Diabetic Retinopathy Classification: ETDRS 7-fields vs Widefield Imaging (ClarusDR)

Diabetic Retinopathy (DR) remains a significant cause of blindness in working-age populations across the world. The gold-standard method for assessing its severity is the modified Airlie House classification developed for the Early Treatment Diabetic Retinopathy Study (ETDRS), still used by clinical trials for developing of DR therapeutic and management guidelines.

ETDRS Diabetic Retinopathy Severity Score (DRSS) is based on the identification of DR lesions on colour fundus photography's (CFP) obtained in different locations of the retina. The photography acquisition protocol consists in acquiring seven stereoscopic pairs of overlapping 30° fields images of the ocular fundus to map out the macula and mid-peripheral retina. Photographers and fundus camera systems usually need training and certification by external reading centres to guarantee proper images quality and fields definition in a process that can be challenging with a considerable learning curve. Patient collaboration to follow a fixation point in different locations and withstand intense flashes of light, a good dilation of eye pupil and the difficulty to obtain well focused images in peripherical gaze positions are some of the main difficulties for obtaining gradable images using this 7-fields acquisition protocol.

On the other side, the grading process of ETDRS 7-fields images can be extremely labour-intensive and strongly dependent on the quality of the images, presence of artifacts and the definition of the peripherical fields, requiring well-trained people to identify and recognise features that can be very subtle or easily get unnoticed.

Moreover, the retinal area documented with the ETDRS 7-fields protocol represents approximately only 35% of the retina surface. Substantial diabetic retinal pathology can exist in the retinal periphery located outside this area which is being emphasized by advanced retinal imaging technology. Predominantly Peripheral Lesions (PPL) like the presence of venous beading, new-vessels, haemorrhages and microaneurysms in the extreme periphery have been correlated with peripheral non-perfusion, neurodegenerative changes and consequent increase of DR progression.

Recent instrumentation like OptosTM (Optos, Dunfermline, UK) or Clarus 500TM (Carl Zeiss Meditech Inc., Dublin, USA) allow wide-field acquisitions that document up to 90% of the retina surface in just one or two images decreasing patients tiredness and discomfort and overcoming most of the quality and fields definition issues described above.

OptosTM equipment can acquire almost 200º of the retina in just one picture without the need for pupil dilation and using ultra-widefield scanning laser ophthalmoscopy (SLO) technology. Its final image is based on the superimposition of two images acquired with 2 different laser wavelengths: a green and a red wavelength, giving a semirealistic colour image that despite its high contrast and sharpness, gives the retina a greenish and unreal aspect. Also, its 200º field amplitude is usually disturbed by artifacts caused by the presence of eyelashes or eyelids, that mainly obscure the peripherical area of the retina.

On the other hand, Clarus 500 TM equipment uses an imaging technique called Broad Line Fundus Imaging that is a hybrid of confocal SLO (cSLO) and traditional fundus photography. This technology provides higher resolution images with more accurate coloration of the fundus. A single image capture with this system obtains 133-degrees of view but with the acquisition of just two pictures (a temporal and nasal image of the retina), a 200-degree of view can be achieved.

Several studies have suggested moderate to substantial agreement between Optos ultra-wide field (UWF) and ETDRS 7-fields imaging and have shown that DR occurs in areas peripheral to the ETDRS fields in up to 40% of eyes, which may imply a more severe level of DR in 9% to 15% of eyes. However, there is only limited data regarding the validity of DR assessment using Clarus 500TM imagine technique obtained in cross-sectional studies with few patients.

On this basis, the aim of this work is to identify and compare the ETDRS severity level of diabetic patients using 3 different images modalities: the 30º ETDRS 7-fields colour fundus photography's protocol, the 2 wide-field images technique with Clarus 500 TM and the 1 wide-field image of OptosTM, in a prospective, longitudinal, and randomized study. The ETDRS 7-fields area will be superimposed on both wide-field equipment's images so equivalent retinal areas will be analysed. Our goal is to demonstrate that ETDRS severity level can be accurately evaluated using only two Clarus 500TM high quality images with wider amplitude, decreasing the effort and collaboration skills that are required in older techniques, while improving the quality and identification capability of key disease features with less artifacts than other wide-field systems. Additionally, peripheral retina outside the ETDRS 7-fields area will also be analysed to assess the presence of PPL and evaluate its relevance and association with disease severity level and progression.

Finally, the investigators expect to evaluate patients' opinion about their experience in each imaging modality, evaluating their discomfort and satisfaction degree while submitted to each procedure.

Inclusion criteria

• DM type 2 according to 1985 World Health Organization (WHO) criteria
• Age between 35 and 80 years
• BCVA ≥ 75 letters (20 /32)
• Refraction with a spherical equivalent less than 5 Diopters
• NPDR levels 35 and 43-47 (based on the ETDRS criteria - 7 fields CFP)

Exclusion criteria

• Cataract or other eye disease that may interfere with fundus examinations
• Glaucoma
• Other retinal vascular disease than DR
• Refractive errors > to + or - 5 Diopters
• Pupil diameter of less than 5 mm
• Poor quality images due to artifacts, movements, or media opacities