Diminished Foveal Sensitivity May Predict the Development of Advanced Age-related Macular Degeneration

Diminished Foveal Sensitivity May Predict the Development of Advanced Age-related Macular Degeneration JANET S. SUNNESS, MD, ROBERT W. MASSOF, PhD, MARY A. JOHNSON, PhD, NEIL M. BRESSLER, MD, SUSAN B. BRESSLER, MD, STUART L. FINE, MD

Abstract: Visual function testing was performed on one eye with drusen from each of 18 elderly patients in 1984. Eleven patients had advanced age-related macular degeneration (AMD) in the fellow eye, and seven had only bilateral drusen. These patients were all followed prospectively (median, 45 months), at which time one eye had a new vessel membrane, three eyes had pigment epithelial detachments, and one eye had geographic atrophy. Only two of these five eyes had AMD-related visual loss in the fellow eye in 1984. The degree of loss of foveal dark-adapted sensitivity in 1984 predicted which patients developed advanced AMD with 100% sensitivity and 92% specificity. The presence of high-risk drusen characteristics in 1984 predicted the development of advanced AMD with 100% sensitivity but only 55% specificity. For this small group of patients, foveal dark-adapted sensitivity loss was an excellent predictor of the subsequent development of advanced AMD in eyes with drusen. A prospective study of a large group of patients with drusen is being undertaken to validate this finding. Ophthalmology 96:375-381, 1989

Age-related macular degeneration (AMD) is the leading cause of new legal blindness in the United States population over 60 years of age. 1 Laser treatment may reduce or slow the rate of development of severe central visual loss associated with well-defined extrafoveal choroidal neovascular membranes (CNVMs).2 There are currently no other treatments available for the many other forms of AMD causing visual loss, including poorly defined Originally received: September 8, 1988. Revision accepted: November 14, 1988. From the Laboratory of Physiological Optics and the Retinal Vascular Center, The Wilmer Institute. The Johns Hopkins University School of Medicine. Baltimore. Presented at the American Academy of Ophthalmology Annual Meeting, Las Vegas. October 1988. Supported by National Eye Institute grant EY05675 and Physician Scientist award EY00259 (Dr. Sunness). Reprint requests to Janet S. Sunness, MD. Wilmer B-27. The Johns Hopkins Hospital, 600 N. Wolfe St, Baltimore. MD 21205.

CNVMs, well-defined CNVMs involving the foveal region, retinal pigment epithelial (RPE) detachments, and geographic atrophy of the RPE. This greatly limited therapeutic armamentarium forces us to focus our attention on preventing the development of these advanced stages ofAMD. In order to target patients for careful monitoring or for trials of preventive therapy, one must first be able to estimate the risk of advanced AMD developing for each patient. Although it is known that drusen are present in eyes with vision loss from AMD,3 they are present in at least 25% of patients older than 60 years of age by ophthalmoscopy,l and in the entire elderly population by histopathologic study.4 In contrast, only 0.5% of these patients will have severe visual loss associated with AMD.l Thus, the presence of drusen alone does not separate patients at risk for visual loss associated with AMD. A number of studies have examined, in a retrospective3,5-7 or cross-sectiona1 8 fashion, drusen characteristics which appear to be associated with the subsequent development of AMD-related visual loss, and the drusen characteristics

375

OPHTHALMOLOGY

•

MARCH 1989 •

present in one eye of patients whose fellow eyes have AMD-related·visualloss. Some characteristics of drusen, such as large size, confluence, softness, large number, and focal hyperpigmentation have been found to be related to the risk of development of exudative AMD (CNVMs and RPE detachments). However, this sort of analysis is still not adequately informative, and one cannot predict with any certainty whether a patient with certain drusen characteristics will have advanced AMD associated with visual loss. For this reason, we began looking at certain aspects of visual function in patients with drusen and AMD. We tested a group of eyes with drusen at the beginning of 1984 and found that all drusen eyes, despite good acuity, had a reduction in dark-adapted foveal sensitivity compared with young normals. 9 There was a wide range of magnitude of sensitivity loss, which suggested to us on a statistical basis that perhaps we were detecting a small age-related sensitivity loss in most patients, with some patients having additional dysfunction which could not be detected on the basis of fundus examination alone. 10 We measured retinal sensitivity over drusen and non drusen areas at equal distance from the fovea using our fundus camera stimulator, and determined that drusen are not directly responsible for the loss in sensitivity. I I These studies suggested to us that those patients with large losses of sensitivity might have "subclinical pathology" which could possibly be a harbinger of the development of AMD and vision loss. We have been able to obtain full follow-up of our patients a median 45 months later. Our results suggest that for this small group of patients, the magnitude of darkadapted foveal sensitivity was a very good predictor of the subsequent development of advanced AMD.

SUBJECTS AND METHODS 1984 STUDY

In 1984, patients were identified from the files of the Retinal Vascular Center of the Wilmer Ophthalmological Institute. Patients identified as having drusen or AMD on the chart were selected if they met the following criteria: visual acuity of 20/50 or better in at least one eye, fundus manifestations of AMD limited to drusen only, no other retinal disease or significant media opacity, 55 years of age or older, residence near Baltimore, and able to cooperate for vision function testing. Eighteen patients who met these criteria were tested between January and March 1984. All 18 patients had only drusen in at least one eye. Written informed consent was obtained from all patients. Of the 18 patients studied, 7 had drusen only in both eyes, and 11 had drusen only in one eye and AMD-related visual loss in the fellow eye. Seven of the fellow eyes had CNVMs, three had RPE detachments, and one had geographic atrophy. Best-corrected visual acuity ranged from 20/16 to 20/50 in the drusen eye tested (median visual acuity, 20/25). The age range for this group was 57 to 81 years (median, 69 years). Table 1 gives clinical information for each patient individually. 376

VOLUME 96

•

NUMBER 3

Foveal dark-adapted sensitivity was measured in a drusen only eye of each patient. For those patients with bilateral drusen, the eye with better acuity was selected, and if both eyes had equal acuity an eye was selected at random. The pupil of each eye was dilated with 1% tropicamide. The eye to be tested was patched for 1 hour to achieve full dark adaptatiop.. Absolute sensitivity was measured on the Tubinger perimeter for a 1.8° red (656nm) 500-msec stimulus centered at the fovea. (This measurement was part of a more comprehensive study of retinal function which included static perimetry of the full horizontal meridian with red and blue-green lights, scotopic electroretinogram, intensity-response functions, flicker electroretinogram phase functions, and electrooculograms.) These results have been presented previously.9 Absolute sensitivity was obtained using a method of ascending limits. Three separate measurements were made at the foveal location, and these were averaged for the final foveal sensitivity measure. The sensitivity value indicates the amount of neutral density filter interposed, so that the higher the sensitivity the more dense the filter and the more dim the light seen. Absolute sensitivity is expressed in decibels (dB), relative to a maximum luminance (0 dB) of 8.4 cd/m2. After visual function testing, each patient underwent a slit-lamp examination and direct and indirect fundus examination. Fundus photographs were obtained for each patient. FOLLOW-UP

For the purpose of this study, we chose as an endpoint the development of one of the three severe vision-threatening aspects of AMD: CNVMs, RPE detachments, or geographic atrophy. We will refer to this endpoint as advancedAMD. In October 1987, all participants in the 1984 study were asked to return to Wilmer for a follow-up evaluation; 11 returned. Five additional patients were willing to participate but could not attend because of illness or difficulties in traveling, and these patients were therefore examined in their homes by one of us (JS). One patient whose status had remained stable did not wish to participate, and information was obtained from her ophthalmologist. One patient died, and information about his most recent examination was obtained from his ophthalmologist. Thus, follow-up was obtained for all patients. Duration offollowup was 34 months for the deceased patient, and ranged from 42 to 50 months for the remaining patients (overall median follow-up time, 45 months). On follow-up examination, written informed consent was obtained. Visual acuity with correction and pinhole was measured, and dilated direct and indirect fundus examination was performed on all patients. For those patients who returned for follow-up, slit-lamp examination and fundus photographs were obtained. DRUSEN RISK CLASSIFICATION

In 1988, the original 1984 photographs for each study eye were reviewed by two independent fundus photograph

SUNNESS et al

•

PREDICTING AGE-RELATED MACULAR DEGENERATION

Table 1. Individual Patient Data Age (yrs)

Visual Acuity in 1984

Patient No_

Sex/Eye

1 2 3 4 5

F/OS F/OS M/OD F/OD M/OD

74 69

72

20/25 20/20 20/25 20/30 20/30

6 7 8 9 10 11 12 13 14

M/OS F/OD F/OD F/OD M/OD F/OD F/OS F/OD F/OS

64 57 67 73 71 72 64 80 69

15 16 17 18

M/OD F/OD F/OS F/OD

74 69 66 81

57 57

Fellow Eye

Foveal Sensitivity* (dB)

Drusen Grade

Follow-up (mos)

Final Visual Acuity

Outcome

28 32 32 30 25

0 0

44 47 44 45 45

20/25 20/20 20/30 20/30 20/30

RTS RTS RTS RTS RTS

20/25 20/16 20/50 20/16 20/30 20/25 20/25 20/20 20/25

Drusen Drusen Drusen Drusen Geographic atrophy NVM NVM NVM PED NVM NVM PED NVM Drusen

30 34 31 32 31 29 36 34 27

6 1 2 5 1 5

6

34 48 44 47 48 43 47 47 43

20/20 20/25 20/25 20/20 20/20 20/25 20/25 20/50t 20/120

20/30 20/30 20/25 20/25

Drusen Drusen PED NVM

25 27 18 26

6 5 6+* 3

45 50 45 42

2/200 20/40 20/40 20/100

RTS RTS RTS RTS RTS RTS RTS RTS Geographic atrophy PED PED PED NVM

2 4 6

Poor photograph Poor photograph

OS = lef~ eye; RTS = remained the same; OD = right eye; NVM = choroidal neovascular membrane; PED = retinal pigment epithelial detachment * Relative log foveal sensitivity. t Lens changes developed, account for the drop in visual acuity. * Fundus photograph in 1984 suspicious for a pigment epithelial detachment

graders (NB and SB), who were masked as to the identity of each patient, the status of the fellow eye, the visual acuity, and the results of the visual function testing. The photographs were graded using a modified method that has been described recently. 12 The photographs were first assessed as to their quality and adequacy. The presence of exudative disease or atrophy was noted. Drusen within 1500 ~m of the fovea (inner zone) and drusen between 1500 and 3000 ~m from the fovea (outer zone) were assessed separately. For each zone, the number, size, and confluence of the drusen, and the presence of focal hyperpigmentation of the RPE were evaluated. A risk factor grade from 0 to 6 was assigned to each fundus based on the number and nature of risk factors present. The photographs of two patients were graded as poor quality and these patients could not be assigned a risk factor grade. Two patients received a rating of 0 (no drusen ::> 125 ~m in diameter), two received a 1 (two with no drusen> 125 ~m, but one had pigment in the outer zone and one had confluence in the inner zone), two received a 2 (large drusen in the inner zone, and moderate confluence), one received a 3 (large drusen in the inner zone, and extensive confluence), one received a 4 (large drusen in the inner and outer zones, moderate confluence in the inner and outer zones), three received a 5 (large drusen in the inner and outer zones, extensive confluence in the inner zone and moderate confluence in the outer zones), and four received a 6 (large drusen in the inner and outer zones, and extensive confluence in the inner and outer

zones). In addition, one patient's photographs were suspicious for the presence of exudative disease.

RESULTS Table 1 gives the results of follow-up examination for each patient. At follow-up, five patients had advanced AMD in the eye with only drusen when tested in 1984. (This will be called the advanced AMD group.) Patient 18 had a new vessel membrane, patients 15, 16, and 17 had RPE detachments, and patient 14 had geographic atrophy. Thirteen patients remained essentially the same with only drusen, and at most mild pigmentary change and minimal atrophy. These patients continued to have acuity in the 20/20-to-20/30 range, with the exception of one patient whose lens opacity accounted for a visual acuity of20/50. (This will be called the unchanged group, in the sense that they did not change in their AMD status.) Figure 1 shows the fundus appearance in 1984 and in 1988, and the fluorescein angiographic appearance in 1988, of the eye which had a new vessel membrane over the follow-up period (patient 18). In 1984, the fundus photograph showed large drusen and extensive confluence in the inner zone, which received a drusen grade of 3. In 1988, a poorly defined new vessel membrane is present, with lipid and hemorrhage seen on the fundus photograph, and early hyperfluorescence and late leakage on the flu-

377

OPHTHALMOLOGY

•

MARCH 1989

•

VOLUME 96

•

NUMBER 3

Fig 1. Patient 18, right eye. Top left. in 1984, large confluent drusen were present in the region within 1500!lm of the foveal center. Second row left. in 1988, a new vessel membrane was present, as evidenced by the presence of subretinal lipid and blood. Third row left and bottom left. early and late frames of fluorescein angiogram in 1988. A poorly defined new vessel membrane is seen, with early hyperfluorescence and continued late leakage. Fig 2. Patient 15, right eye. Top right. in 1984, large drusen with a high degree of confluence were detected within 1500 !lm of the fovea, and also between 1500 and 3000 !lm of the fovea. Second row right. in late 1985, a pigment epithelial detachment was seen. Third row right and bottom right. early and late frames of fluorescein angiogram in 1985. A typical pigment epithelial detachment is noted, with early and sustained hyperfluorescence of a circumscribed area.

orescein angiogram. The presence of a membrane had first been noted in 1987, 39 months after the visual function testing. Figure 2, top right, shows the fundus appearance in 1984 of the right eye of patient 15. The fundus photograph showed large drusen in the inner and outer zones, and extensive confluence in the inner and outer zones, for a 378

drusen grade of 6. Figure 2, second row right to bottom right, taken 20 months later, shows the typical fundus and angiographic appearance of an RPE detachment at the time it was first diagnosed. A dense vitreous hemorrhage developed in this patient shortly before his 1988 follow-up visit, believed to be on the basis of his advanced AMD.

SUNNESS et al

•

PREDICTING AGE-RELATED MACULAR DEGENERATION

The ages of those patients who developed advanced AMD ranged from 66 to 81 years, so that there>was considerable age overlap and no significant difference in age between the advanced AMD and unchanged groups by t test. Three of the patients with advanced AMD in the tested eye had a visual acuity of 20/25 in that eye at the time of initial examination and two had a visual acuity of 20/30 at that time. This did not differ from the group which remained the same (t = 1.039, df= 16, P = 0.314). Two of the five patients (40%) in the advanced AMD group had exudative AMD in the fellow eye in 1984, whereas 8 (62%) of the 13 patients who remained unchanged had exudative AMD in the fellow eye. One patient in the unchanged group had geographic atrophy in the fellow eye in 1984. Thus, the status of the fellow eye did not help in differentiating which patients would go on to advanced AMD in the eye with only drusen. Figure 3 shows, for each outcome group, the foveal dark-adapted sensitivities measured in 1984. The group with advanced AMD had significantly lower foveal darkadapted sensitivities than the group of patients who remained the same (mean advanced group, 24.6 dB; mean unchanged group, 31.1 dB; t = 3.97; df= 16; P = 0.001). There was only minimal overlap between the groups, in that there was one patient who remained unchanged who had a foveal sensitivity within the range of the group with subsequent development of AMD. Choosing an optimal criterion level of foveal sensitivity, this visual function measure achieved a sensitivity of 100% and specificity of 92% in predicting the development of advanced AMD. Figure 4 shows, for each outcome group, the drusen risk classification based on the 1984 photographs. The 16 sets offundus photographs graded included all the patients who developed AMD visual loss. The most severe group (group 6 plus suspicious case) included three of the patients who had advanced AMD, but also two patients who did not. Including group 5 with the more severe group, four of the new advanced AMD eyes are included, but also four eyes that did not develop advanced AMD. In order to identify all five eyes which developed advanced AMD, all eyes with grade 3 and worse would be included, which also includes five eyes without advanced AMD. Thus, to achieve 100% sensitivity, one must accept a specificity of only 55%. To achieve a maximal specificity of 82%, one has a sensitivity of only 60%.

---"e

40

~

'-'

35

..

30

ClJ

25

.... :.: .:: .~

=

00

.

eo

o

•

-; ClJ

0

20

o

~

I

I

Advanced AMD

No Change

15

Outcome Fig 3. Dark-adapted foveal sensitivity, measured in 1984, displayed with respect to diagnosis at the 1988 follow-up visit. Closed circle represents eye with geographic atrophy, open circles represent eyes with RPE detachments, closed square represents eye with CNVM, and dotted open squares represent eyes remaining with only drusen. The five eyes with new advanced AMD had significantly lower foveal sensitivities when tested in 1984, at a time when these eyes had only drusen, than the eyes which remained unchanged in their drusen status over the follow-up period. There is only one point of overlap between the groups.

high-risk drusen characteristics, so that the eyes tested, as a group, were not representative of the general elderly population with drusen. The optimal criterion level chosen for defining risk of advanced AMD based on foveal sensitivity was derived from this data set, and could change as more patients are tested. We are now in the process of

ClJ

"e

= \.0

~

= = ClJ

(IJ

\.0

6+

0

6

eo

1:11:1

5

0

1:11:1

4 r 3

2

1:1

• 1:11:1

~

DISCUSSION

Illlll

0

Illlll

Advanced AMD

For this small group ofpati~nts followed for a median of 45 months, foveal dark-adapted sensitivity to a 1.8 0 red target was an excellent predictor of the subsequent development of advanced AMD in eyes with drusen. It performed better than knowledge offellow eye status, initial visual acuity, and evaluation of drusen risk characteristics in separating the group destined to go on to the vision-threatening aspects of AMD. These results, though quite promising, must be interpreted with a good deal of caution. The number of patients tested and followed was small. Many of the patients had

No Change

Outcome Fig 4. Drusen grade, based on 1984 fundus photographs, displayed with respect to diagnosis at the 1988 follow-up visit. Closed circle represents eye with new geographic atrophy, open circles represent eyes with new RPE detachments, closed square represents eye with new CNVM, and dotted open squares represent eyes remaining with only drusen. Drusen grade 6+ represents a photograph suspicious for the presence of exudative disease. All the eyes with advanced AMD had drusen grades of 3 or greater, but so did 5 of the 11 eyes unchanged in their AMD status. (Two eyes are not included because photographic quality did not permit adequate grading.)

379

OPHTHALMOLOGY

•

MARCH 1989

•

VOLUME 96

•

NUMBER 3

methods that potentially could be performed in a general ophthalmologist's office. The exciting aspect of this measure is its potential to provide additional and independent information not available from evaluation of drusen characteristics of fundus photographs. In particular, the fact that our measure detected the patient who went on to develop a CNVM, who was not at the highest risk on the basis of the photographic classification, suggests that this test might be useful for screening elderly patients in the general population for the risk of vision-threatening AMD developing. Testing of a larger group of patients with follow-up over time should provide an answer.

testing a large group of patients with good acuity and drusen, whom we will then follow prospectively to validate our finding of the value of dark-adapted foveal sensitivity. An interesting feature of our study is that three of the five patients had the RPE detachment form of advanced AMD. Because there is a continuum between soft confluent drusen and RPE detachments, one might argue that in fact what was being achieved by the foveal sensitivity measure was an early detection of an RPE detachment rather than prediction of something which had not yet evolved. This might be particularly true of patient 17, who carried a diagnosis of drusen in the tested eye when seen by us, but masked evaluation of the fundus photographs was suspicious for an RPE detachment already being present. However, even if what is being detected is early disease, the test is still providing helpful information because four other patients in drusen photograph risk categories equal to those patients with RPE detachments did not go on to develop advanced AMD. If one excludes the patient with fundus photographs suspicious for an RPE detachment from this study, one still achieves a sensitivity of 100% using both visual function and fundus photograph criteria, with specificities of 92% for the visual function criterion and 55% for the fundus photograph criterion. A third consideration is that it is known that both AMD and high-risk drusen characteristics are more common in the more elderly patients,5,8 so that in any study of AMD one must watch for confounding age effects. In this study, the patient who developed a CNVM was 81 years of age and was the oldest patient in our study. However, within our group of elderly patients there was not a significant correlation offoveal sensitivity with age (correlation coefficient = 0.13), and the next oldest patient, who was 80 years of age, had among the highest sensitivity values we measured. It is possible that if fluorescein angiography had been performed at the time of the initial visit, it might have facilitated the identification of eyes at high risk for the development of advanced AMD. However, fluorescein angiograms were not available for most of the patients studied, and no angiograms had been performed within 6 months before the initial visit for any of the patients who went on to develop advanced AMD. The goal of the study was to compare visual function performance with evaluation of fundus characteristics, both noninvasive

REFERENCES 1, Leibowitz HM, Krueger DE, Maunder LR, et al. The Framingham Eye Study Monograph. Surv Ophthalmol1980; 24(Suppl):335-610. 2. Macular Photocoagulation Study Group. Argon laser photocoagulation for senile macular degeneration: results of a randomized clinical trial. Arch Ophthalmol1982; 100:912-8. 3. Gass JDM. Drusen and disciform macular detachment and degeneration. Arch Ophthalmol1973; 90:206-17. 4. Coffey AJH, Brownstein S. The prevalence of macular drusen in postmortem eyes. Am J Ophthalmol1986; 102:164-71. 5. Smiddy WE, Fine SL. Prognosis of patients with bilateral macular drusen. Ophthalmology 1984; 91:271-7. 6. Gregor Z, Bird AC, Chisholm IH. Senile disciform macular degeneration in the second eye. Br J OphthalmoI1977; 61:141-7. 7. Strahlman ER, Fine SL, Hillis A. The second eye of patients with senile macular degeneration. Arch Ophthalmol 1983; 101: 1191-3. 8. Bressler NM, Bressler SB, Seddon JM, et al. Drusen characteristics in patients with exudative versus non-exudative age-related macular degeneration. Retina 1988; 8:109-14. 9. Sunness JS, Massof RW, Johnson MA, et al. Peripheral retinal function in age-related macular degeneration. Arch Ophthalmol 1985; 103: 811-6. 10. Massof RW, Choy D, Sun ness JS, et al. Foveal threshold elevations associated with age-related drusen. Clin Vis Sci (in press). 11. Sun ness JS, Johnson MA, Massof RW, Marcus S. Retinal sensitivity over drusen and nondrusen areas: a study using fundus perimetry. Arch Ophthalmol1988; 106:1081-4. 12. Bressler SB, Bressler NM, West S, et al. Senile eye changes: the. grading and prevalence of macular degeneration. ARVO Abstracts. Invest Ophthalmol Vis Sci 1987; 28(Suppl):106.

Discussion by

John R. Heckenlively, MD The authors have presented longitudinal data from a group of patients with early age-related macular degeneration (AMD) evaluating whether foveal sensitivity testing has predictive value in forecasting severe macular degeneration. The endpoints which the authors regarded as significant signs of severe macular deFrom the Jules Stein Eye Institute, UCLA School of Medicine, Los Angeles.

380

generation were: focal retinal pigment epithelial (RPE) detachments, subretinal neovascularization, and geographic atrophy. Originally, the authors selected 18 patients who could be reliably followed, Seven of these had less severe disease in that their eyes only had bilateral drusen; II others had age-related visual loss in the fellow eye. Foveal dark-adapted sensitivity testing and fundus photographs were obtained initially and fundus evaluation alone was performed approximately 3 years later.

SUNNESS et al

•

PREDICTING AGE·RELATED MACULAR DEGENERATION

The final macular status was compared with the initial foveal sensitivity and amount of d r u s e n . , The authors found that the five patients whose macular tests showed the poorest thresholds of sensitivity to the red stimulus were the ones who had the most severe macular problems. Unfortunately, the lower foveal sensitivity was not specific for the appearance of subretinal neovascularization, thus this test was not predictive in helping to catch this potentially treatable complication of AMD. The results suggest that patients with mild macular degeneration who later had severe macular problems can be reliably detected early with the foveal sensitivity test. With further research, this finding may prove helpful in developing new modalities of therapy for these age-related diseases. It should be noted that the authors were using focal psychophysical testing and the cone and rod systems from all the retina were not evaluated. I have had the opportunity to study a number of patients in the senior citizen age range who had complaints of adaptation problems or whose poor visual acuity was not explained by the macular appearance. On electroretinographic testing, many of these patients had partial or full cone dysfunction. Here are several examples: A 58-year-old woman had a 10year history of night blindness and photosensitivity. Her visual acuity was 20/100 in the right eye and 20/30 in the left, and she had numerous irregularly shaped drusen of the posterior pole. On electroretinographic (ERG) testing, she had a severe cone degeneration with a flat photopic ERG, whereas the rod ERG was excellent. Her electrooculogram results, visual field, and

final rod threshold values were within normal limits. The ERG tracing demonstrated no cone response, whereas the scotopic responses were fully recordable. Her nyctalopia was a form of urban night blindness and was due to a lack of cone dark adaptation in the partial darkness characteristic of the city environment. Another example is that of a 71-year-old patient who had had a successful cataract extraction but who complained of problems adapting to darkened areas. Although her macular areas were flat in appearance, they were otherwise unremarkable. On ERG testing, she was found to have an unsuspected partial cone degeneration with a 50% reduction of her photopic b-wave amplitudes. The diagnosis of cone degeneration or dysfunction frequently is difficult to make, particularly in adult-onset cases because fundus changes may be difficult to find. Occasionally, the patients will have subtle foveal atrophy as in these two cases, or may have drusen or RPE changes associated with age effects. They are likely to have severe cone dysfunction on ERG testing. In summary, the authors have reported that foveal sensitivity testing appears to be an accurate predictor of severe macular degeneration in eyes with mild changes that were retested 3 years later. Foveal sensitivity testing may be useful in helping to assess research studies in the treatment of macular disease. It should be remembered that foveal sensitivity testing does not give direct information about the cone system which can be abnormal in some cases of macular degeneration and must be tested by using electoretinography.

381