Ophthalmic Surgery, Lasers and Imaging Retina

Clinical Science 

Normative Data on Foveal Avascular Zone Dimensions in Four Macular Vascular Layers

Roy Schwartz, MD; Akanksha Bagchi, MBBS, MS; Adam Dubis, PhD; Philip Hykin, FRCOphth; Sobha Sivaprasad, FRCOphth

Abstract

BACKGROUND AND OBJECTIVES:

To provide normative data on the size of the foveal avascular zone (FAZ) in the four histological retinal vascular layers as measured using optical coherence tomography angiography (OCTA) and evaluate interobserver variability.

PATIENTS AND METHODS:

Two graders measured the FAZ area in each layer (nerve fiber layer [NFL], ganglion cell layer [GCL], inner plexiform layer to inner nuclear layer [INL], and INL to outer plexiform layer), as well as the superficial and deep capillary plexuses.

RESULTS:

Forty-seven eyes of 25 subjects were included. The FAZ was not clearly delineated in the NFL or GCL layers. There was high agreement between measurements of the FAZ area of the different layers. Inter- and intraobserver agreements were high.

CONCLUSION:

This is the first study to measure FAZ size across the retina. There was a high correlation between FAZ sizes among the different layers on OCTA.

[Ophthalmic Surg Lasers Imaging Retina. 2018;49:580–586.]

Abstract

BACKGROUND AND OBJECTIVES:

To provide normative data on the size of the foveal avascular zone (FAZ) in the four histological retinal vascular layers as measured using optical coherence tomography angiography (OCTA) and evaluate interobserver variability.

PATIENTS AND METHODS:

Two graders measured the FAZ area in each layer (nerve fiber layer [NFL], ganglion cell layer [GCL], inner plexiform layer to inner nuclear layer [INL], and INL to outer plexiform layer), as well as the superficial and deep capillary plexuses.

RESULTS:

Forty-seven eyes of 25 subjects were included. The FAZ was not clearly delineated in the NFL or GCL layers. There was high agreement between measurements of the FAZ area of the different layers. Inter- and intraobserver agreements were high.

CONCLUSION:

This is the first study to measure FAZ size across the retina. There was a high correlation between FAZ sizes among the different layers on OCTA.

[Ophthalmic Surg Lasers Imaging Retina. 2018;49:580–586.]

Introduction

The recent introduction of optical coherence tomography angiography (OCTA) has provided a novel and noninvasive technique for demonstrating the microvascular blood flow within the retina, previously mainly possible with the more invasive fluorescein angiography. OCTA produces a depth-resolved evaluation of the reflectance data from retinal tissue, providing a three-dimensional volume of information.1

Early studies on primate histology demonstrated up to four retinal vascular networks in the macula.2–5 The superficial vascular plexus (SVP) is composed of blood vessels which are located primarily in the ganglion cell layer (GCL). Two deeper capillary networks are found above and below the inner nuclear layer (INL): the intermediate capillary plexus (ICP) and deep capillary plexus (DCP), respectively. The fourth layer is a regional layer named the radial peripapillary capillary plexus (PRCP), running in parallel with nerve fiber layer (NFL) axons.

Owing to the ability of OCTA to resolve the vascular layers of the retina in three dimensions, most of the software analysis tools for OCTA have separated the inner retinal capillaries into two plexuses: a superficial capillary plexus (SCP), also named superficial vascular plexus (SVP), and a deep capillary plexus (DCP), also named deep vascular plexus (DVP). Using customized segmentation analysis of OCTA images obtained using the RTVue XR Avanti OCTA (Optovue, Fremont, CA), Park et al. were able to also demonstrate the ICP.6 In their study, this plexus was qualitatively and functionally distinct, clearly distinguishable from the SCP and DCP.

Recently a new OCTA module was introduced for the Spectralis OCT machine (Heidelberg Engineering, Heidelberg, Germany). This module offers the option to better segment the retinal vasculature into four different plexuses, corresponding to the four histological locations described above (NFL, GCL, inner plexiform layer [IPL]-INL, and INL-outer plexiform layer). It also allows the segmentation of the SCP and DCP, as is commonly used in other machines.

The foveal avascular zone (FAZ) is a capillary-free area in the center of the macula, which is surrounded by interconnected capillary beds.7 Changes to its size have been shown to be indicative of the microcirculation state of the central fovea. It has been shown to enlarge in conditions such as diabetic retinopathy and branch retinal vein occlusion.8,9

The aim of our study was to provide normative data on the size of the FAZ in the four different retinal vascular layers and to evaluate interobserver variability in measurement of the FAZ area.

Patients and Methods

Healthy volunteers with no previous ophthalmic or medical history underwent OCTA examination at the clinical research facility, Moorfields Eye Hospital. The study was approved by the institutional review board (ROAD No. 17/013). Exclusion criteria included any medical history that may affect the retinal vasculature (eg, diabetes mellitus, hypertension), or previous ocular disease or surgery.

Image Acquisition

A commercially available Heidelberg Spectralis spectral-domain OCTA (SD-OCTA) machine was used to acquire the images. It acquires 85,000 A-scans per second, with an axial resolution of 3.9 μm and a lateral resolution of 5.7 m. The software is based on a probability algorithm. Unless the direction of flow is measured simultaneously, OCTA is a qualitative determination of local blood flow (ie, flow or no flow). To take this into account optimally, in the Spectralis, a probability value for the presence of flow is derived at every location within the OCTA B-scan based on a statistical model of the changes of the OCT signal over time (repeats). The contrast increases with the number of repeated measurements at each B-scan location.

Study participants underwent SD-OCT imaging following a protocol that included a three-dimensional volume set of 10° × 10° (2.9 mm × 2.9 mm), consisting of 512 section images, and an automatic real-time mode of seven frames per scan. The scan overlay was centered over the fovea.

Following image acquisition, the machine segments the retinal vasculature into several distinct presets: SVP – from the ILM to the IPL; DVP – from the IPL to the OPL; NFL — from the inner limiting membrane (ILM) to the NFL; GCL — from the NFL to the GCL; IPL-INL — from the IPL to the INL; and INL-OPL — from the INL to the OPL. For each participant, en face OCTA images of each preset were exported as a tag image file format file.

Image Analysis

The exported images were imported into an image analysis software (ImageJ 1.51j8; National Institutes of Health, Bethesda, MD). Two graders (RS and AB) marked the area of the FAZ in each image, as shown in Figure 1. The graders performed the assessments independently and were masked to each other's results. If the FAZ area was not delineated clearly, brightness was adjusted. Images in which the quality of the scan did not allow for correct measurement of the FAZ were not used. For each eye, the FAZ area in each plexus was measured three times by each grader. The size given by the software was converted to a millimeter area using the Gullstrand Eye conversion (0.291 mm/degree).

Optical coherence tomography angiography (OCTA) of the left eye of a 33-year-old, healthy male showing the foveal avascular zone before (A) and after (B) delineation with ImageJ software (National Institutes of Health, Bethesda, MD).

Figure 1.

Optical coherence tomography angiography (OCTA) of the left eye of a 33-year-old, healthy male showing the foveal avascular zone before (A) and after (B) delineation with ImageJ software (National Institutes of Health, Bethesda, MD).

Statistical Analysis

The qualitative characteristics of all subjects were summarized as frequencies and percentages. The quantitative characteristics were summarized as means and standard deviations. Repeatability analysis was performed by calculating the intraclass correlation coefficient (ICC). A Bland-Altman plot was used to assess the repeatability of the method by comparing repeated measurements for each single examiner. The interobserver reproducibility was evaluated with a two-way mixed model with the subjects as a random effect and the observers as the fixed effect. All statistical analyses were performed by SPSS-24 (IBM, Armonk, NY) and MedCalc statistical software version 12 (MedCalc, Mariakerke, Belgium).

Results

A total of 50 eyes of 25 subjects were included in the study. FAZ measurement was not done in three eyes due to insufficient image quality. Therefore, 47 eyes were included in the analysis. Mean subject age was 34 years ± 7.3 years (range: 22 years to 55 years), and 16 (64%) were women.

The FAZ area was not clearly delineated in the NFL and GCL layers, and therefore not measured. The mean FAZ area for each layer, as measured by each grader, along with ICCs are shown in Table 1. The interobserver agreement between graders was high for all layers (ICC > 0.98) (Figure 2).

Mean ± SD of Foveal Avascular Zone Area (mm2) for Each Grader

Table 1:

Mean ± SD of Foveal Avascular Zone Area (mm2) for Each Grader

Bland-Altman Plot showing the interobserver agreement for foveal avascular zone area measurement between graders. The X-axis is the mean between grader measurements, and the Y-axis is the difference between measurements. The bold lines indicate the average absolute differences between the grader measurements, and the fine lines show the upper and lower bounds of the 95% consistency intervals. (A) Agreement for measurement of the superficial vascular plexus. (B) Agreement for measurement of the deep vascular plexus. (C) Agreement for measurement of the vascular layer from the inner plexiform layer to the inner nuclear layer. (D) Agreement for measurement of the vascular layer from the inner nuclear layer to the outer plexiform layer. FAZ = foveal avascular zone; SVP = super vascular plexus; DVP = deep vascular plexus; IPL = inner plexiform layer; INL = inner nuclear layer; OPL = outer plexiform layer

Figure 2.

Bland-Altman Plot showing the interobserver agreement for foveal avascular zone area measurement between graders. The X-axis is the mean between grader measurements, and the Y-axis is the difference between measurements. The bold lines indicate the average absolute differences between the grader measurements, and the fine lines show the upper and lower bounds of the 95% consistency intervals. (A) Agreement for measurement of the superficial vascular plexus. (B) Agreement for measurement of the deep vascular plexus. (C) Agreement for measurement of the vascular layer from the inner plexiform layer to the inner nuclear layer. (D) Agreement for measurement of the vascular layer from the inner nuclear layer to the outer plexiform layer. FAZ = foveal avascular zone; SVP = super vascular plexus; DVP = deep vascular plexus; IPL = inner plexiform layer; INL = inner nuclear layer; OPL = outer plexiform layer

The intraobserver values for measurement of each layer are presented in Table 2. The intraobserver agreement was high for all layers (ICC ≥ 0.998).

Intraobserver Agreement by Layer

Table 2:

Intraobserver Agreement by Layer

Table 3 shows the mean FAZ area in each layer by eye. There was no statistically significant difference between eyes in each of the layers.

Mean Foveal Avascular Zone Area (mm2) by Eye

Table 3:

Mean Foveal Avascular Zone Area (mm2) by Eye

In order to estimate the similarity between the four measurable layers, we examined the agreement between FAZ measurement of all layers for each grader. The ICC was high for both graders: 0.999 for the first grader, and 0.996 for the second grader.

Discussion

To the best of our knowledge, this is the first study assessing normative data and the reliability of FAZ area measurements in vascular layers other than the SCP and DCP. In this study, the mean FAZ area was similar between the different retinal layers, as segmented by the device. There was no significant difference in FAZ area between the eyes. The interobserver and intraobserver agreements for FAZ measurement were excellent.

Several studies measured the FAZ area in healthy eyes using different OCTA devices, employing different image acquisition algorithms.10–14 These studies have provided data on the dimensions of the FAZ in the SCP and DCP, as delineated by the various systems. Previous histological studies have demonstrated four different vascular networks: The SCP in the GCL; the ICP, above the INL; the DCP, below the INL; and a regional layer – the PRCP, in the NFL layer.2–5 The recently introduced Heidelberg Spectralis OCTA device has the ability to segment the retinal vasculature as in other devices, namely SCP and DCP, but also according to the four histologic layers as demonstrated by histology. Although in theory, this segmentation would have enabled the measurement of the FAZ area in six different layer configurations, in our study, the FAZ was not measurable in two layers. In the NFL layer the FAZ was almost indiscernible, possibly due to the location of the vessels mainly peripapillary (Figure 3A). In the GCL, the FAZ was not clearly delineated (Figure 3B).

Nerve fiber layer (NFL) and ganglion cell layer (GCL) vascular networks, as acquired by the Heidelberg optical coherence tomography angiography machine (Heidelberg Engineering, Heidelberg, Germany). (A) The NFL vascular network of the left eye of a 29-year-old woman. The foveal avascular zone (FAZ) is barely visible. (B) The GCL vascular network in the right eye of a 35-year-old woman. Although the FAZ is seen more clearly, it was not delineated clearly enough to allow for its correct measurement.

Figure 3.

Nerve fiber layer (NFL) and ganglion cell layer (GCL) vascular networks, as acquired by the Heidelberg optical coherence tomography angiography machine (Heidelberg Engineering, Heidelberg, Germany). (A) The NFL vascular network of the left eye of a 29-year-old woman. The foveal avascular zone (FAZ) is barely visible. (B) The GCL vascular network in the right eye of a 35-year-old woman. Although the FAZ is seen more clearly, it was not delineated clearly enough to allow for its correct measurement.

We found excellent agreement between FAZ areas in all layers, as measured by both graders (ICC = 0.999 and 0.996). This finding demonstrates that despite the different segmentations of the vascular layers, the difference in FAZ size in the various layers as measured by the Heidelberg OCTA in healthy subjects is negligible. The significance of segmentation of the macular retinal vasculature to four layers using OCTA has never been examined. Further studies on eyes with vascular pathologies (eg, vein occlusion, diabetic retinopathy) are needed to assess the benefit of such segmentation.

The intraobserver and interobserver agreement for FAZ measurement were excellent in our study (ICC ≥ 0.998, ICC > 0.98, respectively). The interobserver agreement was high in all layers (0.984 to 0.991). Previous studies have shown a lower agreement for measurement of the DCP in comparison with the SCP. In a study by Shahlaee et al.,11 the ICC for interobserver agreement for FAZ measurements in the SCP was ≥ 0.9 but did not meet the lowest acceptable grader agreement for the measurements made on the DCP (ICC < 0.85). According to the authors, the reason for the difference may be the better definition of FAZ outlines at the level of the SCP in comparison with the DCP using the AngioVue OCTA system. A study by Magrath et al.14 conducted on the AngioVue OCTA and Zeiss Cirrus Angioplex (Zeiss, Oberkochen, Germany) also mentioned the better anatomical definition of the FAZ at the SCP as a reason for less interobserver variability in this layer.

Previous OCTA studies demonstrated a larger mean FAZ area in the DCP than in the SCP in normal eyes. In a study by Samara et al.12 using the Optovue RTVue XR 100 Avanti, the mean FAZ area in the SCP was 0.266 mm2 ± 0.097 mm2, whereas in the DCP it was 0.459 mm2 ± 0.227 mm2. In the study by Magrath et al.14 comparing the Optovue Avanti and Zeiss Cirrus Angioplex, the mean FAZ size in the DCP was larger in an automated measurement by the Avanti machine, and manual measurements done on the Avanti and Cirrus machines (0.3565 mm2, 0.3637 mm2, and 0.39926 mm2, respectively) than in the SCP (0.2855 mm2, 0.2739 mm2, and 0.26572 mm2). In the study by Shahlaee et al. using the AngioVue OCTA system,11 the mean FAZ size was 0.27 mm2 ± 0.101 mm2 in the SCP, and 0.34 mm2 ± 0.116 mm2 in the DCP. In the current study, the mean FAZ area was similar in the SCP (0.323 mm2 ± 0.101 mm2 as measured by grader 1, 0.324 mm2 ± 0.104 mm2 as measured by grader 2) and the DCP (0.323 mm2 ± 103 mm2 by grader 1, 0.322 mm2 ± 0.103 mm2 by grader 2).

The discrepancy between the current study and previous studies in relation to the difference of the FAZ area in the SCP and DCP may be due to several reasons. First, the different segmentation methods in the various machines. In the AngioVue system, the SCP extends from 3 μm below the ILM to 15 μm below the IPL. The DCP extends from 15 μm to 70 μm below the IPL.11 In the Zeiss system, the superficial retinal layer slab is defined by the ILM as the inner surface, and an approximation of the IPL as the outer surface, as estimated by the location of the ILM plus 70% of the thickness between the ILM and OPL. The deep retina layer slab is defined by the IPL as the inner surface, and the OPL, as approximated by the retinal pigment epithelium minus 110 μm, as the outer surface.15 In the Heidelberg system, the device delineates the layers automatically and uses the following conventions for layer segmentation: SCP is defined as the ILM as the inner surface, and the IPL as the outer surface. DCP is defined by the IPL as the inner surface and the OPL as the outer surface.16

Another reason may be related to image resolution. As previously mentioned, there was lower interobserver agreement in measurement of the DCP in comparison with SCP in previous studies. This trend, along with the explanation given by different authors regarding the more difficult identification of FAZ in the DCP may result in a larger measurement of the DCP. We did not notice a difference in the quality of FAZ delineation between the SCP and the DCP in our study. The Heidelberg OCTA scan consists of 512 section images, which produces a high-resolution image. In addition, the Heidelberg OCTA employs a probability algorithm in which a probability value for the presence of flow is derived at every location within the OCTA B-scan based on a statistical model of the changes of the OCT signal over time. This approach may have a higher discriminatory power compared to conventional OCTA algorithms, leading to high-contrast images. It is possible that the difference in resolution and contrast allows for better delineation of the DCP with this system, as manifested by a similar FAZ area as in the SCP, and by the very high interobserver agreement in DCP measurement.

However, a third reason could be projection artifacts from the SCP that may, in fact, limit accurate measurement of the DCP using the Heidelberg system. The light beam that encounters the superficial retinal plexus may pass through the moving blood cells, and the projection of the scanned vessels, including those in the FAZ, may appear on the reconstruction of the DCP.17 It is possible that the higher resolution provided by the Heidelberg OCTA system results in more projection artifacts, thus leading to a false measurement of the FAZ at the level of the DCP.

The mean FAZ size, as averaged between all layers and with both graders, was 0.322 ± 0.101. Table 4 shows the range of FAZ areas measured in various studies with different OCTA systems. The mean value found in our current study is similar to the mean FAZ area found using other OCTA systems. Variations between the different studies may be related to differences in the age, sex, and race of the populations in the different studies, as well as the algorithms and methods used to counter projection artifacts employed by the OCTA system and the measurement techniques used.

Comparison of Studies Measuring FAZ Area in Healthy Eyes Using Various OCTA Systems

Table 4:

Comparison of Studies Measuring FAZ Area in Healthy Eyes Using Various OCTA Systems

We found no difference in the FAZ size of each layer when the two eyes of each subject were compared. This is in line with findings in previous studies.14

A limitation of our study is that measurement of the FAZ area was only done manually, as the Heidelberg system does not allow for automatic measurements of the FAZ. However, the interobserver agreement in the study was excellent.

In conclusion, our study showed that in healthy subjects, there was high agreement between the FAZ area in the different retinal vascular layers. The intraobserver and interobserver agreement for FAZ area measurements were excellent. Further studies on different pathologies are needed to evaluate the utility of measurements in the four vascular layers in our daily practice.

References

  1. Spaide RF, Fujimoto JG, Waheed NK. Optical coherence tomography angiography. Retina. 2015;35(11):2161–2162. doi:10.1097/IAE.0000000000000881 [CrossRef]
  2. Provis JM. Development of the primate retinal vasculature. Prog Retin Eye Res. 2001;20(6):799–821. doi:10.1016/S1350-9462(01)00012-X [CrossRef]
  3. Snodderly DM, Weinhaus RS, Choi JC. Neural-vascular relationships in central retina of macaque monkeys (Macaca fascicularis). J Neurosci. 1992;12(4):1169–1193. doi:10.1523/JNEUROSCI.12-04-01169.1992 [CrossRef]
  4. Stone J, van Driel D, Valter K, Rees S, Provis J. The locations of mitochondria in mammalian photoreceptors: Relation to retinal vasculature. Brain Res. 2008;1189:58–69. doi:10.1016/j.brainres.2007.10.083 [CrossRef]
  5. Campbell JP, Zhang M, Hwang TS, et al. Detailed vascular anatomy of the human retina by projection-resolved optical coherence tomography angiography. Sci Rep. 2017;7:42201. doi:10.1038/srep42201 [CrossRef]
  6. Park JJ, Soetikno BT, Fawzi AA. Characterization of the middle capillary plexus using optical coherence tomography angiography in healthy and diabetic eyes. Retina. 2016;36(11):2039–2050. doi:10.1097/IAE.0000000000001077 [CrossRef]
  7. Tick S, Rossant F, Ghorbel I, et al. Foveal shape and structure in a normal population. Invest Ophthalmol Vis Sci. 2011;52(8):5105–5110. doi:10.1167/iovs.10-7005 [CrossRef]
  8. Takase N, Nozaki M, Kato A, Ozeki H, Yoshida M, Ogura Y. Enlargement of foveal avascular zone in diabetic eyes evaluated by en face optical coherence tomography angiography. Retina. 2015;35(11):2377–2383. doi:10.1097/IAE.0000000000000849 [CrossRef]
  9. Wons J, Pfau M, Wirth MA, Freiberg FJ, Becker MD, Michels S. Optical coherence tomography angiography of the foveal avascular zone in retinal vein occlusion. Ophthalmologica. 2016;235(4):195–202. doi:10.1159/000445482 [CrossRef]
  10. Carpineto P, Mastropasqua R, Marchini G, Toto L, Di Nicola M, Di Antonio L. Reproducibility and repeatability of foveal avascular zone measurements in healthy subjects by optical coherence tomography angiography. Br J Ophthalmol. 2016;100(5):671–676. doi:10.1136/bjophthalmol-2015-307330 [CrossRef]
  11. Shahlaee A, Pefkianaki M, Hsu J, Ho AC. Measurement of foveal avascular zone dimensions and its reliability in healthy eyes using optical coherence tomography angiography. Am J Ophthalmol. 2016;161:50–55.e1. doi:10.1016/j.ajo.2015.09.026 [CrossRef]
  12. Samara WA, Say EAT, Khoo CTL, et al. Correlation of foveal avascular zone size with foveal morphology in normal eyes using optical coherence tomography angiography. Retina. 2015;35(11):2188–2195. doi:10.1097/IAE.0000000000000847 [CrossRef]
  13. Guo J, She X, Liu X, Sun X. Repeatability and reproducibility of foveal avascular zone area measurements using angioplex spectral domain optical coherence tomography angiography in healthy subjects. Ophthalmologica. 2017;237(1):21–28. doi:10.1159/000453112 [CrossRef]
  14. Magrath GN, Say EAT, Sioufi K, Ferenczy S, Samara WA, Shields CL. Variability in foveal avascular zone and capillary density using optical coherence tomography angiography machines in healthy eyes. Retina. 2017;37(11):2102–2111. doi:10.1097/IAE.0000000000001458 [CrossRef]
  15. Cirrus HD-OCT [user manual]. Oberkochen, Germany: Zeiss; 139–140.
  16. Spectralis OCT Angiography Module [user manual - Software version 6.7]. Heidelberg, Germany: Heidelberg Engineering; 31.
  17. Spaide RF, Fujimoto JG, Waheed NK. Image artifacts in optical coherence tomography angiography. Retina. 2015;35(11):2163–2180. doi:10.1097/IAE.0000000000000765 [CrossRef]

Mean ± SD of Foveal Avascular Zone Area (mm2) for Each Grader

LayerGrader 1Grader 2ICC
SVP0.323 ± (0.101)0.324 ± (0.104)0.991
DVP0.323 ± (0.103)0.322 ± (0.103)0.987
IPL-INL0.318 ± (0.100)0.314 ± (0.102)0.994
INL-OPL0.325 ± (0.102)0.327 ± (0.105)0.984

Intraobserver Agreement by Layer

LayerGrader 1 ICCGrader 2 ICC
SVP1.0001.000
DVP0.9991.000
IPL-INL1.0001.000
INL-OPL0.9990.998

Mean Foveal Avascular Zone Area (mm2) by Eye

Grader 1Grader 2
SVPDVPIPL-INLINL-OPLSVPDVPIPL-INLINL-OPL
OD0.3130.3140.3090.3150.3140.3220.3050.321
OS0.3330.3320.3280.3350.3310.3260.3220.333
Mean of Both Eyes0.3230.3230.3180.3250.3220.3240.3140.327
P Value.516.547.520.512.582.881.580.700

Comparison of Studies Measuring FAZ Area in Healthy Eyes Using Various OCTA Systems

StudyYearOCTA SystemN (Eyes)Mean FAZ Area ± SD (mm2)

Current study2017Heidelberg OCTA470.322 ± 0.101

Guo et al.132017Zeiss AngioPlex250.375 ± 0.110

Shahlaee et al.112016AngioVue OCTA340.27 ± 0.101

Carpineto et al.102016XR Avanti600.2515 ± 0.096
AngioVue

Magrath et al.2016XR Avanti500.319
AngioVue
Zeiss AngioPlex500.332

Samara et al.122015XR Avanti700.266 ± 0.097
AngioVue
Authors

From NIHR Biomedical Research Centre, Moorfields Eye Hospital, London (RS, AB, PH, SS); and UCL Institute of Ophthalmology, London (AD).

This manuscript was presented at the European Society of Ophthalmology Conference, in Barcelona, Spain, in June 2017.

Dr. Hykin has received grants, personal fees, and non-financial support from Bayer, Novartis, and Allergan outside the submitted work. Dr. Sivaprasad has received grants and personal fees from Bayer, Novartis, Allergan, and Boehringer Ingelheim, as well as personal fees from Heidelberg Engineering, outside the submitted work. The remaining authors report no relevant financial disclosures.

Address correspondence to Roy Schwartz, MD, NIHR Biomedical Research Centre, Moorfields Eye Hospital, 162 City Road, London, EC1V 2PD, United Kingdom; email: royschwartz@gmail.com.

Received: August 22, 2017
Accepted: November 02, 2017

10.3928/23258160-20180803-05

Sign up to receive

Journal E-contents