Purpose
To describe the natural course, visual outcomes, and anatomic changes and provide histologic correlates in eyes with intraretinal hyperreflective foci associated with acquired vitelliform lesions.
Design
Retrospective cohort study and imaging-histology correlation in a single donor eye.
Methods
participants : Patients with intraretinal hyperreflective foci and acquired vitelliform lesions from 2 tertiary referral centers were evaluated from January 2002 to January 2014. main outcome measures : The chronology of clinical and imaging features of retinal anatomic changes and the pattern of intraretinal hyperreflective foci migration were documented using spectral-domain optical coherence tomography (OCT). One donor eye with intraretinal hyperreflective foci was identified in a pathology archive by ex vivo OCT and was studied with high-resolution light and electron microscopic examination.
Results
Intraretinal hyperreflective foci were associated with acquired vitelliform lesions in 25 of 254 eyes (9.8%) with a strong female preponderance (86% of patients). Focal disruptions to the ellipsoid zone and external limiting membrane overlying the acquired vitelliform lesions were observed prior to the occurrence of intraretinal hyperreflective foci in 75% of cases. Histologic evaluation showed that intraretinal hyperreflective foci represent cells of retinal pigment epithelium origin that are similar to those found in the vitelliform lesions themselves and contain lipofuscin granules, melanolipofuscin granules, and melanosomes. The occurrence of intraretinal hyperreflective foci was not a significant determinant of final visual acuity ( P = .34), but development of outer retinal atrophy was ( P = .003).
Conclusions
Intraretinal hyperreflective foci associated with acquired vitelliform lesions are of retinal pigment epithelium origin, and the natural course and functional changes are described.
Acquired vitelliform lesions are associated with a range of macular disorders including age-related macular degeneration (AMD), adult-onset foveomacular dystrophy (AOFVD), cuticular drusen, chronic central serous chorioretinopathy (CSC), epiretinal membrane (ERM), vitreomacular traction (VMT), and deferoxamine toxicity. Gass was one of the earliest investigators to describe the clinical and histologic aspects of acquired vitelliform lesions, noting focal loss of retinal receptors, atrophy of the retinal pigment epithelium (RPE), and accumulation of pigment-laden cells in the subretinal space. Later histologic studies confirmed that acquired vitelliform lesions are characterized by subretinal material consisting of photoreceptor debris and pigment accompanied by an underlying atrophic RPE containing variable levels of lipofuscin granules. However, many questions regarding the natural course and pathophysiology of acquired vitelliform lesions remain unanswered.
Intraretinal hyperreflective foci have been observed in some eyes with acquired vitelliform lesions, and their clinical significance is unclear. It is not known if the occurrence of intraretinal hyperreflective foci has a bearing on long-term visual potential, or if retinal anatomic changes are associated with the formation of intraretinal hyperreflective foci. The cellular source of intraretinal hyperreflective foci is currently uncertain. Histopathologic correlations before the spectral-domain optical coherence tomography (SD OCT) era using medium-resolution paraffin sections and light microscopy were collectively noncommittal in answering this question. Investigators using these modalities described intraretinal hyperreflective foci as abnormal RPE, pigment-laden macrophages, or, simply, pigmented cells. Whether these cells originate from RPE or from monocytes has major ramifications for theories of disease pathogenesis, choice of experimental model systems, and therapeutic strategies.
The purpose of this study was to document the natural course of eyes with acquired vitelliform lesions and intraretinal hyperreflective foci using SD OCT and to define the retinal anatomic changes that are associated with intraretinal hyperreflective foci formation and migration. Building on prior findings by Freund and associates that disrupted photoreceptor outer segments account for vision loss in eyes with acquired vitelliform lesions, we explored the determinants of final visual acuity in these cases. Correlating high-resolution histology to ex vivo OCT imaging in an archival postmortem specimen provided new information regarding the origin and composition of intraretinal hyperreflective foci. The results of this study provide clinical and histologic information that is expected to aid the management and pathophysiological understanding of intraretinal hyperreflective foci and acquired vitelliform lesions.
Methods
This retrospective study was approved by the Institutional Review Board of the Manhattan Eye, Ear and Throat Hospital/North Shore Long Island Jewish Hospital, New York, New York and the University and Polytechnic Hospital La Fe, Valencia, Spain. The histology study was approved by the Institutional Review Board at the University of Alabama at Birmingham. Both studies complied with the Health Insurance Portability and Accountability Act of 1996 and followed the tenets of the Declaration of Helsinki.
Clinical Data Collection and Imaging
A retrospective chart review of all consecutive patients with acquired vitelliform lesions seen between January 1, 2002 and January 1, 2014 was conducted at the Vitreous Retina Macula Consultants of New York (New York, New York, USA) and the Department of Ophthalmology, University and Polytechnic Hospital La Fe (Valencia, Spain). An acquired vitelliform lesion was defined as a round, yellow-colored, subretinal structure that was hyperautofluorescent on fundus autofluorescence (FAF) imaging. The temporal course of intraretinal hyperreflective foci on SD OCT was studied. Patients with Best vitelliform dystrophy were excluded. Eyes that demonstrated macular atrophy, choroidal neovascularization, or subretinal fibrosis at the first clinic visit were excluded. Demographic information including sex, eye, and age at acquired vitelliform lesion diagnosis was collected. Clinical examination including best-corrected visual acuity (BCVA) as evaluated on Snellen charts on first visit; at acquired vitelliform lesion diagnosis; at first noted intraretinal hyperreflective foci; at follow-up intervals of 6 months, 1 year, and 2 years; and at last follow-up was recorded. Data from color fundus photography, FAF, and SD OCT imaging was recorded from baseline and follow-up visits. Color fundus photography and FAF images were obtained with a Topcon TRC-50DX fundus camera (Topcon USA, Paramus, New Jersey, USA). SD OCT images using 19–37 horizontal linear B-scans (15–30 degrees by 10–20 degrees) with 113–250 μm spacing were obtained using the Heidelberg Spectralis (Heidelberg Engineering, Heidelberg, Germany). Eye-tracking and image registration software were enabled during image acquisition at each visit.
Two independent reviewers at each of the 2 clinical institutions measured the choroidal thickness at each of these stages using the calipers on the SD OCT software (Spectralis version 3.2; Heidelberg Engineering). An average was calculated for each measurement. Central macular thickness (CMT) in the central 1-mm-diameter circle of the ETDRS thickness map was recorded with the Spectralis software. The following determinations were made using SD OCT in eyes that demonstrated intraretinal hyperreflective foci: interval between baseline visit and the time of acquired vitelliform lesion detection; interval between acquired vitelliform lesion detection and the time when migration of intraretinal hyperreflective foci was first observed; presence or absence of an acquired vitelliform lesion at the final visit; the integrity of the external limiting membrane and ellipsoid zone above the acquired vitelliform lesion; presence of intraretinal hyperreflective foci above the acquired vitelliform lesion and its location in proximity to the foveal center (categorized as foveal [beneath foveal center] or juxtafoveal [within 1–199 μm of foveal center]); and presence of outer retinal atrophy (loss of external limiting membrane and ellipsoid zone hyperreflective bands on SD OCT).
Statistical Analysis
All statistical analyses were performed using SPSS software version 22.0 (IBM Corp, Armonk, New York, USA). The Shapiro-Wilk test was used to verify normality of the parameters analyzed. The analysis of variance test was used to evaluate normally distributed parameters such as BCVA converted to the logarithm of the minimal angle of resolution (logMAR) at different time points and the Greenhouse-Geisser correction was used for asphericity. Independent t test was used to compare logMAR BCVA between eyes with and without intraretinal hyperreflective foci. The Friedman test was used for parameters that were not normally distributed, including CMT and choroidal thickness. The null hypothesis for BCVA, CMT, and choroidal thickness was no difference among time points. One-way random intraclass correlation coefficient tested reliability between measurers. Point-biserial correlation was used when comparing continuous variables such as logMAR BCVA to binary nominal variables such as the presence of outer retinal atrophy. The null hypothesis was no difference in BCVA between patients with and without outer retinal atrophy. A P value <.05 was deemed statistically significant. Results were expressed as mean ± standard deviation.
Ex Vivo Imaging and Histopathology
AMD eyes were identified by ex vivo imaging of >900 pairs of eyes accessioned in 1996–2012 for research purposes from nondiabetic white donors to the Alabama Eye Bank. Eyes were preserved by immersion in 1% paraformaldehyde and 2.5% glutaraldehyde in 0.1 M phosphate buffer following anterior segment excision. Starting in 2011, in creating the National Eye Institute–funded Project MACULA website of AMD histopathology, we performed ex vivo imaging on eyes remaining from this collection, prior to histologic processing and analysis, as described. In brief, eyes underwent multimodal ex vivo imaging including SD OCT using Spectralis (Heidelberg Engineering, Heidelberg, Germany). From each globe, an 8-mm-diameter full-thickness tissue punch containing the fovea and temporal portion of the optic nerve head was removed with a trephine. This punch was held in a closed chamber with a 60 diopter lens in the front mounted on a Spectralis. We performed a 30 × 20-degree SD OCT volume (143 scans, 30 μm apart, with Automatic Real Time averaging set at 25) and red-free scanning laser ophthalmoscopy. The right eye of an 85-year-old female donor recovered at 3:10 hours postmortem exhibiting ex vivo SD OCT signatures consistent with acquired vitelliform lesion and intraretinal hyperreflective foci in the setting of AMD was studied in detail. The tissue punch was postfixed by osmium tannic acid paraphenylenediamine to accentuate extracellular lipid in AMD lesions and impart polychromaticity in subsequent toluidine blue staining. Submicrometer-thick epoxy resin sections were reviewed and photodocumented with a 60× oil-immersion objective (numerical aperture = 1.4) and digital camera (XC10, Olympus; 1900× viewing magnification; available at http://projectmacula ). For transmission electron microscopy, the epoxy block was thin-sectioned at silver-gold and viewed with a 1200 EXII electron microscope (JEOL USA, Peabody, Massachusetts, USA) and an AMTXR-40 camera (Advanced Microscopy Techniques, Danvers, Massachusetts, USA). Images were composited with adjustments for exposure, contrast, and background color correction only (Photoshop CS6; Adobe Systems, San Jose, California, USA). We use the nomenclature of Zanzottera and associates for RPE morphology. A combined population of RPE-specific spindle-shaped melanosomes and melanolipofuscin can be recognized in this preparation.
Results
Of the 254 eyes with acquired vitelliform lesions from both centers, 25 eyes (9.8%) from 22 patients demonstrated intraretinal hyperreflective foci at diagnosis or during the follow-up interval on SD OCT and met the inclusion criteria for this study. Nineteen out of 22 patients (86.3%) were female. Average age ± standard deviation was 71.2 ± 15.1 years (range 30–90 years). Mean follow up ± standard deviation was 71.1 ± 33.0 months (range 16–147 months). Fourteen eyes had AOFVD, 8 eyes had AMD, 2 eyes had VMT, and 1 eye had ERM. Of the 229 eyes with acquired vitelliform lesions but without intraretinal hyperreflective foci that were screened, 108 (47.2%) had AMD, 90 (39.3%) had AOFVD, 15 (6.6%) had ERM, 10 (4.4%) had CSC, and 6 (2.6%) had VMT.
Illustrative Case
The natural course of an acquired vitelliform lesion and intraretinal hyperreflective foci is presented in the eye of a 61-year-old female patient with adult-onset foveomacular vitelliform dystrophy. At baseline visit, vitelliform material is seen as a collection of hyperreflective material with distinct borders, confined by the ellipsoid zone internally and the RPE–Bruch membrane band externally ( Figure 1 ). At this visit, the foveal contour of the overlying retina is preserved and cystoid spaces are not present. Eye-tracked SD OCT demonstrates focal consolidation of hyperreflective material within the acquired vitelliform lesion with movement of this material out of the vitelliform lesion in an inward direction into the overlying retina ( Figure 1 ). Focal disruption of the ellipsoid zone and external limiting membrane is evident at sites of intraretinal hyperreflective foci. Migrating material is characterized by hyperreflective foci (red arrowhead) that shadow the underlying tissue. Color photographs with corresponding SD OCT B-scans show focal pigment in the same location as the migrated hyperreflective material ( Figure 2 ). On FAF, small areas of intraretinal hyperreflective foci appear to be hypoautofluorescent but are difficult to evaluate owing to the large hyperautofluorescent acquired vitelliform lesion beneath ( Figure 2 ). SD OCT demonstrates that once within the retina, the material dissipates laterally within the Henle fiber layer and does not traverse inwardly beyond the outer plexiform layer ( Figure 1 ). Eventually the vitelliform lesion collapses, resulting in diffuse outer retinal atrophy ( Figure 1 , Bottom right).
Lesion Presence and Visual Acuity at Different Time Points
There was no difference between final BCVA of the 229 eyes with acquired vitelliform lesions that did not develop intraretinal hyperreflective foci and the 25 that did ( P = .522).
In the 10 eyes lacking a vitelliform lesion on the baseline visit, the mean time between the first visit and first noted vitelliform lesion was 55.3 ± 32.4 months (range 9–104 months). In these 10 eyes, intraretinal hyperreflective foci formed 5.2 ± 6.4 months (range 0–18 months) after vitelliform formation. In the 7 eyes with acquired vitelliform lesions at baseline, the mean time between first noted acquired vitelliform lesion and intraretinal hyperreflective foci was 27.3 ± 17.8 months (range 6–50 months). The remaining 8 eyes presented with both acquired vitelliform lesions and intraretinal hyperreflective foci.
Best-corrected visual acuity (logMAR) was 0.32 ± 0.16, 0.36 ± 0.20, and 0.46 ± 0.30 when the acquired vitelliform lesion was first observed, when the intraretinal hyperreflective foci were first observed, and at last follow-up, respectively ( P = .005). CMT was 295.2 ± 50.2 μm, 238.8 ± 103.5 μm, and 219.2 ± 94.8 μm, respectively, at these time points ( P = .010). Choroidal thickness was 254.3 ± 116.8 μm, 298.2 ± 50.0 μm, and 268.1 ± 50.2 μm, respectively ( P = .170). The measured intraclass correlation coefficient between reviewers for choroidal thickness at initial SD OCT was 0.867, which indicated good agreement.
Acquired Vitelliform Lesions, Intraretinal Hyperreflective Foci, and Outer Retinal Changes
When intraretinal hyperreflective foci were first noted, 18 eyes (72%) had only foveal intraretinal hyperreflective foci, 6 eyes (24%) had juxtafoveal intraretinal hyperreflective foci, and 1 eye (4%) had both. The intraretinal hyperreflective foci were not observed to migrate anterior to the outer plexiform layer in any of the eyes. Ellipsoid zone and external limiting membrane focal defects were identified in SD OCT scans prior to the occurrence of intraretinal hyperreflective foci in 8 of 12 eyes (75%). Twenty-three of 25 eyes (92%) showed disruption of the ellipsoid zone, and 18 eyes (72%) had focal disruption of the external limiting membrane above the acquired vitelliform lesion at any time during the clinical course. On last follow-up, 8 of 25 eyes (32%) had an intact ellipsoid zone and external limiting membrane. These changes seemed more pronounced at the location where intraretinal hyperreflective foci first occurred ( Figures 1 and 3 ). Anatomic features at the site of the acquired vitelliform lesion as imaged by SD OCT at various time points after presentation are summarized in the Table .
| Presentation of Vitelliform Lesion (n = 25) | 6 Months (n = 23) | 1 Year (n = 25) | 2 Years or Longer (n = 23) | |
|---|---|---|---|---|
| Presence of intraretinal hyperreflective foci | 8 (32%) | 19 (82.6%) | 24 (96%) | 23 (100%) |
| Intact ELM | 18 (72%) | 6 (26.1%) | 8 (32%) | 4 (17.4%) |
| Intact EZ | 15 (60%) | 5 (21.7%) | 3 (12%) | 1 (4.3%) |
| Collapse of vitelliform lesion | 0 (0%) | 4 (17.4%) | 9 (36%) | 11 (47.8%) |
| Outer retinal atrophy | 0 (0%) | 1 (4.3%) | 3 (12%) | 4 (17.4%) |
a Time point 6 months defined as scan closest to 6 months after presentation of vitelliform lesion between 6 months and 1 year. Time point 1 year taken from closest scan 1 year from presentation of vitelliform lesion between 1 and 2 years. Time point 2 years or longer taken from closest scan to 2 years from presentation of vitelliform lesion.
BCVA at last follow-up was significantly and negatively correlated with the presence of outer retinal atrophy (r = 0.57, n = 25, P = .0030), but not with partial or complete acquired vitelliform lesion collapse ( P = .16), resolution of intraretinal hyperreflective foci ( P = .34), or persistent focal disruption of the external limiting membrane ( P = .33) or ellipsoid zone ( P = .33).
Histopathology
Figures 4 and 5 show a donor eye with ex vivo imaging and histologic evidence of focal disruption of RPE and photoreceptors and intraretinal material of RPE origin, associated with AMD. This eye was chosen for analysis because ex vivo SD OCT scans revealed a phenotypic resemblance to the assembled clinical cases. Ex vivo color fundus photography revealed a subfoveal lesion that appeared yellow through the macular pigment with clumps of brown pigment ( Figure 4 , Top). By ex vivo SD OCT, a large RPE detachment (pigment epithelial detachment; PED) lacking internal reflectivity was apparent. Owing to its size, this PED corresponded to the subfoveal lesion seen by color photography, and the clumps corresponded to the RPE changes described below. PED contents were mostly lost in histologic preparation, and the AMD diagnosis was supported by soft druse material adherent to the overlying RPE basement membrane ( Figure 5 , Top left). The PED had several inward perturbations of the reflective RPE band ( Figure 4 , Middle) that were identifiable by subsequent histology ( Figure 4 , Bottom). From nasal to temporal (left to right in Figure 4 , Middle), these were focal disruptions (pink arrowhead), a thickening into the subretinal space with punctate hyperreflective spots (green arrowhead), and intraretinal hyperreflective foci (yellow arrowhead). The convexities of these perturbations matched concavities in the outer segment layer, which was artifactually detached. The zones of reflectivity correlated to epithelial RPE with shed granule aggregates at the basal aspect, an explosion of granules and nucleated pigmented cells in the subretinal space, and pigmented cells with nuclei within the retina, respectively. In these areas, abundant RPE granules (lipofuscin, melanolipofuscin, and melanosomes) predominated ( Figure 5 , bottom 3 panels). Subretinal RPE-derived material ( Figure 4 , Bottom; Figure 5 , Top left) was readily distinguishable from both photoreceptor outer segments and subretinal drusenoid deposits present elsewhere in this eye ( Supplemental Figure , available at AJO.com ). Evidence of photoreceptor degeneration included shortened outer segments overlying the RPE perturbations ( Figure 4 , Bottom), outer segments with disoriented disks delimiting granule explosions ( Figure 5 , Bottom middle), and absent outer segments with shortened inner segments near the intraretinal RPE ( Figure 5 , Bottom middle).
