Original Paper Received: September 25, 2013 Accepted after revision: January 15, 2014 Published online: March 20, 2014
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
Reis-Bücklers Corneal Dystrophy: A Reappraisal Using in vivo and ex vivo Imaging Techniques Qingfeng Liang a Zhiqiang Pan a Xuguang Sun a Christophe Baudouin b, c Antoine Labbé a–c
a
Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing, China; b Quinze-Vingts National Eye Center, Paris and Versailles Saint-Quentin-en-Yvelines University, Versailles, and c INSERM U968, UPMC Université Paris-6, UMRS968, Institut de la Vision, CNRS, UMR7210, Paris, France
Key Words Reis-Bücklers corneal dystrophy · In vivo confocal microscopy · Anterior segment optical coherence tomography · Histology · Transmission electron microscopy · Epithelium · Cornea
Abstract Purpose: To characterize the phenotype of Reis-Bücklers corneal dystrophy (RBCD) using in vivo and ex vivo imaging techniques. Methods: Five RBCD patients with penetrating keratoplasty (PK) were enrolled. Before surgery, all patients underwent a complete ophthalmological examination including slitlamp biomicroscopy, in vivo confocal microscopy (IVCM) and anterior segment (AS) optical coherence tomography (OCT). After PK, corneal buttons were examined by light and transmission electron microscopy (TEM). Correlations between in vivo and ex vivo images were analyzed. Results: In all cases, irregular geographic-like subepithelial gray-white opacities were observed in the central and mid-peripheral cornea. AS-OCT images of the cornea of all patients revealed hyperreflective homogeneous and continuous deposits concentrated at the level of Bowman’s layer and anterior stroma. Using IVCM, a highly reflective irregular amorphous material was observed from intermediate epithelial cells to the ante-
© 2014 S. Karger AG, Basel 0030–3747/14/0514–0187$39.50/0 E-Mail [email protected] www.karger.com/ore
rior stroma. Sparse deposits of highly reflective material were also detected in the posterior stroma. TEM showed in all specimens basal epithelial cells containing small vesicles with rodshaped dense material. Conclusions: IVCM and AS-OCT may be a useful adjunct to biomicroscopy for the diagnosis and management of RBCD. The correlations between the different in vivo and ex vivo imaging techniques emphasize the hypothesis of an epithelial origin for RBCD. © 2014 S. Karger AG, Basel
Introduction
Reis-Bücklers corneal dystrophy (RBCD) is an autosomal dominant dystrophy caused by mutations in the TGFBI gene on chromosome 5 at q31 [1]. RBCD is characterized clinically by bilateral superficial corneal opacities, recurrent erosions and significant visual impairment [1]. RBCD affects primarily Bowman’s layer with the presence of band-shaped granular and subepithelial deposits that begin to appear during the first or second decade of life [2]. Although it is generally accepted that a mutated transforming growth factor-β-induced protein accumulated in RBCD, its exact pathophysiological mechanism remains unknown [3, 4]. Prof. Zhiqiang Pan Beijing Tongren Eye Center, Beijing Tongren Hospital Capital Medical University Beijing, 100005 (China) E-Mail panyj0526 @ sina.com
Table 1. Clinical characteristics of RBCD patients who underwent PK
Case No.
No. of the pedigree
Gender
Age, years
Preoperative BCVA (R/L)
Surgery eye
Follow-up after surgery, months
Recurrence of RBCD
Postoperative BCVA (R/L)
1 2 31 4 5
IV:1 IV:5 IV:14 IV:17 IV:22
M M M F M
44 36 37 45 41
0.1/0.1 0.1/0.2 0.15/0.15 FC/FC 0.05/0.2
OS OS OS OS OD
15 6 8 8 12
no no yes no no
0.1/0.5 0.1/0.4 0.15/0.2 FC/0.4 0.2/0.2
BCVA = Best-corrected visual acuity; FC = finger counting. 1 Case 3 had already undergone lamellar keratoplasty in his right eye 7 years before with recurrence of RBCD 2 years before the study. He eventually developed recurrence 3 months after PK in his left eye.
A specific clinical and histological phenotype associated with the identification of the gene mutation makes a precise diagnosis of RBCD possible [5]. However, in clinical practice, corneal dystrophies are mostly diagnosed and differentiated using siltlamp biomicroscopy. Because the slitlamp cannot provide details of corneal structures at the cellular level, many corneal dystrophies and most particularly granular dystrophies are difficult to distinguish [5, 6]. Recently developed in vivo imaging techniques such as in vivo confocal microscopy (IVCM) and anterior segment (AS) optical coherence tomography (OCT) are providing new insights into the clinical evaluation of corneal dystrophies [7, 8]. In RBCD, IVCM studies have shown an abnormal hyperreflective dystrophic material in both the epithelium and Bowman layers [6, 7, 9, 10]. Similarly, AS-OCT is also providing high-resolution in vivo cross-sections of the cornea, and granular dystrophies have already been analyzed using this technique [11]. However, to our knowledge, no detailed study aiming to correlate in vivo and histopathological findings in RBCD has been published to date. Thus, the purpose of this study was to characterize the phenotype of RBCD using slitlamp biomicroscopy, IVCM, AS-OCT, light (LM) and transmission electron microscopy (TEM) in order to help clinicians manage this rare corneal disease and to better understand its pathophysiology. Patients and Methods Patients Five patients (4 men and 1 woman; mean age, 40.6 ± 3.5 years; range, 36–45 years) from a Chinese family with RBCD scheduled to undergo penetrating keratoplasty (PK) were enrolled in the
188
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
present study. A p.Arg124Leu mutation of the TGFBI gene was detected in that family, and their pedigree, clinical manifestations and molecular genetic analysis have already been published elsewhere [12]. The Medical Ethics Committee of the Beijing Tongren Hospital approved the study protocol, and all participants gave their informed consent, according to the Declaration of Helsinki. Before surgical treatment, a second laboratory (SinoGenoMax Ltd., Chinese National Human Genome Center, Beijing, China) confirmed the previous genetic analysis with a p.Arg124Leu mutation of the βIGH3 gene. All patients underwent a complete ophthalmological examination including slitlamp biomicroscopy, IVCM and AS-OCT analysis. Except for 1 patient (case 3) who had already had lamellar keratoplasty 7 years before in his right eye and developed recurrence, all patients had primary RBCD in both eyes. After PK, corneal buttons were examined by LM and TEM. One patient (case 3) developed recurrence in his left eye 3 months after PK and was re-examined with the slitlamp, IVCM and AS-OCT at this time (table 1). In vivo Confocal Microscopy IVCM images were obtained using the new Rostock Cornea Module® of the Heidelberg Retina Tomograph® III (Heidelberg Engineering GmbH, Heidelberg, Germany) [13]. The laser source employed in the Heidelberg Retina Tomograph III/Rostock Cornea Module is a diode laser with a wavelength of 670 nm. Images consist of 384 × 384 pixels covering an area of 400 × 400 μm with transversal optical resolution of approximately 1 μm/pixel and an acquisition time of 0.024 s (Heidelberg Engineering). The lens of the microscope is an immersion lens (Olympus, Hamburg, Germany), magnification ×60, covered by a sterile polymethylmethacrylate cap. The x-y position of the image and the section depth are controlled manually. The eye is adjusted by means of the live image controlled by a CCD color camera (640 × 480 pixels, RGB, 15 frames/s). Before examination, a drop of topical anesthetic (proparacaine hydrochloride 0.5%) was instilled in the lower conjunctival fornix. For all eyes examined, 50–100 IVCM images parallel to the corneal epithelium, stroma and endothelium were obtained. IVCM images showing oblique sections of the cornea were also obtained whenever possible. Each eye was examined for less than 5 min. Images were then analyzed using Image J® software (NIH, Bethesda, Md., USA).
Liang/Pan/Sun/Baudouin/Labbé
Color version available online
Fig. 1. Slitlamp biomicroscopy images of the cornea in RBCD. a Diffuse, geographic subepithelial opacification with a clear peripheral cornea in a primary case of RBCD. b Slitlamp photography showing abnormal material predominantly at the level of Bowman’s layer. c Similar geographic opacification in a recurrent case after PK. d Early recurrence with corneal opacification beginning inferiorly at Bowman’s layer (white arrow).
a
b
c
d
Anterior Segment Optical Coherence Tomography A spectral-domain optical coherence tomograph fitted with an AS module (Optovue Corporation, Fremont, Calif., USA) was used. The OCT axial and lateral optical resolutions were 18 and 60 μm, respectively. All acquisitions were taken using the high-resolution mode with an acquisition time of 0.125 s/cross-section for overall AS examinations. For each eye, 2 images of the central cornea and 1 image of the limboconjunctival area in each quadrant were acquired. Deposit thickness and central corneal thickness were also measured in all eyes. Images were analyzed using Image J® software. Light and Electron Microscopy All corneal buttons obtained from the 5 RBCD patients were divided into 2 pieces. One half was examined by LM and the other half by TEM. Corneal tissues for LM were fixed in 10% formalin in 0.1 M phosphate buffer and dehydrated through a graded series of ethanol, cleared with xylene, and embedded in paraffin wax. Finally they were cut into 3.5-μm sections. Then the corneal sections were stained with hematoxylin and eosin (HE), Masson trichrome, Congo red and periodic acid-Schiff (PAS) stains, and examined by LM (Olympus DP72, Tokyo, Japan). Corneal tissues for TEM were fixed in 2.5% glutaraldehyde in 0.1 M phosphate buffer and were dissected in segments under 1 mm3, immersed in 2.5% glutaraldehyde in 0.1 M phosphate buffer and postfixed in 1% osmium tetroxide in 0.05 M phosphate buffer for 1 h. Then the tissues were dehydrated through a graded ethanol series and embedded in EPON resin. Resin-embedded blocks were cut into semithin (1-μm) and thin (65-nm) sections. Semithin sections were stained with toluidine blue and thin sections with uranyl acetate and lead citrate. All of them were examined with a transmission electron microscope (DSM 962, Zeiss, Oberkochen, Germany) operating at 20 kV. Images were digitized and analyzed using Image J software.
Reis-Bücklers Corneal Dystrophy
Results
Biomicroscopy In all cases of primary RBCD, irregular geographiclike subepithelial gray-white opacities were observed in the central and mid-peripheral corneas of both eyes. They were confluent with various densities. Opacities were separated from the limbus by a clear band of peripheral cornea (fig. 1a). No abnormal corneal fluorescein staining was detected. These opacities were predominantly detected at the level of Bowman’s layer (fig. 1b). Dot-like lesions were also observed in the anterior stroma. The posterior stromal layers, Descemet membrane and endothelium were apparently normal. One patient (case 3) had already undergone lamellar keratoplasty 7 years before in his right eye, but he developed recurrence 2 years before the study. Slitlamp examination revealed multiple opacities in the subepithelial layer and anterior stroma of the donor cornea similar to those observed in primary RBCD (fig. 1c). The same patient developed recurrence in his left eye 3 months after PK. There was no sign of acute rejection. Small, irregular and superficial corneal opacities began to recur at the edge of the grafted cornea (fig. 1d). These opacities extended from the subepithelial area to the anterior stroma.
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
189
a
b
c
d
e
Fig. 2. AS-OCT images of the central and peripheral cornea of patients with RBCD. a Hyperreflective material at Bowman’s layer
extending into epithelial layers with a saw-tooth-like border. b, c The dystrophic material became thinner in the peripheral cor-
Anterior Segment Optical Coherence Tomography AS-OCT images of the cornea of all patients before surgery revealed hyperreflective homogeneous and continuous deposits concentrated at the level of Bowman’s layer and anterior stroma (fig. 2a). The dystrophic material had an anterior saw-tooth border and a well-limited posterior edge within the anterior stroma (fig. 2a). This material was predominantly observed in the central and mid-peripheral cornea with a thickness of central corneal deposits varying from 72 to 132 μm (mean, 96 ± 12 μm, n = 5) for a mean corneal thickness of 566 ± 21 μm (n = 5; fig. 2a). The deposits became thinner in the periphery of the cornea, and this hyperreflective material tended to disappear before the limbus (fig. 2b, c). Deeper corneal layers were unremarkable. In case 3, AS-OCT images of the right eye (with recurrence) showed similar hyperreflective deposits located at the level of Bowman’s 190
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
nea and disappeared before the limbus (white arrows). d, e Image of late-recurrence case (left eye of case 3) after lamellar keratoplasty. The abnormal epithelial material lined the grafted stroma without penetrating it (asterisks).
layer. Interestingly, the border of the grafted corneal stroma was clearly delineated by the deposits without any visible material in the grafted stroma (fig. 2d). Moreover, at the limit between the host and the grafted cornea, the abnormal dystrophic material was delineating the corneal stroma and was limited to the epithelial layer (fig. 2e). In vivo Confocal Microscopy In all primary cases, almost identical IVCM images were obtained. A highly reflective irregular amorphous material was observed starting at the level of intermediate and basal epithelial cells (fig. 3a). Epithelial cell layer thickness was also irregular with fewer epithelial layers underlying the dystrophic material (fig. 3a). Some enlarged and cystic epithelial cells were observed above and surrounding the abnormal material (fig. 3b). In the Liang/Pan/Sun/Baudouin/Labbé
a
b
c
d
e
f
g
h
i
Fig. 3. IVCM images of the cornea in RBCD (400 × 400 μm). a Oblique view of the cornea showing the abnormal hyperreflective material at Bowman’s layer and protruding into epithelial layers. b Cystic and enlarged epithelial cells above the dystrophic material (arrows). c, d Highly reflective, irregular dystrophic material in the basal epithelial layer. e Bowman’s layer was replaced by the dystrophic material without visible subbasal nerves or inflamma-
tory cells. This material is highly reflective with no dark shadow. Abnormal deposits (arrows) were also detected in the anterior stroma (f) and in the posterior stroma but were sparse (g). h Late recurrence showed similar material at Bowman’s layer. i Early recurrence 3 months after PK revealed small accumulation of dystrophic material at Bowman’s layer.
most superficial layers, this material was focally distributed with irregular but well-defined borders (fig. 3c). In deeper layers, these deposits were diffusely distributed without dark shadows (fig. 3d). Pronounced changes were observed in Bowman’s layer, which was replaced by this highly reflective material (fig. 3e). The reflectivity of this material was much higher than the reflectivity of the anterior stroma keratocyte nuclei. Subbasal nerve
plexus were not detected in areas of abnormal material deposition in all cases. No inflammatory dendriticshaped cell was observed within or surrounding these deposits. Within the anterior stroma, fine, diffuse, round deposits and spindle-shaped materials were also observed between keratocyte nuclei (fig. 3f). Sparse deposits of highly reflective material were also detected in the posterior stroma (fig. 3g). No abnormalities were ob-
Reis-Bücklers Corneal Dystrophy
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
191
Color version available online
a
b
c
d
Fig. 4. LM images of the cornea in RBCD. a Epithelium thickness was irregular with some cystic epithelial cells, stained with HE. ×200. b Stained with PAS. ×200. c Bowman’s layer was replaced
with sheet-like deposits positive with Masson trichrome staining (arrows). ×200. d Similar sparse deposits were located in the middle and posterior stroma (arrows). ×400.
served at the level of the Descemet membrane and endothelium. In case 3, similar IVCM features were detected in the right eye 2 years after recurrence (fig. 3h). In the left eye (early recurrence), IVCM showed highly reflective granular materials at the basal epithelial cells and Bowman’s layer (fig. 3i). Although the reflectivity and density of these deposits were lower than those observed after long-term recurrence, their IVCM aspect was similar. There was no abnormal infiltration of dendrite form cells or tissue edema suggesting graft rejection in the left eye.
to Bowman’s layer and superficial stroma. Deposits did not stain with PAS (fig. 4b), but they were eosinophilic on HE staining and red with Masson trichrome (fig. 4a, c). A similar material was also detected as several round and sparse deposits in the middle and posterior stroma using Masson trichrome and HE staining (fig. 4a, d). Using TEM, accumulations of electron-dense rod- or trapezoidal-shaped deposits were seen in locations corresponding to LM observations of dystrophy material deposition (fig. 5a). Fine microfibril material and abnormal proteoglycans were also present in the vicinity of these deposits. In all specimens, basal epithelial cells contained small vesicles enclosing rod-shaped deposits (fig. 5b). Most of the vesicles were attached to the electron-dense deposits. The deposits with rod-shaped, trapezoidal and moth-eaten substances (fig. 5c–e) were surrounded by microfibrillar materials and large proteoglycan filaments (fig. 5c). Similar electron-dense deposits were also detected in the deep stroma but were sparse (fig. 5d).
Light and Electron Microscopy In all corneal specimens, LM revealed an irregular and focally thin epithelium (1 or 2 layers of epithelial cells) with enlarged and cystic epithelial basal cells suggesting degeneration (fig. 4a, b). Bowman’s layer was absent and replaced by band-shaped subepithelial deposits (fig. 4c). These deposits were observed from the basal epithelium 192
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
Liang/Pan/Sun/Baudouin/Labbé
c
a
b
d
e
Fig. 5. TEM images of the cornea in RBCD. SD = Subepithelial deposits; Bo = Bowman’s layer; Ep = epithelial cells; N = nucleus. a Subepithelial deposits are located above Bowman’s layer close to darkly stained epithelial cells. b Abnormal deposits with a rod-
shaped appearance. c Multiple trapezoidal and rod-shaped deposits appeared between epithelial cells. d Some deposits appeared in the epithelium (white rectangle). e Deposits with a ‘moth-eaten’ morphology.
Reis-Bücklers Corneal Dystrophy
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
193
Discussion
In the present study, IVCM, AS-OCT, LM and TEM analyses of the cornea of patients with genetically proven RBCD were performed to provide characteristic images of RBCD and their correlations, but also a reappraisal of its pathophysiology. Each imaging technique has its own limitations, such as the slitlamp or AS-OCT, which cannot provide details of the structures at the cellular level, and IVCM, which has no available in vivo staining procedures. Although the resolution of histopathology, in particular with TEM, is much higher than with in vivo techniques, information could be obtained only after surgery and sampling/staining procedures that may alter corneal tissues [9]. Consequently, the use of both in vivo and ex vivo imaging techniques is valuable when analyzing corneal dystrophies. Only a morphological description of the corneal changes and an epithelial, stromal or endothelial location could be obtained with the slitlamp. Therefore, IVCM was developed to overcome this limitation, and it now provides better resolution and image contrast for the in vivo analysis of corneal structures [14, 15]. With IVCM, an amorphous hyperreflective material was observed in RBCD at the basal epithelium, Bowman’s layer and anterior stroma without inflammatory cells or signs of stromal activation [6, 7, 9, 10]. We originally also observed epithelial cell alterations above this abnormal hyperreflective material. This in vivo observation was confirmed by histology showing similar epithelial cell changes. Although IVCM can provide oblique sections of the cornea (fig. 3a), they are difficult to obtain and may result in false images depending on the obliquity of the focal plane. The correlation with AS-OCT, showing the anterior saw-tooth border of the abnormal deposits, helped us interpret the IVCM images. The focal epithelial deposits previously described with IVCM [7, 9], and also observed in the present study, are actually in continuity with and similar to the abnormal material replacing Bowman’s layer (fig. 3c, d). Histopathology findings on corneal specimens confirmed our IVCM and AS-OCT results by showing similar continuous subepithelial deposits above and near Bowman’s layer. In clinical practice, granular corneal dystrophies are difficult to differentiate, and Thiel-Behnke corneal dystrophy often appears similar to RBCD. The IVCM appearance of the dystrophic tissue in Thiel-Behnke disease with round edges and dark shadows, which is also less reflective than in RBCD, may help differentiate the two dystrophies [9]. Recurrence and graft rejection are the two most frequent complications after PK in RBCD [16]. Using IVCM, a case of early recurrence showed highly 194
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
reflective granular materials at the level of the basal cell layer and Bowman’s layer. This material was similar to the primary cases and the older recurrence in the same patient but with lower density and reflectivity. IVCM images of corneal allograft rejections are different, with focal accumulations of hyperreflective dendritic-like cells at the basal epithelium and Bowman’s layer associated with altered keratocytes [17]. Therefore, IVCM might be a useful tool to help clinicians differentiate between recurrence and graft rejection in RBCD patients. The histopathology examination (LM and TEM) of corneal specimens also demonstrated that the abnormal deposits were not only limited to the anterior part of the cornea as previously thought, but were also present in the posterior stroma [6, 9, 10, 16]. These deposits appeared positive with Masson trichrome and HE staining, like the deposits observed at Bowman’s layer. Similarly, IVCM images also showed corresponding fine, diffuse and granular hyperreflective deposits in the middle and posterior stroma (fig. 3g). These deposits, observed in the posterior stroma with both in vivo and ex vivo techniques, are very similar to those found in lattice type-1 dystrophy [6]. Werner et al. [6] also used IVCM to describe diffuse small dystrophic material in the middle and posterior stroma in cases of lattice type-1 dystrophy and granular dystrophy. However, these authors observed dystrophic material only in the anterior part of the cornea in RBCD, but these observations were not confirmed by histology [6]. The relationship between RBCD, granular dystrophy and lattice type-1 dystrophy has been further explained by molecular genetic studies. These dystrophies could be different phenotypic expressions of the same gene mutation (βIGH3) on chromosome 5q31 [5, 18]. Our data showing that genetically proven RBCD patients have both anterior and posterior stromal involvement may also support this finding. Although RBCD is described as a dystrophy of Bowman’s layer [4, 5], some authors have suggested an epithelial origin [4, 19]. Several observations from the present study seem to confirm this hypothesis. With AS-OCT, the dystrophic material became thinner in the periphery of the cornea and tended to disappear before the limbus, explaining the clear band of tissue observed clinically in the peripheral cornea. The vicinity of this corneal ring without dystrophic material and the limbus is interesting because the limbus is a vascularized area with possibly greater ability to clear abnormal deposits. Nevertheless, the proliferative activity and the lower differentiation of the limbus epithelial cells [20] may also be a possible explanation to the absence of dystrophic material in this particular area. Both IVCM and LM on histological corLiang/Pan/Sun/Baudouin/Labbé
neal sections demonstrated epithelial cell abnormalities surrounding the dystrophic material. Similarly, TEM showed numerous electron-dense and rod-shaped deposits characteristic of RBCD among epithelial cells. Irregular oval precipitates of noncollagenous protein also appeared on the epithelial side of the basement membrane. The pathophysiology of recurrence after corneal grafting in RBCD also suggests an epithelial origin, given that the new corneal tissue is made of the grafted stroma and the host epithelium. The AS-OCT images of the patient who had already experienced recurrence showed that the border of the grafted stroma was clearly delineated by dystrophic material with no visible stromal material. Moreover, with IVCM we also observed the first deposits at the level of basal epithelial cells and Bowman’s layer at the edge of the graft cornea in the early recurrence case. The epithelial origin of RBCD could also explain the variable risk and delay of recurrence after corneal graft. Majo et al. [21] demonstrated that some epithelial stem cells (or oligopotent cells) could be present in the central cornea of mammals and may participate in the renewal of the corneal epithelium. Similarly, several studies found that donor epithelial cells may be present in grafted cornea more
than 1 year after PK [22]. Nonrecurrent cases after PK for RBCD may be grafted corneas with persistent donor epithelial cells. According to this hypothesis, the protection of the donor corneal epithelium when performing PK for RBCD may be of great importance in reducing the rate of recurrence by inhibiting host epithelial cell migration over the graft tissue. Catanese et al. [22] found that 2% cyclosporine A eyedrops may delay the disappearance of the donor epithelial cells after PK. Further studies should therefore evaluate such treatments to reduce the rate of RBCD recurrence after PK or lamellar keratoplasty. IVCM and AS-OCT are noninvasive and suitable imaging tools for the in vivo evaluation of corneal diseases and may assist clinicians not only in the diagnosis, but also in the treatment choice with the depth of involvement in RBCD. The correlations between the different in vivo and ex vivo imaging techniques argue in favor of the hypothesis of an epithelial origin for RBCD.
Disclosure Statement No conflicting relationship exists for any author.
References 1 Li D, Qi Y, Wang L, et al: An atypical phenotype of Reis-Bucklers corneal dystrophy caused by the G623D mutation in TGFBI. Mol Vis 2008;14:1298–1302. 2 Cho KJ, Mok JW, Na KS, et al: TGFBI gene mutations in a Korean population with corneal dystrophy. Mol Vis 2012;18:2012–2021. 3 Klintworth GK: Corneal dystrophies. Orphanet J Rare Dis 2009;4:7. 4 Ridgway AE, Akhtar S, Munier FL, et al: Ultrastructural and molecular analysis of Bowman’s layer corneal dystrophies: an epithelial origin? Invest Ophthalmol Vis Sci 2000; 41: 3286–3292. 5 Weiss JS, Moller HU, Lisch W, et al: The IC3D classification of the corneal dystrophies. Cornea 2008;27(suppl 2):S1–S83. 6 Werner LP, Werner L, Dighiero P, et al: Confocal microscopy in Bowman and stromal corneal dystrophies. Ophthalmology 1999; 106:1697–1704. 7 Shukla AN, Cruzat A, Hamrah P: Confocal microscopy of corneal dystrophies. Semin Ophthalmol 2012;27:107–116. 8 Simpson T, Fonn D: Optical coherence tomography of the anterior segment. Ocul Surf 2008;6:117–127.
Reis-Bücklers Corneal Dystrophy
9 Kobayashi A, Sugiyama K: In vivo laser confocal microscopy findings for Bowman’s layer dystrophies (Thiel-Behnke and Reis-Bucklers corneal dystrophies). Ophthalmology 2007; 114:69–75. 10 Kobayashi A, Fujiki K, Fujimaki T, et al: In vivo laser confocal microscopic findings of corneal stromal dystrophies. Arch Ophthalmol 2007;125:1168–1173. 11 Miura M, Mori H, Watanabe Y, et al: Threedimensional optical coherence tomography of granular corneal dystrophy. Cornea 2007; 26:373–374. 12 Liang Q, Sun X, Jin X: TGFBI gene mutation in a Chinese pedigree with Reis-Bucklers corneal dystrophy. Ophthalmic Physiol Opt 2012;32:74–80. 13 Labbe A, Nicola RD, Dupas B, et al: Epithelial basement membrane dystrophy: evaluation with the HRT II Rostock Cornea Module. Ophthalmology 2006;113:1301–1308. 14 Niederer RL, McGhee CN: Clinical in vivo confocal microscopy of the human cornea in health and disease. Prog Retin Eye Res 2010; 29:30–58. 15 Labbe A, Kallel S, Denoyer A, et al: Corneal imaging. J Fr Ophtalmol 2012;35:628–634. 16 Vemuganti GK, Rathi VM, Murthy SI: Histological landmarks in corneal dystrophy: pathology of corneal dystrophies. Dev Ophthalmol 2011;48:24–50.
17 Niederer RL, Sherwin T, McGhee CN: In vivo confocal microscopy of subepithelial infiltrates in human corneal transplant rejection. Cornea 2007;26:501–504. 18 Moller HU: Granular corneal dystrophy Groenouw type I (GrI) and Reis-Bucklers’ corneal dystrophy (R-B). One entity? Acta Ophthalmol (Copenh) 1989;67:678–684. 19 Akhtar S, Meek KM, Ridgway AE, et al: Deposits and proteoglycan changes in primary and recurrent granular dystrophy of the cornea. Arch Ophthalmol 1999;117:310–321. 20 Thoft RA, Friend J: The X, Y, Z hypothesis of corneal epithelial maintenance. Invest Ophthalmol Vis Sci 1983;24:1442–1443. 21 Majo F, Rochat A, Nicolas M, et al: Oligopotent stem cells are distributed throughout the mammalian ocular surface. Nature 2008;456: 250–254. 22 Catanese M, Popovici C, Proust H, et al: Fluorescent in situ hybridization (FISH) on corneal impression cytology specimens (CICS): study of epithelial cell survival after keratoplasty. Invest Ophthalmol Vis Sci 2011; 52: 1009–1013.
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
195
Copyright: S. Karger AG, Basel 2014. Reproduced with the permission of S. Karger AG, Basel. Further reproduction or distribution (electronic or otherwise) is prohibited without permission from the copyright holder.
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
Reis-Bücklers Corneal Dystrophy: A Reappraisal Using in vivo and ex vivo Imaging Techniques Qingfeng Liang a Zhiqiang Pan a Xuguang Sun a Christophe Baudouin b, c Antoine Labbé a–c
a
Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing, China; b Quinze-Vingts National Eye Center, Paris and Versailles Saint-Quentin-en-Yvelines University, Versailles, and c INSERM U968, UPMC Université Paris-6, UMRS968, Institut de la Vision, CNRS, UMR7210, Paris, France
Key Words Reis-Bücklers corneal dystrophy · In vivo confocal microscopy · Anterior segment optical coherence tomography · Histology · Transmission electron microscopy · Epithelium · Cornea
Abstract Purpose: To characterize the phenotype of Reis-Bücklers corneal dystrophy (RBCD) using in vivo and ex vivo imaging techniques. Methods: Five RBCD patients with penetrating keratoplasty (PK) were enrolled. Before surgery, all patients underwent a complete ophthalmological examination including slitlamp biomicroscopy, in vivo confocal microscopy (IVCM) and anterior segment (AS) optical coherence tomography (OCT). After PK, corneal buttons were examined by light and transmission electron microscopy (TEM). Correlations between in vivo and ex vivo images were analyzed. Results: In all cases, irregular geographic-like subepithelial gray-white opacities were observed in the central and mid-peripheral cornea. AS-OCT images of the cornea of all patients revealed hyperreflective homogeneous and continuous deposits concentrated at the level of Bowman’s layer and anterior stroma. Using IVCM, a highly reflective irregular amorphous material was observed from intermediate epithelial cells to the ante-
© 2014 S. Karger AG, Basel 0030–3747/14/0514–0187$39.50/0 E-Mail [email protected] www.karger.com/ore
rior stroma. Sparse deposits of highly reflective material were also detected in the posterior stroma. TEM showed in all specimens basal epithelial cells containing small vesicles with rodshaped dense material. Conclusions: IVCM and AS-OCT may be a useful adjunct to biomicroscopy for the diagnosis and management of RBCD. The correlations between the different in vivo and ex vivo imaging techniques emphasize the hypothesis of an epithelial origin for RBCD. © 2014 S. Karger AG, Basel
Introduction
Reis-Bücklers corneal dystrophy (RBCD) is an autosomal dominant dystrophy caused by mutations in the TGFBI gene on chromosome 5 at q31 [1]. RBCD is characterized clinically by bilateral superficial corneal opacities, recurrent erosions and significant visual impairment [1]. RBCD affects primarily Bowman’s layer with the presence of band-shaped granular and subepithelial deposits that begin to appear during the first or second decade of life [2]. Although it is generally accepted that a mutated transforming growth factor-β-induced protein accumulated in RBCD, its exact pathophysiological mechanism remains unknown [3, 4]. Prof. Zhiqiang Pan Beijing Tongren Eye Center, Beijing Tongren Hospital Capital Medical University Beijing, 100005 (China) E-Mail panyj0526 @ sina.com
Table 1. Clinical characteristics of RBCD patients who underwent PK
Case No.
No. of the pedigree
Gender
Age, years
Preoperative BCVA (R/L)
Surgery eye
Follow-up after surgery, months
Recurrence of RBCD
Postoperative BCVA (R/L)
1 2 31 4 5
IV:1 IV:5 IV:14 IV:17 IV:22
M M M F M
44 36 37 45 41
0.1/0.1 0.1/0.2 0.15/0.15 FC/FC 0.05/0.2
OS OS OS OS OD
15 6 8 8 12
no no yes no no
0.1/0.5 0.1/0.4 0.15/0.2 FC/0.4 0.2/0.2
BCVA = Best-corrected visual acuity; FC = finger counting. 1 Case 3 had already undergone lamellar keratoplasty in his right eye 7 years before with recurrence of RBCD 2 years before the study. He eventually developed recurrence 3 months after PK in his left eye.
A specific clinical and histological phenotype associated with the identification of the gene mutation makes a precise diagnosis of RBCD possible [5]. However, in clinical practice, corneal dystrophies are mostly diagnosed and differentiated using siltlamp biomicroscopy. Because the slitlamp cannot provide details of corneal structures at the cellular level, many corneal dystrophies and most particularly granular dystrophies are difficult to distinguish [5, 6]. Recently developed in vivo imaging techniques such as in vivo confocal microscopy (IVCM) and anterior segment (AS) optical coherence tomography (OCT) are providing new insights into the clinical evaluation of corneal dystrophies [7, 8]. In RBCD, IVCM studies have shown an abnormal hyperreflective dystrophic material in both the epithelium and Bowman layers [6, 7, 9, 10]. Similarly, AS-OCT is also providing high-resolution in vivo cross-sections of the cornea, and granular dystrophies have already been analyzed using this technique [11]. However, to our knowledge, no detailed study aiming to correlate in vivo and histopathological findings in RBCD has been published to date. Thus, the purpose of this study was to characterize the phenotype of RBCD using slitlamp biomicroscopy, IVCM, AS-OCT, light (LM) and transmission electron microscopy (TEM) in order to help clinicians manage this rare corneal disease and to better understand its pathophysiology. Patients and Methods Patients Five patients (4 men and 1 woman; mean age, 40.6 ± 3.5 years; range, 36–45 years) from a Chinese family with RBCD scheduled to undergo penetrating keratoplasty (PK) were enrolled in the
188
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
present study. A p.Arg124Leu mutation of the TGFBI gene was detected in that family, and their pedigree, clinical manifestations and molecular genetic analysis have already been published elsewhere [12]. The Medical Ethics Committee of the Beijing Tongren Hospital approved the study protocol, and all participants gave their informed consent, according to the Declaration of Helsinki. Before surgical treatment, a second laboratory (SinoGenoMax Ltd., Chinese National Human Genome Center, Beijing, China) confirmed the previous genetic analysis with a p.Arg124Leu mutation of the βIGH3 gene. All patients underwent a complete ophthalmological examination including slitlamp biomicroscopy, IVCM and AS-OCT analysis. Except for 1 patient (case 3) who had already had lamellar keratoplasty 7 years before in his right eye and developed recurrence, all patients had primary RBCD in both eyes. After PK, corneal buttons were examined by LM and TEM. One patient (case 3) developed recurrence in his left eye 3 months after PK and was re-examined with the slitlamp, IVCM and AS-OCT at this time (table 1). In vivo Confocal Microscopy IVCM images were obtained using the new Rostock Cornea Module® of the Heidelberg Retina Tomograph® III (Heidelberg Engineering GmbH, Heidelberg, Germany) [13]. The laser source employed in the Heidelberg Retina Tomograph III/Rostock Cornea Module is a diode laser with a wavelength of 670 nm. Images consist of 384 × 384 pixels covering an area of 400 × 400 μm with transversal optical resolution of approximately 1 μm/pixel and an acquisition time of 0.024 s (Heidelberg Engineering). The lens of the microscope is an immersion lens (Olympus, Hamburg, Germany), magnification ×60, covered by a sterile polymethylmethacrylate cap. The x-y position of the image and the section depth are controlled manually. The eye is adjusted by means of the live image controlled by a CCD color camera (640 × 480 pixels, RGB, 15 frames/s). Before examination, a drop of topical anesthetic (proparacaine hydrochloride 0.5%) was instilled in the lower conjunctival fornix. For all eyes examined, 50–100 IVCM images parallel to the corneal epithelium, stroma and endothelium were obtained. IVCM images showing oblique sections of the cornea were also obtained whenever possible. Each eye was examined for less than 5 min. Images were then analyzed using Image J® software (NIH, Bethesda, Md., USA).
Liang/Pan/Sun/Baudouin/Labbé
Color version available online
Fig. 1. Slitlamp biomicroscopy images of the cornea in RBCD. a Diffuse, geographic subepithelial opacification with a clear peripheral cornea in a primary case of RBCD. b Slitlamp photography showing abnormal material predominantly at the level of Bowman’s layer. c Similar geographic opacification in a recurrent case after PK. d Early recurrence with corneal opacification beginning inferiorly at Bowman’s layer (white arrow).
a
b
c
d
Anterior Segment Optical Coherence Tomography A spectral-domain optical coherence tomograph fitted with an AS module (Optovue Corporation, Fremont, Calif., USA) was used. The OCT axial and lateral optical resolutions were 18 and 60 μm, respectively. All acquisitions were taken using the high-resolution mode with an acquisition time of 0.125 s/cross-section for overall AS examinations. For each eye, 2 images of the central cornea and 1 image of the limboconjunctival area in each quadrant were acquired. Deposit thickness and central corneal thickness were also measured in all eyes. Images were analyzed using Image J® software. Light and Electron Microscopy All corneal buttons obtained from the 5 RBCD patients were divided into 2 pieces. One half was examined by LM and the other half by TEM. Corneal tissues for LM were fixed in 10% formalin in 0.1 M phosphate buffer and dehydrated through a graded series of ethanol, cleared with xylene, and embedded in paraffin wax. Finally they were cut into 3.5-μm sections. Then the corneal sections were stained with hematoxylin and eosin (HE), Masson trichrome, Congo red and periodic acid-Schiff (PAS) stains, and examined by LM (Olympus DP72, Tokyo, Japan). Corneal tissues for TEM were fixed in 2.5% glutaraldehyde in 0.1 M phosphate buffer and were dissected in segments under 1 mm3, immersed in 2.5% glutaraldehyde in 0.1 M phosphate buffer and postfixed in 1% osmium tetroxide in 0.05 M phosphate buffer for 1 h. Then the tissues were dehydrated through a graded ethanol series and embedded in EPON resin. Resin-embedded blocks were cut into semithin (1-μm) and thin (65-nm) sections. Semithin sections were stained with toluidine blue and thin sections with uranyl acetate and lead citrate. All of them were examined with a transmission electron microscope (DSM 962, Zeiss, Oberkochen, Germany) operating at 20 kV. Images were digitized and analyzed using Image J software.
Reis-Bücklers Corneal Dystrophy
Results
Biomicroscopy In all cases of primary RBCD, irregular geographiclike subepithelial gray-white opacities were observed in the central and mid-peripheral corneas of both eyes. They were confluent with various densities. Opacities were separated from the limbus by a clear band of peripheral cornea (fig. 1a). No abnormal corneal fluorescein staining was detected. These opacities were predominantly detected at the level of Bowman’s layer (fig. 1b). Dot-like lesions were also observed in the anterior stroma. The posterior stromal layers, Descemet membrane and endothelium were apparently normal. One patient (case 3) had already undergone lamellar keratoplasty 7 years before in his right eye, but he developed recurrence 2 years before the study. Slitlamp examination revealed multiple opacities in the subepithelial layer and anterior stroma of the donor cornea similar to those observed in primary RBCD (fig. 1c). The same patient developed recurrence in his left eye 3 months after PK. There was no sign of acute rejection. Small, irregular and superficial corneal opacities began to recur at the edge of the grafted cornea (fig. 1d). These opacities extended from the subepithelial area to the anterior stroma.
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
189
a
b
c
d
e
Fig. 2. AS-OCT images of the central and peripheral cornea of patients with RBCD. a Hyperreflective material at Bowman’s layer
extending into epithelial layers with a saw-tooth-like border. b, c The dystrophic material became thinner in the peripheral cor-
Anterior Segment Optical Coherence Tomography AS-OCT images of the cornea of all patients before surgery revealed hyperreflective homogeneous and continuous deposits concentrated at the level of Bowman’s layer and anterior stroma (fig. 2a). The dystrophic material had an anterior saw-tooth border and a well-limited posterior edge within the anterior stroma (fig. 2a). This material was predominantly observed in the central and mid-peripheral cornea with a thickness of central corneal deposits varying from 72 to 132 μm (mean, 96 ± 12 μm, n = 5) for a mean corneal thickness of 566 ± 21 μm (n = 5; fig. 2a). The deposits became thinner in the periphery of the cornea, and this hyperreflective material tended to disappear before the limbus (fig. 2b, c). Deeper corneal layers were unremarkable. In case 3, AS-OCT images of the right eye (with recurrence) showed similar hyperreflective deposits located at the level of Bowman’s 190
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
nea and disappeared before the limbus (white arrows). d, e Image of late-recurrence case (left eye of case 3) after lamellar keratoplasty. The abnormal epithelial material lined the grafted stroma without penetrating it (asterisks).
layer. Interestingly, the border of the grafted corneal stroma was clearly delineated by the deposits without any visible material in the grafted stroma (fig. 2d). Moreover, at the limit between the host and the grafted cornea, the abnormal dystrophic material was delineating the corneal stroma and was limited to the epithelial layer (fig. 2e). In vivo Confocal Microscopy In all primary cases, almost identical IVCM images were obtained. A highly reflective irregular amorphous material was observed starting at the level of intermediate and basal epithelial cells (fig. 3a). Epithelial cell layer thickness was also irregular with fewer epithelial layers underlying the dystrophic material (fig. 3a). Some enlarged and cystic epithelial cells were observed above and surrounding the abnormal material (fig. 3b). In the Liang/Pan/Sun/Baudouin/Labbé
a
b
c
d
e
f
g
h
i
Fig. 3. IVCM images of the cornea in RBCD (400 × 400 μm). a Oblique view of the cornea showing the abnormal hyperreflective material at Bowman’s layer and protruding into epithelial layers. b Cystic and enlarged epithelial cells above the dystrophic material (arrows). c, d Highly reflective, irregular dystrophic material in the basal epithelial layer. e Bowman’s layer was replaced by the dystrophic material without visible subbasal nerves or inflamma-
tory cells. This material is highly reflective with no dark shadow. Abnormal deposits (arrows) were also detected in the anterior stroma (f) and in the posterior stroma but were sparse (g). h Late recurrence showed similar material at Bowman’s layer. i Early recurrence 3 months after PK revealed small accumulation of dystrophic material at Bowman’s layer.
most superficial layers, this material was focally distributed with irregular but well-defined borders (fig. 3c). In deeper layers, these deposits were diffusely distributed without dark shadows (fig. 3d). Pronounced changes were observed in Bowman’s layer, which was replaced by this highly reflective material (fig. 3e). The reflectivity of this material was much higher than the reflectivity of the anterior stroma keratocyte nuclei. Subbasal nerve
plexus were not detected in areas of abnormal material deposition in all cases. No inflammatory dendriticshaped cell was observed within or surrounding these deposits. Within the anterior stroma, fine, diffuse, round deposits and spindle-shaped materials were also observed between keratocyte nuclei (fig. 3f). Sparse deposits of highly reflective material were also detected in the posterior stroma (fig. 3g). No abnormalities were ob-
Reis-Bücklers Corneal Dystrophy
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
191
Color version available online
a
b
c
d
Fig. 4. LM images of the cornea in RBCD. a Epithelium thickness was irregular with some cystic epithelial cells, stained with HE. ×200. b Stained with PAS. ×200. c Bowman’s layer was replaced
with sheet-like deposits positive with Masson trichrome staining (arrows). ×200. d Similar sparse deposits were located in the middle and posterior stroma (arrows). ×400.
served at the level of the Descemet membrane and endothelium. In case 3, similar IVCM features were detected in the right eye 2 years after recurrence (fig. 3h). In the left eye (early recurrence), IVCM showed highly reflective granular materials at the basal epithelial cells and Bowman’s layer (fig. 3i). Although the reflectivity and density of these deposits were lower than those observed after long-term recurrence, their IVCM aspect was similar. There was no abnormal infiltration of dendrite form cells or tissue edema suggesting graft rejection in the left eye.
to Bowman’s layer and superficial stroma. Deposits did not stain with PAS (fig. 4b), but they were eosinophilic on HE staining and red with Masson trichrome (fig. 4a, c). A similar material was also detected as several round and sparse deposits in the middle and posterior stroma using Masson trichrome and HE staining (fig. 4a, d). Using TEM, accumulations of electron-dense rod- or trapezoidal-shaped deposits were seen in locations corresponding to LM observations of dystrophy material deposition (fig. 5a). Fine microfibril material and abnormal proteoglycans were also present in the vicinity of these deposits. In all specimens, basal epithelial cells contained small vesicles enclosing rod-shaped deposits (fig. 5b). Most of the vesicles were attached to the electron-dense deposits. The deposits with rod-shaped, trapezoidal and moth-eaten substances (fig. 5c–e) were surrounded by microfibrillar materials and large proteoglycan filaments (fig. 5c). Similar electron-dense deposits were also detected in the deep stroma but were sparse (fig. 5d).
Light and Electron Microscopy In all corneal specimens, LM revealed an irregular and focally thin epithelium (1 or 2 layers of epithelial cells) with enlarged and cystic epithelial basal cells suggesting degeneration (fig. 4a, b). Bowman’s layer was absent and replaced by band-shaped subepithelial deposits (fig. 4c). These deposits were observed from the basal epithelium 192
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
Liang/Pan/Sun/Baudouin/Labbé
c
a
b
d
e
Fig. 5. TEM images of the cornea in RBCD. SD = Subepithelial deposits; Bo = Bowman’s layer; Ep = epithelial cells; N = nucleus. a Subepithelial deposits are located above Bowman’s layer close to darkly stained epithelial cells. b Abnormal deposits with a rod-
shaped appearance. c Multiple trapezoidal and rod-shaped deposits appeared between epithelial cells. d Some deposits appeared in the epithelium (white rectangle). e Deposits with a ‘moth-eaten’ morphology.
Reis-Bücklers Corneal Dystrophy
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
193
Discussion
In the present study, IVCM, AS-OCT, LM and TEM analyses of the cornea of patients with genetically proven RBCD were performed to provide characteristic images of RBCD and their correlations, but also a reappraisal of its pathophysiology. Each imaging technique has its own limitations, such as the slitlamp or AS-OCT, which cannot provide details of the structures at the cellular level, and IVCM, which has no available in vivo staining procedures. Although the resolution of histopathology, in particular with TEM, is much higher than with in vivo techniques, information could be obtained only after surgery and sampling/staining procedures that may alter corneal tissues [9]. Consequently, the use of both in vivo and ex vivo imaging techniques is valuable when analyzing corneal dystrophies. Only a morphological description of the corneal changes and an epithelial, stromal or endothelial location could be obtained with the slitlamp. Therefore, IVCM was developed to overcome this limitation, and it now provides better resolution and image contrast for the in vivo analysis of corneal structures [14, 15]. With IVCM, an amorphous hyperreflective material was observed in RBCD at the basal epithelium, Bowman’s layer and anterior stroma without inflammatory cells or signs of stromal activation [6, 7, 9, 10]. We originally also observed epithelial cell alterations above this abnormal hyperreflective material. This in vivo observation was confirmed by histology showing similar epithelial cell changes. Although IVCM can provide oblique sections of the cornea (fig. 3a), they are difficult to obtain and may result in false images depending on the obliquity of the focal plane. The correlation with AS-OCT, showing the anterior saw-tooth border of the abnormal deposits, helped us interpret the IVCM images. The focal epithelial deposits previously described with IVCM [7, 9], and also observed in the present study, are actually in continuity with and similar to the abnormal material replacing Bowman’s layer (fig. 3c, d). Histopathology findings on corneal specimens confirmed our IVCM and AS-OCT results by showing similar continuous subepithelial deposits above and near Bowman’s layer. In clinical practice, granular corneal dystrophies are difficult to differentiate, and Thiel-Behnke corneal dystrophy often appears similar to RBCD. The IVCM appearance of the dystrophic tissue in Thiel-Behnke disease with round edges and dark shadows, which is also less reflective than in RBCD, may help differentiate the two dystrophies [9]. Recurrence and graft rejection are the two most frequent complications after PK in RBCD [16]. Using IVCM, a case of early recurrence showed highly 194
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
reflective granular materials at the level of the basal cell layer and Bowman’s layer. This material was similar to the primary cases and the older recurrence in the same patient but with lower density and reflectivity. IVCM images of corneal allograft rejections are different, with focal accumulations of hyperreflective dendritic-like cells at the basal epithelium and Bowman’s layer associated with altered keratocytes [17]. Therefore, IVCM might be a useful tool to help clinicians differentiate between recurrence and graft rejection in RBCD patients. The histopathology examination (LM and TEM) of corneal specimens also demonstrated that the abnormal deposits were not only limited to the anterior part of the cornea as previously thought, but were also present in the posterior stroma [6, 9, 10, 16]. These deposits appeared positive with Masson trichrome and HE staining, like the deposits observed at Bowman’s layer. Similarly, IVCM images also showed corresponding fine, diffuse and granular hyperreflective deposits in the middle and posterior stroma (fig. 3g). These deposits, observed in the posterior stroma with both in vivo and ex vivo techniques, are very similar to those found in lattice type-1 dystrophy [6]. Werner et al. [6] also used IVCM to describe diffuse small dystrophic material in the middle and posterior stroma in cases of lattice type-1 dystrophy and granular dystrophy. However, these authors observed dystrophic material only in the anterior part of the cornea in RBCD, but these observations were not confirmed by histology [6]. The relationship between RBCD, granular dystrophy and lattice type-1 dystrophy has been further explained by molecular genetic studies. These dystrophies could be different phenotypic expressions of the same gene mutation (βIGH3) on chromosome 5q31 [5, 18]. Our data showing that genetically proven RBCD patients have both anterior and posterior stromal involvement may also support this finding. Although RBCD is described as a dystrophy of Bowman’s layer [4, 5], some authors have suggested an epithelial origin [4, 19]. Several observations from the present study seem to confirm this hypothesis. With AS-OCT, the dystrophic material became thinner in the periphery of the cornea and tended to disappear before the limbus, explaining the clear band of tissue observed clinically in the peripheral cornea. The vicinity of this corneal ring without dystrophic material and the limbus is interesting because the limbus is a vascularized area with possibly greater ability to clear abnormal deposits. Nevertheless, the proliferative activity and the lower differentiation of the limbus epithelial cells [20] may also be a possible explanation to the absence of dystrophic material in this particular area. Both IVCM and LM on histological corLiang/Pan/Sun/Baudouin/Labbé
neal sections demonstrated epithelial cell abnormalities surrounding the dystrophic material. Similarly, TEM showed numerous electron-dense and rod-shaped deposits characteristic of RBCD among epithelial cells. Irregular oval precipitates of noncollagenous protein also appeared on the epithelial side of the basement membrane. The pathophysiology of recurrence after corneal grafting in RBCD also suggests an epithelial origin, given that the new corneal tissue is made of the grafted stroma and the host epithelium. The AS-OCT images of the patient who had already experienced recurrence showed that the border of the grafted stroma was clearly delineated by dystrophic material with no visible stromal material. Moreover, with IVCM we also observed the first deposits at the level of basal epithelial cells and Bowman’s layer at the edge of the graft cornea in the early recurrence case. The epithelial origin of RBCD could also explain the variable risk and delay of recurrence after corneal graft. Majo et al. [21] demonstrated that some epithelial stem cells (or oligopotent cells) could be present in the central cornea of mammals and may participate in the renewal of the corneal epithelium. Similarly, several studies found that donor epithelial cells may be present in grafted cornea more
than 1 year after PK [22]. Nonrecurrent cases after PK for RBCD may be grafted corneas with persistent donor epithelial cells. According to this hypothesis, the protection of the donor corneal epithelium when performing PK for RBCD may be of great importance in reducing the rate of recurrence by inhibiting host epithelial cell migration over the graft tissue. Catanese et al. [22] found that 2% cyclosporine A eyedrops may delay the disappearance of the donor epithelial cells after PK. Further studies should therefore evaluate such treatments to reduce the rate of RBCD recurrence after PK or lamellar keratoplasty. IVCM and AS-OCT are noninvasive and suitable imaging tools for the in vivo evaluation of corneal diseases and may assist clinicians not only in the diagnosis, but also in the treatment choice with the depth of involvement in RBCD. The correlations between the different in vivo and ex vivo imaging techniques argue in favor of the hypothesis of an epithelial origin for RBCD.
Disclosure Statement No conflicting relationship exists for any author.
References 1 Li D, Qi Y, Wang L, et al: An atypical phenotype of Reis-Bucklers corneal dystrophy caused by the G623D mutation in TGFBI. Mol Vis 2008;14:1298–1302. 2 Cho KJ, Mok JW, Na KS, et al: TGFBI gene mutations in a Korean population with corneal dystrophy. Mol Vis 2012;18:2012–2021. 3 Klintworth GK: Corneal dystrophies. Orphanet J Rare Dis 2009;4:7. 4 Ridgway AE, Akhtar S, Munier FL, et al: Ultrastructural and molecular analysis of Bowman’s layer corneal dystrophies: an epithelial origin? Invest Ophthalmol Vis Sci 2000; 41: 3286–3292. 5 Weiss JS, Moller HU, Lisch W, et al: The IC3D classification of the corneal dystrophies. Cornea 2008;27(suppl 2):S1–S83. 6 Werner LP, Werner L, Dighiero P, et al: Confocal microscopy in Bowman and stromal corneal dystrophies. Ophthalmology 1999; 106:1697–1704. 7 Shukla AN, Cruzat A, Hamrah P: Confocal microscopy of corneal dystrophies. Semin Ophthalmol 2012;27:107–116. 8 Simpson T, Fonn D: Optical coherence tomography of the anterior segment. Ocul Surf 2008;6:117–127.
Reis-Bücklers Corneal Dystrophy
9 Kobayashi A, Sugiyama K: In vivo laser confocal microscopy findings for Bowman’s layer dystrophies (Thiel-Behnke and Reis-Bucklers corneal dystrophies). Ophthalmology 2007; 114:69–75. 10 Kobayashi A, Fujiki K, Fujimaki T, et al: In vivo laser confocal microscopic findings of corneal stromal dystrophies. Arch Ophthalmol 2007;125:1168–1173. 11 Miura M, Mori H, Watanabe Y, et al: Threedimensional optical coherence tomography of granular corneal dystrophy. Cornea 2007; 26:373–374. 12 Liang Q, Sun X, Jin X: TGFBI gene mutation in a Chinese pedigree with Reis-Bucklers corneal dystrophy. Ophthalmic Physiol Opt 2012;32:74–80. 13 Labbe A, Nicola RD, Dupas B, et al: Epithelial basement membrane dystrophy: evaluation with the HRT II Rostock Cornea Module. Ophthalmology 2006;113:1301–1308. 14 Niederer RL, McGhee CN: Clinical in vivo confocal microscopy of the human cornea in health and disease. Prog Retin Eye Res 2010; 29:30–58. 15 Labbe A, Kallel S, Denoyer A, et al: Corneal imaging. J Fr Ophtalmol 2012;35:628–634. 16 Vemuganti GK, Rathi VM, Murthy SI: Histological landmarks in corneal dystrophy: pathology of corneal dystrophies. Dev Ophthalmol 2011;48:24–50.
17 Niederer RL, Sherwin T, McGhee CN: In vivo confocal microscopy of subepithelial infiltrates in human corneal transplant rejection. Cornea 2007;26:501–504. 18 Moller HU: Granular corneal dystrophy Groenouw type I (GrI) and Reis-Bucklers’ corneal dystrophy (R-B). One entity? Acta Ophthalmol (Copenh) 1989;67:678–684. 19 Akhtar S, Meek KM, Ridgway AE, et al: Deposits and proteoglycan changes in primary and recurrent granular dystrophy of the cornea. Arch Ophthalmol 1999;117:310–321. 20 Thoft RA, Friend J: The X, Y, Z hypothesis of corneal epithelial maintenance. Invest Ophthalmol Vis Sci 1983;24:1442–1443. 21 Majo F, Rochat A, Nicolas M, et al: Oligopotent stem cells are distributed throughout the mammalian ocular surface. Nature 2008;456: 250–254. 22 Catanese M, Popovici C, Proust H, et al: Fluorescent in situ hybridization (FISH) on corneal impression cytology specimens (CICS): study of epithelial cell survival after keratoplasty. Invest Ophthalmol Vis Sci 2011; 52: 1009–1013.
Ophthalmic Res 2014;51:187–195 DOI: 10.1159/000358805
195
Copyright: S. Karger AG, Basel 2014. Reproduced with the permission of S. Karger AG, Basel. Further reproduction or distribution (electronic or otherwise) is prohibited without permission from the copyright holder.
