection (Fig. 1). The total radioactivity injected was approximately 100 (3.7), 20 (0.5), and 20 (0.5) μCi (MBq) for the H₂¹⁵O, HbO¹⁵O, and HbC¹⁵O injections, respectively. The injected volume was between 30 and 50 μL.

Measurement and analysis

Sixteen male Sprague-Dawley rats (366 ± 43 g) were used for the study. Three rats were used to compare the kinetics of HbC¹⁵O, HbO¹⁵O, and H₂¹⁵O, and the remaining 13 were used to compare the clearance slopes of ¹⁵O radioactivity after intracarotid injections of HbO¹⁵O and H₂¹⁵O. Measurement began at the commencement of bolus injection. Monitoring cortical CBF with the LDF checked for possible perturbation of the cerebral circulation by the intracarotid injection. The time between frame acquisitions was 4 seconds and the measurement lasted for 120 seconds (30 frames in total). Multiple measurements of H₂¹⁵O, HbO¹⁵O, and/or HbC¹⁵O activity were repeated for the same rat with a minimum interval of about 20 minutes. The administration of injections was randomized to help avoid any systematic physiologic change between injections. Arterial blood gas tension was measured prior to each injection. The animals were killed with an overdose of intravenous pentobarbital after the last measurement.

The time course of ¹⁵O radioactivity at the brain surface was measured with a homemade β detector (Yamamoto, 1997). The detector consisted of a thin CaF₂(Eu) scintillator, a tapered fiberoptic plate, and a position-sensitive photomultiplier tube. The output signal consisted of a 64 × 64-pixel image fed to a CAMAC-based data acquisition system attached to a Macintosh computer running K-max software. A 10-mm-diameter detector was used in this study so that 1 pixel corresponded to an area of 0.156 × 0.156-mm². Performance tests revealed that 66% of the signal originates within a depth of 0.5 mm and 97% from within 2 mm of the detector surface.

Acquired images were transferred to a workstation (Indy, SGI) for further analysis. An elliptical ROI was selected from each data set by visual inspection of the part of the image considered to constitute cortex. Approximately 2,000 pixels covering an area of about 50 mm² were included in each ROI. The mean and standard deviation of the activity–time course was obtained from the ROIs using the Dr. View (Asahi Kasei Jyoho System, Tokyo) image analysis software. The physical decay of ¹⁵O (T_1/2 = 123 seconds) was corrected for throughout all procedures.

RESULTS

Tracer time–activity curves

The mean time courses of H₂¹⁵O, HbO¹⁵O, and HbC¹⁵O radioactivity were compared. Figure 2 shows a typical set of time–activity curves acquired after each of the three tracers was injected into a rat. Data were corrected for the physical decay of ¹⁵O. The time–activity curve of HbC¹⁵O decreased to less than 1% of the peak within 30 seconds of the injection. Thus, the vascular transit of red blood cells requires that at least the first 30 seconds of data after injection is neglected during washout analysis.

FIG. 2.

Logarithmic display of time–activity curves after intracarotid bolus injection of ¹⁵O tracers in rat. HbC¹⁵O (circle) does not leak out from the vascular space and only passed through the inside of the vessel after it was injected. H₂¹⁵O (cross) diffused into the brain tissue and was washed out by cerebral blood flow. The HbO¹⁵O (triangle) curve showed a small peak early on corresponding to nonextracted tracer. Extracted ¹⁵O was washed out afterward, but with a slightly faster slope than that measured for the H₂¹⁵O curve.

The curve after the H₂¹⁵O injection had a linear slope in the logarithmic scale, consistent with the expected monoexponential clearance of the H₂¹⁵O tracer. On the other hand, the curve after the HbO¹⁵O injection showed a small peak compatible with nonextracted oxygen passing through vessels in the region, followed by the clearance phase. Upon visual inspection, the part of this curve after the transit of intravascular tracer also appears to have an approximately linear behavior in the logarithmic scale. The gradient of the logarithmic curve can be compared with that of the H¹⁵O data.

Comparison of ¹⁵O radioactivity clearance rates

In order to quantitatively compare the kinetics of ¹⁵O radioactivity after bolus injection of H₂¹⁵O and HbO¹⁵O, we calculated the slope of the portion of the curve between 60 and 120 seconds with a least-squares fit of a straight line to the logarithmic data. Even though the HbC¹⁵O measurements showed that intravascular tracer can be neglected after only 30 seconds, the data during this later period was selected to make certain that all unextracted tracer had passed. Estimates of the HbO¹⁵O and H₂¹⁵O clearance rates for each rat are presented in Table 1. The mean ± SD HbO¹⁵O clearance rate (0.494 ± 0.07 min^-1) is larger than the H₂¹⁵O rate (0.406 ± 0.037 min⁻¹).

TABLE 1.

Clearance rate of ¹⁵O radioactivity after intracarotid bolus injection of HbO¹⁵O and H₂¹⁵O during normcapnia in 13 rats

		PaCO₂ (mm Hg)		Clearance (min⁻¹)

Rat #	Body weight (g)	H₂¹⁵O	HbO¹⁵O	H₂¹⁵O	HbO¹⁵O
1	360	44.6	48.5	0.450	0.640
2	360	41.1	41.5	0.393	0.452
3	330	44.3	37.0	0.397	0.445
4	350	37.6	35.2	0.326	0.426
5	310	40.8	40.3	0.427	0.359
6	380	39.0	35.7	0.435	0.499
7	320	43.3	43.3	0.383	0.510
8	450	46.1	48.5	0.405	0.447
9	450	35.8	44.7	0.377	0.506
10	320	43.3	41.1	0.420	0.531
11	400	36.7	38.9	0.362	0.512
12	350	40.4	42.8	0.465	0.518
13	360	35.6	38.1	0.434	0.575
Mean ± SD	365 ± 45	40.7 ± 3.5	41.2 ± 4.3	0.406 ± 0.038	0.494 ± 0.071

A pair of HbO¹⁵O and H₂¹⁵O measurements were carried out in each rat. PaCO₂ (mm Hg) and clearance slope (min⁻¹) values are listed. The clearance slope for HbO¹⁵O was systematically larger than that for H₂¹⁵O in each rat; a Wilcoxon test (n = 13) showed that this difference was significant (P <0.001).

The estimates were pooled and a Wilcoxon test (n = 13) was used to test for a significant difference between the HbO¹⁵O and H₂¹⁵O groups. The radioactivity clearance was significantly larger after HbO¹⁵O injection than after H₂¹⁵O injection (P<0.001). At the same time, the mean ± SD PaCO₂ levels of the H₂¹⁵O and HbO¹⁵O injections were 40.7 ± 3.5 mm Hg and 41.2 ± 4.3 mm Hg, respectively, and another Wilcoxon test on this data showed that there was no significant difference between the two groups. Therefore, the clearance rates of ¹⁵O radioactivity after bolus injection of HbO¹⁵O and H₂¹⁵O were significantly different under the same normocapnic Paco₂ level.

Additional observations

No sharp peak corresponding to unextracted tracer in the early part of the H₂¹⁵O kinetics was observed. It is possible that the limited temporal resolution (8 s/frame) of the β camera may have allowed a small peak to pass unresolved, but the absence of any peak was consistent across all H₂¹⁵O experiments. Therefore, it appears that E_w ≈ 1 and [¹⁵O]water behaved as a freely diffusible tracer under the current experimental conditions. With this information and p_w = 0.95, the CBF value was calculated from the [¹⁵O]water activity curves to be approximately 0.39 ± 0.035 mL·100 g⁻¹·min⁻¹, which is low compared with normal rat values (Sakurada et al., 1978) and might be due to the effects of the anesthesia.

DISCUSSION

Possible explanations for the difference in clearance rates

In the current study, we tested the hypothesis that oxygen diffusing into brain is immediately metabolized into water so that the oxygen concentration of brain tissue is negligible. If this hypothesis is correct, then the clearance rates of ¹⁵O radioactivity after bolus injection of HbO¹⁵O and H₂¹⁵O into the intracarotid artery should be identical. However, the logarithmic slope of the curve after HbO¹⁵O injection was significantly steeper than that after H₂¹⁵O injection (Fig. 2, Table 1).

One explanation for the larger HbO¹⁵O clearance slope could be the significant presence of nonmetabolized oxygen in the brain tissue. If this is truly the case, then there are two molecular forms of ¹⁵O radioactivity in brain tissue after HbO¹⁵O administration: diffused O¹⁵O and metabolized H₂¹⁵O. Under this assumption, the unbound O¹⁵O behaves as a diffusible tracer (it may diffuse into the interstitial space and back into the bloodstream without being metabolized), but the partition coefficient is unknown. If oxygen behaves as a lipophilic substance and the solubility is proportional to the lipid fraction of each component, brain tissue should have a large capacity for oxygen because it contains a significant fraction of lipid. Nevertheless, there is a much larger capacity for oxygen in red blood cells, which leads to a smaller brain–blood partition coefficient of oxygen than unity. On the other hand, the brain–blood partition coefficient of water is almost unity (Herscovitch and Raichle, 1985). Therefore, the O¹⁵O clearance should be much faster than H₂¹⁵O clearance. This means that the steeper slope of the curve after HbO¹⁵O injection might be explained as an additional relatively rapid O¹⁵O clearance added to the metabolized H₂¹⁵O clearance curve, assuming that the latter behaves in the same manner as externally supplied H₂¹⁵O.

Another potential explanation for the larger slope is the contribution of recirculating ¹⁵O radioactivity returned from systemic circulation. At maximum, the amount of returning H¹⁵O activity is expected to be about 10% (Høedt-Rasmussen et al., 1966). For the case of the HbO¹⁵O tracer, under the original hypothesis recirculating radioactivity includes both a H₂¹⁵O component and a nonmetabolized O¹⁵O component that arises exclusively from O¹⁵O radioactivity that has not diffused into the brain. The H₂¹⁵O component should produce approximately the same contribution as a H₂¹⁵O injection, and the nonmetabolized O¹⁵O component will contribute an additional factor. Therefore, the recirculation fraction after a HbO¹⁵O injection will increase the radioactivity concentration, corresponding to slower clearance than for an injection of H₂¹⁵O. It follows that the clearance curve after a HbO¹⁵O injection should be less steep than that for H₂¹⁵O in the logarithmic scale, which is contrary to our observation. Consequently, a recirculating component does not explain the difference in slope.

The possibility that some change occurred in the physiologic condition of the rats after injection cannot be ignored. However, as shown in Table 1, PaCO₂ levels were carefully monitored during all experiments and no significant differences were observed. Furthermore, the order of injection administration was randomized. Another possibility is that the injectates might have affected the cerebral circulation or global CBF (e.g., hemorheologic factors like viscosity). Monitoring CBF with LDF eliminated this possibility, and no systematic change in CBF level was observed after injection with either tracer.

Finally, as the field of view of the β detector was relatively large (approximately 10 mm in diameter) with respect to the rat brain, the clearance might have suffered from a heterogeneity problem due to the inclusion of multiple brain tissue components. However, as the geometry was identical for both injections, this heterogeneity problem cannot explain the difference in clearance slopes. Additionally, as the depth response of the detector is limited to around 2 mm and the β rays were detected through the partially thinned skull, the detected radioactivity is confined to the cortex. That is, the detector overwhelmingly detected signal from gray matter so that there was very little mixing of the signals from white and gray matter.

Consequences of nonmetabolized tissue oxygen

Assuming that the above reasoning is correct in rejecting most of the potential confounding factors discussed in the previous section, the finding of a steeper logarithmic clearance curve after HbO¹⁵O injection than after H₂¹⁵O injection suggests that there is some unknown component contributing to the former curve. If this unknown component is the result of a relatively rapid clearance of nonmetabolized O¹⁵O radioactivity from tissue, then the original hypothesis of a negligible brain oxygen concentration must be reconsidered. Abandoning the hypothesis means that it is necessary to divide the tissue activity into metabolized and nonmetabolized oxygen compartments. A suggested model is shown in Fig. 3. In this model, the radioactivity in the water compartment is unidirectionally supplied from the nonmetabolized oxygen compartment, and a fraction of the nonmetabolized O¹⁵O is returned directly to the blood stream. A brief derivation of the compartmental concentrations is included in the Appendix. The resultant tissue radioactivity during clearance is a biexponential curve with two intrinsic clearance rates.

FIG. 3.

Tracer model for radioactivity kinetics after HbO¹⁵O injection. The model consists of a vascular compartment, a nonmetabolized oxygen compartment, and a metabolized water compartment. C_n and C_m denote the radioactivity concentrations in the nonmetabolized and metabolized brain tissue compartments, respectively. C_a^O and C_a^W are the arterial nonmetabolized oxygen and metabolized water in the vascular compartment. K₁^O and k₂^O are the rate constants of radioactive nonmetabolized oxygen transfer between the arterial blood and nonmetabolized oxygen compartment. K₁^W and k₂^W are the rate constants of radioactive metabolized water transfer between arterial blood and the metabolized water compartment. k₃ is the rate constant for metabolism from nonmetabolized oxygen to metabolized water.

It is not our intention to conclusively show that biexponential behavior explains the contradiction observed in the results. After all, the two extra degrees of freedom allowed by a biexponential fitting function will obviously account for more of the variance in the data. Also, the monoexponential fits to the O¹⁵O data were very good, with an F test showing that there is a probability of less than 0.1% that a linear fit is incompatible with the logarithmic data. Nevertheless, the low temporal resolution and limited range of the data do not rule out possible biexponential behavior. Using the H¹⁵O clearance estimates as the rate of water transfer from brain tissue to vessel, k₂^W, a nonlinear least-squares calculation estimated the other rate, k₂^O + k₃, to be 1.58 (0.03, 2.03) min⁻¹. This estimate is quoted as the median and interquartile range of the estimates from all rats. Although the error in this estimate is large, it meets the expectation that k₂^O + k₃ must be the faster rate because the [¹⁵O]water clearance (k₂^W) dominates at long times.

If a biexponential interpretation of the data is correct, there may be serious consequences for the estimation of physiologic parameters with PET. For example, the assumption that nonmetabolized oxygen is negligible has been widely utilized when estimating CMRo₂. The existence of a nonmetabolized oxygen tissue compartment will result in an overestimation of the CMRo₂ determined using O¹⁵O tracer and PET. In calculating CMRo₂ with autoradiographic imaging methods like single-breath (Mintun et al., 1984) or steady-state (Frackowiak et al., 1980) inhalation, the size of the error can be simply estimated by the fraction of nonmetabolized [¹⁵O]radioactivity with respect to metabolized water. However, in order to quantify the magnitude of the error in CMRo₂, further study is necessary to measure the concentration of nonmetabolized oxygen.

CONCLUSION

The clearance of ¹⁵O radioactivity after an intracarotid bolus injection of HbO¹⁵O into rat was found to be faster than that observed after a bolus injection of H₂¹⁵O. One interpretation of this finding is that a nonnegligible presence of oxygen exists in the extravascular space of the brain. If this is true, then further study is needed to quantify the oxygen concentration in brain tissue and to estimate the error in CMRo₂ measurement with PET.

Footnotes

APPENDIX

Tracer kinetics for a model including a nonmetabolized oxygen compartment, a metabolized water compartment, and a blood compartment was derived using the following parameters:

C_a^O, arterial concentration of radioactive oxygen

C_a^W, arterial concentration of radioactive water

C_n, concentration of activity in the nonmetabolized compartment

C_m, concentration of activity in the metabolized compartment

K₁^O, rate of oxygen transfer from vessel to brain tissue

k₂^O, rate of oxygen transfer from brain tissue to vessel

K₁^W, rate of water transfer from vessel to brain tissue

k₂^W, rate of water transfer from brain tissue to vessel

k₃, transfer rate from nonmetabolized to metabolized compartment

Ignoring the contribution from the physical decay of ¹⁵O radioactivity, the equations for the model in Fig. 3 are

Assuming C_n(0) = C_m(0) = 0 and k₂^O + k₃ – k₂^W ≠ 0, these equations can be solved by Laplace transform to give

where * denotes a convolution integral. Hence, the total brain radioactivity concentration during washout (i.e., after vascular transit and with negligible recirculation) can be written as the sum of the two compartments

For the simplest case of a bolus injection of HbO¹⁵O; i.e., C_a^O = B_Oδ(t) and C_a^W = 0, the concentration of radioactivity is

which is biexponential.

Acknowledgments:

The authors thank Ms. Yuko Miura for her critical suggestions during the preparation of this paper, Dr. Kazuhiro Takahashi for valuable advice on the experimental setup, Mr. Yozo Ito for rat surgery, and the Department of Radiology and Nuclear Medicine staff for technical assistance.

Abbreviations used

References

Ances

Buerk

Greenberg

Detre

(2001) Temporal dynamics of the partial pressure of brain tissue oxygen during functional forepaw stimulation in rats. Neurosci Lett 306:106–110

Chance

(1965) Reaction of oxygen with the respiratory chain in cells and tissues. J Gen Physiol 49:163–188

Duong

Iadecola

Kim

(2001) Effect of hyperoxia, hypercapnia and hypoxia on cerebral interstitial oxygen tension and cerebral blood flow in the rat brain: An ¹⁹F/¹H MR study. Magn Reson Med 45:61–70

Erdmann

Kunke

Krell

(1973) Tissue pO₂ and cell function: an experimental study with multi microelectrodes in the rat brain, In: Oxygen supply, theoretical and practical aspect of oxygen supply and microcirculation of tissue ( Kessler

Bruley

Clark

Jr Lubbers

Silver

Strauss

, eds), Munchen-Berlin-Wein: Urban & Schwarzenberg, pp167–174

Frackowiak

Lenzi

Jones

Heather

(1980) Quantitative measurement of regional cerebral blood flow and oxygen metabolism in man using 15O and positron emission tomography: Theory, procedure, and normal values. J Comput Assist Tomogr 4:727–736

Hagami

Murakami

Takahashi

Kanno

Aizawa

Hachiya

Shoji

Shishido

Uemura

(1986) Studies on the direct synthesis of [O-15]-H₂O. Kaku Igaku 23:351–358

Herscovitch

Raichle

(1985) What is the correct value for the brain-blood partition coefficient for water? J Cereb Blood Flow Metab 5:65–69

Hempel

Jobsis

LaManna

Rosenthal

Saltzman

(1997) Oxidation of cerebral cytochrome aa3 by oxygen plus carbon dioxide at hyperbaric pressures. J Appl Physiol 43;873–879

Høedt-Rasmussen

Sveinsdottir

Lassen

(1966) Regional cerebral blood flow in man determined by intra-arterial injection of radioactive inert gas. Circ Res 18:237–247

10.

Mintun

Raichle

Martin

WRW

Herscovitch

(1984) Brain oxygen utilization measured with O-15 radiotracers and positron emission tomography. J Nucl Med 25:177–187

11.

Mintun

Lundstrom

Snyder

Vlassenko

Shulman

Raichle

(2001) Blood flow and oxygen delivery to human brain during functional activity: Theoretical modeling and experimental data. Proc Natl Acad Sci U S A 98:6859–6864

12.

Sakurada

Kennedy

Jehle

Brown

Carbin

Sokoloff

(1978) Measurement of local blood flow with iodo[¹⁴C]antipyrine. Am J Physiol 234:H59–H66

13.

Yamamoto

Seki

Kashikura

Fujita

Matsuura

Ban

Kanno

(1997) Development of high resolution beta camera for direct measurement of positron distribution on brain surface. IEEE Trans Nucl Sci 44:1538–1542

15 O Radioactivity Clearance is Faster after Intracarotid Bolus Injection of 15 O-Labeled Oxyhemoglobin than after 15 O-Water Injection

Abstract

Keywords

THEORY

H215O kinetics

HbO15O kinetics

MATERIALS AND METHODS

Animal preparation

Preparation of the 15O-labeled tracers

Measurement and analysis

RESULTS

Tracer time–activity curves

Comparison of 15O radioactivity clearance rates

Additional observations

DISCUSSION

Possible explanations for the difference in clearance rates

Consequences of nonmetabolized tissue oxygen

CONCLUSION

Footnotes

APPENDIX

Acknowledgments:

Abbreviations used

References

H₂¹⁵O kinetics

HbO¹⁵O kinetics

Preparation of the ¹⁵O-labeled tracers

Comparison of ¹⁵O radioactivity clearance rates