Static and Flicker Perimetry in Age-Related Macular Degeneration

Abstract

Purpose.: The relationship between clinical severity of age-related macular degeneration (AMD) and macular function has not been well established. In this study, we investigated the correlation between clinical severity and functional deficits as detected by static and flicker perimetry.

Methods.: This cross-sectional study consisted of 279 AMD subjects and 24 control participants. AMD subjects were allocated into 1 of 10 AMD severity groups depending on the status of the designated study eye and the fellow eye, as assessed by color fundus photographs. Visual acuity, and static and flicker perimetry were tested on one eye during the same session. The geometric means, SDs, and percentage of abnormal eyes of static and flicker sensitivity of each AMD severity group were determined and compared.

Results.: The pattern of change in sensitivity and percentage of abnormal eyes for static perimetry across all AMD severity groups were similar to flicker perimetry. Eyes with drusen > 125 μm (P[static] = 0.018, P[flicker] = 0.024), drusenoid epithelial detachment (P[static and flicker] < 0.001) and noncentral geographic atrophy (GA; P[static and flicker] < 0.001) had significant reductions in static and flicker sensitivities compared to normal eyes. Static (β-coefficient −1.59, 95% confidence interval [CI] −4.78–1.60) and flicker (β-coefficient −1.29, 95% CI −4.66–2.08) sensitivities declined at a similar rate in eyes that showed clinical signs of progression.

Conclusions.: Static and flicker perimetry were affected similarly across the spectrum of AMD severity, and methods appeared to be valid techniques for assessing retinal sensitivity in AMD once drusen > 125 μm are present, but before the development of late AMD.

Introduction

The clinical severity of age-related macular degeneration (AMD) currently is determined on the basis of the fundus appearance. Fundus signs, such as drusen type, size and number, and the presence or absence of pigmentary abnormalities, are considered the hallmarks of the early stages of AMD. These clinical features are used to determine a severity scale of AMD, which then is used to predict the risk of progression to the advanced forms of AMD; geographic atrophy (GA) and choroidal neovascularization (CNV). ¹ However, monitoring only these clinical changes is not sufficient as an outcome measures in the era where we wish to determine if novel interventions, used early in the disease, can prevent progression to vision loss. There is an urgent need for other markers of disease severity that can be monitored to follow disease progression. 

Visual acuity traditionally has been used in clinical trials as the only measure of macular function. However, it is well established that visual function, as determined by a variety of psychophysical tests, such as perimetry, ^2–5 dark adaptation, ^6,7 and fine matrix mapping, ⁸ is reduced in patients with early stages of AMD. These functional changes occur before a loss in visual acuity or progression to advanced disease is detected. Efforts are being made to determine which of these tests might be useful markers of early disease, which can be used to monitor progression toward vision loss. These functional biomarkers might provide additional endpoints for new intervention trials in the early stages of AMD. 

Flicker perimetry has been a functional measure evaluated by many, including our group, where the flicker sensitivity within the central 6° has been found to be reduced in the early stages of AMD. ^3,9–13 In a large cross-sectional study we have found that the sensitivity to an achromatic flickering stimulus was a clinically applicable test and was reduced in patients with the early stages of AMD, with further reduction in sensitivity observed as AMD severity increased. ¹⁴ Recently, we also have shown that flicker sensitivities also can be used to monitor the rate of progression with time and predict the development of late complications, especially GA and to a lesser extent CNV. ¹⁵  

It has been suggested that retinal responses to flickering stimuli require much greater local metabolic demand compared to the responses to static stimuli. ¹⁶ In AMD, it is believed that thickening of Bruch's membrane creates a diffusion barrier between the choroid and retina, leading to a decrease in the metabolic supply to the retina. Thus, it has been suggested that the sensitivity of flickering stimuli would decrease earlier than would static stimuli in AMD. ³ However, to our knowledge a direct comparison has not been made between these two stimulus paradigms in a large sample nor in a longitudinal study. In our study, we investigated the ability of static and flicker perimetry to detect changes in sensitivity over the spectrum of AMD severity groups and over time. 

Methods

Participants

Individuals with AMD were recruited from the Royal Victorian Eye and Ear Hospital clinics, and from private ophthalmology practices. Control subjects were recruited from unrelated family members and friends of the cases. The study was approved by the Human Research Ethics Committee of the Royal Victorian Eye Hospital and conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants. 

The inclusion criteria for the study eye were best corrected visual acuity of 20/60 or better and any early stage of AMD (including noncentral GA). ¹⁷ If both eyes satisfied the inclusion criteria, the eye with more advanced fundus changes was designated the study eye. If the changes in fundi were similar, then the eye with better visual acuity was chosen. In cases when visual acuity and the fundus status were the same, the right eye was assigned as the study eye. Clinical status of the fellow eye ranged from presence of any signs of the early stages of AMD through to late stage AMD (central GA or CNV). Control participants were of similar age range, and had normal fundi and a visual acuity of 20/20 or better. The eye with better visual acuity was designated the study eye. If acuity was the same in both eyes then the right eye was chosen. 

Exclusion criteria included presence of cataract, glaucoma, amblyopia, color deficiency, systemic hypertension uncontrolled by medication (systolic > 150 and diastolic > 90 mm Hg), or presence of physical or mental impairment, neurologic or systemic diseases, and medication that might alter retinal function. 

All participants underwent a standardized examination procedure, including assessment of best corrected visual acuity (BCVA) using 4-meter logMAR chart, clinical eye examination using slit-lamp biomicroscopy of the anterior and posterior segments, digital photography using a nonmydriatic nonstereoscopic 45° retinal camera (Canon CR6-45NM Retinal Camera; Canon, Saitama, Japan), and interview regarding history and current symptoms of eye diseases, medical history, and medication. When needed, an additional examination in the retinal clinic, including fluorescein angiography, was undertaken to verify the presence of CNV. Perimetry tests were conducted only in the study eye, before any imaging or clinical examination was performed. 

Static and flicker perimetry was performed in all patients at baseline. Participants with any early stage of AMD in the study eye were invited to attend 6-monthly repeat testing as part of a longitudinal component to the study. Participants who returned for at least 2 consecutive 6-monthly visits (equivalent to 1 year follow-up) were included in our analysis for the longitudinal component of the study. 

Fundus Grading

The fundus images were graded according to the International Classification and Grading System for AMD ¹⁷ by two independent graders using “OptoMize PRO” software from Digital Healthcare Image Management System (Digital Healthcare Ltd., Cambridge, UK). Where disagreement occurred between the two graders, the results were adjudicated by a senior retinal specialist (RHG). The data on grading agreement between the two graders have been reported previously. ¹⁸  

The grading classification was based on the morphologic fundus changes with drusen defined by their size, such as ≤63 μm (small), between 63 and 125 μm (intermediate), and >125 μm (large). Pigmentary abnormality was defined as either increased (black or grey patches) or decreased (whitish patches) pigmentation of the RPE. Drusenoid pigment epithelial detachment (DPED) was defined as areas of elevated central RPE caused by the confluence of soft drusen. Geographic atrophy was defined as sharply delineated areas of RPE hypopigmentation, larger than 175 μm with visible choroidal vessels in its base and was either noncentral GA where the atrophy was located outside the central field of 1000 μm diameter and not considered advance AMD, or central GA, which was one of the two forms of advanced AMD. Choroidal neovascularization included serous detachment of the sensory retina, or hemorrhagic or serous pigment epithelial detachment. Any suggestion of CNV was confirmed by fluorescein angiography. A disciform scar was referred as a well defined stable scar. 

Perimetry

Static and flicker perimetry was performed with an automated perimeter (model M-700; Medmont International Pty Ltd., Vermont, Victoria, Australia) as described previously. ^3,15 The sensitivity of the Medmont automated perimeter has been reported to be comparable to that of the Humphrey field analyser. ¹⁹ The M-700 is a bowl perimeter that uses light-emitting diodes (LEDs) of 565 nm wavelength. The bowl has a background luminance of 3.2 cd·m⁻² and a maximum spot luminance of 320 cd·m⁻². The LEDs subtend 0.43° (Goldmann size III) and are arranged concentrically at various eccentricities from 1° to 50°. It has several test patterns, but in our study we used the Macular test protocol. This protocol consisted of 48 test points located at 1°, 3°, 6°, and 10° from central fixation (Fig. 1). Using this test protocol, we have shown previously a greater loss in sensitivity within the central 6° in early AMD. ^3,20 The test uses the Zippy Estimation by Sequential Testing (ZEST) algorithm, adapted from the Bayesian method, and takes approximately 6 minutes to complete per eye. Static stimuli were presented with durations of 200 ms, while the flickering stimuli were presented with durations of 800 ms. The longer duration for the flickering stimulus is necessary to ensure that several flicker cycles are presented and there is enough time for the subject to differentiate static from flickering stimulus. In addition, previous work from our group showed that the 800 ms stimulus duration used in the Medmont perimeter (model M-700; Medmont International Pty Ltd.) was suitable to determine stable flicker thresholds. ²¹ For the static test, subjects were asked to respond by button press when a “spot of light” stimulus appears. Additional exposure and pseudo exposure were inserted randomly during testing for checking false-negative and false-positive, respectively. 

Figure 1

 

Stimulus pattern used in the study. The test grid consisted of 48 points located at 1°, 3°, 6°, and 10° from the central fixation.

Flicker perimetry was presented as an entirely separate test. In the flicker test, the stimulus is flashed on and off at equal intervals with a flicker rate. This test varies the temporal frequency of the stimulus with eccentricity to enhance the dynamic range of the test. ²² In the auto-flicker mode, thresholds are obtained at 1° and 3° with stimuli flickering at 18 Hz, 16 Hz at 6°, and 12 Hz at 10°. Flickering stimuli were delivered using the luminance-pedestal flicker method. Flicker thresholds were determined using the auto-flicker test. The patient was required to respond only if the stimulus was perceived to be flickering. If the target was visible but flicker was not apparent, then the patient should not respond. To assist the patient in making this discrimination, a high rate of nonflickering targets was presented at the start of the test. If a response was made to these static targets, a false-positive error was recorded. The participants were allowed a 2-minute practice trial during which errors made on nonflickering presentations were pointed out and positive reinforcement given for correct responses. Thereafter, the perimetric testing was performed without any feedback from the operator. Test reliability was assessed using nonflickering false-positive and flickering false-negative trials that were presented randomly during testing. Where the rate of false-positive or false-negative indices exceeded 20%, the results were discarded and the test was repeated. 

Fixation was checked by a blind-spot monitor algorithm. At the start of the test, the Medmont system (model M-700; Medmont International Pty Ltd.) performed a detection of the blind spot and used it to monitor fixation throughout the test. Blind spot fixation monitoring checked that the patient was fixated correctly on the central target by periodically exposing a light point, located at the blind spot of the eye being tested, and checking that the patient did not respond. The number of times a blind spot light point was seen by the patient as a ratio of the total number of blind spot checks performed represent the percentage of fixation loss. By convention, a fixation loss higher than approximately 20% is considered to indicate low patient reliability. None of the patients in our study had fixation loss indices more than 20%. 

Analyses

Participants were allocated to a severity group based upon the AMD status in the study and fellow eyes at baseline (see Table). These groups were arranged in a hierarchy of severity of clinical fundus features recognized as increasing the risk of progression to advanced stages of the disease. ^1,23–26  

Table. 

 

Study Group Classification

Group	Fundus Status		N	Age		Visual Acuity*
	Study Eye	Fellow Eye		Mean	SD	Mean	SD
1	No fundus pathology	No fundus pathology	24	61.0	12.4	−0.07	0.07
2	Hard drusen only (≤63 μm)	Same or no fundus changes	13	66.9	10.1	−0.07	0.07
3	Intermediate drusen (>63 to ≤124 μm) no pigment	Same as study eye or as in the group above	7	69.9	13.1	−0.02	0.08
4	Any soft drusen (≥125 μm) and no pigmentary changes ≥ 125 μm	Same as study eye or as in the groups above	66	67.3	11.3	−0.02	0.11
5	Any soft drusen (≥125 μm) and no pigmentary changes ≥ 125 μm	Any pigmentary abnormalities ≥ 125 μm with/without any drusen	12	73.9	8.6	0.02	0.13
6	Any drusen > 63 μm and pigmentary abnormalities ≥ 125 μm	Any early AMD changes	72	69.7	10.5	0.01	0.12
7	Drusenoid pigment epithelial detachment	Same as study eye or as in the groups above	17	71.4	8.8	0.11	0.18
8	Nonfoveal GA outside 1000 μm central circle and >175 μm in size	Same as study eye or as in the groups above	8	74.6	9.3	0.14	0.19
9	Any soft drusen with/without any pigmentary abnormality	GA	11	72.8	12.8	0.08	0.17
10	Any soft drusen with/without any pigmentary abnormality	CNV (including disciform scar)	64	75.1	9.0	0.04	0.15
11	Nonfoveal GA outside 1000 μm central circle and >175 μm in size	CNV (including disciform scar) or central GA > 175 μm	9	74.6	10.1	0.1	0.16

*  Visual acuity of the study eye in log unit.

To minimize learning effect, we discarded the result of the first test and used the results of the second test as baseline measurements. The geometric mean of static and flicker sensitivity of each concentric ring and the worst performing point (the test point with the lowest sensitivity) of each study eye was used for the analysis. Data of static and flicker responses were distributed normally. Thus, comparisons between the normal and AMD groups for static and flicker perimetry were performed with analysis of covariance (ANCOVA), with age as a covariate, and Dunnett post hoc tests. 

To over come the potential problems with different dynamic ranges and scales between the static and flicker test, the comparison between static versus flicker results was studied further by examining the number of SDs from normal and the percentage of eyes with abnormal sensitivity (eyes with more than 2 SDs from normal) for both perimetric methods for each AMD severity group. By using the number of SDs from normal and percentage of abnormal eyes, measurements for static and flicker were converted to a common scale, which allowed for a direct comparison between the two tests. The difference in the number of SDs between static and flicker at each AMD severity group was evaluated with Student's t-test. The percentage of eyes with abnormal sensitivity and 95% confidence intervals (CI) within each AMD severity group were calculated for each perimetric method. 

Those participants who completed 1-year follow-up clinical examination, photography, and perimetry were included in our longitudinal analysis. Participants were assessed for progression using our previously described severity scale ²⁷ by two independent graders. Eyes then were categorized into those with or without evidence of clinical progression. Linear regression analysis was used to compare the rate of change in static and flicker sensitivity with time in the stable early AMD and progressed groups. 

To allow some determination of test–retest reproducibility for static and flicker results, data at baseline and 6-month follow-up of the stable earliest AMD eyes (groups 2–4 in the Table) in were examined using the Bland-Altman method. ²⁸  

Results

A total of 303 subjects participated in the cross-sectional study. Participants were divided into 11 clinical groups based upon the clinical appearances of their fundi. A summary of the study groups is presented in the Table. One year follow-up data, with 3 consecutive perimetry results and color fundus images gradable for progression, were available on 129 eyes of 129 patients. Of these, 119 eyes had remained clinically stable and 10 eyes were graded as having progression within the early stages of AMD (increased number of drusen or development of pigmentary abnormalities) at the 1-year follow-up. Test–retest reproducibility data were available for analyses in 58 stable eyes of 58 subjects in groups 2 to 4. 

The mean static and flicker sensitivities of normal eyes (group 1) within the central 10° were 25.9 ± 3.1 and 22.4 ± 3.3 dB, respectively. On average, the flicker sensitivity was approximately 3.4 ± 2.9 dB lower than the static sensitivity in normal eyes. Reduction in retinal function in AMD eyes was detected mainly within the central 3 rings (central 6°) with the difference between static and flicker sensitivity being greater at the inner rings compared to the outer rings (Fig. 2). 

Figure 2

 

Static and flicker sensitivities of various AMD severity groups. Static and flicker sensitivity changed in a similar pattern across AMD severity groups. Error bars represent 1 SD. Asterisks indicate significance difference in sensitivity compared to the normal, nonAMD group (group 1).

The mean static and flicker sensitivity of various AMD severity groups is shown in Figure 2. The pattern of change in static sensitivity across all AMD severity groups was similar to that of the flicker sensitivity. Although there was a trend for reduction in static and flicker sensitivity in the group of eyes with drusen < 63 μm (group 2) and 63 to 125 μm drusen (group 3), the loss of sensitivity was not significant (Fig. 2). In ring 1, there was a significant reduction in flicker sensitivity in eyes with drusen > 125 μm (groups 4 and 5) compared to controls. Interestingly, eyes with any size of drusen and presence of pigmentary abnormalities (group 6) did not display worse function than eyes with drusen > 125 μm but no pigmentary abnormalities (groups 4 and 5), suggesting that the presence of pigment disturbances did not indicate a further reduction in the sensitivity over and beyond the presence of drusen. In eyes where the drusen size was large enough to consider the lesion a drusenoid PED (group 7), the responses in the central ring 1 were among the lowest sensitivities among the groups. The presence of GA in the fellow eye (group 9) did not result in lower sensitivity in the study eye in any ring compared to the sensitivities from study eyes with the same features, but without GA in the fellow eye (groups 5 and 6). Worse sensitivity might have been expected in the study eye in group 9 as those eyes are considered at increased risk for progression due to the status of the fellow eye. When the study eye had noncentral GA (groups 8 and 11), the presence of CNV in the fellow eye (group 11) was associated with a poorer sensitivity in the study eyes. This suggests that an eye may be further along the process of progression, as depicted by worse function, when CNV is present in the fellow eye. These findings indicated that the functional status of an eye does not always correspond with the predicted hierarchy of risk of vision loss based on clinical fundus signs. 

The number of SDs from a normal response was compared between static and flicker sensitivity for each AMD severity group (Fig. 3). It was apparent that by the time drusen > 125 μm (group 5) were present, the sensitivities were significantly reduced (>2 SDs) with flicker and static stimuli. In the cases of noncentral GA (groups 8 and 11) there was a further dramatic deviation from normal in flicker and static sensitivities at the outer rings. The pattern of change in the SDs across all AMD severity groups and eccentricities was similar for the static and flicker. In certain retinal locations and AMD severity groups, the static sensitivity was more reduced than was flicker sensitivity, while in other retinal locations, the flicker sensitivity was more reduced than static. For example, in ring 1, flicker results deviated more from normal compared to static, with this difference being significant in group 4 with drusen > 125 μm, but this difference was not seen at further eccentricities. Static sensitivities were more abnormal when compared to flicker sensitivities in ring 2 in all AMD groups. There did not appear to be a consistent difference between the two methods of testing retinal sensitivity, with no significant difference in the number of SDs between static and flicker in the vast majority of the AMD groups at all concentric rings nor when the mean of the worst performing points was considered. 

Figure 3

 

Mean SDs of static and flicker perimetry across various AMD severity groups. There was no significant difference in the numbers of SDs between static and flicker in the vast majority of AMD groups at all concentric rings, nor at the worst performing point. Error bars represent the SEM. Asterisks indicate significance difference in mean SD between static and flicker tests.

When considering the percentage of eyes with abnormal sensitivity detected by each perimetric method, we found that groups with early stages of disease (groups 2–4) had a low percentage of abnormal eyes, whereas groups with later stages of disease (groups 7–11) had a much higher percentage of abnormal eyes. When the results between the two methods were compared, again there was no significant difference in the percentage of abnormal eyes between the two perimetric methods in most groups. Flicker perimetry appeared to define a higher percent of abnormal eyes compared to static, particularly in ring 1 of the groups with early stages (3–4), but these differences were not statistically significant (Fig. 4). In ring 2, however, the percentage of eyes with abnormal static sensitivity was significantly greater than that of the flicker sensitivity in groups 6 and 10 (Fig. 4). A significantly greater percentage of abnormal eyes also was detected by static perimetry in ring 3, group 10 compared to flicker perimetry. 

Figure 4

 

The percentage of eyes with abnormal sensitivity for static and flicker perimetry across various AMD severity groups. There was significantly greater percentage of eyes with abnormal static sensitivity compared to that of the flicker sensitivity in groups 6 (ring 2) and 10 (rings 2 and 3). The differences in the percentage of abnormal eyes detected by the two methods in the remaining severity groups and rings were not significant. Error bars represent 95% CIs.

Of the 129 eyes with data available for the longitudinal component of the study, 119 eyes remained clinically stable and 10 eyes progressed within the early stages of AMD at 1 year follow-up. The changes in static and flicker sensitivity with time in the stable early stages of AMD group and progressed group are shown in Figure 5. Linear regression analysis results showed that the β-coefficients (slope of the linear function) for the static and flicker in the stable group were 0.54 (95% CI −0.58–1.71) and 0.28 (95% CI −0.99–1.56), respectively. In the progressed within the early stages of AMD group, the β-coefficient for the static was −1.59 (95% CI −4.78–1.60) and for the flicker it was −1.29 (95% CI −4.66–2.08). Therefore, static and flicker sensitivities appear to change more rapidly in those eyes that show clinical signs of progression, but there was no significant difference in the β-coefficient between static and flicker in the stable and progressed groups. 

Figure 5

 

Static and flicker sensitivity at baseline and 1-year follow-up in the stable early AMD and progression groups. Static and flicker exhibit similar patterns of change in sensitivity in the stable group as well as the progression group. Error bars represent 95% CI.

Test–retest reproducibility was examined in 58 eyes of 58 subjects in the stable group. The Bland-Altman plots for static and flicker are shown in Figure 6. The mean difference for static perimetry was −0.18 dB, and the limits of agreement (95%) were −5.45 and 5.09 dB. The mean difference for flicker perimetry was −0.41 dB, and the limits of agreements were −4.96 and 4.15 dB. These findings indicated that in this cohort of participants the test–retest reproducibility of the flicker test is similar to that of static. 

Figure 6

 

Bland Altman plots showing the test–retest repeatability for static and flicker perimetry.

Discussion

In this study, we examined the static and flicker sensitivities using a Medmont perimeter (model M-700; Medmont International Pty Ltd.) across a range of AMD severity groups to determine whether static and flicker test results were relative to each other in each severity grade of AMD, and whether clinical severity, as determined by color fundus photographic grade, correlated with the recorded functional deficits. 

With regards to disease severity, both perimetric measures changed broadly with severity grade. We found that static and flicker sensitivities decreased as drusen size increased in the study eyes. However, we did not always find a reduction in sensitivity when clinical fundus signs, such as the addition of pigmentary disturbance or in the presence of GA, were seen in the fellow eye, indicating a higher clinical risk for progression. Consistent with our earlier reports using other functional modalities, ¹⁴ we found that the group with drusen 63 to 125 μm in size had most variation in sensitivity in the inner rings, in both test modalities, suggesting that at this stage of AMD some people will have normal function while others will have quite abnormal results. Perhaps at this stage we might be able to differentiate those destined to have progression and those less likely to do so. Thus, functional testing could provide additional information to the traditional grading of AMD severity using color fundus photographs and may aid in the classification of the early stages of AMD. 

It has been reported that flicker sensitivities are affected more than static sensitivities in AMD and, as such, potentially would make for a better functional assay. ^3,11 In our study, we found that while flicker sensitivities always were lower than static, they were consistently so and both perimetric results were affected similarly across AMD severity groups. Occasionally, one parameter may have been more abnormal compared to the other, but there were no consistent patterns and the differences were not statistically significant in the majority of groups. When determining the ability to discriminate eyes with abnormal results, both tests were not particularly sensitive at the early stages of disease where these tests are more likely to be used. However, it is possible that those in whom an abnormality is found (around 10%–40% of cases in both modalities) may well be the people more likely to have progression to vision loss. Examining their rate of change in sensitivity with time would allow us to differentiate those with progression from those without, as reported here and in a recent publication. ¹⁵ The test–retest reliability for static and flicker perimetry found in our study was approximately ±5 dB. Thus, small changes in sensitivity should be interpreted with caution. 

When considering the use of a functional measure in AMD, one must consider the purpose of the test. The criteria of the test for investigating the very earliest changes in a small group of subjects in a laboratory setting would be different from that for use in a large clinical trial to monitor functional changes with time. In large scale clinical trials, many facets will need to be considered in addition to the output indices, such as the sensitivity. Parameters, such as reliability indices, ease of use, and availability of the test equipment, also are important. Thus, although static and flicker stimuli delivered by the Medmont system (model M-700; Medmont International Pty Ltd.) appear to be equally sensitive in detecting changes in retinal function in AMD, patients preferred the static perimeter as they reported that it is easier to perform, the test duration is shorter compared to flicker perimetry, and flicker perimetry required a greater level of cognitive processing. Furthermore, unpublished data from our laboratory shows that reliability indices, such as false-negatives, were greater in the flicker tests compared to static testing (5.8% in flicker vs. 1.6% in static), even in experienced participants. We often found that in the flicker perimetry test participants were unsure whether the target flickered and so produced more false-negative responses and, consequently, producing more points with false-positive defect. In addition, all perimeters currently manufactured have the static stimulus, but very few perimeters have flicker testing capacity, making static testing widely applicable. It also is worth noting that a few static perimeters currently available have a fundus tracking capability to ensure the same retinal location is being tested over time. When taking into account all these additional considerations, static perimetry appears to be a practical and useful test for assessing visual function, at least in large scale clinical trials of the early stages of AMD. However, in using static perimetry on the Medmont (model M-700; Medmont International Pty Ltd.), significant differences appeared only once drusen > 125 μm were present. This technique was not able to discriminate between earlier stages of the disease and normal age-matched controls. 

The main strength of our study was that static and flicker perimetry was performed at the same session with the same perimetry devices, by a single experienced examiner throughout the study. This minimized the variation between tests. The study also consisted of a large sample of AMD patients. A limitation of this study was that the presence or absence of GA could be determined only on the bases of digital color fundus images as fundus autofluorescent images were not available. Furthermore, the lack of a fundus tracking system during visual field testing using the Medmont perimeter (model M-700; Medmont International Pty Ltd.) meant that unreliable responses due to poor fixation had to be discarded and the test needed to be repeated to improve the reliability of the tests. However, all of the subjects had good vision (>20/40) so that they had no problem seeing the central fixation spot. While we were unable to detect a significant difference in sensitivity to a static or flickering stimulus as presented on a Medmont perimeter (model M-700; Medmont International Pty Ltd.), it might be possible to detect difference using other stimulus parameters. In addition, the flicker stimulus generated by the Medmont may not produce a pure temporal signal and it is possible that, because of this limitation, the static and flicker tests showed similar changes in AMD in our study. However, our group has shown previously that this type of flicker stimulus is tuned temporally and suitable to determine stable flicker thresholds. ^21,29–31  

In conclusion, retinal dysfunction in AMD can be measured by static and flicker perimetry, and both show trends of worsening sensitivities as the clinical stage of the disease progresses. The size of the drusen appeared to be important in determining the worsening function when compared to the influence of pigmentary abnormalities. The increased risk of progression in an eye due to the advanced nature of the fellow eye was not reflected consistently by a poorer function, when comparing eyes of similar clinical features, but without advanced fellow eye disease. Static and flicker perimetry can detect functional changes associated with progression. Neither test detected abnormalities in a high percentage of early stages of disease, where drusen was <125 μm, but those in whom a defect was detected may be the ones most likely to have progression. Flicker perimetry has been shown to be useful in monitoring disease progression and predicting the development of advanced AMD, and from the longitudinal results presented here, it would appear that static perimetry also would be useful. When considering all aspects of a test that might be used to monitor change with time in large scale clinical trials, static perimetry has several advantages. 

Acknowledgments

Presented in part at the annual meeting of the Association for Research in Vision and Ophthalmology, Fort Lauderdale, Florida, May 5–10, 2012. 

Supported by the National Health and Medical Research Council (NH and MRC) Project Grant 350224, Australian Research Council Linkage Project (ARC-LP0211474), Macular Degeneration Foundation Research Grant, NH and MRC practitioner fellowship (RHG, #529905). CERA receives Operational Infrastructure Support from the Victorian Government and is supported by a NHMRC Centre for Clinical Research Excellence Award (#529923). 

Disclosure: C.D. Luu, None; P.N. Dimitrov, None; Z. Wu, None; L.N. Ayton, None; G. Makeyeva, None; K.-Z. Aung, None; M. Varsamidis, None; L. Robman, None; A.J. Vingrys, Medmont International Pty Ltd. (F, C); R.H. Guymer, None