Signs and symptoms of age-related macular degeneration

Undetectable and detectable signs of early AMD

Age-related macular degeneration (AMD) encompasses a spectrum of clinical signs ranging from undetectable to overly manifested macular disorder. The funduscopically undetected signs may overlap with the aging changes of macula. The aging changes of macula were not systemically studied in vivo until the advent of optical coherence tomography (OCT) in recent decades. The availability of a new generation of OCT such as spectral-domain OCT (SD-OCT) enables a powerful optical imaging technique to provide high-resolution optical slides in vivo of the human retina. Nowadays, SD-OCT and other newer OCT techniques can obtain multiple scans over retinal areas and generate quantitative maps of retinal thickness with high spatial resolution. By using OCT, the retinal layers at macula and the volume of macula and the architecture of macula at different locations of the posterior pole can be visualized and quantified. The OCT study on macula creates an opportunity of distinguishing the age-related macular diseases from the aging macula. The following representative SD-OCT slide shows the segmentation of retinal layers, which is compatible with histologic slides ( Fig. 4.1 ) .

Normal aging changes of macula can be identified by both histologic and OCT approaches. Previous histologic studies showed the dynamic changes of human retinal photoreceptors and retinal pigment epithelium (RPE) cells throughout the life span (ranging from the second to the ninth decade). At the foveal center, no significant differences were found in cone or RPE cell densities from the second to ninth decade. It suggests that the densities of foveal cones and RPE cells are kept stable throughout this age period. These histologic findings support the OCT findings. Specifically, the OCT measurement of center point foveal thickness on RPE, fovea cones, and their synapses with inner nuclear layers seems unchanged with age. In addition, OCT measurements also showed that aging does not significantly change central subfield thickness.

However, histologically measured rod density decreases by 30%, beginning from inferior to the fovea in midlife and continuously losing in an annulus between 0.5 and 3 mm eccentricity by the ninth decade. Being equivalent to OCT parafoveal and perifoveal regions by early treatment diabetic retinopathy study (ETDRS) macular mapping, the inner and outer macula were defined as the circular zones around the macula. In fact, OCT measurements of macular regional thickness were in agreement with an age-related decrease in both inner and outer macula-subfield thickness in histology slides. A recent study on the drusen evolution of intermediate AMD patients showed that the emergence and regression of drusen are associated with RPE thickness changes. Therefore, longitudinal measurement of RPE thickness may be a potential marker for predicting drusen emergence/regression and for AMD prognostic development.

A decrease in the inner retina thickness of the aging retina is observed by OCT studies, specifically in the retinal nerve fiber layer (RNFL) and the retinal ganglion cell layer (GCL). Because RNFL and GCL constitute a relatively smaller proportion of the overall thickness at the fovea, the fovea is largely spared from thickness reduction due to loss of GCL and RNFL in the aging retina ( Fig. 4.1 ). ^{, ,} Recent OCT findings reveal the significant changes of GCL thickness within the macula of intermediate AMD eyes. Both age-dependent and AMD disease-dependent GCL loss were found in a location-dependent fashion. ^, The GCL cell loss was faster between the second and fourth decades than between the fourth and ninth decades. Meanwhile, the rod and GCL cell densities at the temporal equator maintained a constant ratio during aging, which suggests an interplay relationship between outer retinal photoreceptor loss and inner retinal neuron death, although they only have trans-synaptic connections.

Both histologic and in vivo OCT findings pointed out that the early AMD signs overlap with naturally aged retina, although the aging retina does not necessarily develop into early AMD. The overlapping could be clinically undetectable in the past, but it now should be explored in the OCT era. If some OCT findings can be used as biomarkers for early AMD study, a key question is that whether the aging changes and AMD changes of chorioretinal layers can be differentiated based on location-specificity? Inferentially, when OCT measurements were used as biomarkers for the study of Alzheimers disease (AD) and other neurodegenerative diseases, distinct OCT features have been found among these diseases. Because β-amyloid plaques in the retina of AD patients were located within sites of GCL degeneration and occurring in clusters in the mid- and far-periphery of the superior and inferior quadrants, ^, OCT findings of decreased GCL at these disease-specific regions other than generalized aging changes of GCL are very informative for helping AD diagnosis. In fact, the distinct pattern of GCL thickness in intermediate AMD patients has been found that the GLC thickness alters in macula spatial clusters, with thinning toward the fovea and thickening toward the peripheral macula. Whether these OCT findings of GCL are AMD-specific, and could be used as disease-specific biomarkers in clinic, merits further study.

AMD is considered as a disease of macular neurovascular complex (MNC), because there is intimate anatomic relationship among PRs, RPE, and CC from early embryonic development. The choroid is a vascular tissue comprising three distinct layers, that is, choriocapillaris, Sattler’s layer, and Haller’s layer. The three layers of choroidal vessels begin to be identified at 21-week gestation, which coincides with the beginning of photoreceptor cell differentiation. As the choroidal vascular system is the only blood supply for oxygen and nutrient demanded by MNC, the diminishing choroidal layers, namely a decrease in choroidal blood flow observed in aging and diseases, will result in damage of this complex. The observation of aging changes of CC in vivo has become accessible because of the emerging of OCT and OCT angiography (OCTA). ^, OCTA utilizes motion contrast technology that can detect the movement of red blood cells in consecutive scans. When OCTA is combined with OCT B-scans, both blood flow and chorioretinal structure can be simultaneously obtained. When swept-source OCT was utilized, choroidal thickness at superior, inferior, nasal, and temporal sites to the fovea could be determined. The choroidal thickness was anatomically thinnest at the nasal site, followed by the order of temporal, inferior, superior, and foveal sites. The choroidal thickness overall decreased with age ranging from 21 up to 85 years.

In a large cohort study measuring multiple chorioretinal segmentation parameters in elderly peoples, only a reduction of subfoveal choroidal thickness was found to be significantly associated with CFH risk genotype (rs1061170), a high-risk genotype of AMD development. ^, In a same-aged group, ranging from 21 to 82 years, choriocapillaris flow density was determined by OCTA. The choriocapillaris flow density was found to be negatively associated with advancing age. Therefore, OCTA-measured choroidal signs are promising to become applicable for early AMD detection.

After teasing out aging chorioretinal changes, the location-dependent and chororetinal-layer-specific imaging markers may find detectable signs in early AMD.

Drusen and pigmentary changes

Drusen [singular: druse] are the hallmark of AMD and the most common early sign of nonexudative AMD. They are visible yellowish deposits under the retina. Generally speaking, the typical location of drusen is along the basal surface of the RPE and corresponds to the thickening of the inner layer of Bruch’s membrane in the posterior pole.

This clinical classification of AMD heavily depends on drusen characteristics (see Table 3.1 in Chapter 3 ). ^, Drusen are characterized by their size; their location related to the cellular structure of RPE cells, for example, subplasma membrane, and above or below basement membrane; type of drusen based on the distinct or ill-defined margin, that is, hard or soft drusen, position of predilection for macula, and association with pigmentary changes.

The impact of druse size on the integrity of the overlying RPE layer and photoreceptor changes is clearly demonstrated. In a cohort study, 5933 drusen of 25 patients with early or intermedium AMD along with their adjacent RPE and photoreceptors were analyzed by polarization-sensitive OCT. In general, when the mean of druse size was 97 μm, the overlying RPE layer was still intact. If the mean druse size increased to 134 μm, the irregular RPE band appeared. It is also notable that a gradually loss of RPE integrity was correlated with the increased size of adjacent drusen. When the mean druse size reached 196 μm, up to one-third of the overlying RPE signal was missing. In this study, although the sample size of drusen with a discontinuous ellipsoid zone (EZ) was too small for statistical analysis, there was a trend with disrupted EZ more often with larger drusen. These data support the guideline of druse-size dependent AMD classification because the size of drusen can determine the degree of photoreceptor-RPE integrity. In addition to the drusen size, the location of drusen related to cellular structures of RPE is illustrated in Fig. 4.2 . The druse location can be precisely identified by histological and ultrastructural studies. The clinical characteristics of drusen and pigmentary changes are identifiable signs correlating with the evolution of subretinal deposits and RPE function. Histologically, the drusen with trichrome present collagen-like materials, while with periodic acid Schiff stain they appear to be glycoproteins. Ultrastructurally, these small deposits between the plasma membrane and RPE basement membrane (basal lamina) are named basal laminar deposits (BlamD). Whereas, the materials as membranous collections visualized by electron microscopy between the basement membrane and the inner collagenous layer of Bruch’s membrane are termed as basal linear deposits (BLinD) ( Fig. 4.2B ). Based on a histological study with postmortem eyes, Sarks et al. found that a continuous layer of early BlamD precedes the first appearance of BLinD. They incisively pointed out that the presence of both BLinD and early BlamD represents the threshold of early AMD. At this point, the fundus is still normal. Because AMD progression depends on the degree of membranous debris accumulation, the large membranous deposit manifests as intermediate or large drusen clinically. When late BlamD appears, it heralds severe RPE damage and corresponds to clinical pigmentary changes. Whereas, if the membranous deposit process is limited to BLinD and basal mounds that are focal membranous accumulations internal to the basement membrane, the fundus may remain normal or present as pigmentary change alone, because RPE integrity may be still intact. On the other hand, the progressively thickening of BLinD may lead to clinically visible soft drusen ( Fig. 4.2C ). While, membranous debris production in some eyes could not progress beyond the formation of basal mounds ( Fig. 4.2C ). In summary, different characteristics of drusen including intermediate/large drusen, pigmentary changes, and soft drusen represent membranous accumulation at different cellular structures of RPE and indication of disease progression. Epidemiologic studies showed that a large amount of membranous debris, in the appearance of large drusen and soft drusen are at the highest risk of developing late AMD over a 5- or 10-year period. If the clinical sign is pigmentary changes alone, the lower rate of late AMD development was observed over the same period. ^, The latter suggests that the membranous deposit process is limited to the formation of BLinD and basal mounds.

Drusen can be further defined by their boundaries including hard drusen with distinct demarcation, soft drusen with poor demarcation, and confluent drusen without clear boundaries ( Fig. 4.3 ). Hard drusen are discrete and well-defined focal areas containing hyalinization of the RPE-Bruch’s membrane complex. Soft drusen share the same materials with BLinD, but BLinD is diffusely distributed. Soft drusen can coalesce to form confluent drusen. Both soft and confluent drusen are associated with a high risk of late AMD development.

Drusenoid pigment epithelial detachment

Clinically observed soft drusen represent the progressive process of membranous deposit, because soft drusen composition resembles both plasma lipoproteins and outer segment. ^, Histologically soft drusen is the same as BLinD ( Fig. 4.4 ).

The progressive BLinD forming thickened inner portion of Bruch’s membrane with overlying RPE can separate from the rest of Bruch’s membrane, resulting in pigment epithelial detachment (PED). This pathological splitting between the inner layer and the rest layers of Bruch’s membrane demonstrated by OCT and histologic study is named as drusenoid PED ( Fig. 4.5 ). Drusenoid PED can be seen in both nonexudative and neovascular AMD.

Curcio raised a very important question that whether soft drusen has position predilection for macula. Sarks et al. observed that soft drusen of elderly patients are located within or just beyond the inner macula (within 3-mm diameter area). In a histology study of postmortem eyes with nonexudative AMD, BLinD is more abundant than subretinal drusenoid deposits under the fovea. Based on the pathology data of 760 eyes with AMD from 450 patients, both BlamD and BLinD at macula are positively associated with choroidal neovascularization. Although the underlying mechanisms of soft drusen’s predilection for macula is currently unknown, longitudinal observation of BLinD/soft drusen at macula is critical for disclose of AMD progression.