August 2011
Volume 52, Issue 9
Free
Retina  |   August 2011
The Effect of Hypercarbia and Hyperoxia on the Total Blood Flow to the Retina as Assessed by Magnetic Resonance Imaging
Author Affiliations & Notes
  • Nasim Maleki
    From the Department of Radiology, Children's Hospital Boston and Harvard Medical School, Boston, Massachusetts;
  • David C. Alsop
    the Department of Radiology, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, Massachusetts;
  • Weiying Dai
    the Department of Radiology, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, Massachusetts;
  • Christopher Hudson
    the University of Waterloo, School of Optometry, Waterloo, Ontario, Canada;
    the Departments of Ophthalmology & Vision Science and
  • Jay S. Han
    Medical Imaging, The Toronto Western Hospital and the University of Toronto, Toronto, Ontario, Canada;
  • Joe Fisher
    the Department of Anesthesiology, University Health Network, University of Toronto, Toronto, Ontario, Canada;
    Thornhill Research Inc., Toronto, Ontario, Canada; and
    the Department of Physiology, University of Toronto, Toronto, Ontario, Canada.
  • David Mikulis
    Medical Imaging, The Toronto Western Hospital and the University of Toronto, Toronto, Ontario, Canada;
Investigative Ophthalmology & Visual Science August 2011, Vol.52, 6867-6874. doi:10.1167/iovs.10-6762
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Nasim Maleki, David C. Alsop, Weiying Dai, Christopher Hudson, Jay S. Han, Joe Fisher, David Mikulis; The Effect of Hypercarbia and Hyperoxia on the Total Blood Flow to the Retina as Assessed by Magnetic Resonance Imaging. Invest. Ophthalmol. Vis. Sci. 2011;52(9):6867-6874. doi: 10.1167/iovs.10-6762.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: The feasibility of measuring total blood flow to the retina with Arterial Spin Labeling Magnetic Resonance Imaging (ASL-MRI) has been described previously. In the present study, the hypothesis was that the reactivity that the ASL-MRI detects at the human retina is dominated by the choroidal blood flow, and thus it may serve as a useful tool for quantitative assessment of the choroidal vascular reactivity.

Methods.: Before imaging, the intraocular pressure (IOP) was measured in the study eye of nine clinically healthy subjects (four males) while the subject performed the ventilatory protocol subsequently imaged by the scanner. End-tidal CO2 partial pressure (PETCO2) was increased to target 45 mm Hg, (baseline PETCO2 = 40 mm Hg and baseline PETO2 = 100 mm Hg). PETO2 was then increased to target 300 and 500 mm Hg while keeping PETCO2 constant at 45 mm Hg. A background-suppressed, pulsed-continuous ASL sequence was used for blood flow imaging.

Results.: The measured total blood flow increased significantly from 1.55 ± 0.17 μL/mm2/min at the baseline to 1.96 ± 0.18 μL/mm2/min during hypercarbia. With increasing PETO2, the measured blood flow did not change significantly relative to the hypercarbia condition but remained significantly elevated relative to the baseline. There were no significant changes in systolic, diastolic, or mean blood pressure, heart rate, or IOP during all four breathing conditions.

Conclusions.: The lack of change in the ASL signal under hyperoxic conditions is consistent with the hypothesis that this noninvasive assessment technique is predominantly weighted by choroidal blood flow. The results indicate that a CO2 provocation challenge, in combination with ASL-MRI, is a promising noninvasive approach for investigating choroidal vascular reactivity under normal and disease states.

Within the eye, there are two distinct vascular beds that differ with respect to function, responsivity, and structure. The inner (relative to vitreous) retinal layers are served by the inner retinal circulation which is supplied by the central retinal artery (CRA), and the outer retinal layers, including the photoreceptors, are served by the choroidal circulation, which is supplied by ciliary arteries. The choroid has the highest perfusion rate of any vascular bed within the human body, 1 being 10 to 15 times that of the cerebral cortex vasculature. The high choroidal blood flow reflects the extraordinary metabolic activity of the photoreceptors, providing for all the metabolic requirements of the foveal avascular zone (FAZ). Various studies 2 5 have shown that increases in the arterial partial pressure of O2 (Pao 2) promote local vasoconstriction of the arterioles, venules, and capillaries in the retina, whereas increases in the arterial partial pressure of CO2 (Paco 2) result in vasodilation in the retina. In the choroid, on the other hand, there is relatively little change in response to increases in Pao 2, whereas increases in Paco 2 result in vasodilation. 6 10  
Choroidal blood flow outside of FAZ cannot be readily studied using current quantitative optical imaging methods, 11 We have recently shown the feasibility of measuring the total blood flow to the retina in humans with arterial spin labeling-magnetic resonance imaging (ASL-MRI). 12 With this method, inflowing unlabeled arterial blood acts as an intrinsic contrast agent washing out magnetically labeled blood from a demarcated tissue region. The use of ASL for quantifying flow has been validated for the brain by comparisons with microspheres 13 and diffusible radiotracers 14 in animals and with positron emission tomography in humans, 15,16 and it has been used for studying retinal perfusion in animals. 17 19  
A new method of automated targeted control of end-tidal gases 20 has recently been shown to precisely target Paco 2. 21 A device based on these methods (Respiract; Thornhill Research, Inc. [TRI], Toronto, ON, Canada) has been used to study response characteristics of the changes in Paco 2 and Pao 2 of ocular vessels with respect to diameter and blood flow. 22 24 The method is also compatible with the MRI environment and has been used to provide stimuli for the study of cerebrovascular reactivity (CVR) with blood oxygen level–dependent (BOLD) and ASL-MRI. 25 28  
We hypothesized that the reactivity of the ASL-MR signal detected at the retina is dominated by the choroidal blood flow contribution, and thus it may serve as a useful tool for quantitative assessment of choroidal blood flow reactivity. To test this hypothesis, we used our ASL-MRI technique to measure the changes in retinal blood flow in response to hypercarbic and hyperoxic stimulation, to determine the relative contribution of the retinal and choroidal circulation to the ASL-MR signal. 
Material and Methods
In this article, we will refer to the partial pressures in arterial blood (PaCO2 and PaO2) when referencing the independent variable affecting tissue blood flow at the site of action and the end-tidal values (PETCO2, PETO2), when referencing measured variables. 
Subjects
The study protocol was approved by the Research Ethics Board of the University Health Network (UHN), and all subjects gave written informed consent after explanation of the nature and possible consequences of the study, according to the Declaration of Helsinki. Nine clinically healthy subjects of mean age 34 ± 7 years were recruited (four males). Exclusion criteria included smoking, history of cardiovascular disease, asthma, systemic hypertension, refractive error greater than ±6.00 DS and greater than ± 1.50 DC, any ocular pathology, family history of glaucoma or diabetes, or treatment with any systemic medication(s) in the past 2 weeks. Subjects were asked to refrain from caffeine at least 12 hours before the study, as caffeine may influence vascular tone in the brain and eye. 
Sample
The required sample size was calculated according to the effect size that was estimated by conducting a pilot study to determine the magnitude of the response to increased level of PETCO2 at normoxia. For this pilot study, four healthy subjects were recruited. The breathing protocol consisted of the following sequence: while maintaining normoxia (PETO2 = 100) and 4 minutes of normocarbia (PETCO2 = 40 mm Hg), followed by 4 minutes of hypercarbia (PETCO2 = 50 mm Hg), followed by 4 minutes of normocarbia. To calculate the sample size, two independent sources of variability were considered. First, the variability in response to hypercarbia, as measured with MRI in the pilot study described above. Second, the variability in the baseline blood flow among different normal subjects, as estimated from our previous MRI study data. 12 We found that a sample size of nine subjects was needed to identify the significance of flow differences between different states of the end-tidal gases. Sample size was calculated for an α of 5% and power of 90%. 
Recording Physiologic Data
Blood pressure and pulse rate were digitally sampled with an MRI-compatible monitoring instrument (Veris; Medrad Inc., Indianola, PA) throughout the imaging procedure. Systolic, diastolic, and mean arterial blood pressures were measured in 2-minute intervals, and blood oxygen saturation and heart rate were monitored continuously by blood oximetry. Before the MRI, intraocular pressure (IOP) was measured while the subjects performed the same gas control protocol. The IOP measurements were performed with the subject supine, the same position assumed while scanned in the MR scanner. After instillation of a topical anesthetic (one drop of tropicamide 1%) into the study eye, IOP was measured with a handheld tonometer (Tono-Pen XL; Medtronic Solan, Jacksonville, FL) at each phase of the gas provocation protocol. Both IOP measurement and MR imaging were performed after subjects rested for approximately 15 minutes to achieve stability of ocular perfusion pressure, heart rate, and blood pressure. 
Protocols
The protocol was performed in one visit. PETCO2 and PETO2 were attained using a series of prospective end-tidal gas-targeting algorithms 20,27 applied via a computer-controlled gas blender (Respiract; TRI). This method allows the independent accurate (±1 mm Hg) targeting of PETCO2 and PETO2, irrespective of minute ventilation, by blending specific source gases into the breathing circuit. 20 The subject breathed via a sequential gas delivery breathing circuit secured to the face in an air-tight manner via a mask (Fig. 1). PETCO2 was increased from the baseline normoxia (PETO2 100 mm Hg) and normocarbia (PETCO2 40 mm Hg), to normoxia and the target 45 mm Hg (hypercarbia). PetO2 was then increased to the target 300 mm Hg (hypercarbia/hyperoxia300) and 500 mm Hg (hypercarbia/hyperoxia500). During hyperoxia, PETCO2 was kept constant at 45 mm Hg. With respect to Pco 2, it has been shown that when an individual breathes via a sequential rebreathing circuit, PETCO2 provides an accurate measurement of arterial Pco 2 (PaCO2). 21 Subjects were exposed to the end-tidal gas challenges outside the imaging unit before the MRI study, to familiarize them with the breathing conditions. Gas was sampled continuously from inside the mask and analyzed for PETCO2 and PETO2. All the subjects easily tolerated the increase in the PETCO2 without any discomfort. 
Figure 1.
 
The sequential rebreathing system. The input (fresh) gas reservoir and the exhaled gas reservoir are connected to the face mask with separate one-way valves. The two reservoirs are connected via a positive end-expiratory pressure (PEEP) valve that allows the subject to rebreathe the exhaled gas when the fresh gas reservoir is emptied.
Figure 1.
 
The sequential rebreathing system. The input (fresh) gas reservoir and the exhaled gas reservoir are connected to the face mask with separate one-way valves. The two reservoirs are connected via a positive end-expiratory pressure (PEEP) valve that allows the subject to rebreathe the exhaled gas when the fresh gas reservoir is emptied.
MR Imaging
Studies were performed on a high-definition 3-Tesla MR system (GE Healthcare, Milwaukee, WI) with an 8-channel array head coil. A background-suppressed PCASL (BS-PCASL) sequence 29 was used for blood flow imaging. The timing of the BS-PCASL pulse sequence is shown in Figure 2. A near axial slice passing through the optic nerve heads was prescribed graphically based on the sagittal localizer image. Images were acquired with a 2D, half-Fourier, single-shot fast spin-echo (SSFSE) sequence. This imaging sequence is much less affected by the nonuniform magnetic fields in the orbits caused by air in the nearby sinuses than is the echoplanar imaging sequence most often used for brain blood flow imaging. Imaging parameters were field of view (FOV), 24 cm; matrix, 96 × 96; bandwidth, 20.83 kHz; and slice thickness, 10 mm. The linear ordered partial k-space acquisition used six extra echoes for a total echo train length of 54. The echo spacing was 5.2 ms. A long TR of 7000 ms was selected to minimize any effects of the inversion pulses on blood magnetization for the next repetition, and a minimum effective TE of 36.5 ms was chosen to minimize the T2 contribution to the signal intensity. The labeling slab was also in the axial plane, approximately 5 cm below the slice. This location labels blood in the carotid and other arteries below the Circle of Willis. Image pairs (label and control) were acquired at a postlabeling delay of 1.25 seconds every 16 seconds throughout the whole gas provocation protocol. Tissue T 1 and M 0 reference images were also obtained during imaging. Furthermore, to minimize systematic errors resulting from eye motion, the volunteers were asked to fix on a landmark and blink only during the silent periods of the sequence. 30 The noise and silent periods were practiced with the subjects before data acquisition. Images that contained eye movements were excluded during postacquisition analysis. At the end of the study, two images were acquired with all the labeling and background suppression pulses turned off. The average of these two images was used for normalizing the signal for blood flow quantification and to provide greater anatomic information. 12  
Figure 2.
 
The pulse-sequence timing of the background-suppressed PCASL sequence used for retinal perfusion imaging.
Figure 2.
 
The pulse-sequence timing of the background-suppressed PCASL sequence used for retinal perfusion imaging.
MR Data Analysis
All image data were saved as raw echo intensities and were transferred to a workstation for further processing. Images for each coil of the eight-channel phased array coil were reconstructed offline with custom tools developed in the IDL programming environment. The data for each coil were averaged for all the repetitions, and the resulting average was phase corrected. A near-optimal combination of the complex coil image data was achieved by sum of squares of individual coil images weighted by the coil sensitivity ratios derived from smoothing the non–background-suppressed image with a 9 × 9 Gaussian filter. 31 This method decreased the bias due to the rectification of negative noise contributions in the reconstructed perfusion images. Baseline detrending was not a necessary preprocessing step for ASL perfusion data as drift effects are minimized in ASL. This is primarily a result of successive pair-wise subtraction between images acquired with and without labeling. 32  
Images of all subjects were analyzed with the same procedure. The ASL perfusion images were co-registered to the anatomic reference scan and analyzed in ImageJ (http://rsb.info.nih.gov/ij/ developed by Wayne Rasband, National Institutes of Health, Bethesda, MD). For each breathing period, images free of motion artifacts were averaged, to improve the signal-to-noise ratio. Rectangular regions of interest (ROIs) were defined similarly for both eyes. An ROI of 1 to 1.25 cm in length along the retina (based on the subject) and 2.5 mm in width and 10 mm in depth (imaging slice thickness) centered on the fovea was defined. After definition of the retinal ROIs, they were overlaid on the reference anatomic image to confirm location at the posterior edge of the eye. 12 For each ROI, the flow was calculated as the flow per unit surface area to make the measurement independent of the ROI width and the resolution. Basically, by converting to μL/mm2/min, the impact of partial volume perpendicular to the retina was negated. 12 Microsphere studies 33 suggest a rapid decrease in flow with distance from the fovea, and there is partial volume impact in this direction that was uncorrected because of the absence of a model for this change. 
Blood flow was quantified in a single-compartment model, 12,34 in which the observed signal is considered to come from a well-mixed tissue compartment with intra- and extravascular water in perfect communication:   where M 0 is the equilibrium brain tissue magnetization; f is blood flow; λ is the blood/tissue water partition coefficient; R 1a is the longitudinal relaxation rate of blood; α is the inversion efficiency; R 1 is the longitudinal relaxation rate of the retinal tissue in the absence of blood flow; δ is the transit time from the labeling region to the tissue compartment (exchange time); τ is the labeling duration; and w is the delay introduced between the end of the labeling pulse and image acquisition, to minimize transit time artifacts. 35 We assumed a labeling efficiency of 0.85; however, as inversion pulses used for background suppression also attenuate the ASL signal, an additional loss of efficiency of 75% was accounted for in the flow quantification. 36 We assumed the average brain value of 0.9 g/mL to apply to the retina. 37 Although λ for the retina is uncertain, most neural and vascular tissues have λ between 0.8 and 1, 37 and so the error should not be extreme. We also used R 1a = 0.67 seconds−1. 38 The signal intensity in the retina was normalized to the intensity of the vitreous in an unsuppressed reference image. The water density of the vitreous is close to 100%, and so it serves as a reference for sensitivity to water. We assumed average values of δ = 1.1 second and R 1 = 0.76 seconds−1 for all subjects who were estimated in our previous study. 39  
Changes in blood flow values between respective ROI in the images obtained at baseline and the images obtained during the hyperoxic and hypercarbic inhalation challenges were compared by using analysis of variance (ANOVA) and the Tukey honest significant difference (HSD) post hoc test to determine the significance of any change between different states, at the level of P < 0.05 (analyses performed in Statistica 6.0; StatSoft, Tulsa, OK). 
Ventilation with high O2 concentrations has been shown to decrease T1 in blood. The decrease of T1 in blood depends strongly on the fraction of dissolved O2 in solution and is dependent only on the Po 2, and independent of the hemoglobin content. 40 Reduction of the blood relaxation rate as a result of increased Po 2 was modeled according to in vitro results reported in Janne d'Othée et al., 40 where the relationship between blood T1 and percentage of O2 in the gas mixture is found to be linear and almost independent of the field strength. At 8.45 T and 100% O2, there is a 22.5% reduction of the blood T1 value compared with the reduction of the blood T1 in room air (20% O2). At 1.5 T the reduction is 25.7%. We selected a similar reduction of 25.7% for the 3.0 T analysis. 
Gas Data Analysis
The analogue signals acquired from continuous monitoring of PETCO2 and PETO2 were digitized at 50 Hz and recorded. PETCO2 and PETO2 were obtained from the recorded data by selecting the peak values at the end of expiration. The resulting waveform was used for the rest of the analysis. To investigate the effect of PETCO2 and PETO2 on the blood flow, we calculated the differences between PETCO2 and PETO2 and baseline. End-tidal values from corresponding periods were compared with one-way ANOVA and Tukey HSD post hoc tests. All data were presented as the mean ± SD for each breathing period. 
Results
Respiratory Parameters
During hypercarbia, PETCO2 increased from baseline 39.3 ± 0.4 to 45.36 ± 0.3 mm Hg (P < 0.0001), a group mean increase of 15.4%. Once established, PETCO2 remained unchanged during each stage of hyperoxia, 45.4 ± 0.9 mm Hg at hypercarbia/hyperoxia300 and 45.6 ± 1.1 mm Hg at hypercarbia/hyperoxia500. Attained PETCO2 and PETO2 values are shown in Figure 3. All the targeted end-tidal values were achieved with minimal group variability. The largest variability was in PETO2 at hypercarbia/hyperoxia500 (481 ± 32.5 mm Hg) where most subjects were not able to reach or sustain a PETO2 of 500 mm Hg. 
Figure 3.
 
The group mean PETCO2 (left) and PETO2 (right) in response to normocarbia/normoxia (a), hypercarbia/normoxia (b), hypercarbia/hyperoxia300 (c), and hypercarbia/hyperoxia500 (d).
Figure 3.
 
The group mean PETCO2 (left) and PETO2 (right) in response to normocarbia/normoxia (a), hypercarbia/normoxia (b), hypercarbia/hyperoxia300 (c), and hypercarbia/hyperoxia500 (d).
Examples of the time courses of normalized PETCO2 and PETO2 changes from a representative subject are shown in Figure 4. The subject was 27 years old, and the plots represent the PETCO2 and PETO2 changes during the MR scan. The four breathing periods of normocarbia/normoxia, hypercarbia/normoxia, hypercarbia/hyperoxia300, and hypercarbia/hyperoxia500 are readily distinguishable in this plot. 
Figure 4.
 
Time course of normalized PETCO2 (left) and PETO2 (right) for a single subject.
Figure 4.
 
Time course of normalized PETCO2 (left) and PETO2 (right) for a single subject.
Systemic Hemodynamic Parameters
Systemic hemodynamic changes are summarized in Table 1. There was no significant change in systolic, diastolic, or mean blood pressure during all four breathing conditions. Heart rate did not change during any of the hypercarbia/hyperoxia states. 
Table 1.
 
Systemic Hemodynamic Changes with the Various Breathing Conditions
Table 1.
 
Systemic Hemodynamic Changes with the Various Breathing Conditions
Normocarbia/Normoxia Hypercarbia Hypercarbia/Hyperoxia300 Hypercarbia/Hyperoxia500
SP, mm Hg 103 (6) 105 (5) 105 (5) 104 (4)
DP, mm Hg 62 (5) 62 (9) 63 (6) 61 (6)
MAP, mm Hg 81 (7) 81 (8) 81 (7) 81 (5)
HR, beats/min 70 (8) 71 (8) 69 (8) 67 (8)
Vascular Reactivity
The measured total blood flow changed significantly with hypercarbia (P < 0.0001). Group mean blood flow changes in response to different hypercapnic/hyperoxic breathing conditions are shown in Figure 5. Blood flow increased from 1.55 ± 0.17 μL/mm2/min at the baseline to 1.96 ± 0.18 μL/mm2/min during hypercarbia. With increasing PETO2, blood flow did not change significantly relative to the hypercarbia condition, but remained significantly elevated relative to baseline. The difference between hypercarbia/hyperoxia300 (1.92 ± 0.21 μL/mm2/min) and hypercarbia/hyperoxia500 (1.84 ± 0.19 μL/mm2/min) was not significant either, although there was a trend toward a reduction in flow. The increase in flow between normoxia/normocarbia and hypercarbia/hyperoxia300 and between hypercarbia and hypercarbia/hyperoxia500 were significant (P < 0.0002). Hypercarbia caused a 26% relative increase (i.e., relative to baseline) in blood flow. There was a 6.7% increase in the blood flow per 1 mm Hg. There were no significant correlations between systolic, diastolic, or mean blood pressure and the blood flow. 
Figure 5.
 
The group mean blood flow values in response to normocarbia/normoxia (a), hypercarbia/normoxia (b), hypercarbia/hyperoxia 300 (c), and hypercarbia/hyperoxia 500 (d).
Figure 5.
 
The group mean blood flow values in response to normocarbia/normoxia (a), hypercarbia/normoxia (b), hypercarbia/hyperoxia 300 (c), and hypercarbia/hyperoxia 500 (d).
Figure 6 shows ASL perfusion images (ΔM) at baseline, hypercarbia/normoxia, hypercarbia/hyperoxia300, and hypercarbia/hyperoxia500 for a single subject. Flow and perfusion signals were linearly related by a constant, as described in equation 1. Therefore, these images are representative of blood flow changes during each of the breathing conditions as well. In these images, the signal is highest in the hypercarbia/normoxia condition and decreases during the next two hyperoxia states while hypercarbia is maintained. However, this signal reduction may not be entirely related to the reactivity response to the hyperoxic challenge, as the longitudinal relaxation rate in the blood depends strongly on the fraction of dissolved O2 in solution, and it declines as a result of increased Po 2
Figure 6.
 
Left to right: ASL perfusion-weighted images at baseline, hypercarbia/normoxia, hypercarbia/hyperoxia 300, and hypercarbia/hyperoxia500 in one subject.
Figure 6.
 
Left to right: ASL perfusion-weighted images at baseline, hypercarbia/normoxia, hypercarbia/hyperoxia 300, and hypercarbia/hyperoxia500 in one subject.
Figure 7 shows the blood flow values for the same subject calculated by modeling the reduction of blood relaxation rate as a result of increased Po 2 and including that in the model for the blood flow calculation. 
Figure 7.
 
Effect of correcting for the reduction of the blood T1 as a result of increased Po 2 on the estimation of blood flow in the retina of one subject during normocarbia/normoxia (a), hypercarbia/normoxia (b), hypercarbia/hyperoxia300 (c), and hypercarbia/hyperoxia500 (d).
Figure 7.
 
Effect of correcting for the reduction of the blood T1 as a result of increased Po 2 on the estimation of blood flow in the retina of one subject during normocarbia/normoxia (a), hypercarbia/normoxia (b), hypercarbia/hyperoxia300 (c), and hypercarbia/hyperoxia500 (d).
Figure 8 shows ASL perfusion images (ΔM) for a subject at normocarbia/normoxia, hypercarbia (50 mm Hg)/normoxia, and normocarbia/normoxia that were acquired during the pilot study. The increase in blood flow in the retina and the much more dramatic increase in the brain gray matter during hypercarbia is noticeable in these images. Normalized MR signal change and the corresponding PETCO2 signal change in an ROI in the gray matter of a subject are shown in Figure 9. Perfusion signal changes signifying vascular reactivity closely followed the changes in PETCO2
Figure 8.
 
Left to right: ASL perfusion images for a subject at baseline, during hypercarbic/normoxic challenge (PETCO2 = 50 mm Hg), and at return to the baseline conditions.
Figure 8.
 
Left to right: ASL perfusion images for a subject at baseline, during hypercarbic/normoxic challenge (PETCO2 = 50 mm Hg), and at return to the baseline conditions.
Figure 9.
 
Normalized PETCO2 and MR signal changes in an ROI in the gray matter of a single subject during hypercarbia challenge (PETCO2 = 50 mm Hg).
Figure 9.
 
Normalized PETCO2 and MR signal changes in an ROI in the gray matter of a single subject during hypercarbia challenge (PETCO2 = 50 mm Hg).
Intraocular Pressure
There was no change in the measured IOP, MAP, and ocular perfusion pressure values among the four breathing conditions. 
Discussion
This study showed that using a clinical MRI scanner and without a customized coil and ASL perfusion imaging can measure, in humans, the changes in the combined retinal and choroidal blood flow induced by hypercarbic and hyperoxic stimuli. The spatial resolution of the method is not sufficient to distinguish between the retinal and choroidal contributions to the ASL signal, and the detected signal is therefore a weighted combination of both. However, by provoking vascular reactivity in response to hypercarbic and hyperoxic stimuli and comparing the responses with the results from optical measurement, 24 we confirm the dominance of the choroidal circulation's contribution to the detected MR signal with background-suppressed PCASL. The measured total blood flow increased significantly from baseline during hypercarbia. With progressive hyperoxia, total blood flow did not change significantly relative to the hypercarbia condition, indicating the immeasurably small contribution of the reduction in inner retinal blood flow to the ASL signal. These findings are consistent with the relative CO2-responsiveness, and O2 insensitivity of choroidal vasculature, which appears to provide the dominant component of the ASL signal. 
Choroidal blood flow provides the bulk of the oxygen to the outer retina, and it is necessary for maintaining retinal oxygenation and for regulating the temperature of the retina and dissipating heat produced primarily by phototransduction and by the effect of incident light on the retina. 41,42 Choroidal blood flow is measurably responsive to changes in Pco 2, but shows relatively little reactivity to changes in Po 2. 6,10 The increase in choroidal blood flow in the FAZ region has been reported to be 1.5% per 1 mm Hg increase in Pco 2. 10 The choroidal circulation has at best a limited capability to respond to intrinsic O2 requirements, whereas the retinal circulation does effectively vasoconstrict in response to hyperoxia. 43,44 Increases in Pao 2 have been reported to decrease retinal blood flow by 30% using the Heidelberg Retina Flowmeter 45 and by 60% using laser Doppler flowmetry. 46,47  
The results of this study follow up on the findings of our earlier study where blood flow to the retina was estimated by repeated measurements with different ASL timing, and we attributed our findings to the dominance of choroidal blood flow. 12 Kisilevsky et al. 24 observed an almost complete reversal of the effect of hypercarbia/normoxia on the retinal blood flow with PETCO2 increased 23% above baseline by hypercarbia/hyperoxia300, which we did not observe with MR. They also found a significant difference between reactivity response of the retinal blood flow to the hypercarbia/hyperoxia300 and hypercarbia/hyperoxia500 conditions, which was not evident with our ASL-MRI technique. That we did see an increase in blood flow with hypercapnia gives us confidence that the lack of reduction in blood flow with the superimposed hyperoxia was not as a result of lack of resolution and therefore supports our conclusion that the detected ASL signal is predominantly weighted by choroidal blood flow. As our study size was determined based on the effect size measured by a previous pilot study and data reported by Kisilevsky et al., 24 we are confident that our study had sufficient statistical power for our conclusions. 
An important feature of our gas provocation protocol was to maintain isocapnia at hypercarbic levels during the hyperoxic states. The administration of oxygen results in hyperventilation and a reduction in PETCO2. 48 First, without maintenance of isocapnia, it would not be possible to attribute changes in ocular blood flow to the effects of PaCO2 or PaO2. Second, isocapnia was maintained at hypercarbic levels during the hyperoxic challenges, to make certain that any lack of constriction was not due to the vessels' already being maximally constricted at baseline. 24 Finally, by maintaining hyperoxia and hypercarbia simultaneously and independently with an accuracy of (±1 mm Hg), we were able to investigate the relative effect of both acting independently on the retinal and choroidal vasculature. 
Although the difference between hypercarbia/normoxia, hypercarbia/hyperoxia300, and hypercarbia/hyperoxia500 was not found to be significant, there was a trend toward a reduction in blood flow with increased PETO2 (Fig. 5). We used this trend, and the values reported by Kisilevsky et al., 24 to estimate the magnitude of inner retinal vasoconstriction in response to hyperoxic provocation to the overall signal intensity. The relative contribution of retinal and choroidal blood flow to the total estimated blood flow with our technique was found to be 5% and 95%, respectively. 
The change in the relaxation time of blood with increased levels of hyperoxia could confound the vascular reactivity measured by MRI. 27 Failure to accurately account for these changes when calculating blood flow with the ASL technique will lead to an overestimation of the changes induced by gas provocation. It can be seen in Figure 7 that without considering the effect of hyperoxygenation on blood T1, the flow changes would be significantly overestimated. Applying the same correction of the effect of hyperoxia on the estimation of the cerebral blood flow during the hyperoxic challenges showed that CBF decreased only slightly between 5% and 10% in the various subjects. Other studies have reported the same level of reduction in CBF in response to similar levels of hyperoxia. 49 More precise estimates of the relaxation times in arterial blood at different levels of hyperoxia should lead to more accurate reactivity estimates. 
We did not account for a change in retinal T1 during hyperoxia. There are two reasons for this: one is the expectation that T1 changes would be small and the other is uncertainty in the retinal oxygenation changes and the corresponding effects on relaxation. Hyperoxia greatly increases the oxygen tension in blood but does not proportionately increase the oxygen content of blood because of hemoglobin saturation. Hence, the changes in tissue oxygenation, which reflect a competition between inflowing oxygen content and oxidative metabolism, are more muted compared with the changes in arterial blood R 1. 50 This is highlighted for example in Tadamura et al. 51 where only tissues with very high blood content showed any detectable change in R 1 with hyperoxia. We are unaware of any direct demonstration of retinal T1 changes with hyperoxia, but the ability to detect retinal oxygenation changes by measuring changes in T1 of the nearby vitreous has been emphasized. 30 Since the vitreous is in close contact with the retinal blood supply, has a very long intrinsic T1, and very little intrinsic oxygen utilization, the effects of oxygen changes are amplified. 
A single-compartment model 34 was used to model the blood flow to the retina, in which the observed signal is considered to come from a well-mixed tissue compartment with intra- and extravascular water in perfect communication. Using a single-compartment model is a better approach for modeling blood flow to the retina because of the unique circulation system of retina and more specifically because of the choroidal circulation, which feeds the outer retinal layers. The innermost vascular layer of the choroid, the choriocapillaris, is composed of richly anastomotic, fenestrated capillaries beginning at the optic disc margin. On the other hand, choroidal blood flow is around 10 times higher than the inner retinal blood flow, and as such the ASL signal should be more dominated by the choroidal blood flow for which the rate of exchange of water between extra- and intravascular pools is very fast since it is richly fenestrated. 
The relaxation rate of the blood is dependent on the hematocrit fraction, temperature, and the oxygen saturation level of the blood, 38 and it could vary from subject to subject. The R 1a on 3.0 T, as a function of discussed variables, ranges between 0.57 and 0.67 seconds−1 38 under normal physiological conditions. We also used R 1a = 0.67 seconds−1. Lower R 1a values would lead to an approximate 17% increase in the estimated blood flow. As we corrected for the effect of oxygenation on the R 1a and the experimental conditions remain the same throughout the provocation testing, the reported changes in the reactivity results remain the same and just the baseline will be shifted which does not have any effect on the comparisons. The same argument holds true for other assumed values (δ = 1.1 second and R 1 = 0.76 seconds−1). 
Table 2.
 
Group mean IOP with the Various Breathing Conditions
Table 2.
 
Group mean IOP with the Various Breathing Conditions
Normocarbia/Normoxia Hypercarbia Hypercarbia/Hyperoxia300 Hypercarbia/Hyperoxia500
IOP, mm Hg 20 (2) 19 (2) 20 (2) 20 (1)
We used a 1.25-second delay for imaging the rational behind this choice is that in our prior study, 12 we measured the average transit time to the retina to be 1.1 second so a 1.25-second delay is longer and nearly optimal for imaging. Since we were imaging a comparable group of young volunteers, it was not necessary to make it longer. The transit time to the eye may well be shorter than that to many regions of the brain, where a longer postlabeling delay is necessary. 
Perfusion imaging by ASL can be highly sensitive to the transit time from the labeling site to the imaged tissue. Gonzalez-At et al. 52 have assessed the relative contribution of changes in arterial transit time and perfusion to the ASL images during motor and visual tasks. They have shown that the fractional transit time changes accompanying activation are approximately equal in magnitude and opposite in sign to fractional perfusion changes. Hypercarbic provocation, as a stimulus, could cause a similar effect in reducing transit time. We did not assess this effect experimentally in this study. However, we theoretically modeled transit delay reductions ranging from 10% to 90% in estimated flow values. We found that the maximum error in flow estimation, in the event that there is a reduction in transit time, would be only 2.5%. 
ASL-MRI in general is not immune to movement artifacts. The retina is at the edge of a large signal-intensity gradient between fat and vitreous. Very small movements can overwhelm the vascular effects under study. We used a fixation landmark and cued eye blink 12,30,53 to avoid eye movement. However, our effective solution to negate this problem was to add an optimized background-suppression module to our sequence, thereby facilitating successful imaging of retinal/choroidal blood flow. The suppressed static tissue signal was less than 2% of the unsuppressed signal. Inflowing labeled blood was minimally attenuated by the multiple-inversion scheme 54 of the background-suppression pulses. Images were acquired using a 2D, half-Fourier, SSFSE sequence. This imaging sequence is less affected by the nonuniform magnetic fields in the orbits caused by air in the nearby sinuses than is the echoplanar imaging sequence most often used for brain blood flow imaging. 
Detailed analysis of retinal and choroidal reactivity would be possible using stronger magnetic fields because of an improved signal-to-noise ratio, along with a dedicated surface coil that would permit high spatial resolution imaging and the discrimination of inner retinal and choroidal blood flow. In this study, an eight-channel array head coil was used for image acquisition at a 3.0-T magnet, a system that is not optimized for studying the eye. However, vascular layer–specific and differential responses of the retinal and choroidal blood flow to hyperoxia and hypercapnia have been reported in anesthetized rats at 4.7-T when using BOLD fMRI 55 and when using a continuous ASL technique at 7 T in rats. 17 Li et al., 17 using a small circular surface coil (inner diameter, ∼7 mm), have successfully differentiated retinal and choroidal blood flow. Both studies have found that inner retinal vessels and outer-layer choroidal vessels responded differently to the hyperoxic and hypercarbic inhalation stimuli, indicative of differential blood-flow regulation. Hyperoxia induces a larger BOLD response in the outer layer and hypercapnia induces a smaller BOLD response in the outer layer than in the inner layer. 55 The former effect is due to the reduction of the retinal blood flow that counteracts the BOLD signal increase from hyperoxia. The increased BOLD signal in response to hypercapnia results from the reduction of the fractional oxygen extraction and increased capillary and venous oxygen saturation as a consequence of vasodilation. Our study findings are in agreement with both of the discussed studies. The results of our study, however, show that on a clinical scanner and with an eight-channel array head coil, the choroidal reactivity to a proactive stimulation can be captured if an optimized background suppression–ASL sequence is used. This technique permits the noninvasive assessment of retinal/choroidal blood flow changes in humans directly and quantitatively. 
Conclusion
Measures of blood flow in the retina are currently derived primarily from optical techniques that relate dye transit time or velocity distributions to blood flow. These techniques have extended markedly our understanding of the vascular physiology of the retina and optic nerve in humans. For the choroidal circulation, however, a valid technique is lacking that reliably quantifies choroidal hemodynamics in terms of absolute blood flow units. Apart from indocyanine green angiography, the only other optical technique that has been widely used for choroidal blood flow measurement is laser Doppler flowmetry which is suitable for flow measurements only within the restricted region of the FAZ. 11 The present study indicates that arterial Pco 2 is an important determinant of choroidal blood flow in humans, whereas Po 2 has little impact on this flow. Our results indicate that a CO2 provocation challenge in combination with the ASL-MR perfusion imaging is a promising noninvasive approach to investigate choroidal vascular reactivity under normal and disease states, such as diabetic retinopathy, glaucoma, and age-related macular degeneration. 
Footnotes
 Disclosure: N. Maleki, None; D.C. Alsop, None; W. Dai, None; C. Hudson, P; J.S. Han, None; J. Fisher, P; D. Mikulis, None
References
Kaufman P Alm A eds. Adler's Physiology of the Eye: Clinical Application. 10th Edition. St. Louis: Mosby; 2003.
Johnson PC . Autoregulation of blood flow. Circ Res. 1986;59:483–495. [CrossRef] [PubMed]
Riva CE Falsini B Logean E . Flicker-evoked responses of human optic nerve head blood flow: luminance versus chromatic modulation. Invest Ophthalmol Vis Sci. 2001;42:756–762. [PubMed]
Dorner GT Garhofer G Zawinka C Kiss B Schmetterer L . Response of retinal blood flow to CO2-breathing in humans. Eur J Ophthalmol. 2002;12:459–466. [PubMed]
Michelson G Patzelt A Harazny J . Flickering light increases retinal blood flow. Retina. 2002;22:336–343. [CrossRef] [PubMed]
Alm A Bill A . The oxygen supply to the retina. II. Effects of high intraocular pressure and of increased arterial carbon dioxide tension on uveal and retinal blood flow in cats. Acta Physiol Scand. 1972;84:306–319. [CrossRef] [PubMed]
Friedman E Chandra SR . Choroidal blood flow III. Effects of oxygen and carbon dioxide. Arch Ophthalmol. 1972;87:70–71. [CrossRef] [PubMed]
Riva CE Cranstoun SD Grunwald JE Petrig BL . Choroidal blood flow in the foveal region of the human ocular fundus. Invest Ophthalmol Vis Sci. 1994;35:4273–4281. [PubMed]
Riva CE Cranstoun SD Mann R Barnes GE . Local choroidal blood flow in the cat by laser Doppler flowmetry. Invest Ophthalmol Vis Sci. 1994;35:608–618. [PubMed]
Geiser MH Riva CE Dorner GT Diermann U Luksch A Schmetterer L . Response of choroidal blood flow in the foveal region to hyperoxia and hyperoxia-hypercapnia. Curr Eye Res. 2000;21(2):669–676. [CrossRef] [PubMed]
Harris A Kagemann L Ehrlich R Rospigliosi C Moore D Siesky B . Measuring and interpreting ocular blood flow and metabolism in glaucoma. Can J Ophthalmol. 2008;43:1–9. [CrossRef]
Maleki N Dai W Alsop DC . Quantification of retinal blood flow using a pseudo-continuous arterial spin labeling technique. NMR Biomed. 2011;24(1):104–111. [CrossRef] [PubMed]
Walsh EG Minematsu K Leppo J Moore SC . Radioactive microsphere validation of a volume localized continuous saturation perfusion measurement. Magn Reson Med. 1994;31:147–153. [CrossRef] [PubMed]
Ewing JR Wei L Knight RA . Direct comparison of local cerebral blood flow rates measured by MRI arterial spin-tagging and quantitative autoradiography in a rat model of experimental cerebral ischemia. J Cereb Blood Flow Metab. 2003;23(2):198–209. [CrossRef] [PubMed]
Ye FQ Berman KF Ellmore T . H215O PET validation of Steady-state arterial spin tagging cerebral blood flow measurements in humans. Magn Reson Med. 2000;44:450–456. [CrossRef] [PubMed]
Kimura H Kado H Koshimoto Y . Multislice continuous arterial spin-labeled perfusion MRI in patients with chronic occlusive cerebrovascular disease: a correlative study with CO2 PET validation. J Magn Reson Imaging. 2004;22(2):189–198. [CrossRef]
Li Y Cheng H Duong T . Blood-flow magnetic resonance imaging of the retina. Neuroimage. 2008;39(4):1744–1751. [CrossRef] [PubMed]
Li Y Cheng H Shen Q . Blood flow magnetic resonance imaging of retinal degeneration. Invest Ophthalmol Vis Sci. 2009;50(4):1824–1830. [CrossRef] [PubMed]
Muir ER Duong TQ . MRI of retinal and choroidal blood flow with laminar resolution. NMR Biomed. 2011;24:216–223. [CrossRef] [PubMed]
Slessarev M Han J Mardimae A . Prospective targeting and control of end-tidal CO2 and O2 concentrations. J Physiol. 2007;581:1207–1219. [CrossRef] [PubMed]
Ito S Mardimae A Han J . Non-invasive prospective targeting of arterial PCO2 in subjects at rest. J Physiol. 2008;15:3675–3682. [CrossRef]
Gilmore ED Hudson C Preiss D Fisher J . Retinal arteriolar diameter, blood velocity, and blood flow response to an isocapnic hyperoxic provocation. Am J Physiol Heart Circ Physiol. 2005;288:H2912–H2917. [CrossRef] [PubMed]
Venkataraman ST Hudson C Fisher JA Flanagan JG . The impact of hypercapnia on retinal capillary blood flow assessed by scanning laser Doppler flowmetry. Microvasc Res. 2005;69:149–155. [CrossRef] [PubMed]
Kisilevsky M Hudson C Mardimae A Wong T Fisher J . Concentration-dependent vasoconstrictive effect of hyperoxia on hypercarbia-dilated retinal arterioles. Microvasc Res. 2008;75(2):263–268. [CrossRef] [PubMed]
Vesely A Sasano H Volgyesi G Somogyi R . MRI mapping of cerebrovascular reactivity using square wave changes in end-tidal PCO2. Magn Reson Med. 2001;45:1011–1013. [CrossRef] [PubMed]
Mikulis DJ Krolczyk G Desal H . Preoperative and postoperative mapping of cerebrovascular reactivity in moyamoya disease by using blood oxygen level-dependent magnetic resonance imaging. J Neurosurg. 2005;103(2):347–355. [CrossRef] [PubMed]
Prisman E Slessarev M Han J . Comparison of the effects of independently-controlled end-tidal PCO(2) and PO(2) on blood oxygen level-dependent (BOLD) MRI. J Magn Reson Imaging. 2008;27(1):185–191. [CrossRef] [PubMed]
Mandell DM Han JS Poublanc J . Mapping cerebrovascular reactivity using blood oxygen level-dependent MRI in Patients with arterial steno-occlusive disease: comparison with arterial spin labeling MRI. Stroke. 2008;39(7):2021–2028. [CrossRef] [PubMed]
Dai W Garcia D de Bazelaire C Alsop DC , Continuous flow driven inversion for arterial spin labeling using pulsed radiofrequency and gradient fields. Magn Reson Med. 2008;60(6):1488–1497. [CrossRef] [PubMed]
Berkowitz BA McDonald C Ito Y Tofts PS Latif Z Gross J . Measuring the human retinal oxygenation response to a hyperoxic challenge using MRI: eliminating blinking artifacts and demonstrating proof of concept. Magn Reson Med. 2001;46:412–416. [CrossRef] [PubMed]
Bydder M Larkman DJ Hajnal JV . Combination of signals from array coils using image-based estimation of coil sensitivity profiles. Magn Reson Med. 2002;47(3):539–548. [CrossRef] [PubMed]
Aguirre GK Detre JA Zarahn E Alsop DC . Experimental design and the relative sensitivity of BOLD and perfusion fMRI. Neuroimage. 2002;15(3):488–500. [CrossRef] [PubMed]
Nork MT Kim CY Shanmuganayagam D Van Lysel MS Ver Hoeve N Folts JD . Measurement of retinal choroidal blood flow in rabbits and monkeys using fluorescent microspheres. Arch Ophthalmol. 2006;124:860–868. [CrossRef] [PubMed]
Buxton RB Frank LR Wong EC Siewert B Warach S Edelman RR . A general kinetic model for quantitative perfusion imaging with arterial spin labeling. Magn Reson Med. 1998;40:383–396. [CrossRef] [PubMed]
Alsop DC Detre JA . Reduced transit-time sensitivity in noninvasive magnetic resonance imaging of human cerebral blood flow. J Cereb Blood Flow Metab. 1996;16:1236–1249. [CrossRef] [PubMed]
Garcia DM Duhamel G Alsop DC . Efficiency of inversion pulses for background suppressed arterial spin labeling. Magn Reson Med. 2005a;54, 366–372. [CrossRef]
Herscovitch P Raichle ME . What is the correct value for the brain-blood partition coefficient for water? J Cereb Blood Flow Metab. 1985;5:65–69. [CrossRef] [PubMed]
Lu H Clingman C Golay X van Zijl PC . Determining the longitudinal relaxation time (T1) of blood at 3.0 Tesla. Magn Reson Med. 2004;52:679–682. [CrossRef] [PubMed]
Maleki N Dai W Alsop DC . Blood flow quantification of the human retina with MRI. NMR Biomed. 2008;24:104–111. [CrossRef]
Janne d'Othée B Rachmuth G Munasinghe J Lang EV . The effect of hyperoxygenation on T1 relaxation time in vitro. Acad Radiol. 2003;10(8):854–860. [CrossRef] [PubMed]
Parver LM . Choroidal blood flow as a heat dissipating mechanism in the macula Am J Ophthalmol. 1980;89:641–646. [CrossRef] [PubMed]
Parver LM Auker CR Carpenter DO . Choroidal blood flow. Arch Ophthalmol. 1982;100:1327–1330. [CrossRef] [PubMed]
Kiel JW van Heuven WA . Ocular perfusion pressure and choroidal blood flow in the rabbit. Invest Ophthalmol Vis Sci. 1995;36:579–585. [PubMed]
Kiel JW Shepherd AP . Autoregulation of choroidal blood flow in the rabbit. Invest Ophthalmol Vis Sci. 1992;33:2399–2410. [PubMed]
Sternn K Manapace R Rainer G . Reproducibility and sensitivity of scanning laser Doppler flowmetry using graded changes in PO2. Br J Ophthalmol. 1997;81:360–364. [CrossRef] [PubMed]
Trokel S . Effect of respiratory gases upon choroidal hemodynamics. Arch Ophthalmol. 1965;73:838–842. [CrossRef] [PubMed]
Riva CE Grunwald JE Sinclair SH . Laser Doppler velocimetry study of the effect of pure oxygen breathing on retinal blood flow. Invest Ophthalmol Visual Sci. 1983;24:47–51.
Becker HF Polo O McNamara SG Berthon-Jones M Sullivan CE . Effect of different levels of hyperoxia on breathing in healthy subjects. J Appl Physiol. 1996;81(4):1683–1690. [PubMed]
Bulte DP Chiarelli PA Wise RG Jezzard P . Cerebral perfusion response to hyperoxia. J Cereb Blood Flow Metab. 2007;27(1):69–75. [CrossRef] [PubMed]
Berkowitz BA . Role of dissolved plasma oxygen in hyperoxia-induced contrast. Magn Reson Imaging. 1997;15(1):123–126. [CrossRef] [PubMed]
Tadamura E Hatabu H Li W Prasad PV Edelman RR . Effect of oxygen inhalation on relaxation times in various tissues. J Magn Reson Imaging. 1997;7(1):220–225. [CrossRef] [PubMed]
Gonzalez-At JB Alsop DC Detre JA . Cerebral perfusion and arterial transit time changes during task activation determined with continuous arterial spin labeling. Magn Reson Med. 2000;43(5):739–746. [CrossRef] [PubMed]
Zhang Y Peng Q Kiel JW Rosende CA Duong TQ . Magnetic resonance imaging of vascular oxygenation changes during hyperoxia and carbogen challenges in the human retina. Invest Ophthalmol Vis Sci. 2010;52:286–291. [CrossRef]
Garcia DM de Bazelaire C Alsop DC . Pseudo-continuous flow driven adiabatic inversion for arterial spin labeling. Proceedings of the 13th Scientific Meeting of the International Society for Magnetic Resonance in Medicine. Berkeley, CA: ISMRM; 2005:37.
Cheng H Nair G Walker TA . Structural and functional MRI reveals multiple retinal layers. Proc Natl Acad Sci U S A. 2006;103(46):17525–17530. [CrossRef] [PubMed]
Figure 1.
 
The sequential rebreathing system. The input (fresh) gas reservoir and the exhaled gas reservoir are connected to the face mask with separate one-way valves. The two reservoirs are connected via a positive end-expiratory pressure (PEEP) valve that allows the subject to rebreathe the exhaled gas when the fresh gas reservoir is emptied.
Figure 1.
 
The sequential rebreathing system. The input (fresh) gas reservoir and the exhaled gas reservoir are connected to the face mask with separate one-way valves. The two reservoirs are connected via a positive end-expiratory pressure (PEEP) valve that allows the subject to rebreathe the exhaled gas when the fresh gas reservoir is emptied.
Figure 2.
 
The pulse-sequence timing of the background-suppressed PCASL sequence used for retinal perfusion imaging.
Figure 2.
 
The pulse-sequence timing of the background-suppressed PCASL sequence used for retinal perfusion imaging.
Figure 3.
 
The group mean PETCO2 (left) and PETO2 (right) in response to normocarbia/normoxia (a), hypercarbia/normoxia (b), hypercarbia/hyperoxia300 (c), and hypercarbia/hyperoxia500 (d).
Figure 3.
 
The group mean PETCO2 (left) and PETO2 (right) in response to normocarbia/normoxia (a), hypercarbia/normoxia (b), hypercarbia/hyperoxia300 (c), and hypercarbia/hyperoxia500 (d).
Figure 4.
 
Time course of normalized PETCO2 (left) and PETO2 (right) for a single subject.
Figure 4.
 
Time course of normalized PETCO2 (left) and PETO2 (right) for a single subject.
Figure 5.
 
The group mean blood flow values in response to normocarbia/normoxia (a), hypercarbia/normoxia (b), hypercarbia/hyperoxia 300 (c), and hypercarbia/hyperoxia 500 (d).
Figure 5.
 
The group mean blood flow values in response to normocarbia/normoxia (a), hypercarbia/normoxia (b), hypercarbia/hyperoxia 300 (c), and hypercarbia/hyperoxia 500 (d).
Figure 6.
 
Left to right: ASL perfusion-weighted images at baseline, hypercarbia/normoxia, hypercarbia/hyperoxia 300, and hypercarbia/hyperoxia500 in one subject.
Figure 6.
 
Left to right: ASL perfusion-weighted images at baseline, hypercarbia/normoxia, hypercarbia/hyperoxia 300, and hypercarbia/hyperoxia500 in one subject.
Figure 7.
 
Effect of correcting for the reduction of the blood T1 as a result of increased Po 2 on the estimation of blood flow in the retina of one subject during normocarbia/normoxia (a), hypercarbia/normoxia (b), hypercarbia/hyperoxia300 (c), and hypercarbia/hyperoxia500 (d).
Figure 7.
 
Effect of correcting for the reduction of the blood T1 as a result of increased Po 2 on the estimation of blood flow in the retina of one subject during normocarbia/normoxia (a), hypercarbia/normoxia (b), hypercarbia/hyperoxia300 (c), and hypercarbia/hyperoxia500 (d).
Figure 8.
 
Left to right: ASL perfusion images for a subject at baseline, during hypercarbic/normoxic challenge (PETCO2 = 50 mm Hg), and at return to the baseline conditions.
Figure 8.
 
Left to right: ASL perfusion images for a subject at baseline, during hypercarbic/normoxic challenge (PETCO2 = 50 mm Hg), and at return to the baseline conditions.
Figure 9.
 
Normalized PETCO2 and MR signal changes in an ROI in the gray matter of a single subject during hypercarbia challenge (PETCO2 = 50 mm Hg).
Figure 9.
 
Normalized PETCO2 and MR signal changes in an ROI in the gray matter of a single subject during hypercarbia challenge (PETCO2 = 50 mm Hg).
Table 1.
 
Systemic Hemodynamic Changes with the Various Breathing Conditions
Table 1.
 
Systemic Hemodynamic Changes with the Various Breathing Conditions
Normocarbia/Normoxia Hypercarbia Hypercarbia/Hyperoxia300 Hypercarbia/Hyperoxia500
SP, mm Hg 103 (6) 105 (5) 105 (5) 104 (4)
DP, mm Hg 62 (5) 62 (9) 63 (6) 61 (6)
MAP, mm Hg 81 (7) 81 (8) 81 (7) 81 (5)
HR, beats/min 70 (8) 71 (8) 69 (8) 67 (8)
Table 2.
 
Group mean IOP with the Various Breathing Conditions
Table 2.
 
Group mean IOP with the Various Breathing Conditions
Normocarbia/Normoxia Hypercarbia Hypercarbia/Hyperoxia300 Hypercarbia/Hyperoxia500
IOP, mm Hg 20 (2) 19 (2) 20 (2) 20 (1)
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×