Impaired Retinal Capillary Function in Patients With Alzheimer Disease

Background: Extensive evidence indicates that vasculopathy, especially the level of microcirculation, contributes to neurodegeneration in Alzheimer disease (AD). However, it is not easy to directly monitor cerebral microcirculation. The retinal microvasculature has been proposed as a surrogate measure to study cerebral vascular changes. Indeed, decreased retinal microvascular network densities were reported in patients with AD. We sought to determine the retinal capillary function (RCF, the efficiency of blood flow transferring in the capillary network) in patients with AD. Methods: Twenty patients (age 60-84 years, mean ± SD: 72.8 ± 7.7 years) with AD and 14 age-matched cognitively normal controls (CN, age 62-81 years, mean ± SD: 68.6 ± 6.7 years.) were recruited. There were no differences in vascular risk factors, including smoking, hypertension, hyperlipidemia, Type 2 diabetes, and cardiovascular disease, between the groups. One eye of each subject in both groups was imaged. Retinal blood flow (RBF) was measured using a retinal function imager, and retinal capillary density (RCD, expressed as fractal dimension Dbox) was measured using optical coherence tomography angiography. RCF was defined as the ratio of RBF to RCD. Results: RCF was 1.62 ± 0.56 nl/s/Dbox (mean ± SD) in the AD group, which was significantly lower than that (2.56 ± 0.25 nl/s/Dbox, P < 0.01) in the CN group. The change of RCF in the AD group represented 28% lower than in the CN group. RCF was significantly and positively correlated with RBF in the AD group (r = 0.98, P < 0.05) and in the CN group (r = 0.65, P < 0.05). Conclusions: Our study is the first to demonstrate impaired retinal capillary function in patients with AD. The alteration of RCF was mainly due to decreased retinal blood flow, which is transferred by the capillary network. The RCF may be developed as a biomarker of impaired cerebral microcirculation in patients with AD.

Original Contribution Section Editors: Clare Fraser, MD Susan Mollan, MD Impaired Retinal Capillary Function in Patients With Alzheimer Disease Jack A. Cipolla, BS, Hong Jiang, MD, PhD, Ava-Gaye Y. Simms, MD, Bernard Baumel, MD, Tatjana Rundek, MD, PhD, Jianhua Wang, MD, PhD Background: Extensive evidence indicates that vasculopathy, especially the level of microcirculation, contributes to neurodegeneration in Alzheimer disease (AD). However, it is not easy to directly monitor cerebral microcirculation. The retinal microvasculature has been proposed as a surrogate measure to study cerebral vascular changes. Indeed, decreased retinal microvascular network densities were reported in patients with AD. We sought to determine the retinal capillary function (RCF, the efficiency of blood flow transferring in the capillary network) in patients with AD. Methods: Twenty patients (age 60–84 years, mean ± SD: 72.8 ± 7.7 years) with AD and 14 age-matched cognitively normal controls (CN, age 62–81 years, mean ± SD: 68.6 ± 6.7 years.) were recruited. There were no differences in vascular risk factors, including smoking, hypertension, hyperlipidemia, Type 2 diabetes, and cardiovascular disease, between the groups. One eye of each subject in both groups was imaged. Retinal blood flow (RBF) was measured using a retinal function imager, and retinal capillary density (RCD, expressed as fractal dimension Dbox) was measured using optical coherence tomography angiography. RCF was defined as the ratio of RBF to RCD. Results: RCF was 1.62 ± 0.56 nl/s/Dbox (mean ± SD) in the AD group, which was significantly lower than that (2.56 ± 0.25 nl/s/Dbox, P , 0.01) in the CN group. The change of RCF in the AD group represented 28% lower than in the CN group. RCF was significantly and positively correlated with RBF in the AD group (r = 0.98, P , 0.05) and in the CN group (r = 0.65, P , 0.05). Conclusions: Our study is the first to demonstrate impaired retinal capillary function in patients with AD. The alteration of RCF was mainly due to decreased retinal blood flow, which is transferred by the capillary network. The RCF may University of Miami Miller School of Medicine (JAC, HJ, A-GYS, BB, TR, JW), Miami, Florida; Department of Ophthalmology (HJ, A-GYS, JW), Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, Miami, Florida; Department of Neurology (HJ, BB, TR), University of Miami Miller School of Medicine, Miami, Florida; and The University of Miami Evelyn F. McKnight Brain Institute (HJ, TR, JW), Miami, Florida. Supported by NIH Center Grant P30 EY014801, NINDS 1R01NS111115-01 (J. Wang), the Ed and Ethel Moor Alzheimer’s Disease Research Program (Florida Health, 20A05, to H. Jiang), and a grant from Research to Prevent Blindness. The authors report no conflicts of interest. Address correspondence to Hong Jiang, MD, PhD, Bascom Palmer Eye Institute, University of Miami, Miller School of Medicine, 1638 NW 10th Avenue, McKnight Building—Room 202A, Miami, FL 33136; E-mail: [e-mail] 30 be developed as a biomarker of impaired cerebral microcirculation in patients with AD. Journal of Neuro-Ophthalmology 2024;44:30–34 doi: 10.1097/WNO.0000000000001954 © 2023 by North American Neuro-Ophthalmology Society A lzheimer disease (AD) negatively affects the lives of many patients around the world. As of 2021, the estimated prevalence of AD worldwide is nearing 50 million people.1 In the United States alone, more than 6.2 million people aged 65 years and older have AD.2 Extensive evidence indicates that vasculopathy, especially at the level of microcirculation, contributes to neurodegeneration in AD.3 Most patients with AD have degeneration and vascular alterations affecting the capillaries.4 Although the brain microvasculature can be imaged and studied through MRI and transcranial Doppler imaging techniques, these imaging modalities are time-consuming and expensive and, thus, are not practical for early detection and monitoring of cerebral microvascular alterations.5 One potential solution is to use imaging of the microvasculature in the retina, which shares similar embryologic, anatomic, and physiologic features with the brain, but it is easily accessible for imaging6 and cost-effective. Indeed, the retinal microvasculature has been used to study cerebral vascular changes in patients with AD.7 Retinal tissue hypoperfusion and decreased retinal microvascular network densities were reported in patients with AD.7 Furthermore, retinal microvascular density was decreased in AD patients with brain MRI white matter T2 hyperintense lesions,8 mirroring the decreased capillary density seen in the brains of patients with AD.9 The retinal capillary is generally understood to have 2 primary functions: to transport blood and transfer oxygen and nutrients to the surrounding tissues. Studying capillary function may provide further insight into the role of the capillary in the onset and progression of AD. Retinal capillary function (RCF) has been defined as the efficiency of blood flow through the capillary network10 and calculated from retinal blood flow (RBF) and retinal capillary density (RCD). RCF has been validated previously in a study that determined the changes of the RCF after physical exercise in cognitively normal individuals.10 RCF has not been studied in patients with AD. The goal of this project was to determine the RCF in patients with AD. Cipolla et al: J Neuro-Ophthalmol 2024; 44: 30-34 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution METHODS Patients with AD and age-matched cognitively normal controls (CN) were recruited. This study was approved by the institutional review board for human research at the University of Miami, and written informed consent was obtained from each participant. The right eye of each subject in both groups was scanned. Two imaging modalities and analyses were performed for each group to determine RBF and RCD in each group, and the results were used to calculate RCF. RBF was measured using the retinal function imager (RFI, Optical Imaging Ltd, Rehovot, Israel). The details of the RFI and its application were previously reviewed.11 In brief, the RFI system captures RBF by using a fundus camera with a high-speed digital camera and stroboscopic flash lamp system. Blood flow was tracked using the motion of red blood cells and then used to calculate blood flow velocity (Fig. 1). Blood flow in the arterioles and venules crossing a 2.5-mm circle centered around the macular region were measured separately and summed (Fig. 1). The average blood flow of arterioles and venules was calculated to represent RBF. AngioVue optical coherence tomography angiography (OCTA, AngioVue, Optovue, Inc, Fremont, CA) was used to measure RCD (RCD; Fig. 2). RCD, its image protocol and analysis, and the OCTA imaging system have been well-described in the literature.7 Scanning of the macula centered on the fovea was completed using an Angio retina of the 3 · 3 protocol. The retinal vascular network defined the vessels from the inner limiting membrane (ILM) to the outer plexiform layer (OPL), and angiographic en face images of the total retinal vascular network were exported and analyzed using fractal analysis.7 Small vessels in the annulus from 0.6 to 2.5 mm were analyzed after the removal of vessels with a diameter of larger than 25 mm and a series of image processing procedures. The avascular zone is 0.6 mm in the center of the fovea, and thus, the annulus described encompasses most of the capillaries in calculating RCD. Dbox, representing RCD, was calculated using the box-counting method in fractal analysis. RCF, defined as the efficiency of blood flow transferring in the capillary network, was calculated as the ratio of RBF to RCD, resulting in a unit of nL/s/Dbox for RCF and equating to blood flow volume per second per unit of capillaries. RCF was calculated, averaged, and compared between the AD and CN groups. A statistical software package (STATISTICA, StatSoft, Inc, Tulsa, OK) was used for descriptive statistics and data analysis. The categorical variables, including confounders such as sex, were analyzed using a x2 test. The student t tests were used to analyze the differences in RCF between groups. Pearson correlation was used to evaluate the relationship among the retinal parameters. A P , 0.05 was considered statistically significant. Cipolla et al: J Neuro-Ophthalmol 2024; 44: 30-34 FIG. 1. Retinal blood flow imaged using the retinal function imager. The arterioles with the negative values were marked in red, which indicates that the blood flow moves toward the fovea (moving away from the heart). The venules with positive values were marked in purple, which indicated the blood was flowing away from the fovea. The diameters of the vessels crossing the 2.5-mm circle outlined in blue were measured. The blood flow rate of each arteriole (green dots) and venule (yellow dots) was calculated using its velocity and diameter. The arteriolar blood flow (RBF) entering the circle to the macula was the sum of the blood flow rates of all arterioles crossing the 2.5-mm circle (i.e., all green dots). The venular RBF was calculated as the sum of the blood flow rates of all venules crossing the 2.5-mm circle (i.e., yellow dots). RBF, retinal blood flow. Data are represented as mean ± SD or percentages, where appropriate. RESULTS Twenty patients (age 60–84 years, mean ± SD: 72.8 ± 7.7 years) with AD and 14 age-matched cognitively normal controls (CN, age 62–81 years, mean ± SD: 68.6 ± 6.7 years) were recruited for participation in this study. Vascular risk factors, including smoking, hypertension, hyperlipidemia, Type 2 diabetes, and cardiovascular disease, were similar between the 2 groups (Table 1). RCF in the AD group was 28% lower than in the CN group. RCF was 1.62 ± 0.56 nL/s/Dbox (mean ± SD) in the AD group, which was significantly lower than 2.56 ± 0.25 nL/s/Dbox (P = 0.0001) in the CN group (Fig. 3). In addition, RCF was highly correlated to RBF in the AD group (r = 0.98, P , 0.05) and the CN group (r = 0.65, P , 0.05) (Fig. 4). However, RCF was not correlated to mini-mental state examination (MMSE) in patients with AD (r = 0.03, P . 0.05). 31 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 2. Segmentation of the OCTA image. The raw image (A) acquired using OCTA was processed, and any vessels with a diameter of 25 mm (B) were extracted. The remaining vessels, defined as microvessels (i.e., mainly capillaries), were then skeletonized (C). The circular zone, with a diameter of 2.5 mm centered on the fovea, was analyzed to yield retinal capillary density (RCD). OCTA, optical coherence tomography angiography. DISCUSSION This study demonstrates impaired retinal capillary function in patients with AD. The alteration of RCF was mainly due to decreased RBF, which the capillary network can transfer. An MRI perfusion study of the brain in patients with AD has demonstrated capillary dysfunction in the brain, associated with impaired cerebral microvascular perfusion, suggesting the link between capillary dysfunction and AD pathophysiology.12 Given the evidence that capillary dysfunction plays a role in AD disease progression,12 RCF may be developed as a biomarker to study cerebral microcirculation in patients with AD. The capillary system serves to both maintain blood flow in the tissue and facilitate the diffusion of oxygen and nutrients to the tissue.13 Thus, impairment of either of these functions can lead to degeneration of the tissue supplied by the capillaries. Blood flow must be maintained at a certain threshold to supply the tissue with sufficient oxygen. It is worth noting that this study did not determine the ability of the capillary to deliver oxygen to tissues (the process of diffusion), which is a key function of capillaries. Previous data demonstrate diminished diffusion of oxygen in patients with AD,14 possibly due in part to b amyloidinduced thickening of capillary vessel walls.15 This study demonstrates diminished RCF in patients with AD. RCF has been defined as the ratio of RBF to RCD, and its decrease indicates that RBF and RCD are changing at different rates. A decline in RCF is either a result of a decrease in RBF that is greater than the RCD decrease or an increase in RCD that is greater than the increase in RBF. We found RCF to be significantly related to RBF in both groups, indicating that the decline in RCF is a result of a decline in RBF greater than the change in RCD. This is consistent with previous data that demonstrated diminished RBF in patients with AD.16 In addition, retinal arteriolar dilation, and thus maximum diameter, is diminished in patients with AD,17 indicating that AD may impair the ability of retinal capillaries to deliver 32 effective blood flow.17 The mechanism underlying these observations may be due to the accumulation of phosphorylated tau protein in the brain and the retina.17 Phosphorylated tau protein accumulation is a primary driver in AD, and amyloid b CSF levels have been shown to be positively correlated with the ability of retinal arterioles to dilate and inversely correlated with levels of phosphorylated tau protein in the brain and the retina.17 Furthermore, patients with AD have shown phosphorylated tau protein-driven capillary kinking, looping, and twisting, all of which may impede blood flow.18 Hence, the increased presence of phosphorylated tau protein in AD may drive retinal arteriole and capillary dysfunction, resulting in diminished RBF and RCF.17,18 We did not find a relationship between RCF and the MMSE scores in patients with AD. This may be due to MMSE being an unreliable indicator of AD progression, variability in patient baseline RCD values, measurement TABLE 1. Demographics and clinical manifestations of Alzheimer disease and cognitively normal participants N Age (mean ± SD, yr) Sex MMSE Disease duration (yr) Smoking Hypertension Dyslipidemia Diabetes Cardiovascular disease AD CN P 20 72.7 ± 7.7 9M/11F 21.9 ± 4.6 3.8 ± 1.6 0 11 9 3 5 14 68.6 ± 6.7 8M/6F 29.4 ± 0.8 — 0 8 8 2 1 0.24 0.48 ,0.01 NS NS 0.90 0.48 0.95 0.17 There were no differences in vascular risk factors, including smoking, hypertension, hyperlipidemia, Type 2 diabetes, and cardiovascular disease, between the 2 groups (All chi-square P . 0.05). AD, Alzheimer disease; CN, cognitively normal; MMSE, mini-mental state examination. Cipolla et al: J Neuro-Ophthalmol 2024; 44: 30-34 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 3. RCF in AD and CN. RCF was 28% lower in the AD group than in the CN group (P , 0.01). Bars = standard error. AD, Alzheimer disease; CN, cognitively normal; RCF, retinal capillary function. error, or the complex stepwise pathophysiology of AD progression outlined above.14,15,17,18 Previous data have demonstrated that MMSE is also not related to retinal tissue perfusion or retinal microvascular network density in patients with AD.7,16 However, MMSE has been shown to have a large measurement error and significant variation in change in annual score.19 The retinal vessel density of the superficial vascular plexus (SVP) in patients with CN, MCI, and AD has been shown to be positively correlated with the Montreal Cognitive Assessment (MoCA), which is a more sensitive test for the cognitive function that includes additional examination.20,21 Of note, deep vascular plexus (DVP) density and total retinal density were not found to be correlated with MoCA.21 Furthermore, the absence of a relationship between the RCF and MMSE may be due to differing patient baseline RCD and RBF values. Baseline values for RCD and RBF have been shown to vary among healthy individuals,22,23 with both decreasing in patients with AD.3,7 However, RCF changes over time have been shown to be positively correlated with cognitive function changes in CN individuals.10 Thus, it is possible that a longitudinal comparison of RCF to MMSE or MoCA over time may demonstrate a stronger relationship than the cross-sectional study presently conducted. Finally, given that the RCD was determined through OCTA imaging, it is possible that measurement error is contributing to the lack of this relationship. There are limitations that may have affected the results of this study. The study sample size was small (N = 20), and we did not include direct brain imaging to confirm that the observations in the retina are mirrored in the brain. Despite these limitations, this study demonstrates a significant decrease in RCF in patients with AD and a relationship between RBF and RCF. Thus, future studies with larger sample sizes and longitudinal designs, which include both MoCA testing and imaging of the brain microvasculature, are needed to assess simultaneous retinal, cerebral, and cognitive changes. Cipolla et al: J Neuro-Ophthalmol 2024; 44: 30-34 FIG. 4. Relationship between RCF and RBF in the AD and CN groups. RCF was significantly correlated with RBF in both the AD (A) and CN (B) groups. AD, Alzheimer disease; CN, cognitively normal; RBF, retinal blood flow; RCF, retinal capillary function. Conclusions This study uniquely demonstrates decreased RCF in patients with AD. The decline in RBF, which the retinal capillary network can transfer, is primarily responsible for the observed changes in RCF. Future longitudinal studies with a larger sample size, which include both retinal and brain microvasculature imaging, may further characterize changes in RCF in AD. With further characterization, RCF may be proven to be a useful tool for early diagnosis and monitoring of the progression of AD. STATEMENT OF AUTHORSHIP Conception and design: H. Jiang, J. Wang; Acquisition of data: A.-G. Y. Simms, B. Baumel, J. Wang, H. Jiang; Analysis and interpretation of data: J. A. Cipolla, A.-G. Y. Simms, B. Baumel, J. Wang, H. Jiang. Drafting the manuscript: J. A. Cipolla, J. Wang, H. Jiang; Revising the manuscript for intellectual content: J. A. Cipolla, A.-G. Y. Simms, B. Baumel, J. Wang, H. Jiang, T. Rundek. Final approval of the completed manuscript: J. A. Cipolla, A.-G. Y. Simms, B. Baumel, J. Wang, H. Jiang, T. Rundek. REFERENCES 1. Guzman-Martinez L, Calfío C, Farias GA, Vilches C, Prieto R, Maccioni RB. New frontiers in the prevention, diagnosis, and treatment of Alzheimer’s disease. J Alzheimers Dis. 2021;82:S51–S63. 2. 2021 Alzheimer’s disease facts and figures. Alzheimers Demen. 2021;17:327–406. 33 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution 3. Jiang H, Liu Y, Wei Y, et al. Impaired retinal microcirculation in patients with Alzheimer’s disease. PLoS One. 2018;13:e0192154. 4. Love S, Miners JS. Cerebrovascular disease in ageing and Alzheimer’s disease. Acta Neuropathol. 2016;131:645–658. 5. Chandra A, Dervenoulas G, Politis M. Magnetic resonance imaging in Alzheimer’s disease and mild cognitive impairment. J Neurol. 2019;266:1293–1302. 6. London A, Benhar I, Schwartz M. The retina as a window to the brain—from eye research to CNS disorders. Nat Rev Neurol. 2013;9:44–53. 7. Jiang H, Wei Y, Shi Y, et al. Altered macular microvasculature in mild cognitive impairment and Alzheimer disease. J Neuroophthalmol. 2018;38:292–298. 8. Wang X, Zhao Q, Tao R, et al. Decreased retinal vascular density in Alzheimer’s disease (AD) and mild cognitive impairment (MCI): an optical coherence tomography angiography (OCTA) study. Front Aging Neurosci. 2020;12:572484. 9. Baloyannis SJ. Brain capillaries in Alzheimer’s disease. Hell J Nucl Med. 2015;18(suppl 1):152. 10. Jiang H, Signorile JF, Simms AG, Wang J. Improvement of retinal capillary function after high-speed circuit resistance training in healthy older adults. J Neuroophthalmol. 2023;43:180–184. 11. Wang L, Jiang H, Grinvald A, Jayadev C, Wang J. A Mini review of clinical and research applications of the retinal function imager. Curr Eye Res. 2018;43:273–288. 12. Madsen LS, Nielsen RB, Parbo P, et al. Capillary function progressively deteriorates in prodromal Alzheimer’s disease: a longitudinal MRI perfusion study. Aging Brain. 2022;2:100035. 13. Østergaard L, Jespersen SN, Engedahl T, et al. Capillary dysfunction: its detection and causative role in dementias and stroke. Curr Neurol Neurosci Rep. 2015;15:37. 34 14. Olafsdottir OB, Saevarsdottir HS, Hardarson SH, et al. Retinal oxygen metabolism in patients with mild cognitive impairment. Alzheimers Dement (Amst). 2018;10:340–345. 15. Thal DR, Griffin WST, De Vos RAI, Ghebremedhin E. Cerebral amyloid angiopathy and its relationship to Alzheimer’s disease. Acta Neuropathol. 2008;115:599–609. 16. Gameiro GR, Jiang H, Liu Y, et al. Retinal tissue hypoperfusion in patients with clinical Alzheimer’s disease. Eye Vis (Lond). 2018;5:21. 17. Querques G, Borrelli E, Sacconi R, et al. Functional and morphological changes of the retinal vessels in Alzheimer’s disease and mild cognitive impairment. Sci Rep. 2019;9:63. 18. Buée L, Hof PR, Delacourte A. Brain microvascular changes in Alzheimer’s disease and other dementias. Ann N Y Acad Sci. 1997;826:7–24. 19. Clark CM. Variability in annual mini-mental state examination score in patients with probable Alzheimer disease. Arch Neurol. 1999;56:857. 20. Zhang YS, Zhou N, Knoll BM, et al. Parafoveal vessel loss and correlation between peripapillary vessel density and cognitive performance in amnestic mild cognitive impairment and early Alzheimer’s Disease on optical coherence tomography angiography. PLoS One. 2019;14:e0214685. 21. Fang M, Strand K, Zhang J, et al. Retinal vessel density correlates with cognitive function in older adults. Exp Gerontol. 2021;152:111433. 22. Giocanti-Aurégan A, Gazeau G, Hrarat L, et al. Ethnic differences in normal retinal capillary density and foveal avascular zone measurements. Int Ophthalmol. 2020;40:3043–3048. 23. Garhofer G, Werkmeister R, Dragostinoff N, Schmetterer L. Retinal blood flow in healthy young subjects. Invest Opthalmol Vis Sci. 2012;53:698. Cipolla et al: J Neuro-Ophthalmol 2024; 44: 30-34 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.