New Trends for Early Diabetic Retinopathy Diagnosis

Joana Tavares Ferreira

1, 2

and Luís Abegão Pinto

3, 4

Department of Ophthalmology, Central Lisbon Hospital Center, Lisbon, Portugal

NOVA Medical School, Universidade Nova de Lisboa, Lisbon, Portugal

Department of Ophthalmology, Northern Lisbon Hospital Center, Lisbon, Portugal

Visual Sciences Study Center, Faculty of Medicine, Lisbon University, Lisbon, Portugal

Keywords: Diabetic Retinopathy, Optical Coherence Tomography, Retinal Layers, Neurodegeneration, Choroidal

Thickness.

Abstract: Diabetes Mellitus is one of the most common chronic diseases in the world and is a critical public health

problem that could even be considered a pandemic. The diabetic retinopathy is the leading cause of blindness

in adults. Diabetic retinopathy is now considered to be a new neurodegenerative disease. In fact, retinal

neurodegeneration is present before any microcirculatory abnormalities can be detected in ophthalmoscopy.

Functional studies documenting electroretinogram abnormalities, loss of dark adaptation, contrast sensitivity

and colour vision and abnormal microperimetry that occur before any vascular abnormality. Novel imaging

optical devices have allowed that this pre-vascular damage to be quantified in a non-invasive and reproducible

way with retinal layer and choroidal thickness measurement.

1 INTRODUCTION

Diabetic retinopathy (DR) is the leading cause of

legal blindness among working-aged adults in the

United States (Klein, 2007). Of the 415 million

diabetic patients worldwide in 2015, over one-third

will develop DR in their lifetime (International

Diabetes Federation (IDF), 2015).

The RETINODIAB study, an epidemiologic

study that investigated the prevalence and

progression rates of DR based on a national screening

community program in Portugal, identified a 16.3%

prevalence rate of DR and a 4.6% incidence rate of

any DR within the first year in diabetic patients

without retinopathy at baseline (M. Dutra Medeiros et

al., 2015; Marco Dutra Medeiros et al., 2015).

The International Clinical Classification of DR is

based on the observation of microvascular retinal

changes (‘Grading diabetic retinopathy from

stereoscopic color fundus photographs--an extension

of the modified Airlie House classification. ETDRS

report number 10. Early Treatment Diabetic

Retinopathy Study Research Group.’, 1991).

However, diabetic neuroretinal degeneration has been

demonstrated in histological studies and through the

measurement of functional loss with a number of

functional tests, including contrast vision, color

vision, visual field, dark adaptation and

electroretinogram. These retinal neurodegenerative

changes include apoptosis of several populations of

retinal cells (e.g., photoreceptors, bipolar cells,

ganglion cells and astrocytes) with consequent effects

on the thickness of different retinal layers in the

earliest stages of DR or when DR cannot be detected

by ophthalmologic examination (Barber et al., 1998;

Carrasco et al., 2007, 2008; Garcia-Ramírez et al.,

2009). Furthermore, it has been hypothesized that

changes in the choriocapillaris may precede the

development of DR (Nagaoka et al., 2004). However,

the relationship between DR and diabetic

choroidopathy remains unclear.

Recently, optical coherence tomography (OCT)

has been introduced into clinical practice as the most

non-invasive and objective method to visualize the

retina, showing an amount of detail that resembles

histological specimens (Fischer et al., 2009; van Dijk

et al., 2011). Initially, OCT was applied to detect

complications of DR (edema macular or epiretinal

membrane) (Ceklic, Maár and Neubauer, 2008). Later

on, it allowed to perform quantitative and qualitative

measurements of retinal thickness and segmentation

of all intraretinal layers (van Dijk et al., 2009, 2010,

2012; Vujosevic and Midena, 2013; Tavares Ferreira

J et al., 2016), Figure 1. The new Spectralis Spectral

Tavares Ferreira J. and AbegÃ

co Pinto L.

New Trends for Early Diabetic Retinopathy Diagnosis.

DOI: 10.5220/0006328304020406

Domain-OCT automatic segmentation software

demonstrated excellent repeatability and

reproducibility of each of the eight individual retinal

layer thickness measurements (Ctori and Huntjens,

2015). Potentially, OCT might detect early retinal

changes, and thus help define which diabetic patients

may be at-risk to develop DR. Ultimately, it could be

used to plan preventive therapy before the

development of vascular lesions detectable by

ophthalmoscopy (Simó and Hernández, 2014).

However, up until now, the smaller scale, mostly pilot

studies or only focusing on specific retinal layers on

this topic in OCT image analysis did not show a

temporal relationship between DM duration or arising

DR and the changes observed in retinal layers.

Figure 1: Retinal layer segmentation. Figure adapted to

(Tavares Ferreira J, Alves M, Dias-Santos A, Costa L,

Santos BO, Cunha JP, Papoila AL, 2016)

This review aims to summarize recent literature

concerning retinal structure changes in diabetic

patients and its relationship with diabetic retinal

disease progression.

2 OPTICAL COHERENCE

TOMOGRAPHY IN EARLY

DIAGNOSIS OF DIABETIC

RETINOPATHY

Several studies using Spectral Domain (SD)-OCT,

showed a decreased retinal nerve fiber layer (RNFL)

or (ganglion cell layer) GCL thickness in diabetic

patients without DR (Vujosevic and Midena, 2013;

Chhablani et al., 2015; Carpineto et al., 2016).

However, Hille van Dijk et al. did not find differences

in any inner layer thickness between non-diabetic and

type 1 or type 2 diabetic patients even without DR

(van Dijk et al., 2010, 2012). Nevertheless, different

SD-OCT devices were used (Cirrus, Topcon or

Nidek) and the diabetic patient samples were very

small (30 (Vujosevic and Midena, 2013), 20

(Chhablani et al., 2015), 19 (van Dijk et al., 2010) or

39 (van Dijk et al., 2012) patients to perform a

reliable multivariable analysis. Only Carpineto et al.

studied 131 type 2 diabetic patients without DR using

Cirrus SD-OCT, and identified a reduced ganglion

cell-inner plexiforme layer (GC-IPL) and RNFL

thickness compared with healthy controls (Carpineto

et al., 2016). Vujosevic et al. studied both inner and

outer layers but in opposition to this work they did not

find any differences in the retinal pigment epithelium

(RPE) and photoreceptor (PR) layers thickness.

However, these authors have studied the RPE and PR

layers together not individualizing them in two

different layers.

Tavares Ferreira et al, used SD-OCT to compare

the retinal layers thickness between non-diabetic

subjects and type 2 diabetic patients without DR and

with different Diabetes Mellitus (DM) duration

(Tavares Ferreira J et al., 2016). In their multivariable

regression models, after adjusting for age, gender,

intraocular pressure (IOP) and axial length, and

correcting for multiple testing, no difference in the

overall retinal total (RT) thickness throughout the

Early Treatment Diabetic Retinopathy Study

(ETDRS) areas was found. Interestingly, the patterns

of layer distribution were not the same in the two

samples. The PR layer was the most consistent

finding, with a smaller thickness in diabetic patients

when compared to their non-diabetic controls, Figure

Figure 2: Photoreceptor layer thickness in all groups,

determined automatically by SD-OCT in nine ETDRS areas

in the macula. Figure adapted to (Tavares Ferreira J, Alves

M, Dias-Santos A, Costa L, Santos BO, Cunha JP, Papoila

AL, 2016).

Nevertheless, the pattern of thickness in this layer

differs with disease duration. Once stratified diabetic

patients according to this parameter, the thinner

layers could be found in patients with both an early

(group I) and longer known diabetes diagnosis (group

III) (p<0.001). On the other hand, the thinning in PR

in diabetic patients with moderate duration (group II)

did not reach statistical significance when compared

to the healthy controls. The remaining layers (outer

nuclear layer - ONL, outer plexiform layer - OPL,

inner nuclear layer - INL and GCL) showed an overall

tendency towards a thicker layer in diabetic retinas

when compared to non-diabetic patients, but did not

reached statistical significance (Tavares Ferreira J et

al., 2016).

PR layer was not uniform throughout disease

duration. This could be interpreted as a temporary

cellular swelling due to a number of reasons, ranging

from the diabetic induced hypoxia (Kern and

Berkowitz, 2015), oxidative stress with increased

generation of superoxide and other reactive oxygen

species in the retina (Du et al., 2013) which induces

the release of pro-inflammatory molecules and

changes in retinal vasculature. Ultimately, the

continuous cellular swelling is known to lead to a

cellular atrophy (Kern and Berkowitz, 2015),

potentially explaining the thinnest PR layer in the

patients with longer disease duration. This non-linear

behaviour is important as it can explain the

contradictory results in this field, as each study may

be recruiting patients with a different disease

duration. Additionally, it could be clinically relevant

as studies have suggested the importance of the PR

layer in the development of DR, loss of PR reduced

the severity of vascular degeneration in DR (Arden,

2001; De Gooyer et al., 2006). Further studies would

be needed to interpret such findings.

These same authors did a longitudinal study,

based on the baseline study referred in which the aim

was to evaluate which diabetic patients without DR

would develop DR after one year and to use SD-OCT

to detect changes in retinal and choroidal layers over

a period of one year (Tavares Ferreira J et al., 2016).

These 125 type 2 diabetic patients without DR

showed that after one year, independent of the

development of DR, the choroidal thickness (CT)

increased between 10 and 17 µm (p<0001 to 0003),

and there was a decrease in the GCL (I3 and N6

sectors), IPL (S6 and N6 sectors), INL (T6 and N6

sectors), OPL (S6 sector) and overall RT (S3, N3, I3,

S6 and T6 sectors) (p<0.001). Interestingly, in this

study, the variable retinopathy was negatively

associated with the overall RT (central, S3, T3 I3 and

N3 sectors), ONL (T6 and I6 sectors) and PR layer

(N6 sector). In the span of just one year, the presence

of DR decreased the overall RT in the studied

locations between 13.04 and 16.63 µm (Tavares

Ferreira J et al., 2016).

Overall these results may be explained by a

process of inflammation which accompanies or

precedes the early cell apoptosis of the DR. Thus,

before there is a thinning of the different inner retinal

layers, there is a significant thickening compared to

non-diabetic patients. When some cells are in

apoptotic stage and others in inflammatory phase

probably we will not find any significant differences

in the thickness of the retinal layers. Further, when

analysing the RPE layer, a significant association

between the RPE thickness and the CT was noticed,

so it is important to include the variable CT in the

regression models. Ferreira et al. found that the CT in

diabetic patients without DR was increased in

comparison with non-diabetic subjects, (Ferreira et

al., 2015) and since the choroid supplies the RPE the

referred association between CT and RPE would be

expected.

This last study had some limitations (Tavares

Ferreira J et al., 2016). Firstly, despite including 125

diabetic patients without DR, when divided into

groups according to diabetes duration, their sample

sizes became small. However this is the first study

with a considerable sample that splits the diabetic

patients according to disease duration, and yet finding

differences in these subgroups compared to non-

diabetic subjects. Secondly, retinal measurements

were done with automatic software. The ideal retinal

layers segmentation is one that involves automatic

segmentation with supervision and manual correction

when necessary. In this way, a manual correction was

performed when the segmentation was inaccurate by

an ophthalmologist masked to the patients’ diagnosis.

Thirdly, all diabetic patients had type 2 DM meaning

that the onset of diabetes was self-reported and could

thus be underestimated.

3 CONCLUSIONS

In conclusion, diabetic patients without DR have a

thinning of some retinal layers (inner retinal and PR

layers), when compared to a non-diabetic group.

There are early changes in retinal layers of diabetic

patients even without clinical signs of DR that

probably correspond to an inflammatory and

apoptotic process of the retina as neurovascular unit.

In diabetic patients without DR at the one-year

follow-up point, was observed overall thickening of

the choroid and decreases in the thickness of the inner

retinal layers (GCL, IPL and INL) and overall retinal

thickness (RT). Thus, when patients develop DR, the

choroid begins to decrease along with the overall RT

and PR layer thickness.

The OCT is a new technology that allows the

diagnosis of early retinal and choroid structural

changes of diabetic patients before the onset of

diabetic retinopathy.

REFERENCES

Arden, G. B. (2001) ‘The absence of diabetic retinopathy in

patients with retinitis pigmentosa: implications for

pathophysiology and possible treatment.’, The British

journal of ophthalmology, 85(3), pp. 366–70.

Barber, A. J., Lieth, E., Khin, S. A., Antonetti, D. A.,

Buchanan, A. G. and Gardner, T. W. (1998) ‘Neural

apoptosis in the retina during experimental and human

diabetes. Early onset and effect of insulin.’, The Journal

of clinical investigation, 102(4), pp. 783–91.

Carpineto, P., Toto, L., Aloia, R., Ciciarelli, V., Borrelli, E.,

Vitacolonna, E., Di Nicola, M., Di Antonio, L. and

Mastropasqua, R. (2016) ‘Neuroretinal alterations in

the early stages of diabetic retinopathy in patients with

type 2 diabetes mellitus.’, Eye (London, England),

30(5), pp. 673–9.

Carrasco, E., Hernández, C., Miralles, A., Huguet, P.,

Farrés, J. and Simó, R. (2007) ‘Lower somatostatin

expression is an early event in diabetic retinopathy and

is associated with retinal neurodegeneration.’, Diabetes

care, 30(11), pp. 2902–8.

Carrasco, E., Hernández, C., de Torres, I., Farrés, J. and

Simó, R. (2008) ‘Lowered cortistatin expression is an

early event in the human diabetic retina and is

associated with apoptosis and glial activation.’,

Molecular vision, 14, pp. 1496–502.

Ceklic, L., Maár, N. and Neubauer, A. S. (2008) ‘Optical

coherence tomography fast versus regular macular

thickness mapping in diabetic retinopathy.’,

Ophthalmic research, 40(5), pp. 235–40.

Chhablani, J., Sharma, A., Goud, A., Peguda, H. K., Rao,

H. L., Begum, V. U. and Barteselli, G. (2015)

‘Neurodegeneration in Type 2 Diabetes: Evidence

From Spectral-Domain Optical Coherence

Tomography.’, Investigative ophthalmology & visual

science, 56(11), pp. 6333–8.

Ctori, I. and Huntjens, B. (2015) ‘Repeatability of Foveal

Measurements Using Spectralis Optical Coherence

Tomography Segmentation Software.’, PloS one, 10(6),

p. e0129005.

van Dijk, H. W., Kok, P. H. B., Garvin, M., Sonka, M.,

Devries, J. H., Michels, R. P. J., van Velthoven, M. E.

J., Schlingemann, R. O., Verbraak, F. D. and Abràmoff,

M. D. (2009) ‘Selective loss of inner retinal layer

thickness in type 1 diabetic patients with minimal

diabetic retinopathy.’, Investigative ophthalmology &

visual science, 50(7), pp. 3404–9.

van Dijk, H. W., Verbraak, F. D., Kok, P. H. B., Garvin, M.

K., Sonka, M., Lee, K., Devries, J. H., Michels, R. P. J.,

van Velthoven, M. E. J., Schlingemann, R. O. and

Abràmoff, M. D. (2010) ‘Decreased retinal ganglion

cell layer thickness in patients with type 1 diabetes.’,

Investigative ophthalmology & visual science, 51(7),

pp. 3660–5.

van Dijk, H. W., Verbraak, F. D., Kok, P. H. B., Stehouwer,

M., Garvin, M. K., Sonka, M., Hans Devries, J.,

Schlingemann, R. O. and Abràmoff, M. D. (2012)

‘Early neurodegeneration in the retina of type 2 diabetic

patients’, Investigative Ophthalmology and Visual

Science, 53(6), pp. 2715–2719.

van Dijk, H. W., Verbraak, F. D., Stehouwer, M., Kok, P.

H. B., Garvin, M. K., Sonka, M., DeVries, J. H.,

Schlingemann, R. O. and Abràmoff, M. D. (2011)

‘Association of visual function and ganglion cell layer

thickness in patients with diabetes mellitus type 1 and

no or minimal diabetic retinopathy.’, Vision research,

51(2), pp. 224–8.

Du, Y., Veenstra, A., Palczewski, K. and Kern, T. S. (2013)

‘Photoreceptor cells are major contributors to diabetes-

induced oxidative stress and local inflammation in the

retina.’, Proceedings of the National Academy of

Sciences of the United States of America, 110(41), pp.

16586–91.

Dutra Medeiros, M., Mesquita, E., Gardete-Correia, L.,

Moita, J., Genro, V., Papoila, A. L., Amaral-Turkman,

A. and Raposo, J. F. (2015) ‘First incidence and

progression study for diabetic retinopathy in Portugal,

the RETINODIAB study: Evaluation of the screening

program for Lisbon region’, Ophthalmology. Elsevier

Inc, 122(12), pp. 2473–2481.

Dutra Medeiros, M., Mesquita, E., Papoila, a. L., Genro, V.

and Raposo, J. F. (2015) ‘First diabetic retinopathy

prevalence study in Portugal: RETINODIAB Study--

Evaluation of the screening programme for Lisbon and

Tagus Valley region’, British Journal of

Ophthalmology, 99(10), pp.1328-33.

Ferreira, J., Vicente, A., Proença, R., Dias-Santos, A.,

Santos, B., Cunha, J. P. and Abegão Pinto, L. (2015)

‘Choroidal Thickness in Diabetic Patients without

Diabetic Retinopathy’, Acta Ophthalmologica,

93(S255). doi:10.1111/j.1755-3768.2015.0488.

Fischer, M. D., Huber, G., Beck, S. C., Tanimoto, N.,

Muehlfriedel, R., Fahl, E., Grimm, C., Wenzel, A.,

Remé, C. E., van de Pavert, S. A., Wijnholds, J., Pacal,

M., Bremner, R. and Seeliger, M. W. (2009)

‘Noninvasive, in vivo assessment of mouse retinal

structure using optical coherence tomography.’, PloS

one, 4(10), p. e7507.

Garcia-Ramírez, M., Hernández, C., Villarroel, M., Canals,

F., Alonso, M. A., Fortuny, R., Masmiquel, L., Navarro,

A., García-Arumí, J. and Simó, R. (2009)

‘Interphotoreceptor retinoid-binding protein (IRBP) is

downregulated at early stages of diabetic retinopathy.’,

Diabetologia, 52(12), pp. 2633–41.

De Gooyer, T. E., Stevenson, K. A., Humphries, P.,

Simpson, D. A. C., Gardiner, T. A. and Stitt, A. W.

(2006) ‘Retinopathy is reduced during experimental

diabetes in a mouse model of outer retinal

degeneration’, Investigative Ophthalmology and Visual

Science, 47(12), pp. 5561–5568.

‘Grading diabetic retinopathy from stereoscopic color

fundus photographs--an extension of the modified

Airlie House classification. ETDRS report number 10.

Early Treatment Diabetic Retinopathy Study Research

Group.’ (1991) Ophthalmology, 98 (5 Suppl), pp. 786–

806.

International Diabetes Federation (IDF) (2015) ‘IDF

Diabetes Atlas 7th edition’, idf.org.

Kern, T. S. and Berkowitz, B. A. (2015) ‘Photoreceptors in

diabetic retinopathy.’, Journal of diabetes

investigation, 6(4), pp. 371–80.

Klein, B. E. K. (2007) ‘Overview of epidemiologic studies

of diabetic retinopathy.’, Ophthalmic epidemiology,

14(4), pp. 179–183.

Nagaoka, T., Kitaya, N., Sugawara, R., Yokota, H., Mori,

F., Hikichi, T., Fujio, N. and Yoshida, a (2004)

‘Alteration of choroidal circulation in the foveal region

in patients with type 2 diabetes.’, The British journal of

ophthalmology, 88(8), pp. 1060–1063.

Simó, R. and Hernández, C. (2014) ‘Neurodegeneration in

the diabetic eye: New insights and therapeutic

perspectives’, Trends in Endocrinology and

Metabolism, 25(1), pp. 23–33.

Tavares Ferreira J, Alves M, Dias-Santos A, Costa L,

Santos BO, Cunha JP, Papoila AL, Abegão Pinto L.

(2016) ‘Retinal Neurodegeneration in Diabetic Patients

without Diabetic Retinopathy’, Investigative

Ophthalmology and Visual Science, 57(14), pp. 6455-

6460.

Vujosevic, S. and Midena, E. (2013) ‘Retinal layers

changes in human preclinical and early clinical diabetic

retinopathy support early retinal neuronal and müller

cells alterations’, Journal of Diabetes Research, 2013,

pp. 905058.