In the United States, age-related macular degeneration (AMD) is the leading cause of blindness in the elderly population.¹ AMD causes visual impairment in an estimated 1.7 million Americans over the age of 65.¹

The good news is that new drugs that target vascular endothelial growth factor (VEGF) allow for more effective treatment of AMD patients. The prognosis, however, largely relies on early detection of the disease. By contrast, delays in AMD treatment can result in significant vision loss within a short period of time.

Advances in retinal technology have allowed for early detection and treatment of AMD. These include, but are not limited to, optical coherence tomography (OCT) and preferential hyperacuity perimetry (PHP). Either device may result in earlier diagnosis and treatment of AMD patients. In this paper, we will discuss the role of these new technologies in the detection of exudative, or wet, AMD.

Two Forms of AMD

Age-related macular degeneration is characterized by pathological changes associated with deterioration of the retinal pigment epithelium (RPE), Bruch’s membrane and choriocapillaris. Further damage to the overlying sensory retina results in decreased visual function.

There are two forms of the disease: nonexudative and exudative. The end stage of either form is associated with severe, irreversible vision loss.

Nonexudative AMD, also referred to as atrophic or dry AMD, is more prevalent than wet AMD.² Dry AMD is associated with both drusen formation and RPE alteration. While there is no effective cure for AMD, studies have shown that supplementation with antioxidants and zinc can decrease the likelihood of progression to an advanced form of the disease.^3,4

Exudative AMD, also called neovascular or wet AMD, affects only 10% to 20% of patients with AMD, yet it accounts for the majority of severe vision loss among all AMD patients.^1,2 Wet AMD is characterized by choroidal neovascularization (CNV), which may be accompanied by subretinal leakage, edema and hemorrhage. This leads to damage of the neurosensory retina. Irreversible, severe vision loss is frequently linked to fibrovascular scar formation, marking the end stage of the disease. Treatments, such as laser, intravitreal steroid injection and anti-VEGF agents, are typically intended to stabilize vision and prevent disease progression, although they may actually improve visual function.

According to the Treatment of AMD with Photodynamic Therapy Trial (TAP), choroidal neovascular lesion size is the single most important feature associated with the post-treatment prognosis.⁵ Treatment of smaller lesions results in better postoperative visual acuity. CNV averages a growth rate of 10µm to 20µm per day.⁶ At the time of diagnosis, the average size of a CNV membrane is 3,300µm.^5,7 Furthermore, treatment is less likely to be effective for larger lesions.

Cross-Sectional Imaging

During the last decade, clinicians have used OCT to evaluate and monitor patients with macular disease. OCT is a non-invasive, non-contact imaging technique that uses low-coherence laser interferometry to construct a cross-sectional image of the retina.⁸ The 10µm axial resolution is suitable for visualization of the major retinal layers, providing valuable diagnostic data.

1. An optical coherence tomography (OCT) image shows choroidal neovascularization associated with wet AMD.

Using the principles of interferometry, the time-delay light signals reflected from various ocular tissues are compared with a moving reference light of known path length. A detector analyzes the interference patterns to generate a two-dimensional cross-sectional image, or tomograph, of the retina. This image is displayed in “false” colors, with bright colors representing high optical reflectivity and dim colors showing minimal reflectivity.

OCT calculates retinal thickness maps and charts from six radial line scans, which all intersect through the fovea.⁹ Retinal thickness is estimated as the distance between the internal limiting membrane (ILM)/nerve fiber layer (NFL) and the RPE. In accordance with standards from the Early Treatment Diabetic Retinopathy Study, OCT extrapolates retinal thickness values and displays them within the nine macular zones.^8,10 These measurements are reliable and reproducible, making them very useful resources for monitoring macular diseases.

Although fluorescein angiography (FA) remains the best method for evaluating CNV, non-invasive OCT has expanded the potential to better manage AMD patients. A recent report cited that 37,000 OCT scans are being performed daily in the United States.¹¹ The dramatic increase in the number of OCTs performed has been linked to a decreased use of FA.¹²

FA classifies two distinct types of CNV. The classic choroidal neovascular membrane appears as a lacy, hyperfluorescent lesion during the early stages of angiography and exhibits intensified leakage during the later stages.

2. OCT reveals hyper-reflective anomalies within the retinal pigment epithelium resulting from choroidal neovascularization.

The pattern observed with occult CNV, however, is not as obvious. There is less leakage as well as indistinct boundaries and irregular hyperfluorescence in the late stages of the angiogram.

Confounding factors, such as retinal hemorrhages, may further obscure significant FA findings. Other disadvantages of FA include cost and the potential for complications associated with the procedure.

Patients with AMD face a wide range of potential morphological change, such as RPE atrophy, CNV and retinal edema. Unlike FA, OCT provides structural information, allowing the clinician to visualize such morphological changes.

3. OCT is capable of detecting a subfoveal retinal detatchment caused by cystoid macular edema.

Categorizing Wet AMD

Using the OCT, researchers classified wet AMD into two distinct categories.¹³ The classic CNV membrane appears as a fusiform (spindle-like), hyper-reflective lesion either above or below the RPE (figure 1). Poorly defined CNV lesions present as hyper-reflective irregularities within the RPE and are typically associated with subretinal edema or detachments (figure 2).

The OCT’s discrimination of details can also aid in the diagnosis of associated findings. Specifically, OCT evaluation of a patient with wet AMD can help the clinician:

• Detect and localize a subfoveal retinal detachment under cystoid macular edema (CME) (figure 3).

• Determine the stage of associated retinal angiomatous proliferation.

• Detect RPE tears and differentiate serous RPE detachments from neurosensory detachments.

Serous RPE detachment is observed as an elevated, hyper-reflective pigmented epithelium with an underlying optically clear space and sharp borders (figure 4).

Limitations of OCT View of the subretinal area is limited. Dilation is typically required. Image quality is influenced by several factors, including: • Poor tear layer. • Cataract. • Dirty lens instrument. • Lateral eye movement.

Structural changes associated with wet AMD may develop prior to significant visual changes or ophthalmoscopic signs (see “Case Report: Detachment Suggests Conversion to Wet AMD,” below). Thus, the OCT can provide an early diagnosis, which should lead to a timely referral for treatment.

Current guidelines for therapeutic intervention of wet AMD are mandated by findings observed on FA.

However, some clinicians argue that FA alone cannot adequately confirm the type, size or even presence of a CNV lesion following treatment.¹⁴

4. A serous detatchement of the retinal pigment epithelium is observed on OCT.

For example, FA findings may show late hyperfluorescence associated with a fibrovascular disciform scar following treatment for wet AMD. This late staining pattern should be indistinguishable from true leakage associated with active CNV.

The morphological changes related to a fibrovascular scar are more clearly demarcated on OCT. A fibrovascular scar on OCT is de-depicted as hyper-reflective RPE with moderate backscattering. Retinal profile changes, such as loss of normal foveal contour, are also typical. The associated overlying retinal atrophy is delineated as intraretinal thinning.

Retinal Thickness

Activity associated with a new onset or recurrence of CNV is indirectly measured through qualitative and quantitative increases in retinal thickness. Accumulation of subretinal fluid directly associated with recurrence of CNV almost always accompanies either a neurosensory or RPE detachment.

OCT can be used to monitor progression or recurrence of CNV associated with wet AMD through the use of quantitative indices associated with increased retinal thickness. Following treatment, OCT can also be used to observe a decrease in retinal thickness (figures 5 and 6).

Both the restoration of normal foveal depression and improvement of visual acuity following treatment show how OCT is able to correlate structural changes and visual function. This promotes a better comprehension of the pathogenesis and mechanism of visual disturbances.

With the advent of anti-VEGF therapy, OCT has optimized “retreatment” protocols. Several clinical trials, including the Prospective OCT Imaging of Patients With Neovascular AMD Treated With IntraOcular Lucentis (PrONTO) Study from the Bascom Palmer Eye Institute in Miami, have shown the benefits of OCT imaging as a guideline for additional treatment, such as early visualization of CNV recurrence.^15,16

5,6. OCT images show a recurrance of choroidal neovascularization associated with wet AMD. The image above was taken before the patient underwent treatment, and the image below was taken following treatment.

Although OCT alone has been proven to be very thorough in the evaluation of CNV, it is only moderately effective at the actual detection of CNV.¹⁷ The OCT is a powerful screening tool that has valuable diagnostic potential when used in conjunction with FA and/or indocyanine green angiography.

During the last few years, improved image quality and resolution has increased the sensitivity of OCT. Higher resolution is achieved through the use of broader bandwidth. Ultra-high resolution (UHR) OCT uses a titanium-sapphire short pulse laser, which produces a resolution of 3µm. Microscopic structures, such as the photoreceptor layer, can be visualized at such higher resolutions.

UHR OCT primarily enhances localization, enabling the user to distinguish subretinal layers from the RPE and intraretinal layers.¹⁸ However, the laser used for UHR OCT is expensive and hard to maintain, which greatly limits its commercial availability.

Spectral Domain OCT

Despite the value of OCT in patients who have AMD, it also has several limitations (see “Limitations of OCT,” above). Because of OCT’s low acquisition speed, lateral eye movement can degrade or distort the image. However, recent advances in OCT technology have focused on improving both image resolution and speed, without sacrificing image quality.

The most technologically advanced form of OCT, spectral domain, can scan a large area in a relatively short period of time, yielding 3µm axial resolution.

These high-resolution images reveal better structural details. This technology makes a clear distinction between subtle pathological changes and normal variation. Carl Zeiss, Optivue, Topcon and Heidelberg are just a few manufactures that have or will soon introduce spectral domain OCT to the commerical marketplace.

Spectral Domain in Detail Spectral domain uses a spectrometer as a detector along with a stationary reference mirror.27 The lack of moving parts yields exceptional speed, capable of acquiring 30,000 A-scans per second. An increased number of scans set at different locations and orientations allows for coverage of the entire retina. Spectral domain OCT is capable of producing high-resoultion imagery. Also, the user can focus on an exact area of interest. Each cross-sectional scan is registered to a spatial location on a fundus image, allowing for a specific, line-by-line comparison, with precise localization.27 These scans can be used to accurately monitor macular disease over time. The high-speed potential allows for re-creation of a 3-D retinal image. This technology produces more clinical information regarding normal and pathological changes. Specifically, the 3-D retinal maps provide precise retinal thickness assessment and allow for a more accurate follow-up examination. The location, depth and extent of a CNV can be easily demarcated using a 3-D retinal map. In addition, volumetric measurements can be used to identify small changes over time. Another important feature: the ability to segment the retinal layers. Improved clinical diagnosis can be achieved by separating retinal layers, allowing for exact visualization of the pathological changes occurring in each individual layer. Improved image quality, increased image acquisition speed and software features will provide more detailed information, further enhancing clinical diagnosis.

The current commercially available OCT (Stratus OCT 3, Carl Zeiss Meditec Inc.) relies on the principles of time domain, requiring movement of a reference mirror and the use of a single detector to generate 2-D imagery. Time domain is capable of acquiring 512 A-scans per 1.3 seconds; these are extrapolated from six radial scans intersecting through the fovea.⁹

Preferential Hyperacuity Perimetry

The Amsler grid traditionally has been used for the evaluation and self-monitoring of AMD, but several studies have found limitations of this method.^19-21 Amsler grid tests are less than effective in the early detection of CNV lesions.²² In one documented case series, the limitations of the Amsler grid may have been partially responsible for late detection of CNV lesions.⁷ At the time of detection, the CNV lesions appeared relatively large, and were located in the subfoveal area. This finding contributed to the severe decrease in visual function.⁷

However, there are two major problems with Amsler grid testing. One is patients’ inability to properly maintain fixation and noncompliance in performing this test. Peripheral scanning of the entire grid is also difficult to control, masking non-foveal defects.

Other limiting factors, such as “crowding effect,” occur when adjacent lines impinge on the patient’s fixation line(s).¹⁹

The crowding effect decreases spatial discrimination, which makes detection of peripheral abnormalities difficult. Also, cortical completion, the process in which the brain artificially “fills in” distorted or missing segments of imagery, usually causes central visual defects to be entirely overlooked.

AMD patients are unaware of a small or subtle central defect until it is large enough to affect visual acuity. Nearly 80% of eyes with CNV have worse than 20/40 vision at the time of diagnosis.⁷

New technologies, such as the computer-automated threshold Amsler grid (TAG) perimeter, improve sensitivity and can display a 3-D map of the central visual field. The TAG perimeter achieves higher resolution by varying perceived luminance, using continuous-motion touch screen LED display of differing wavelengths.²³ Three-dimensional mapping allows for visualization of depth, extent and shape of the defect.

More recent advancements in patient evaluation, such as the Preferential Hyperacuity Perimeter (PHP), have addressed the aforementioned inherent problems associated with the Amsler grid testing. The PHP can quantify and map central defects associated with CNV in AMD patients (figure 7). Using 500 data points, PHP tests the central 14 degrees of visual field.

Visual field defects are mapped through the use of a touch screen to identify flashing dots, with a deviating signal on distinct areas of the macula.

PHP is analogous to a Humphrey visual field perimeter, but PHP uses hyperacuity, or Venier acuity, rather than white light. Hyperacuity is the eye’s ability to detect subtle, relative spatial localization of two objects. Hyperacuity is 10 times more sensitive than Snellen visual acuity.²⁴

RPE elevation, which may occur in AMD patients, likely causes a shift in the regular position of the photoreceptors. Such a shift may cause the afflicted individual to see an object at a different location than its actual location in space.²⁵ The PHP records this perceived shift, posing an anatomical explanation for the metamorphopsia.

7. A hyperacuity deviation map as measured by preferential hyperacuity perimetry (PHP). The darker shaded areas of red illustrate where visual field defects are the most severe and pronounced.

Other Recent Technologies Retinal thickness measurements enable the identification and tracking of structural changes associated with various retinal diseases, including AMD. The following technologies can be useful in measuring retinal thickness and macular changes: • HRT 3. The Heidelberg Retina Tomograph 3 (Heidelberg Engineering) uses confocal scanning laser technology to combine retinal thickness and edema indices (Edema Index).28 The Edema Index is a relative indicator of fluid accumulation based on changes in light reflectance. Index values higher than 2.0 likely indicate fluid accumulation. The HRT 3 captures three full scans, compiling a total of 192 single images, which measure the entire retina. Then, the TruTrack technology software checks and aligns the images, removes images with questionable quality, and combines the sets into one 3-D composite image. It also provides the Edema Index and retinal thickness maps. • RTA 5. The Retinal Thickness Analyzer (Talia/Marco) offers the ability to quantitatively document anatomical changes in retinal and subretinal tissues by measuring retinal thickness variations and topographic changes of the chorioretinal interface using a scanning laser ophthalmoscope (SLO).29 The RTA 5 presents data as color-coded 2-D and 3-D thickness and topography maps, deviation probability maps (from a normative database), numerical values, interactive 3-D cut sections and digital fundus images. • Stereo fundus cameras. Stereo fundus photography is also useful when evaluating macular pathology. The Automated Retinal Imaging System (ARIS, Visual Pathways, Inc.) is a patented, fully automated camera system that images the fundus stereoscopically.30 ARIS can obtain infrared, red-free and color images to evaluate deep retinal tissues. The Nidek 3Dx stereo fundus camera and 3Dx/F fluorescent stereo fundus camera are also newly available for stereoscopic macular imaging. The 3Dx has the capability for stereo color photography and FA.31

The PHP analysis provides reliability indices to indicate if the defect is outside the normal limits. Additionally, the analysis reveals the existence of a corresponding defect zone within the macular area.

Finally, it provides general recommendations for further action. A recent study found that PHP had an 88% specificity rate in differentiating CNV from the intermediate stage of dry AMD, and an 82% sensitivity rate in accurately detecting newly diagnosed CNV.²⁴

Another advantage of PHP is its ability to provide reproducible quantitative measurements, which may be used to monitor patients over time. The results are not dependent on the size or location of the lesion, deceased contrast sensitivity, media opacities or increased patient age. Additionally, PHP is a non-invasive, cost-effective method of monitoring AMD.

However, there are some disadvantages to PHP. Elderly patients with poor manual dexterity may find the touch screen difficult to use. Additionally, false-positive readings can be mistakenly associated with other causes of metamorphopsia.^25,26 Selecting appropriate patients for the test typically decreases the number of false-positive results. Patients that have geographic atrophy, worse than 20/200 visual acuity, or a previous diagnosis of wet AMD should be excluded from the test.

False-negative readings with the PHP can also occur (see “Case Report: Detachment Suggests conversion to Wet AMD,” below). Although a visual field defect was noted in this case, it did not appear outside of normal limits. This may have occurred because we did not have a baseline negative PHP visual field for comparison. Using baseline and serial PHP tests is crucial for proper monitoring. In the presence of a recent onset CNV lesion, growth can occur rapidly, which can cause devastating vision loss.

Researchers and clinicians are debating ideal serial testing time. The manufacturer recommends that the test be performed once the patient is diagnosed with intermediate dry AMD, and should be repeated every three to four months, thereafter.

While the patient may appear asymptomatic, and has a negative Amsler grid test, an undetected CNV lesion may be causing irreversible retinal damage. PHP provides early detection and reliable quantification of hyperacuity peri-metric defects consistent with the progression of intermediate dry to early wet AMD. PHP may identify high-risk patients, even when symptoms are minimal or absent.

Case Report: Detachment Suggests Conversion to Wet AMD A 57-year-old white female with dry AMD in both eyes presented for a three-month follow-up. She reported no visual or ocular complaints. Her systemic history was unremarkable. Her ocular history was significant for posterior chamber intraocular lenses O.U. She also had glaucoma O.U. that was controlled with Cosopt (dorzolamide and timolol maleate, Merck). Her dry AMD O.U. had been diagnosed two years prior. Best-corrected visual acuity was 20/30 O.D. and 20/70 O.S. Her acuity three months earlier was 20/25 O.D. and 20/50 O.S. Amsler grid tests revealed mild metamorphopsia O.S., which was unchanged from her previous examination. The preferential hyperacuity perimetry (PHP) revealed a hyperacuity defect in the superior-nasal field (6° x 4°) O.S.; however, the defect was not shown to be outside the normal limits. 8. This graphic reveals an increase in retinal thickness caused by a recurrence of choroidal neovascularization. All other aspects of the preliminary examination were unremarkable. Intraocular pressure measured 12mm Hg O.U. Anterior segment biomicroscopy revealed no pathology. Dilated fundus examination revealed healthy optic nerves with pink, distinct neuroretinal rims, and 0.50/0.50 cup-to-disc ratios O.U. Both maculae appeared flat and intact with moderate drusen. There was no ophthalmoscopic evidence of chorodial neovascularization (CNV) or subretinal fluid in either eye. Macular optical coherence tomography (OCT), however, revealed a shallow serous retinal detachment O.S.; this was associated with an optically clear space below. The foveal contour was intact. Also, a well-defined, hyper-reflective fusiform lesion was noted, which was protruding through the RPE (figures 8 and 9). We referred the patient for FA, which confirmed the diagnosis of early-stage wet AMD. The patient was scheduled for treatment with an anti-VEGF agent. In this case, we do not know the patient’s outcome, as we lost her to follow-up. 9. OCT shows a serous neurosensory retinal detatchment caused by choroidal neovascularization. However, for the purposes of this article, the most important item to note is that these retinal technological advancements enhanced our ability to diagnose and manage our patient with AMD, ultimately resulting in the preservation of her vision.

 

Clinicians today have more options than ever before to both monitor dry AMD and provide early detection of its conversion to the wet form. New, non-invasive technologies facilitate more accurate examinations and can help identify high-risk patients, even when symptoms are minimal. Ultimately, new retinal technologies expedite the treatment process, which may prevent associated visual damage.

Dr. Shechtman is an associate professor of optometry at Nova Southeastern University College of Optometry. Dr. Reynolds has published and lectured on various ocular disease topics and has an interest in the eye care of minority populations. Dr. Pizzimenti is a frequent speaker, author and clinical researcher in the area of age-related macular degeneration.