miophthalmology

Print this page

Diabetic Retinopathy and its Complications

Dr. Smita Agarwal | 26 June 2017
Patients with uncontrolled diabetes are at high risk of developing a wide range of ocular complications that can affect almost all structures of the eye.

Diabetes is a major medical problem that is increasing in prevalence throughout the world.1 Diabetic patients often develop ophthalmic complications, such as corneal surface disease, glaucoma, cataract, iris neovascularisation, and neuropathies. The most common and potentially most blinding of these, however, is diabetic retinopathy (DR).2,3 

Diabetic retinopathy remains the leading cause of preventable blindness in working aged people.4 The Eye Diseases Prevalence Research Group defined vision threatening DR as the presence of severe non-proliferative DR (NPDR) or proliferative DR (PDR) and / or diabetic macular oedema (DMO).5 In Australia, approximately 6 per cent of people have been diagnosed with diabetes.6 Of these, 4.2 per cent will develop DMO and 15 per cent will develop DR, with 60 per cent of these experiencing reduced vision.7 The rates of death and hospitalisation resulting from diabetic complications are approximately four-fold higher for the indigenous population and two-fold higher for those of rural or lower socioeconomic background.6

Signs and Symptoms

Early stage DR is generally asymptomatic. At more advanced stages, patients may experience symptoms including floaters, blurred vision, distortion, and progressive visual acuity loss. Signs of DR include the following:

  • Micro-aneurysms (M/A): The earliest clinical sign, these occur due to capillary wall outpouching as a result of pericyte loss, and appear as red dots in the superficial layers of the retina.
  • Dot and blot haemorrhages: These appear similar to M/A if small, however occur in the deeper layers, inner nuclear and outer plexiform layers of the retina.
  • Flame-shaped haemorrhages: Splinter haemorrhages in the superficial layers.
  • Cotton-wool spots: Occlusion of precapillary arterioles cause nerve fibre layer infarcts. These are frequently bordered by M/A and vascular hyper permeability.
  • Venous loops and beading: These are an indication of ischaemia, and are generally located adjacent to areas of non-perfusion. They can be used as a predictor of pre-proliferative changes.
  • Intra-retinal microvascular abnormalities (IRMA): Remodelling of capillary beds without proliferative changes.
  • Macular oedema: Observed as macular thickening, fluid collection or exudation, this is the leading cause of visual impairment and can occur at any stage of NPDR or PDR.
  • Hard exudates: The deposition of solidified lipids and proteins caused by the absorption of fluids following DMO.

Non-proliferative Diabetic Retinopathy

Vascular changes associated with NPDR are limited to the retina and do not extend beyond the internal limiting membrane. Indications include:

  • Mild: the presence of at least one M/A (Figure 1A, B).
  • Moderate: the presence of haemorrhages, M/A and hard exudates (Figure 1B, 2B).
  • Severe (4-2-1 rule): haemorrhages and M/A in four quadrants, with venous beading in at least two quadrants and IRMA in at least one quadrant.

Patients with severe NPDR have a 15 per cent chance of progression to PDR within a year, with very severe NPDR increasing the chances to 45 per cent.8

Proliferative Diabetic Retinopathy

The hallmark of PDR is neovascularisation (Figure 2A), the formation of new blood vessels that extend beyond the internal limiting membrane. Other indications include macular oedema (Figure 2B, C), pre-retinal and vitreous haemorrhages (Figure 2D), fibro-vascular fronds (Figure 2E) and tractional detachments.

Diagnosis

Diagnosis and monitoring of DR involves both imaging and lab studies. Imaging studies such as fluorescein angiography (Figure 3A), fundus photography (Figure 3B) and optical coherence tomography (OCT) (Figure 3C) allow the user to quantify the macular thickness and to monitor the response to treatment.

Recently, OCT-angiography (OCT-A) has provided a non-invasive method of detecting areas of non-perfusion (Figure 4A) and neovascularisation of the retina without the risk associated with venepuncture or dye exposure linked to fluorescein angiography. However, the method is unable to show vessel leakage, for which fluorescein angiography is still the only method available.

In the occurrence of vitreous haemorrhage, B-scan ultrasonography can be used to assess the integrity of the retina. Lab studies to monitor glycated haemoglobin (HbA1c) levels are also important in the long term follow-up care of patients with diabetes and DR.

Etiology

  • Duration of diabetes plays a significant role in the development of DR. Approximately 25–50 per cent of Type 1 diabetics show some signs of DR. This increases to 75–95 per cent after 15 years and to 100 per cent after 30 years. Approximately 23 per cent of Type 2 diabetics have NPDR after 11–13 years and this increases to 41 per cent after 14–16 years, and 60 per cent after 16 years.10
  • Hypertension and hyper-lipidaemia can lead to more extensive retinal vessel leakage and hard exudate formation.
  • Pregnant women without DR have a 10 per cent risk of developing NPDR during pregnancy and of those with pre-existing NPDR, 4 per cent progress to PDR.10

Management of Diabetic Retinopathy

Laser Photocoagulation: Retinal photocoagulation, which was first shown to be therapeutically beneficial in 1954,11 uses a retinal laser to destroy pathological vessels. It can be applied evenly across the non–macular retina (pan-retinal photocoagulation), or at specific vessels (focal photocoagulation). Following its introduction, subsequent trials by the Early Treatment Diabetic Retinopathy Study (ETDRS) group8 and the Diabetic Retinopathy Study Research (TDRSR) group12 confirmed its effectiveness, leading to its extensive use over many decades. However, retinal photocoagulation only prevents further loss of vision rather than restoring vision, and causes side effects including night vision and peripheral vision loss.13

Glucose Control

Historically, DR and DMO have been treated by tighter glycaemic control in the form of insulin sensitisers, systemic insulin administration, or improved diet and exercise. The United Kingdom Prospective Diabetes Study (UKPDS) revealed that the risk of DR was reduced through both improved glycaemic control and blood pressure control in patients with Type 2 diabetes. A 1 per cent reduction in HbA1c reduced the risk of retinopathy by 31 per cent, and a 10 mm Hg reduction in systolic blood pressure reduced photocoagulation or vitreous haemorrhage by 11 per cent.14

More recently, control of the underlying disease through intensive glycaemic control has been recommended to reduce the incidence and severity of diabetic complications, including DR. The Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial set forth a protocol which lowered the targeted HbA1c in patients with Type 2 diabetes to below 6.0 per cent from a standard range of 7.0–7.9 per cent and found that intensive glycaemic control significantly reduced DR progression.15 This data is consistent with previous clinical trials showing reduced progression of DR following intensive glycaemic control in Type 2 diabetics16 and in Type 1 diabetics.17 It is also important to note the importance of controlling hypoglycaemia, with the intensive control group (i.e. target HbA1c of <6.0 per cent) of the ACCORD trial reporting a 22 per cent higher rate of deaths than the standard treatment group (i.e. target HbA1c of 7.0-7.9 per cent).15 This is particularly pertinent in cases of uncontrolled diabetes.

Fenofibrate (LIPIDIL)

The Field and Accord studies,18,19 report that fenofibrate reduces the rate of progression of DR and the need for laser treatment in patients with type 2 diabetes and pre-existing DR. However, patients with no signs of DR or with severe DR showed no beneficial effect.

Therapeutic intervention

DR and DMO manifest as vascular dysfunction through inflammation, pericyte loss and breakdown of the blood retinal barrier, and in PDR through neovascularisation. One of the primary factors in the development of DR and DMO is the pathological release of various growth factors that trigger new blood vessel formation. The most noted of these is vascular endothelial growth factor (VEGF). Anti-VEGF treatments target this vascular dysfunction, preventing neovascularisation and disease progression, leading to significant improvement in visual function.

In a clinical trial that compared the anti-VEGF drugs aflibercept (Eylea), ranibizumab (Lucentis), and bevacizumab (Avastin; off label) for treatment of DMO, aflibercept provided greater visual improvement on average than the other two drugs when visual acuity was 20/50 or worse at the onset. However, when visual acuity prior to the commencement of treatment was between 20/40 and 20/32, all three drugs achieved similar improvement in vision. No major differences in safety were found.20 While anti-VEGF treatments have changed the treatment paradigm of DR and DMO, they are invasive, require frequent administration, and are not effective in certain people. The common side effects associated with anti-VEGF agents include eye pain, vitreous floaters, cataract, conjunctival haemorrhage, and increased intraocular pressure (IOP).21 There is also a risk of endophthalmitis and retinal detachment, thus patients should be monitored closely for blurred vision, eye pain or redness, photophobia, or vision changes.21,22

Figure 5 shows regression of oedema in a patient with DMO following treatments with anti-VEGF agents.

Intravitreal steroids such as triamcinolone acetonide (Triesence) and dexamethasone (Ozurdex) have also shown a beneficial effect in the management of DMO and PDR, owing to their anti-inflammatory and anti-angiogenic effects. In a comparative study of triamcinolone acetonide and bevacizumab in the treatment of persistent diffuse DMO, triamcinolone acetonide showed significantly better results in regards to foveal thickness and visual acuity following treatment.23 Although both treatments led to regression of oedema, the effect was temporary with both treatment groups displaying recurrence of oedema within 24 weeks following cessation of treatment. Another study that compared the outcomes of dexamethasone and bevacizumab treatments showed that although treatment with dexamethasone led to a greater decrease in average central foveal thickness compared to bevacizumab, both led to a similar increase in visual acuity.24 Although fewer treatments were necessary for the dexamethasone treatment group, some patients lost visual acuity due to cataract formation. These studies suggest that steroid treatments are more effective against DMO compared to anti-VEGF treatments, possibly due to their additional anti-inflammatory properties. However, steroid use is also associated with an increased incidence of elevated IOP, cataracts and infection.25

Vitrectomy

Vitrectomy can be used for PDR in cases of long standing vitreous haemorrhage, tractional retinal detachment, and combined tractional and rhegmatogenous retinal detachment.

Patient education

Optometrists and local health practitioners play a significant role in educating patients. Good control of risk factors including blood glucose, hypertension, renal disease and hyperlipidemia are crucial in delaying the onset and progression of DR. Smoking may additionally exacerbate the condition by further compromising oxygen delivery to the retina. Regular eye checks are important to detect for any early changes and intervention needs to be started early to prevent irreversible damage. Any new visual symptoms warrant urgent review.

Prognosis

Favourable prognostic factors include well defined leakage, good peri-foveal perfusion and recent onset circinate exudates. Poor prognostic factors include diffuse oedema, multiple leaks, lipid deposition in the fovea, macular ischaemia, cystoid macular oedema (CMO), preoperative vision of less than 20/200 and hypertension.

Other Diabetic Complications

Diabetic patients are at high risk of developing xanthelesma, conjunctival infections, contact lens related microbial keratitis, miotic pupils with delayed response to mydriatics, uveitis, retinal vascular occlusions, ocular ischemic syndrome, glaucoma microvascular cranial nerve palsies and anterior ischaemic optic neuropathy.

Diabetes and Cataract

Diabetic patients are at increased risk of developing cataracts, typically cortical and posterior sub-capsular cataracts,26,27 The lens rapidly becomes cataractous with dense white anterior and posterior sub-capsular cortical cataract resembling a snowstorm i.e. ‘snowflake cataracts (Figure 6). Age contributes an additional risk, with a three to four-fold higher risk in patients with diabetes who are older than 65 years compared to patients without diabetes.28 Although rapid glycaemic control has been reported to slow DR progression, it may also lead to irreversible opacification of the lens.29

Unfortunately, cataract surgery may itself cause pseudo-phakic CMO or exacerbate DR as a result of the impaired blood-aqueous barrier in diabetic patients,31 although it has significantly reduced with the advent of modern small incision phacoemulsification cataract surgery. In a recent study identifying the incidence of oedema following cataract surgery, it was found that diabetic patients were at increased risk, even in the absence of preoperative DR.32 The risk was further increased in patients with DR, with incidences increasing proportional to severity.

Prophylaxis and management of pseudo-phakic CMO remain challenging owing to a lack of diagnostic criteria and prospective randomised clinical trials.33 Non-invasive OCT and OCT-A are relatively new diagnostic tools, which provide high resolution imaging of retina to detect any subtle changes in foveal thickness and hence in better predicting the prognosis and further treatment following cataract surgery. Macular micro-perimetry has also been used to objectively quantify macular sensitivity as an effective and accurate assessment of retinal function in patients with DMO and DR.

There has traditionally been a more cautious approach towards cataract surgery for diabetic patients, with many delaying the surgery due to the uncertainty of postoperative results. However, it can be quite difficult to monitor retinal changes with the rapidly progressive cortical cataract. It is generally therefore recommended to perform cataract surgery for these patients, with postoperative reassessment and treatment of DR and DMO if necessary. For early cataracts, it is vital that glucose levels and DR are stable before surgical intervention. It is also recommended that PDR and DMO are adequately treated preoperatively via photocoagulation and/or anti-VEGF treatment as the conditions can exacerbate following surgery.

For such cases, new treatment regimens of steroidal or non-steroidal anti-inflammatory drugs, at or around the time of surgery, can reduce the risk of oedema following the surgery. The intraoperative insertion of intravitreal dexamethasone implants have been shown to reduce central macular thickness following cataract surgery in patients with DMO, with patients reporting improved visual acuity postoperatively.34 However, oedema generally recurred after an average of 21 weeks postoperatively, requiring further treatment.

Daily administration of nepafenac for three months following surgery also significantly decreased the incidence of macular oedema in diabetic patients, with an increase in visual acuity compared to patients that did not receive the treatment.35 However, another study reported that topical NSAIDs were only effective at preventing postoperative macular oedema in healthy and diabetic patients with no pre-existing DR, with no benefit observed for patients with DR.36 It has therefore been suggested that topical NSAIDs such as nepafenac be used for all diabetic patients following cataract surgery, with a combination of intravitreal anti-VEGF drugs and steroids such as dexamethasone being reserved for patients with pre-existing DMO.37

Post-operative Complications and Outcomes

Despite these advancements, diabetic patients are at additional risk of postoperative complications including neovascularisation of the iris, epithelial defects, posterior synechiae, pupillary block, pigmented precipitates on the lens implant and iritis following cataract surgery.38 Tear film abnormalities are common in diabetic patients with reduced tear breakup times, reduced corneal sensitivity and reduced goblet cell density in conjunctiva. Stopping topical medications and using lubricating eye drops may help avoid such complications,39 while treatments such as the topical application of insulin may also aid corneal healing in diabetic patients.40

Patients with mild NPDR, generally have a good prognosis following surgery,41 but still require frequent monitoring and control of risk factors. One study suggests that micro-perimetry could be considered as an additional functional test, as it is superior to testing CDVA in patients with diabetic macular oedema.42 It should be noted that in a separate study following patients with advanced DR postoperatively, patient quality of life did not improve following surgery despite improvements in visual acuity.43 Good communication between the patient and the health care practitioner is therefore vital in order to manage patient expectations.

In conclusion, it is clear that patients with uncontrolled diabetes are at high risk of developing a wide range of ocular complications. If diabetes is not adequately controlled, it can affect almost all the structures of the eye including the cornea, uveal tract, lens, vitreous, orbital tissues, cranial nerves, optic nerve and can also pose challenging issues with cataract surgery. 

The author thanks Dr. Erin Thorwell for her assistance in preparing this manuscript.

Dr Smita Agarwal MBBS, MS, Grad Dip Med-Ref Surgery (USyd), FRANZCO, is a comprehensive ophthalmologist with special interests in refractive cataract surgery, glaucoma, retinal and anterior segment eye diseases. Dr. Agarwal is the Head of Ophthalmology Department at Wollongong and Shellharbour Public Hospitals, a senior lecturer at University of Sydney and University of Wollongong and a Visiting Medical Officer at a number of private hospitals. She is actively involved in the teaching of optometrists, GPs and medical students. She also performs laser eye surgery at Vision Eye Institute, Hurstville.

Dr. Agarwal has authored several research papers and case reports in Australia and overseas, including Journal of Cataract and Refractive Surgery, Journal of Ophthalmic and Plastic Surgery and Medical Journal of Australia.

References
1 Federman JL, Gouras P, Schubert H. Systemic diseases. In: Podos SM, Yanoff M eds, Textbook of Ophthalmology: Retina and Vitreous Maryland Heights, Mosby, 1994;7-24.
2 Cai X, McGinnis JF. Diabetic retinopathy: animal models, therapies and perspectives. J Diabetes Res 2016; 2016:9 pages.
3 Aiello LM, Cavallerano JD, Aiello LP et al. Diabetic Retinopathy. In: Guyer DR, Yannuzzi LA, Chang S eds, Retina, Vitreous, Macular. Philadelphia, Saunders, 1999;316-344.
4 Cheung N, Mitchell P, Wong TY. Diabetic Retinopathy. Lancet 2010; 376:124-136.
5 TEDPR Group. The prevalence of diabetic retinopathy among adults in the United States. Arch Ophthalmol 2004; 122:552-563.
6 AIHW. Diabetes. 2017. Available at: www.aihw.gov.au/diabetes/ (accessed 05/05/2017).
7 Diabetes Australia. Diabetes in Australia. 2015. Available at: www.diabetesaustralia.com.au/diabetes-in-australia (accessed 05/05/2017).
8 ETDRS Group. Early photocoagulation for diabetic retinopathy: EDTRS report number 9. Ophthalmology 1991; 98:766-785.
9 Medscape. Diabetic Retinopathy. 2017. Available at: emedicine.medscape.com/article/1225122-overview#a1 (accessed 05/05/2017).
10 Arevalo JF, Lasave AF, Zeballos DG et al. Diabetic Retinopathy. In: Arevelo JF ed, Retinal and Choroidal Manifestations of Selected Systemic Diseases. New York, Springer Science and Business Media, 2012;387-416.
11 Meyer-Schwickerath G. Light coagulation: a method for treatment and prevention of retinal detachment. Albert Von Graefes Arch Ophthalmol 1954; 156:2-34.
12 TDRSR Group. Photocoagulation treatment of prolifertive diabetic retinopathy. Clinical application of Diabetic Retinopathy study (DRS) findings, DRS report number 8. Ophthalmology 1981; 88:583-600.
13 Fong DS, Girach A, Boney A. Visual side effects of successful scatter laser photocoagulation surgery for proliferative diabetic retinopathy: A literature review. Retina 2007; 27:816-824.
14 Kohner EM. Microvascular disease: what does the UKPDS tell us about diabetic retinopathy? Diabet Med 2008; 25:20-24.
15 Gerstein HC, Miller ME, Byington RP et al. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med 2008; 358:2545-2559.
16 UKPDS Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes. Lancet 1998; 352:837-853.
17 DCCT Group. The effect of intensive treatment of diabetes on the development and progression of long term complications in insulin-dependent diabeted mellitus. N Engl J Med 1993; 329:978-986.
18 Keech AC, Mitchell P, Summanen PA et al. Effect of Fenofibrate on the need for laser treatment for diabetic retinopathy (Field Study): a randomised controlled trial. Lancet 2007; 370:1687-1697.
19 Chew EY, Davis MD, Danis RP et al. The effects of medical management on the progression of diabetic retinopathy in persons with type 2 diabetes: The Action to Control Cardiovascular Risk in Diabetes (ACCORD) eye study. Ophthalmology 2014; 121:2443-2451.
20 Wells JA, Glassman AR, Ayala AR et al. Aflibercept, bevacizumab, or ranibizumab for diabetic macular oedema. N Engl J Med 2015; 372:1193-1203.
21 Sharma YR, Tripathy K, Venkatesh P et al. Aflibercept: how does it compare with other anti-VEGF drugs? Austin J Clin Ophthalmol 2014; 1:1016.
22 Artunay O, Yuzbasioglu E, Rasier R et al. Incidence and management of acute endoophthalmitis after intravitreal bevacizumab (Avastin) injection. Eye 2009; 23:2187-2193.
23 Shimura M, Nakazawa T, Yasuda K et al. Comparative therapy evaluation of intravitreal bevacizumab and triamcinolone acetamide on persistent diffuse diabetic macular oedema. Am J Ophthalmol 2008; 145:854-861.
24 Gillies MC, Lim LL, Campain A et al. A randomised clinical trial of intravitreal bevacizumab versus intravitreal dexamethasone for diabetic macular oedema. Ophthalmology 2014; 121:2473-2481.
25 Sayin N, Kara N, Pekel G. Ocular complications of diabetes mellitus. World J diabetes 2015; 6:92-108.
26 Delcourt C, Cristol JP, Tessier F et al. Risk factors for cortical, nuclear, and posterior subcapsular cataracts; the POLA study. Am J Epidemiol 2000; 151:497-504.
27 Hennis A, Wu SY, Nemesure B et al. Risk factors for incident cortical and posterior subcapsular lens opacities in the Barbados Eye Studies. Arch Ophthalmol 2004; 122:525-530.
28 Ederer F, Hiller R, Taylor HR. Senile lens changes and diabetes in two populations studies. Am J Ophthalmol 1981:381-395.
29 Kato S, Oshika T, Numaga T et al. Influence of rapid glycemic control on lens opacity in patients with diabetes mellitus. Am J Ophthalmol 2000; 130:354-355.
30 National Eye Institute. Cataract. 2017. Available at: nei.nih.gov/photo/cataract (accessed 10/05/2017).
31 Ferguson VMG, Spalton DJ. Continued breakdown of the blood aqueous barrier following cataract surgery. Br J Ophthalmol 1992; 76:453-456.
32 Chu CJ, Johnston RL, Buscombe C et al. Risk factors and incidence of macular edema after cataract surgery: a data base study of 81984 eyes. Ophthalmology 2016; 123:316-323.
33 Yonekawa Y, Kim IK. Pseudophakic cystoid macular edema. Curr Opin Ophthalmol 2012; 23:26-32.
34 Sze AM, Luk FO, Yip TP et al. Use of intravitreal dexamthasone implant in patients with cataract and macular edema undergoing phacoemulsification. Eur J Ophthalmol 2015; 25:168-172.
35 Singh RP, Lehmann R, Martel J et al. Nepafenac 0.3% after cataract surgery in patients with diabetic retinopathy: results of two randomised phase 3 trials. Ophthalmology 2017;
36 Modjtahedi BS, Paschal JF, Batech M et al. Perioperative topical nonsteroidal anti-inflammatory drugs for macular edema prophylaxis following cataract surgery. Am J Ophthalmol 2017; 176:174-182.
37 Boscia F, Giancipoli E, D'Amico Ricci G et al. Management of macular oedema in diabetic patients undergoing cataract surgery. Curr Opin Ophthalmol 2017; 28:23-28.
38 Krupsky S, Zalish M, Oliver M et al. Anterior segment complications in diabetic patients following extracapsular cataract extraction and posterior chamber intraocular lens implantation. Ophthalmic Surg 1991; 22:526-530.
39 Javadi M-A, Zarei-Ghanavati S. Cataracts in diabetic patients: a review article. J Ophthalmic Vis Res 2008; 3:52-65.
40 Bastion ML, Ling KP. Topical insulin for healing of diabetic epithelial defects?: a retrospective review of corneal debridement during vitreoretinal surgery in Malaysian patients. Med J Malaysia 2013; 68:208-216.
41 Krepler K, Biowski R, Schrey S et al. Cataract surgery in patients with diabetic retinopathy: visual outcome, progression of diabetic retinopathy, and incidence of diabetic macular edema. Graefes Arch Clin Exp Ophthalmol 2002; 240:735-738.
42 Edington M, Sachdev A, Morjaria R et al. Structural-functional correlation in patients with diabetic macular edema. Retina 2017; 37:881-885.
43 Mozaffarieh M, Heinzl H, Sacu S et al. Clinical outcomes of phacoemulsification cataract surgery in diabetes patients: visual function (VF-14), visual acuity and patient satisfaction. Acta Ophthalmol 2005; 83:176-183.
  • Figure 2: (A) Fluorescein angiogram demonstrating foveal dye leakage indicative of macular oedema and fundus photographs demonstrating (B) retinal exudates within the fovea, (C) retinal neovascularisation, (D) boat-shaped  pre-retinal haemorrhage and
  • Figure 3: (A) Fundus photograph, (B) fluorescein angiogram showing dye leakage and (C) OCT of patient with DMO.
  • Figure 4: OCT-A image of (A) vessels within an avascular zone (indicated by a dashed red circle) and (B) the macula of patient with DMO.
  • Figure 5: OCT image of the macula of a patient with DMO before treatment and following anti-VEGF treatment.
  • Figure 6: Patient with cortical cataract. Image obtained from National Eye Institute.
  • Figure 1: Fundus photograph of a retina with early background DR showing (A) multiple microaneurysms and (B) blot haemorrhages (long arrow), microaneurysms (short arrow), and hard exudates (arrowhead). Images were obtained and modified from Medscape.

' Patients with severe NPDR have a 15 per cent chance of progression to PDR within a year '