Egyptian Retina Journal

: 2019  |  Volume : 6  |  Issue : 2  |  Page : 33--37

Correlation between visual acuity and diabetic macular ischemia using optical coherence tomography angiography

Lydia Maged Louis1, Alaa Eldin Fayed2, Mohamed Salah Helal3, Ahmed Mohammed Habib4, Mohammed Hanafy Hashem5, Sherif Nabil Embabi1,  
1 Department of Ophthalmology, Faculty of Medicine, Ain Shams University; Retina Unit, Al Watany Eye Hospital, Cairo, Egypt
2 Retina Unit, Al Watany Eye Hospital; Department of Ophthalmology, Faculty of Medicine, Cairo University, Cairo, Egypt
3 Retina Unit, Al Mashreq Eye Center, Cairo, Egypt
4 Department of Ophthalmology, Faculty of Medicine, Ain Shams University; Retina Unit, Al Mashreq Eye Center, Cairo, Egypt
5 Department of Ophthalmology, Faculty of Medicine, Ain Shams University, Cairo, Egypt

Correspondence Address:
Dr. Lydia Maged Louis
13 Dorrat Al Kahera Compound, New Cairo


Context: Diabetic macular ischemia affects visual function to a variable degree. Aims: This study aims to find whether parameters of ischemia measured on optical coherence tomography angiography (OCTA) correlate with best spectacle-corrected visual acuity (BCVA). Settings and Design: A cross-sectional study was done on 33 eyes with diabetic retinopathy (DR) and 11 nondiabetic control eyes. Subjects and Methods: A 3 mm × 3 mm OCTA images of the superficial capillary plexus (SCP), intermediate capillary plexus (ICP), deep capillary plexus (DCP), and full-thickness retinal slabs were obtained and used to measure foveal avascular zone (FAZ) area and circularity index (CI) manually. Statistical Analysis Used: Statistical comparison between the diabetic and control groups and correlation with BCVA was calculated. Results: CI was significantly higher in controls in each of the SCP (P > 0.001), ICP (P > 0.05), DCP (P = 0.005), and full retinal thickness (P < 0.05) slabs. Compared to the moderate nonproliferative DR group, the control group had a smaller FAZ (P = 0.04) and higher CI (P = 0.01) in the DCP slab. Neither the FAZ area nor the CI was significantly correlated with BCVA in any of the slabs. Conclusions: OCTA can detect a measurable degree of ischemia in diabetic eyes compared to controls. CI is the most sensitive parameter for this purpose. The DCP shows the earliest affection with increasing grades of DR. FAZ area and CI are not good indicators of VA.

How to cite this article:
Louis LM, Fayed AE, Helal MS, Habib AM, Hashem MH, Embabi SN. Correlation between visual acuity and diabetic macular ischemia using optical coherence tomography angiography.Egypt Retina J 2019;6:33-37

How to cite this URL:
Louis LM, Fayed AE, Helal MS, Habib AM, Hashem MH, Embabi SN. Correlation between visual acuity and diabetic macular ischemia using optical coherence tomography angiography. Egypt Retina J [serial online] 2019 [cited 2020 Aug 6 ];6:33-37
Available from:

Full Text


Diabetic macular ischemia (DMI), a complication of diabetic retinopathy (DR), is a leading cause of visual impairment in diabetic patients.[1] It is irreversible and results in poorer visual improvement in patients treated for diabetic macular edema (DME) by intravitreal injections.[2],[3] There is a need to accurately diagnose and quantify DMI in order to predict the visual prognosis for those patients. The Early Treatment DR Study clinical protocols for diagnosing and grading DMI used fundus fluorescein angiography which is still the current gold standard.[4] The introduction of optical coherence tomography (OCT) allowed revolutionary advancements in the management of cases of DME but did not aid in the diagnosis or quantification of DMI. With the recent emergence of OCT angiography (OCTA), which has the ability to view the vascular plexuses of the retina separately, more research is directed toward using OCTA for diagnosing DMI.

The aim of this study was to correlate the LogMAR best spectacle-corrected visual acuity (BCVA) with the degree of DMI measured by OCTA in eyes with diabetic maculopathy.

 Subjects and Methods

This study was approved by the Research Ethics Committee at the Faculty of Medicine at Ain Shams University, followed the Tenets of the Declaration of Helsinki and was performed in accordance with Health Insurance Portability and Accountability Act regulations. Written informed consent was obtained from all the participants.

Thirty-three eyes of 21 diabetic patients with DR and 11 eyes of six healthy age-matched controls were included in the study.

BCVA was measured using a Snellen chart and converted to LogMAR for statistical analysis. All patients were referred for OCT and OCTA imaging.

Patients with diabetic maculopathy were recruited. Clinical stratification was done by an experienced retina consultant (S.N.E.) according to the International Clinical DR Scale[5] into mild, moderate, and severe nonproliferative DR (NPDR) and proliferative DR (PDR). All included patients were at least 18 years old, with a minimum BCVA of 6/60 (LogMAR = 1). Exclusion criteria were significant media opacities and extensive subfoveal hard exudates which may reduce visual acuity and any associated optic nerve or other retinal disorders. OCTA images, in which identification of the foveal avascular zone (FAZ), were not possible due to significant motion or projection artifact or signal strength index below 50 was discarded.

The AngioVue OCTA system (RTVue-XR Avanti; Optovue, Fremont, CA, USA) with a split-spectrum amplitude-decorrelation angiography software algorithm (version 2017.1.0.49) was used.

A 3 mm × 3 mm OCTA images centered on the fovea for each of the superficial capillary plexus (SCP), intermediate capillary plexus (ICP), and deep capillary plexus (DCP) layers were acquired, as well as a full retinal thickness slab. Custom slabs were obtained using the Optovue software such that the upper boundary of the SCP slab was the internal limiting membrane (ILM), and the lower limit 9 μm above the inner plexiform layer (IPL) and the ICP starts at 9 μm above the IPL up to 30 μm below the IPL, the DCP lies between a line 30 μm below the IPL and another 10 μm below the outer plexiform layer (OPL), and the full retinal thickness slab extends from the ILM to 10 μm below the OPL. This was done to enable analysis of each of the three plexuses separately. No manual manipulation was done to the slab boundaries. The projection artifact was removed.

Manual analysis of the OCTA images to quantify DMI was done by one masked grader (L. M. L.) using Fiji image analysis software (National Institutes of Health, Bethesda, MD, USA). Quantification of DMI was done using two FAZ metrics, namely the area and regularity. The FAZ was defined as the central avascular region of the macula devoid of any blood vessels. Freehand selection of this area was done using Fiji, and then, the FAZ area (in mm2) was calculated [Figure 1]. The same selection was used to FAZ regularity. In this study, FAZ regularity was measured as the circularity index (CI). CI is defined as the similarity of the FAZ circumference to that of a custom circle having the same area. It is calculated as 4 π* (area/perimeter2). A perfect circle would have a CI = 1, and the higher the irregularity, the lower the CI.[6] CI is not a measure of the shape of the FAZ, but rather, the regularity of the outline of the FAZ. It has been previously reported that an ischemic FAZ has a less regular outline, most probably due to capillary pruning and dropout. CI and acircularity index (the reciprocal of CI) have been measured in previous studies aiming to assess the degree of ischemic changes.[7],[8] It has been claimed to be less affected by magnification errors due to refractive errors than FAZ area, making it a more accurate parameter to quantify ischemia.[7] These calculations were performed automatically using Fiji software [Figure 1]. The previous steps were repeated for all four slabs obtained for each eye.{Figure 1}

Eyes were included if they had any degree of diabetic changes affecting the macula. Eyes were not diagnosed as having DMI, but rather, a possible correlation between a larger and more irregular FAZ and poorer visual acuity was investigated, similar to the previous studies.[9]

Data were coded and entered using the statistical package SPSS version 25 (IBM Corp., IBM SPSS Statistics for Windows, Armonk, NY). Data were summarized using mean, standard deviation, median, minimum and maximum for quantitative variables and frequencies (number of cases), and relative frequencies (percentages) for categorical variables. Comparisons between groups were done using unpaired t-test when comparing two groups and analysis of variance (ANOVA) with multiple comparisons post hoc test when comparing ≥2 groups.[10] For comparison of paired measurements within each group, repeated-measures ANOVA was used.[11] For comparing categorical data, Chi-square test was performed. Exact test was used instead when the expected frequency is [12] Correlations between quantitative variables were done using Pearson correlation coefficient.[13] Testing for interrater reliability was done using the intra-class coefficient and Cronbach's alpha reliability coefficient with their 95% confidence interval.[14] P < 0.05 was considered as statistically significant.


All recruited patients were Egyptian. [Table 1] and [Table 2] show the distribution of demographic and clinical parameters among the studied 33 diabetic and 11 control eyes. 36% of each group were male, and no significant difference was found in age between the two groups. Signal strength was significantly higher in controls than diabetic patients (P < 0.05). Of the diabetic eyes, 21% had mild NPDR, 36% had moderate NPDR, 30% had severe NPDR, and 12% had PDR.{Table 1}{Table 2}

[Table 3] compares diabetic patients and controls for the parameters of DMI in each of the slabs. Mean CI was found to be significantly higher in the control eyes in each of the SCP (P < 0.001), ICP (P < 0.05), DCP (P = 0.005), and full retinal thickness (P < 0.05) slabs. Further, stratification into the DR stages [Table 4] and [Table 5] found a statistically significant difference between the controls and moderate NPDR groups for each of the FAZ area (P = 0.04) and CI (P = 0.01) in the DCP slab only.{Table 3}{Table 4}{Table 5}

[Table 6] shows no correlation between the FAZ area or CI and LogMAR BCVA in the diabetic eyes.{Table 6}


In this study on 33 diabetic eyes and 11 control eyes, CI was found to be a good indicator of ischemia due to the significant difference between diabetic and control eyes (P < 0.001). FAZ area was also larger in diabetic eyes in all slabs, but the difference did not achieve statistical significance. It has been previously demonstrated that a measurable degree of ischemia exists in diabetic eyes compared to healthy eyes which can be assessed by OCTA.[15],[16],[17] The DCP slab is the most severely affected by diabetic ischemic changes given the fact that both FAZ area (P = 0.04) and CI (P = 0.01) in this slab were more significantly affected in the moderate NPDR group compared to the control group. This is in accordance with the previous publications.[16],[18]

However, FAZ area and CI were not found to be good predictors of VA in any of the slabs, despite the fact that multiple publications have established FAZ area as a good predictor of VA in both the SCP and DCP in eyes with DMI[9],[19] and in eyes with DR.[15],[16],[17],[20] However, it should be noted that in these previous studies, segmentation was different such that the ICP was included in either the SCP or DCP slab; while in our study, each plexus was studied separately providing more clinically relevant results. Furthermore, FAZ size reportedly has a high variability even among normal individuals.[9],[21],[22] Moreover, the use of different machines and image analysis protocols may have led to different results. We conclude that although FAZ area may not be valuable in predicting visual function, it may be useful in monitoring the progression of disease over time.[9]

We realize a number of limitations to our study. Eyes with DR were included regardless of the type of diabetes, duration of diagnosis, type of treatment administered, or tightness of blood sugar control. Therefore, no data can be inferred about whether these factors affect the parameters measured in this study. Since only 3 mm × 3 mm images were used in this study to ensure higher image resolution, these results cannot be generalized to a larger perimeter of the macula. We realize that no information can be inferred from about the utility of these parameters in following up disease progression or response to treatment, as the images were obtained at one visit.

This study emphasizes the role of OCTA in assessing each vascular plexus of the retina individually and establishes CI as a useful parameter in predicting ischemic changes in diabetic eyes.

To our knowledge, this is thefirst study in its field done on an Egyptian cohort. Further, a study in Egypt and the Middle East may aid in finding ethnic and global trends in macular ischemic changes related to diabetes. Multicenter prospective studies are also needed to establish accurate figures for the diagnosis of DMI using OCTA parameters and to evaluate their prognostic value.


Diabetic eyes have significantly more ischemia than control eyes on OCTA which can best be measured by FAZ area and CI in the DCP slab. FAZ area and CI are not good predictors of LogMAR BCVA.


We would like to extend our gratitude to Dr. Rania Estawro, Dr. Therese Kamal, Dr. Dina Baddar, Dr. Rabab Emara, Dr. Sahar Metwally, Dr. Mohammed Mekkawy, and Dr. Nouran El Beltagi for their wholesome support, help, and advice.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Mansour AM, Schachat A, Bodiford G, Haymond R. Foveal avascular zone in diabetes mellitus. Retina 1993;13:125-8.
2Chung EJ, Roh MI, Kwon OW, Koh HJ. Effects of macular ischemia on the outcome of intravitreal bevacizumab therapy for diabetic macular edema. Retina 2008;28:957-63.
3Jonas JB, Martus P, Degenring RF, Kreissig I, Akkoyun I. Predictive factors for visual acuity after intravitreal triamcinolone treatment for diabetic macular edema. Arch Ophthalmol 2005;123:1338-43.
4Classification of diabetic retinopathy from fluorescein angiograms. ETDRS report number 11. Early treatment diabetic retinopathy study research group. Ophthalmology 1991;98:807-22.
5Wilkinson CP, Ferris FL 3rd, Klein RE, Lee PP, Agardh CD, Davis M, et al. Proposed international clinical diabetic retinopathy and diabetic macular edema disease severity scales. Ophthalmology 2003;110:1677-82.
6Rasband W. Circularity; 2000. Available from: [Last accessed on 2019 Jan 23].
7Krawitz BD, Mo S, Geyman LS, Agemy SA, Scripsema NK, Garcia PM, et al. Acircularity index and axis ratio of the foveal avascular zone in diabetic eyes and healthy controls measured by optical coherence tomography angiography. Vision Res 2017;139:177-86.
8Ashraf M, Nesper PL, Jampol LM, Yu F, Fawzi AA. Statistical model of optical coherence tomography angiography parameters that correlate with severity of diabetic retinopathy. Invest Ophthalmol Vis Sci 2018;59:4292-8.
9Samara WA, Shahlaee A, Adam MK, Khan MA, Chiang A, Maguire JI, et al. Quantification of diabetic macular ischemia using optical coherence tomography angiography and its relationship with visual acuity. Ophthalmology 2017;124:235-44.
10Chan YH. Biostatistics 102: Quantitative data parametric non-parametric tests. Singapore Med J 2003;44:391-6.
11Chan YH. Biostatistics 301. Repeated measurement analysis. Singapore Med J 2004;45:354-68.
12Chan YH. Biostatistics 103: Qualitative data tests of independence. Singapore Med J 2003;44:498-503.
13Chan YH. Biostatistics 104: Correlational analysis. Singapore Med J 2003;44:614-9.
14Rankin G, Stokes M. Reliability of assessment tools in rehabilitation: An illustration of appropriate statistical analyses. Clin Rehabil 1998;12:187-99.
15Hwang TS, Gao SS, Liu L, Lauer AK, Bailey ST, Flaxel CJ, et al. Automated quantification of capillary nonperfusion using optical coherence tomography angiography in diabetic retinopathy. JAMA Ophthalmol 2016;134:367-73.
16Freiberg FJ, Pfau M, Wons J, Wirth MA, Becker MD, Michels S, et al. Optical coherence tomography angiography of the foveal avascular zone in diabetic retinopathy. Graefes Arch Clin Exp Ophthalmol 2016;254:1051-8.
17Takase N, Nozaki M, Kato A, Ozeki H, Yoshida M, Ogura Y, et al. Enlargement of foveal avascular zone in diabetic eyes evaluated by en face optical coherence tomography angiography. Retina 2015;35:2377-83.
18Salz DA, de Carlo TE, Adhi M, Moult E, Choi W, Baumal CR, et al. Select features of diabetic retinopathy on swept-source optical coherence tomographic angiography compared with fluorescein angiography and normal eyes. JAMA Ophthalmol 2016;134:644-50.
19Sim DA, Keane PA, Fung S, Karampelas M, Sadda SR, Fruttiger M, et al. Quantitative analysis of diabetic macular ischemia using optical coherence tomography. Invest Ophthalmol Vis Sci 2014;55:417-23.
20Balaratnasingam C, Inoue M, Ahn S, McCann J, Dhrami-Gavazi E, Yannuzzi LA, et al. Visual acuity is correlated with the area of the foveal avascular zone in diabetic retinopathy and retinal vein occlusion. Ophthalmology 2016;123:2352-67.
21Samara WA, Say EA, Khoo CT, Higgins TP, Magrath G, Ferenczy S, et al. Correlation of foveal avascular zone size with foveal morphology in normal eyes using optical coherence tomography angiography. Retina 2015;35:2188-95.
22Shahlaee 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;161:50-5.