Regional Differences in Cerebral Vascular Response to P a co 2 Changes in Humans Measured by Positron Emission Tomography

Abstract

Hypercapnia and hypocapnia produce cerebral vasodilation and vasoconstriction, respectively. However, regional differences in the vascular response to changes in P_aco₂ in the human brain are not pronounced. In the current study, these regional differences were evaluated. In each of the 11 healthy subjects, cerebral blood flow (CBF) was measured using ¹⁵O-water and positron emission tomography at rest and during hypercapnia and hypocapnia. All CBF images were globally normalized for CBF and transformed into the standard brain anatomy. t values between rest and hypercapnia or hypocapnia conditions were calculated on a pixel-by-pixel basis. In the pons, cerebellum, thalamus, and putamen, significant relative hyperperfusion during hypercapnia was observed, indicating a large capacity for vasodilatation. In the pons and putamen, a significant relative hypoperfusion during hypocapnia, that is, a large capacity for vasoconstriction, was also observed, indicating marked vascular responsiveness. In the temporal, temporo-occipital, and occipital cortices, significant relative hypoperfusion during hypercapnia and significant relative hypoperfusion during hypocapnia were observed, indicating that cerebral vascular tone at rest might incline toward vasodilatation. Such regional heterogeneity of the cerebral vascular response should be considered in the assessment of cerebral perfusion reserve by hypercapnia and in the correction of CBF measurements for variations in subjects' resting P_aco₂.

Keywords

CBF Human PET

Hypercapnia produces cerebral vasodilation and increases CBF, whereas hypocapnia produces cerebral vasoconstriction and decreases CBF (Kety and Schmidt, 1948; Raper et al., 1971; Shimosegawa et al., 1995). Because the vascular responsiveness to vasodilator in the brain with insufficient cerebral perfusion pressure is reduced by autoregulatory vasodilation (Kanno et al., 1988), hypercapnia can be used to estimate cerebral perfusion reserve in occlusive cerebrovascular disease (Herold et al., 1988; Levine et al., 1991; Kuwabara et al., 1997). Although hypercapnia is widely used to estimate cerebral perfusion reserve, there are only a few reports concerning cerebral vascular response to change in P_aco₂ in normal subjects (Levine et al., 1991; Shimosegawa et al., 1995; Kuwabara et al., 1997). To the authors' knowledge, there has been no detailed investigation of regional differences in cerebral vascular response to P_aco₂ changes in the human brain. Regional differences in changes in blood oxygenation level-dependent contrast (BOLD) signal during hypercapnia has been investigated in humans using magnetic resonance imaging (MRI), however, they did not directly measure CBF (Kastrup et al., 1999).

Fox et al. (1988) reported that intersubject averaging of CBF images, a technique requiring transformation of brain images of individual subjects into a standard brain shape and size in three dimensions (anatomic standardization), provides enhanced detection of focal brain responses (Friston et al., 1990). To investigate regional differences in cerebral vascular response to changes in P_aco₂, the authors measured CBF using ¹⁵O-water (H₂¹⁵O) and positron emission tomography (PET) in each of the 11 healthy subjects at rest and during hypercapnia and hypocapnia. The regional differences in cerebral vascular response were evaluated on a pixel-by-pixel basis with an anatomic standardization technique. In this analysis, it was determined whether there were brain regions that have a larger or smaller CBF response to change in P_aco₂ than the whole brain on average, that is, whether there is heterogeneity of CBF response.

METHODS

Subjects

The current study was approved by the Ethics Committee of the Akita Research Institute of Brain and Blood Vessels. Eleven healthy men (51 to 71 years of age; mean ± SD, 59.8 ± 5.5 years) were recruited and provided written informed consent. The subjects were judged healthy based on medical history, physical examination, blood screening analysis, electrocardiogram, echocardiography, magnetic resonance (MR) imaging of the brain, and MR angiography of the brain and the neck.

PET procedures

The Headtome V dual PET system (Shimadzu, Kyoto, Japan) used for all studies provided 47 sections with center-to-center distances of 3.125 mm (Iida et al., 1996). The intrinsic spatial resolution was 4.0 mm in-plane and 4.3 mm full width at half maximum axially. Reconstruction with a Butterworth filter resulted in a final in-plane resolution of approximately 8 mm full width at half maximum.

The dual PET system allowed brain and heart studies to be performed simultaneously (Iida et al., 1998). After 1-minute continuous inhalation of C¹⁵O gas (approximately 5 GBq total supplied to the mouth), 4-minute static scanning was performed and three arterial blood samples were taken. The C¹⁵O PET data of the brain were used to calculate cerebral blood volume (Martin et al., 1987). The C¹⁵O PET data of the heart were used to derive the arterial input function for the brain study (Iida et al., 1992). After transmission scanning, H₂¹⁵O PET studies were performed at rest and during hypercapnia and hypocapnia. The intervals between H₂¹⁵O PET studies were at least 15 minutes. The scanning protocol consisted of 180-second static scanning of the brain and 360-second dynamic scanning of the heart after continuous intravenous infusion of H₂¹⁵O over 2 minutes. The dose of radioactivity was 1.1 to 1.4 GBq at the start of the scanning. Cerebral blood flow was estimated using the dual PET system as previously described (Iida et al., 1998). Using the arterial input function derived from the left ventricular time-activity curve measured by the PET camera ring positioned over the heart (Iida et al., 1992), CBF images were calculated from the brain PET camera data by the autoradiographic method (Raichle et al., 1983; Kanno et al., 1987; Iida et al., 1998).

Hypercapnia was induced by inhalation of 7% CO₂ gas, starting 1 minute before scanning and continuing until the end of scanning of the heart (360 seconds). Hypocapnia was induced in the same way by hyperventilation (Kanno et al., 1988). Three arterial blood samples were taken during each H₂¹⁵O PET scanning of the heart (360 seconds) to measure arterial gaseous pressure. A head-fixation system with individual molds for each subject was used to minimize head movement during the PET measurements. H₂¹⁵O PET studies were performed at rest, during hypercapnia, and during hypocapnia in six subjects; and they were performed at rest, during hypocapnia, and during hypercapnia conditions in the other five subjects.

Regions of interest

Regions of interest (ROIs) were drawn on the original PET images. Circular ROIs were defined for the pons, thalamus, and putamen (16 mm in diameter), and elliptical ROIs were defined for the cerebellar cortex, centrum semiovale, and four neocortical regions representing the frontal, temporal, parietal, and occipital lobes (16 mm × 32 mm). Each ROI was drawn in three adjacent sections, and data were pooled to obtain the average concentration of radioactivity for the whole volume of interest.

The vascular response to change in P_aco₂ was calculated as the percent change in CBF per absolute change in P_aco₂ (mm Hg) in response to hypercapnia and hypocapnia as follows (Kanno et al., 1988):

100 \times \frac{{C B F}_{a} - {C B F}_{r}}{{C B F}_{r}} \times \frac{1}{| P_{a} {C O}_{2_{a}} - P_{a} {C O}_{2_{r}} |}

(1)

where the subscripts r and a denote rest and activation conditions (hypercapnia or hypocapnia), respectively.

Anatomic standardization

All CBF images were globally normalized to 1000 counts/pixel for CBF, and transformed into the standard brain size and shape by linear and nonlinear parameters with the system for anatomic standardization developed by Minoshima et al. (1994) (Senda et al., 1998). After anatomic standardization, the brain images of all subjects had the same anatomic format. Average images for each condition—that is, rest, hypercapnia and hypocapnia conditions—were calculated and compared on a pixel-by-pixel basis.

From the standardized CBF images, descriptive three-dimensional t-maps (paired t-test) of hypercapnia minus rest, rest minus hypercapnia, hypocapnia minus rest, and rest minus hypocapnia were calculated. In the t-maps, the areas with a significance level of P < 0.05 adjusted for multiple comparison on a pixel-by-pixel basis were considered statistically significant. These comparisons were performed to determine whether there were brain regions that have a larger or smaller CBF response to a change in P_aco₂ than the whole brain on average, that is, whether there is heterogeneity of CBF response.

RESULTS

Table 1 shows P_aco₂, P_ao₂, pH, blood pressure, and heart rate during each H₂¹⁵O PET scanning. P_aco₂ values during hypercapnia and hypocapnia were significantly greater and smaller, respectively, than values at rest. Because blood pressures were measured at the ankle in these studies, the values were approximately 25 mm Hg greater than pressures measured at the brachium.

TABLE 1.

P_aco₂, P_ao₂, pH, blood pressure, and heart rate for each H₂¹⁵O PET scan

Condition	P_aco₂(mm Hg)	P_ao₂(mm Hg)	pH	BP (systole/diastole)(mm Hg)	HR(beats/min)
Rest	39.9 ± 2.6	81.6 ± 9.1	7.419 ± 0.011	143 ± 22/82 ± 11	60 ± 8
Hypercapnia	42.8 ± 2.9*	105.2 ± 7.5*	7.391 ± 0.013*	156 ± 26*/84 ± 11	62 ± 9†
Hypocapnia	28.8 ± 2.4*	109.2 ± 8.8*	7.520 ± 0.036*	142 ± 29/76 ± 13†	62 ± 8

Values are shown as mean ± SD. BP, blood pressure; HR, heart rate.

Significance of difference from rest studies (paired t-test): *P < 0.001; †P < 0.01.

Cerebral blood flow values during rest, hypercapnia, and hypocapnia, and vascular responses to P_aco₂ changes are given in Table 2. Cerebral blood flow in all regions were significantly increased during hypercapnia and decreased during hypocapnia. The average vascular response to the change in P_aco₂ in the cerebral cortices was 6.7 ± 5.8% per mm Hg for hypercapnia and −3.2 ± 0.7% per mm Hg for hypocapnia. Vascular responses of more than 10% per mm Hg during hypercapnia were observed in the pons, cerebellum, thalamus, and putamen; lowest vascular responses during hypercapnia were observed in the temporal and occipital cortices. Responses during hypercapnia were almost identical between the cerebral cortices and the deep cerebral white matter. The greatest absolute values of vascular response during hypocapnia were observed in the pons and putamen; lowest absolute value of vascular response during hypocapnia was observed in the cerebellum.

TABLE 2.

CBF values under rest, hypercapnia, and hypocapnia conditions and the vascular response to changes in P_aco₂

	CBF (mL/100 mL · min)			Vascular response (%/mm Hg)
Region	Rest	Hypercapnia	Hypocapnia	Hypercapnia	Hypocapnia
Pons	42.4 ± 10.4	51.0 ± 10.2†	26.2 ± 5.3†	11.4 ± 14.9	−3.5 ± 1.3
Cerebellum	47.6 ± 10.8	61.0 ± 14.3†	32.6 ± 5.4†	17.5 ± 21.9	−2.8 ± 0.9
Thalamus	59.0 ± 15.6	70.7 ± 13.8†	37.7 ± 5.6†	14.2 ± 21.2	−3.1 ± 0.6
Putamen	57.7 ± 11.8	70.1 ± 16.2†	36.2 ± 5.2†	12.2 ± 14.9	−3.6 ± 1.9
Frontal cortex	43.5 ± 8.8	50.9 ± 10.8†	28.4 ± 3.5†	9.3 ± 8.8	−3.1 ± 0.9
Temporal cortex	50.1 ± 6.0	56.4 ± 8.7‡	31.6 ± 4.4†	5.3 ± 7.0	−3.4 ± 0.6
Occipital cortex	49.2 ± 7.2	54.1 ± 9.7‡	31.5 ± 3.9†	3.7 ± 4.0	−3.3 ± 1.2
Parietal cortex	44.5 ± 6.3	51.8 ± 9.2†	30.0 ± 3.0†	9.8 ± 11.9	−2.9 ± 0.7
Cerebral cortices*	46.8 ± 6.3	53.3 ± 9.0†	30.4 ± 3.2†	6.7 ± 5.8	−3.2 ± 0.7
Centrum semiovale	22.7 ± 4.0	26.2 ± 5.1†	14.4 ± 2.7†	6.7 ± 7.5	−3.3 ± 1.1

Values are shown as mean ± SD. CBF, cerebral blood flow.

Average of frontal, temporal, occipital, and parietal cortices.

Significance of difference from rest studies (paired t-test): †P < 0.001; ‡P < 0.01.

Average PET images of rest, hypercapnia, and hypocapnia conditions are shown in Fig. 1. The t-maps of hypercapnia minus rest, rest minus hypercapnia, hypocapnia minus rest, and rest minus hypocapnia are shown in Fig. 2. The t values of the pixels, in which significant relative CBF changes were identified, are shown in Table 3. During hypercapnia, significant relative hyperperfusion was observed in the pons, cerebellum, thalamus, and putamen (Fig. 2A, Table 3). Significant relative hypoperfusion during hypercapnia was observed in the temporal, temporo-occipital, and occipital cortices (Fig. 2B, Table 3) that still show hyperperfusion with less magnitude (Table 2). During hypocapnia, significant relative hyperperfusion was observed in the precentral gyrus, the prefrontal cortex, and a part of the cerebellum (Fig. 2C, Table 3) that still show hypoperfusion with less magnitude (Table 2). Significant relative hypoperfusion during hypocapnia was observed in the pons, hippocampal region, putamen, cingulate gyrus, temporal cortex, temporo-occipital cortex, and occipital cortex (Fig. 2D, Table 3).

FIG. 1.

Average positron emission tomography images of under rest, hypercapnia, and hypocapnia conditions. Slice positions are expressed in millimeters from the anterior—posterior commissure line in the standard brain. Pixel values are globally normalized to 1000 counts/pixel. Scale maximum and minimum values are 1800 and 0, respectively.

FIG. 2.

The t-maps of hypercapnia minus rest (A), rest minus hypercapnia (B), hypocapnia minus rest (C), and rest minus hypocapnia (D). For the t-maps, areas with a positive t value corresponding to the significance level of P < 0.05 are shown in white on the magnetic resonance image of the standard brain. Slice positions are expressed in millimeters from the anterior—posterior commissure line in the standard brain.

TABLE 3.

Coordinates of the pixels in which significant relative CBF changes were identified

	Coordinate
Region	x	y	z	t value
Hypercapnia minus rest
Pons	4	−20	−31.5	3.7
Cerebellum
(right)	32	−56	−31.5	6.9
(left)	−22	−60	−31.5	7.7
Thalamus
(right)	8	−8	9	6.8
(left)	−6	−14	9	7.2
Putamen
(right)	24	2	2.25	5.0
(left)	−22	−2	2.25	5.4
Rest minus hypercapnia
Temporal cortex
(right)	54	−20	2.25	2.3
(left)	−42	−18	2.25	2.9
Temporooccipital cortex
(right)	46	−60	2.25	2.6
(left)	−38	−62	2.25	3.3
Occipital cortex
(right)	24	−76	2.25	4.5
(left)	−16	−80	2.25	4.1
Hypocapnia minus rest
Precentral gyrus
(right)	50	2	31.5	3.6
(left)	−50	−2	31.5	4.6
Rest minus hypocapnia
Pons	2	−22	−31.5	3.7
Putamen
(right)	20	8	2.25	3.5
(left)	−22	10	2.25	3.5
Temporal cortex
(right)	48	−16	2.25	3.4
(left)	−44	−12	2.25	5.8
Temporooccipital cortex
(right)	42	−54	2.25	4.0
(left)	−32	−64	2.25	4.3
Occipital cortex
(right)	24	−76	2.25	2.8
(left)	−16	−78	2.25	3.6

Coordinates (x, y, z) are expressed in millimeters from the anterior commissure with positive directions toward right, anterior, and superior (Talairach and Tournoux, 1988). CBF, cerebral blood flow.

DISCUSSION

Hypercapnia

The average vascular response to the change in P_aco₂ in the cerebral cortices that we observed under conditions of hypercapnia agreed with values previously reported (Shimosegawa et al., 1995; Kuwabara et al., 1997). Significant relative hyperperfusion was observed in the pons, cerebellum, thalamus, and putamen (Fig. 2A, Table 3), and the vascular responses in these regions were more than 10% per mm Hg (Table 2). Kuwabara et al. (1997) reported a marked vascular response in the human cerebellum and putamen. A marked increase in BOLD signal during hypercapnia has been observed in human cerebellum (Kastrup et al., 1999). A marked vascular response in the pons, cerebellum, thalamus, and putamen has also been reported in fetal sheep (Helou et al., 1991) and in rat (Pelligrino et al., 1993). These regions have a large capacity for vasodilatation, that is, a large perfusion reserve. This suggests that these regions might be tolerant to reduction in cerebral perfusion pressure. The significant relative hypoperfusion that we observed during hypercapnia in the temporal, temporo-occipital, and occipital cortices (Fig. 2B, Table 3), and the low vascular responses that we observed in the temporal and occipital cortices (Table 2) suggest a low capacity for vasodilatation. Although hypercapnia is widely used for estimation of cerebral perfusion reserve in patients with occlusive cerebrovascular disease (Herold et al., 1988; Levine et al., 1991; Kuwabara et al., 1997), these regional differences in vascular response should be considered in the assessment of cerebral perfusion reserve.

The vascular responses that we observed during hypercapnia were almost identical between the cerebral cortices and the deep cerebral white matter (Table 2), as previously reported (Ramsay et al., 1993b). However, in an animal study, vascular response during hypercapnia was greater in the cerebral cortex than in white matter (Heistad et al., 1976; McPherson et al., 1995). One reason for this may be the mixing of gray and white matter and cerebrospinal fluid in the cerebral cortex ROI caused by the limited spatial resolution of PET.

During hypercapnia, significant relative hyperperfusion in the anterior part of the left cingulate gyrus was also observed (Fig. 2A). Neural activation in the thalamus and limbic system including cingulate gyrus during CO₂-stimulated breathing has been reported in humans (Corfield et al., 1995). Significant relative hyperperfusion in the thalamus and anterior cingulate cortex during hypercapnia may be related to CO₂-stimulated breathing.

In the current study, the subjects were in an older age group. The decrease in vascular response to hypercapnia during aging has been observed in rats (Lartaud et al., 1993), whereas no aging effects have been observed in humans (Levine et al., 1994). The aging effects in vascular responses to changes in P_aco₂ including the regional differences should be investigated.

Hypocapnia

The average vascular response to the change in P_aco₂ in the cerebral cortices that we observed under conditions of hypocapnia agreed with values previously reported (Shimosegawa et al., 1995; Ishii et al., 1998).

During hypocapnia, a significant relative hyperperfusion was observed in the precentral gyrus, the prefrontal cortex, and a part of the cerebellum (Fig. 2C, Table 3). A significant relative hyperperfusion during voluntary hyperventilation has been reported in humans in the bilateral primary motor, premotor cortices, and cerebellum, indicating neural activation induced by hyperventilation (Ramsay et al., 1993a; Ishii et al., 1998). In the current study, approximately the same regions were activated by hyperventilation.

During hypocapnia, a significant relative hypoperfusion was observed in the pons, hippocampal region, putamen, cingulate gyrus, temporal cortex, temporo-occipital cortex, and occipital cortex (Fig. 2D, Table 3), and large absolute values of vascular response were observed in the pons and putamen (Table 2). These regions, including the limbic system, might have a large capacity for vasoconstriction. In the pons and putamen, a large capacity for vasodilatation was also observed (Fig. 2A, Table 2), indicating that these regions might have marked vascular responsiveness. It has been reported that no regional difference in vascular response during hypocapnia was observed in elder normals (Tsuda and Hartmann, 1989), however, CBF was measured by ¹³³Xe with multidetector, which could measure for only cerebral cortical regions.

Because the cerebral vascular response to a change in P_aco₂ was heterogeneous among the brain regions, the correction of CBF measurements for variations in subjects' resting P_aco₂ are not applicable on a regional basis.

Cerebral vascular tone

In the temporal, temporo-occipital, and occipital cortices, a low capacity for vasodilatation (Fig. 2B) and a large capacity for vasoconstriction (Fig. 2D) were observed, possibly indicating that the cerebral vascular tone at rest inclines toward vasodilatation in the temporal, temporo-occipital, and occipital cortices.

The regional heterogeneity of sympathetic innervation of the intracranial arterioles has been reported, that is, the occipital lobe is less by the sympathetic innervation than the other brain regions (Edvinsson et al., 1976). Less sympathetic innervation in the occipital lobe may be related to the tendency of the cerebral vascular tone at rest to incline toward vasodilatation. Other determinants of cerebral vascular tone have been proposed, including nitric oxide (Faraci and Brian, 1994) and potassium (K⁺) channels (Faraci and Sobey, 1998). The regional differences in cerebral vascular tone measured by PET in the current study might also be related to the regional differences in nitric oxide synthase and K⁺ channel activity in the brain. However, such regional differences are currently unknown.

In conclusion, the regional differences in cerebral vascular response to a P_aco₂ change were evaluated on a pixel-by-pixel basis by an anatomic standardization technique. In the pons, cerebellum, thalamus, and putamen, a significant relative hyperperfusion during hypercapnia was observed, indicating a large capacity for vasodilatation. In the pons and putamen, a significant relative hypoperfusion during hypocapnia, that is, a large capacity for vasoconstriction, was also observed, indicating a marked vascular responsiveness. In the temporal, temporo-occipital, and occipital cortices, a significant relative hypoperfusion during hypercapnia and a significant relative hypoperfusion during hypocapnia were observed, indicating that the cerebral vascular tone at rest might incline toward vasodilatation. Such regional heterogeneity of the cerebral vascular response should be considered in the assessment of cerebral perfusion reserve and in the correction of CBF measurements for variations in subjects' resting P_aco₂.

Footnotes

Acknowledgments

The authors gratefully acknowledge the assistance of the members of the Akita Research Institute of Brain and Blood Vessels in performing the PET experiments.

References

Corfield

Fink

Ramsay

Murphy

Harty

Watson

Adams

Frackowiak

Guz

(1995) Evidence for limbic system activation during CO₂-stimulated breathing in man. J Physiol (Lond) 488:77–84

Edvinsson

Owman

Sjoberg

(1976) Autonomic nerves, mast cells, and amine receptors in human brain vessels. A histochemical and pharmacological study. Brain Res 115:377–393

Faraci

Brian

Jr (1994) Nitric oxide and the cerebral circulation. Stroke 25:692–703

Faraci

Sobey

(1998) Role of potassium channels in regulation of cerebral vascular tone. J Cereb Blood Flow Metab 18:1047–1063

Fox

Mintun

Reiman

Raichle

(1988) Enhanced detection of focal brain responses using intersubject averaging and change-distribution analysis of subtracted PET images. J Cereb Blood Flow Metab 8:642–653

Friston

Frith

Liddle

Dolan

Lammertsma

Frackowiak

(1990) The relationship between global and local changes in PET scans. J Cereb Blood Flow Metab 10:458–466

Heistad

Marcus

Ehrhardt

Abboud

(1976) Effect of stimulation of carotid chemoreceptors on total and regional cerebral blood flow. Circ Res 38:20–25

Helou

Hudak

Jones

Jr (1991) Cerebral blood flow response to hypercapnia in immature fetal sheep. Am J Physiol 261:H1366–1370

Herold

Brown

Frackowiak

Mansfield

Thomas

Marshall

(1988) Assessment of cerebral haemodynamic reserve: correlation between PET parameters and CO₂ reactivity measured by the intravenous ¹³³xenon injection technique. J Neurol Neurosurg Psychiatry 51:1045–1050

10.

Iida

Miura

Kanno

Ogawa

Uemura

(1996) A new PET camera for noninvasive quantitation of physiological functional parametric images: Headtome-V-dual. In: Quantification of brain function using PET ( Myers

Cunningham

Bailey

Jones

, eds), San Diego, CA: Academic Press, pp 57–61

11.

Iida

Miura

Shoji

Ogawa

Kado

Narita

Hatazawa

Eberl

Kanno

Uemura

(1998) Non-invasive quantitation of CBF using oxygen-15-water and a dual-PET system. J Nucl Med 39:1789–1798

12.

Iida

Rhodes

de Silva

Araujo

Bloomfield

Lammertsma

Jones

(1992) Use of the left ventricular time-activity curve as a noninvasive input function in dynamic oxygen-15-water positron emission tomography. J Nucl Med 33:1669–1677

13.

Ishii

Sasaki

Yamaji

Sakamoto

Maeda

(1998) Cerebral blood flow changes in the primary motor and premotor cortices during hyperventilation. Ann Nucl Med 12:29–33

14.

Kanno

Iida

Miura

Murakami

Takahashi

Sasaki

Inugami

Shishido

Uemura

(1987) A system for cerebral blood flow measurement using an H₂¹⁵O autoradiographic method and positron emission tomography. J Cereb Blood Flow Metab 7:143–153

15.

Kanno

Uemura

Higano

Murakami

Iida

Miura

Shishido

Inugami

Sayama

(1988) Oxygen extraction fraction at maximally vasodilated tissue in the ischemic brain estimated from the regional CO₂ responsiveness measured by positron emission tomography. J Cereb Blood Flow Metab 8:227–235

16.

Kastrup

Kruger

Glover

Neumann-Haefelin

Moseley

(1999) Regional variability of cerebral blood oxygenation response to hypercapnia. Neuroimage 10:675–681

17.

Kety

Schmidt

(1948) The effects of altered arterial tensions of carbon dioxide and oxygen on cerebral blood flow and cerebral oxygen consumption of normal young men. J Clin Invest 27:484–492

18.

Kuwabara

Ichiya

Sasaki

Yoshida

Masuda

Matsushima

Fukui

(1997) Response to hypercapnia in moyamoya disease. Cerebrovascular response to hypercapnia in pediatric and adult patients with moyamoya disease. Stroke 28:701–707

19.

Lartaud

Bray-des-Boscs

Chillon

Atkinson

Capdeville-Atkinson

(1993) In vivo cerebrovascular reactivity in Wistar and Fischer 344 rat strains during aging. Am J Physiol 264:H851–858

20.

Levine

Dobkin

Rozental

Satter

Nickles

(1991) Blood flow reactivity to hypercapnia in strictly unilateral carotid disease: Preliminary results. J Neurol Neurosurg Psychiatry 54:204–209

21.

Levine

Hanson

Nickles

(1994) Cerebral vasocapacitance in human aging. J Neuroimaging 4:130–136

22.

Martin

Powers

Raichle

(1987) Cerebral blood volume measured with inhaled C¹⁵O and positron emission tomography. J Cereb Blood Flow Metab 7:421–426

23.

McPherson

Kirsch

Ghaly

Traystman

(1995) Effect of nitric oxide synthase inhibition on the cerebral vascular response to hypercapnia in primates. Stroke 26:682–687

24.

Minoshima

Koeppe

Frey

Kuhl

(1994) Anatomic standardization: linear scaling and nonlinear warping of functional brain images. J Nucl Med 35:1528–1537

25.

Pelligrino

Koenig

Albrecht

(1993) Nitric oxide synthesis and regional cerebral blood flow responses to hypercapnia and hypoxia in the rat. J Cereb Blood Flow Metab 13:80–87

26.

Raichle

Martin

Herscovitch

Mintun

Markham

(1983) Brain blood flow measured with intravenous H₂¹⁵O. II. Implementation and validation. J Nucl Med 24:790–798

27.

Ramsay

Adams

Murphy

Corfield

Grootoonk

Bailey

Frackowiak

Guz

(1993a) Regional cerebral blood flow during volitional expiration in man: a comparison with volitional inspiration. J Physiol (Lond) 461:85–101

28.

Ramsay

Murphy

Shea

Friston

Lammertsma

Clark

Adams

Guz

Frackowiak

(1993b) Changes in global cerebral blood flow in humans: effect on regional cerebral blood flow during a neural activation task. J Physiol (Lond) 471:521–534

29.

Raper

Kontos

Patterson

Jr (1971) Response of pial precapillary vessels to changes in arterial carbon dioxide tension. Circ Res 28:518–523

30.

Senda

Ishii

Oda

Sadato

Kawashima

Sugiura

Kanno

Ardekani

Minoshima

Tatsumi

(1998) Influence of ANOVA design and anatomical standardization on statistical mapping for PET activation. Neuroimage 8:283–301

31.

Shimosegawa

Kanno

Hatazawa

Fujita

Iida

Miura

Murakami

Inugami

Ogawa

Itoh

(1995) Photic stimulation study of changing the arterial partial pressure level of carbon dioxide. J Cereb Blood Flow Metab 15:111–114

32.

Talairach

Tournoux

(1988) Co-planar stereotaxic atlas of the human brain. Stuttgart, Germany: Georg Thieme Verlag

33.

Tsuda

Hartmann

(1989) Changes in hyperfrontality of cerebral blood flow and carbon dioxide reactivity with age. Stroke 20:1667–1673