Ординатура / Офтальмология / Английские материалы / Diabetic Retinopathy_Lang_2007

Table 1. Classification of severity of diabetic retinopathy

1–7 number of fields (fig. 3); D definitely present; DA disc area; DD disc diameter; DR diabetic retinopathy; HE hard exudates; H/MA hemorrhages and microaneurysms; M moderate; NVD neovascularisation of the disc; NVE neovascularization elsewhere; PRH preretinal hemorrhage; Q questionable; S severe; SE soft exudates; VH vitreous hemorrhage.

A total of 50.2% of eyes with severe NPDR will develop proliferative retinopathy within 1 year, and 14.6% will develop proliferative retinopathy with high-risk characteristics. In eyes with very severe NPDR (severe hemorrhages in 4 quadrants, VB in 2 quadrants, and IRMA in 1 quadrant), the risk of developing high-risk PDR is 45% [3]. PDR is categorized into three levels: mild, moderate and high risk (table 1).

Macular edema can be associated with any severity level of diabetic retinopathy. It is the most common cause of visual loss in NPDR. The Early Treatment of Diabetic Retinopathy Study (ETDRS) investigators classified the severity by three characteristics concerning the relation to the center of the macula. The macular edema is defined as clinically significant macular edema if any of the three features described in table 2 is present.

Fig. 3. Seven standard fields of the ETDRS classification, with x representing the fovea [3].

Table 2. Classification of diabetic clinically significant macular edema

Thickening of the retina at or within 500 m of the center of the macula

Hard exudates at or within 500 m of the center of the macula associated with thickening of the adjacent retina

A zone or zones of thickening of 1 disc area or larger, any part of which is within 1 disc diameter of the center of the macula

A more simplified international classification of diabetic retinopathy and macular edema was proposed by the Global Diabetic Retinopathy Project Group (tables 3, 4) in order to improve communication between ophthalmologists and primary care physicians [4]. This classification simplifies the ETDRS classification of diabetic retinopathy and diabetic macular edema for clinical use improving ophthalmological screening of diabetic patients and guiding the definition for laser treatment.

Epidemiology

The Centers of Disease Control estimate that 18.2 million Americans have diabetes mellitus, of whom 90% have type 2 diabetes [5]. Diagnosed diabetes mellitus is most prevalent in the middle-aged and elderly populations.

Table 3. Diabetic retinopathy disease severity scale [4]

About 4.1 million adults over the age of 40 years have diabetic retinopathy [6]. In type 1 diabetic patients, ocular involvement occurs as early as 3–5 years after onset of diabetes mellitus. When the diagnosis of type 2 diabetes is made, up to 15% already have diabetic retinopathy. Occasionally, diabetic retinopathy is the initial sign of type 2 diabetes. The prevalence of any diabetic retinopathy is about 98% after 20 years and about 50% for PDR after 15 years of type 1 diabetes mellitus [7]. About 15% of type 1 diabetic patients have diabetic macular edema after 15 years. In type 2 diabetic patients, 50–80% have diabetic retinopathy after 20 years and 10–30% have PDR. A clinically significant macular edema is found in 25% of type 2 diabetic patients after 15 years [8]. However, the incidence, especially of PDR, seems to be decreasing in recent years due to better glycemic control.

Risk Factors

In several observational studies, poorer glycemic control is associated with increased severity of diabetic retinopathy. When the Diabetes Control and Complications Trial [9] results were stratified by hemoglobin A1c (HbA1c) levels, there was a 40% reduction in the risk of retinopathy progression for every 10% decrease in HbA1c [2]. Therefore, the recommended levels of HbA1c are below 7%.

In the UK Prospective Diabetes Study, a comparison of more intensive blood pressure control versus less intensive blood pressure control in type 2 diabetics demonstrated that better blood pressure control was associated with a decreased risk of retinopathy progression [10]. There might also be a benefit on the progression of diabetic retinopathy of angiotensin-converting enzyme inhibition and blood pressure reduction, even in normotensive persons [11]. The UK Prospective Diabetes Study compared -blockers and angiotensin-converting enzyme inhibitors in tight blood pressure control and found that benefits were present in both treatment groups. Good blood pressure control is considered below 130/80 mm Hg [10].

The Wisconsin Epidemiological Study of Diabetic Retinopathy and ETDRS found elevated serum levels of cholesterol being associated with increased severity of hard exudates [1]. The severity of retinal hard exudates is associated with decreased visual acuity and is a significant risk factor for moderate visual loss [12]. The strongest risk factor for the development of subretinal fibrosis in patients with diabetic macular edema was the presence of severe hard exudates [13]. The Diabetes Control and Complications Trial [14] found that the severity of retinopathy was associated with increasing triglycerides and inversely associated with high-density lipoprotein cholesterol. Elevated triglycerides and lowdensity lipoprotein cholesterol are associated with PDR [15]. These data suggest that lowering elevated serum lipids might reduce the risk of visual loss.

Pregnancy may accelerate the progression of diabetic retinopathy [16], even though this seems to be rare. Pregnant diabetic patients with mild or moderate NPDR should be examined every 3 months, and those with severe NPDR every 1–3 months.

Examination of Diabetic Patients

Clinical examination should include best corrected visual acuity and intraocular pressure to rule out glaucoma. Slit-lamp examination is necessary to detect iris neovascularization.

Fundus examination should be performed with ophthalmoscopy, a fundus contact lens, or with a 78or 90-diopter fundus lens at the slit lamp with dilated

Table 5. Ophthalmological examination in patients with diabetes mellitus

CME Cystoid macular edema; CSME clinically significant macular edema; DR diabetic retinopathy.

pupils to assess the severity of diabetic retinopathy. Pharmacological mydriasis improves image quality, allows better identification of maculopathy and the grading of diabetic retinopathy stage [17]. The risk of angle-closure glaucoma, that might be caused by mydriasis, is extremely rare, especially if the anterior chamber depth is examined by slit-lamp biomicroscopy prior to pupil dilation [18].

Gonioscopy can rule out iris neovascularization and angle closure (table 5). If indicated, fluorescein angiography and optical coherence tomography should be performed (tables 5, 6). If dense vitreous hemorrhage blocks the view on the fundus, retinal detachment should be ruled out by ultrasonography.

This clinical evaluation in diabetic patients is essential to make the correct treatment decisions. Follow-up of the patients depends on the stage of retinopathy and macular involvement (table 6).

Treatment Recommendations

In patients with mild and moderate NPDR, the risk of progression to proliferative retinopathy is very low. Therefore, scatter photocoagulation is not recommended in eyes with mild or moderate NPDR, provided that careful followup can be maintained. In this group, the 5-year rate of severe visual loss is 1–3%.

Good glycemic and blood pressure control and treatment of dyslipidemia, if present, are recommended [19].

Panretinal Laser Treatment of Diabetic Retinopathy

The laser treatment recommendations for diabetic retinopathy are based on the results of two randomized clinical trials of laser photocoagulation, the Diabetic Retinopathy Study (DRS) [20] and the ETDRS [21].

The results of the DRS showed a 50% reduction in severe visual loss in eyes with severe NPDR or PDR and visual acuity of 20/100 or better that had received photocoagulation compared with eyes that were not treated. The overall risk of severe visual loss with PDR at the 2-year follow-up examination was 6% in the treated eyes compared with 16% in the control group. With the DRS high-risk characteristics, the risk increased to 11% in the treated eyes compared with 26% in the control group.

The reports of the DRS identified retinopathy characteristics with a high risk of severe visual loss which are neovascularization on the disc or any neovascularization accompanied by vitreous hemorrhage [1]. Panretinal laser treatment considerably improves the visual prognosis, especially in PDR. Wei et al. [22] found a complete resolution of neovascularization in 67% of eyes, and a partial resolution in 33% which needed additional photocoagulation. Postoperative preretinal and vitreous hemorrhage occurred in 9%. Qian et al. [23] report that panretinal argon laser photocoagulation is effective in 85% of eyes with NPDR and in 77% with PDR. Visual acuity improved in 23% and was unchanged in 61%. Rema et al. [24] studied the outcome of patients with type 2 diabetes and PDR after panretinal laser treatment. Seventy-three percent of patients with visual acuity of 6/9 or better maintained their vision. Of patients with visual acuity 6/12–6/36, 59% maintained the same vision and 20% improved their vision at 1-year follow-up. Of patients with visual acuity 6/60 or less, 70%

Fig. 4. a PDR with vitreous hemorrhage before laser treatment. b The same eye after panretinal laser treatment.

maintained their vision and 30% improved. On multiple regression analysis, diastolic blood pressure, duration of diabetes, fasting blood glucose and nephropathy were associated with decreased vision after panretinal laser treatment.

Scatter laser photocoagulation should be considered in severe and very severe NPDR, especially in patients with poor compliance, proliferative disease in the fellow eye, pending cataract surgery, poor glycemic control, high blood pressure, advanced renal disease and extensive capillary closure [25]. In eyes with severe and very severe NPDR and clinically significant macular edema, the macular edema is treated first and panretinal scatter photocoagulation should be delayed until the macular edema has improved.

In mild and moderate PDR, scatter laser photocoagulation should be performed because it reduces the risk of severe visual loss, especially in type 2 diabetes. The clinically significant macular edema and cystoid macular edema should be treated first, before performing panretinal laser treatment.

In PDR with high-risk characteristics, extensive scatter laser photocoagulation should be performed immediately because of the high risk of visual loss (fig. 4).

In the standard full-scatter panretinal photocoagulation, 1,200–1,600 burns of 500- m spot size on the retina are applied to the retina (table 7; fig. 4). The burns are placed from the vascular arcades to the equator, nasally 500 m apart from the optic disc and temporally 2 disc diameters temporal to the macular center (3,000 m). Treatment has to be extended to within the vascular arcades, if retinal neovascularization is located within this area with a spot size of 200 m when treating within 1,500 m from the center of the foveal avascular zone. Treatment should not be extended to closer than 500 m from the macular center. If a grid laser treatment has been performed for clinically significant macular edema, the panretinal burns should be placed temporally adjacent to the macular burns, if there are retinal edema or areas of nonperfused

Table 7. Treatment of diabetic retinopathy

Diabetic retinopathy

Mild and moderate PDR

LP laser photocoagulation; ME macular edema; VH vitreous hemorrhage.

retina that carry the risk of later development of neovascularization. The burns should have moderate intensity with a spacing of about 1 burn width apart. In patients with high-risk PDR and iris neovascularization, a number of 2,000 burns is recommended. Panretinal photocoagulation should be completed in two or more sessions within 3–6 weeks. One session should not extend 500–600 burns to avoid side effects that are described below. Further division of panretinal treatment sessions can be considered when clinically significant macular edema is present, because this reduces the risk of visual loss. The order of sessions in which the retina is treated is optional. If there is a risk of vitreous hemorrhage, the inferior quadrants should be treated first. Disc neovascularization and elevated neovascularization elsewhere are treated with scatter photocoagulation in an attempt to get a regression of the new vessels. Flat neovascularization elsewhere is directly treated with confluent laser burns with 200–500 m in size on the retina to close the new vessels. Panretinal scatter photocoagulation is very effective in iris neovascularization (at least 2,000 laser burns), especially if treatment is given prior to the development of neovascular glaucoma. In neovascular glaucoma, panretinal laser treatment should be combined with a cyclodestructive procedure. Preferably, the scatter treatment should be applied first, because after the cyclodestructive procedure, fibrin and cells in the anterior chamber might hinder laser treatment.

Table 8. Panretinal laser treatment

NVE Neovascularization elsewhere.

Panretinal photocoagulation significantly reduces the risk of severe visual loss, but in some patients, vision may worsen in spite of laser treatment. If in patients with disk neovascularization the regression of the neovascular vessels is slow, the risk of vitreous hemorrhage persists. The patients must return for follow-up visits at 3-month intervals for a successful result, since additional laser is needed in at least 30% of patients, because of the insufficient regression of neovascularization in at least one third of the patients. Additional fill-in laser treatment between prior laser scars and photocoagulation of the peripheral retina are required, if the new vessels do not regress sufficiently.

The Goldmann three-mirror laser lens gives an upright image and can be used for retinal posterior pole and periphery treatment. For panretinal treatment, wide-angle lenses are usually employed. They provide a wider view and have an invert image. They differ in image magnification, field of view and laser spot magnification factor (table 8). The wider the field, the smaller the image magnification.

The panretinal laser treatment can usually be performed under topical anesthesia. If a patient is experiencing more severe pain, the pace of the applications can be lowered, and the ciliary nerves at 3 and 9 o’clock should be avoided or subconjunctival anesthesia can be used.

Laser lenses suitable for panretinal laser treatment are wide field lenses with a magnification of 1.4 for the area up to the equator and lenses with a magnification of 1.8 for the retinal periphery.

The size of the laser burns in the retina depends on the laser spot size, the magnification of the lens, the power and duration of the applied laser burn, the transparency of the media (cataract requires more energy, pseudophakia less energy), and the pigmentation of the eye. The power of the laser burn is the minimum necessary to obtain a burn of medium, gray-white intensity.

However, in a prospective randomized controlled clinical trial, Bandello et al. [26] found that light panretinal photocoagulation with a very light effect of the burn on the retina on biomicroscopy had the same efficacy in comparison with classic treatment with white burns in eyes with high-risk PDR. Light coagulation was associated with fewer complications and allowed reduction in number of treatment sessions. The regression of neovascularization was the same in both groups, the total mean session number was 7.4 for the light and 9.9 for the classic treatment group. Therefore, light effect of burns can also be considered for panretinial laser treatment.

Zaninetti et al. [27] emphasize that eyes requiring vitrectomy because of vitreous hemorrhage or retinal detachment in PDR after panretinal laser photocoagulation are often a result of incomplete photocoagulation. Therefore, sufficient panretinal laser treatment is mandatory.

Strong positive predictors of post-panretinal photocoagulation visual acuity outcome are pretreatment visual acuity and low age. Diabetes type and diabetes duration have no influence on visual outcome [28]. The visual prognosis is inversely related to the number of treatment sessions required.

Exploration of symptoms revealed that the most frequently reported symptom due to diabetic retinopathy is blurred vision (55% of patients). First-time laser-treated patients report fewer symptoms than multi-treated patients. The patients’ expectations were basically met; however, the treatment had less of an impact than they had hoped for. Patients would have laser treatment again if they needed [29].

Laser Treatment of Diabetic Macular Edema

Diabetic macular edema may be present at any level of NPDR and PDR. It is caused by either focal or diffuse leakage. As the severity of diabetic retinopathy increases, the proportion of eyes with macular edema increases, ranging from 38% in eyes with moderate to severe NPDR to 71% in eyes with PDR [30].

The diagnosis of macular edema requires stereoscopic examination of the macula at the slit lamp with a fundus lens with dilated pupil. If the diagnosis of diabetic macular edema is made, a fluorescein angiography should be performed to rule out additional ischemic maculopathy. With the OCT, the macular edema can be quantified and posterior hyaloid traction can be detected. Secondary epiretinal membranes can also be associated with diabetic macular edema.

Diabetic macular edema is classified into clinically significant and clinically insignificant according to the ETDRS (table 2) or, by the more simplified scale for diabetic macular edema, into mild, moderate and severe diabetic macular edema (table 4).