Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Jan 15.
Published in final edited form as: Neuroimage. 2010 Sep 17;54(2):1063–1069. doi: 10.1016/j.neuroimage.2010.09.020

MRI Reveals Differential Regulation of Retinal and Choroidal Blood Volume in Rat Retina

Govind Nair 1,2, Yoji Tanaka 2, Moon Kim 5, Darin E Olson 3,5, Peter M Thulé 3,5, Machelle T Pardue 4,5, Timothy Q Duong 6,7
PMCID: PMC3190232  NIHMSID: NIHMS237988  PMID: 20850550

Abstract

The retina is nourished by two unique (retinal and choroidal) circulations. The lack of depth-resolved blood-volume (BV) imaging techniques hampers investigation of vascular-specific regulation of the retina in vivo. This study presents a high-resolution, laminar-specific magnetic resonance imaging (MRI) study to image retinal and choroidal BV, their responses to physiologic challenges in normal and Royal-College-of-Surgeons (RCS) rats (a model of retinal degeneration). Retinal and choroidal BV's were imaged by MRI (3×3×800μm) with intravascular administration of monocrystalline iron oxide nanocolloid (MION) contrast agent. Relative baseline BV and BV changes due to physiologic challenges were calculated in normal and RCS rat retinas. BV-MRI revealed two well-resolved retinal and choroidal vascular layers located on either side of the retina and an intervening avascular layer. The ratio of choroidal:retinal BV in normal rats at baseline was 9.8±3.2 in control rat retinas (N=7). Hyperoxia decreased retinal BV (-51±17%, p<0.05) more than choroidal BV (-28±14%), and hypercapnia increased retinal BV (52±11%, p<0.01) more than choroidal BV (12±11%). BV-MRI in degenerated retinas of RCS rats (N=7) revealed thinning of the avascular layer and an increase in relative baseline retinal and choroidal BV. Only hypercapnia-induced BV changes in the retinal vasculature of RCS rats were significantly different (smaller) from controls (p<0.05). These findings suggest that BV in both retinal vasculatures is regulated. The relative baseline BV in both vasculatures increased in retinal degeneration. BV-MRI provides clinically relevant data that may prove useful for early detection and longitudinal probing of retinal diseases, and could complement optical imaging techniques.

Introduction

The retina is nourished by two distinct – retinal and choroidal – vasculatures. Retinal vessels – coursing on the surface of the retina, within the ganglion cell layer, inner plexiform layer and inner nuclear layer – are characterized by relatively low blood flow (BF) and a large arterio-venous pO2 difference (similar to the brain). Choroidal vessels – located external to the photoreceptor layer, and sandwiched between the retinal pigment epithelium and sclera – are characterized by relatively high BF and a low arterio-venous pO2 difference. The outer nuclear layer and the photoreceptor segments are avascular (Harris et al., 1998). Importantly, sympathetic autonomous innervations seem to play a significant role in choroidal BF regulation, while the retinal vasculature is autoregulated in a manner similar to cerebral vessels (Laties, 1967; Steinle et al., 2000a). Microsphere measurements showed that basal choroidal BF is many times larger than retinal BF (Bill, 1984). Analysis by oxygen polarographic electrodes showed that tissue oxygenation profiles differ between the two vasculatures in responses to physiologic challenges (Yu et al., 2000). These findings suggest divergent regulation of retinal and choroidal vessels. Moreover, retinal and choroidal regulation may respond differently to various retinal diseases, such as retinal ischemia, diabetic retinopathy, and retinitis pigmentosa.

The lack of non-invasive, depth-resolved BF or blood-volume (BV) imaging techniques has limited investigation of hemodynamic regulation in the two vasculatures in vivo. Fluorescein angiography, indocyanin-green angiography, and laser Doppler flowmetry (LDF) (Riva et al., 1994) are capable of providing information on BF velocity and relative BF in the retina. Intrinsic optical imaging can measure relative BV changes in the retina (Nelson et al., 2005). Optical coherence tomography can visualize blood velocity in large retinal and choroidal vessels (Drexler and Fujimoto, 2008). While these optical approaches have contributed remarkably to our understanding of retinal pathophysiology, they cannot unambiguously resolve retinal and choroidal BV or BF at the capillary/tissue level. Optical techniques are also often hampered by disease-induced opacity of the vitreous humor, cornea and lens.

In contrast, magnetic resonance imaging (MRI) provides in vivo structural, physiological (i.e., BF, BV and oxygenation) and functional information without depth limitation. Although the spatial resolution of MRI is low compared to optical methods, advances in MRI technologies have made possible high resolution MRI of the retina. Anatomical (Berkowitz et al., 2006; Shen et al., 2006), BF (Li et al., 2008), oxygenation (Duong et al., 2002), and functional (Duong et al., 2002) MRI of the retina, which is ∼270 μm thick including the choroid (Cheng et al., 2006), have been reported. A recent blood-oxygenation-level-dependent (BOLD) functional MRI (fMRI) study reveals differential responses of the two vasculatures when challenged with hyperoxia or hypercapnia (Cheng et al., 2006). During hyperoxic challenge, choroidal BOLD responses were larger than retinal responses. In contrast, during hypercapnic challenge, the BOLD choroidal responses were minimal, while retinal responses were large. However, because the BOLD signal is a convolution of competing changes in oxygen metabolism, including BV, and BF, among other measurement parameters, deciphering the root causes of altered BOLD signals remains difficult (Ogawa et al., 1993). Direct measurement of a single physiological parameter (such as BV and BF which are tightly coupled to each other under normal physiological conditions) may illuminate the phenomenon underlying differential vascular-specific responses in the retina.

The present study describes a high-resolution (30×30×800μm) BV MRI approach using a blood-pool MRI contrast agent – monocrystalline iron oxide nanocolloid (MION) (Mandeville et al., 1998) which reduces blood water signals via enhanced magnetic susceptibility and thus allows highly sensitive measurement of total relative BV at the tissue level without the need to visualize individual vessels. BV MRI was applied to investigate layer-specific baseline BV and BV fMRI changes associated with hyperoxic and hypercapnic challenges in normal rat retina and in a retinal degeneration model – the Royal College of Surgeons (RCS) rat (Dowling and Sidman, 1962) which has a genetic defect shared by many patients with autosomal-recessive retinitis pigmentosa (Gal et al., 2000).

Materials and Methods

Animal preparations

Experiments were performed on normal adult male Long Evans rats (350-450 g, N = 7) and RCS rats (350-450 g, N = 7). Femoral vein was catheterized under 2% isoflurane for the administration of the intravascular BV contrast agent, MION (5 mg/kg), and a femoral artery was catheterized for blood-pressure monitoring. The animal was placed in a MRI-compatible stereotaxic headset with a feedback-regulated warm-water circulating pad. A thin layer of methylcellulose was applied to the corneal surface to prevent desiccation. During MRI, the animal was maintained with ∼1% isoflurane anesthesia, mechanically ventilated, and paralyzed with pancuronium bromide (1mg/kg/hr, ip). End-tidal CO2 (Surgivet capnometer), heart rate and arterial oxygen saturation (Nonin-8600), and rectal temperature (Digisense from Cole Palmer) were maintained within normal physiological ranges unless otherwise perturbed. This protocol yielded stable animal preparation for prolonged multiple measurements (Cheng et al., 2006; Li et al., 2008).

Inhalation stimuli

Hyperoxic (100% O2) and hypercapnic (5% CO2, 21% O2, balance N2) challenges were used to modulate BV. Ambient air was used as baseline. Images were acquired continuously for 6 mins during baseline and 6 mins during hyperoxic or hypercapnic challenge. A break of 10-15 minutes was given between each stimulus. This break has been shown previously to be more than sufficient for the systemic circulation to return to baseline as demonstrated by MRI monitoring of BF and oxygenation (i.e., BOLD), exhaled O2 and CO2 monitoring as well as blood-gas measurements in rat brain studies. Typically, two trials of both hyperoxia and hypercapnia were studied on the same animal and the presentation order of different physiologic challenges was random, with the entire study lasting ∼4 hours including animal preparations. The relative baseline BV was calculated from the baseline measurement of hypercapnia and hyperoxia, before and after MION injection. The percentage change in BV during stimulus was calculated from images acquired during baseline and gas challenge, before and after MION administration.

MRI methods

MRI studies were performed on a Bruker 7-Tesla/30-cm magnet and a 40 G/cm B-GA12 gradient insert (Billerica, MA). A small circular surface coil (inner diameter ∼7 mm) was placed on the left eye. Magnetic field homogeneity was optimized on an isotropic voxel encompassing the entire eye. A single sagittal imaging slice bisecting the center of the eye and the optic nerve head was used to minimize partial-volume effect. Serial BV MRI was acquired before and after MION injection using T2*-weighted images with a gradient-echo sequence, 200 ms repetition time (TR), 6.5 ms echo time (TE), 7.7 × 7.7 mm field of view, 25×256 matrix giving an in-plane resolution of 30×30 μm, and 800 μm slice thickness.

Histology

Standard histology of the retina was analyzed on slices carefully chosen to match the MRI slices. Eyes were enucleated following anesthetic overdose and the sagittal plane marked on the eyeball using permanent marker. The eyes were immersion fixed overnight in 2% paraformaldehyde/ 2% glutaraldehyde and subsequently rinsed in 0.1 M phosphate buffer, dissected to isolate the posterior eyecup, divided into two halves along the along the sagittal plane, embedded in epoxy-resin, and sectioned at 5 microns for toluidine blue staining. Images of the sections at the level of the optic nerve, which approximately corresponded to the MRI slices, were captured under 2× magnification using an image analysis program (Image Pro, Cybernetics).

Data analysis

Image analysis employed codes written in Matlab (MathWorks Inc) and STIMULATE software (University of Minnesota). Images were corrected for possible motion and drift (Cheng et al., 2006). BV index was calculated pixel-by-pixel from T2*-weighted MRI before and after MION injection, as the total concentration of MION in a pixel, and therefore the change in R2* after injection, was proportional to the blood volume in the pixel (Mandeville et al., 1998). Changes in transverse relaxation rate (Δ(1/T2*) or ΔR2*) acquisitions were calculated as ΔR2* = -ln(S/So)/TE, where S/So is the signal relative to the value before MION injection, and TE the echo time.

The magnitude BV change associated with stimulation was calculated from images before and after stimulus as ΔBV = (ΔR2*[post-MION] - ΔR2*[pre-MION])/ΔR2*[pre-MION], where the pre-MION signal change arises from the stimulus-evoked BOLD (deoxyhemoglobin) effect (Mandeville et al., 1998) and the post-MION signal change arises from both the stimulus-evoked MION and BOLD effects. Magnitude changes assumed that blood MION concentrations between animals were similar, which was achieved by using weight based dosage (5 mg/kg) in rats of similar size.

To minimize errors due to partial-volume, automated profile analysis (Cheng et al., 2006) of ΔR2* image (blood-volume index) was performed, instead of region-of-interest analysis. The retina was first detected using an edge-detection technique. Radial projections perpendicular to the vitreous boundary were then obtained with (3×) spatial interpolation. Some interpolations were necessary for automation and such spatial interpolation does not significantly alter the magnitude or width of the signals being measured (Cheng et al., 2006). BV was determined as a function of distance from the optic nerve head. BV profiles were also plotted across the retinal thickness. The peak ΔR2* profile was taken as BV index, and the thickness of individual peaks at half-maximum height (full-width-at-half-maximum, FWHM) was determined to be the layer thickness. Blood volume percent changes (i.e., changes in ΔR2*) due to hyperoxia and hypercapnia relative to baseline (air) were tabulated. All reported values and error bars on graphs are mean ± SD. A two-tailed student's t-test was performed for statistical comparison between groups. Paired t-test was used for comparison of choroidal to retinal BV and unpaired t-test was used for comparison between control and RCS groups. A p < 0.05 was taken to be statistically significant.

Results

Rat physiological parameters (body temperature, respiration rate, blood pressure, pO2, and pCO2) were maintained within normal values over the course of the experiment, unless intentionally perturbed using hyperoxic or hypercapnic challenge. There were no significant changes to baseline physiologic parameters due to the injection of MION. The end-tidal CO2 (EtCO2) measured before the injection of MION (30 ± 2 mmHg) was not different from that measured after the injection (29.9 ± 1.9 mmHg, p > 0.9). The mean arterial blood pressure before MION administration (MABP: 109 ± 20 mmHg) was not statistically different from after the MION injection (115 ± 11 mmHg, p > 0.8). Similarly, heart rate was not statistically different before and after the MION injection (397 ± 37 bpm before versus 406 ± 37 after, p > 0.6). The EtCO2 decreased by 10 ± 3% (p < 0.01) during hyperoxia and increased by 148 ± 2% during hypercapnic challenges, as expected (p < 0.01). There were no significant changes in any other measured physiologic parameter measured during the systemic gas challenges. Rats were allowed to recover from inhalation for 10-15 min after each physiologic stimulus.

Basal BV in control and P90 RCS rats

Figure 1 shows a T2*-weighted MRI before and after intravenous MION injection in a normal rat, and the calculated relative baseline BV image. T2*-weighted MRI signals on either side of the retina were attenuated after MION injection. The relative basal BV image, derived from changes in the spin-spin relaxation rate constant (ΔR2*) before and after MION injection, delineated the high BV signals on either side of the retina, corresponding to the retinal and choroidal vascular layers. The middle avascular layer in the retina, lens and vitreous or aqueous humor showed no significant BV signals. However, significant BV signals were detected in the iris, ciliary body and optic nerve head as expected because these structures have substantial blood flow. Some signal contrasts were observed on the corneal surface, likely due to subtraction errors in regions of high magnetic susceptibility artifacts.

Figure 1.

Figure 1

Open in a new tab

T2*-weighted MRI at 30×30 μm in-plane resolution of a rat eye (a) before and (b) after MION from a representative normal rat. (c) Relative baseline blood volume (ΔR2*) map clearly delineates the retinal and choroidal vascular layer and the avascular layer in between. Arrow indicates the location of the optic nerve head (ONH).

Figure 2 shows the relative baseline BV as a function of distance from the optic nerve head in a normal rat. Choroidal BV was relatively uniform along the length of the retina with no significant difference between the central and peripheral retina. In contrast, retinal BV was maximal around the optic nerve head. To avoid averaging over heterogeneous regions, group data were obtained from two regions of ∼0.3 mm each at 1 mm superior and inferior to the optic nerve head (see Figure 3a, inset).

Figure 2.

Figure 2

Open in a new tab

Layer-specific baseline blood volume as a function of distance from the optic nerve head (ONH) from a representative control rat. Choroidal blood volume was substantially higher than retinal blood volume. The retinal profile peaks at the optic nerve head whereas the choroidal profile is relatively uniform. Data were obtained from superior to inferior retina (point a to b).

Figure 3.

Figure 3

Open in a new tab

(a) Profile plots of the baseline blood volume across the retinal thickness from a representative control and Royal-College-of-Surgeons (RCS) rats at P90. Inset: lines indicate radial projections across the retina. To avoid averaging spatial heterogeneity, BV indices were tabulated over 0.3 mm strip at ± 1mm away from the optic nerve head (red lines). (b) Group-averaged relative baseline blood volume values were taken at the peak of the ΔR2* profiles. RCS retinas were thinner than normal age-matched controls as seen by the proximity of the blood volume peaks from retinal and choroidal vasculatures compared to age-matched controls. The relative baseline blood volume was higher in both retinal and choroidal vascular layers in the RCS rats. (c) Toluidine-stained histological sections show the thinning of an RCS rat retina compared to an age-matched control. (GCL: ganglion cell layer, IPL: inner plexiform layer; INL: inner nuclear layer; OPL: outer plexiform layer; ONL: outer nuclear layer; IS+OS: inner and outer segments of the photoreceptors; CH: choroidal vasculature). * p < 0.05, ** p < 0.01

In control rats the BV profile taken across the retinal thickness showed a retinal and choroidal layer thickness (FWHM) of 60 ± 24 and 83 ± 11 μm (mean ± SD, N = 7), respectively, (Figure 3a), with clear visualization of the avascular layer possible. The peak-to-peak separation was 203 ± 25 μm. By contrast, in P90 RCS rats MRI revealed marked thinning of the avascular photoreceptor layer, the two peaks being separated by 131 ± 18 μm (in grey, p < 0.01). The retinal and choroidal layer thicknesses (as derived by extending the individual peaks in the profile plots) in the RCS rats by MION MRI were 50 ± 11 and 73 ± 7 μm (N = 7), respectively, which were not statistically different from those of the age-matched controls.

The group-averaged relative basal BV (defined as the peak value of ΔR2* profile) of the retinal and choroidal vascular layers in normal and age matched RCS rats are shown in Figure 3b. The BV peak ratio of choroidal to retinal profile was 9.8 ± 3.2 in control rats, and 8.9 ± 2.8 in P90 RCS rats (p>0.5). The area-under-the-curve ratio of the choroidal to retinal BV profile was 12.7 ± 4.3 in control rats and 13 ± 4 in P90 RCS rats, which was also not statistically different from controls. Importantly, baseline retinal and choroidal BV was markedly higher in RCS (71% and 55%, respectively) compared to age-matched controls. Retinal degeneration in P90 RCS rats was confirmed by histology, in which only a thin debris layer was observed in place of the photoreceptors and photoreceptor segments and the total thickness was markedly reduced (Figure 3c).

Stimulus-induced BV changes in control rats

Vascular layer-specific BV changes associated with hyperoxia and hypercapnia relative to air (baseline) were analyzed (Figure 4). Both choroidal and retinal vessels exhibited hyperoxia-induced vasoconstriction and hypercapnia-induced vasodilation. Hyperoxia-induced BV percent change was larger in the retinal vasculature (51 ± 17%) compared to the choroidal vasculature (28 ± 14%) (p < 0.01). Similarly, hypercapnia-induced BV percent change was larger in the retinal vasculature (52 ± 11%) compared to the choroidal vasculature (12 ± 11%) (p < 0.01). There were no statistically significant stimulus-induced changes in the thickness of either the retinal or choroidal vascular layer (p > 0.5).

Figure 4.

Figure 4

Open in a new tab

Layer-specific blood-volume changes responding to hyperoxia and hypercapnia from (a) a representative control rat. (b) Group-averaged retinal and choroidal blood-volume changes deu to hyperoxia and hypercapnia. (* p < 0.05, ** p < 0.01).

Stimulus-induced BV changes in RCS rats

BV fMRI associated with hyperoxia and hypercapnia was analyzed in RCS rat retinas in comparison with controls. Hyperoxia-induced BV changes in RCS rat retinas decreased by 51 ± 12% in the retinal vasculature and 22 ± 11% in the choroidal vasculature compared to baseline (p < 0.01), neither of these changes were significantly different from age-matched controls (p > 0.05). Hypercapnia-induced BV changes increased by 32 ± 18% in the retinal vasculature, and 9 ± 10% in the choroidal vasculature during hypercapnic stimulus (p < 0.01 vs baseline) in RCS rats. Only the hypercapnia-induced BV changes in the retinal vasculature of RCS rats were significantly different (smaller) from age-matched controls (p < 0.05).

Discussion

Blood-volume MRI revealed two distinct laminar signals that correspond to the retinal and choroidal vascular layers bounding the retina, separated by the avascular layer in between. Choroidal BV was about ten times higher than retinal BV, as expected. fMRI based on BV changes detects vascular layer-specific responses to physiologic challenges and reveals that the choroidal blood vessels were less responsive to physiologic challenge than retinal vessels, indicative of differential hemodynamic regulation of the two vasculatures. In an accepted animal model of photoreceptor degeneration, MRI confirmed the disappearance of the outer nuclear layer and photoreceptor segments, as seen previously (Cheng et al., 2006). However, the baseline BV values in the retinal and choroidal vasculatures were elevated compared to age-matched controls. Only the hypercapnia-induced BV percent changes in the retinal vascular layer of the RCS rats were significantly different from controls. Choroidal and retinal vascular layer thickness and choroidal:retinal BV ratio were unaltered by retinal degeneration. To our knowledge this is the first in vivo report of vascular layer-specific BV MRI and BV fMRI of vascular coupling to physiologic challenges in normal retinas and in a clinically relevant model of retinal disease.

Basal BV in normal animals

Non-invasive measurements of retinal and choroidal blood volume in vivo have not been reported to our knowledge. Consistent with the vascular layout of the retina, BV-MRI signal of the retinal vasculature was highest close to the optic nerve head and declined toward the peripheral retina. All retinal vessels originate from the optic nerve head and surface retinal vascular density progressively decreases toward the periphery. In contrast, choroidal BV was relatively uniform across the entire retinal length. Choroidal vessels originate from multiple medial and lateral posterior ciliary arteries entering the choroidal at various locations (Hayreh, 1974). Consequently, the BV findings and the thicknesses of the retinal and choroidal vascular layers were consistent with known retinal anatomy.

While quantitative blood volume in the vasculature was not determined herein, the choroidal BV was determined to be ten times higher than retinal BV. This result is in good agreement with the higher blood flow observed in the choroidal than the retinal vasculature using microspheres as tracers in rhesus monkey retina (Bill, 1984). This finding is also in agreement with an MRI study utilizing an intravascular contrast agent (Gd-DTPA, gadolinium diethylenetriamine penta-acetic acid) in which the subtraction of post and pre contrast images showed the choroidal vascular layer to be significantly more enhanced than the retinal vascular layer, albeit no quantitative analysis was performed (Cheng et al., 2006). The functional necessity of a high choroidal BV remains unclear. High BF and BV may be required to dissipate heat produced by incident light (Parver, 1980) and/or to maintain a large oxygen gradient to facilitate oxygen delivery to the avascular outer nuclear layer (Linsenmeier and Padnick-Silver, 2000) where the oxygen tension is close to zero under normal physiologic conditions. However, these hypotheses remain to be validated.

Stimulus-evoked BV changes in normal retinas

Only a few BF studies have utilized physiological challenges to characterize the retinal vasculatures. Hyperoxia has consistently been reported to decrease BF in the retina by 30% using the Heidelberg Retina Flowmeter (Sternn et al., 1997), and 60% using LDF (Eperson et al., 1975; Riva et al., 1983; Trokel, 1965). However, these techniques do not provide vascular layer specificity and are mostly sensitive to surface retinal vessels with unknown contributions from the choroid. Hyperoxia has also been reported to have little effect on choroidal BF when measured using LDF in the human macula where retinal vessels are largely absent (Riva et al., 1994). Retinal blood flow increases more significantly than choroidal blood flow post exercise in humans (Okuno et al., 2006).

The literatures on hypercapnic responses, on the other hand, are inconsistent. Inhalation of 10% CO2 in air showed no significant vasodilation in the retinal vessels (Frayser and Hickam, 1964). Inhalation of carbogen (95% O2 + 5% CO2) increased choroidal BF by 12.5 ± 11.7% but inter-subject variations were large (Geiser et al., 2000). At higher CO2 concentrations, however, retinal BF was observed to increase 240% and choroidal BF was observed to increase 150% (arterial pCO2 = 80.9 mmHg, which we estimate to be > 15% CO2, effectively) (Alm and Bill, 1972).

Oxygen electrode measurements (Yu et al., 2000) suggest differential tissue pO2 responses in the two vasculatures when challenged with hyperoxia and carbogen (5% CO2, 95% O2). Hyperoxia markedly enhanced pO2 in the choroid but only mildly enhanced pO2 in the retinal region. Tissue oxygenation under carbogen inhalation compared to oxygen inhalation was slightly higher in the choroidal region, but markedly higher in the retinal region.

A recent BOLD fMRI study (Cheng et al., 2006) also suggested differential oxygen responses to hyperoxia and hypercapnia in the two vascular layers. Hyperoxia induced a larger BOLD response in the outer choroidal band (12 ± 2%, N = 8) than the inner retinal band (7 ± 2%, p < 0.01). In contrast, hypercapnia induced a smaller BOLD response in the outer choroidal band (1.6 ± 1%, N = 8) than in the inner retinal band (10 ± 2%, p < 0.01). The BOLD signal is a convolution of competing changes in oxygen metabolism, BV, and BF. The current study provides direct evident that BV in the two vasculatures is indeed differentially regulated, and BV MRI provides a novel perspective on ocular hemodynamics. Based on the current data, we conclude that choroidal vessels vasoconstrict in response to hyperoxia and vasodilate in response to hypercapnia less than retinal vessels, but do autoregulate to a degree. Our findings support the notion that the retinal vessels are strongly autoregulated.

It is unknown how BV and BF are coupled in the retina. It is possible that BF-BV coupling in the retina is similar to that in the brain, and can be described via Grubb's relation (ΔBF = ΔBV0.38) (Grubb et al., 1974). However, the BF-BV coupling in the retina may differ from that in the brain because of the unique characteristics of the retinal and choroidal vasculature as discussed above (Laties, 1967; Steinle et al., 2000a). BF measurements are needed.

BV and BV responses in RCS rat retinas

Photoreceptor degeneration in RCS rats is well characterized with prominent degeneration of the avascular layer as confirmed by BV peak separation, providing a good model to corroborate layer-specific BV findings. In addition, peak-ΔR2* in RCS retinal and choroidal vessels were 46% to 80% higher than age-matched wild-type rats in this study. Neovascularization has been reported in retinal vasculature of P120 RCS rats resulting in an almost 50% increase in superficial retinal capillary density (Seaton and Turner, 1992). Fluorescein and indocyanin green angiography studies in ∼P70 RCS rats reveal hyperfluorescent regions, suggesting regions of neovascularization and formation of coiled vascular tuffs (Seaton and Turner, 1992). Consistent with these observations, our studies indicate that the choroidal BV is 46%-61% higher and retinal BV is 51%-80% higher in the RCS retina than normal retina.

It should be noted that the steady-state BV calculation by MRI assumes that there is no leakage of MION from either the retinal or choroidal vasculature in RCS rats, which could potentially result in an overestimation of the BV. The blood-retinal barrier in normal retinal vessels and the retinal pigment epithelium form the tight junctions to prevent leakage of large molecules such as indocyanin-green and MRI contrast agents. However, retinal vascular permeability has been shown to increase with age in RCS retina using fluorescein angiography (Satoh and Yamaguchi, 2000). A previous MRI study on RCS rats using Gd-DTPA as a vascular contrast agent revealed an apparent thickening of the choroid, suggesting this debris layer is permeable to Gd-DTPA in retinal degeneration.

Our current finding of BV increase in RCS rats appears to be inconsistent with a previous study which reported a blood flow decrease in RCS rats using arterial spin-labeling technique without a use of contrast agent (Li et al., 2009). A possible explanation could be due to methodological differences, namely, that blood volume and blood flow parameters are different aspects of the physiology. Another possible explanation is that MION leakage overestimated BV, but should not significantly affect. However, lack of ΔR2* changes due to MION in the vitreous, as well as lack of any time dependent ΔR2* in the choroids in either the control or RCS rats suggests that MION does not leak out of the retinal vasculature, up to 2 hours following MION injection. Further studies are needed to clarify these inconsistencies. Nonetheless, the implication is that BV MRI, which measures tissue BV without the need to visualize individual vessels, can be utilized to image BV abnormalities in vivo.

BV fMRI responses to hyperoxia in both retinal and choroidal vessels in RCS rat retinas are similar to normal controls, in reasonable agreement with a BOLD fMRI study which showed that hyperoxic responses in RCS rats were only slightly attenuated compared to controls (Cheng et al., 2006). Hypercapnia-induced BV changes in the retinal vasculature of RCS rats were significantly different (smaller) from controls (p < 0.05), consistent with BOLD fMRI findings (Cheng et al., 2006). Hypercapnia-induced BV changes in the choroidal vessels in RCS rats were not statistically different from controls (p > 0.05). In contrast, a previous report found that hypercapnia-induced BOLD changes in choroidal vasculature of RCS rats were significantly different from controls although the hypercapnia-induced BOLD changes in normal choroid is already small (Cheng et al., 2006). Further studies are needed to address this discrepancy. We found no other publication using other imaging techniques to study hemodynamics and vascular responses of retinal degeneration. These findings underscore the utility of BV fMRI to image changes in BV and its vascular reactivity of the two vasculatures associated with retinal diseases.

In summary, we demonstrate a novel MRI application to image BV in the retina, providing important insights into the unique retinal and choroidal hemodynamic regulation in normal and diseased retinas in vivo. This approach may prove useful to monitor BV, its vascular reactivity, permeability and neovascularization associated with diseases in the two vasculatures of the retina.

Highlights.

  • Blood-volume MRI revealed two distinct laminar signals that correspond to the retinal and choroid vascular layers bounding the retina.

  • Choroid BV was about ten times higher than retinal BV.

  • fMRI based on BV changes detects vascular layer-specific responses to physiologic challenges and reveals that the choroid blood vessels were less responsive to physiologic challenge than retinal vessels.

  • In an accepted animal model of photoreceptor degeneration, MRI confirmed the disappearance of the outer nuclear layer and photoreceptor segments.

  • The baseline BV in the retinal and choroid vasculature were elevated compared to age-matched controls.

Acknowledgments

Funding sources: NIH/NEI (R01EY014211, R01EY018855) to TQD, VA Career Development (TQD and DEO), VA MERIT (TQD, PMT and MTP).

Abbreviations

BF

blood flow

BV

blood volume

BOLD fMRI

blood-oxygenation-level-dependent functional MRI

MION

monocrystalline iron oxide nanocolloid

MRI

magnetic resonance imaging

RCS

Royal College of Surgeons

effective spin-spin relaxation time (T2*) or rate constant (ΔR2*).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. 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 Physiologica Scand. 1972;84:306–319. doi: 10.1111/j.1748-1716.1972.tb05182.x. [DOI] [PubMed] [Google Scholar]
  2. Berkowitz BA, Roberts R, Goebel DJ, Luan H. Noninvasive and simultaneous imaging of layer-specific retinal functional adaptation by manganese-enhanced MRI. Invest Ophthalmol Vis Sci. 2006;47:2668–2674. doi: 10.1167/iovs.05-1588. [DOI] [PubMed] [Google Scholar]
  3. Bill A. Circulation in the eye. In: Renkin EM, Michel CC, editors. Handbook of physiology Part 2 in Microcirculation. American Physiological Society; Bethesda, MD: 1984. pp. 1001–1035. [Google Scholar]
  4. Caldwell RB, Roque RS, Solomon SW. Increased vascular density and vitreo-retinal membranes accompany vascularization of the pigment epithelium in the dystrophic rat retina. Curr Eye Res. 1989;8:923–937. [PubMed] [Google Scholar]
  5. Cheng H, Nair G, Walker TA, Kim MK, Pardue MT, Thule PM, Olson DE, Duong TQ. Structural and functional MRI reveals multiple retinal layers. Proc Natl Acad Sci USA. 2006;103:17525–17530. doi: 10.1073/pnas.0605790103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Dowling JE, Sidman RL. Inherited retinal dystrophy in the rat. J Cell Bio. 1962;14:73–109. doi: 10.1083/jcb.14.1.73. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Drexler W, Fujimoto JG. State-of-the-art retinal optical coherence tomography. Prog Retin Eye Res. 2008;27:45–88. doi: 10.1016/j.preteyeres.2007.07.005. [DOI] [PubMed] [Google Scholar]
  8. Duong TQ, Ngan SC, Ugurbil K, Kim SG. Functional Magnetic Resonance Imaging of the Retina. Invest Ophthalmol Vis Sci. 2002;43:1176–1181. [PMC free article] [PubMed] [Google Scholar]
  9. Eperson G, Johnson M, David NJ. The effect of arterial PO2 on relative retinal blood flow in monkeys. Invest Ophthalmol Vis Sc. 1975;14:342–352. [PubMed] [Google Scholar]
  10. Frayser R, Hickam JB. Retinal vascular response to breathing increased carbon dioxide and oxygen concentrations. Invest Ophthalmol Vis Sci. 1964;3:427–431. [PubMed] [Google Scholar]
  11. Gal A, Li Y, Thompson DA, Weir J, Orth U, Jacobson SG, Apfelstedt-Sylla E, Vollrath D. Mutations in MERTK, the human orthologue o the RCS rat retinal dystrophy gene, cause retinitis pigmentosa. Nat Genet. 2000;26:270–271. doi: 10.1038/81555. [DOI] [PubMed] [Google Scholar]
  12. Geiser MH, Riva CE, GDorner GT, Diermann U, Luksch A, Schmetterer L. Response of choroidal blood flow in the foveal region to peroxia and hyperoxia-hypercapnia. Current Eye Res. 2000;21:669–676. [PubMed] [Google Scholar]
  13. Grubb RL, Jr, Raichle ME, Eichling JO, Ter-Pogossian MM. The effects of changes in PaCO2 on cerebral blood volume, blood flow, and vascular mean transit time. Stroke. 1974;5:630–639. doi: 10.1161/01.str.5.5.630. [DOI] [PubMed] [Google Scholar]
  14. Harris A, Kagemann L, Cioffi GA. Assessment of human ocular hemodynamics. Surv of Ophthalmol. 1998;42:509–533. doi: 10.1016/s0039-6257(98)00011-3. [DOI] [PubMed] [Google Scholar]
  15. Hayreh SS. Submacular choroidal vascular pattern. Experimental fluorescein fundus angiographic studies. Albrecht Von Graefes Arch Klin Exp Ophthalmol. 1974;192:181–196. doi: 10.1007/BF00416865. [DOI] [PubMed] [Google Scholar]
  16. Laties A. Central retinal artery innervation. Absence of adrenergic innervation to the intraocular branches. Arch Ophthalmol. 1967;77:405–409. doi: 10.1001/archopht.1967.00980020407021. [DOI] [PubMed] [Google Scholar]
  17. LaVail MM, Sidman RL, O'Neil D. Photoreceptor-pigment epithelial cell relationships in rats with inherited retinal degeneration. Radioautographic and electron microscope evidence for a dual source of extra lamellar material. J Cell Biol. 1972;53:185–209. doi: 10.1083/jcb.53.1.185. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Li Y, Cheng H, Duong TQ. Blood-flow magnetic resonance imaging of the retina. Neuroimage. 2008;39:1744–1751. doi: 10.1016/j.neuroimage.2007.10.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Li Y, Cheng H, Shen Q, Kim M, Thule PM, Olson DE, Pardue MT, Duong TQ. Blood-Flow Magnetic Resonance Imaging of Retinal Degeneration. Invest Ophthalmol Vis Sci. 2009;50:1824–1830. doi: 10.1167/iovs.08-2188. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Linsenmeier RA, Padnick-Silver L. Metabolic dependence of photoreceptors on the choroid in the normal and detached retina. Invest Ophthalmol Vis Sci. 2000;41:3117–3123. [PubMed] [Google Scholar]
  21. Mandeville JB, Marota JJ, Kosofsky BE, Keltner JR, Weissleder R, Rosen BR. Dynamic functional imaging of relative cerebral blood volume during rat forepaw stimulation. Magn Reson Med. 1998;39:615–624. doi: 10.1002/mrm.1910390415. [DOI] [PubMed] [Google Scholar]
  22. Nelson DA, Krupsky S, Pollack A, Aloni E, Belkin M, Vanzetta I, Rosner M, Grinvald A. Special report: Noninvasive multi-parameter functional optical imaging of the eye. Ophthalmic Surg Lasers Imaging. 2005;36:57–66. [PubMed] [Google Scholar]
  23. Ogawa S, Menon RS, Tank DW, Kim SG, Merkle H, Ellermann JM, Ugurbil K. Functional brain mapping by blood oxygenation level-dependent contrast magnetic resonance imaging. Biophys J. 1993;64:800–812. doi: 10.1016/S0006-3495(93)81441-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Okuno T, Sugiyama T, Kohyama M, Kojima S, Oku H, Ikeda T. Ocular blood flow changes after dynamic exercise in humans. Eye. 2006;20:796–800. doi: 10.1038/sj.eye.6702004. [DOI] [PubMed] [Google Scholar]
  25. Parver LM. Choroidal blood flow as a heat dissipating mechanism in the macula. Am J Ophthalmol. 1980;89:641–646. doi: 10.1016/0002-9394(80)90280-9. [DOI] [PubMed] [Google Scholar]
  26. 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] [Google Scholar]
  27. Riva CE, Grunwald JE, Sinclair SH. Laser Doppler velocimetry study of the effect of pure oxygen breathing on retinal blood flow. Invest Ophthalmol Vis Sci. 1983;24:47–51. [PubMed] [Google Scholar]
  28. Satoh T, Yamaguchi K. Ocular fundus abnormalities detected by fluorescein and indocyanine green angiography in the Royal College of Surgeons dystrophic rat. Exp Anim. 2000;49:275–280. doi: 10.1538/expanim.49.275. [DOI] [PubMed] [Google Scholar]
  29. Seaton AD, Turner JE. RPE transplants stabilize retinal vasculature and prevent neovascularization in the RCS rat. Invest Ophthalmol Vis Sci. 1992;33:83–91. [PubMed] [Google Scholar]
  30. Shen Q, Cheng H, Pardue MT, Chang TF, Nair G, Vo VT, Shonat RD, Duong TQ. Magnetic resonance imaging of tissue and vascular layers in the cat retina. J Magn Reson Imaging. 2006;23:465–472. doi: 10.1002/jmri.20549. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Steinle JJ, Krizsan-Agbas D, Smith PG. Regional regulation of choroidal blood flow by autonomic innervation in the rat. Am J Physiol Regul Integr Comp Physiol. 2000a;279:202–209. doi: 10.1152/ajpregu.2000.279.1.R202. [DOI] [PubMed] [Google Scholar]
  32. Steinle JJ, Krizsan-Agbas D, Smith PG. Regional regulation of choroidal blood flow by autonomic innervation in the rat. Am J Physiol Regul Integr Comp Physiol. 2000b;279:R202–209. doi: 10.1152/ajpregu.2000.279.1.R202. [DOI] [PubMed] [Google Scholar]
  33. Sternn K, Manapace R, Rainer G, Findl O, Wolzt M, Schmetterer L. Reproducibility and sensitivity of scanning laser Doppler flowmetry using graded changes in PO2. Br J Ophthalmol. 1997;81:360–364. doi: 10.1136/bjo.81.5.360. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Trokel S. Effect of respiratory gases upon choroidal hemodynamics. Arch Ophthalmol. 1965;73:838–842. doi: 10.1001/archopht.1965.00970030840016. [DOI] [PubMed] [Google Scholar]
  35. Yu DY, Cringle SJ, Su EN, Yu PK. Intraretinal oxygen levels before and after photoreceptor loss in the RCS rat. Invest Ophthalmo Vis Sci. 2000;41:3999–4006. [PubMed] [Google Scholar]

RESOURCES