Disorganization of Inner Retina and Outer Retinal Morphology inDiabetic Macular Edema

Das, R., Spence, G., Hogg, R. E., Stevenson, M., & Chakravarthy, U. (2018). Disorganization of Inner Retinaand Outer Retinal Morphology in Diabetic Macular Edema. JAMA Ophthalmology, 1-7.https://doi.org/10.1001/jamaophthalmol.2017.6256

Published in:JAMA Ophthalmology

Document Version:Peer reviewed version

Queen's University Belfast - Research Portal:Link to publication record in Queen's University Belfast Research Portal

Publisher rights© 2018 American Medical Association. All rights reserved. This work is made available online in accordance with the publisher’s policies.Please refer to any applicable terms of use of the publisher.

General rightsCopyright for the publications made accessible via the Queen's University Belfast Research Portal is retained by the author(s) and / or othercopyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associatedwith these rights.

Take down policyThe Research Portal is Queen's institutional repository that provides access to Queen's research output. Every effort has been made toensure that content in the Research Portal does not infringe any person's rights, or applicable UK laws. If you discover content in theResearch Portal that you believe breaches copyright or violates any law, please contact openaccess@qub.ac.uk.

Download date:03. Apr. 2022

1

Title- Disorganization of inner retina and outer retinal morphology in diabetic 1

macular edema 2

Authors 3

Radha Das FRCS, Queens University, Belfast 4

Email- radhardas@gmail.com 5

Gareth Spence MB BCh, Belfast health and Social Care Trust 6

Email- gspence1990@gmail.com 7

Ruth E Hogg, PhD, Queens University, Belfast 8

Email- r.e.hogg@qub.ac.uk 9

Michael Stevenson, PhD, Queens University, Belfast 10

Email- wt55@hotmail.com 11

Usha Chakravathy, PhD, Queens University, Belfast 12

Email- u.chakravarthy@qub.ac.uk 13

14

Corresponding Author 15

Professor Usha Chakravarthy 16

Centre for Public health 17

Institute of Clinical Sciences A 18

Queen’s University of Belfast 19

Royal Victoria Hospital, Belfast BT12 6BA 20

Email – u.chakravarthy @qub.ac.uk 21

Phone- 028 9097 1654 22

23

Manuscript word count- 2993 24

2

Question: To characterize retinal morphology using spectral domain optical 25

coherence tomography in a large series of eyes with diabetic macular edema 26

(DME). 27

Findings: In this cross sectional observational case series that included one 28

hundred and two eyes from 80 individuals, disorganization of inner retinal layer 29

(DRIL) was strongly associated with the disruption of outer retinal layers. DRIL 30

was more frequently observed in eyes with increasing severity of diabetic 31

retinopathy and associated with poorer visual acuity. 32

Meaning: DRIL was associated with morphological changes in the outer retina 33

and worse levels of DR in eyes with DME. 34

Abstract 35

36

Importance: In diabetic macular edema (DME), identification of baseline markers 37

on spectral domain optical coherence tomography (SD-OCT) and their 38

relationship with severity of diabetic retinopathy (DR) might aid in the 39

management and in designing future trials. 40

Objective: To examine relationships between DR severity, retinal morphology on 41

SD-OCT and visual acuity (VA) in participants with DME. 42

Design: Cross sectional observational case series. 43

Setting: A single tertiary care referral center. 44

3

Participants: Demographics, VA, SD-OCT and color fundus photographs of 80 45

individuals with DME (102 eyes) seen between December 28th 2013 and 31st 46

April 2014 with data analysed between May and July 2016. 47

On SD-OCT we graded type and shape of DME, shape and presence of septae 48

within the intra retinal cystoid abnormalities, presence of hyper reflective dots and 49

foci, integrity of the external limiting membrane (ELM) and ellipsoid zone (EZ), 50

presence and extent of disorganization of inner retinal layer (DRIL), the status of 51

the vitreomacular interface and epiretinal membrane (ERM). We measured retinal 52

thickness at the fovea (RTF) and at the site of maximum pathology, choroidal 53

thickness (CT) at the fovea and 1000µ temporal and nasal to fovea. Color 54

photographs were graded to derive a DR severity stage. 55

Main Outcomes and Measures: SD-OCT features and thickness metrics. 56

Results: The mean age (SD) was 63 +/-11 and 37.5% were females. The odds of 57

having DRIL were greater in eyes with disrupted ELM (OR 4.4, 95% CI (1.6 to12), 58

P =. 003), disrupted EZ (OR 2.7, 95% CI (1 to 7.2), P = .03), presence of ERM 59

(OR 2.8, CI (1 to 7.4), P =. 03) and increase in RTF (OR 1.6, 95% CI (1.1 to 2.2), 60

P<. 001). DRIL was more likely to occur in eyes with proliferative DR (OR 7.3, 61

95% CI (1.7 to 31.4), P =. 007). Mean VA decreased by approximately 4.7 letters 62

for each 100-μm increase in the average global DRIL (95% CI (-7.9 to 1.4), P=. 63

006). 64

65

Conclusions and Relevance: DRIL was associated with disruption of outer 66

retina and increasing DR severity. Further studies are warranted to establish 67

DRIL as a non-invasive morphological marker in eyes with DME. 68

4

Introduction 69

Diabetic retinopathy (DR) represents a spectrum of pathological changes that 70

occur in the microvasculature of the eye in patients with Diabetes Mellitus (DM). 71

Diabetic macular edema (DME) is a characteristic feature of DR and an important 72

cause of visual loss in people with DM 1. DME is an accumulation of fluid within 73

the macular tissue layers, which arises as a consequence of failure of the blood-74

retinal barrier (BRB). Typically DME causes blurring and distortion of vision, 75

which is reflected in a reduction in visual acuity (VA). The Wisconsin 76

Epidemiologic Study of Diabetic Retinopathy (WESDR) found that 20% of 77

patients with type 1 DM and 25% of those with type 2 DM will develop DME after 78

10 years of follow up 2. 79

Although DME is detectable as increases in macular thickness based on 80

biomicroscopic examination and through stereoscopic assessments of fundus 81

color images, the introduction of time domain optical coherence tomography 82

(OCT) enabled a more precise quantification of overall retinal thickness but this 83

measure was found to correlate only modestly with VA 3. The subsequent 84

availability of spectral domain OCT (SD-OCT) with superior resolution and 85

reproducibility has resulted in improved ability to differentiate early DME from 86

normality and also permits characterization of pathology on a retinal layer by 87

layer basis. Features such as hyper reflective foci can be localized to individual 88

retinal layers or as recently shown to be within the walls of microaneurysms 4. 89

Studies have been undertaken and have established correlations between VA 90

and certain features of retinal morphology such as the intactness of the ellipsoid 91

zone (EZ) and external limiting membrane (ELM) 5-7. A recent report from Sun et 92

al described an OCT feature termed disorganization of the retinal inner layers 93

5

(DRIL), improvement in which was predictive of better VA outcomes 8,9. Notably 94

DRIL was found to be associated with VA after resolution of center-involving DME 95

10. Although relationships between DRIL and VA have been explored, to date no 96

data exists on the relationships between DRIL and other OCT features of DME or 97

with severity of DR. The present study represents a systematic examination of 98

the relationships between DRIL and SD-OCT features of DME, severity of DR 99

and VA in a large series of eyes with DME. 100

101

Methods 102

This was a cross-sectional retrospective observational case series at a single 103

clinical center. The study adhered to the tenets of the Declaration of Helsinki and 104

full ethical approval from the local ethics committee (REC 105

reference14/NI/1025 ). Details of patients attending this clinic recorded in the 106

electronic medical records (EMR) were reviewed. 107

A standardized protocol using trained examiners is used to test and record VA, 108

which is measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) 109

chart. Tomographic images were captured (Spectralis, Heidelberg Engineering, 110

Heidelberg, Germany) on both eyes at all visits and consists of a single horizontal 111

and vertical line scan through the fovea followed by a 6 x 6 mm macular raster 112

scan (37 raster lines spacing of 120μm, 20o *15o ). Color photographs (2 fields -113

macula centered and optic disc centered) were captured on a Visucam 500 (Carl 114

Zeiss Meditec, Oberkochen, Germany) after pupillary dilation. 115

The EMR was interrogated for patients with a diagnosis of DME who attended a 116

retina service between December 28th 2013 and 31st April 2014. The data 117

6

analysis was performed from May 2016 July 2016. 118

From the 94 unique cases we selected the records of 80 patients with a 119

confirmed diagnosis of DM type 1 or 2, with OCT features of DME, color fundus 120

photography and a VA record corresponding to the same visit. Exclusions were 121

diagnosis not DM, age < 18 years, OCT or fundus photographs not available. 122

We extracted information on age, sex, duration of DM, history of hypertension, 123

hyperlipidemia and cigarette smoking from the EMR for each case. 124

Image analysis 125

Grading was performed by two of the authors (R.D and G.S) after training by the 126

senior clinician (U.C.) who checked 20% of the grading outputs. Intergrader 127

variability was assessed by kappa statistics. The images were scrutinized on the 128

proprietary software provided by the manufacturer of the OCT acquisition system. 129

All of the B scans of the 6x6 macular raster were examined and we recorded the 130

type and shape of DME, shape of the intra retinal cystoid abnormalities and 131

septae if seen within the cystoid abnormalities, presence or absence of hyper 132

reflective dots and foci, and the status of the vitreomacular interface and 133

presence of an epiretinal membrane (ERM). We identified the foveal B scan 134

(defined as the scan passing through the maximal foveal depression) and 135

measured the horizontal extent of DRIL and disruption of the ELM and EZ. In the 136

dataset fewer than 10% of eyes were classified as not showing a foveal dip. In 137

these cases where the foveal scan was not clearly identifiable we used the scan, 138

which is offset by 5 degrees from the horizontal as the scan of choice. We also 139

measured the retinal thickness at the fovea (RTF) and at the site of maximum 140

pathology (maximum retinal thickness; MRT). We measured choroidal thickness 141

7

(CT) at the fovea and 1000µ temporal and nasal to fovea. The grading form is 142

shown in eTable 1 in the supplement. 143

We defined DRIL as the presence of a region on the B scan where the 144

boundaries between the ganglion cell–inner plexiform layer complex, inner 145

nuclear layer, and outer plexiform layer could not be separately identified. The 146

method used to determine the horizontal extent of DRIL in each of 7 B-scans was 147

identical that described by Sun et al in that we only measured the diameter upto a 148

maximum of 1000 μM. The B scans that were selected were the foveal scan and 149

the 3 immediately superior and inferior to the foveal scan. The measurements 150

from these 7 scans were summed to derive an average global DRIL measure for 151

each eye 8. 152

We graded the color photographs according to the ETDRS grading protocol 11 to 153

derive a DR severity stage. 154

Statistical analysis 155

We analyzed the data using SPSS version 22. We performed kappa statistics to 156

test the intergrader variability. We examined relationships between VA and age 157

and duration of diabetes using linear regression. We performed one-way 158

analysis of variance (ANOVA) for associations between VA with hypertension, 159

hyperlipidemia and smoking. We also used ANOVA to test relationships 160

between VA and the various categorical variables (retinal morphological features 161

observed on OCT and DR severity staging). . We used Pearson’s coefficient to 162

test for correlations between VA and OCT linear variables (diameter of DRIL, 163

disrupted ELM and EZ, height of RTF, MRT, and CT). With VA as the dependent 164

variable, those variables that reached significance in the preceding steps were 165

8

then tested using general linear regression modelling (GLM) after controlling for 166

RTF, which had been previously identified as a confounder factor by Sun et al 8. 167

As DRIL had been identified as a key determinant of VA outcome8,we performed 168

binary logistic regression to seek associations between presence or absence of 169

DRIL with SD OCT features and DR severity. We tested pairs of variables and we 170

did not create a multivariable model as the modest sample size could have lead 171

to invalidation of the independence of the observations on the variable of interest. 172

The interclass correlation of eyes with the same person was 0.08 and so we did 173

not deem it necessary to adjust for correlations between 2 eyes. We considered 174

P ≤. 05 as statistically significant in these analyses. 175

Results 176

Descriptive characteristics of the study cohort are shown in Table 1. 177

One hundred and two eyes of 80 individuals were analysed. Due to poor quality 178

OCT images, one eye was not graded for OCT morphology and metrics. The 179

mean age was 63 (SD, 11) years, 30 (37.5%) were females, with the majority of 180

the sample having type 2 DM (Table 1). The frequency of the DR severity stages 181

in the 100 eyes (2 eyes were not graded due to poor quality colour images) is 182

shown in Table 1. Less than half of the included eyes had been previously treated 183

with macular laser and none had received an anti vascular endothelial growth 184

factor (VEGF) treatment. The intergrader variability was tested using kappa and 185

as large number of variables were graded, we provide the range of kappa values, 186

which was from 0.6 to 1. Linear regression of VA against age and duration of DM 187

showed no significant associations. ANOVA also confirmed the absence of 188

associations with key systemic features of hypertension, hyperlipidemia and 189

9

smoking. (e Table 2). Therefore these variables were not included in any of the 190

subsequent analyses. 191

Associations between VA and retinal morphological features identified specific 192

variables that reached statistical significance on the ANOVA. These were, shape 193

of DME, presence of septae in intraretinal cystoid abnormalities, presence of 194

DRIL and disruption of ELM and EZ and DR severity stage. Other variables that 195

were tested [type and symmetry of intra retinal fluid, presence of cystoid 196

abnormalities and ERM] did not reach significance (eTable 3). Pearsons 197

correlation coefficients (e Table 4) showed significant correlations between VA 198

and horizontal extent of DRIL, ELM and EZ, RTF and MRT. 199

The results from GLM adjusted for RTF are shown in Table 2. Variables that lost 200

significance in the model were shape of DME and presence of septae in cystoid 201

abnormalities (Table 2) while all others remained significant. The parameter 202

estimates show that better VA is associated with absence of DRIL, intact ELM 203

and EZ and less severe stages of DR (Table 2). Presence of intact ELM is 204

associated with a 14 letter better mean VA (14 (95 % CI 7.9 to 19.8) P < 0.001) 205

compared to eyes with disrupted ELM. Similar differences in VA were also found 206

when the EZ was intact (10.5(4.7 to 16.4) P = 0.001) and when DRIL was absent 207

(9.8 (3 to 16.6) P < 0.001). 208

The parameter estimates indicate that for each 100-μm increase in the average 209

global DRIL, there is a decrease in the mean VA of approximately 4.6 letters 210

(Table 3). With each 100 μm increase in disruption of ELM and EZ, the mean VA 211

decreased by 0.4 letters. For each 100 μm of increase in RTF and MRT, VA 212

decreased by 2.4 and 0.04 letters respectively. After adjustment for RTF, all the 213

variables remained significant. The MRT variable showed a strong impact on VA 214

10

with each 100-μm increase in retinal thickness resulting in a decrease of 13.1 215

letters (-19.6 to -6.6, P <0.001). Figure 1A shows the presence of disrupted ELM 216

and EZ. 217

218

Relationship between DRIL and other OCT features of DME and DR 219

Binary logistic regression showed that the odds of having DRIL was greater in 220

eyes with disrupted ELM (odds ratio 4.4, 95% CI (1.6 to12), P=. 003) and 221

disrupted EZ (OR 2.7,1 to 7.2, P =. 03), presence of ERM (OR 2.8, 1 to 7.4, P =. 222

03) and increase in RTF (OR 1.6,1.1 to 2.2, P <. 001) (Table 4). DRIL was more 223

likely to occur in eyes with PDR (OR 7.3, 1.7 to 31.4, P = .007) (Table 4). Figure 224

1B is an example of DRIL in an eye with ERM. 225

Discussion 226

We systematically graded the tomographic images of the retina and sought 227

relationships between the presence and severity of a number of morphological 228

features of DME including DRIL a novel and recently described biomarker with 229

severity of DR and visual function. Results from this study suggest that centrally 230

located DRIL is correlated with VA in eyes with center-involved DME and was 231

strongly associated with the disruption of ELM and EZ layers and presence of 232

ERM. Worsening of VA was seen with increasing DR severity and may be related 233

to the presence of DRIL. The presence of central location of DRIL and the extent 234

of disruption were both associated with worse VA and these findings are similar 235

to previous studies, which found DRIL to be a predictive marker for VA outcomes 236

in eyes with resolved DME after treatment 8-10. A an important finding from our 237

analysis has been the elucidation of the relationship between the horizontal 238

extent of DRIL and VA and we have shown that for each 100 micron increase in 239

11

DRIL there is a negative impact of around 6 letters which is more than one line on 240

the Early Treatment Diabetic Retinopathy Study (ETDRS) chart. It has been 241

hypothesized that disorganization of the inner retina occurs when bipolar axons 242

snap at a point when their elasticity limit has been exceeded due to edema 12. It 243

has also been suggested that DRIL represents loss of bipolar, amacrine or 244

horizontal cells within the inner retinal layers 8. Key findings from the present 245

analysis also include the highly significant relationships between presence of 246

DRIL and outer retinal changes. DRIL was strongly associated with the disruption 247

of ELM and EZ layers. In our study RTF was significantly increased in the 248

presence of DRIL and taken together suggest that the mechanisms that create 249

the conditions for inner retinal disorganization may also be responsible for 250

concurrent disruption of the outer retinal architecture. 251

We noted that these outer retinal changes were also associated with poor VA and 252

our findings are consistent with those from previous studies 5,7,13,14. . Breakdown 253

of the BRB in DME could create the conditions for damage to the ELM and EZ. 254

Unlike DRIL where mechanical forces have been considered in its pathogenesis, 255

it has been postulated that the disrupted ELM occurs as a consequence of 256

impaired retinal function 14. It is presently unknown if the EZ line seen on OCT 257

images truly corresponds to the histologic junction of the inner and outer 258

segments. An important consideration with the growing literature regarding the 259

EZ is how to objectively and consistently evaluate disruption. Maheshwary et al 5 260

concluded that not only is the status of Inner segment/Outer segment, disruption 261

important, but the percentage disruption is a predictor of visual acuity in a patient. 262

Spaide and Curcio speculated that this highly reflective band was located at the 263

ellipsoid in the inner segments, considering the correlation between the 264

12

microstructure on the SD-OCT images and the histologic findings 15. Our present 265

work not only reinforces the important clinical relevance of an intact EZ but also 266

suggests that the same pathogenic pathways that create conditions for DRIL may 267

also disrupt the outer retinal architecture. 268

Cystic changes that appear within the macula represent focal areas of coalesced 269

extra cellular fluid. These foci of fluid likely result from disturbed cellular function 270

of the Müller cells, which are thought to play an important role in acting as 271

metabolic pumps to keep the macula dehydrated 16,17. Based on the univariate 272

analysis undertaken in the present study we noted that VA was better in eyes 273

with septae within the cystoid abnormalities compared to those without and our 274

findings support those of previous investigators. 275

Many studies have shown that ERM is common in eyes with DME and similarly 276

we observed a high frequency of ERM in the present series. We systematically 277

graded for ERMs that are recognized on OCT as thin, hyperreflective bands 278

anterior to the retina or bright red bands in the pseudocolor representation on 279

OCT 18. Gandorfer et al showed that ERMs in DME consist of multilayered 280

membranes situated on a layer of native vitreous collagen, predominantly 281

coupled with fibroblasts and fibrous astrocytes 19. Removal of the ERM by 282

vitrectomy and membrane peel can result in resolution of the DME and 283

improvements in VA 19,20. To date no study has examined the associations 284

between ERM, DRIL and VA and our observations of a strong relationship 285

between these features suggest a common pathological process. Histological 286

studies are required to understand the pathogenetic mechanisms underpinning 287

these inner retinal morphological manifestations. 288

13

A prospective study investigated the correlation between the features of OCT 289

and the severity of retinopathy 21. This study found that the prevalence of DME 290

with serous retinal detachment and with vitreomacular traction was higher in eyes 291

with severe NPDR to PDR than in eyes with mild-to-moderate 21. It is also known 292

that increasing DR severity is associated with macular thickening 22. While the 293

present study also shows similar findings it is the first to report an association 294

between DRIL and increasing severity of DR. Our finding of worsening of VA with 295

increasing DR severity may be related to the presence of DRIL. 296

A biomarker that is of interest is the visibility of the cone outer receptor segments 297

(COST). It has been shown that there is reduced visibility of COST in patients 298

with DME. During grading we observed that COST lines could not be delineated 299

consistently therefore we did not grade for this biomarker. Our experience is 300

similar to that of Rii et al who found that these outer retinal layers could not be 301

segmented even in normal eyes scans 23. 302

The strengths of this study are the systematic grading of color and OCT images 303

and the homogenous sample of patients with DME naïve to treatment with anti 304

VEGF. An important limitation of the study is its cross-sectional design and 305

hence we are unable to predict the impact on function over time of the various 306

OCT markers that we identified. Also our study is retrospective and, the sample 307

size is modest. Furthermore the data were acquired as part of routine clinical care 308

and thus many patients had prior exposure to laser photocoagulation, which can 309

alter the appearance of the retinal layers. However we believe our findings do not 310

arise as a consequence of laser treatment since the foveal retina is avoided while 311

undertaking macular laser. 312

313

14

In summary we have systematically studied both inner and outer retinal 314

morphology in DME and relationships both with VA and increasing severity of DR. 315

316

Conclusion 317

DRIL correlated with the integrity of outer retinal architecture and the presence of 318

ERM. The odds of having DRIL increased with DR severity. DRIL requires further 319

investigation in large clinical trials across multiple sites with prospective data 320

collection with adjustment for multiple potential confounders including VA and 321

other morphological features on OCT to be considered as a non-invasive marker. 322

323

Acknowledgement- 324

Dr Alyson Muldrew, senior grader, netWORC UK 325

Professor Usha Chakravarthy had full access to all the data in the study and 326

takes responsibility for the integrity of the data and the accuracy of the data 327

analysis. 328

329

330

331

332

333

334

335

336

337

338

15

References 339

1. Moss SE, Klein R, Klein BE. The 14-year incidence of visual loss in a diabetic population. 340

Ophthalmology. 1998;105(6):998-1003. doi: 10.1016/S0161-6420(98)96025-0. 341

2. Klein R, Klein BE, Moss SE, Cruickshanks KJ. The wisconsin epidemiologic study of diabetic 342

retinopathy. XV. the long-term incidence of macular edema. Ophthalmology. 1995;102(1):7-16. 343

3. Diabetic Retinopathy Clinical Research Network, Browning DJ, Glassman AR, et al. 344

Relationship between optical coherence tomography-measured central retinal thickness and visual 345

acuity in diabetic macular edema. Ophthalmology. 2007;114(3):525-536. doi: 346

10.1016/j.ophtha.2006.06.052. 347

4. Bolz M, Schmidt-Erfurth U, Deak G, et al. Optical coherence tomographic hyperreflective foci: 348

A morphologic sign of lipid extravasation in diabetic macular edema. Ophthalmology. 349

2009;116(5):914-920. doi: 10.1016/j.ophtha.2008.12.039; 10.1016/j.ophtha.2008.12.039. 350

5. Maheshwary AS, Oster SF, Yuson RM, Cheng L, Mojana F, Freeman WR. The association 351

between percent disruption of the photoreceptor inner segment-outer segment junction and visual 352

acuity in diabetic macular edema. Am J Ophthalmol. 2010;150(1):63-67.e1. doi: 353

10.1016/j.ajo.2010.01.039; 10.1016/j.ajo.2010.01.039. 354

6. Otani T, Yamaguchi Y, Kishi S. Correlation between visual acuity and foveal microstructural 355

changes in diabetic macular edema. Retina. 2010;30(5):774-780. doi: 356

10.1097/IAE.0b013e3181c2e0d6; 10.1097/IAE.0b013e3181c2e0d6. 357

7. Shin HJ, Lee SH, Chung H, Kim HC. Association between photoreceptor integrity and visual 358

outcome in diabetic macular edema. Graefes Arch Clin Exp Ophthalmol. 2012;250(1):61-70. doi: 359

10.1007/s00417-011-1774-x; 10.1007/s00417-011-1774-x. 360

16

8. Sun JK, Lin MM, Lammer J, et al. Disorganization of the retinal inner layers as a predictor of 361

visual acuity in eyes with center-involved diabetic macular edema. JAMA Ophthalmol. 362

2014;132(11):1309-1316. doi: 10.1001/jamaophthalmol.2014.2350 [doi]. 363

9. Sun JK, Radwan SH, Soliman AZ, et al. Neural retinal disorganization as a robust marker of 364

visual acuity in current and resolved diabetic macular edema. Diabetes. 2015;64(7):2560-2570. 365

doi: 10.2337/db14-0782 [doi]. 366

10. Radwan SH, Soliman AZ, Tokarev J, Zhang L, van Kuijk FJ, Koozekanani DD. Association 367

of disorganization of retinal inner layers with vision after resolution of center-involved diabetic 368

macular edema. JAMA Ophthalmol. 2015;133(7):820-825. doi: 369

10.1001/jamaophthalmol.2015.0972 [doi]. 370

11. ETDRS report number 10. Grading diabetic retinopathy from stereoscopic color fundus 371

photographs--an extension of the modified airlie house classification. Early treatment diabetic 372

retinopathy study research group. Ophthalmology. 1991;98(5 Suppl):786-806. 373

12. Pelosini L, Hull CC, Boyce JF, McHugh D, Stanford MR, Marshall J. Optical coherence 374

tomography may be used to predict visual acuity in patients with macular edema. Invest 375

Ophthalmol Vis Sci. 2011;52(5):2741-2748. doi: 10.1167/iovs.09-4493 [doi]. 376

13. Forooghian F, Stetson PF, Meyer SA, et al. Relationship between photoreceptor outer segment 377

length and visual acuity in diabetic macular edema. Retina. 2010;30(1):63-70. doi: 378

10.1097/IAE.0b013e3181bd2c5a; 10.1097/IAE.0b013e3181bd2c5a. 379

14. Murakami T, Nishijima K, Sakamoto A, Ota M, Horii T, Yoshimura N. Association of 380

pathomorphology, photoreceptor status, and retinal thickness with visual acuity in diabetic 381

retinopathy. Am J Ophthalmol. 2011;151(2):310-317. doi: 10.1016/j.ajo.2010.08.022; 382

10.1016/j.ajo.2010.08.022. 383

17

15. Spaide RF, Curcio CA. Anatomical correlates to the bands seen in the outer retina by optical 384

coherence tomography: Literature review and model. Retina. 2011;31(8):1609-1619. doi: 385

10.1097/IAE.0b013e3182247535 [doi]. 386

16. Rotsos TG, Moschos MM. Cystoid macular edema. Clin Ophthalmol. 2008;2(4):919-930. 387

17. Deak GG, Bolz M, Ritter M, et al. A systematic correlation between morphology and 388

functional alterations in diabetic macular edema. Invest Ophthalmol Vis Sci. 2010;51(12):6710-389

6714. doi: 10.1167/iovs.09-5064 [doi]. 390

18. Panozzo G, Parolini B, Gusson E, et al. Diabetic macular edema: An OCT-based 391

classification. Semin Ophthalmol. 2004;19(1-2):13-20. 392

19. Gandorfer A, Rohleder M, Grosselfinger S, Haritoglou C, Ulbig M, Kampik A. Epiretinal 393

pathology of diffuse diabetic macular edema associated with vitreomacular traction. Am J 394

Ophthalmol. 2005;139(4):638-652. doi: S0002-9394(04)01398-4 [pii]. 395

20. Sakimoto S, Saito Y, Nakata K, Sakamoto Y, Tatebayashi M. Surgical outcomes of epiretinal 396

membrane removal after successful pars plana vitrectomy for retinal diseases. Jpn J Ophthalmol. 397

2008;52(3):227-230. doi: 10.1007/s10384-008-0539-5 [doi]. 398

21. Alkuraya H, Kangave D, Abu El-Asrar AM. The correlation between optical coherence 399

tomographic features and severity of retinopathy, macular thickness and visual acuity in diabetic 400

macular edema. Int Ophthalmol. 2005;26(3):93-99. doi: 10.1007/s10792-006-9007-8 [doi]. 401

22. Browning DJ, Fraser CM, Clark S. The relationship of macular thickness to clinically graded 402

diabetic retinopathy severity in eyes without clinically detected diabetic macular edema. 403

Ophthalmology. 2008;115(3):533-539.e2. doi: S0161-6420(07)00749-X [pii]. 404

18

23. Rii T, Itoh Y, Inoue M, Hirakata A. Foveal cone outer segment tips line and disruption 405

artifacts in spectral-domain optical coherence tomographic images of normal eyes. Am J 406

Ophthalmol. 2012;153(3):524-529.e1. doi: 10.1016/j.ajo.2011.08.021 [doi]. 407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

19

Table 1. Descriptive characteristics of the study cohort 430

431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471

Abbreviations- DM, diabetes mellitus; NPDR, non proliferative diabetic 472retinopathy; PDR, proliferative diabetic retinopathy; PRP, pan retinal 473photocoagulation. 474a Total of 100 as 2 eyes were ungradeable. 475 476

477

Characteristic of study population

Total cohort n= 80

Age, mean +/-SD, years 63 +/-11

Females, n (%) 30(37.5)

DM duration, mean +/-SD, years 16.74 +/-10.84

DM type, n (%)

Type 1 17(21.3)

Type 2 63 (78.8) Hypertension n (%) 51(63.7)

Hyperlipidemia n (%) 54(67.5)

Smoking n (%) 33(41.3)

Ocular Characteristic n=102

Diabetic retinopathy stage n (%) a

Mild NPDR 53(52)

Moderate NPDR 37(35.3)

Severe NPDR 0(0)

PDR 10(10.8)

Previous laser surgery n (%)

None 27(26.5)

Macular laser 44(43.1)

PRP 14(13.7)

Both Macular laser and PRP 17(16.7)

Visual acuity, mean +/-SD, letters

54.37 +/-16.13

20

Table 2. Associations between VA and retinal morphological features of DME 478and DR a 479Variable VA (letters)

mean +/-SD B (95%CI) P value

Shape of DME Dome 54.7 +/- 14.4 -1.7 (-9.1 to 5.5) .62 Fusiform 56.9 +/- 15.3 reference Septae in cystoid abnormalities

Yes 57.15 +/- 13.5 reference .19 No 49.95 +/-15.0 -4.7 (-11.9 to 2.4) DRIL Yes 46.1+/-14.9 reference No 58.1+/-13.5 9.8 (3.0 to 16.6) <.001 b ELM disrupted Yes 46.1+/-14.8 reference No 61.0+/-11.5 13.9(7.9 to 19.8) <.001 b EZ disrupted Yes 48.2+/-14.8 reference No 60.6+/-12.4 10.5(4.7 to 16.4) .001 b DR severity stage Mild NPDR 59.9+/-14.09 11.6 (2.0 to 21.1) .01 b Moderate NPDR 55.0+/-14.6 8.8(-1.0 to 18.7) .07 PDR 45.1+/-14.2 reference Abbreviations- VA, visual acuity; DME, diabetic macula edema; DRIL, 480disorganization of inner retinal layers; ELM, external limiting membrane; EZ, 481ellipsoid zone; DR, diabetic retinopathy; NPDR, non proliferative diabetic 482retinopathy; PDR, proliferative diabetic retinopathy. 483a Results are reported from General linear regression model adjusted for Retinal 484thickness at fovea. 485b P <.05. 486

487

488

489

490

491

492

493

494

495

21

Table 3- Correlations between VA and OCT derived linear variables a 496

Variable B (95%CI) P value

B (95%CI) P value ( adjusted for RTF)

RTF per 100 μm -2.4(-4 to -0.7), .005b MRT per 100 μm -0.04 (-.08 to-.01), .01 b -13.1(-19.6 to -6.6),

<.001 b Average global DRIL per 100 μm

-4.6(-8 to -1.3), .009 b -4.7 (-7.9 to -1.4), .006 b

ELM disruption at fovea per 100 μm

-0.4(-0.8 to -0.1), .01 b -0.5 (-0.9 to -0.1), .01 b

EZ disruption at fovea per 100 μm

-0.4(-0.7 to -0.1), .003 b -0.4(-0.7 to -0.1), .008 b

Abbreviations- VA, visual acuity; OCT, optical coherence tomography; RTF, retinal 497thickness at fovea; MRT, maximum retinal thickness; DRIL, disorganization of 498inner retinal layers; ELM, external limiting membrane; EZ, ellipsoid zone. 499a Results are reported from General linear regression model. 500b P<.05 501

502

503

504

505

506

507

508

509

510

511

512

513

514

515

516

517

22

Table 4- Relationship between DRIL with other retinal morphological 518features of DME and DR a 519 DRIL OR (95% CI) P value Variable n

Yes n

No n

Type of DME IRF only 17 55 IRF and SRF 6 23 0.8(.2 to 2.4) .7 Type of IRF Diffuse 4 17 reference Cystoid 5 21 1(0.2 to 4.3) .9 Mixed 14 40 1(0.2 to 4.3) .5 Cystoid abnormalities in inner retina

Yes 20 61 1.8(.4 to 7) .3

No 3 17 reference Septae in the cystoid abnormalities

Yes 13 46 0.6(.2-1.7) .3

No 7 15 reference IRF symmetry at fovea

Symmetrical 17 52 0.7(.2 to 2) .5

Asymmetrical 6 26 reference DME shape Dome 14 37 0.5(0.2 to1.4) .2 Fusiform 9 41 reference ERM Yes 11 19 2.8(1 to 7.4) .03b No 12 59 reference ELM disrupted

Yes 15 23 4.4(1.6 to12) .003 b

No 8 55 reference EZ disrupted Yes 14 28 2.7(1 to 7.2) .03 b No 9 50 reference RTF (linear variable)

NA NA 1.6(1.1 to 2.2)

<.001 b

DR severity scale

Mild NPDR 9 44 reference

Moderate NPDR

8 29 1.3(.4 to 3.8) .5

PDR 6 4 7.3(1.7 to 31.4)

.007 b

Abbreviations- DME, diabetic macular edema;DR, diabetic retinopathy; IRF, 520intraretinal fluid; SRF, subretinal fluid; ERM, epiretinal membrane;ELM, external 521limiting membrane; EZ, ellipsoid zone;RTF, retinal thickness at fovea; NPDR, non 522proliferative diabetic retinopathy; PDR, proliferative DR. 523a Results are reported form Binary logistic regression with DRIL presence as the 524dependent variable. 525b P <.05 526 527

23

528

Figure 1 A and B 529

530

531

532

A- SD-OCT image showing disrupted ELM and EZ in DME. The yellow line 533

denotes disruption of the ELM and EZ 534

B- SD-OCT image showing the presence of DRIL and ERM in DME. 535

Arrow depicts the ERM. The area of DRIL is denoted by a line where 536

the retinal boundaries cannot be identified. 537

538

539

540

541

24

542

543

544

545

546

547

548