New detection technologies important in management of AMD

ByNeil M. Bressler, MD

Of the 8 million people in the United States who are older than 55 years and have been diagnosed with intermediate age-related macular degeneration, over the next 5 years, 200,000 people annually will develop the neovascular or geographic atrophic form of AMD.^1,2 Early detection of choroidal neovascularization is a vital part of treatment. If CNV can be detected early, before significant loss of visual acuity, steps can be taken to reduce the risk of further progression and prevent further loss of visual acuity.^3-5

Obstacles to early detection

A number of limitations hinder early detection of CNV. Patients often recognize symptoms, such as distortion of lines, blurred or reduced vision or dark areas in their visual field, late in the course of the disease when visual acuity is 20/100 or worse and a scotoma has already formed.

It also can be difficult to diagnose CNV even when patients visit their ophthalmologists once or twice a year. Often, no blood or lipids, which suggest CNV, can be detected when looking at the back part of the eye. In these patients, however, a fluorescein angiogram may show features of CNV, in which CNV is present on the angiogram and a change in elevation of the retinal pigment epithelium (RPE) is appreciated, but the change can be subtle and may not have caused any changes a physician can detect biomicroscopically or that the patient would notice in terms of vision loss.

Patients may not notice a minor loss of visual acuity from the development of CNV because images “seen” in the brain are not exactly the images formed on the retina. The brain ignores many retinal defects. For example, everyone has a blind spot created by the optic nerve in each eye, but it usually is not recognized because the brain completes the image.

When CNV has been detected in the past, the average size of the neovascular lesion was approximately 3,300 µ.⁶ Eighty percent of new cases were subfoveal, and about 40% of the cases had a visual acuity of less than 20/200.⁶ Data from the Complications of Age-Related Macular Degeneration Prevention Trial (CAPT) showed that even with annual angiograms and intensive monitoring every 6 months, two-thirds of patients had a visual acuity of 20/40 or better when they developed CNV, and 60% of CNV developed under the center of the fovea.⁷

The Amsler grid in CNV detection

Data have proved that the Amsler grid is not sensitive at detecting CNV in patients with AMD.⁸ There are three theories as to why this is so:

The Amsler grid does not force fixation. People have a natural tendency to scan with the fovea. Focusing on the center of the grid and relating to the periphery is an unnatural action. Because the Amsler grid does not force fixation, a slight distortion may “move” around the grid as the patient scans. Subsequently, the distortion may not remain fixed long enough for the brain to detect it.

The Amsler grid does not overcome cortical completion. Completion is when the brain “fills in” the rest of an image, like filling in the blind spot created by the optic nerve or creating an image of a white triangle when viewing a Kaniza triangle (Figure) instead of seeing the true image on the retina. The brain is capable of filling in distortions, making them unrecognizable to the patient.

Inhibition by neighboring peripheral lines, or crowding, reduces detection. When viewing multiple data points simultaneously, the brain has reduced ability to isolate a single data point. Similarly, neighboring lines in the Amsler grid reduce the ability to isolate distortions.

Preferential Hyperacuity Perimeter

The Preferential Hyperacuity Perimeter (PreView PHP, Carl Zeiss Meditec, Dublin, Calif.) was developed to detect CNV earlier and monitor patients better than common methods.

The PHP is a perimetry machine similar to that used in a Humphrey visual field examination for detecting glaucoma. Both diagnostic tools measure the peripheral visual field, are automated, have a normative database for comparison and provide a report. The PHP also measures the central visual field in addition to the periphery. The PHP produces a hyperacuity stimulus (as opposed to the Humphrey visual field perimeter’s white-on-white stimulus), which it compares to the normative database and provides a report on whether abnormality consistent with the presence of CNV exists.

The PHP is based on Vernier hyperacuity, or the human ability to perceive minute differences in the relative spatial localization of two objects. Because the brain is sensitized to the detection of small shifts in the co-linear arrangement of photoreceptors and slight distortions of photoreceptors in relation to each other, minute differences in the placement of objects in space in relation to one another are readily noticed. This hypersensitivity is subsequently used to detect the presence of drusen or CNV, as these different lesions cause different elevations in the RPE.

How PHP works

The PHP shows the patient a dotted line, with some of the dots deviating, introducing an artificial distortion that the patient should easily see if the distortion is great enough. The set of dots appears too briefly to allow completion from the brain or steady fixation to be required and, because only one set of dots appears at a time, crowding cannot occur. The patient then points out on the touch-sensitive screen where the distortion of the line occurred. This artificial distortion is progressively made smaller. When the elevation caused by CNV (due to the lifting up of the photoreceptor from the RPE) is larger than that of the artificial distortion, the patient will preferentially point out the area of true distortion.

The PHP goes though about 500 data points tested along the central 14° where CNV is likely to be present. The test takes approximately 5 minutes per eye and provides reliability parameters that indicate the likelihood of a false-negative or false-positive, based on the patient’s test taking. The machine notes the results of any true distortions recorded by the patient, compares them with the competing artificial distortion and calculates the likelihood of CNV.

How does PHP measure up in detecting CNV?

In a study of an early version of the PHP, the technology detected CNV in 94% of the 32 participants with CNV, compared with 34% detected with a supervised Amsler grid.⁸ A high number of false-positives were also recorded, however. Subsequently, the algorithm used to compare the test results against the normative database was changed. In another trial performed on subjects with a range of AMD lesions, 100% of patients with CNV were detected with the PHP vs. 53% detected with a supervised Amsler grid.⁹ A high number of false-positives were recorded,⁷ however, and the algorithm was changed to what is used in the technology today.

“Proper management includes relating to the patient the importance of self-monitoring and periodic exams, although these are limited in their ability to detect CNV.”
– Neil M. Bressler, MD

The most recent study was designed to determine whether the PHP can distinguish patients with CNV of a recent onset (within the last 60 days) from those with intermediate AMD. Participants underwent fundus photography, fluorescein angiography, a PHP examination and a visual acuity examination. Fundus photographs and fluorescein angiograms were analyzed by a centralized reading center for the gold standard, in addition to being analyzed by retinal specialists.

According to the gold standard, 65 participants had recent onset CNV and 57 participants had intermediate AMD. According to the data, the sensitivity, or the probability that the PHP result will be correctly positive, was 81.5%.¹⁰ The PHP’s specificity at discerning intermediate AMD from recent onset CNV was 87.7%.¹⁰ The sensitivity of having a retinal specialist study the color fundus photos was 70% (95%, CI: 58% to 81%) and the specificity was 95% (95%, CI: 85% to 99%).¹⁰ In a subgroup of 66 patients with visual acuity of 20/40 or better, the sensitivity of the retinal specialists was 53% (95%, CI: 30% to 76%), compared with 60% of the PHP. Specificity of the retinal specialists was 94% (95%, CI: 83% to 99%), compared with 90% of the PHP.¹⁰

Early detection of CNV and AMD management

Therapeutic management of patients with intermediate AMD includes making them aware of the potential prophylactic benefits of dietary supplements, such as those used in the Age-Related Eye Disease Study. Proper management includes relating to the patient the importance of self-monitoring and periodic exams, although these are limited in their ability to detect CNV. Other methods of detection, including optical coherence tomography and the PHP, can result in early detection and treatment and better visual outcomes.

References

Age-Related Eye Disease Study Research Group. A randomized, placebo-controlled, clinical trial of high-dose supplementation with vitamins C and E, beta carotene, and zinc for age-related macular degeneration and vision loss: AREDS report no. 8. Arch Ophthalmol. 2001;119(10):1417-1436.

Bressler NM, Bressler SB, Congdon NG, et al. Potential public health impact of Age-Related Eye Disease Study results: AREDS report no. 11. Arch Ophthalmol. 2003;121(11):1634-1636.

Miller JW. Randomized, controlled phase III study of ranibizumab (Lucentis) for minimally classic or occult neovascular age-related macular degeneration. Presented at the annual meeting of the American Society of Retina Specialists; July 18, 2005; Montreal, Quebec, Canada.

Gragoudas ES. VEGF Inhibition Study in Ocular Neovascularization-1 (VISION-1): Efficacy results from phase II/III Macugen (pegaptanib sodium) clinical trials. Presented at the annual meeting of the Association for Research in Vision and Ophthalmology; April 27, 2004; Fort Lauderdale, Fla.

D’Amico DJ, Bird AC. VEGF Inhibition Study in Ocular Neovascularization-1 (VISION-1): Safety evaluation from the pivotal Macugen (pegaptanib sodium) clinical trials. Abstract 2363, presented at the annual meeting of the Association for Research in Vision and Ophthalmology; April 27, 2004; Fort Lauderdale, Fla.

Olsen TW, Feng X, Kasper TJ, Rath PP, Steuer ER. Fluorescein angiographic lesion type frequency in neovascular age-related macular degeneration. Ophthalmology. 2004;111(2):250-255.

Maguire MG. Characteristics of choroidal neovascularization in the untreated eye of CAPT patients. Data presented at the 28^th annual meeting of the Macula Society; February 24, 2005; Key Biscayne, Fla.

Loewenstein A, Malach R, Goldstein M, et al. Replacing the Amsler grid: A new method for monitoring patients with age-related macular degeneration. Ophthalmology. 2003;110(5):966-970. Comment and author reply in Ophthalmology. 2005;112(2):357.

Preferential Hyperacuity Perimeter Research Group. Results of a multicenter clinical trial to evaluate the preferential hyperacuity perimeter for detection of age-related macular degeneration. Retina. 2005;25(3):296-303.

Preferential Hyperacuity Perimetry Research Group. Preferential Hyperacuity Perimeter (PreView PHP) for detecting choroidal neovascularization study. Ophthalmology. 2005;112(10):1758-1765.

A summary of alternative methods in detecting AMD

by Linda Christian

In addition to use of an Amsler grid and Preferential Hyperacuity Perimeter (PreView PHP, Carl Zeiss Meditec, Dublin, Calif.) technology, other methods have been proven to be effective or are undergoing research for detecting choroidal neovascularization.

Fluorescein angiography and OCT

Considered by many ophthalmologists to be the gold standard to diagnose wet age-related macular degeneration, fluorescein angiography is usually performed after a patient reports symptoms, such as distortion of vision, or when a physician suspects abnormalities in blood vessel growth. The photographs reveal any retinal defects, swelling, leaking of blood vessels or circulation problems.

Noninvasive and noncontact, optical coherence tomography provides cross-sectional images of the retina and retinal pigment epithelium (RPE)-choriocapillary complex. Although fluorescein angiography is considered by many to be the mainstay of diagnosing patients with AMD, OCT has been proven useful in assessing certain aspects of the disease, including the state of the vitreoretinal interface and sites of the attachment to the macula; intraretinal pathology in terms of thickness and presence of cysts; and subretinal space for neovascularization, geographical atrophy and the presence of blood or fluid.¹

Efficacy of fluorescein angiography/OCT

In a head-to-head comparison published last year and conducted on 10 patients with predominantly classic CNV, 10 patients with serous RPE detachment and 10 patients with geographic RPE atrophy, fluorescein angiography visualized the vascular configuration and perfusion and leakage changes and detected neovascular structures and delineated their configuration in patients with classic CNV.² In patients with serous RPE detachment, fluorescein angiography detected a pooling of extravascular fluid. In patients with geographic RPE atrophy, fluorescein angiography showed reduced choroidal perfusion.²

Using OCT, intra-RPE, subretinal and sub-RPE fluid accumulation secondary to CNV was documented. Classic CNV was represented as a hyperreflective band at the level of the RPE-choriocapillary complex, and RPE detachment was represented as a dome-shaped detachment. OCT images of patients with geographic RPE atrophy showed a loss of the RPE band and had an increased depth resolution.²

In other studies, fluorescein angiography proved less effective than OCT in identifying lesion boundaries, and neither was 100% effective.³ Pigmentary abnormalities, soft drusen and detachments of the neurosensory retina and RPE had distinct presentations on OCT.^4,5 OCT was also effective in evaluating subretinal and intraretinal fluid, assessing subfoveal involvement of CNV and monitoring CNV before and after laser photocoagulation.⁴ OCT was unable to detect CNV beneath serous RPE detachments or identify basil linear deposits.^4,5 Although CNV was evident and classic CNV membranes had well-defined boundaries, occult CNV membranes showed a less defined structure, and researchers determined that OCT cannot replace conventional diagnostic techniques.⁵

Indocyanine green angiography as a means of detection

Indocyanine green (ICG) angiography is often used in addition to fluorescein angiography, as it extends the scope of therapy by increasing occult new vessels delimitation poorly defined by fluorescein angiography. With this better choroidal visualization, ICG is an effective means of further exploring AMD.⁶ Although ICG has shown limited diagnostic value in the presence of subretinal or intraretinal bleedings, data have shown subtraction ICG is an effective clinical tool in detecting CNV concomitant with subretinal hemorrhage.^7,8 In addition, data have shown fluorescein angiography-guided ICG to be effective in the detection of feeder vessels of subfoveal CNV, minimizing the amount of ICG injected and the examination time compared to conventional ICG.⁹

Emerging methods of detection

Emerging methods of detection include dark adaptation, autofluorescence imaging and clinical tests of ocular function.

Dark adaptation is based on the hypothesis that in early AMD, the topography of rod dysfunction and loss based on dark adaptation matches the location of pathology in the RPE and Bruch’s membrane complex, possibly due to retinoid deficiency of the photoreceptors due to impaired retinoid translocation.¹⁰ In a study conducted to clarify the pathogenesis of late-onset retinal degeneration, which exhibits sub-RPE deposits similar to those of AMD, abnormal dark adaptation mirrored the regional distribution of the sub-RPE deposits in the eye.¹¹ Further research is needed to evaluate the efficacy of dark adaptation in identifying AMD.

Taken with a confocal scanning laser ophthalmoscope, fundus autofluorescence images may offer physicians a means to detect abnormal phenotype in retinal disease when other techniques might not.¹² Autofluorescence, implied to be derived from lipofuscin at the RPE level, occurs in the presence of retinal photoreceptor loss, which would result in abnormal autofluorescence levels over lesions associated with AMD.¹³ In an analysis of autofluorescence characteristics in 65 eyes with CNV at various stages of evolution, autofluorescence characteristics were used to effectively discern and study patients with recent-onset CNV from patients already diagnosed with CNV and with late-stage CNV.¹⁴ However, more research is needed to evaluate the efficacy of autofluorescence in identifying and increasing our understanding of CNV, although studies have shown a classification system based on patterns of abnormal autofluorescence can be applied to CNV and used to identify risk factors for AMD or monitoring therapies.^15,16

Researchers are also studying whether clinical tests of ocular function and macular appearance can help predict the development of CNV. Studies have shown that a slower recovery from glare and more extensive fundoscopic changes appear to be independent risk factors for the development of a CNV membrane in eyes of patients with unilateral AMD, and a slower foveal electroretinogram implicit time may be a sign of early CNV membrane development or abnormal choroidal perfusion.^17,18 Further research is needed to determine what implications this may have.

References

Rashed H, Izatt J, Toth C. Optical coherence tomography of the retina. Optics & Photonics News. 2002;13(4):48-51.

Ahlers C, Michels S, Elsner H, Birngruber R, Pruente C, Schmidt-Erfurth U. Topographic angiography and optical coherence tomography: A correlation of imaging characteristics. Eur J Ophthalmol. 2005;15(6):774-781.

Kim SG, Lee SC, Seong YS, Kim SW, Kwon OW. Choroidal neovascularization characteristics and its size in optical coherence tomography. Yonsei Med J. 2003;44(5):821-827.

Hee MR, Baumal CR, Puliafito CA, et al. Optical coherence tomography of age-related macular degeneration and choroidal neovascularization. Ophthalmology. 1996;103(8):1260-1270.

Spraul CW, Lang GE, Lang GK. Value of optical coherence tomography in diagnosis of age-related macular degeneration. Correlation of fluorescein angiography and OCT findings [article in German]. Klin Monatsbl Augenheilkd. 1998;212(3):141-148.

Mauget-Faysse M. Indocyanine green angiography and occult neovascularization in age-related macular degeneration [article in Spanish]. J Fr Ophtalmol. 2001;24(4):401-410.

Schutt F, Fischer J, Kopitz J, Holz FG. Indocyanine green angiography in the presence of subretinal or intraretinal haemorrhages: Clinical and experimental investigations. Clin Experiment Ophthalmol. 2002;30(2):110-114.

Matsumoto M. Shiraki K, Obana A. Detection of choroidal neovascularization by subtraction indocyanine green angiography. Osaka City Med J. 2003;49(2):85-91.

Yanagi Y, Tamaki Y, Sekine H. Fluorescein angiography-guided indocyanine green angiography for the detection of feeder vessels in subfoveal choroidal neovascularization. Eye. 2004;18(5):474-477.

Jackson GR, Owsley C, Curcio CA. Photoreceptor degeneration and dysfunction in aging and age-related maculopathy. Ageing Res Rev. 2002;1(3):381-396.

Milam AH, Curcio CA, Cideciyan AV, et al. Dominant late-onset retinal degeneration with regional variation of sub-retinal pigment epithelium deposits, retinal function, and photoreceptor degeneration. Ophthalmology. 2000;107(12):2256-2266.

von Ruckmann A, Fitzke FW, Bird AC. Distribution of fundus autofluorescence with a scanning laser ophthalmoscope. Br J Ophthalmol. 1995;79(5):400-401.

von Ruckmann A, Fitzke FW, Bird AC. Fundus autofluorescence in age-related macular disease imaged with a laser scanning ophthalmoscope. Invest Opthalmol Vis Sci. 1997;38(2):478-486.

Dandekar SS, Jenkins SA, Peto T, et al. Autofluorescence imaging of choroidal neovascularization due to age-related macular degeneration. Arch Ophthalmol. 2005;123(11):1507-1513.

Blindewald A, Bird AC, Dandekar SS, et al. Classification of fundus autofluorescence patterns in early age-related macular disease. Invest Ophthalmol Vis Sci. 2005;46(9):3309-3314.

Trieschmann M, Spital G, Lommatzsch A, et al. Macular pigment: Quantitative analysis on autofluorescence images. Graefes Arch Clin Exp Ophthalmol. 2003;241(12):1006-1012.

Sandberg MA, Weiner A, Miller S, Gaudio AR. High-risk characteristics of fellow eyes of patients with unilateral neovascular age-related macular degeneration. Ophthalmology. 1998;105(3):441-447.

Remulla JF, Gaudio AR, Miller S, Sandberg MA. Foveal electroretinograms and choroidal perfusion characteristics in fellow eyes of patients with unilateral neovascular age-related macular degeneration. Br J Ophthalmol. 1995;79(6):558-561.