Classification of age-related macular degeneration

AMD classification based on drusen and pigmentary changes before OCT era

Understanding the nature and developing effective treatments for age-related macular degeneration (AMD), a consensus classification for defining this leading cause of blindness in the elderly is imperative. First, AMD is a spectrum macular disorder covering both the early and late stages of the disease with a wide range of clinical signs and symptoms. Before optical coherence tomography (OCT) era, as the early AMD is usually asymptomatic, the classification is mainly based on the presence of drusen and pigmentary changes within two disc-diameters of the fovea. Therefore, various criteria including the size of drusen (e.g., small vs. large), the type of drusen (e.g., hard vs. soft), the anatomic position (subretinal vs. sub-RPE), the number, and the area of drusen as well as the location, size, and area of pigmentary changes had been used to classify different stages of AMD among ophthalmologists in clinical practice. Based on these features of drusen and pigmentary changes in the literature, AMD classification and the severity scale in terms of evaluation from early to late AMD are summarized in this section. In addition, the severity scales of 5-year and 10-year risks of advanced AMD in one eye are shown in Tables 2.1 and 2.2 in Chapter 2 . In Chapter 2, Chapter 4 of this book, the morphologic and clinical characteristics of drusen including cuticular drusen and reticular pseudodrusen have been introduced by multimodal imaging techniques.

Conventionally, AMD has been divided into two main types. Dry AMD is the most common type, consisting of 90% of diagnosed cases. Geographic atrophy (GA) is the advanced stage of dry AMD. Wet (neovascular) AMD is less common than dry, but is associated with more rapid loss of sight. The main manifestations are choroidal neovascularization (CNV), subretinal fluid, and pigment epithelial detachment (PED). In recent years, two additional conditions, retinal angiomatous proliferation and polypoidal choroidal vasculopathy (PCV) have been included under the term neovascular AMD, which is collectively characterized as macular neovascularization (MNV).

Based on the age-related eye disease study (AREDS) data by using color fundus photography (CFP), an expert consensus committee pointed out that identification of clinical risks at the early stage that leading to late AMD development is the most important goal for the consensus of classification system. This system unifies vocabulary of practice into a common nomenclature for all clinical providers and researchers. According to this guideline, patients at early stages with the potential loss of vision could be identified. For these patients, additional and/or specific therapy might be discovered and utilized to prevent blindness. The consensus of the classification system is a guideline of all clinical providers and researchers for the communication with their patients and among themselves. In this classification, the size of the drusen and the presence of pigmentary alterations are the most important factors. When the size of drusen is smaller than 63 μm, the term “drupelet” is used as a distinct type of drusen, because drusen of this size contribute no risk of subsequent late AMD. Therefore, the presence of drupelets is considered a normal aging change. The presence of medium drusen (63–125 μm) is a risk factor for the development of large drusen. By calculation of the 5-year rate of progression to large drusen, >50% for patients with medium drusen in both eyes and ∼25% for patients with medium drusen in one eye may develop large drusen. In contrast, the 5-year risk for the development of large drusen in eyes with small or no drusen is less than 5%. The overall 10-year risk for the development of late AMD in patients with bilateral medium drusen is ∼13% and those with unilateral medium drusen is ∼5%, respectively. Based on a post hoc analysis of fundus images taken during the AREDS trial, the risk factors for CNV development correlated with the presence of large drusen (>125 m) within the 2-disc diameter of the fovea were statistically significant.

The pigmentary changes at the macula were analyzed in nine subfields inside the circle with a 2-disc diameter of the fovea. The pigmentary changes were categorized as (1) retinal pigment epithelial degeneration defined as the percentage area of each subfield; (2) increased retinal pigment defined as the presence of granules or clumps of gray or black pigment in or beneath the retina; (3) pigmentary abnormality defined as the presence of either retinal pigment epithelium (RPE) degeneration or increased retinal pigment in the central macular subfield. By using these three criteria, the prevalence of various signs of maculopathy including these three pigmentary changes was estimated by using the Beaver Dam Eye Study (1988–90) data. Retinal pigment epithelial degeneration was present in 20.7%, increased retinal pigment in 24.5%, and pigmentary abnormalities of the RPE in 26.6% of the persons aged 75 years or older, as compared to 3.7%, 6.9%, and 7.3% of persons with age ranging 43–54, respectively. The post hoc analyses of the AREDS trial support these findings that the presence of pigment at the macula is the risk factor for CNV development. The clinical classification of AMD is provided in Table 3.1 .

The classification of AMD and the disease severity scale based on the consensus of experts’ opinions have provided a guideline for follow-up of the disease course from early and intermediate to late AMD. As both large drusen and pigmentary changes in one or both eyes are individually proved to be risk factors for late AMD development, it is reasonable to deduce that patients with an increasing combination of these risk factors are at increased risk. If a risk score of 1 for each risk factor in each eye is assigned, the maximum risk score of 2 can be obtained per eye or 4 per patient. For example, a person who has large drusen and pigmentary abnormalities in both eyes receives a score of 4. Therefore, a 5-step severity scale of 0 (no risk factor) to 4 (both risk factors in both eyes) can be established. For instance, the 5-year risk of late AMD development increases from a score of 0 to a score of 4 (see Table 2.1 in Chapter 2 and Fig. 3.1 ). Most importantly, the prediction of late AMD development according to the severity scale can be understood by both providers and patients. During the follow-up, the modifiable risk factors and dietary/nutritional regimen could be discussed with the patients. The necessary genetic tests and possible future treatment can be conducted based on accurate classification of AMD.

Validation and revision of the AREDS classification of AMD in OCT era

It is clear that the AREDS classification of AMD was established by data mainly obtained by CFP and fluorescein angiography (FA) in a semiquantitative manner known as “a study in FA era.” For more than a decade, the AREDS classification system has been the gold standard for AMD grading and risk stratification. However, this classification system of AMD and treatment algorithms for late AMD has been revised with OCT imaging. In fact, some studies have described confounding errors in the CFP classification of intermediate AMD. In a comparative study, druse areas identified by CFP were compared to those measured by SD-OCT. There was general agreement between CFP and SD-OCT in identifying the presence and absence of drusen. While, the disagreement of these two methods occurred mainly at how to determine druse margins. As CFP-measured drusen may be influenced by the bias of graders, the calculated drusen burden was underestimated particularly in larger drusen with pigmentation. Nowadays, it is achievable to quantify RPE and RPE drusen complex (RPEDC) in eyes of intermediate AMD with semi- or automated OCT. By delineating the inner aspect of the RPE plus drusen material and the outer aspect of Bruch’s membrane, the RPEDC volume containing drusen materials (including subretinal drusenoid deposits) can be measured ( Fig. 3.2 ). ^, Although during the measurement process, the thickness of RPEDC in AMD and the control are largely overlapping, the SD-OCT images successfully distinguished RPEDC thickness of patients with intermediate AMD from the control eyes. Therefore, practically OCT-measured RPEDC thickness became the single most discriminative biomarker of intermediate AMD. In addition, based on OCT measurement, the modified new nomenclature of atrophic AMD is described (see Chapter 5 , Table 5.1).