Leonard H. Ginsburg, MD, CDE

Patients with diabetes develop diabetic retinopathy (DR) 97% over time. At the time of initial diagnosis, about 30% of patients have manifestation of DR and 3% diagnosed ≥30 years of age will have clinically significant macular edema or high-risk features. ^1–3 Within 20 years of diagnosis, diabetic macular edema (DME) effects almost 30% of patients, resulting in DME being the leading cause of vision loss in working adults. Left untreated, 29% of eyes with DME will experience moderate vision loss 3 years after onset.

Properly staging DR is critical in identifying the risk of progression so that treatment can be instituted at the proper time and provide guidance to the primary care physician.

Anterior Features of Critical Importance

While retinopathy is the major cause for vision loss, cataracts occur 5 times more frequently. However, the conditions often missed include microrubeosis (clusters of blood vessels with grape-like appearances) found on the iris margin and often not identified after dilation to obvious rubeosis on the iris or in the angle. In proliferative diabetic retinopathy (PDR), angle vessels can occur and are more common in patients with carotid disease. Gonioscopic evaluation is critical and it is incumbent upon eye doctors to check the angle, especially in a hyperope. The eye doctor is especially helpful in ruling out narrow angles before the patient is sent to a retina specialist, who often sees patients after dilation and may miss the rubeosis or narrow angles.

Classification and Staging Diabetic Retinopathy and Its Predictive Value

Diabetic retinopathy has been traditionally classified as background, preproliferative, or proliferative with or without high-risk characteristics. However, the Early Treatment Diabetic Retinopathy Study (ETDRS) established a more specific classification system based on the relative severity of the various diabetic retinal lesions, defining DR as either nonproliferative (NPDR) or proliferative (PDR). Nonproliferative diabetic retinopathy is further distinguished as either mild, moderate, or severe whereas PDR is distinquished as either early, high-risk, or advanced (Table 1).

Properly staging DR is critical in determining risk of progression to sight-threatening vision loss and serves as a guide for determining frequency of exams and the need for diagnostic tests and treatment. Additionally, discovering PDR and notifying the patient and primary care physician of this fact can have predictive value with regard to kidney disease (95% of patients with PDR have renal disease).

Hemorrhages and microaneurysms (H/Ma), intraretinal microvascular abnormalities (IRMA), and venous beading were the most significant lesions for predicting progression to PDR within 12 months. Nerve fiber layer infarcts of increasing severity were only half as important a predictor as IRMA, H/Ma, or venous beading. The severity of hard exudates and macular edema had little predictive value for developing PDR and can be present at any level of retinopathy, although DME did have a significant effect on central vision.

Importantly, underestimating the severity of retinopathy by only 1 level underestimates the risk of progression to PDR within 12 months by 50% or more. The ETDRS demonstrated that without treatment, 50% of eyes with PDR lose their sight within 5 years.

Clinical Pearls in Diagnosing and Staging Nonproliferative Diabetic Retinopathy

Intraretinal Hemorrhages/Microaneurysms

The severity of H/Ma is a strong predictor for progression of retinopathy. Their number and distribution is critical in moderate, severe, and very severe retinopathy. Striking asymmetry in 1 quadrant could suggest a process, such as branch retinal vein occlusion (BRVO), or if unilateral, carotid disease.

The appearance depends on location: in the retinal nerve fiber layer (RNFL), the appearance is flame-shaped; in deeper layers confined by retinal cells the appearance is sharply demarcated, dot-blot hemorrhages. Hemorrhages on the optic nerve are not typical for retinopathy and if present, raise suspicion for neovascularization (NVD) or a comorbid condition affecting the optic nerve head. Microaneurysms light up on a fluorescein angiogram (FA) as compared to hemorrhages, but are grouped together when staging retinopathy.

Nerve Fiber Layer Infarcts

The result of infarction in the nerve fiber layer appears clinically as grayish-white spots with ill-defined feathery edges that suggests a 95% association with systemic disease. If the patient does not have diabetes, the eye doctor should always recommend a systemic workup.

Major Vessel Changes

In advanced retinopathy, arterioles may appear thin or white. Venules may appear dilated and tortuous and venous loops may be present.

Venous Beading

These localized irregularities in venous caliber are the most powerful predictors for the development of PDR.

Intraretinal Microvascular Abnormalities

Tortuous vascular segments resemble new vessels distinguished by intraretinal location, situated at borders of areas of capillary non-perfusion. Intraretinal microvascular abnormalities (IRMA) are strong predictors for the progression of diabetic retinopathy.

Clinical Pearls in Diagnosing and Staging Proliferative Diabetic Retinopathy

New Vessels

Unlike IRMA, neovascularization rises on the retinal surface and may be pulled into the vitreous cavity. If new vessels appear on or within one disc diameter of the disc margin, they are known as NVD; any other location, it is referred to as neovascularization elsewhere (NVE).

Fibrovascular Proliferation

Growth of new vessels is accompanied by proliferation of fibrous tissue between the posterior vitreous gel and the internal limiting membrane.

Vitreoretinal Interface (Traction) and Posterior Hyaloid Face

Whether the posterior hyaloid face (PHF) is attached or detached often determines progression and extent of symptoms. New vessels can break as the vitreous detaches or tractional retinal detachments can occur as fibrous proliferation adheres to PHF. A preretinal hemorrhage is between attached and partially detached PHF when the VH does not break through. When the PHF detaches or VH breaks through, the PHF symptoms can be positional.

Clinical Pearls in Diagnosing and Staging Macular Edema

Hard Exudates

The location of hard exudates in relation to the fovea is important and has an impact on vision, but is irrelevant when staging retinopathy, as exudates have no predictive value for progression.

Subretinal Fibrosis

Subretinal fibrosis is an uncommon feature associated with severe exudates or disruption of the retinal pigment epithelium (RPE) by an over-aggressive laser. In the ETDRS, very severe hard exudates were present in the macula (74%) of these eyes prior to the appearance of subretinal fibrosis. Very severe, hard exudates in the ETDRS resulted in subretinal fibrosis (31%), while clinically significant macular edema (CSME) and lesser levels of hard exudation, such as fibrosis developed in only 0.05%.

Classification of Diabetic Maculopathy

Nearly half of those developing DME will lose more than 2 lines of vision within 2 years without treatment. While complications of PDR, vitreous hemorrhage, and tractional detachment may cause the most severe loss of vision in retinopathy, maculopathy is the main cause of visual impairment in patients with type 2 diabetes.

In 1987, prior to the optical coherence tomography (OCT) era, the ETDRS defined strictly through retinal exam or photography, CSME. Laser treatment for CSME was shown to reduce moderate vision loss by 50%. CSME is defined as either retinal thickening at, or within, 500 mm of the center of the macula; or hard exudates at, or within, 500 mm of the center of the macula, if associated with thickening of the adjacent retina; or zone(s) of retinal thickening ≥1 disc area in size, at least part of which is within 1 disc diameter of the center of the macula.

CSME does not define axial thickening, whereas OCT measures this with precision. Axial thickening more than any other measured variable, has been shown to correlate with visual acuity in DME.⁶

CSME does not define subtypes of DME. Diabetic macular edema can be grouped into 4 categories (Table 2).

Diffuse diabetic macular edema (DDME) is a different disease process than focal edema and usually proves more difficult to manage. The majority of the macula is often involved; cystoid macular edema and even serous macular detachment may be prominent features.

Subclinical diabetic macular edema (SCDME), defined as edema recognized clinically but not yet meeting the definition of clinically significant. The OCT can identify patients with intraretinal or even subretinal fluid that is not appreciated clinically.

The progression of SCDME to CSDME using OCT has been shown to occur in 1 study in 31.4% of patients over a median of 14 months.⁷

Should edema discovered only on OCT be considered the new CSME “OCT significant macular edema?” There is evidence of a paradigm shift. Many recent diabetic retinopathy clinical research network studies have used mildly increased central retinal thickness as a treatment criterion, especially when evaluating anti-VEGF therapy. In these studies, CSME has been neither an inclusion criterion nor an outcome measure. Earlier diagnosis of DME with OCT could result in closer monitoring, more intensive systemic diabetes management, and earlier more successful treatments.

Methods of Decreasing Progression

Is There a Systemic Relationship to DME?

While local retinal findings indicate progression to DME, ie, MAs, it is likely that diffuse macular edema has systemic relationships. The pathogenesis of DME is multifactorial with relationships to: duration of diabetes, high blood pressure, high glycosylated HgB, cardiac and renal failure, obesity, proteinuria, pregnancy, sleep apnea, and ischemic peripheral retinal – (VEGF). A recent study of diabetic patients with DME showed much higher rates of complications by 3 years than in patients with no DME: myocardial infarction (5.4% vs 3.0%), stroke (4.2% vs 1.6%), congestive heart failure (11.9% vs 4.2%), peripheral vascular disease (11.7% vs 4.5%), cerebrovascular disease, (14.6% vs 7.5%), and renal disease (22% vs 4.9%). Systemic fluid overload clearly plays a role in DME. Additionally, the mean total number of outpatient visits in 1 year was more than 10 days greater in DME patients than in non-DME patients (25.5 vs 14.9 days).⁸

In addition to panretinal photocoagulation and intravitreal interventions that will be discussed at length in a future issue, clinicians and patients should be aware of the impact that better maintenance of systemic factors through diet, exercise, and systemic medications, when necessary, can have on ocular health. Each 1% decrease of HgbA1C reduces DR progression by 50% and results in a 30% to 50% reduction in the development of DME.^9–11 Blood pressure control: each decrease of 10 mm of SBP results in a decrease of 10% DR progression.¹¹ The ACCORD¹² and FIELD¹³ studies showed that lipid management can decrease progression of DR by 30%. Staging DR truly can help guide a primary care physician, as even mild DR has been found to increase the risk of cardiovascular disease.¹⁴

Conclusion

With improved glycemic control, complications from diabetes can be reduced by 76% and improvement in other systemic factors may reduce vision-threatening complications as well. Therefore, all eye doctors must encourage their patients to control their blood sugar and become involved in the medical management of their care. Who better to motivate a patient than the doctor responsible for their most precious gift – vision? With proper classification that predicts progression – timely care is given. Diabetic macular edema and PDR have systemic relationships. Therefore, it is incumbent upon the eye doctor to inform physicians of the potential relationships, such as renal disease with PDR and to urge patients to adhere to the primary care physicians’ guidance to optimize metabolic control.

References

1. Klein R, et al. Arch Ophthalmol. 1984;102(4):520–526.
2. Klein R, et al. Arch Ophthalmol. 1984;102(4):527–532.
3. Diabetes Prevention Program Research Group. Diabet Med. 2007;24(2):137–144.
4. Wilkinson CP, et al. Ophthalmology. 2003;110(9):1677–1682.
5. The Eye Disease Prevalence Research Group. Arch Ophthamol. 2004;122(4):552–563.
6. Diabetic Retinopathy Clinical Research Network, et al. Ophthalmology. 2007;114(3):525–536.
7. Browning DJ, et al. Am J Ophthalmol. 2008;145(1):149–154.
8. Wallick C, et al. Presented at the Association for Research in Vision and Ophthalmology, 2014 Annual Meeting; May 4–8, 2014; Orlando, FL, USA.
9. Ophthalmology. 1984;91(12):1464–1474.
10. Diabetes Control and Complications Trial Research Group. N Engl J Med. 1993;329(14):977–986.
11. UK Prospective Diabetes Study (UKPDS) Group. Lancet. 1998; 352(9131):837–853.
12. ACCORD Study Group, et al. N Engl J Med. 2010;362(17):1563–1574.
13. Scott R, et al. Cardiovasc Diabetol. 2005;4:13.
14. Kawasaki R, et al. Ophthalmology. 2013; 120(3):574–582.