CPD Modules Available

Print this page

Diabetic Macular Oedema: Detection and Treatment, To Laser or Inject.

2 CPD in Australia | 1CD in New Zealand | 1 February 2018

By Associate Professor Gerald Liew

It is an exciting time to be involved in the diagnosis and management of diabetic macular oedema (DME). Major advances in imaging technology have revolutionised the detection of DME, and new anti-Vascular Endothelial Agents (VEGF) have revolutionised its treatment. This article will summarise how DME is now diagnosed, classified and treated, and illustrate these with some patient examples.


1. Diagnose diabetic macular oedema on colour photographs and clinical examination.
2. Diagnose diabetic macular oedema on OCT
3. Classify diabetic macular oedema on OCT
4. Understand that treatment options include focal, grid laser and anti-VEGF injections
5. Understand how classification of diabetic macular oedema is related to choice of treatment
6. Understand how diabetic macular oedema responds to treatment with focal laser
7.  Understand how diabetic macular oedema responds to treatment with anti-VEGF agents, and why prolonged treatment is required.


Diabetes is a major health challenge world-wide, and diabetic eye disease is the major complication of diabetes. It is estimated that there are currently 415 million adults living with diabetes globally, of whom one third (145 million) have some form of diabetic eye disease.1 One third of those with diabetic eye disease (45 million) have disease that is severe enough to be vision threatening. Translated into an Australian context, there are an estimated one million Australians living with diagnosed and undiagnosed diabetes, of whom 100,000 have some form of diabetic eye disease.2

Many of these patients will be asymptomatic, as most forms of diabetic eye disease do not cause symptoms until late in the disease. Nonetheless this offers an avenue for early intervention, hence the importance of regular screening of patients with diabetes. In countries which have introduced nation-wide screening for diabetic eye disease, rates of blindness from diabetes have reduced quite dramatically. In the UK for example, blindness from diabetic retinopathy has reduced by 30 per cent over the last decade. Such results show that much of the vision impairment from diabetic eye disease is preventable.3

With improved understanding of the importance of good glycaemic, blood pressure and lipid control on the development and progression of diabetic complications, rates of diabetic eye disease have reduced in many parts of the world. In developed countries, rates of diabetic retinopathy have reduced by approximately two to three fold over the last three decades.4

This optimism however, is balanced by the rapid increase in prevalence of type 2 diabetes in developing countries, which now account for the majority of the global burden of diabetes and diabetic complications.1

Diabetic eye disease refers generally to the presence of diabetic retinopathy and diabetic macular oedema (DME). Diabetic retinopathy is characterised by the presence of retinal microaneurysms, haemorrhages, venous abnormalities such as loops, cotton wool spots, and later neovascularisation, vitreous haemorrhage and tractional retinal detachment. It is classified as nonproliferative (none, mild, moderate) and proliferative (low risk and high risk).

Proliferative diabetic retinopathy refers to the development of new vessels (neovascularisation), which can occur at the optic disc (neovascularisation at the disc), or elsewhere in the retina, most often along the vascular arcades (neovascularisation elsewhere). NVD and NVE represent friable, easily damaged new blood vessels, which can bleed causing vitreous haemorrhage and preretinal haemorrhages. Repeated cycles of bleeding lead to fibrotic scarring, which can then develop into tractional retinal detachment. Once tractional retinal detachment occurs, the only treatment option is surgical correction and removal of the fibrotic membranes through vitrectomy surgery. If left untreated, the end stage of proliferative diabetic retinopathy is often rubeosis of the anterior segment, whereby new blood vessels invade the drainage angle of the eye, causing angle closure and rapid increase in intraocular pressure. The resulting rubeotic glaucoma is difficult to treat and can lead to hand movements or worse vision that is irreversible. In the early stages, prior to vitreous or pre-retinal haemorhage, diabetic retinopathy can be asymptomatic, hence the need to screen patients at regular intervals.

In contrast, DME refers to the presence of oedema (tissue fluid) in the retina which often takes the form of cystic cavities. This causes the retina, particularly the macula region, to thicken and distorts the architecture of the photoreceptors, bipolar cells and other retinal cells. As macular oedema develops slowly over time, it can remain asymptomatic until severe.

The pathophysiology of DME is poorly understood but involves a breakdown of the blood retinal barrier, with cytokines such as VEGF playing a major role. In early stages, DME has a minimal impact on vision with many patients retaining good vision which slowly deteriorates as the DME worsens. Hard exudates and microaneurysms,together with retinal thickening from the oedema, are the hall marks of DME. Their presence distorts the normal retinal anatomy and may be directly toxic to retinal neural tissue, causing vision loss. Chronic DME can lead to atrophy and irreversible loss of vision, even after the DME has resolved. For all these reasons, early treatment is usually indicated.

Detection and Classification of DME

Traditionally, DME has been detected clinically on slit lamp bimicroscopy  as elevation of the central macula. The condition can also be detected through use of high resolution retinal photographs, ideally stereoscopic to allow appreciation of retinal thickening. Slit lamp bimicroscopy and stereoscopic fundus photographs remained the method of diagnosing DME for many decades until around the mid-2000s.

A clinical/photographic classification of DME was developed, which was referred to as Clinically Significant Macular Oedema (CSME) and referred to the presence of one or more of the following features:
1. Retinal thickening involving the centre of the fovea, or within 500 μm of it;
2. Retinal thickening of least one disc diameter (1500 μm), any part of which
was within one disc diameter of the centre of the fovea; or
3. Yellow, hard exudates within 500 μm of the centre of the fovea with associated
retinal thickening.

Figure 1 shows the presence of CSME which fulfils criteria three. The source of the CSME is the small microaneurysms at the macula which leak, forming the yellow hard exudates. The presence of CSME can be confirmed with fundus fluorescein angiography, which shows leakage of fluorescein dye into surrounding retinal tissue. The presence of the leaking microaneurysms responsible for the CSME can often be located with fluorescein angiography if not visible clinically. CSME was the definition of macular oedema used in the early seminal clinical trials, the Diabetic Retinopathy Study and Early Treatment of Diabetic Retinopathy Study (ETDRS), which laid down many of the clinical trial guidelines for studying retinal diseases still in use today.

Figure 1. Clinically significant macular oedema

In the mid-2000s, Optical Coherence Tomography (OCT) scanning became commercially available and is now considered the gold standard for diagnosing DME. OCT uses infra-red laser interferometry to provide cross-sectional scans of the retina. Earlier generations of OCT scanning used Time domain fourier analysis, which generated images based on the time taken to capture reflected light from the retina. Newer generations use Spectral domain fourier analysis where a spectrometer measures the spectrum of light reflected and generates images based on this. Spectral domain OCT can produce images of 10 μm axial resolution or less, which can be considered an in-vivo histological section of retinal tissue. DME on OCT is generally classified as centre involving DME and non-centre involving DME, examples of which are shown in Figure 2. Centre involving DME refers to macular oedema that involves the centre of the fovea, whereas non-centre involving DME refers to macular oedema that does not involve the centre. This classification system is useful for decision making algorithms and is the main criteria for deciding between laser and anti-VEGF treatment of DME.

Figure 2. OCT images of centre involving DME and non-centre involving DME.

Treatment of DME

Focal and Grid Laser The traditional treatment of DME involved optimising systemic control of blood glucose, blood pressure and lipids, with focal laser to leaking microaneurysms, or grid laser to areas of CSME. Focal and grid laser is effective at reducing mild DME but takes months and visual gains are often modest, of the order of one ETDRS line of vision or less.

Due to the risk of foveal injury and laser scar expansion, focal laser is contraindicated in patients with leaking microaneurysms closer than 500 μm to the centre of the fovea. Several laser machines are in clinical use. The most widely used and validated is the green argon laser, but other laser machines are also used, such as the micropulse and yellow lasers. Laser is performed with a portable module at the slit lamp or at a dedicated laser machine under topical anaesthesia and is considered a very low risk procedure in trained hands.

Example Case One: CSME treated with focal laser.

Figure 3 shows an example of a patient with type 2 diabetes of 15 years’ duration with CSME in her right eye. She was treated with one session of focal laser and her CSME slowly reduced over the next few months, resolving completely by one year. Her best corrected visual acuity improved from 6/9 to 6/6 and the hard exudates resolved completely. Laser is however, destructive to retinal tissue as shown by the laser scars in the temporal macula.

Figure 3. Treatment of CSME.

Anti-VEGF Agents

Anti-VEGF agents have now become first line treatments for DME.5 They were first introduced for treating neovascular macular degeneration in 2006, and have revolutionised treatment of macular diseases with a prominent cystoid macular oedema component such as macular degeneration, DME, and retinal vein occlusion. All work by targeting VEGF, the central mediator in diabetic macular oedema. VEGF over secretion in the retina is believed to lead to DME through breakdown of the blood retinal barrier, leading to cystoid fluid accumulation in the retina. There are five forms of VEGF: VEGF-A; VEGF-B; VEGF-C; VEGF-D; and placenta growth factor (PGF). VEGF-A is the main cytokine implicated in retinal pathology. There are multiple isoforms of VEGF-A which are derived from alternative splicing of mRNA from a single, 8-exon VEGF-A gene. Three anti-VEGF agents are currently in wide clinical use – bevacizumab (Avastin), ranibizumab (Lucentis) and Aflibercept (Eylea). Bevacizumab is an anti-VEGF-A full-length antibody that targets all VEGF-A isoforms; ranibizumab was derived from bevacizumab and is an anti-VEGF-A antibody fragment that also targets all VEGF-A isoforms; aflibercept is an anti-VEGF-A recombinant fusion protein that targets all VEGF-A isoforms as well as VEGF-B and Placental Growth Factor (PIGF).

The first large scale studies to demonstrate efficacy were the RISE and RIDE studies which showed ranibizumab was highly effective at treating DME.6 This was followed by other studies such as RESTORE,7 which demonstrated superiority of ranibizumab over focal and grid laser in treating DME. This was followed by the VIVID and VISTA studies which showed aflibercept was also highly effective at treating DME.8 More recently, the DRCR.net research collaboration in the United States has shown in their Protocol T study9 that over one year of treatment, aflibercept appeared to be superior to ranibizumab and bevacizumab in treating severe DME where baseline visual acuity was worse than 6/12. In patients with baseline visual acuity of 6/12 or better, aflibercept, ranibizumab and bevacizumab were of similar efficacy. In patients with baseline visual acuity worse than 6/12, those treated with aflibercept achieved an average of +18.9 ETDRS letters gain, while those treated with ranibizumab achieved +14.2 ETDRS letters, and bevacizumab +11.7 ETDRS letters. In those with better baseline vision of 6/12 or better, all three agents achieved visual gains of between +7.5 to 8.3 ETDRS letters.9

It should be noted that after two years, these differences disappeared, and ranibizumab caught up with aflibercept, with ETDRS letter gains of +12.3 and +12.8 letters respectively in all participants, while bevacizumab achieved gains of +10.0 letters at two years.10

With regards to the Australian context, bevacizumab is not PBS listed or TGA approved for treatment of DME, and hence is rarely used. Both aflibercept and ranibizumab are PBS listed and TGA approved for treatment of DME. Another factor to bear in mind when interpreting the results of Protocol T above is that the dose of ranibizumab approved for use in Australia is 0.5mg, higher than the dose of 0.3mg licenced in the United States and used in the study.

Most clinical trials including RISE and RIDE, VIVID and VISTA used a monthly injection regime which is difficult to replicate in real world clinical settings. In contrast to treatment of neovascular macular degeneration where three loading doses is the norm, for treatment of DME more than three is desirable. In the RESTORE study7, three loading doses of ranibizumab achieved a +6.1 EDTRS letter gain, while five loading doses of aflibercept in VIVID and VISTA8 achieved a +12.5 ETDRS letter gain, and six loading doses
in Protocol T9 achieved +13.3 ETDRS letter gains. It has thus now become commonplace for six loading doses to initiate treatment of DME for all three anti-VEGF agents.

The International Council of Ophthalmology DME Guidelines 201711 recommend treating patients with
non-centre involving DME with focal laser as a first line treatment, while other patients with centre involving DME are suitable for anti-VEGF therapy. The effectiveness of intravitreal therapy over focal laser does come at a price, as intravitreal delivery of the agent can very rarely be associated with ocular complications including endophthalmitis, leading to blindness.

Further, the burden of treatment with anti-VEGF agents is high, as in order to receive the maximum benefits from the treatment, most patients needed nine to 10 injections in the first year of treatment per eye.9 With both eyes undergoing treatment on alternate visits, this would translate to an injection almost every two weeks. Fortunately, in the second year of treatment, most patients require five to six injections. Another
trial of anti-VEGF agents, Protocol I,12 has followed patients for five years, and suggests that the treatment burden continues to reduce, with many patients requiring only two injections in the third year, one in the fourth year, and none in the fifth year.

As VEGF has multiple systemic functions, including maintaining vascular health, there were concerns that prolonged anti-VEGF administration could increase the risk of cardiovascular events. Although very small amounts are administered intravitreally, some anti-VEGF does escape into the systemic circulation. Recent
evidence, including meta analyses, have found no evidence of increased risk of cardiovascular events or mortality with anti-VEGF agents.13

Example Case 2: Gradual reduction in DME with anti-VEGF

Figure 4 shows a female patient aged 53 years with type 2 diabetes for 15 years, which was poorly controlled. She developed centre-involving DME with reduction in her left eye vision to 6/48 and gross thickening of her macula and central macular thickness (CMT) of 695 μm (normal 250 μm). After her first three ranibizumab injections over three months, there was little change in her DME which remained persistent although there was an improvement in visual acuity to 6/24.

Figure 4. Slow reduction of centre involving DME with anti-VEGF treatment.

Figure 5. Anti-VEGF injections clear hard exudates and improve severity of diabetic retinopathy.

After six injections over six months, her DME reduced to 524 μm and visual acuity improved to 6/18. After 12 injections over 12 months, her DME had reduced to 295 μm, and visual acuity had improved to 6/15. This example shows the importance of continuing with treatment as progress is slow and each injection slowly reduces the degree of DME.

Example Case 3. Anti-VEGF injections are effective at clearing hard exudates and reversing severity of diabetic retinopathy.

Figure 4 shows a male patient aged 57 years with type 2 diabetes for 15 years with numerous hard exudates at the fovea. These are a sign of chronic and severe DME and can cause irreversible vision loss if hard exudate plaques deposit at the fovea. After nine monthly anti-VEGF injections of aflibercept, most of the hard
exudates have cleared and visual acuity has improved from 6/24 to 6/9.

Note as well that anti-VEGF treatment has a disease modifying effect, with improvement in the severity of non-proliferative diabetic retinopathy. At initiation of treatment the patient had severe nonproliferative diabetic
retinopathy in the form of multiple large retinal haemorrhages. After treatment with anti-VEGF agents, the diabetic retinopathy has regressed and is now moderate, and the retinal haemorrhages have resolved.

Anti-VEGF Agents and Diabetic Retinopathy

Anti-VEGF agents have a disease modifying effect and regular use causes regression of diabetic retinopathy in most, but not all patients.14 The effect is strongest in patients with severe levels of diabetic retinopathy. The DRCR.net Protocol T study has recently published results showing 22–38 per cent of patients treated with anti-VEGF agents experience improvements in their degree of severity of non-proliferative diabetic retinopathy.14

Higher percentages of regression were observed for patients with proliferative diabetic retinopathy. The CLARITY15 and DRCR.net trials14 examining use of anti-VEGF to treat proliferative diabetic retinopathy have also shown similar findings; that anti-VEGF is as effective at treating proliferative disease as the traditional treatment of panretinal photocoagulation laser. It should however, be noted that currently no anti-VEGF
agent is licenced anywhere in the world for the purpose of treating diabetic retinopathy, as the area is still an active area of research. Protocols, loading doses and follow up are issues that need to be considered, as well as the cost of long term anti-VEGF treatment. The role of anti-VEGF is thus continuing to evolve and will
likely play a major role in management of proliferative diabetic retinopathy.

Example 5. Macular ischaemia is a possible reason for non-response to treatment. This can be diagnosed with OCT-angiography.

Figure 6 shows a 65 year old male with poorly controlled type 2 diabetes for 30 years who had over a year of treatment with anti-VEGF therapy. The DME has now resolved but his vision remains reduced at 6/36. OCT-angiography is a new technology that can image the in vivo flow of blood in the retinal arterioles, venules and capillaries, and shows an increased foveal avascular zone in the macula responsible for the chronic
reduced vision. This was confirmed on fundus fluorescein angiography. This case illustrates the usefulness of OCT-angiography in imaging retinal vascular disorders such as diabetic retinopathy and reducing the need for invasive investigations such as fluorescein angiography.


Figure 6. Macular ischaemia

Other Treatments for DME

Intravitreal steroid injections are also effective in treating DME. Intravitreal triamcinolone has been used to treat DME and was examined in the Protocol I trial.12 In patients who were pseudophakic, the efficacy of intravitreal triamcinolone was comparable to intravitreal ranibizumab but was associated with increased intraocular pressure as well as the need for glaucoma medications and glaucoma filtering surgery. In phakic patients, intravitreal triamcinolone was also associated with increased risk of developing cataract and
needing cataract surgery. Intravitreal triamcinolone is not PBS listed or TGA approved for managing DME.

Dexamethasone implant is another option for treatment of DME. The MEAD study16 has shown the implant is effective at treating DME although there is an increased risk of raised intraocular pressure and cataract. The implant has recently become PBS listed for treatment of DME in patients who are pseudophakic (or are
scheduled to have cataract surgery), and have a contraindication to, or have failed to respond to, or are otherwise unsuitable for, treatment with an anti-VEGF agent.


DME is a major cause of vision loss, but fortunately new advancements in OCT imaging and anti-VEGF therapy have markedly improved management of the condition. Focal laser retains a role in managing milder, non-centre involving DME while anti-VEGF agents are now first line agents for centre involving DME. Anti-VEGF agents require a loading dose of up to six months and treatment continues for several years. Anti-VEGF agents are effective at clearing hard exudates and regressing diabetic retinopathy. Other options for treating DME include steroid implants which require monitoring for raised intraocular pressure and cataract.



Gerald Liew, MBBS, MMed, PhD, FRANZCO is an Associate Professor and Consultant Ophthalmologist. He practices at Retina Associates Sydney at Liverpool, Chatswood, Norwest; at Westmead Hospital; and Children’s Hospital at Westmead. He is Associate Professor with the Westmead Clinical School, University of Sydney, and South West Clinical School, UNSW, and a member of the Association for Research in Vision and Ophthalmology (ARVO), American Academy of Ophthalmologists (AAO), and European Society of Retinal Specialists (EuRetina).

A medical retina subspecialist, A/Prof. Liew is heavily involved in research including clinical trials of treatments for retinal conditions, methods to improve diagnosis and monitoring of macular diseases such as diabetic retinopathy, and investigating the genetic basis of retinal diseases such as macular degeneration. He has published over 140 peer reviewed research articles in major journals with leading international specialists from Australia, Singapore, USA and UK.

A/Prof. Liew has no financial disclosures relevant to subject matter.   
 This education article was sponsored by Bayer
1. Yau JW, Rogers SL, Kawasaki R, et al. Global prevalence and major risk factors of diabetic retinopathy. Diabetes care 2012; 35(3): 556-64.
2. Diabetes Australia. Diabetes in Australia. 2017. www.diabetesaustralia.com.au.
3. A comparison of the causes of blindness certifications in England and Wales in working age adults (16-64 years),1999-2000 with 2009-2010. Liew G, Michaelides M, Bunce C. BMJ Open. 2014 Feb 12;4(2):e004015. doi: 10.1136/bmjopen-2013-004015.
4. Mini Review: Changes in the Incidence of and Progression to Proliferative and Sight-Threatening Diabetic Retinopathy Over the Last 30 Years. Liew G, Wong VW, Ho IV. Ophthalmic Epidemiol. 2017 Apr;24(2):73-80. doi:10.1080/09286586.2016.1259638. Epub 2017 Jan 19.
5. Liew G, Mitchell P, Ho IV. Management of Diabetic Eye Disease - To Laser or Inject? Diabetes Management Journal 2016; (November).
6. Nguyen QD, Brown DM, Marcus DM, et al. Ranibizumab for diabetic macular edema: results from 2 phase III randomized trials: RISE and RIDE. Ophthalmology 2012; 119(4): 789-801.
7. Mitchell P, Bandello F, Schmidt-Erfurth U, et al. The RESTORE study: ranibizumab monotherapy or combined with laser versus laser monotherapy for diabetic macular edema. Ophthalmology 2011; 118(4): 615-25.
8. Korobelnik JF, Do DV, Schmidt-Erfurth U, et al. Intravitreal aflibercept for diabetic macular edema. Ophthalmology 2014; 121(11): 2247-54.
9. Wells JA, Glassman AR, Ayala AR, et al. Aflibercept, bevacizumab, or ranibizumab for diabetic macular edema. The New England journal of medicine 2015; 372(13):1193-203.
10. Wells JA, Glassman AR, Ayala AR, et al. Aflibercept, Bevacizumab, or Ranibizumab for Diabetic Macular Edema:Two-Year Results from a Comparative Effectiveness Randomized Clinical Trial. Ophthalmology 2016; 123(6):1351-9.
11. Ophthalmology ICo. Guidelines for Management of Diabetic Macular Oedema. wwwicophorg 2017.
12. Bressler SB, Glassman AR, Almukhtar T, et al. Five-Year Outcomes of Ranibizumab With Prompt or Deferred Laser Versus Laser or Triamcinolone Plus Deferred Ranibizumab for Diabetic Macular Edema. American journal of ophthalmology 2016; 164: 57-68.
13. Vascular Safety of Ranibizumab in Patients With Diabetic Macular Edema: A Pooled Analysis of Patient-Level Data From Randomized Clinical Trials. Zarbin MA, Dunger-Baldauf C, Haskova Z, Koovejee P, Mousseau MC, Margaron P, Snow H, Beaumont PE, Staurenghi G, Francom S.
14. Bressler SB, Liu D, Glassman AR, et al. Change in Diabetic Retinopathy Through 2 Years: Secondary Analysis of a Randomized Clinical Trial Comparing Aflibercept, Bevacizumab, and Ranibizumab. JAMA ophthalmology 2017; 135(6): 558-68.
15. Sivaprasad S, Prevost AT, Vasconcelos JC, et al. Clinical efficacy of intravitreal aflibercept versus panretinal photocoagulation for best corrected visual acuity in patients with proliferative diabetic retinopathy at 52 weeks (CLARITY): a multicentre, single-blinded, randomised,
controlled, phase 2b, non-inferiority trial. Lancet (London, England) 2017; 389(10085): 2193-203.