Clinical and angiographic characterization of choroidal neovascularization in diabetic retinopathy

Abstract

Background:

To report the clinical and angiographic characteristics of choroidal neovascularization in patients with diabetic retinopathy.

Methods:

Patients of type 2 diabetes mellitus with presence of choroidal neovascularization in at least one eye were retrospectively analyzed. The study eyes were divided into three groups based on presence (active or scarred) or absence of choroidal neovascularization (fellow eyes). Imaging characteristics of active choroidal neovascularization were recorded using optical coherence tomography, fluorescein, and indocyanine angiography. Central macular thickness, subfoveal choroidal thickness, and large choroidal vessel layer thickness were compared at baseline and final visit.

Results:

Our study reports the prevalence rate of choroidal neovascularization in eyes with diabetic retinopathy (0.27%; 36 out of 13,382 eyes). A total of 64 eyes of 32 patients (age, mean ± standard deviation: 68.5 ± 9.3 years) with baseline visual acuity of 0.69 ± 0.69 logarithm of minimum angle of resolution (Snellen equivalent 20/100) were included. Nonproliferative diabetic retinopathy (57 eyes) comprised the majority followed by proliferative diabetic retinopathy (7 eyes). Eyes with choroidal neovascularization (36, 56.25%) included both active (25) and scarred (11) choroidal neovascularization, with bilateral choroidal neovascularization in 4 patients. Type 1 choroidal neovascularization was the most common subtype of choroidal neovascularization on optical coherence tomography. Common etiologies for active choroidal neovascularization included age-related macular degeneration (3; 12%), myopia (1; 4%), and inflammatory choroidal neovascularization secondary to chorioretinitis (1; 4%). In the remaining 20 eyes, choroidal neovascularization formation was primarily due to diabetic choroidopathy.

Conclusion:

The prevalence of choroidal neovascularization in eyes with diabetic retinopathy is very low, with a lower prevalence of age-related macular degeneration. Diabetic choroidopathy plays a significant role in formation of choroidal neovascularization in eyes with diabetic retinopathy.

Keywords

Diabetic retinopathy diabetic choroidopathy choroidal neovascular membrane

Introduction

Diabetes mellitus is a systemic disorder, which affects a large population worldwide.^1,2 Diabetic retinopathy (DR) and choroidopathy are characterized by specific changes at the level of retina and choroid, respectively, and play a major role in visual morbidity of diabetic patients. Diabetic choroidopathy has been recently described in patients with diabetes mellitus.^3–6 There have been ample angiographic and histologic evidence of diabetic choroidopathy in eyes with DR.^3,4,7,8 Multiple choroidal changes, including presence of microaneurysms (MAs), loss of choriocapillaris, capillary dropout areas, and choroidal neovascularization (CNV), have been shown in eyes with diabetes mellitus.^3,9 However, the variability in the results in different studies shows that choroidal changes do not follow a similar pattern as seen in retinopathy changes. Moreover, diabetic choroidopathy changes have also been noticed in eyes without any clinical evidence of DR.^3,10,11 This finding suggests that choroidopathy probably precedes the development of retinopathy and could be an etiological factor in a subset of eyes without DR suffering from an unexplained vision loss.^3,7,8

Previous authors have shown a reduced incidence of CNV in eyes with DR.^12–15 The role of DR in the development and progression of CNV is, however, not clearly elucidated. Several reports have failed to show any association between age-related macular degeneration (AMD) and diabetes mellitus, while others have shown that the incidence of neovascular AMD was lower in eyes with DR.^12,13,16 Hua et al.¹⁷ have reported a case of CNV in an elderly patient with diabetic choroidopathy with no evidence of AMD changes.

The aim of the current study was to investigate patients with DR with concomitant diagnosis of CNV in either eye. We analyzed the clinical and angiographic characteristics of CNV and their response to treatment, and identified the possible correlation between DR and CNV.

Methods

The study was a retrospective chart review including 64 eyes of 32 patients of diabetes mellitus with DR and presence of CNV in at least one eye during the study period of January 2015 to June 2018. Eyes without DR, or with advanced stages of DR i.e. tractional retinal detachment or neovascular glaucoma, media opacities such as significant cataract, or vitreous hemorrhage, were excluded from the study. An informed written consent was obtained from all the patients, and the study was performed in accordance with the Declaration of Helsinki. The study approval was received from the local Institutional Ethics Committee (IEC).

A detailed systemic and ocular history related to diabetes mellitus was obtained from all participants. A comprehensive ophthalmic evaluation was done, which included slit-lamp examination, dilated fundus examination using indirect ophthalmoscopy and 90-diopter lens, and intraocular pressure measurements using Goldmann applanation tonometer. Best-corrected visual acuity (BCVA) recorded in Snellen chart was converted to logarithm of minimum angle of resolution (logMAR) for statistical analysis.

Patients received a comprehensive multimodal imaging assessment, which included optical coherence tomography (swept-source-OCT or enhanced depth imaging-OCT), fundus fluorescein angiography (FFA), and indocyanine green angiography (ICGA) in order to establish the diagnosis and document the various diabetes-related retinal and choroidal changes. FFA/ICGA was performed in select patients with suspicious CNV based on the physician discretion. Multiple OCT characteristics were recorded, including the type of CNV based on the localization (above or below the retinal pigment epithelium (RPE)), central macular thickness (CMT), height of neurosensory detachment or subretinal fluid (NSD/SRF), length of SRF, presence of hard exudates, hyperreflective foci within the retina and choroid, intraretinal fluid (IRF), cystoid degeneration, double-layer sign, presence or absence of pigment epithelial detachment (PED), PED height if present, subfoveal choroidal thickness (SFCT), and subfoveal large choroidal vessel layer thickness (SF-LCVT). Number of eyes which maintained integrity of more than 50% of central 1000 µm of external limiting membrane (ELM)/ellipsoid zone (EZ) at fovea was also recorded. CMT was defined as the distance between internal limiting membrane (ILM) and anterior border of RPE–Bruch’s membrane complex at the fovea. SFCT was measured as the distance between Bruch’s membrane and choroid–scleral interface. SF-LCVT was measured at the nearest point to the fovea where a large choroidal vessel of diameter ⩾100 µm was present. This was calculated between choroid–scleral interface and inner margin of the large choroidal vessel.^18,19 FFA and ICGA were used to identify MAS, focal or diffuse leakage, neovascularization, capillary nonperfusion areas in the retina and choroid, and choroidal vascular abnormalities, and classify the CNV as classic (type II) or occult (type I) neovascularization. Diagnosis of CNV and presence of disease activity were defined on the basis of multimodal imaging, which comprised OCT and FFA/ICGA (in select patients).

DR grading was done as per Diabetic Retinopathy Disease Severity Scale into mild, moderate, and severe nonproliferative diabetic retinopathy (NPDR) and proliferative diabetic retinopathy (PDR).²⁰ Diabetic macular edema (DME) was classified as the presence or absence of IRF or SRF accumulation. CNVs were further grouped into either active or scarred lesions. A total of three groups were analyzed—eyes with active CNV, scarred CNV, and the fellow eyes of CNV. Patients treated with intravitreal anti-vascular endothelial growth factors (VEGFs) were followed up on pro-re-nata (PRN) protocol, and the mean number of injections during the follow-up duration was calculated. BCVA (baseline and final) and CMT (baseline and final) were recorded at the baseline and last follow-up visit.

Statistical analysis

Statistical analysis was done using SPSS software, version 23 (SPSS, Inc., Chicago, IL, USA). The data were tabulated as mean ± standard deviation (SD). The comparison of OCT, FFA, and ICGA parameters was done using one-way analysis of variance (ANOVA). Baseline and final parameters in active CNV group were compared using paired t test. Tukey’s post hoc test was used to compare the statistical significance between two groups if ANOVA results were statistically significant. A p value of <.05 was considered as statistically significant.

Results

The data of 6691 patients (13,382 eyes) with presence of DR were retrieved from the electronic medical records system. These included eyes with presence of either DR–NPDR (6923; 51.73%), PDR (6459; 48.27%), or DME (1499; 11.2%). The maximum number of eyes was present in NPDR group (mild, 2566; moderate, 3091; severe, 1266). A total of 64 eyes of 32 patients fulfilled the inclusion criteria and were subsequently analyzed (Table 1). Among 64 eyes of 32 patients, 36 eyes had CNV (25 active, 11 scarred), with a prevalence of 0.27% of CNV among DR eyes.

Table 1.

The baseline demographic parameters of eyes with DR and CNV.

Baseline demographic data	Number (n)
No. of patients (eyes)	32 (64)
Age (M ± SD; years)	68.5 ± 9.3
Gender	24 males, 8 females
BCVA (logMAR)	0.69 ± 0.69
Duration of diabetes (M ± SD; years)	12.1 ± 7.15
Stages of DR
Mild NPDR	32
Moderate NPDR	25
PDR	7
Active CNV	25
Scarred CNV	11
Fellow eyes of CNV with DR	28
No. of patients with bilateral CNV	4

DR: diabetic retinopathy; CNV: concomitant choroidal neovascularization; SD: standard deviation; BCVA: best-corrected visual acuity; logMAR: logarithm of minimum angle of resolution; NPDR: nonproliferative diabetic retinopathy; PDR: proliferative diabetic retinopathy.

Mean age of study subject was 68.5 ± 9.3 years with a predominant male distribution (24 patients, 75%) and 8 females (25%). Mean duration of diabetes mellitus was 12.1 ± 7.5 years. The enrolled eyes were mostly in NPDR group (mild = 32, moderate = 25) compared to the PDR group (7). Four eyes among the seven eyes of PDR group had previously received panretinal photocoagulation (PRP). None of the eyes had received grid or focal laser photocoagulation. The mean baseline BCVA of 64 eyes was 0.69 ± 0.69 logMAR (Snellen equivalent visual acuity 20/100). Mean BCVA values in active CNV (25 eyes), scarred CNV (11 eyes), and fellow eye of CNV (28 eyes) were 0.71 ± 0.5 logMAR (20/100), 1.56 ± 0.64 logMAR (20/720), and 0.35 ± 0.58 logMAR (20/40), respectively. The differences between the groups were statistically significant (p < .001). Four patients showed presence of bilateral CNV.

OCT characteristics

The mean CMT (active CNV, 410.5 ± 259.6 µm; scarred CNV, 345.9 ± 220.3 µm; fellow eyes, 211.3 ± 67.02 µm) was significantly different across the three groups (p = .001). Pairwise comparison using Tukey’s post hoc test showed significant differences in CMT between active CNV eyes and fellow eye of CNV only (p < 0.01). However, there was no significant variation of SFCT (p = .56) and LCVT (p = .21) among the three groups. Type 1 CNV was present in a total of 21 eyes (58.3%, out of 36 eyes with CNV), with 16 (64%) and 5 (45.4%) eyes in active and scarred CNV groups, respectively (Table 2).

Table 2.

The clinical and multimodal imaging parameters in eyes with DR and CNV.

Baseline demographic data	Active CNV	Scarred CNV	Fellow eyes	p value
No. of eyes (n)	25	11	28
BCVA (logMAR) (M ± SD)	0.71 ± 0.5	1.56 ± 0.64	0.35 ± 0.58	<.001
Stages of DR
Mild NPDR	11	7	14
Moderate NPDR	11	4	10
PDR	3	0	4
DME	0	1	1
NSD	6	0	0
Additional fundus features
Drusen	3	3	7
Chorioretinitis	1	1	0
Myopia	1	0	1
OCT characteristics
Type of CNV on OCT (type 1/2)	16/9	5/6	NA
CMT (µm)	410.5 ± 259.6	345.9 ± 220.3	211.3 ± 67.02	.001
NSD height (µm)	96.7 ± 51.5	NA	NA
SRF length	847.3 ± 920.4	NA	NA
Hard exudates	13	5	7
Hyperreflective dots in retina	19	7	15
IRF	14	4	3
Cystoid degeneration	4	4	2
DLS	15	4	0
ELM/EZ integrity (>50%) at central 1000 µm	16/16	2/2	28/28
PED^*	FVPED (n = 17)	FVPED (n = 3)	PED* (n = 5)
Height of PED (µm)	215.6 ± 175.1	202.7 ± 192.7	61.0 ± 63.8	<.001
SFCT (µm)	230.1 ± 112.5	199.4 ± 53.9	236.3 ± 95.4	.56
SF-LCVT (µm)	120.4 ± 65.7	93.4 ± 23.5	131.6 ± 64.8	.21
Hyperreflective dots in choroid	17	7	10
Angiography characteristics (FFA/ICGA)
Microaneurysms	15	2	16
Diffuse leak at macula	3	0	0
Petalloid filling	0	0	0
CNP areas	3	0	0
NVD/NVE	1	0	1
Size of CNV (mm²)	5.6 ± 3.6	12.98 ± 2.15	NA
Type of CNV (classic/occult)	7/8	NA	NA
Mean follow-up (M ± SD; months)	10.0 ± 12.14
Mean number of injections (M ± SD)	5.1 ± 4.3	NA	NA
BCVA (logMAR) at last follow-up	0.63 ± 0.51	NA	0.23 ± 0.3
CMT (M ± SD; µm) at last follow-up	331.4 ± 218.4	NA	219.5 ± 24.8

DR: diabetic retinopathy; CNV: choroidal neovascularization; BCVA: best-corrected visual acuity; logMAR: logarithm of minimum angle of resolution; SD: standard deviation; NPDR: nonproliferative diabetic retinopathy; PDR: proliferative diabetic retinopathy; DME: diabetic macular edema; NSD: neurosensory detachment; OCT: optical coherence tomography; CNV: choroidal neovascularization; CMT: central macular thickness; SRF: subretinal fluid; IRF: intraretinal fluid; DLS: double-layer sign; ELM/EZ: external limiting membrane/ellipsoid zone; PED: pigment epithelial detachment; FVPED: fibrovascular pigment epithelial detachment; PED*: drusenoid PED; SFCT: subfoveal choroidal thickness; SF-LCVT: subfoveal large choroidal vessel layer thickness; FFA: fundus fluorescein angiography; ICGA: indocyanine angiography; CNP: capillary nonperfusion; NVD/NVE: neovascularization of disc/elsewhere. Significant p values are highlighted in bold.

Fibro-vascular PED (FVPED) was present in 17 of 25 (68%) and 3 of 11 (27.3%) eyes in active and scarred CNV, respectively, whereas 5 of 28 (17.9%) eyes had drusenoid PED in fellow eyes. PED height was significantly different across the groups (p < .001). A total of 16 (36%) and 2 (18.2%) eyes maintained integrity of >50% length of ELM/EZ within 1000 µm of fovea in active and scarred CNV groups, respectively. Hyperreflective dots in retina and choroid were most commonly seen in eyes with active CNV (19/25; 76% and 17/25; 68%) followed by eyes with scarred CNV (7/11 in both retina and choroid; 63.6%) with the least prevalence in fellow eyes of CNV (15/28; 53.6% and 10/28; 35.7%).

CNV group (active and scarred)

Common etiologies for active CNV included AMD (3; 12%), myopia (1; 4%), and inflammatory CNV secondary to chorioretinitis (1; 4%). In the remaining 20 eyes, another etiology could not be established, suggestive of CNV formation primarily due to diabetic choroidopathy. Active CNV group received a mean of 5.1 ± 4.3 intravitreal anti-VEGF injections during the mean follow-up duration of 10.0 ± 12.1 months. Change in BCVA from baseline (0.71 ± 0.5 logMAR) to last follow-up visit (0.63 ± 0.51 logMAR) was not significant (p = .57). The mean change in CMT was not significant at the last follow-up (p = .24). FFA and ICGA were available for 15 eyes among the 25 eyes of active CNV. FFA/ICGA descriptors included seven and eight eyes with classic and occult CNV, respectively, with a mean size of 5.6 ± 3.6 mm². All the eyes showed presence of MAs (15/15 eyes; 100%), whereas capillary nonperfusion areas and neovascularization were noted in three eyes and one eye, respectively. Three eyes in scarred CNV group showed the presence of drusen. NPDR was present in seven (mild NPDR) and four (moderate NPDR) eyes, respectively, whereas one eye showed presence of DME in the scarred CNV group. Hard exudates (5), IRF (4), cystoid degeneration (4), and double-layer sign (4) were present with variable frequency in scarred CNV eyes. ELM/EZ integrity (>50%) was present only in 2/11 (18.2%) patients. Representative cases are shown in Figures 1 and 2.

Figure 1.

A 65-year-old male presented with diminution of vision in the left eye and BCVA of 20/100. (a) Left eye fundus showed subretinal hemorrhage superior to fovea (arrow), drusen (arrowhead), and microaneurysm (asterisk). (b) Early phase FFA showed leakage (arrow), drusen (arrowhead), and microaneurysm (asterisk). (c) Mid-phase indocyanine green angiography showed vascular network (arrow) and microaneurysm (arrowhead). (d) OCT showed subretinal hyperreflectivity suggestive of type 2 CNV (arrow). (e) Left eye received 1 anti-vascular endothelial growth factor (VEGF) injection. Post-injection BCVA was 20/100 with OCT showing scarred CNV (arrow).

Figure 2.

A 67-year-old male presented with right eye diminution of vision and BCVA of 20/30. (a) Right eye fundus showed subretinal hemorrhage with grayish membrane involving the fovea (arrow), superficial hemorrhage (arrowhead), and microaneurysms (asterisk). (b) Late phase FFA showed hyperfluorescence with ill-defined margins corresponding to the area of grayish membrane suggestive of CNV, blocked fluorescence due to hemorrhages (arrowhead), and hyperfluorescence due to microaneurysms (asterisk). (c) OCT showed subretinal hyperreflectivity with fibrovascular pigment epithelial detachment (arrow) and small pocket of subretinal fluid (asterisk). Post six intravitreal anti-VEGF injections, BCVA was 20/30 with persistent activity and (d) OCT showed subretinal hyperreflectivity (arrow) and subretinal fluid (asterisk).

Fellow eyes of CNV

Twenty-four eyes had presence of NPDR (mild, 14; moderate, 10) with one eye showing presence of DME. CMT was significantly lower in eyes with fellow eye compared to eyes with active CNV (p = .001), while SFCT and LCVT difference between these groups was not significant. Outer retinal layers, including ELM/EZ, were intact in all (28; 100%) eyes. FFA showed presence of MAs in all 16 eyes (100%) in which FFA was performed.

Discussion

We analyzed the eyes with DR and presence of CNV in at least one eye. There was a significant difference of BCVA in the three groups (active CNV, scarred CNV, and the fellow eyes of CNV; p < .001). While there was a significant variation in CMT among the three groups (p = .001), the choroidal parameters such as SFCT and SF-LCVT were comparable among the groups. The majority of CNV cases were type 1 CNV (21 eyes; 58.3%) on OCT or occult CNV (8 eyes; 53.3%). Yoshikawa et al.²¹ showed higher prevalence of classic CNV in AMD patients with DR. Other authors have reported a comparatively lower rate of classic CNV (17%–21%) in eyes with AMD.^22,23 There was improvement in both clinical and anatomical outcomes in terms of BCVA gain (p = .57) and CMT reduction (p = .24), which was not statistically significant. The mean number of anti-VEGF injections based on PRN protocol was 5.1 ± 4.3.

As alluded earlier, histopathological studies have shown multiple changes in the choroid of diabetic patients.^4,7,8,24 Diabetic choroidopathy is a recently coined terminology, which incorporates these pathologic changes in these eyes with DR. It has been proposed to play a far significant role in the development of DR with few authors hypothesizing that choroidopathy can be a precursor lesion to DR.³

In our series, the prevalence of neovascular AMD in eyes with DR was low (6/13,382 eyes; 0.04%). Previous reports have also shown a lower prevalence of AMD (0.4%) including neovascular AMD in eyes with DR as compared to the general population where the prevalence has been reported to be as high as 8% to 15%.^{12–14,25–27} Authors have hypothesized that the breakdown of inner blood retinal barrier in DR leads to a compensatory upregulation of RPE pump (outer blood retinal barrier), thus protecting against development and progression of AMD.^12,28 Moreover, the eyes with DR treated with laser photocoagulation were also considered at a lesser risk for the development of neovascular AMD while others refute this hypothesis.^13,21 On the contrary, there is ample literature to suggest that DM is a risk factor for AMD with odds ratio ranging from 1.05 to 1.87.^29,30

Hua et al.¹⁷ have described a case of CNV in patient with NPDR. They attributed the CNV formation to an entity termed as proliferative diabetic choroidopathy, which represents the choroidal microvascular changes in diabetes mellitus.¹⁷ Despite the high incidence of DR throughout the world, diabetic choroidopathy leading to CNV formation has been rarely reported in the literature. Therefore, the point which merits consideration is whether DR and/or choroidopathy is an independent risk factor for CNV formation. Intrachoroidal microangiopathy or neovascularization has been reported in eyes with or without DR with a predisposition toward outer choroid.²⁴ Previous publications have described diverse location for these CNV ranging from posterior pole to equator and beyond.^4,7,8,24 These CNV originated at the sites with diffuse choriocapillaris degeneration compared to sites with focal loss.⁷ Reports suggest increased levels of polymorphonuclear (PMN) cells in diabetic choroid.³¹ Increased proteolytic enzymes along with increased expression of molecules such as intracellular adhesion molecule 1 (ICAM-1) in these eyes lead to oxidative damage to endothelium, therefore compounding ischemia and ultimately leading to CNV formation.^7,31,32

Another observation was that PDR eyes (7; 10.9%) were far lesser compared to NPDR eyes (57; 89.1%) in our study cohort of 64 eyes, while the number of patients with PDR seen during the study period was much higher (6459; 48.27%). A possibility of protection against CNV formation especially with higher grades of DR needs to be ruled out. Anti-VEGF agents commonly used nowadays to treat DME and/or PDR may also lead to regression of nascent, asymptomatic CNV. On the contrary, a higher rate of auto-infarction (up to 50%) of these CNV combined with their peripheral location may not lead to clinically manifest disease.^7,8

Studies have shown reduced subfoveal choroidal blood flow in eyes with diabetes, which, along with the histopathological changes described earlier, may change the natural history of CNV.^3,10 In our study, we could not identify any attributable cause to CNV formation in 20 (80%) eyes with active CNV. Other identifiable causes include AMD (3; 12%), myopia (1; 4%), and chorioretinitis (1; 4%) among 20% of the eyes. Thus, the majority of CNV was probably associated with diabetic choroidopathy.

The limitations of the study are the retrospective nature, limited number of patients, and a short follow-up. The number of patients in the PDR group was lesser compared to NPDR group. Similarly, only two patients in the entire cohort had presence of DME. Hypothetically, the natural history of CNV formation and progression may not be significantly altered in the early stages of DR or absence of DME. Due to small sample size, we could not evaluate the difference between the various etiologies of CNV in our cohort. Moreover, the effect of different anti-VEGF injections in the anatomical and visual outcomes especially the long term was not analyzed.

Although patients with DR may have CNVs of varied etiology including AMD, a subset of eyes with DR is predisposed to develop CNV, and retinochoroidal changes may act as an independent risk factor for CNV formation. The discrete evidence of retinopathy and choroidopathy acting as an independent factor for CNV formation is still lacking. Future, long-term longitudinal studies may provide a discrete evidence regarding the association of CNV and diabetic retino-choroidopathy.

Footnotes

Author contributions

S.R.S., M.L., and J.C. were involved in the designing of the study. D.C.P. and S.R.S. were involved in the collection of data. S.R.S., D.C.P., V.G., and J.C. were responsible for interpretation and analysis of the data. S.R.S., D.C.P., V.G., M.L., and J.C. were involved in manuscript writing and reviewed the article. All the authors conducted the study and equally contributed in the preparation, review, and approval of the manuscript.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Vishal Govindahari

References

King

Aubert

Herman

WH.

Global burden of diabetes, 1995-2025: prevalence, numerical estimates, and projections. Diabetes Care 1998; 21(9): 1414–1431.

NCD Risk Factor Collaboration. Worldwide trends in diabetes since 1980: a pooled analysis of 751 population-based studies with 4.4 million participants. Lancet 2016; 387(10027): 1513–1530.

Melancia

Vicente

Cunha

, et al. Diabetic choroidopathy: a review of the current literature. Graefes Arch Clin Exp Ophthalmol 2016; 254(8): 1453–1461.

Lutty

GA.

Diabetic choroidopathy. Vision Res 2017; 139: 161–167.

Wang

Lains

Providencia

, et al. Diabetic choroidopathy: choroidal vascular density and volume in diabetic retinopathy with swept-source optical coherence tomography. Am J Ophthalmol 2017; 184: 75–83.

Hua

Liu

Wang

, et al. Imaging evidence of diabetic choroidopathy in vivo: angiographic pathoanatomy and choroidal-enhanced depth imaging. PLoS ONE 2013; 8(12): e83494.

Cao

McLeod

Merges

, et al. Choriocapillaris degeneration and related pathologic changes in human diabetic eyes. Arch Ophthalmol 1998; 116(5): 589–597.

McLeod

Lutty

GA.

High-resolution histologic analysis of the human choroidal vasculature. Invest Ophthalmol Vis Sci 1994; 35(11): 3799–3811.

Borrelli

Palmieri

Viggiano

, et al. Photoreceptor damage in diabetic choroidopathy. Retina 2020; 40(6): 1062–1069.

10.

Nagaoka

Kitaya

Sugawara

, et al. Alteration of choroidal circulation in the foveal region in patients with type 2 diabetes. Br J Ophthalmol 2004; 88(8): 1060–1063.

11.

Schocket

Brucker

Niknam

, et al. Foveolar choroidal hemodynamics in proliferative diabetic retinopathy. Int Ophthalmol 2004; 25(2): 89–94.

12.

Cummings

Cunha-Vaz

Treatment of neovascular age-related macular degeneration in patients with diabetes. Clin Ophthalmol 2008; 2(2): 369–375.

13.

Zylbermann

Landau

Rozenman

, et al. Exudative age-related macular degeneration in patients with diabetic retinopathy and its relation to retinal laser photocoagulation. Eye 1997; 11(Pt. 6): 872–875.

14.

Voutilainen-Kaunisto

Terasvirta

Uusitupa

, et al. Age-related macular degeneration in newly diagnosed type 2 diabetic patients and control subjects: a 10-year follow-up on evolution, risk factors, and prognostic significance. Diabetes Care 2000; 23(11): 1672–1678.

15.

Minakaran

Vafidis

Mensah

Proliferative diabetic retinopathy, maculopathy and choroidal neovascularization: concurrent pathology. Invest Ophthalmol Vis Sci 2013; 54: 2433.

16.

Proctor

Ambati

Age-related macular degeneration and diabetic retinopathy: is diabetic retinopathy protective against ARMD?

Invest Ophthalmol Vis Sci 2007; 48(Suppl. 1): 2149.

17.

Hua

Wong

, et al. Choroidal microvascular proliferation secondary to diabetes mellitus. Oncotarget 2017; 8(2): 2034–2036.

18.

Rasheed

Singh

Invernizzi

, et al. Wide-field choroidal thickness profile in healthy eyes. Sci Rep 2018; 8(1): 17166.

19.

Branchini

Adhi

Regatieri

, et al. Analysis of choroidal morphologic features and vasculature in healthy eyes using spectral-domain optical coherence tomography. Ophthalmology 2013; 120(9): 1901–1908.

20.

Wilkinson

Ferris

3rd Klein

, et al. Proposed international clinical diabetic retinopathy and diabetic macular edema disease severity scales. Ophthalmology 2003; 110(9): 1677–1682.

21.

Yoshikawa

Ogata

Wada

, et al. Characteristics of age-related macular degeneration in patients with diabetic retinopathy. Jpn J Ophthalmol 2011; 55(3): 235–240.

22.

Maruko

Iida

Saito

, et al. Clinical characteristics of exudative age-related macular degeneration in Japanese patients. Am J Ophthalmol 2007; 144(1): 15–22.

23.

Cohen

Creuzot-Garcher

Darmon

, et al. Types of choroidal neovascularisation in newly diagnosed exudative age-related macular degeneration. Br J Ophthalmol 2007; 91(9): 1173–1176.

24.

Fukushima

McLeod

Lutty

GA.

Intrachoroidal microvascular abnormality: a previously unrecognized form of choroidal neovascularization. Am J Ophthalmol 1997; 124(4): 473–487.

25.

Huang

Lai

, et al. Prevalence and risk factors for age-related macular degeneration in the elderly Chinese population in south-western Taiwan: the Puzih eye study. Eye 2014; 28(6): 705–714.

26.

Zhang

Liu

, et al. Prevalence of age-related macular degeneration in an elderly urban Chinese population in China: the Jiangning Eye Study. Invest Ophthalmol Vis Sci 2014; 55: 6374–6380.

27.

Wong

, et al. Global prevalence of age-related macular degeneration and disease burden projection for 2020 and 2040: a systematic review and meta-analysis. Lancet Glob Health 2014; 2: e106–e116.

28.

Sander

Larsen

Moldow

, et al. Diabetic macular edema: passive and active transport of fluorescein through the blood-retina barrier. Invest Ophthalmol Vis Sci 2001; 42(2): 433–438.

29.

Choi

Lym

Moon

, et al. Diabetes mellitus and early age-related macular degeneration. Arch Ophthalmol 2011; 129: 196–199.

30.

Chen

Rong

, et al. Diabetes mellitus and risk of age-related macular degeneration: a systematic review and meta-analysis. PLoS ONE 2014; 9(9): e108196.

31.

Lutty

Cao

McLeod

DS.

Relationship of polymorphonuclear leukocytes to capillary dropout in the human diabetic choroid. Am J Pathol 1997; 151(3): 707–714.

32.

Fletcher

Haynes

Crouch

SM.

Acquired abnormalities of polymorphonuclear neutrophil function. Blood Rev 1990; 4(2): 103–110.