Investigation of blood flow regulation and oxygen saturation of the retinal vessels in primary open-angle glaucoma

Evaluation of Reproducibility of Measurements of Peripapillary Blood Flow with Laser Speckle Flowgraphy in Healthy and POAG Subjects

BACKGROUND

Peripapillary blood flow is of diagnostic value in glaucoma.

MATERIAL AND METHODS

We investigated the reproducibility of peripapillary blood flow measurements in healthy individuals and in open-angle glaucoma (OAG) patients with the Nidek laser speckle flowgraphy (LSFG)-NAVI system. Blood flow was calculated as the mean of vascular area (MV), mean of tissue area (MT), and mean of overall flow (MA). Intra-rater, inter-session, and inter-rater reproducibility were assessed by determining the intraclass correlation coefficient (ICC) and coefficient of variance (COV).

RESULTS

There were 28 healthy subjects and 16 patients with OAG who participated in this prospective single-center trial. ICC and COV of intra-rater reproducibility were assessed in MV as 0.901; 6.424, MT as 0.909; 4.025, and MA as 0.965; 4.168. Inter-session reproducibility yielded ICC and COV values of MV (0.834; 9.223), MT (0.904; 6.215), MA (0.909; 7.069). Inter-rater reproducibility was computed for MV as 0.78; 9.758, MT as 0.890; 7.210, and MA as 0.888; 6.942. There was a significant difference in MA (p = 0.0184) between the healthy individuals (mean 36.299) and the OAG group (mean 24.944), which was not observed for MV (p = 0.0735) or MT (p = 0.1351).

CONCLUSION

High reproducibility of measurements of peripapillary blood flow with the LSFG system was demonstrated in both groups. There is a difference in peripapillary blood flow between healthy controls and OAG patients.

Macular Oxygen Saturation in Glaucoma Using Retinal Oximetry of Visible Light Optical Coherence Tomography: A Pilot Study

Purpose

A cross-sectional pilot study to compare macular oxygen saturation (sO2) and associated clinical measurements between normal and glaucoma subjects and to evaluate whether macular sO2 can be a diagnostic metric for early-stage glaucoma.

Methods

Forty-eight eyes of 35 subjects from three groups were included: normal subjects (16 eyes, 10 subjects), suspect/pre-perimetric glaucoma (GS/PPG) subjects (17 eyes, 12 subjects), and perimetric glaucoma (PG) subjects (15 eyes, 13 subjects). We performed retinal oximetry of visible light optical coherence tomography (VIS-OCT) in macular vessels, with 512 × 256 sampling points over a 5 × 5 mm2 area. Zeiss Cirrus OCT scans and a 24-2 visual field test (VFT) were conducted. Statistical analysis was conducted.

Results

Significant differences were observed among the three groups for all VIS-OCT, Zeiss OCT, and VFT variables. As glaucoma severity increased, macular AsO2 (arterial sO2) and A-V sO2 (arteriovenous sO2 difference) decreased, whereas macular VsO2 (venous sO2) increased. Macular AsO2 and A-V sO2 were found to be statistically correlated with ganglion cell layer + inner plexiform layer (GCL+IPL) and circumpapillary retinal nerve fiber layer in all eyes, as well as in PG eyes. Within the PG group, a dominant correlation between AsO2 and ganglion cell layer + inner plexiform layer was observed in the more damaged lower hemifield.

Conclusions

Glaucoma subjects showed altered macular sO2, indicating reduced oxygen consumption. The sO2 measured by VIS-OCT could be a potential metric for early glaucoma diagnosis.

Translational Relevance

This study shows macular sO2 measurements via VIS-OCT could bridge advanced imaging technology and clinical glaucoma detection.

Assessment of choriocapillaris and choroidal vascular changes in posterior uveitis using swept-source wide-field optical coherence tomography angiography

PURPOSE

To evaluate choriocapillaris (CC) and choroidal vascular changes in patients with posterior uveitis using swept-source (SS) wide-field optical coherence tomography angiography (OCTA).

METHOD

Consecutive patients with posterior uveitis were evaluated using 3×3 mm and 12×12 mm OCTA scan patterns and montage images of 5×12×12 mm or 2×15×9 mm, covering approximately 70°-90° of the retina. The images were quantitatively and qualitatively analysed and compared with healthy controls.

RESULTS

Eighty-six eyes of 56 patients with posterior uveitis (mean age 45.2±19.9 years; 58.9% female), and 38 eyes of 19 age-matched healthy controls (57.9% female) were included. The mean perfusion density (PD) in 3×3 mm and 12×12 mm CC scans was significantly lower in eyes with posterior uveitis compared with those of healthy controls. However, no significant difference in the mean PD of choroidal scans was found comparing eyes with posterior uveitis and healthy controls. The mean PD in eyes with active disease was significantly higher compared with the inactive eyes on 12×12 mm choroidal scans (55.61% vs 51.25%, p=0.02), while no difference was found in the CC slabs.

CONCLUSION

CC and choroidal assessment using OCTA provides useful information in patients with posterior uveitis. SS-OCTA metrics of the CC and choroidal slabs are promising tools in uveitis patients in the future.

TRIAL REGISTRATION NUMBER

NCT02811536.

Optical Coherence Tomography Angiography Vessel Density in Healthy, Glaucoma Suspect, and Glaucoma Eyes

Purpose

To evaluate and compare optical coherence tomography angiography (OCTA) retinal vasculature measurements in healthy, glaucoma suspect (GS), and glaucoma patients.

Methods

One hundred fourteen eyes with good quality OCTA pictures were included from 38 healthy, 38 GS, 38 and primary open-angle glaucoma (POAG) participants. The information on retinal vasculature was summarized as a vessel density map and as vessel density (%), which is the fraction of the flowing vessel area over the total area examined. The superior, inferior, nasal, and temporal quadrants, as well as whole vessel density (wVD) and peripapillary vessel density (ppVD) data taken from the retinal nerve fiber layer, were studied. Global indices of the visual field were correlated with vessel density measurement.

Results

Mean vessel density was significantly lower in POAG eyes compared with GSs and healthy eyes (wVD) 45.34% ± 6.64%, 50.06% ± 1.97% and 53.06% ± 2.12%, respectively (P < 0.001), and ppVD 47.42% ± 7.73%, 47.42% ± 7.73% and 56.074% ± 2.71%, respectively (P < 0.001). A linear relationship between vessel density (wVD and ppVD) and global indices of the visual field (mean deviation [MD] and pattern standard deviation [PSD]) shows a significant (P < 0.001) relation. R 2 is 0.35, 0.36, 0.36, and 0.33 for MD and wVD, PSD and wVD, PSD and ppVD, and MD and ppVD, respectively.

Conclusions

For distinguishing between healthy and glaucoma eyes, OCTA vessel density demonstrated near similar diagnostic accuracy to visual field tests. These findings imply that OCTA measurements reflect damage to tissues important in the pathogenesis of POAG.

Macular Oxygen Saturation in Glaucoma Using Retinal Oximetry of Visible Light Optical Coherence Tomography

Purpose

Oxygen saturation (sO2) plays a critical role in retinal pathophysiology, especially at the macula, which undergoes significant energy consumption. While macular damage has been suggested to be involved in early-stage glaucoma, there has been no report to date on non-invasive macular sO2 in glaucoma. Therefore, we conducted this study to compare macular sO2 associated with other clinical measurements between normal and glaucoma subjects and evaluate whether there are significant differences.

Method

This is a cross-sectional study. We used visible light optical coherence tomography (VIS-OCT) for retinal oximetry in perifoveal vessels. The subjects from groups of normal, suspect/pre-perimetric glaucoma (GS/PPG) and perimetric glaucoma (PG) were scanned using VIS-OCT in the macular region with a sampling density of 512×256 in an area of 5×5 mm2. 48 eyes (16 normal, 17 GS/PPG and 15 PG) were included for the analysis. For each eye, we measured the sO2 of arterioles (AsO2), venules (VsO2), and calculated the difference between arterioles and venules (A-V sO2=AsO2-VsO2), oxygen extraction (OE=(AsO2-VsO2)/AsO2 ×100%). Additionally, we included Zeiss Cirrus OCT scans and 24-2 visual field test (VFT) for clinical benchmark. One-way ANOVA was used to compare the differences among the three groups. Spearman correlation tests were used for correlation sO2 markers to standard metrics including the thickness of ganglion cell layer and inner plexiform layer (GCL+IPL), circumpapillary retinal nerve fiber layer (cpRNFL) and mean deviation (MD) in VFT.

Result

Significant differences were found among three groups for all VIS-OCT, Zeiss OCT, and VFT variables. Macular AsO2, A-V sO2, OE decreased, and VsO2 increased along with severity. Macular AsO2 and A-V sO2 were statistically correlated with GCL+IPL and cpRNFL in all eyes, as well as only PG eyes. Within PG eyes, the correlation between AsO2 and GCL+IPL is dominant in more damaged lower hemifield.

Conclusion

The GS/PPG and PG subjects had significantly higher macular VsO2, lower A-V sO2 and OE indicating less oxygen consumption. The sO2 measured by retinal oximetry of VIS-OCT can be a potential metric for the early diagnosis of glaucoma.

Dual wavelength retinal vessel oximetry - influence of fundus pigmentation

BACKGROUND

Clinical methods examining oxygenation parameters in humans have been used in many different care settings, but concerns have been raised regarding their clinical utility when assessing people with darker skin pigmentation. While saturation values can be crucial in emergency medicine, they are equally valuable in assessing disease mechanisms and monitoring change in disease progression. Retinal pigmentation varies across individuals and hence, can impact on retinal oxygen parameters. The objective of this study was to quantify and eliminate the impact of retinal pigmentation on retinal vessel oxygen saturation parameters measured in the superficial retinal arterioles and venules.

METHODS

105 healthy individuals of varying skin colour, iris colour and heritage were included. Following a full eye exam to exclude any ocular abnormality, all participants underwent intraocular pressure, systemic blood pressure measurements and dilated dual wavelength retinal photography. Rotation matrices were employed to minimise the dependency of retinal pigmentation on arterial and venous oxygen saturation measurements determined in a concentric measurement annulus.

RESULTS

Retinal oxygen saturation in venules showed a linear correlation with retinal pigmentation (y = 0.34 × x + 38.598), whereas arterial saturation followed a polynomial pattern (y = 0.0089 × x2 + 0.7499 × x + 85.073). Both arterial and venous saturation values were corrected using local fundus pigmentation. Pre-correction retinal arterial and venous oxygen saturation were 89.0% (±13.1) and 43.7% (±11.5), respectively, and post- correction values were 94.8% (±8.7) for arteries and 56.3% (±7.0) veins.

CONCLUSIONS

When assessing multi-ethnic cohorts, it is important to consider the impact of pigmentation on imaging parameters and to account for it prior to clinical interpretation.

Peripapillary Oxygenation and Retinal Vascular Responsiveness to Flicker Light in Primary Open Angle Glaucoma

The aim of our study was to evaluate peripapillary oxygenation and its relationship to retinal vascular responsiveness to flicker light in patients with primary open angle glaucoma (POAG). Retinal vessel oxygen saturation was measured in 46 eyes of 34 Caucasian patients with POAG and in 21 eyes of 17 age-matched controls using the oximetry tool of Retinal Vessel Analyser (RVA: IMEDOS Systems UG, Jena, Germany). The mean oxygen saturation of the major arterioles (A-SO2; %) and venules (V-SO2; %), as well as the corresponding arterio−venular difference (A-V SO2; %), were calculated. We also measured retinal vascular responsiveness (RVR) to flicker light by means of RVA. Glaucoma patients were divided in two subgroups according to their median arteriolar and venular vascular responsiveness to flicker light (AFR and VFR). Glaucomatous damage was assessed by optical coherence tomography (Carl Zeiss Meditec, Dublin, CA, USA) and static automated perimetry (Octopus, program G2/standard strategy: Haag-Streit International, Köniz, Switzerland). In addition, we calculated the mean peripapillary oxygen exposure [ppO2E; %/µm] by dividing the mean A-V SO2 with the mean retinal nerve fibre layer (RNFL) thickness. In glaucoma patients, A-SO2 and V-SO2 values were significantly increased, and their difference decreased when compared to controls (p < 0.017; linear mixed-effects model). Grouped with respect to retinal vascular responsiveness to flicker light, subjects with reduced VFR (≤2.9%) had significantly higher ppO2E (0.49 ± 0.08%/µm, respectively, 0.43 ± 0.06%/µm; p = 0.027). Additionally, higher ppO2E in glaucoma patients correlated negatively with the neuroretinal rim area (p < 0.001) and the RNFL thickness (p = 0.017), and positively with the mean defect of the visual field (p = 0.012). Reduced venular vascular responsiveness in our glaucoma patients was associated with increased peripapillary oxygenation exposure. Thus, ganglion cells and their axons in glaucomatous eyes with reduced retinal vascular responsiveness are prone to be more exposed to higher oxidative stress, probably contributing to the further progression of glaucomatous damage.

Optic disc blood perfusion and oxygenation in glaucoma

Purpose

To investigate the haemoglobin concentration and oxygenation in the optic disc in glaucoma patients vs. controls.

Methods

Thirty-one eyes of primary open angle glaucoma patients (mean age: 64.9 ± 2.1 years) and 31 eyes of 31 healthy controls (65.5 ± 2.0 years) were included. Perimetry, optical coherence tomography (OCT), and OCT angiography were performed. Multispectral imaging was used to record the optic disc reflectance at wavelengths 522 nm, 548 nm, 555 nm, 586 nm, and 610 nm, and haemoglobin concentration and oxygenation (SO₂) were calculated from these measures. This was done in the rest and under stimulation of neuronal activity by flicker light.

Results

The haemoglobin concentration was significantly lower (p < 0.001) in the rim (40.0 ± 6.3) and the excavation (35.7 ± 8.0) of the glaucoma patients’ discs than in controls (45.7 ± 7.5). SO₂ was not different in general, but lower in a subgroup of 18 glaucoma patients with ischaemic disc rims than in non-ischaemic ones (median 26.8%, interquartile range (IQR): 29.5% vs. 51.9%, IQR 32.0%, p = 0.02) as well as in controls (41.0%, IQR 30.6%, p = 0.01). Flicker light stimulation significantly increased the haemoglobin concentration in the controls (+ 1.3 ± 3.6, p = 0.048) as well as in the rim of glaucoma discs (+ 2.6 ± 5.0, p = 0.006) and SO₂ in the controls only (+ 15.4 ± 23.6%, p = 0.001). The haemoglobin concentration was significantly correlated with the perimetric mean defect, retinal nerve fibre layer (RNFL) thickness and para-papillary perfusion density.

Conclusions

The optic disc haemoglobin concentration and oxygenation are quantifiable from multispectral imaging and reduced in glaucoma. The correlation of haemoglobin concentration with perfusion density, RNFL thickness and visual field loss indicates its implication in glaucoma pathology.

Effect of Latanoprostene Bunod on Optic Nerve Head Blood Flow

SIGNIFICANCE

Topical latanoprostene bunod increases capillary oxygen saturation and blood volume at the optic nerve head in healthy individuals.

PURPOSE

This study aimed to evaluate the effect of topical latanoprostene bunod on optic nerve blood volume and oxygen saturation in a population of healthy participants.

METHODS

In this prospective double-blind crossover study, 23 healthy participants aged from 21 to 62 years were recruited. Optic nerve head capillary blood volume (ONHvol) and oxygen saturation (ONHSaO2) baselines were measured over a period of 2 hours using multichannel spectroscopic reflectometry and were remeasured after a 7-day once-daily instillation regimen of either latanoprost 0.005% or latanoprostene bunod 0.024%. After a 30-day washout period, participants were crossed over to the alternate product, and measurements were repeated. Participants were used as their own baselines to calculate variation in ONHvol and ONHSaO2 across time and pharmacological agents. The Friedman test was used to establish significant differences in optic nerve head parameters from baseline values, and Conover post hoc analysis was carried for multiple between-group comparisons.

RESULTS

Latanoprostene bunod 0.024% induced a significant increase of 4% in ONHSaO2 compared with latanoprost 0.005% (P < .001). Furthermore, latanoprostene bunod increased ONHvol levels by more than twofold at all time points (P < .001 at T60, T90, and T120). The increase in ONHvol was 66.2% higher than levels achieved with latanoprost at T60 (P = .001), 47% higher at T90 (P < .001), and 45% higher at T120 (P < .01).

CONCLUSIONS

Latanoprostene bunod 0.024% induces a significant increase in optic nerve head blood volume and oxygen saturation in healthy subjects, when compared with latanoprost 0.005%. Future studies are needed to evaluate whether similar responses are elicited in patients suffering from glaucomatous optic neuropathy.

Clinical implications of retinal oximetry in retinal vein occlusion: a review

PURPOSE

To review retinal vein occlusion (RVO) and its relationship with retinal oxygen saturation via automated retinal oximetry in eyes with RVO.

METHODS

A literature review was performed in PubMed and Medline until October 2021 utilizing specific keywords and cross-matched reference lists.

RESULTS

This review found RVO to be associated with risk factors including age, hypertension, cardiovascular and metabolic conditions, male gender, and glaucoma. These may be attributed to a breakdown of regulatory processes in the retina. Retinal venous oxygen saturation (SvO₂ ) and arteriovenous difference in eyes with central RVO have been found to be reduced and elevated, respectively. The literature indicates variable and contradictory findings in regard to branch RVO and retinal oxygenation. Additionally, ischaemic eyes have been found to have elevated retinal arterial oxygen saturation; however, the literature reports variable results regarding SvO₂ levels. Medications have been suggested to increase SvO₂ in RVO eyes, which may represent an important mechanism for disease management. Ranibizumab is currently the most studied drug for retinal oxygenation in RVO and has been suggested to increase SvO₂ in RVO eyes. In comparison, dexamethasone was found to decrease SvO₂ .

CONCLUSION

The current literature on retinal oxygenation in ischaemic subtypes of RVO and in drug therapies is minimal, and further work is required to expand upon our understanding of how ischaemia and drugs influence retinal oxygenation and clinical outcomes.

Simultaneous segmentation and classification of the retinal arteries and veins from color fundus images

BACKGROUND AND OBJECTIVES

The study of the retinal vasculature represents a fundamental stage in the screening and diagnosis of many high-incidence diseases, both systemic and ophthalmic. A complete retinal vascular analysis requires the segmentation of the vascular tree along with the classification of the blood vessels into arteries and veins. Early automatic methods approach these complementary segmentation and classification tasks in two sequential stages. However, currently, these two tasks are approached as a joint semantic segmentation, because the classification results highly depend on the effectiveness of the vessel segmentation. In that regard, we propose a novel approach for the simultaneous segmentation and classification of the retinal arteries and veins from eye fundus images.

METHODS

We propose a novel method that, unlike previous approaches, and thanks to the proposal of a novel loss, decomposes the joint task into three segmentation problems targeting arteries, veins and the whole vascular tree. This configuration allows to handle vessel crossings intuitively and directly provides accurate segmentation masks of the different target vascular trees.

RESULTS

The provided ablation study on the public Retinal Images vessel Tree Extraction (RITE) dataset demonstrates that the proposed method provides a satisfactory performance, particularly in the segmentation of the different structures. Furthermore, the comparison with the state of the art shows that our method achieves highly competitive results in the artery/vein classification, while significantly improving the vascular segmentation.

CONCLUSIONS

The proposed multi-segmentation method allows to detect more vessels and better segment the different structures, while achieving a competitive classification performance. Also, in these terms, our approach outperforms the approaches of various reference works. Moreover, in contrast with previous approaches, the proposed method allows to directly detect the vessel crossings, as well as preserving the continuity of both arteries and veins at these complex locations.

The Oxygen Saturation in the Retinal Vessels of Patients with Diabetes Mellitus - The Search for Determinants

INTRODUCTION

Oxygen saturation in retinal vessels can be non-invasively measured by a new method. It is known that oxygen extraction is lowered in diabetic retinopathy. In the cross-sectional study presented here, it has been investigated whether diabetes-specific changes and typical accompanying diseases could influence the measurements in oximetry.

METHODS

During a time span of seven months, 237 diabetics were included who came to our clinic for an intravitreal injection, who met the inclusion criteria, who showed no exclusion criteria and gave written consent. 203 eyes of 203 patients were evaluated. The oxygen saturation (SO₂) was measured in digitally stored fundus images taken with a fundus camera (Zeiss 450 FF). In its illumination beam, a filter was inserted with two transmission maxima at 584 nm and 610 nm (bandwidth 10 nm). With the program "Oximetry" (Imedos Systems, Jena, Germany), the venous and the arterial SO₂ were measured. The values were stored in a spreadsheet. Diabetes related data, clinical parameters and accompanying diseases were documented in the predefined scheme.

RESULTS

In the whole patient sample, the venous oxygen saturation (vSO₂) was 69 ± 12%, the arterial saturation (aSO₂) 99 ± 5% and the arteriovenous difference 31 ± 9%. A statistically significant association could not be found between the measurement values of oximetry and age, duration of diabetes and HbA_1c (Kruskal-Wallis ANOVA: p > 0.05). The differences in oximetry values between units of analysis (UOA) e.g. hypertension and non units of analysis (NOA) were analysed. Descriptive testing showed significant differences in vSO₂% in the following units of analysis: allergies n = 47: UOA vs. NUOA: 64 vs. 69 (t test: p < 0.02); sartan therapy n = 46: UOA vs. NUOA: 64 vs. 70 (p = 0.003); state after intravitreal injections n = 144: 68 vs. 71 (p = 0.02). Oxygen extraction from the retinal capillaries was improved in patients who had injections with VEGF inhibitors. An improvement in oxygen extraction is also seen in patients with allergies and hypertension in comparison to those without these units of analysis.

CONCLUSION

The clinically observed improvement in the retinal state situation after intravitreal injections with VEGF inhibitors may be seen in association with the improved oxygen extraction. It seems reasonable to study whether a switch to sartans in the therapy of hypertension may improve retinal function.

Morphologic Analysis of Peripapillary Retinal Arteriole Using Adaptive Optics in Primary Open-angle Glaucoma

PURPOSE

The purpose of this study was to better understand the role of vascular risk factors in the pathogenesis of primary open-angle glaucoma (POAG), a detailed analysis of retinal arterial wall thickness is needed. The purpose of the present study was to make a morphologic analysis of peripapillary arteriole in POAG using adaptive optics (AO) technology.

PATIENTS AND METHODS

We included otherwise healthy subjects with an isolated confirmed diagnosis of bilateral POAG. Patients' clinical characteristics were noted. AO imaging followed by a complete ophthalmic examination was performed. A single operator masked to clinical data performed 5 measurements at different locations of each analyzed vessel. For each location, lumen diameter and wall thickness were measured. Total diameter, wall-to-lumen ratio (WLR), and whole cross-sectional area were calculated.

RESULTS

Lumen diameter and total diameter were significantly lower in the glaucoma group (n=31) than in the control group (n=29): [median (interquartile ranges)] 88.3 (82.6-99.2) versus 102.3 (87-113.1) (P=0.03) and 121.1 (109.3-130.5) versus 134.4 (112.7-144.4), respectively (P=0.015). Wall thickness, WLR, and whole cross-sectional area were not significantly different. Apart from a significantly higher WLR in subjects with reported high cholesterol levels, we did not observe any correlation between patients' clinical characteristics and any of the parameters.

CONCLUSIONS

We observed in POAG a narrowing of the arteriolar lumen without modification of the vessel wall thickness. To date, it is the first time that these data are obtained using AO. This suggests that the vascular risk factor in POAG only reduces the vascular caliber without inducing any patent atherosclerosis of the retinal arterial wall.

Retinal oximetry in glaucoma: investigations and findings reviewed

Abnormalities of the retinal blood supply have been widely implicated in primary open-angle glaucoma (POAG). Impaired blood supply to the retina and optic nerve head (ONH) may be a primary pathophysiologic mechanism contributing to POAG ('vascular hypothesis'). However, the decreased metabolic activity of atrophic tissue is itself known to induce both vascular changes and decreased blood flow due to reduced oxygen demand. Therefore, primary nonvascular factors could potentially induce glaucomatous atrophy, with subsequent secondary vascular pathology ('mechanical hypothesis'). Retinal oximetry holds great promise in the investigation of glaucoma pathogenesis, as it can provide useful data on retinal metabolic oxygen demand, especially when combined with measurements of retinal blood flow. This review surveys the research on retinal metabolism in POAG using spectroscopic retinal oximetry. The use of mathematical models in combination with oximetric data to investigate the role of retinal metabolism and oxygen supply in POAG is also discussed.

Ocular blood flow as a clinical observation: Value, limitations and data analysis

Alterations in ocular blood flow have been identified as important risk factors for the onset and progression of numerous diseases of the eye. In particular, several population-based and longitudinal-based studies have provided compelling evidence of hemodynamic biomarkers as independent risk factors for ocular disease throughout several different geographic regions. Despite this evidence, the relative contribution of blood flow to ocular physiology and pathology in synergy with other risk factors and comorbidities (e.g., age, gender, race, diabetes and hypertension) remains uncertain. There is currently no gold standard for assessing all relevant vascular beds in the eye, and the heterogeneous vascular biomarkers derived from multiple ocular imaging technologies are non-interchangeable and difficult to interpret as a whole. As a result of these disease complexities and imaging limitations, standard statistical methods often yield inconsistent results across studies and are unable to quantify or explain a patient's overall risk for ocular disease. Combining mathematical modeling with artificial intelligence holds great promise for advancing data analysis in ophthalmology and enabling individualized risk assessment from diverse, multi-input clinical and demographic biomarkers. Mechanism-driven mathematical modeling makes virtual laboratories available to investigate pathogenic mechanisms, advance diagnostic ability and improve disease management. Artificial intelligence provides a novel method for utilizing a vast amount of data from a wide range of patient types to diagnose and monitor ocular disease. This article reviews the state of the art and major unanswered questions related to ocular vascular anatomy and physiology, ocular imaging techniques, clinical findings in glaucoma and other eye diseases, and mechanistic modeling predictions, while laying a path for integrating clinical observations with mathematical models and artificial intelligence. Viable alternatives for integrated data analysis are proposed that aim to overcome the limitations of standard statistical approaches and enable individually tailored precision medicine in ophthalmology.

Relating glaucomatous visual field loss to retinal oxygen delivery and metabolism

PURPOSE

To test the hypothesis that visual field (VF) loss is associated with decreased retinal blood flow, oxygen delivery and metabolism, and with increased retinal oxygen extraction fraction (OEF) in glaucomatous individuals.

METHODS

Glaucomatous subjects underwent automated perimetry, dual wavelength scanning laser ophthalmoscopy and Doppler optical coherence tomography imaging in order to measure mean deviation, central retinal arterial and venous diameter equivalents (CRAE and CRVE), oxygen saturation levels and total retinal blood flow (TRBF), respectively. Retinal oxygen delivery, metabolism and extraction fraction were derived from measurements of oxygen saturation and blood flow.

RESULTS

Twenty eyes of 14 subjects were included in the study. Mean deviation was on average -13.76 ± 8.70 dB. Mean CRAE and CRVE were 126 ± 28 and 191 ± 35 μm, respectively. Mean TRBF and oxygen delivery were 34.3 ± 11.7 μl/min and 6.5 ± 2.6 μl O₂ /min, respectively. Mean oxygen metabolism and extraction fraction were 2.1 ± 0.94 μl O₂ /min and 0.34 ± 0.15, respectively. Visual field (VF) loss was associated with reduced CRAE and CRVE, TRBF and oxygen delivery and associated with increased OEF.

CONCLUSION

Combined measurements of retinal blood flow and oxygen saturation in glaucomatous individuals suggest VF loss is associated with impaired oxygen delivery and augmented OEF.

Artery-vein segmentation in fundus images using a fully convolutional network

Epidemiological studies demonstrate that dimensions of retinal vessels change with ocular diseases, coronary heart disease and stroke. Different metrics have been described to quantify these changes in fundus images, with arteriolar and venular calibers among the most widely used. The analysis often includes a manual procedure during which a trained grader differentiates between arterioles and venules. This step can be time-consuming and can introduce variability, especially when large volumes of images need to be analyzed. In light of the recent successes of fully convolutional networks (FCNs) applied to biomedical image segmentation, we assess its potential in the context of retinal artery-vein (A/V) discrimination. To the best of our knowledge, a deep learning (DL) architecture for simultaneous vessel extraction and A/V discrimination has not been previously employed. With the aim of improving the automation of vessel analysis, a novel application of the U-Net semantic segmentation architecture (based on FCNs) on the discrimination of arteries and veins in fundus images is presented. By utilizing DL, results are obtained that exceed accuracies reported in the literature. Our model was trained and tested on the public DRIVE and HRF datasets. For DRIVE, measuring performance on vessels wider than two pixels, the FCN achieved accuracies of 94.42% and 94.11% on arteries and veins, respectively. This represents a decrease in error of 25% over the previous state of the art reported by Xu et al. (2017). Additionally, we introduce the HRF A/V ground truth, on which our model achieves 96.98% accuracy on all discovered centerline pixels. HRF A/V ground truth validated by an ophthalmologist, predicted A/V annotations and evaluation code are available at https://github.com/rubenhx/av-segmentation.

Cataract significantly influences quantitative measurements on swept-source optical coherence tomography angiography imaging

Purpose

To analyze retinal blood flow before and after cataract surgery using swept-source optical coherence tomography angiography (SS-OCTA).

Methods

Prospective observational study. Consecutive patients were recruited and scanned using SS-OCTA before and after cataract surgery. Laser flare photometry were performed post-surgery. Perfusion and vessel density of superficial (SCP) and deep capillary plexuses (DCP) of the 3 × 3 mm images as well as foveal avascular zone (FAZ) measurements were assessed. Vessel continuity, vessel visibility and presence of artefacts were evaluated by two blinded graders using a predefined grading protocol.

Results

Thirteen eyes of 12 patients met the inclusion criteria. There was a significant increase of perfusion and vessel densities in both the SCP and the DCP after cataract surgery within the 3 × 3 mm images. Significantly better distinguishability of FAZ border was observed postoperatively in both SCP and DCP, however, FAZ area and perimeter measurements did not significantly change after cataract surgery. Mean number of motion artifacts in SCP and DCP numerically decreased by 37% (P = .089) and 42% (P = .080).

Conclusions

Lens opacities have a significant influence on retinal blood flow measurements in SS-OCTA and should be considered in quantitative vessel analysis. Inflammation may also impact the assessment of density parameters. FAZ measurements seems to be the most robust parameters in terms of media opacity.

Vessel density in OCT angiography permits differentiation between normal and glaucomatous optic nerve heads

AIM

To evaluate whether optical coherence tomography angiography (OCTA) can detect altered vessel density (VD) at the optic nerve head (ONH) in glaucoma patients. Special attention is paid to the accuracy of the OCTA technique for distinguishing healthy from glaucomatous eyes.

METHODS

A total of 171 eyes were examined by the OCTA system AngioVue™ (Optovue): 97 eyes diagnosed with glaucoma and 74 healthy control eyes. The papillary and peripapillary VD was measured. Furthermore, the VD was correlated with different structural and functional measurements. In order to test the accuracy of differentiation between eyes with and without glaucoma, we calculated the receiver operating characteristic curve (ROC) and the area under the curve (AUC).

RESULTS

The papillary and peripapillary VD in glaucomatous eyes was significantly lower than in healthy eyes (P<0.05). The VD of the nasal peripapillary sector was significantly lower than in the other sectors. The further the disease had progressed [measured by determining the thickness of the ganglion cell complex (GCC) and the retinal nerve fiber layer (RNFL)] the greater the VD reduction. The AUC discriminated well between glaucomatous and normal eyes (consensus classifier 94.2%).

CONCLUSION

OCTA allows non-invasive quantification of the peripapillary and papillary VD, which is significantly reduced in glaucomatous eyes and accurately distinguishes between healthy and diseased eyes. OCTA expands the spectrum of procedures for detecting and monitoring glaucoma.

Assessing oxygen saturation in retinal vessels in high myopia patients pre- and post-implantable collamer lens implantation surgery

PURPOSE

To determine whether posterior chamber phakic implantable collamer lens (ICL) surgery in high myopia patients impedes oxygen saturation of retinal vessels.

METHODS

Mean oxygen saturation and diameter in retinal blood vessels were measured before and after ICL implantation surgery to correct high myopia refractive errors (i.e. -6.00 to -20.25 dioptres [D]), using an Oxymap T1 retinal oximeter.

RESULTS

In 17 eyes of 17 patients, the Oxymap T1 retinal oximeter detected a small but significant decrease in oxygen saturation of retinal venules, 1-week postoperatively (compared to preoperative measurements). Moreover, at 1 week after ICL implantation, the diameter of patient retinal vessels had consistently contracted, compared to preoperative measurements. By 1 month after ICL surgery, however, both the oxygen saturation and retinal vessel diameter had returned to preoperative levels. Otherwise, no statistically significant difference in oxygen saturation and diameter of retinal arterioles was found when comparing their measurements before and 1 week after implantation.

CONCLUSION

Stable levels of oxygen saturation in retinal vessels, as detected by the Oxymap T1 oximeter, show ICL implantation would not leave lasting impact or adverse effects to retina oxygen saturation in high myopia patients.

Retinal and choroidal oxygen saturation of the optic nerve head in open-angle glaucoma subjects by multispectral imaging

The aim of this study was to determine whether differences exist in oxygen supply to the optic nerve head (ONH) from the retinal and choroidal vascular layers in patients with primary open angle glaucoma (POAG) using multispectral imaging (MSI).This ia an observational, cross-sectional study.Multispectral images were acquired from 38 eyes of 19 patients with POAG, and 42 healthy eyes from 21 matched volunteers with Annidis' RHA multispectral digital ophthalmoscopy. Superficial and deeper oxygen saturation of the optic disc was represented by the mean gray scale values on the retinal and choroidal oxy-deoxy maps, respectively. Statistical analysis was performed to detect differences in ONH oxygen saturation between the 2 groups. Oxygen saturation levels in the eyes of POAG patients with severe glaucoma were compared to those of fellow eyes from the same subjects. Linear correlation analysis was performed to assess the association between ONH oxygen saturation and systemic and ocular parameters.No statistical difference was found in retinal and choroidal oxygen saturation between the POAG and control groups. In the glaucoma patients, retinal oxygen saturation was lower for eyes with worse visual fields than in those with good visual fields (t = 4.009, P = 0.001). In POAG patients, retinal oxygen saturation was dependent on mean defect of visual field and retinal nerve fiber layer thickness (RNFLT) (r = 0.511, 0.504, P = 0.001, 0.001, respectively), whereas the choroid vasculature oxygen saturation was inversely related to RNFLT (r = -0.391, P = 0.015). An age-dependent increase in retinal oxygen saturation was found for both the POAG and control groups (r = 0.473, 0.410, P = 0.007, 0.003, respectively).MSI revealed a significant correlation between functional and structural impairments in glaucoma and retinal oxygen saturation. MSI could provide objective assessments of perfusion impairments of the glaucomatous ONH. This is a preliminary indication of the effectiveness of MSI for studying POAG.

Inter-eye comparison of retinal oximetry and vessel caliber between eyes with asymmetrical glaucoma severity in different glaucoma subtypes

Background

To compare retinal vessel oxygenation and vessel caliber in primary angle-closure glaucoma (PACG), primary open-angle glaucoma (POAG), normal-tension glaucoma (NTG), and normal controls, as well as between eyes of asymmetrical glaucoma severity.

Methods

This was a prospective, cross-sectional study. The 159 subjects (PACG, n=39; POAG, n=41; NTG, n=41; normal controls, n=38) underwent retinal oxygen saturation measurements using the Oxymap T1 Retinal Oximeter, optical coherence tomography, and Humphrey visual field testing. Retinal oxygen saturation and vessel diameter were compared between the glaucoma groups and normal controls, as well as between eyes of asymmetrical glaucoma severity. Kruskal–Wallis test was performed for comparison among different subtypes of glaucoma. Wilcoxon signed-rank test was used to compare the inter-eye differences.

Results

Compared to normal controls, arteriolar oxygen saturation was increased in PACG eyes (P=0.048) but not in POAG or NTG eyes. There were no significant differences in oxygen saturation in venules or arteriovenous (AV) difference in all three glaucoma groups. Venular diameter was significantly reduced in all glaucoma groups compared to normal controls (P<0.001), but no such change was observed in arteriolar diameter (P=0.10). When comparing between eyes of asymmetrical glaucoma severity, arteriolar oxygen saturation (P=0.03) and AV difference (P=0.04) were significantly higher, while arteriolar diameter was significantly lower (P=0.001) in the worse eye in PACG group. There were no significant differences in oximetric parameters or vessel calibers between the worse and the better eyes in POAG and NTG groups.

Conclusion

Eyes with PACG showed increased arteriolar oxygen saturation and increased AV difference. This was not observed in POAG and NTG eyes. Arteriolar diameter in PACG and venular diameter in all three glaucoma groups were reduced. The difference observed in PACG eyes may be due to an increased metabolic demand in the disease process compared to open-angle glaucoma.

Aqueous humor concentration of VEGF and retinal oxygen saturation after unilateral acute primary angle closure

PURPOSE

To measure the levels of vascular endothelial growth factor (VEGF) in aqueous humor in patients with unilateral acute primary angle closure (APAC) and retinal oxygen saturation (SO2 ) after trabeculectomy.

METHODS

Twelve patients, with unilateral trabeculectomy-required APAC, were recruited as the study group. Aqueous humor samples were collected prospectively for every subject. VEGF concentrations were analysed via enzyme-linked immunosorbent assay (ELISA), and retinal SO2 was measured using oximeter (Reykjavik, Iceland) at 1 month after trabeculectomy.

RESULTS

In the APAC eyes, the mean aqueous humor concentration of VEGF was 388.4 ± 260.1 pg/ml and positively correlated with preoperative intraocular pressure (ρ = 0.658, p = 0.020). In the normal controls, the mean SO2 levels in the larger arterioles and venules were 93.0 ± 5.9% and 59.5 ± 5.4% and the arteriovenous [A-V] difference was 33.5 ± 6.3%. In the study group, the arteriolar SO2 at 1 month after surgery (97.6 ± 19.4%, p = 0.147) did not differ significantly from the values in the control group. However, the mean venular SO2 was significantly lower than the control eyes (50.0 ± 9.1% versus 59.5 ± 5.4%, p = 0.004), and a remarkable increased A-V difference (47.6 ± 22.7% versus 33.5 ± 6.3%) was found in the APAC eyes (p = 0.037).

CONCLUSIONS

Lower venular SO2 and increased A-V difference existed in the APAC eyes after surgery, and it is possible that the saturation was even lower during the attack. Together with the elevated VEGF concentrations in aqueous humor, these provided indirect evidence that ocular hypoxia was constantly present during APAC attack.