Multifocal Electroretinography in Eyes with Reticular Pseudodrusen

Abstract

Purpose.: The aim of our study was to evaluate the impact of reticular pseudodrusen (RPD) on retinal function by multifocal electroretinography (mfERG), and combined simultaneous confocal scanning laser ophthalmoscopy (cSLO) and spectral-domain optical coherence tomography (SD-OCT).

Methods.: We included 19 eyes of 15 patients with RPD in the posterior pole and no other phenotypic retinal alteration were included (7 females and 8 males, age 77.2 ± 5.1 years) as well as 24 eyes of 17 healthy control subjects (7 females and 10 males, age 73.2 ± 5.9 years). All patients underwent fundus photography, SD-OCT, fluorescence angiography (FA), fundus autofluorescence, and near-infrared reflectance cSLO. mfERG measurements were performed by stimulating the retina by a field of 103 hexagons covering an area of approximately 30°. Amplitudes and latencies of focal retinal responses obtained at affected and nonaffected sites of RPD eyes and retinal responses of healthy control subjects were compared.

Results.: In all included study eyes, RPD stages 1–3 could be demonstrated clearly in SD-OCT, FA, and cSLO. The mean amplitudes measured in the areas affected by RPD were 12.5 to 53.1 nV/deg² (control group 19.4–50.1 nV/deg²). The mean latencies were 33.2 to 41.3 ms (control group 33.6–39.7 ms). mfERG amplitudes and latencies of retinal areas affected by RPD were not altered significantly when compared to corresponding nonaffected areas.

Conclusions.: mfERG measurements did not show a definite influence on electrophysiologic activity in retinal areas affected exclusively with RPD.

Introduction

With the advent and constant development of high resolution retinal imaging, age-related macular degeneration (AMD) patients with reticular pseudodrusen (RPD) received a lot of attention recently. Studies proved high prevalence of RPD for patients with AMD and showed higher incidence of progression to late stage AMD among subjects with compared to those without RPD. ^1–5  

Terminology of these lesions has not been consistent in the literature. Terms, like “reticular drusen,” have been used recently. However, they do not reflect the clinical and histologic findings, and can be confused with early soft drusen, hard drusen, or cuticular drusen. Our study deals with RPD also known as subretinal drusenoid deposits. ⁶  

RPD are identified readily by combined simultaneous confocal scanning laser ophthalmoscopy (cSLO) and spectral-domain optical coherence tomography (SD-OCT), and can be discriminated clearly from other drusen types. ^6,7 Efforts have been made to identify the exact location of RPD and to gain further insight into the pathogenesis of retinal alterations. However, it must be investigated to what extent RPD influence local retinal function. 

A localized functional assessment of retinal function became possible by a multifocal electroretinography (mfERG) technique developed by Sutter and Tran. ^8,9 mfERG has been used for more than 15 years to support diagnosis of retinal pathologies. mfERG allows measurement of local ERG activity of the cone-driven retina under light-adapted conditions. Thus, it provides a topographic map of a retinal electrophysiologic activity. 

The aim of our study was to evaluate the impact of RPD in the macula on retinal function by means of mfERG and combined simultaneous cSLO and SD-OCT. 

Methods

Population

Participants were recruited from the medical retina clinic of the Department of Ophthalmology at University of Münster Medical Center. Only patients with distinct RPD in combined simultaneous cSLO and SD-OCT imaging of the posterior pole in one or both eyes were considered. Eyes were not eligible if any signs of conventional drusen, choroidal neovascularization (CNV), geographic atrophy (GA), or pigment epithelium detachment (PED) due to AMD were observed in funduscopy, SD-OCT, cSLO, or fluorescence angiography (FA). Intravitreal anti-VEGF therapy and vitreoretinal surgery in the medical history as well as other vascular or inflammatory retinal pathologies also led to exclusion. Furthermore, eyes with dense lens opacities, corneal opacities, refractive surgery, or a history of intraocular inflammation were not considered either. 

We included 19 eyes of 15 patients with RPD in the posterior pole (7 females and 8 males, age 76.9 ± 4.8 years). Best corrected visual acuity (BCVA) better than 0.5 of the study eye was required to ensure stable fixation during mfERG measurement. A total of 24 eyes of 17 healthy control subjects (7 females and 10 males, age 73.2 ± 5.9 years) without RPD or any other retinal pathology served as a control group. All control subjects also underwent the same examination protocol as the RPD group. Among the control group, 19 eyes (7 females and 8 males, age 73.5 ± 6.3 years) were matched to RPD eyes based on age and sex. 

Informed consent was obtained from all subjects or patients before testing. The research followed the tenets of the Declaration of Helsinki. 

Imaging

All patients underwent fundus photography (Zeiss FF450 plus IR, Carl Zeiss Meditech, Berlin, Germany), SD-OCT, standardized FA and cSLO (fundus autofluorescence [FAF, λ = 488 nm], near-infrared reflectance [IR, λ = 830 nm]; Spectralis, Heidelberg Engineering, Heidelberg, Germany). cSLO imaging was performed with a minimum resolution of 768 × 768 pixels. The field of view was set at 30° × 30° and centered on the macula. For FAF images, blue laser light at 488 nm was used for illumination and a barrier filter at 500 nm applied to limit captured light to autofluorescent structures. SD-OCT images were obtained by using a 20° × 20° volume scan consisting of 25 equally spaced horizontal averaged B-scans. Scans were saved for evaluation after 100 frames had been averaged using the automatic averaging and eye-tracking feature of the Spectralis device. Fundus color photographs were viewed in the Zeiss Visupac 4.2. Heidelberg software (Heidelberg Eye Explorer 1.6.4.0; Heidelberg Engineering) was used for viewing of SD-OCT and cSLO images. 

Definitions

First and most frequently, RPD are noted between the superior part of the fovea and the superior temporal arcade. ² Funduscopically, they may appear slightly whiter or greyish compared with soft drusen. RPD may have a more punctate appearance closer to the fovea. ¹⁰ Fluorescence angiography was performed to rule out any retinal alteration apart from RPD. Here, RPD can be detected as an area of decreased fluorescence surrounded by a faint halo of increased fluorescence adjacent to a network of reticular hypofluorescence. ¹¹  

In our study, identification of RPD depended on required characteristic features in cSLO FAF, IR, and SD-OCT imaging (Fig. 1). For FAF imaging, RPD were defined as a regular network of uniform round or oval shaped irregularities with a diameter ranging between 50 and 400 μm. Furthermore, lesions were characterized by a decreased FAF signal surrounded by mildly increased intensities. ^2,11 In IR imaging, RPD were identified as a pattern-like grouping of lesions varying in size with decreased reflectivity. For larger lesions, these images may be accompanied by a halo-like appearance exhibiting an increased IR signal in the center, surrounded by a decreased intensity. ^2,11 Querques et al. described this characteristic as a “target” aspect of RPD in IR similar to their appearance observed in FAF. ¹¹ In SD-OCT, RPD appear as hyperreflective material above the RPE in the subretinal space. RPD vary in shape and thickness, appearing as conical or flattened lesions. ^1,7 Each eye was evaluated against RPD characteristics based on the grading system established by Zweifel et al. ⁶ and Querques et al. ¹²  

Figure 1. 

 

(A–H) Images of the eye of a representative patient with exclusively RPD obtained by different imaging techniques. (A) Fundus photography shows RPD with a slightly greyish aspect compared to soft drusen. RPD typically show a more punctate appearance closer to the fovea. (B) cSLO near-infrared reflectance image shows typical pattern-like grouping of hyporeflexive lesions in the posterior pole. (C, D) Fluorescence angiography shows hypofluorescent lesions in the early (C) and late (D) phase. (E–H) Combined simultaneous cSLO FAF mode and SD-OCT demonstrate in vivo images of RPD. (E) cSLO FAF image shows lesions with a decreased signal surrounded by mildly increased intensities. (F–H) SD-OCT scans show characteristic lesions anterior to the retinal pigment epithelium that correspond to the described changes in cSLO FAF image. Arrowheads: mark RPD in the SD-OCT scans. Cyan lines: on the cSLO images mark the location of the SD-OCT scan aside. SD-OCT scans are magnified and, therefore, have a different scaling.

mfERG

Subsequently, mfERG was assessed in both eyes. The recordings were performed with a RetiPort System (Roland Consult, Brandenburg an der Havel, Germany). mfERGs were recorded simultaneously on both eyes using Dawson-Trick-Litzkow (DTL) thread electrodes. Pupils were dilated with 1% tropicamide, and recording started when the pupils were dilated to at least 7 mm. Focusing lenses were used when necessary. mfERG was performed according to latest ISCEV guidelines by an experienced technician. ¹³ The retina was stimulated with an array of 103 hexagonal elements presented on a monitor, each of which has a 50% chance of being illuminated every time the frame changes according to a fixed pseudorandom m-sequence. Fixation was monitored by the recording technician. At every mfERG examination, each patient positively reported that he or she could perceive the cross-shaped fixation target clearly. The stimulus contrast was approximately 98% of the luminance of the white hexagon at 120 cd⋅s/m². Background luminance was set at 2 cd⋅s/m² to minimize the stray light affect. Recorded signals were band pass filtered between 10 and 100 Hz, and amplified 100,000 times. Noise-contaminated segments, due to blinks or small eye movements, were rejected and re-recorded. 

Based on manually controlled cursor placement, the mfERG P1 response amplitude (from first negative to first positive peak) and P1 latency (from stimulus onset to first positive peak) were extracted from the data. 

FAF images were imported into the RetiPort software, and an overlay of the measured waveforms was performed, adjusting the center of the measurement field to the fovea in the FAF image. 

Two methods for the statistical evaluation of the mfERG measurements were chosen. Firstly, responses of single retinal fields with and without RPD were compared. In each eye, 10 of the 103 mfERG fields with RPD and whose opposite fields without RPD were identified. As a result, 10 matched pairs were obtained. Image grading for presence of RPD in the retinal fields was performed independently by two masked observers. If the two observers disagreed, a third one was asked to arbitrate. Observers were instructed to use, apart from FAF images, also corresponding IR and SD-OCT images to improve RPD identification in case discrimination between RPD-bearing and RPD-nonbearing fields was unclear (method 1). 

Secondly, macular quadrants were compared. On the basis of the previously described circular grid of the Early Treatment Diabetic Retinopathy Study (ETDRS) Grid and “The Wisconsin age-related Maculopathy Grading-System,” a line from the center of the optic disc passing through the fovea was drawn. ¹⁴ Two more lines through the fovea were placed in a 45-degree angle to this line dividing the macula into four retinal fields of equal size (Fig. 2). Depending on the individual distribution of RPD, the quadrant mostly affected by RPD and the opposite quadrant least affected by RPD were identified, and matched pairs of mfERG measurement fields lying exactly on the opposite sites within these two quadrants were compared (method 2). In both methods, retinal fields in the study eye that were not affected by RPD served as an inner control. 

Figure 2. 

 

(A–D) Representative mfERG responses of the patient seen in Figure 1 with underlying cSLO FAF. (A) Ten pairs of measured waveforms opposing each other are encircled. Yellow fields: sectors where RPD can be detected. Blue fields: sectors without RPD. (B) Sketch of the grid that was applied to define the superior, inferior, temporal, and nasal quadrants. A horizontal line passes from the center of the optic disc through the fovea. Two more lines through the fovea are placed in a 45° angle to the horizontal line dividing the macula into four equal retinal fields. For data evaluation, areas were chosen with opposing sectors. (C) P1 response amplitudes obtained in the 103 measurement sectors are shown. The scale indicated the amplitude values given in nV/deg². (D) P1 latencies are shown for the 103 sectors. The heights of the columns give the values in ms above 30 ms as indicated by the scale.

Both methods also were applied to the control group. As there was no RPD in the control eyes, eyes of normal subjects were matched by age and sex to eyes of patients, and the same sectors as in the RPD eyes also were used in the control eyes for comparison. Firstly, the corresponding ten retinal fields of the healthy control eyes were compared to the opposite fields of the same eccentricity of the same eye. As a result, ten matched pairs were obtained for comparison (method 1). Secondly, the corresponding quadrants were compared in the control eyes that had been compared previously in the study eyes (method 2). 

Statistical Methods

Data were exported from the RetiPort software as Excel files and imported into the Microsoft Excel program for further analysis. Statistical significance of the differences between data obtained in areas with or without RPD and the differences between the corresponding retinal fields in the control group was calculated by Student's t-test and the Wilcoxon signed-rank test. Statistical significance was set at P < 0.05. 

Results

Four patients showed exclusively RPD in both eyes. Eleven patients had one eye with only RPD alterations, while the fellow eyes showed conventional drusen in two patients, CNV in eight patients and a rip of the retinal pigment epithelium due to AMD in one patient. Of the 19 included eyes 10 showed a mild decrease in visual acuity. Mean BCVA of the studied eyes was 0.87 ± 0.13, ranging from 0.6–1.0. The recording technician observed a good fixation for all 15 patients. 

In SD-OCT, RPD appear as discrete conglomerates of hyperreflective material above the RPE in the subretinal space, contrary to the location of conventional drusen material underneath the RPE. RPD vary in shape and thickness, appearing as conical or flattened lesions. The subretinal location accounts for the typical undulations of the inner segment-outer segment (IS/OS) junctions as seen on SD-OCT scans (Fig. 1). In all included eyes, in addition to RPD stages 1 and 2, at least one SD-OCT section showed RPD breaking through the IS/OS boundary, which corresponds to stage 3 based on the grading system by Zweifel et al. ⁶ and Querques et al. ¹²  

As described in the Methods section, the mfERG were evaluated by two methods. The choice of the areas and the matched pairs of the sectors for data analysis is demonstrated for one typical eye in Figure 2. Whereas ten distinct matched pairs were chosen for data evaluation according to method 1 (Fig. 2A), matched pairs in complete areas were chosen in the upper and lower quadrants according to method 2, and the number of compared areas differed among the eyes, ranging from 10 to 20 (Fig. 2B). 

Considerable differences in amplitudes between patients were noticed. In addition, amplitudes obtained in the single sectors of one eye also were different depending on their distance to the fovea. The mean amplitudes measured in the areas affected by RPD were in the range of 12.5 to 53.1 nV/deg² (control group 19.4–50.1 nV/deg²). The values of latencies did not differ much. The mean latencies were found to be 33.2 to 41.3 ms (control group 33.6–39.7 ms). Beside the statistical tests described above, the percentage differences between the matched pairs were analyzed by performing a box plot and the Wilcoxon signed-rank test (Fig. 3). ¹⁵  

Figure 3. 

 

(A–D) Inner comparisons of amplitudes (A, C) and latencies (B, D) of study group (A, B) and control group (C, D) obtained by mfERG. Box plots are shown based on the percentage of the differences between the values obtained in sectors with and without RPD of all 19 eyes (A, B). For the control group (C, D), comparison was performed between the superior and inferior sectors corresponding to those used in the RPD patients. The sectors to be compared were chosen by method 1 (white symbols) or method 2 (gray symbols). The length of the whiskers was defined by the data still in the 1.5× interquartile range. Diamonds: represent some data points located out of this range. ¹⁵ At zero, there is no difference in the parameter values between areas with or without RPD and superior or inferior sectors. Data above the broken zero line indicate higher values of amplitudes (A, C) or latencies (B, D) in areas with RPD or superior sectors, whereas data below zero line indicate higher values in areas without RPD or in inferior sectors. *Values of the parameters are significantly lower than zero. †Values of the parameters are significantly higher than zero (Wilcoxon signed-rank test).

The mfERG response amplitudes of retinal areas affected by RPD were not reduced significantly when compared to non-affected areas (Fig. 3A). There also was no significant change in the mean latencies in affected retinal areas compared to non-affected areas (Fig. 3B). The comparison between amplitude and latency values obtained in corresponding superior and inferior retinal fields in the healthy eyes did not show any statistically significant differences either (Figs. 3C, D). The results of statistical tests calculated by the Wilcoxon rank test are listed in the Table for all 19 RPD eyes and the matched 19 control eyes. Although the differences in the parameters are statistically significant in a few single eyes, no statistical significance of the differences was found for the mean values calculated in total out of all eyes. Moreover, the statistical results were almost the same when Student's t-test was performed. 

Table. 

 

Results of the Wilcoxon Signed-Rank Test

Eye No.	RPD Group				Control Group
	Larger Amplitudes		Shorter Latencies		Larger Amplitudes		Shorter Latencies
	Method 1 (10 Fields)	Method 2 (Quadrants)	Method 1 (10 Fields)	Method 2 (Quadrants)	Method 1 (10 Fields)	Method 2 (Quadrants)	Method 1 (10 Fields)	Method 2 (Quadrants)
1	n.s. (c)	n.s. (c)	n.s. (d)	n.s. (c)	n.s. (i)	n.s. (i)	n.s. (s)	n.s. (s)
2	n.s. (c)	n.s. (c)	n.s. (d)	RPD	n.s. (s)	n.s. (s)	n.s. (s)	n.s. (s)
3	n.s. (c)	Control	n.s. (d)	RPD	n.s. (i)	n.s. (i)	n.s. (s)	n.s. (s)
4	n.s. (d)	n.s. (d)	Control	n.s. (c)	n.s. (i)	n.s. (s)	n.s. (i)	n.s. (i)
5	n.s. (c)	n.s. (c)	RPD	n.s. (d)	n.s. (i)	n.s. (s)	n.s. (s)	n.s. (s)
6	RPD	RPD	RPD	RPD	n.s. (i)	n.s. (i)	n.s. (i)	n.s. (i)
7	Control	n.s. (c)	n.s. (d)	n.s. (d)	n.s. (s)	n.s. (s)	n.s. (s)	n.s. (s)
8	n.s. (c)	n.s. (c)	n.s. (d)	n.s. (c)	n.s. (s)	n.s. (i)	n.s. (i)	Superior
9	n.s. (d)	RPD	n.s. (d)	n.s. (d)	n.s. (i)	n.s. (s)	n.s. (i)	n.s. (i)
10	RPD	n.s. (d)	Control	n.s. (c)	n.s. (s)	n.s. (i)	n.s. (i)	n.s. (i)
11	RPD	RPD	Control	Control	n.s. (s)	n.s. (s)	n.s. (i)	n.s.*
12	n.s. (d)	n.s. (d)	n.s. (d)	n.s. (c)	n.s. (s)	n.s. (i)	n.s. (i)	n.s. (s)
13	n.s. (c)	n.s. (c)	n.s. (c)	RPD	n.s. (s)	Superior	n.s. (s)	n.s. (i)
14	n.s. (d)	n.s. (d)	n.s. (c)	Control	n.s. (s)	n.s. (s)	n.s. (i)	n.s. (i)
15	n.s. (c)	n.s. (c)	n.s. (c)	n.s. (c)	n.s. (i)	n.s. (i)	n.s. (i)	n.s. (i)
16	n.s. (d)	n.s. (c)	Control	n.s. (c)	n.s. (i)	n.s. (s)	n.s. (s)	n.s. (i)
17	n.s. (c)	n.s. (d)	Control	n.s. (c)	n.s. (i)	n.s. (s)	n.s. (s)	n.s. (s)
18	RPD	RPD	n.s. (d)	n.s. (d)	Superior	Superior	n.s. (s)	n.s. (s)
19	n.s. (c)	n.s. (c)	n.s. (d)	n.s. (d)	n.s. (s)	n.s. (s)	n.s. (s)	Inferior

 RPD, indicates a significantly larger amplitude or shorter latency in the RPD affected retinal fields; Control, indicates a significantly larger amplitude or shorter latency in the RPD nonaffected retinal fields. RPD group: If the difference between measurements was nonsignificant (n.s.) it is marked whether the larger amplitude or shorter latency was measured in the RPD affected retinal fields (d) or in the RPD nonaffected fields. Control group: If the difference between superior and inferior sectors was nonsignificant (n.s.) it is marked whether the larger amplitude or shorter latency was measured in the superior sectors (s) or in the inferior sectors (i). Superior, indicates a significantly larger amplitude or shorter latency in the superior sectors. Inferior, indicates a significantly larger amplitude or shorter latency in the inferior sectors.

*  identical values.

It was evaluated whether age or visual acuity of the patients influenced the differences in mfERG parameters. No such correlations could be determined (not shown). 

Discussion

mfERG proved to be a useful and sensitive functional biomarker in the quantitative assessment of localized retinal function in eyes with early and late stage AMD. Studies have been published on mfERG measurements in eyes with different phenotypic retinal alterations due to AMD. ^16–24 Yet, so far to our knowledge no study exists in which eyes with RPD were analyzed by mfERG measurements. 

Results of BCVA testing in our patients are not surprising as eyes with early AMD also have either normal BCVA or a slight decrease of approximately two or fewer letters. ²⁵ However, BCVA does not necessarily reflect underlying changes in pathology or risk of significant vision loss. Patients suffering from AMD often have impaired central vision and, thus, are prone to have unstable fixation. However, patients with no retinal alteration but RPD usually have normal vision and no central scotoma, which explains the stable fixation in our patients. A stable fixation during mfERG recordings is crucial to assess valid and accurate results. Chu et al. found significantly reduced amplitudes for 4° unsteady fixation in subjects with normal vision, but no substantial affection if fixation is maintained within the central stimulus hexagon. ²⁶  

Previous reports on mfERG in patients with early AMD are inconsistent. Reduced foveal amplitudes and increased latencies have been reported in participants with early AMD compared to control participants. ^16–19 Gin et al. recently analyzed mfERG measurements in patients with early AMD and in a healthy control group, and found that the former showed a significant reduction in mean foveal mfERG amplitude and increase in mean mfERG latency in the central 12° compared to that of the control eyes. ²⁰ Other studies were not able to prove significant mfERG changes in early AMD eyes in comparison to healthy control groups at all. ²¹  

Two reports have been published on the progression of change in mfERG responses in patients with early dry AMD. Feigl et al. could not find a significant progression of mfERG responses among 13 patients with early AMD after one year. ²² On the other hand, Gerth et al. followed 14 patients over 28 to 41 months and found that patients with soft drusen showed a progressive loss in mfERG responses despite stable visual acuity. ²³  

Controversy exists as to the spatial resolution of the mfERG measurements in AMD patients and its capacity to reflect accurately morphologic alterations in the posterior pole by a changed functional read-out. Gerth et al. examined AMD patients with large drusen and found significant localized retinal dysfunction; however, significantly abnormal mfERG responses were not restricted to areas with drusen. ¹⁹ Even in neovascular AMD, only a weak-to-moderate correlation between the greatest linear dimension of a choroidal neovascular lesion and mfERG response density was reported by Jurklies et al. ²⁴  

Data averaging compensates for poor signal-to-noise ratios, but comes with the price of losing the spatial specificity of the mfERG. Spatial averaging and group averaging are known to conceal individual abnormal responses. ²⁵ Spatial averaging and extensive group averaging were avoided so as not to mask any significant alterations in central retinal function. 

It has been shown that RPD are found predominantly in the superior part of the macula. ² Consequently, we compared mostly RPD-affected retinal fields in the superior hemisphere with RPD-nonaffected fields in the inferior hemisphere, which possibly could produce a bias. For this reason, we performed the same measurements and data evaluation in a group of healthy controls to check whether function of retinal fields in the superior and inferior regions may be compared legitimately. In accordance with Li et al., who showed that no significant differences in mfERG measurements exist between the nasal and temporal or the superior and inferior regions of the macula, ¹⁹ we did not find any differences between superior and inferior retinal function either. As can be seen in Figure 3, there are certain variations in amplitudes and latencies between the eyes of the RPD group. Notably, similar variations also were found between the eyes of the healthy control group. 

The mechanisms causing the dysfunction of central photoreceptors in early AMD measured in mfERG so far are not entirely understood. It is hypothesized that conventional soft drusen may induce a mechanical displacement of photoreceptor outer segments, and impede the nutrient exchange between photoreceptors and choriocapillaris. 

A decrease in retinal function would be indicated by a decreased amplitude and/or a prolonged latency. It was not possible to detect a corresponding change of amplitudes or latencies in most of the analyzed eyes. Obviously, RPD do not harm the function of the cone photoreceptors that much in most cases. Similar to the findings of Gerth, a marked decrease of local retinal function in areas without RPD was found in a number of eyes (not shown). ¹⁹ Whether an impairment of local retinal function due to RPD occurs in later stages, remains to be elucidated. 

The main signals in the mfERG are derived from cones, which might be one of the reasons for insignificant changes in mfERG amplitudes, as RPD represent a phenotype that is present primarily in the perifoveal macula. ² Testing only cone function in this phenotype certainly limits the validity of our study, since numerous studies have shown that loss of rod sensitivity and slowing of dark adaptation are early signs of photoreceptor dysfunction in AMD as well. ^27–30  

There are different hypotheses regarding the morphologic correlate of RPD. Sarks et al. proposed that undigested disorganized photoreceptor outer segments represent the distinct findings. ³¹ Rudolf et al. hypothesized that RPD are metabolism products that may be released from the apical as well as the basal surface of the RPE cell. ³² Sohrab et al. recently provided results using point-to-point correlation of registered IR, FAF, and red-free images that showed a colocalization of the reticular pattern of RPD and the intervascular choroidal stroma on en face OCT sections. ³³ Lengyel et al. presented similar evidence that demonstrated a very tight spatial relationship between conventional drusen and the intercapillary spaces in the choroid. ³⁴  

Possibly, RPD may be a secondary pathologic change representing an epiphenomenon of choroidal macular malperfusion. ³⁵ Better imaging techniques may elucidate further choroidal perfusion and the possible association with RPD in the future. 

Hood et al. showed that the first-order kernel response of the mfERG originates from photoreceptor and bipolar cells. ^36,37 As RPD can be localized by high-resolution SD-OCT between the RPE and the IS/OS boundary of photoreceptor cells or even breaking through the IS/OS boundary depending on the different stages of RPD, ^6,12 mfERG appears to be a suitable method to measure a possible functional impact of RPD on cone photoreceptors. 

However, when thinking about RPD and their anatomic relationship to the attributed photoreceptor layers in SD-OCT, one must bear in mind that the exact location of RPD is yet to be proven histopathologically, and that RPD and their location seen in SD-OCT may well be caused by optical interferences, as suggested by Schmitz-Valckenberg et al. ² Furthermore, current anatomic attribution of the four hyperreflective bands, conventionally resolved in the outer retina by commercial SD-OCT instruments, particularly bands two and three, is questionable and needs further clarification regarding subsequent pathophysiologic assessments. ³⁸  

Our study showed that the presence of RPD in a certain retinal area per se does not seem to influence the electrophysiologic activity of cone photoreceptor and bipolar cells. 

Our study was limited by several factors. Firstly, a relatively small sample size limits the data's validity. Nevertheless, the consistency of our findings among our subjects supports our conclusions. Another limitation may be represented by the slight age difference between patients with RPD and control subjects (76.9 ± 4.8 vs. 73.5 ± 6.3 years). 

Whether the different stages of RPD may have a different influence on the retinal function cannot be answered in our study, since all included eyes showed RPD stages 1–3 based on the grading system by Zweifel et al. ⁶ and Querques et al., ¹² which made a comparison between patients with exclusively RPD stage 1 versus exclusively stage 2 or stage 3 impossible. 

To our knowledge, this is the first study to assess retinal function using mfERG in patients with RPD. Contrary to other phenotypic characteristics of AMD, mfERG measurements did not show definite interference of electrophysiologic activity in retinal areas affected exclusively with RPD. Follow-up measurements over time to assess potential changes during disease progression as well as a larger patient collective are required before drawing definite conclusions concerning the functional relevance of RPD. 

Acknowledgments

Verena Lüers and Petra Vogt are acknowledged for their technical assistance. 

Footnotes

 Disclosure: F. Alten, Heidelberg Engineering (C), Novartis (C); P. Heiduschka, Heidelberg Engineering (C); C.R. Clemens, Heidelberg Engineering (C); Novartis (C); N. Eter, Heidelberg Engineering (C), Novartis (C), Bayer (C), Pfizer, (C)