- Email: [email protected]
Familial Exudative Vitreoretinopathy
Familial Exudative Vitreoretinopathy Spectral-Domain Optical Coherence Tomography of the Vitreoretinal Interface, Retina, and Choroid Yoshihiro Yonekawa, MD, Benjamin J. Thomas, MD, Kimberly A. Drenser, MD, PhD, Michael T. Trese, MD, Antonio Capone, Jr., MD Purpose: The in vivo microstructural features of familial exudative vitreoretinopathy (FEVR) have not been well described. We present new anatomic features of FEVR with functional and genetic correlations. Design: Consecutive, retrospective, observational case series. Participants: Patients with FEVR treated from 2009 to 2014. Methods: We identified 346 patients with FEVR. Those imaged with spectral-domain optical coherence tomography (SD OCT) with or without enhanced depth imaging (EDI) were included, and images were correlated with best-corrected visual acuity (BCVA), widefield angiography, fundus autofluorescence (AF), and wnt signaling pathway mutations. Main Outcome Measures: Exploratory SD OCT findings and BCVA. Results: A total of 225 imaging sessions were acquired in 74 eyes from 41 patients. Mean age was 19.0 years. Sixty-seven eyes (91%) had interpretable images, of which 50 (75%) had anomalous microstructural findings; all eyes with FEVR severity of stage 2 or greater had abnormalities. A broad spectrum of features were identified: various forms of posterior hyaloidal organization, vitreomacular traction (VMT), vitreopapillary traction, vitreo-fold traction, vitreo-laser scar adhesion, diminished foveal contour, persistent fetal foveal architecture, cystoid macular edema (CME), intraretinal exudates and subretinal lipid aggregation, dry or edematous radial folds, and disruption of the ellipsoid zone. Mean foveal, central macular, and choroidal thicknesses were 305145 mm, 337160 mm, and 21664 mm, respectively. In stages 1 to 2, greater foveal and central macular thicknesses (Rho ¼ 0.493, 0.544, respectively; both P < 0.001) correlated with poorer BCVA, but not choroidal thickness (Rho ¼ 0.032; P ¼ 0.868). Posterior hyaloidal organization (P < 0.001), VMT (P < 0.001), CME (P < 0.001), exudation (P < 0.001), and disruption of the ellipsoid zone (P < 0.001) were associated with poorer BCVA. Disruption of the ellipsoid zone (b ¼ 0.699; P < 0.001) and posterior hyaloidal organization (b ¼ 0.289; P ¼ 0.011) retained significance in multivariate modeling (R2 ¼ 0.627; P < 0.001). Spectral-domain OCT detected all cases of angiographic edema and areas of outer retinal dysfunction that were hypoautofluorescent on AF. Microstructural-genetic associations were not identified. Conclusions: Spectral-domain OCT imaging identified microstructural anomalies in the majority of patients with FEVR. Ophthalmology 2015;-:1e8 ª 2015 by the American Academy of Ophthalmology.
Familial exudative vitreoretinopathy (FEVR) is a hereditary abnormality in retinal vascular development caused by wnt signaling defects.1e3 Anomalous or incomplete retinal vascularization is the primary pathology, with varying degrees of secondary peripheral ischemia and subsequent complications.4 Classic ophthalmoscopic and angiographic findings include peripheral avascular retina flanked posteriorly by abnormally branching vessels, retinal neovascularization, exudation, dragging of vasculature, hyaloidal contraction, and tractional retinal detachment.4e6 Fluorescein angiography is critical in the diagnosis and management of FEVR, and the recent advent of widefield angiography has permitted further characterization of angiographic phenotypes. This has allowed earlier diagnosis and treatment of patients, and, in particular, their family members, many of whom demonstrate milder phenotypic variants.7 2015 by the American Academy of Ophthalmology Published by Elsevier Inc.
Identifying milder phenotypes remains a significant challenge: although index patients commonly present during childhood with advanced disease, less-advanced cases of FEVR may remain undetected in pedigrees for generations because of variable expressivity and seemingly normal gross anatomy of these mildly affected individuals.8 Perhaps understanding the more subtle changes in these milder stages could allow for improved identification of patients, who could then benefit from widefield angiography for more definitive diagnosis. However, the in vivo microstructural anatomy of FEVR remains largely unknown. Early histopathologic studies reported features of enucleated eyes with end-stage disease,9e11 but histopathologic diagnosis is impractical. The advent of optical coherence tomography (OCT) allows high-resolution in vivo evaluation of posterior segment microstructures. With http://dx.doi.org/10.1016/j.ophtha.2015.07.024 ISSN 0161-6420/15
1
Ophthalmology Volume -, Number -, Month 2015 spectral-domain OCT (SD OCT) and enhanced depth imaging (EDI), our understanding and treatment paradigms for many macular diseases have been enriched. However, OCT imaging in FEVR has remained largely unexplored, presumably because of the rarity of the disease and the dominant focus of previous studies on peripheral findings.5,7,8,12e14 We hypothesized that FEVR would be amenable to exploratory SD OCT analysis, given its known effects on the vitreoretinal interface. We also hypothesized, on the basis of our collective clinical and surgical experiences, that there are heretofore underappreciated pathologic microstructural macular features in FEVR. We report the findings of these analyses and identify microstructural features of FEVR; their functional correlations with vision, angiographic, and autofluorescent findings; and the new implications for current and future management strategies.
Methods This study is a single-center, retrospective, noncomparative case series of patients with a clinical or genetic diagnosis of FEVR who were treated at Associated Retinal Consultants. Participants were identified via billing codes, surgical logs, genetic databases, and image banks, during the 6-year period from January 1, 2009, to December 31, 2014. Patients who underwent SD OCT (Spectralis, Heidelberg Engineering, Heidelberg, Germany) imaging with or without EDI were included. Those with only time-domain OCT were excluded, as were patients with concurrent non-FEVR vitreoretinal pathology. Institutional Review Board/Ethics Committee approval was obtained. The study complied with the Health Insurance Portability and Accountability Act of 1996 and adhered to the tenets of the Declaration of Helsinki. Data collection included demographics, including birth history, at the first imaging session, best-corrected visual acuity (BCVA) using automated refracting Snellen acuity charts, highest FEVR staging, laser and surgical treatment history, and total number of imaging sessions. The first imaging session and BCVA were recorded, but all imaging sessions were reviewed to confirm imaging findings. Staging was determined as previously described: stage 1, avascular periphery; stage 2, avascular periphery with neovascularization; stage 3, macula-sparing retinal detachment; stage 4, macula-involving retinal detachment; stage 5, complete retinal detachment.4,8 Spectral-domain OCT/EDI images were analyzed for vitreoretinal interface, preretinal, intraretinal, subretinal, and choroidal abnormalities on the basis of image interpretations made during clinical encounters by the senior authors (K.A.D., M.T.T., A.C.), reviewed systematically by 1 author (Y.Y.), and examined by all authors if there were any discrepancies. Foveal thickness was measured manually from the internal limiting membrane to the outer boundary of the retinal pigment epithelium (RPE). Central macular thickness was defined as the average thickness of the 1-mm diameter circle centered on the fovea, with autosegmentations manually corrected. Subfoveal choroidal thickness in EDI images was measured from the outer border of the RPE to the choroidal-scleral junction.15 Macular measurements were excluded for stages 3 to 5 because of poor patient fixation that often precluded macular volume scans and the unreliability of determining precise foveal centers because of dragged anatomy. Corresponding BCVA, color photography, fluorescein angiography, and fundus autofluorescence (AF) were examined for functional correlations. Photography and fluorescein angiography
2
Table 1. Demographic and Clinical Features of Patients with Familial Exudative Vitreoretinopathy undergoing Spectral Domain Optical Coherence Tomography Feature Age, yrs (n ¼ 41 patients) Mean (median, range) Sex Male Female Eye imaged (n ¼ 74 eyes) Right Left Highest FEVR staging Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Previous treatment None Laser only Laser and anti-VEGF Incisional surgery
No. (%) 19.0 (15.4, 2.4e57.0) 21 (51) 20 (49) 40 (54) 34 (46) 33 19 2 17 3
(45) (26) (3) (23) (4)
15 38 4 17
(20) (51) (5) (23)
FEVR ¼ familial exudative vitreoretinopathy; VEGF ¼ vascular endothelial growth factor.
during examinations under anesthesia were obtained using the RetCam II (Clarity Medical Systems, Pleasanton, CA) and the 130 widefield D1300 lens. The 200Tx (Optos, Marlborough, MA) was used for ultra-widefield (200 ) photography, fluorescein angiography, and fundus AF for participants who could tolerate outpatient angiography. Genetic sequencing is offered for our patients with suspected FEVR. The exons of FZD4 (11q14.2) and NDP (Xp11.3) are initially sequenced. If mutations are not identified, TSPAN12 (7q31.31) would be sequenced and LRP5 (11q13.2) as needed. Detailed sequencing techniques have been published.2,3 Snellen BCVAs were converted to logarithm of the minimum angle of resolution units for statistical analyses. Spearman’s rank correlation was used to determine the correlation coefficients of continuous variables. Analysis of variance testing was performed to identify differences in gene mutations and dichotomous independent variables. Spectral-domain OCT features associated with poorer BCVA were identified using univariate linear regression, and multivariate linear modeling was performed using variables with P < 0.1 during univariate analysis. Statistical tests were 2-tailed, and significance was defined as P < 0.05. Stata version 9.0 (StataCorp LP, College Station, TX) was used for statistical analyses.
Results A total of 346 patients with FEVR were identified during the study period. Within this cohort, 74 eyes from 41 patients (12%) underwent 225 imaging sessions with SD OCT/EDI imaging. The demographic and clinical features are summarized in Table 1. None of the patients had a history of premature birth. All patients with stage 2 or higher had prior treatment. Interpretable SD OCT images were acquired in 67 eyes (91%). Five noninterpretable imaging sessions were for stage 4 eyes with poor vision and inability to fixate, and 2 were from a patient too young to participate fully. Of the eyes with interpretable images, abnormal findings were identified in 50 eyes (75%). All eyes
Yonekawa et al
Microstructural Features of FEVR
Table 2. Foveal, Central Macular, and Subfoveal Choroidal Thicknesses in Patients with Familial Exudative Vitreoretinopathy Abnormal Microstructure Observed All stages Stage 1 Stage 2 Stages 3e5
50 17 17 17
of of of of
67* (75%) 33 (52%) 17 (100%) 17 (100%)
Mean No. of Abnormalities (Range) 3 1 3 5
Mean Foveal Thickness mm (Range) (n [ 59)
(0e12) (0e5) (1e6) (2e12)
305 271 358 327
(118e973) (196e710) (126e973) (118e545)
Mean Central Macular Thickness mm (Range) (n [ 45) 339 302 401 455
(160e1209) (242e600) (160e1209) (389e516)
Mean Subfoveal Choroidal Thickness mm (Range) (n [ 39) 216 224 220 190
(57e363) (57e306) (107e315) (84e363)
*Of eyes with interpretable images.
(100%) that were stage 2 or higher had 1 or more microstructural abnormalities. Foveal, central macular, and subfoveal choroidal thicknesses are summarized in Table 2. The following 14 findings were noted: posterior hyaloidal organization, vitreoschisis, vitreomacular traction (VMT), vitreopapillary traction, vitreo-fold traction, vitreo-laser scar traction, peripheral traction, retinal fold, macular edema, extramacular edema, intraretinal exudates, persistent inner retinal layers at the fovea, external limiting membrane/ ellipsoid zone/interdigitation zone disruption, and retinal pigment epithelial atrophy.
Vitreoretinal Interface Hyaloidal organization is a key feature of FEVR. Posterior hyaloidal organization has various appearances by SD OCT, ranging from thinner epiretinal membrane-like hyperreflective layers (Fig 1A) to much thicker opacities (Fig 1B). The anomalous posterior vitreous also took the form of vitreoschisis-like sheets (Fig 1C). Posterior hyaloidal organization was evident in 30 eyes (46%): 7 (25%) in stage 1, 14 (74%) in stage 2, and 11 (65%) in stages 3 to 5. These abnormal vitreal scaffolds can cause VMT (Fig 1D) and vitreopapillary traction (Fig 1E), as well as vitreoretinal adhesion in the periphery, especially to laser scars (Fig 1F).
Intraretinal Pathology Persistent inner foveal fetal architecture was observed in 13 eyes (20%): 3 (11%) in stage 1, 9 (47%) in stage 2, and 1 (6%) in stages 3 to 5 (Fig 2A and B). This finding was characterized by persistence of the nerve fiber, ganglion cell, inner plexiform, inner nuclear, and outer plexiform layers. It was a bilateral phenomenon in 5 of 8 patients (63%). Of note, the 2 eyes with isolated persistent inner foveal retinal layers without concurrent
vitreoretinal interface or macular pathology had 20/25 BCVA. Diminished foveal contour also was observed in many stage 1 and 2 eyes, although this was not quantified. Macular edema was seen in 24 eyes (36%): 4 (14%) in stage 1, 9 (47%) in stage 2, and 11 (65%) in stages 3 to 5 (Fig 2C). Extramacular retinal edema also was observed (Fig 2D). Intraretinal exudation accompanied macular edema in 15 eyes (63%), which was seen as hyperreflective foci, often with shadowing, that corresponded to lipid exudation in color photographs (Fig 2E). Subretinal lipid aggregation also was observed (Fig 2E). Disruption of the ellipsoid zone was seen in 25 eyes (37%): 3 (4%) in stage 1, 5 (26%) in stage 2, and 17 (100%) in stage 3 to 5 eyes. The disruptions were patchy or broad (Fig 2F). Eyes with ellipsoid zone abnormalities also had concurrent disruptions of the external limiting membrane and interdigitation zone.
Advanced Pathology Volume scans were not achieved in the majority of stage 3 to 5 eyes, but single line-scans were usually possible and allowed determination of whether radial retinal folds were dry (no edema, Fig 3A) or wet (with edema, Fig 3B). Hyaloidal tractional forces on radial retinal folds could be imaged (Fig 3C), as well as shallow retinal detachments that may be difficult to diagnose by ophthalmoscopy in children (Fig 3D). Spectral-domain OCT was also able to identify pockets of attached retina in the midst of seemingly broad tractional retinal detachments (Fig 3E).
Functional Correlations The macular edema in FEVR is associated with leakage on angiography (Fig 4A and B). The sensitivity of SD OCT in detecting angiographically confirmed that macular and peripheral edema
Figure 1. Variable presentations and effects of posterior hyaloidal organization in familial exudative vitreoretinopathy (FEVR). A, Hyperreflective thin epiretinal membrane-like appearance (arrows). B, Broader matrices of hyaloidal organization. C, Vitreoschisis-like sheets of vitreous (inset, arrows). D, Vitreomacular traction (VMT). E, Vitreopapillary traction (arrows). F, Peripheral vitreoretinal adhesion (red arrows) to an area of photocoagulation (white arrow).
3
Ophthalmology Volume -, Number -, Month 2015
Figure 2. Intraretinal microstructural features of familial exudative vitreoretinopathy (FEVR). Persistent inner foveal retinal layers in the right (A) and left (B) eyes of a patient with stage 2 FEVR. The ganglion cell layer, inner plexiform layer, inner nuclear layer, and outer plexiform layer are retained in the fovea. C, Cystoid macular edema (CME). D, Peripheral retina with laser scars (red arrows) and overlying thin retina are shown on the left. Toward the right in the area without laser treatment, there is retinal thickening with mild intraretinal exudation (yellow arrows). E, Intraretinal exudation (red arrow) and subretinal lipid aggregation (yellow arrow). F, Tapering and disruption of the external limiting membrane, ellipsoid zone, interdigitation zone, and retinal pigment epithelium (RPE), between the arrows, in a broad fashion. EZ ¼ ellipsoid zone; GCL ¼ ganglion cell layer; INL ¼ inner nuclear layer; IPL ¼ inner plexiform layer; IZ ¼ interdigitation zone; OPL ¼ outer plexiform layer.
was 100%. Spectral-domain OCT revealed 3 additional eyes with cystic hyporeflectivities that did not correlate with leakage or staining on angiography, but these cases of retinal thickening were secondary to VMT (Figs 1D, 4D and E). The RPE abnormalities with overlying photoreceptor dysfunction were associated with focal areas of hypo-AF (Fig 4F). Fluorescein angiography AF highlighted areas of prior photocoagulation (Fig 4G). Retinal pathology, such as fronds of neovascularization and hemorrhage, blocked underlying fundus AF. Radial folds demonstrated a bed of underlying hypo-AF, which also affected flanking areas near the base of the fold (Fig 4H). Spectral-domain OCT identified all areas of true (not blocked) hypo-AF as outer retinal/RPE abnormalities.
Visual Correlations Visual acuities ranged from 20/15 to no light perception. Mean logMAR visual acuity was 0.09 (Snellen equivalent 20/25) for stage 1, 0.65 (20/89) for stage 2, and 1.49 (20/618) for stages 3 to 5. The BCVA correlated significantly with FEVR stage (Rho ¼ 0.816, P < 0.001) and the number of abnormal findings on SD OCT (Rho ¼ 0.834,
P < 0.001). For microstructural phenotypes in stages 1 and 2, greater foveal thickness (Rho ¼ 0.493, P < 0.001) and central macular thickness (Rho ¼ 0.544, P < 0.001) correlated with poorer BCVA, but not choroidal thickness (Rho ¼ 0.032, P ¼ 0.868). Of the microstructural phenotypes in stages 1 and 2, univariate analysis demonstrated that posterior hyaloidal organization (P < 0.001), VMT (P < 0.001), macular edema (P < 0.001), exudation (P < 0.001), and disruption of the ellipsoid zone (P < 0.001) were associated with poorer BCVA. Disruption of the ellipsoid zone (b ¼ 0.506; P < 0.001) and posterior hyaloidal organization (b ¼ 0.289; P ¼ 0.011) retained significance in simultaneous multivariate modeling (adjusted R2 ¼ 0.627; P < 0.001).
Genetic Sequencing Genetic sequencing was performed on 27 patients. Wnt signaling pathway mutations were detected in 8 (30%): 5 with FZD4 mutations (stages 1e5), 2 with NDP mutations (stages 1e2), and 1 with a TSPAN12 mutation (stage 1). There were no genotypicmicrostructural phenotypic associations for this small subset of patients.
Figure 3. Microstructural pathology in advanced familial exudative vitreoretinopathy (FEVR). A, Dry radial retinal fold (white arrow) without surrounding edema (inset, red arrow). B, Wet radial retinal fold: The left red arrow points to the upslope of the fold; the right red arrow appears to be pointing to a downward slope, but this is a continuation of the upward slope whose image was reflected down because it was beyond the image capturing range. The fold is flanked by intraretinal fluid (red asterisks). The white arrow points to an artifact from beyond the image capturing range. C, Hyaloidal traction on a radial retinal fold (arrows). D, Shallow retinal detachment. Asterisks indicate the area of subretinal fluid. E, Tractional retinal detachment with focally attached retina (red arrow points to attached retina, yellow arrows point to flanking detached retina).
4
Yonekawa et al
Microstructural Features of FEVR
Figure 4. Fluorescein angiographic and fundus autofluorescence (AF) correlations of microstructural features of familial exudative vitreoretinopathy (FEVR). A, Cystoid macular edema in FEVR demonstrates angiographic leakage (B, 0:48, C, 3:12). D, Infrared image of posterior hyaloidal organization causing vitreomacular traction (VMT) in the eye from Figure 1F, (E) which does not leak on fluorescein angiography because of its mechanical mechanism. F, Outer retinal abnormalities in the eye from Figure 2F results in a corresponding area of hypo-AF. G, Laser photocoagulation scars show sharply demarcated hypo-AF. H, Radial retinal folds have underlying hypo-AF.
5
Ophthalmology Volume -, Number -, Month 2015
Discussion Familial exudative vitreoretinopathy is a relatively rare, yet underdiagnosed disease entity that we have yet to fully elucidate. The seminal 1969 report by Criswick and Schepens12 that identified FEVR as a distinct clinical entity described vitreous membranes, macular heterotopia, exudation, retinal detachment, and neovascularization. Angiographic studies subsequently demonstrated the avascular peripheral retina and anomalous leaking branching vasculature,13 and recent findings using widefield angiography have advanced this angiographic characterization.7 Early histologic studies provided information on end-stage pathology from enucleated eyes,9e11 but little was known about the microstructural features of less-advanced FEVR phenotypes. Only 4 eyes from 3 patients in 2 brief reports have been described with OCT, which showed exudates,16 vitreoretinal adhesion,17 and preretinal vitreal deposits.17 In this report of 72 eyes from 40 patients, we report a broad spectrum of microstructural abnormalities in FEVR, the majority of which have not been reported.
Posterior Hyaloidal Organization Dysgenic vitreous is a salient feature of all pediatric vitreoretinopathies, including FEVR, Norrie disease, persistent fetal vasculature, and retinopathy of prematurity (ROP).6 Clinical and surgical experience have demonstrated that the posterior hyaloid in these cases is pathologically adherent and contractile, serves as a scaffold for fibrovascular proliferation,4 and can cause various forms of acute and late tractional sequelae.6 Our study validates these observations, demonstrating a wide range of its microanatomic manifestations in FEVR that increased in frequency and severity with advancing stage. Only a small number of stage 1 eyes demonstrated a thicker, more disorganized posterior hyaloidal matrix; this finding was more common (and more accentuated) in the higher stages and led to VMT and macular edema. These tractional membranes are not the typical epiretinal membranes seen after posterior vitreous detachments, because they originate from abnormal vitreous that becomes pathologic as the cicatricial disease progresses. We therefore recommend the term posterior hyaloidal organization/contraction. The structured posterior hyaloid and its clinically relevant contraction have several implications based on our surgical experiences. First, it explains that hyaloidal contraction plays a prominent role in the presenting pathology, and it is a target for treatment during primary surgical repairs. The onion-layer-like hyaloidal sheets may be difficult to visualize intraoperatively, so vitreous visualization aides may facilitate the dissection. The sheets are also broader than conventional ERMs. Second, the clinical consequence of a multilayered posterior hyaloid is the high incidence of recurrent premacular proliferation with contraction after vitrectomy for the same, despite the intraoperative impression that the premacular surface had been pristinely peeled. Third, contraction of the posterior hyaloid may occur after peripheral ablation or anti-VEGF treatment. The configuration and potential contractile vectors should be
6
examined before initiating treatment. Fourth, posterior hyaloidal organization of FEVR and other vitreoretinopathies should have a high place in the differential diagnosis of pediatric “ERM,” along with superficial combined hamartoma. In addition to VMT, we also frequently found vitreopapillary traction in all stages of disease. This is not surprising, given the strong adherence of peripapillary posterior vitreous cortex and the propensity of the prepapillary space for cellular proliferation. Further studies would be required to determine how the vitreopapillary tractional vectors contribute to macular and peripheral pathology in FEVR. From our surgical experience, we know that prepapillary stalk tissue is a key element in dissection of advanced tractional FEVR detachments. We also identified hyaloidal adhesion to laser scars and focal areas of traction in the periphery. Future developments in OCT technology to allow peripheral scanning will undoubtedly facilitate more precise delineation of these tractional vectors, which may enhance surgical therapy aimed at removing these vectors.4 Intraoperative OCT technology, paired with our findings, also may facilitate hyaloidal dissection and identification of the hyaloido-retinal surgical plane.
Cystoid Macular Edema Macular edema was first noted in angiographic descriptions of FEVR by Feldman et al.18 Of note, no subsequent reports explain this important finding. Macular edema was seen in 29% of our cohort and correlated with poorer BCVA. There were 2 forms of macular edema observed: first, the more frequently observed cystoid macular edema (CME) with concurrent leakage on fluorescein angiography. This form suggests capillary permeability, likely due to ischemia-driven or inflammatory cytokine-induced vascular decompensation of already anomalous vessels.19 Second, SD OCT detected cystic intraretinal spaces in a small subset of eyes that did not correlate with any angiographic leakage. These maculae all exhibited varying degrees of VMT, indicating that mechanical elevation of the macula was the root cause of these structural cystoid changes.20 Intraretinal cystic spaces also have been identified in ROP,21e26 but they likely represent a different pathogenesis compared with the CME seen in FEVR, for several reasons: first, cystic changes are seen in approximately half of infants with ROP, but angiographic studies have not identified leakage.21,22,26 Second, the intraretinal changes in ROP are transient and resolve over time. Third, most studies do not show an association with ROP severity or systemic comorbidities.23e25 Taken together, these studies seem to suggest that these cystoid changes are structural and developmental in origin. In contrast, the CME identified in FEVR correlates with angiographic leakage, indicating a truly pathophysiologic, reactive macular edema.
Inner Retinal Developmental Abnormalities and Outer Retinal Dysfunction Persistence of the inner retinal layers in the fovea was observed in 20% of stage 1 and 2 eyes. This developmental abnormality does not appear to be visually significant in
Yonekawa et al
Microstructural Features of FEVR
isolation, when underlying outer retinal structures remain intact. An analogous finding has been reported in OCT studies of infants with ROP.25,27e32 Although the morphology generally appears to normalize with age,28,32 persistent inner retinal layers and diminished foveal contour are also reported in older children with histories of ROP.27,29,30 Future studies are warranted to understand this microstructural phenotype further. As with many other macular pathologies, atrophy of the outer retinal structures was the strongest independent risk factor for poor visual acuity in eyes with FEVR and was associated with attenuated fundus AF. The external limiting membrane, ellipsoid zone, and interdigitation zone were affected in the majority of these eyes. Outer retinal deterioration was accompanied by concomitant pathology in all cases, whether it was macular edema, exudation, VMT, or previous retinal detachment. Because FEVR is (presumably) not a primary degeneration of photoreceptors, these outer retinal changes are likely secondary to damage from overlying intraretinal and vitreoretinal interface pathology. Our study introduces a number of SD OCT microstructural findings in patients with FEVR, all of which can serve to improve the diagnosis, treatment, and pathophysiologic understanding of this complex hereditary disease in various ways. We offer a less than exhaustive list: Diagnosis: If a pediatric patient presents with 1 or more of the described SD OCT findings, FEVR should be included in the differential diagnosis, and a thorough peripheral retinal examination and widefield fluorescein angiography should be considered. We anticipate that other vitreoretinopathies may harbor similar microstructural changes. Earlier diagnosis will lead to appropriate screening examinations and prompt treatment if necessary, theoretically resulting in improved outcomes. Treatment: The detailed identification of pathologic microanatomy can potentially open new avenues for treatment; namely, targeted addressing of VMT, vitreopapillary traction, CME, intraretinal exudation, subretinal lipid aggregation, wet radial folds, and subclinical retinal detachments. Optimal management strategies will have to be examined in future studies, including a specific focus on the prevention of macular photoreceptor damagedthe common end point of impaired vision, as seen in this study. Pathophysiology: The provision of new phenotypic end points generates a broader context for elaborating the effects of wnt signaling defects. Although this current study was underpowered for genotypeephenotype correlations, future studies may determine certain phenotypes to be associated with certain mutations. For example, our experience has suggested that FZD4 mutations may lead to more macular microvascular abnormalities, whereas tractional forces seem to play a bigger role in NDP mutations (Nudleman et al, unpublished data, April 2015). Structural choroidal changes were not consistently observed in this cohort, but future longitudinal studies may also be able to associate variable choroidal features with disease progression or treatment.
Study Limitations First, the study design was retrospective and it has its resultant inherent biases. Second, because the data were derived from a surgical referral practice, our patient cohort may be skewed toward higher stages of FEVR (and, thus, a higher frequency of demonstrable findings). Third, intraoperative OCT data from the youngest patients were not available, confining our observations to a subset of patients whose mean age is higher than that of the overall FEVR cohort. Also, the microstructural features identified in this study are in patients with FEVR, but we suspect that they are not necessarily unique to FEVR. We would not be surprised that other pediatric vitreoretinopathies have similar findings. Finally, this is a cross-sectional study, and longitudinal data are not within the study design. Therefore, further studies would have to examine the evolution of these SD OCT findings and analyze the treatment effects and outcomes. In summary, we report that SD OCT imaging allows identification of dysgenic microstructural anomalies in the majority of patients with FEVR, introducing a new dimension in the diagnosis and treatment of vitreoretinal pathology associated with FEVR. We accordingly recommend SD OCT imaging for patients with FEVR when possible, because it may serve to incorporate each patient’s individual microstructural disease features into a more focused management plan.
References 1. Drenser KA, Dailey W, Capone A, Trese MT. Genetic evaluation to establish the diagnosis of X-linked familial exudative vitreoretinopathy. Ophthalmic Genet 2006;27:75–8. 2. Wu WC, Drenser K, Trese M, et al. Retinal phenotypegenotype correlation of pediatric patients expressing mutations in the Norrie disease gene. Arch Ophthalmol 2007;125: 225–30. 3. Drenser KA, Dailey W, Vinekar A, et al. Clinical presentation and genetic correlation of patients with mutations affecting the FZD4 gene. Arch Ophthalmol 2009;127:1649–54. 4. Pendergast SD, Trese MT. Familial exudative vitreoretinopathy. Results of surgical management. Ophthalmology 1998;105:1015–23. 5. Miyakubo H, Hashimoto K, Miyakubo S. Retinal vascular pattern in familial exudative vitreoretinopathy. Ophthalmology 1984;91:1524–30. 6. Joshi MM, Ciaccia S, Trese MT, Capone A. Posterior hyaloid contracture in pediatric vitreoretinopathies. Retina 2006;26: S38–41. 7. Kashani AH, Brown KT, Chang E, et al. Diversity of retinal vascular anomalies in patients with familial exudative vitreoretinopathy. Ophthalmology 2014;121:2220–7. 8. Kashani AH, Learned D, Nudleman E, et al. High prevalence of peripheral retinal vascular anomalies in family members of patients with familial exudative vitreoretinopathy. Ophthalmology 2014;121:262–8. 9. Brockhurst RJ, Albert DM, Zakov ZN. Pathologic findings in familial exudative vitreoretinopathy. Arch Ophthalmol 1981;99:2143–6. 10. Nicholson DH, Galvis V. Criswick-Schepens syndrome (familial exudative vitreoretinopathy). Study of a Colombian kindred. Arch Ophthalmol 1984;102:1519–22.
7
Ophthalmology Volume -, Number -, Month 2015 11. Boldrey EE, Egbert P, Gass JD, Friberg T. The histopathology of familial exudative vitreoretinopathy. A report of two cases. Arch Ophthalmol 1985;103:238–41. 12. Criswick VG, Schepens CL. Familial exudative vitreoretinopathy. Am J Ophthalmol 1969;68:578–94. 13. Canny CL, Oliver GL. Fluorescein angiographic findings in familial exudative vitreoretinopathy. Arch Ophthalmol 1976;94:1114–20. 14. Ranchod TM, Ho LY, Drenser KA, et al. Clinical presentation of familial exudative vitreoretinopathy. Ophthalmology 2011;118:2070–5. 15. Imamura Y, Fujiwara T, Margolis R, Spaide RF. Enhanced depth imaging optical coherence tomography of the choroid in central serous chorioretinopathy. Retina 2009;29:1469–73. 16. Day S, Maldonado RS, Toth CA. Preretinal and intraretinal exudates in familial exudative vitreoretinopathy. Retina 2011;31:190–1. 17. Shimouchi A, Takahashi A, Nagaoka T, et al. Vitreomacular interface in patients with familial exudative vitreoretinopathy. Int Ophthalmol 2013;33:711–5. 18. Feldman EL, Norris JL, Cleasby GW. Autosomal dominant exudative vitreoretinopathy. Arch Ophthalmol 1983;101:1532–5. 19. Zhou Y, Wang Y, Tischfield M, et al. Canonical WNT signaling components in vascular development and barrier formation. J Clin Invest 2014;124:3825–46. 20. Johnson MW. Tractional cystoid macular edema: a subtle variant of the vitreomacular traction syndrome. Am J Ophthalmol 2005;140:184–92. 21. Yokoi T, Hiraoka M, Miyamoto M, et al. Vascular abnormalities in aggressive posterior retinopathy of prematurity detected by fluorescein angiography. Ophthalmology 2009;116:1377–82. 22. Lepore D, Molle F, Pagliara MM, et al. Atlas of fluorescein angiographic findings in eyes undergoing laser for retinopathy of prematurity. Ophthalmology 2011;118:168–75.
23. Vinekar A, Avadhani K, Sivakumar M, et al. Understanding clinically undetected macular changes in early retinopathy of prematurity on spectral domain optical coherence tomography. Invest Ophthalmol Vis Sci 2011;52:5183–8. 24. Maldonado RS, O’Connell R, Ascher SB, et al. Spectraldomain optical coherence tomographic assessment of severity of cystoid macular edema in retinopathy of prematurity. Arch Ophthalmol 2012;130:569–78. 25. Dubis AM, Subramaniam CD, Godara P, et al. Subclinical macular findings in infants screened for retinopathy of prematurity with spectral-domain optical coherence tomography. Ophthalmology 2013;120:1665–71. 26. Henaine-Berra A, Garcia-Aguirre G, Quiroz-Mercado H, Martinez-Castellanos MA. Retinal fluorescein angiographic changes following intravitreal anti-VEGF therapy. J AAPOS 2014;18:120–3. 27. Recchia FM, Recchia CC. Foveal dysplasia evident by optical coherence tomography in patients with a history of retinopathy of prematurity. Retina 2007;27:1221–6. 28. Maldonado RS, O’Connell RV, Sarin N, et al. Dynamics of human foveal development after premature birth. Ophthalmology 2011;118:2315–25. 29. Wu WC, Lin RI, Shih CP, et al. Visual acuity, optical components, and macular abnormalities in patients with a history of retinopathy of prematurity. Ophthalmology 2012;119: 1907–16. 30. Yanni SE, Wang J, Chan M, et al. Foveal avascular zone and foveal pit formation after preterm birth. Br J Ophthalmol 2012;96:961–6. 31. Villegas VM, Capó H, Cavuoto K, et al. Foveal structurefunction correlation in children with history of retinopathy of prematurity. Am J Ophthalmol 2014;158:508–512.e2. 32. Vajzovic L, Rothman AL, Tran-Viet D, et al. Delay in retinal photoreceptor development in very preterm compared to term infants. Invest Ophthalmol Vis Sci 2015;56:908–13.
Footnotes and Financial Disclosures Originally received: April 26, 2015. Final revision: July 9, 2015. Accepted: July 22, 2015. Available online: ---.
Author Contributions:
Manuscript no. 2015-667.
Associated Retinal Consultants, William Beaumont Hospital, Royal Oak, Michigan. Presented at: the Retina Society Annual Meeting, October 7e11, 2015, Paris, France; and at the American Academy of Ophthalmology Annual Meeting, October 14e17, 2015, Las Vegas, Nevada. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Y.Y.: partially funded by the Heed Ophthalmic Foundation. The Foundation had no role in the design or conduct of this research.
8
Conception and design: Yonekawa, Thomas, Drenser, Trese, Capone Data collection: Yonekawa, Thomas, Drenser, Trese, Capone Analysis and interpretation: Yonekawa, Thomas, Drenser, Trese, Capone Obtained funding: Not applicable Overall responsibility: Yonekawa, Thomas, Drenser, Trese, Capone Abbreviations and Acronyms: AF ¼ autofluorescence; BCVA ¼ best-corrected visual acuity; CME ¼ cystoid macular edema; EDI ¼ enhanced depth imaging; FEVR ¼ familial exudative vitreoretinopathy; OCT ¼ optical coherence tomography; ROP ¼ retinopathy of prematurity; RPE ¼ retinal pigment epithelium; SD OCT ¼ spectral-domain optical coherence tomography; VMT ¼ vitreomacular traction. Correspondence: Antonio Capone, Jr., MD, Associated Retinal Consultants, William Beaumont Hospital, 3535 West Thirteen Mile Road, Suite 344, Royal Oak, MI 48073. E-mail: [email protected].
Familial exudative vitreoretinopathy (FEVR) is a hereditary abnormality in retinal vascular development caused by wnt signaling defects.1e3 Anomalous or incomplete retinal vascularization is the primary pathology, with varying degrees of secondary peripheral ischemia and subsequent complications.4 Classic ophthalmoscopic and angiographic findings include peripheral avascular retina flanked posteriorly by abnormally branching vessels, retinal neovascularization, exudation, dragging of vasculature, hyaloidal contraction, and tractional retinal detachment.4e6 Fluorescein angiography is critical in the diagnosis and management of FEVR, and the recent advent of widefield angiography has permitted further characterization of angiographic phenotypes. This has allowed earlier diagnosis and treatment of patients, and, in particular, their family members, many of whom demonstrate milder phenotypic variants.7 2015 by the American Academy of Ophthalmology Published by Elsevier Inc.
Identifying milder phenotypes remains a significant challenge: although index patients commonly present during childhood with advanced disease, less-advanced cases of FEVR may remain undetected in pedigrees for generations because of variable expressivity and seemingly normal gross anatomy of these mildly affected individuals.8 Perhaps understanding the more subtle changes in these milder stages could allow for improved identification of patients, who could then benefit from widefield angiography for more definitive diagnosis. However, the in vivo microstructural anatomy of FEVR remains largely unknown. Early histopathologic studies reported features of enucleated eyes with end-stage disease,9e11 but histopathologic diagnosis is impractical. The advent of optical coherence tomography (OCT) allows high-resolution in vivo evaluation of posterior segment microstructures. With http://dx.doi.org/10.1016/j.ophtha.2015.07.024 ISSN 0161-6420/15
1
Ophthalmology Volume -, Number -, Month 2015 spectral-domain OCT (SD OCT) and enhanced depth imaging (EDI), our understanding and treatment paradigms for many macular diseases have been enriched. However, OCT imaging in FEVR has remained largely unexplored, presumably because of the rarity of the disease and the dominant focus of previous studies on peripheral findings.5,7,8,12e14 We hypothesized that FEVR would be amenable to exploratory SD OCT analysis, given its known effects on the vitreoretinal interface. We also hypothesized, on the basis of our collective clinical and surgical experiences, that there are heretofore underappreciated pathologic microstructural macular features in FEVR. We report the findings of these analyses and identify microstructural features of FEVR; their functional correlations with vision, angiographic, and autofluorescent findings; and the new implications for current and future management strategies.
Methods This study is a single-center, retrospective, noncomparative case series of patients with a clinical or genetic diagnosis of FEVR who were treated at Associated Retinal Consultants. Participants were identified via billing codes, surgical logs, genetic databases, and image banks, during the 6-year period from January 1, 2009, to December 31, 2014. Patients who underwent SD OCT (Spectralis, Heidelberg Engineering, Heidelberg, Germany) imaging with or without EDI were included. Those with only time-domain OCT were excluded, as were patients with concurrent non-FEVR vitreoretinal pathology. Institutional Review Board/Ethics Committee approval was obtained. The study complied with the Health Insurance Portability and Accountability Act of 1996 and adhered to the tenets of the Declaration of Helsinki. Data collection included demographics, including birth history, at the first imaging session, best-corrected visual acuity (BCVA) using automated refracting Snellen acuity charts, highest FEVR staging, laser and surgical treatment history, and total number of imaging sessions. The first imaging session and BCVA were recorded, but all imaging sessions were reviewed to confirm imaging findings. Staging was determined as previously described: stage 1, avascular periphery; stage 2, avascular periphery with neovascularization; stage 3, macula-sparing retinal detachment; stage 4, macula-involving retinal detachment; stage 5, complete retinal detachment.4,8 Spectral-domain OCT/EDI images were analyzed for vitreoretinal interface, preretinal, intraretinal, subretinal, and choroidal abnormalities on the basis of image interpretations made during clinical encounters by the senior authors (K.A.D., M.T.T., A.C.), reviewed systematically by 1 author (Y.Y.), and examined by all authors if there were any discrepancies. Foveal thickness was measured manually from the internal limiting membrane to the outer boundary of the retinal pigment epithelium (RPE). Central macular thickness was defined as the average thickness of the 1-mm diameter circle centered on the fovea, with autosegmentations manually corrected. Subfoveal choroidal thickness in EDI images was measured from the outer border of the RPE to the choroidal-scleral junction.15 Macular measurements were excluded for stages 3 to 5 because of poor patient fixation that often precluded macular volume scans and the unreliability of determining precise foveal centers because of dragged anatomy. Corresponding BCVA, color photography, fluorescein angiography, and fundus autofluorescence (AF) were examined for functional correlations. Photography and fluorescein angiography
2
Table 1. Demographic and Clinical Features of Patients with Familial Exudative Vitreoretinopathy undergoing Spectral Domain Optical Coherence Tomography Feature Age, yrs (n ¼ 41 patients) Mean (median, range) Sex Male Female Eye imaged (n ¼ 74 eyes) Right Left Highest FEVR staging Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Previous treatment None Laser only Laser and anti-VEGF Incisional surgery
No. (%) 19.0 (15.4, 2.4e57.0) 21 (51) 20 (49) 40 (54) 34 (46) 33 19 2 17 3
(45) (26) (3) (23) (4)
15 38 4 17
(20) (51) (5) (23)
FEVR ¼ familial exudative vitreoretinopathy; VEGF ¼ vascular endothelial growth factor.
during examinations under anesthesia were obtained using the RetCam II (Clarity Medical Systems, Pleasanton, CA) and the 130 widefield D1300 lens. The 200Tx (Optos, Marlborough, MA) was used for ultra-widefield (200 ) photography, fluorescein angiography, and fundus AF for participants who could tolerate outpatient angiography. Genetic sequencing is offered for our patients with suspected FEVR. The exons of FZD4 (11q14.2) and NDP (Xp11.3) are initially sequenced. If mutations are not identified, TSPAN12 (7q31.31) would be sequenced and LRP5 (11q13.2) as needed. Detailed sequencing techniques have been published.2,3 Snellen BCVAs were converted to logarithm of the minimum angle of resolution units for statistical analyses. Spearman’s rank correlation was used to determine the correlation coefficients of continuous variables. Analysis of variance testing was performed to identify differences in gene mutations and dichotomous independent variables. Spectral-domain OCT features associated with poorer BCVA were identified using univariate linear regression, and multivariate linear modeling was performed using variables with P < 0.1 during univariate analysis. Statistical tests were 2-tailed, and significance was defined as P < 0.05. Stata version 9.0 (StataCorp LP, College Station, TX) was used for statistical analyses.
Results A total of 346 patients with FEVR were identified during the study period. Within this cohort, 74 eyes from 41 patients (12%) underwent 225 imaging sessions with SD OCT/EDI imaging. The demographic and clinical features are summarized in Table 1. None of the patients had a history of premature birth. All patients with stage 2 or higher had prior treatment. Interpretable SD OCT images were acquired in 67 eyes (91%). Five noninterpretable imaging sessions were for stage 4 eyes with poor vision and inability to fixate, and 2 were from a patient too young to participate fully. Of the eyes with interpretable images, abnormal findings were identified in 50 eyes (75%). All eyes
Yonekawa et al
Microstructural Features of FEVR
Table 2. Foveal, Central Macular, and Subfoveal Choroidal Thicknesses in Patients with Familial Exudative Vitreoretinopathy Abnormal Microstructure Observed All stages Stage 1 Stage 2 Stages 3e5
50 17 17 17
of of of of
67* (75%) 33 (52%) 17 (100%) 17 (100%)
Mean No. of Abnormalities (Range) 3 1 3 5
Mean Foveal Thickness mm (Range) (n [ 59)
(0e12) (0e5) (1e6) (2e12)
305 271 358 327
(118e973) (196e710) (126e973) (118e545)
Mean Central Macular Thickness mm (Range) (n [ 45) 339 302 401 455
(160e1209) (242e600) (160e1209) (389e516)
Mean Subfoveal Choroidal Thickness mm (Range) (n [ 39) 216 224 220 190
(57e363) (57e306) (107e315) (84e363)
*Of eyes with interpretable images.
(100%) that were stage 2 or higher had 1 or more microstructural abnormalities. Foveal, central macular, and subfoveal choroidal thicknesses are summarized in Table 2. The following 14 findings were noted: posterior hyaloidal organization, vitreoschisis, vitreomacular traction (VMT), vitreopapillary traction, vitreo-fold traction, vitreo-laser scar traction, peripheral traction, retinal fold, macular edema, extramacular edema, intraretinal exudates, persistent inner retinal layers at the fovea, external limiting membrane/ ellipsoid zone/interdigitation zone disruption, and retinal pigment epithelial atrophy.
Vitreoretinal Interface Hyaloidal organization is a key feature of FEVR. Posterior hyaloidal organization has various appearances by SD OCT, ranging from thinner epiretinal membrane-like hyperreflective layers (Fig 1A) to much thicker opacities (Fig 1B). The anomalous posterior vitreous also took the form of vitreoschisis-like sheets (Fig 1C). Posterior hyaloidal organization was evident in 30 eyes (46%): 7 (25%) in stage 1, 14 (74%) in stage 2, and 11 (65%) in stages 3 to 5. These abnormal vitreal scaffolds can cause VMT (Fig 1D) and vitreopapillary traction (Fig 1E), as well as vitreoretinal adhesion in the periphery, especially to laser scars (Fig 1F).
Intraretinal Pathology Persistent inner foveal fetal architecture was observed in 13 eyes (20%): 3 (11%) in stage 1, 9 (47%) in stage 2, and 1 (6%) in stages 3 to 5 (Fig 2A and B). This finding was characterized by persistence of the nerve fiber, ganglion cell, inner plexiform, inner nuclear, and outer plexiform layers. It was a bilateral phenomenon in 5 of 8 patients (63%). Of note, the 2 eyes with isolated persistent inner foveal retinal layers without concurrent
vitreoretinal interface or macular pathology had 20/25 BCVA. Diminished foveal contour also was observed in many stage 1 and 2 eyes, although this was not quantified. Macular edema was seen in 24 eyes (36%): 4 (14%) in stage 1, 9 (47%) in stage 2, and 11 (65%) in stages 3 to 5 (Fig 2C). Extramacular retinal edema also was observed (Fig 2D). Intraretinal exudation accompanied macular edema in 15 eyes (63%), which was seen as hyperreflective foci, often with shadowing, that corresponded to lipid exudation in color photographs (Fig 2E). Subretinal lipid aggregation also was observed (Fig 2E). Disruption of the ellipsoid zone was seen in 25 eyes (37%): 3 (4%) in stage 1, 5 (26%) in stage 2, and 17 (100%) in stage 3 to 5 eyes. The disruptions were patchy or broad (Fig 2F). Eyes with ellipsoid zone abnormalities also had concurrent disruptions of the external limiting membrane and interdigitation zone.
Advanced Pathology Volume scans were not achieved in the majority of stage 3 to 5 eyes, but single line-scans were usually possible and allowed determination of whether radial retinal folds were dry (no edema, Fig 3A) or wet (with edema, Fig 3B). Hyaloidal tractional forces on radial retinal folds could be imaged (Fig 3C), as well as shallow retinal detachments that may be difficult to diagnose by ophthalmoscopy in children (Fig 3D). Spectral-domain OCT was also able to identify pockets of attached retina in the midst of seemingly broad tractional retinal detachments (Fig 3E).
Functional Correlations The macular edema in FEVR is associated with leakage on angiography (Fig 4A and B). The sensitivity of SD OCT in detecting angiographically confirmed that macular and peripheral edema
Figure 1. Variable presentations and effects of posterior hyaloidal organization in familial exudative vitreoretinopathy (FEVR). A, Hyperreflective thin epiretinal membrane-like appearance (arrows). B, Broader matrices of hyaloidal organization. C, Vitreoschisis-like sheets of vitreous (inset, arrows). D, Vitreomacular traction (VMT). E, Vitreopapillary traction (arrows). F, Peripheral vitreoretinal adhesion (red arrows) to an area of photocoagulation (white arrow).
3
Ophthalmology Volume -, Number -, Month 2015
Figure 2. Intraretinal microstructural features of familial exudative vitreoretinopathy (FEVR). Persistent inner foveal retinal layers in the right (A) and left (B) eyes of a patient with stage 2 FEVR. The ganglion cell layer, inner plexiform layer, inner nuclear layer, and outer plexiform layer are retained in the fovea. C, Cystoid macular edema (CME). D, Peripheral retina with laser scars (red arrows) and overlying thin retina are shown on the left. Toward the right in the area without laser treatment, there is retinal thickening with mild intraretinal exudation (yellow arrows). E, Intraretinal exudation (red arrow) and subretinal lipid aggregation (yellow arrow). F, Tapering and disruption of the external limiting membrane, ellipsoid zone, interdigitation zone, and retinal pigment epithelium (RPE), between the arrows, in a broad fashion. EZ ¼ ellipsoid zone; GCL ¼ ganglion cell layer; INL ¼ inner nuclear layer; IPL ¼ inner plexiform layer; IZ ¼ interdigitation zone; OPL ¼ outer plexiform layer.
was 100%. Spectral-domain OCT revealed 3 additional eyes with cystic hyporeflectivities that did not correlate with leakage or staining on angiography, but these cases of retinal thickening were secondary to VMT (Figs 1D, 4D and E). The RPE abnormalities with overlying photoreceptor dysfunction were associated with focal areas of hypo-AF (Fig 4F). Fluorescein angiography AF highlighted areas of prior photocoagulation (Fig 4G). Retinal pathology, such as fronds of neovascularization and hemorrhage, blocked underlying fundus AF. Radial folds demonstrated a bed of underlying hypo-AF, which also affected flanking areas near the base of the fold (Fig 4H). Spectral-domain OCT identified all areas of true (not blocked) hypo-AF as outer retinal/RPE abnormalities.
Visual Correlations Visual acuities ranged from 20/15 to no light perception. Mean logMAR visual acuity was 0.09 (Snellen equivalent 20/25) for stage 1, 0.65 (20/89) for stage 2, and 1.49 (20/618) for stages 3 to 5. The BCVA correlated significantly with FEVR stage (Rho ¼ 0.816, P < 0.001) and the number of abnormal findings on SD OCT (Rho ¼ 0.834,
P < 0.001). For microstructural phenotypes in stages 1 and 2, greater foveal thickness (Rho ¼ 0.493, P < 0.001) and central macular thickness (Rho ¼ 0.544, P < 0.001) correlated with poorer BCVA, but not choroidal thickness (Rho ¼ 0.032, P ¼ 0.868). Of the microstructural phenotypes in stages 1 and 2, univariate analysis demonstrated that posterior hyaloidal organization (P < 0.001), VMT (P < 0.001), macular edema (P < 0.001), exudation (P < 0.001), and disruption of the ellipsoid zone (P < 0.001) were associated with poorer BCVA. Disruption of the ellipsoid zone (b ¼ 0.506; P < 0.001) and posterior hyaloidal organization (b ¼ 0.289; P ¼ 0.011) retained significance in simultaneous multivariate modeling (adjusted R2 ¼ 0.627; P < 0.001).
Genetic Sequencing Genetic sequencing was performed on 27 patients. Wnt signaling pathway mutations were detected in 8 (30%): 5 with FZD4 mutations (stages 1e5), 2 with NDP mutations (stages 1e2), and 1 with a TSPAN12 mutation (stage 1). There were no genotypicmicrostructural phenotypic associations for this small subset of patients.
Figure 3. Microstructural pathology in advanced familial exudative vitreoretinopathy (FEVR). A, Dry radial retinal fold (white arrow) without surrounding edema (inset, red arrow). B, Wet radial retinal fold: The left red arrow points to the upslope of the fold; the right red arrow appears to be pointing to a downward slope, but this is a continuation of the upward slope whose image was reflected down because it was beyond the image capturing range. The fold is flanked by intraretinal fluid (red asterisks). The white arrow points to an artifact from beyond the image capturing range. C, Hyaloidal traction on a radial retinal fold (arrows). D, Shallow retinal detachment. Asterisks indicate the area of subretinal fluid. E, Tractional retinal detachment with focally attached retina (red arrow points to attached retina, yellow arrows point to flanking detached retina).
4
Yonekawa et al
Microstructural Features of FEVR
Figure 4. Fluorescein angiographic and fundus autofluorescence (AF) correlations of microstructural features of familial exudative vitreoretinopathy (FEVR). A, Cystoid macular edema in FEVR demonstrates angiographic leakage (B, 0:48, C, 3:12). D, Infrared image of posterior hyaloidal organization causing vitreomacular traction (VMT) in the eye from Figure 1F, (E) which does not leak on fluorescein angiography because of its mechanical mechanism. F, Outer retinal abnormalities in the eye from Figure 2F results in a corresponding area of hypo-AF. G, Laser photocoagulation scars show sharply demarcated hypo-AF. H, Radial retinal folds have underlying hypo-AF.
5
Ophthalmology Volume -, Number -, Month 2015
Discussion Familial exudative vitreoretinopathy is a relatively rare, yet underdiagnosed disease entity that we have yet to fully elucidate. The seminal 1969 report by Criswick and Schepens12 that identified FEVR as a distinct clinical entity described vitreous membranes, macular heterotopia, exudation, retinal detachment, and neovascularization. Angiographic studies subsequently demonstrated the avascular peripheral retina and anomalous leaking branching vasculature,13 and recent findings using widefield angiography have advanced this angiographic characterization.7 Early histologic studies provided information on end-stage pathology from enucleated eyes,9e11 but little was known about the microstructural features of less-advanced FEVR phenotypes. Only 4 eyes from 3 patients in 2 brief reports have been described with OCT, which showed exudates,16 vitreoretinal adhesion,17 and preretinal vitreal deposits.17 In this report of 72 eyes from 40 patients, we report a broad spectrum of microstructural abnormalities in FEVR, the majority of which have not been reported.
Posterior Hyaloidal Organization Dysgenic vitreous is a salient feature of all pediatric vitreoretinopathies, including FEVR, Norrie disease, persistent fetal vasculature, and retinopathy of prematurity (ROP).6 Clinical and surgical experience have demonstrated that the posterior hyaloid in these cases is pathologically adherent and contractile, serves as a scaffold for fibrovascular proliferation,4 and can cause various forms of acute and late tractional sequelae.6 Our study validates these observations, demonstrating a wide range of its microanatomic manifestations in FEVR that increased in frequency and severity with advancing stage. Only a small number of stage 1 eyes demonstrated a thicker, more disorganized posterior hyaloidal matrix; this finding was more common (and more accentuated) in the higher stages and led to VMT and macular edema. These tractional membranes are not the typical epiretinal membranes seen after posterior vitreous detachments, because they originate from abnormal vitreous that becomes pathologic as the cicatricial disease progresses. We therefore recommend the term posterior hyaloidal organization/contraction. The structured posterior hyaloid and its clinically relevant contraction have several implications based on our surgical experiences. First, it explains that hyaloidal contraction plays a prominent role in the presenting pathology, and it is a target for treatment during primary surgical repairs. The onion-layer-like hyaloidal sheets may be difficult to visualize intraoperatively, so vitreous visualization aides may facilitate the dissection. The sheets are also broader than conventional ERMs. Second, the clinical consequence of a multilayered posterior hyaloid is the high incidence of recurrent premacular proliferation with contraction after vitrectomy for the same, despite the intraoperative impression that the premacular surface had been pristinely peeled. Third, contraction of the posterior hyaloid may occur after peripheral ablation or anti-VEGF treatment. The configuration and potential contractile vectors should be
6
examined before initiating treatment. Fourth, posterior hyaloidal organization of FEVR and other vitreoretinopathies should have a high place in the differential diagnosis of pediatric “ERM,” along with superficial combined hamartoma. In addition to VMT, we also frequently found vitreopapillary traction in all stages of disease. This is not surprising, given the strong adherence of peripapillary posterior vitreous cortex and the propensity of the prepapillary space for cellular proliferation. Further studies would be required to determine how the vitreopapillary tractional vectors contribute to macular and peripheral pathology in FEVR. From our surgical experience, we know that prepapillary stalk tissue is a key element in dissection of advanced tractional FEVR detachments. We also identified hyaloidal adhesion to laser scars and focal areas of traction in the periphery. Future developments in OCT technology to allow peripheral scanning will undoubtedly facilitate more precise delineation of these tractional vectors, which may enhance surgical therapy aimed at removing these vectors.4 Intraoperative OCT technology, paired with our findings, also may facilitate hyaloidal dissection and identification of the hyaloido-retinal surgical plane.
Cystoid Macular Edema Macular edema was first noted in angiographic descriptions of FEVR by Feldman et al.18 Of note, no subsequent reports explain this important finding. Macular edema was seen in 29% of our cohort and correlated with poorer BCVA. There were 2 forms of macular edema observed: first, the more frequently observed cystoid macular edema (CME) with concurrent leakage on fluorescein angiography. This form suggests capillary permeability, likely due to ischemia-driven or inflammatory cytokine-induced vascular decompensation of already anomalous vessels.19 Second, SD OCT detected cystic intraretinal spaces in a small subset of eyes that did not correlate with any angiographic leakage. These maculae all exhibited varying degrees of VMT, indicating that mechanical elevation of the macula was the root cause of these structural cystoid changes.20 Intraretinal cystic spaces also have been identified in ROP,21e26 but they likely represent a different pathogenesis compared with the CME seen in FEVR, for several reasons: first, cystic changes are seen in approximately half of infants with ROP, but angiographic studies have not identified leakage.21,22,26 Second, the intraretinal changes in ROP are transient and resolve over time. Third, most studies do not show an association with ROP severity or systemic comorbidities.23e25 Taken together, these studies seem to suggest that these cystoid changes are structural and developmental in origin. In contrast, the CME identified in FEVR correlates with angiographic leakage, indicating a truly pathophysiologic, reactive macular edema.
Inner Retinal Developmental Abnormalities and Outer Retinal Dysfunction Persistence of the inner retinal layers in the fovea was observed in 20% of stage 1 and 2 eyes. This developmental abnormality does not appear to be visually significant in
Yonekawa et al
Microstructural Features of FEVR
isolation, when underlying outer retinal structures remain intact. An analogous finding has been reported in OCT studies of infants with ROP.25,27e32 Although the morphology generally appears to normalize with age,28,32 persistent inner retinal layers and diminished foveal contour are also reported in older children with histories of ROP.27,29,30 Future studies are warranted to understand this microstructural phenotype further. As with many other macular pathologies, atrophy of the outer retinal structures was the strongest independent risk factor for poor visual acuity in eyes with FEVR and was associated with attenuated fundus AF. The external limiting membrane, ellipsoid zone, and interdigitation zone were affected in the majority of these eyes. Outer retinal deterioration was accompanied by concomitant pathology in all cases, whether it was macular edema, exudation, VMT, or previous retinal detachment. Because FEVR is (presumably) not a primary degeneration of photoreceptors, these outer retinal changes are likely secondary to damage from overlying intraretinal and vitreoretinal interface pathology. Our study introduces a number of SD OCT microstructural findings in patients with FEVR, all of which can serve to improve the diagnosis, treatment, and pathophysiologic understanding of this complex hereditary disease in various ways. We offer a less than exhaustive list: Diagnosis: If a pediatric patient presents with 1 or more of the described SD OCT findings, FEVR should be included in the differential diagnosis, and a thorough peripheral retinal examination and widefield fluorescein angiography should be considered. We anticipate that other vitreoretinopathies may harbor similar microstructural changes. Earlier diagnosis will lead to appropriate screening examinations and prompt treatment if necessary, theoretically resulting in improved outcomes. Treatment: The detailed identification of pathologic microanatomy can potentially open new avenues for treatment; namely, targeted addressing of VMT, vitreopapillary traction, CME, intraretinal exudation, subretinal lipid aggregation, wet radial folds, and subclinical retinal detachments. Optimal management strategies will have to be examined in future studies, including a specific focus on the prevention of macular photoreceptor damagedthe common end point of impaired vision, as seen in this study. Pathophysiology: The provision of new phenotypic end points generates a broader context for elaborating the effects of wnt signaling defects. Although this current study was underpowered for genotypeephenotype correlations, future studies may determine certain phenotypes to be associated with certain mutations. For example, our experience has suggested that FZD4 mutations may lead to more macular microvascular abnormalities, whereas tractional forces seem to play a bigger role in NDP mutations (Nudleman et al, unpublished data, April 2015). Structural choroidal changes were not consistently observed in this cohort, but future longitudinal studies may also be able to associate variable choroidal features with disease progression or treatment.
Study Limitations First, the study design was retrospective and it has its resultant inherent biases. Second, because the data were derived from a surgical referral practice, our patient cohort may be skewed toward higher stages of FEVR (and, thus, a higher frequency of demonstrable findings). Third, intraoperative OCT data from the youngest patients were not available, confining our observations to a subset of patients whose mean age is higher than that of the overall FEVR cohort. Also, the microstructural features identified in this study are in patients with FEVR, but we suspect that they are not necessarily unique to FEVR. We would not be surprised that other pediatric vitreoretinopathies have similar findings. Finally, this is a cross-sectional study, and longitudinal data are not within the study design. Therefore, further studies would have to examine the evolution of these SD OCT findings and analyze the treatment effects and outcomes. In summary, we report that SD OCT imaging allows identification of dysgenic microstructural anomalies in the majority of patients with FEVR, introducing a new dimension in the diagnosis and treatment of vitreoretinal pathology associated with FEVR. We accordingly recommend SD OCT imaging for patients with FEVR when possible, because it may serve to incorporate each patient’s individual microstructural disease features into a more focused management plan.
References 1. Drenser KA, Dailey W, Capone A, Trese MT. Genetic evaluation to establish the diagnosis of X-linked familial exudative vitreoretinopathy. Ophthalmic Genet 2006;27:75–8. 2. Wu WC, Drenser K, Trese M, et al. Retinal phenotypegenotype correlation of pediatric patients expressing mutations in the Norrie disease gene. Arch Ophthalmol 2007;125: 225–30. 3. Drenser KA, Dailey W, Vinekar A, et al. Clinical presentation and genetic correlation of patients with mutations affecting the FZD4 gene. Arch Ophthalmol 2009;127:1649–54. 4. Pendergast SD, Trese MT. Familial exudative vitreoretinopathy. Results of surgical management. Ophthalmology 1998;105:1015–23. 5. Miyakubo H, Hashimoto K, Miyakubo S. Retinal vascular pattern in familial exudative vitreoretinopathy. Ophthalmology 1984;91:1524–30. 6. Joshi MM, Ciaccia S, Trese MT, Capone A. Posterior hyaloid contracture in pediatric vitreoretinopathies. Retina 2006;26: S38–41. 7. Kashani AH, Brown KT, Chang E, et al. Diversity of retinal vascular anomalies in patients with familial exudative vitreoretinopathy. Ophthalmology 2014;121:2220–7. 8. Kashani AH, Learned D, Nudleman E, et al. High prevalence of peripheral retinal vascular anomalies in family members of patients with familial exudative vitreoretinopathy. Ophthalmology 2014;121:262–8. 9. Brockhurst RJ, Albert DM, Zakov ZN. Pathologic findings in familial exudative vitreoretinopathy. Arch Ophthalmol 1981;99:2143–6. 10. Nicholson DH, Galvis V. Criswick-Schepens syndrome (familial exudative vitreoretinopathy). Study of a Colombian kindred. Arch Ophthalmol 1984;102:1519–22.
7
Ophthalmology Volume -, Number -, Month 2015 11. Boldrey EE, Egbert P, Gass JD, Friberg T. The histopathology of familial exudative vitreoretinopathy. A report of two cases. Arch Ophthalmol 1985;103:238–41. 12. Criswick VG, Schepens CL. Familial exudative vitreoretinopathy. Am J Ophthalmol 1969;68:578–94. 13. Canny CL, Oliver GL. Fluorescein angiographic findings in familial exudative vitreoretinopathy. Arch Ophthalmol 1976;94:1114–20. 14. Ranchod TM, Ho LY, Drenser KA, et al. Clinical presentation of familial exudative vitreoretinopathy. Ophthalmology 2011;118:2070–5. 15. Imamura Y, Fujiwara T, Margolis R, Spaide RF. Enhanced depth imaging optical coherence tomography of the choroid in central serous chorioretinopathy. Retina 2009;29:1469–73. 16. Day S, Maldonado RS, Toth CA. Preretinal and intraretinal exudates in familial exudative vitreoretinopathy. Retina 2011;31:190–1. 17. Shimouchi A, Takahashi A, Nagaoka T, et al. Vitreomacular interface in patients with familial exudative vitreoretinopathy. Int Ophthalmol 2013;33:711–5. 18. Feldman EL, Norris JL, Cleasby GW. Autosomal dominant exudative vitreoretinopathy. Arch Ophthalmol 1983;101:1532–5. 19. Zhou Y, Wang Y, Tischfield M, et al. Canonical WNT signaling components in vascular development and barrier formation. J Clin Invest 2014;124:3825–46. 20. Johnson MW. Tractional cystoid macular edema: a subtle variant of the vitreomacular traction syndrome. Am J Ophthalmol 2005;140:184–92. 21. Yokoi T, Hiraoka M, Miyamoto M, et al. Vascular abnormalities in aggressive posterior retinopathy of prematurity detected by fluorescein angiography. Ophthalmology 2009;116:1377–82. 22. Lepore D, Molle F, Pagliara MM, et al. Atlas of fluorescein angiographic findings in eyes undergoing laser for retinopathy of prematurity. Ophthalmology 2011;118:168–75.
23. Vinekar A, Avadhani K, Sivakumar M, et al. Understanding clinically undetected macular changes in early retinopathy of prematurity on spectral domain optical coherence tomography. Invest Ophthalmol Vis Sci 2011;52:5183–8. 24. Maldonado RS, O’Connell R, Ascher SB, et al. Spectraldomain optical coherence tomographic assessment of severity of cystoid macular edema in retinopathy of prematurity. Arch Ophthalmol 2012;130:569–78. 25. Dubis AM, Subramaniam CD, Godara P, et al. Subclinical macular findings in infants screened for retinopathy of prematurity with spectral-domain optical coherence tomography. Ophthalmology 2013;120:1665–71. 26. Henaine-Berra A, Garcia-Aguirre G, Quiroz-Mercado H, Martinez-Castellanos MA. Retinal fluorescein angiographic changes following intravitreal anti-VEGF therapy. J AAPOS 2014;18:120–3. 27. Recchia FM, Recchia CC. Foveal dysplasia evident by optical coherence tomography in patients with a history of retinopathy of prematurity. Retina 2007;27:1221–6. 28. Maldonado RS, O’Connell RV, Sarin N, et al. Dynamics of human foveal development after premature birth. Ophthalmology 2011;118:2315–25. 29. Wu WC, Lin RI, Shih CP, et al. Visual acuity, optical components, and macular abnormalities in patients with a history of retinopathy of prematurity. Ophthalmology 2012;119: 1907–16. 30. Yanni SE, Wang J, Chan M, et al. Foveal avascular zone and foveal pit formation after preterm birth. Br J Ophthalmol 2012;96:961–6. 31. Villegas VM, Capó H, Cavuoto K, et al. Foveal structurefunction correlation in children with history of retinopathy of prematurity. Am J Ophthalmol 2014;158:508–512.e2. 32. Vajzovic L, Rothman AL, Tran-Viet D, et al. Delay in retinal photoreceptor development in very preterm compared to term infants. Invest Ophthalmol Vis Sci 2015;56:908–13.
Footnotes and Financial Disclosures Originally received: April 26, 2015. Final revision: July 9, 2015. Accepted: July 22, 2015. Available online: ---.
Author Contributions:
Manuscript no. 2015-667.
Associated Retinal Consultants, William Beaumont Hospital, Royal Oak, Michigan. Presented at: the Retina Society Annual Meeting, October 7e11, 2015, Paris, France; and at the American Academy of Ophthalmology Annual Meeting, October 14e17, 2015, Las Vegas, Nevada. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Y.Y.: partially funded by the Heed Ophthalmic Foundation. The Foundation had no role in the design or conduct of this research.
8
Conception and design: Yonekawa, Thomas, Drenser, Trese, Capone Data collection: Yonekawa, Thomas, Drenser, Trese, Capone Analysis and interpretation: Yonekawa, Thomas, Drenser, Trese, Capone Obtained funding: Not applicable Overall responsibility: Yonekawa, Thomas, Drenser, Trese, Capone Abbreviations and Acronyms: AF ¼ autofluorescence; BCVA ¼ best-corrected visual acuity; CME ¼ cystoid macular edema; EDI ¼ enhanced depth imaging; FEVR ¼ familial exudative vitreoretinopathy; OCT ¼ optical coherence tomography; ROP ¼ retinopathy of prematurity; RPE ¼ retinal pigment epithelium; SD OCT ¼ spectral-domain optical coherence tomography; VMT ¼ vitreomacular traction. Correspondence: Antonio Capone, Jr., MD, Associated Retinal Consultants, William Beaumont Hospital, 3535 West Thirteen Mile Road, Suite 344, Royal Oak, MI 48073. E-mail: [email protected].