- Email: [email protected]
Confocal microscopy of cystic disorders of the corneal epithelium1
Confocal Microscopy of Cystic Disorders the Cornea1 Epithelium Ewrardo Hern&der-Quintela, MD, Frank Mayer, MD, Pablo Dighiero, MD, Mich2le Savoldelli, Jean-Marc Legeais, MD, PhD, Gilles Renard, MD
of
Benoit Briat, MD,
Objective: This study aimed to describe the morphology of cystic disorders of the cornea1 epithelium by confocal microscopy. Design: The study design was a prospective evaluation of confocai microscopic images of patients with cystic cornea1 disorders. Participants: Thirteen patients (19 eyes) were included. The cornea1 disorders included four patients with cornea1 decompensation (Fuchs’ dystrophy), five patients with epithelial basement membrane dystrophy (e.g., Cogan’s microcystic and map-dot dystrophies), one patient with Meesmann’s dystrophy, and three patients with recurrent erosion syndrome of unknown etiology. Confocal images of diseased corneas were compared with those of ten normal control eyes (ten subjects). Intervention: All patients were examined by slit-lamp biomicroscopic analysis and confocal microscopic analysis (Tomey, Erlangen-Temmenlohe, Germany). Image analysis was used to identify the cornea1 epithelial structures correlated with the corresponding pathology. Main Outcomes Measures: Confocal microscopy was used to assess the size, shape, light scatter, and reflection of the cysts. Results: Slit-lamp examination results showed cornea1 epithelial cystic lesions in all cases. Confocal microscopy was able to identify cystic lesions in 9 (69.2%) of 13 patients. Of the four patients in whom lesions could not be found by confocal microscopy, three had recurrent erosion syndrome and the other one had epithelial basement membrane dystrophy. The confocal images were compatible with the clinical and histologic pictures of the disease. Normal control eyes did not show any epithelial lesion, either by biomicroscopy or confocal microscopy. Conclusions: Confocal microscopy provides an in vivo evaluation of cystic epithelial cornea1 lesions. This study shows that confocal microscopy is suitable for examining cystic lesions of the cornea1 epithelium. Nevertheless, it is not as sensitive as biomicroscoov in detectina Y cvstic lesions in certain cornea1 conditions. I Ophthalmology
1998; 105:63 l-636<
Epithelial cysts can be present in a wide variety of cornea1 disorders. They may be a primary component of the disease or associated with it. Their two main effects are patient discomfort and loss of vision. Bron and Tripathi’ published a classification of these diseases some time ago. They included hydrokeratopathy, defined as cornea1 edemadue to endothelial dysfunction (Fuchs’ dystrophy), Meesmann’s dystrophy, recurrent cornea1 erosion syndrome, Cogan’s dystrophy (now included in the epithelial basementmembranedystrophies of the cornea), and other Originally Revision
received: July 9, 1997. accepted: October 16, 1997.
From the Department France.
of Ophthalmology,
Presented m part at the 103rd Congress talmologle, Paris, France, May 1997.
HBtel
Dieu
of the SociCtt
Hospital, FranGaise
Pans, d’Oph-
Supported in part by grants I16681 from the Consejo National de Ciencia y Tecnologia, Mexico D.F. (EH-Q); the Asociaci6n para Evitar la Ceguera en MCxico, I.A.P., Mexico D.F. (EH-Q); and the Delegation g la Recherche Clinique AP-HP, Paris, France. The authors have no commercial interest in the development ing of any products referred to in this article. Reprint requests to Jean-Marc Legeais, MD. PhD, Department thalmology, H&e1 Dieu Hospital, I Place du Parvis Notre-Dame, Paris, France.
or marketof Oph75004
disorders. The histology and ultrastructure of several corneal disorders have been well characterized.*-” More recently, the confocal microscope has been used to obtain images of the human normal cornea”‘-” and diseased corneas.14-17But as far as we are aware, no in vivo correlation of these cystic lesions has been published previously. Details of the structure within the ocular tissues of the cornea may provide new information for the diagnosis of disease.Knowledge of the cellular dynamics of the anterior layers of the cornea also could be useful for understanding the pathogenesisof cystic disorders of the cornea1epithelium. Confocal microscopy is a noninvasive method that produces optical sections of cornea1tissue and can be assembled to form an in vivo image with high resolution and contrast.‘4,‘8 It is an effective tool to examine pathologic processesin the cornea.‘4-2’ The current study using confocal microscopy was performed to provide confocal correlation of cystic epithelial cornea1 lesions.
Materials
and Methods
The study wasconductedon I3 patients(19 eyes) seenat the Departmentof Ophthalmology,HBtel-DieuHospital. Of these,
631
Ophthalmology
Volume 10.5, Number 4, April I998
four patients (five eyes) had cornea1 decompensation due to Fuchs’ dystrophy, five patients (nine eyes) had epithelial basement membrane dystrophy, one patient (two eyes) had Meesmann’s dystrophy, and three patients (three eyes) had recurrent erosion syndrome of unknown etiology. The patients were examined with a 900 slit lamp (Hagg-Streit, Koeniz, Switzerland). Photographs of the cornea were taken with an SL-6E slitlamp camera (Topcon, Tokyo, Japan). The location of the lesion was noted and the topical anesthetic, oxibuprocaine chlorhydrate 0.4% (Novesine, MSD), was instilled into the lower conjunctival fornix. Confocal microscopy was used to examine this site (ConfoScan Model P2, Tomey, Erlangen-Temmenlohe, Germany). A drop of sodium hyaluronate (10 mg/ml Healonid; Pharmacia, Piscataway, NJ) was placed on the microscope objective tip to serve as immersion fluid. The objective used for all images was an Achroplan (Zeiss, Oberkochen, Germany) water immersion lens 40x/NA = 0.75 with a working distance of 2 mm. The position of the objective was adjusted to obtain tangential optical sections of the cornea. Oblique sections were obtained whenever possible. None of the confocal examinations took more than 2 minutes. Images were recorded on an SVO9620 Video Cassette Recorder (Sony, Japan). Selected images from the videotape were captured with a frame grabber and stored in a BMP format file. They subsequently were printed in a CHCS545 video printer (Shinko, Japan). Images were not enhanced with computer software. To compare confocal images of diseased corneas with those of healthy eyes (ten eyes/ten patients), the confocal image database was reviewed.
Results Slit-lamp examination results showed cornea1 epithelial cystic lesions in all patients. Confocal microscopy was able to identify cystic lesions in 9 (69.2%) of 13 patients. Of the four patients in whom lesions could not be found by confocal microscopy, three were suffering from recurrent erosion syndrome and the other one from epithelial basement membrane dystrophy. Confocal images of normal control eyes did not show any epithelial lesion.
Clinical Findings Fuchs’ Dystrophy. In all cases, slit-lamp examination results showed stromal and epithelial edema. The epithelial lesions were characterized by surface irregularities due to epithelial cysts and coalescent bullae. The lesions were mainly in the central cornea. Fluorescein accumulated between bullae, the stroma appeared thickened, and Descemet’s membrane had folds with a beaten-metal appearance. None of the cases had reached the scarring-vascularization stage. Epithelial Basement Membrane Dystrophy. All patients were suffering from painful recurrent epithelial erosions. All patients had multiple clear or white cysts, mainly in the central cornea1 epithelium (dots), and as amorphous subepithelial clouding (map-like pattern) in one case. This latter patient had gray opacities close to map-like lesions, and there also were clear fenestrations within the clouding zone. Fluorescein showed cornea1 surface irregularities overlying the map and cystic lesions. Patients with superficial cornea1 lines (fingerprints) were not included in this study. Meesmann’s Dystrophy. This particular case was an asymptomatic patient in whom clinical features were found during a routine ophthalmologic examination. Small round and oval opaque areas (epithelial cysts) were present in the
632
interpalpebral zone of the cornea1 epithelium of both eyes. The cysts were mainly spherical and occasionally fused. These lesions appeared as gray dots on direct illumination and as transparent vesicles in retroillumination. Not all of the cysts took up fluorescein dye. There were opaque areas on Bowman’s membrane that appeared as a slight haze between the cysts. Recurrent Erosion Syndrome. All patients had repeated episodes of early morning irritation, acute ocular pain, photophobia, and blurring of vision. All patients had intraepithelial corneal cysts. The cysts varied widely in shape and content, but only a few (ten) cysts were found at examination.
Confocal Microscopic Findings Confocal microscopy was used to assess the size, shape, light scatter, and reflection of the cysts. Fuchs’ Dystrophy. The lesions varied from round to elongated and were from 20 pm to at least 400 pm in diameter. Cystic lesions formed three patterns. The most frequently encountered pattern was a globular shape with well-defined borders, with a linear or punctiform reflection surrounded by a dark halo within some lesions (Fig 1A, arrow). The second type of lesion had an irregular reflective image within the cyst, was surrounded by a dark halo, and was outlined by the cytoplasm of an epithelial cell (Fig IA, arrowhead). The third type of lesion was more tubular, elongated, and larger (at least 400 pm) (Fig lB, arrows). They appeared separated from other epithelial cells (Fig lB, asterisk) without any other recognizable structure within the epithelium. Punctiform reflective structures with 20 pm across were seen in the lesion (Fig lB, arrowheads). The endothelium showed an abnormal pattern, in which the normal hexagonal endothelial mosaic was lost (Fig IC). Oblique sections showed all the layers of the cornea1 epithelium, which had lost its normal, regular architecture. At the most superficial level, there was cellular irregularity (Fig 1D, arrows). The nuclei of the basal epithelial cells appeared as reflective round structures arranged in a linear pattern above the stroma (Fig lD, arrowheads). Epithelial Basement Membrane Dystrophy. Confocal microscopic examination results disclosed four main features in all cases. First, the anterior stroma and posterior epithelial interface contained highly reflective irregular material intermixed with the keratocytes (Fig 2A, arrows). This material was present even in zones of apparently unaffected cornea1 tissue and was distributed unevenly. Second, the cells of the basal epithelium have abnormal distended cytoplasm with very reflective nuclei (Fig 2B, arrow). Third, the cystic lesions had a more ovoid shape (Fig 2C, arrows). Borders were poorly defined, and there was irregular reflective material within the cysts (Fig 2C, asterisk). The diameter of cysts was 50 to 400 pm. Fourth, the basement membrane bore some long (at least 300 pm), highly reflective, linear structures (Fig 2D, arrows) that had well-delineated anterior borders and diffuse posterior borders. There was a nonreflective space between this linear lesion and the normal basement membrane. Analysis of confocal images of the patient with maplike lesions showed two patterns. There were short (50- 100 pm) linear structures arranged in parallel below the epithelium (Fig 2E, arrows). Oblique sections of epithelium showed an irregular arrangement of tissue with very reflective, round structures in a wavy-linear configuration (pearl collar-like; Fig 2F, arrows), whereas others formed clusters in the shape of a microcyst (Fig 2F, asterisk). These structures extended up into the cornea1 epithelium from the basal layer and were most probably cell nuclei.
Herndndez-Quintela et al * Confocal Microscopy of the Cornea
Figure 1. Confocal microscopuz images of the cornea of patients with Fuchs’ dystrophy. A, cystic lessons in superfictal eplthelial cells with lmear reflection (arrow) and irregular reflective image wlthm the lesion (arrowhead). B, highly reflective material (arrowheads) wlthin an elongated bulla (arrows) and an abnormal epithelium as background (asterisk). C, optical sectlon at the endothehum showing guttae and enlarged endothelial cells. D, oblique aspect of the eplthelium in which normal stroma (S), epithelial cells (Ep, arrowheads), and irregular, edematous superficial cells (arrows) are observed. Horizontal field width -610 pm. E, semIfine micrograph of the cornea of another patient with Fuchs’ dystrophy to correlate with previous Images. Parts A and B are optlcal sectlons probably done between arrows in part E (stam, tolmdme blue; orlgmal magmfication, X 160). This patlent was not mcluded m the confocal macroscopic exammatlon.
Meesmann’s Dystrophy. The cystic lesions were round, well delineated, and measured 10 to 50 pm across. Some lesions showed reflective points in the cytoplasm, which probably correspond to cell nuclei (Fig 3, arrows). Some normal cell nuclei appeared as reflective round images above an apparently normal basal membrane (Fig 3, arrowheads). Cystic lesions were the only confocal findings seen in this case. No other structural features of the disease were found. Recurrent Erosion Syndrome. There were superficial polygonal low reflective areas surrounded by bright epithelial cells (Fig 4). The superficial epithelial cells were 40 to 60 pm across.
The confocal images of the basementmembrane were normal. No epithelial cystic lesions were found by confocal microscopy.
Discussion We examined the cornea1epithelium of patients with cystic disorders of the cornea by confocal microscopy. Cystic lesions were found in 69.2% of the patients examined. The usefulnessof the confocal microscope in the diagnosis of Fuchs’ dystrophy through an opaque cornea has
633
Ophthalmology
Volume 105, Number 4, April 1998
Figure 2. Confocal microscopic images of the cornea1 eplthelmm of patients with eplthehal basement membrane dystrophy. A, tangential sectlon through the subeplthehal space shows highly reflective material (arrows). B, sectlon at the basal-eplthelial level showmg globular cytoplasm and cell nuclei (arrow). C, confocal image of a cyst wxh a diameter of 200 ,um (arrows) and reflective material wlthm the lesion (asterlqk). D, mtraeptthehal basement membrane (arrows), which appeared separated from normal basal eplthehal cells. E, obhque sectlon showmg an abnormal subeplthehal layer. Multiple fibrdlar structures are present (arrows). F, cell nuclei arranged in lmear pattern (arrows) and others formmg a mxrocyst (asterisk). Normal keratocyte nuclei seen m the antenor stroma (K). Horizontal field width ~610 Wm.
been documented.” These authors found irregularly shaped endothelial cells that were three times their normal size, with a hazy stroma and a disrupted epithelium. The current confocal microscopic examination of the cornea1 epithelium of patients with Fuchs’ dystrophy showed four patterns: (1) cystic lesions with well-delineated borders and low internal reflectivity (Fig lA, arrow); (2) cystic lesions with irregular internal reflectivity surrounded by the cytoplasm of epithelial cells (Fig lA, arrowhead); (3) elongated bullae with tubular shapes and multiple round reflective structures within the lesion (Fig 1B); and (4) abnormal epithelial cells with increased intercellular space (Fig lB, asterisk). Figure 1E is provided
634
to compare the confocal microscope image and the histopathologic picture. Figures 1A and 1B probably correspond to optical sections obtained between arrows in Figure 1E. Two factors must be taken into account before interpretation. First, most confocal microscopic sections were taken at a tangential axis. Second, the micrograph shown is not from any of the patients examined in this study, so that no direct correlation is possible. Nevertheless, it seems clear that confocal microscopy tends to show the superficial cystic lesions. The epithelial cells forming cyst walls are not consistently seen by confocal microscopy. Whenever such cells are seen, they appear flattened and arranged concentrically (Fig lA, arrow-
Herndndez-Quintela
et al * Confocal Microscopy of the Cornea to the epithelium may be abnormal epithelial cells adjacent to an aberrant basement membrane.’ One patient with Meesmann’s dystrophy had intraepithelial cysts. The basement membrane appeared normal, and no other abnormal structures were seen (Fig 3). Pathologic studies have described disordered and thickened epithelium, multilaminar basement membrane, intracystic periodic acid-Schiff- and adenosine monophosphate-positive fibrillogranular material, and glycogen within cell vacuoles 9,22,23
Figure 3. Tangenttal confocal mtcroscoptc tmage of the cornea1 eptthehum of a pattent wtth Meesmann’s dystrophy. The cyst 1s surrounded by the plasma membrane of an epnhelial cell. There are two reflecttve, small (10 pm) round tmages wtthm the cytoplasm (arrows) probably corresponding to cell nuclei. Normal epithehal cells nuclet (arrowheads). Horizontal field wtdth -610 pm.
head). The elongated horizontal profile of some lesions (Fig lB, arrows) probably is because of separation of epithelial layers, as shown in histopathologic studies.’ The internal reflectivity within lesions is caused by degenerated epithelial cells and cell debris (Fig lB, arrowheads).2,9 Figures 1D and 1E can be compared directly because both are sagittal sections. The three clinicopathologic basic elements of epithelial basement membrane dystrophy’ may explain our confocal microscopic findings: 1. The basement membrane is thickened and extends into the epithelium as a dense reflective linear structure lying over normal basal epithelial cells (Fig 2D, arrows). 2. The abnormal epithelial cells and intraepithelial microcysts appear as highly reflective cell nuclei and globular lesions, respectively (Figs 2B and 2C). 3. There is fibrillar material between the epithelial basement membrane and Bowman’s layer (Fig 2E, arrows). As in Fuchs’ dystrophy, images of the cystic lesions probably show degenerating epithelial cells and cell debriszS9The highly reflective material below the epithelium (Fig 2A, arrows) probably is caused by intracystic cell debris, which reaches the subepithelial space via a continuity of the intraepithelial cysts with the subepithelial .zone.6The abnormal appearance of the cornea1 epithelial cell in Figure 2B may be caused by intracellular edema described as “pale” cells by Tripathi and Bror? in 1973. The case with map-like cornea1 lesions had round reflective structures. They differed from normal epithelial cell nuclei in that they were smaller in size and had more uniform appearance. They were arranged in two patterns, one wavy-linear (Fig 2F, arrows) and the other encircling a cyst (Fig 2F, asterisk). These structures probably are the nuclei of multinucleate epithelial giant cells. These cells were found in a histopathologic study of a patient with fingerprint dystrophy.3 The structures extending up
davanagh et al” and Tomii and Kinoshitar’ have described the appearance of the normal human epithelium under the confocal microscope. They found great variations in the reflection of the cells from the objective. This was confirmed by Wiegand et al,” who interpreted the high reflectance as being caused by cells that were about to desquamate. In our study, superficial epithelial cells of normal control eyes were observed as polygonal structures, with reflectance variability, and cell nuclei were not seen consistently. We found this same normal pattern in the confocal microscopic images of patients with recurrent erosion syndrome (Fig 4). Confocal microscopy provided an in vivo evaluation of cystic epithelial cornea1 lesions. The images were compatible with the clinical and histologic picture of the disease. The best images were obtained during the first 40 seconds of examination. After this, eye movement increased and the video recorder did not obtain high-quality images. This instrument can be used to monitor disease progression over time. Evaluation of these changes may increase our understanding of its pathogenesis. New technology requires improved image analysis. For example, the objective analysis of desquamating cells in recurrent erosion syndrome or the observation of cyst development in epithelial basement membrane dystrophy may be a clinical application of the confocal microscope that needs to be developed. Confocal microscopy is a new and evolving technology. As such, there still is much to be learned. We believe it is a useful tool for under-
Figure 4. Confocal microscopic image of the cornea1 superficial epithelium of a patient with recurrent erosion syndrome. A mucus band appears as a bright line over the epithelial mosaic. Horizontal field width -610 Pm.
635
Ophthalmology standing cornea.
how the pathologic
processes
develop
Volume 10.5, Number 4, April 1998 in the
References 1. Bron AJ, Tripathi RC. Cystic disorders of the cornea1 epithelium. I Clinical aspects. Br J Ophthalmol 1973;57:36175. 2. Tripathi RC, Bron AJ. Cystic disorders of the cornea1 epithelium. II Pathogenesis. Br J Ophthalmol 1973;57:37690. 3. Brodrick JD, Dark AJ, Peace GW. Fingerprint Dystrophy of the Cornea. A Histologic Study. Arch Ophthalmol 1974;92:483-9. 4. Cogan DG, Kuwabara T, Donaldson DD, Collins E. Microcystic dystrophy of the cornea. A partial explanation for its pathogenesis. Arch Ophthalmol 1974;92:470-4. 5. Rodrigues MM, Fine BS, Laibson PR, Zimmerman LE. Disorders of the cornea1 epithelium. A clinicopathologic study of dot, geographic, and fingerprint patterns. Arch Ophthalmol 1974;92:475-82. 6. Fogle JA, Kenyon KR, Stark WJ, Green RW. Defective epithelial adhesion in anterior cornea1 dystrophies. Am J Ophthalmol 1975; 79:925 -40. cornea1 dystrophy. Trans Am 7. Laibson PR. Microcystic Ophthalmol Sot 1976; 74:488-53 1. 8. Dark AJ. Cogan’s microcystic dystrophy of the cornea: ultrastructure and photomicroscopy. Br J Ophthalmol 1978;62:821-30. 9. Waring III GO, Rodrigues MM, Laibson PR. Cornea1 dystrophies. I. Dystrophies of the epithelium, Bowman’s layer and stroma. Surv Ophthalmol 1978;23:71- 122. 10. Cavanagh HD, Jester JV, Essepian, et al. Confocal microscopy of the living eye. CLAO J 1990; 16:65-73.
636
11. Tomii S, Kinoshita S. Observation of the cornea1 epithelium by scanning confocal microscope. Scanning 1994; 16:3056. 12. Wiegand W, Thaer AA, Kroll P, et al. Optical sectioning of the cornea with a new confocal in vivo slit-scanning videomicroscope. Ophthalmology 1995; 102:568-75. 13. Prydal JI, Dilly PN. In vivo confocal microscopy of the cornea and tear film. Scanning 1995; 17:133-5. 14. CavanaghHD, Petrol1WM, Alizadeh H, et al. Clinical and diagnostic use of in vivo confocal microscopy in patients with cornea1 disease. Ophthalmology 1993; 100:1444-54. 15. Essepain JP, Rajpal RK, Azar DT, et al. The use of confocal microscopy in evaluating cornea1 wound healing after excimer laser keratectomy. Scanning 1994; 16:300-4. 16. Essepian JP, Rajpal R, O’Brien TP. Tandem scanning confocal microscopic analysis of ciprofloxacin cornea1 deposits in vivo. Cornea 1995; 14:402-7. 17. Corbett MC, Prydal JI, Verma S, et al. An in vivo investigation of the structures responsible for cornea1 haze after photorefractive keratectomy and their effect on visual function. Ophthalmology 1996; 103: 1366-80. 18. Winchester K, Mathers WD, Stuphin JE, Daley TF. Diagnosis of Acanthamoeba keratitis in vivo with confocal microscopy. Cornea 1995; 14:10-7. 19. Kaufman SC, Beuerman RW, Kaufman HE. Diagnosis of advanced Fuchs’ endothelial dystrophy with the confocal microscope [letter]. Am J Ophthalmol 1993; 116:652-3. 20. Cohen RA, Chew SJ, Gebhardt BM, et al. Confocal microscopy of cornea1 graft rejection. Cornea 1995; 14:467-72. 21. Mathers WD, Stuphin JE, Folberg, et al. Outbreak of keratitis presumed to be caused by Acanthamoeba. Am J Ophthalmol 1996; 121:129-42. 22. Kuwabara T, Ciccarelli EC. Meesmann’s dystrophy. A pathological study. Arch Ophthalmol 1964;71:676-82. 23. Fine BS, Yannoff M, Pitts E, Slaughter FD. Meesmann’s epithelial dystrophy of the cornea. Am J Ophthalmol 1977;83:633-42.
of
Benoit Briat, MD,
Objective: This study aimed to describe the morphology of cystic disorders of the cornea1 epithelium by confocal microscopy. Design: The study design was a prospective evaluation of confocai microscopic images of patients with cystic cornea1 disorders. Participants: Thirteen patients (19 eyes) were included. The cornea1 disorders included four patients with cornea1 decompensation (Fuchs’ dystrophy), five patients with epithelial basement membrane dystrophy (e.g., Cogan’s microcystic and map-dot dystrophies), one patient with Meesmann’s dystrophy, and three patients with recurrent erosion syndrome of unknown etiology. Confocal images of diseased corneas were compared with those of ten normal control eyes (ten subjects). Intervention: All patients were examined by slit-lamp biomicroscopic analysis and confocal microscopic analysis (Tomey, Erlangen-Temmenlohe, Germany). Image analysis was used to identify the cornea1 epithelial structures correlated with the corresponding pathology. Main Outcomes Measures: Confocal microscopy was used to assess the size, shape, light scatter, and reflection of the cysts. Results: Slit-lamp examination results showed cornea1 epithelial cystic lesions in all cases. Confocal microscopy was able to identify cystic lesions in 9 (69.2%) of 13 patients. Of the four patients in whom lesions could not be found by confocal microscopy, three had recurrent erosion syndrome and the other one had epithelial basement membrane dystrophy. The confocal images were compatible with the clinical and histologic pictures of the disease. Normal control eyes did not show any epithelial lesion, either by biomicroscopy or confocal microscopy. Conclusions: Confocal microscopy provides an in vivo evaluation of cystic epithelial cornea1 lesions. This study shows that confocal microscopy is suitable for examining cystic lesions of the cornea1 epithelium. Nevertheless, it is not as sensitive as biomicroscoov in detectina Y cvstic lesions in certain cornea1 conditions. I Ophthalmology
1998; 105:63 l-636<
Epithelial cysts can be present in a wide variety of cornea1 disorders. They may be a primary component of the disease or associated with it. Their two main effects are patient discomfort and loss of vision. Bron and Tripathi’ published a classification of these diseases some time ago. They included hydrokeratopathy, defined as cornea1 edemadue to endothelial dysfunction (Fuchs’ dystrophy), Meesmann’s dystrophy, recurrent cornea1 erosion syndrome, Cogan’s dystrophy (now included in the epithelial basementmembranedystrophies of the cornea), and other Originally Revision
received: July 9, 1997. accepted: October 16, 1997.
From the Department France.
of Ophthalmology,
Presented m part at the 103rd Congress talmologle, Paris, France, May 1997.
HBtel
Dieu
of the SociCtt
Hospital, FranGaise
Pans, d’Oph-
Supported in part by grants I16681 from the Consejo National de Ciencia y Tecnologia, Mexico D.F. (EH-Q); the Asociaci6n para Evitar la Ceguera en MCxico, I.A.P., Mexico D.F. (EH-Q); and the Delegation g la Recherche Clinique AP-HP, Paris, France. The authors have no commercial interest in the development ing of any products referred to in this article. Reprint requests to Jean-Marc Legeais, MD. PhD, Department thalmology, H&e1 Dieu Hospital, I Place du Parvis Notre-Dame, Paris, France.
or marketof Oph75004
disorders. The histology and ultrastructure of several corneal disorders have been well characterized.*-” More recently, the confocal microscope has been used to obtain images of the human normal cornea”‘-” and diseased corneas.14-17But as far as we are aware, no in vivo correlation of these cystic lesions has been published previously. Details of the structure within the ocular tissues of the cornea may provide new information for the diagnosis of disease.Knowledge of the cellular dynamics of the anterior layers of the cornea also could be useful for understanding the pathogenesisof cystic disorders of the cornea1epithelium. Confocal microscopy is a noninvasive method that produces optical sections of cornea1tissue and can be assembled to form an in vivo image with high resolution and contrast.‘4,‘8 It is an effective tool to examine pathologic processesin the cornea.‘4-2’ The current study using confocal microscopy was performed to provide confocal correlation of cystic epithelial cornea1 lesions.
Materials
and Methods
The study wasconductedon I3 patients(19 eyes) seenat the Departmentof Ophthalmology,HBtel-DieuHospital. Of these,
631
Ophthalmology
Volume 10.5, Number 4, April I998
four patients (five eyes) had cornea1 decompensation due to Fuchs’ dystrophy, five patients (nine eyes) had epithelial basement membrane dystrophy, one patient (two eyes) had Meesmann’s dystrophy, and three patients (three eyes) had recurrent erosion syndrome of unknown etiology. The patients were examined with a 900 slit lamp (Hagg-Streit, Koeniz, Switzerland). Photographs of the cornea were taken with an SL-6E slitlamp camera (Topcon, Tokyo, Japan). The location of the lesion was noted and the topical anesthetic, oxibuprocaine chlorhydrate 0.4% (Novesine, MSD), was instilled into the lower conjunctival fornix. Confocal microscopy was used to examine this site (ConfoScan Model P2, Tomey, Erlangen-Temmenlohe, Germany). A drop of sodium hyaluronate (10 mg/ml Healonid; Pharmacia, Piscataway, NJ) was placed on the microscope objective tip to serve as immersion fluid. The objective used for all images was an Achroplan (Zeiss, Oberkochen, Germany) water immersion lens 40x/NA = 0.75 with a working distance of 2 mm. The position of the objective was adjusted to obtain tangential optical sections of the cornea. Oblique sections were obtained whenever possible. None of the confocal examinations took more than 2 minutes. Images were recorded on an SVO9620 Video Cassette Recorder (Sony, Japan). Selected images from the videotape were captured with a frame grabber and stored in a BMP format file. They subsequently were printed in a CHCS545 video printer (Shinko, Japan). Images were not enhanced with computer software. To compare confocal images of diseased corneas with those of healthy eyes (ten eyes/ten patients), the confocal image database was reviewed.
Results Slit-lamp examination results showed cornea1 epithelial cystic lesions in all patients. Confocal microscopy was able to identify cystic lesions in 9 (69.2%) of 13 patients. Of the four patients in whom lesions could not be found by confocal microscopy, three were suffering from recurrent erosion syndrome and the other one from epithelial basement membrane dystrophy. Confocal images of normal control eyes did not show any epithelial lesion.
Clinical Findings Fuchs’ Dystrophy. In all cases, slit-lamp examination results showed stromal and epithelial edema. The epithelial lesions were characterized by surface irregularities due to epithelial cysts and coalescent bullae. The lesions were mainly in the central cornea. Fluorescein accumulated between bullae, the stroma appeared thickened, and Descemet’s membrane had folds with a beaten-metal appearance. None of the cases had reached the scarring-vascularization stage. Epithelial Basement Membrane Dystrophy. All patients were suffering from painful recurrent epithelial erosions. All patients had multiple clear or white cysts, mainly in the central cornea1 epithelium (dots), and as amorphous subepithelial clouding (map-like pattern) in one case. This latter patient had gray opacities close to map-like lesions, and there also were clear fenestrations within the clouding zone. Fluorescein showed cornea1 surface irregularities overlying the map and cystic lesions. Patients with superficial cornea1 lines (fingerprints) were not included in this study. Meesmann’s Dystrophy. This particular case was an asymptomatic patient in whom clinical features were found during a routine ophthalmologic examination. Small round and oval opaque areas (epithelial cysts) were present in the
632
interpalpebral zone of the cornea1 epithelium of both eyes. The cysts were mainly spherical and occasionally fused. These lesions appeared as gray dots on direct illumination and as transparent vesicles in retroillumination. Not all of the cysts took up fluorescein dye. There were opaque areas on Bowman’s membrane that appeared as a slight haze between the cysts. Recurrent Erosion Syndrome. All patients had repeated episodes of early morning irritation, acute ocular pain, photophobia, and blurring of vision. All patients had intraepithelial corneal cysts. The cysts varied widely in shape and content, but only a few (ten) cysts were found at examination.
Confocal Microscopic Findings Confocal microscopy was used to assess the size, shape, light scatter, and reflection of the cysts. Fuchs’ Dystrophy. The lesions varied from round to elongated and were from 20 pm to at least 400 pm in diameter. Cystic lesions formed three patterns. The most frequently encountered pattern was a globular shape with well-defined borders, with a linear or punctiform reflection surrounded by a dark halo within some lesions (Fig 1A, arrow). The second type of lesion had an irregular reflective image within the cyst, was surrounded by a dark halo, and was outlined by the cytoplasm of an epithelial cell (Fig IA, arrowhead). The third type of lesion was more tubular, elongated, and larger (at least 400 pm) (Fig lB, arrows). They appeared separated from other epithelial cells (Fig lB, asterisk) without any other recognizable structure within the epithelium. Punctiform reflective structures with 20 pm across were seen in the lesion (Fig lB, arrowheads). The endothelium showed an abnormal pattern, in which the normal hexagonal endothelial mosaic was lost (Fig IC). Oblique sections showed all the layers of the cornea1 epithelium, which had lost its normal, regular architecture. At the most superficial level, there was cellular irregularity (Fig 1D, arrows). The nuclei of the basal epithelial cells appeared as reflective round structures arranged in a linear pattern above the stroma (Fig lD, arrowheads). Epithelial Basement Membrane Dystrophy. Confocal microscopic examination results disclosed four main features in all cases. First, the anterior stroma and posterior epithelial interface contained highly reflective irregular material intermixed with the keratocytes (Fig 2A, arrows). This material was present even in zones of apparently unaffected cornea1 tissue and was distributed unevenly. Second, the cells of the basal epithelium have abnormal distended cytoplasm with very reflective nuclei (Fig 2B, arrow). Third, the cystic lesions had a more ovoid shape (Fig 2C, arrows). Borders were poorly defined, and there was irregular reflective material within the cysts (Fig 2C, asterisk). The diameter of cysts was 50 to 400 pm. Fourth, the basement membrane bore some long (at least 300 pm), highly reflective, linear structures (Fig 2D, arrows) that had well-delineated anterior borders and diffuse posterior borders. There was a nonreflective space between this linear lesion and the normal basement membrane. Analysis of confocal images of the patient with maplike lesions showed two patterns. There were short (50- 100 pm) linear structures arranged in parallel below the epithelium (Fig 2E, arrows). Oblique sections of epithelium showed an irregular arrangement of tissue with very reflective, round structures in a wavy-linear configuration (pearl collar-like; Fig 2F, arrows), whereas others formed clusters in the shape of a microcyst (Fig 2F, asterisk). These structures extended up into the cornea1 epithelium from the basal layer and were most probably cell nuclei.
Herndndez-Quintela et al * Confocal Microscopy of the Cornea
Figure 1. Confocal microscopuz images of the cornea of patients with Fuchs’ dystrophy. A, cystic lessons in superfictal eplthelial cells with lmear reflection (arrow) and irregular reflective image wlthm the lesion (arrowhead). B, highly reflective material (arrowheads) wlthin an elongated bulla (arrows) and an abnormal epithelium as background (asterisk). C, optical sectlon at the endothehum showing guttae and enlarged endothelial cells. D, oblique aspect of the eplthelium in which normal stroma (S), epithelial cells (Ep, arrowheads), and irregular, edematous superficial cells (arrows) are observed. Horizontal field width -610 pm. E, semIfine micrograph of the cornea of another patient with Fuchs’ dystrophy to correlate with previous Images. Parts A and B are optlcal sectlons probably done between arrows in part E (stam, tolmdme blue; orlgmal magmfication, X 160). This patlent was not mcluded m the confocal macroscopic exammatlon.
Meesmann’s Dystrophy. The cystic lesions were round, well delineated, and measured 10 to 50 pm across. Some lesions showed reflective points in the cytoplasm, which probably correspond to cell nuclei (Fig 3, arrows). Some normal cell nuclei appeared as reflective round images above an apparently normal basal membrane (Fig 3, arrowheads). Cystic lesions were the only confocal findings seen in this case. No other structural features of the disease were found. Recurrent Erosion Syndrome. There were superficial polygonal low reflective areas surrounded by bright epithelial cells (Fig 4). The superficial epithelial cells were 40 to 60 pm across.
The confocal images of the basementmembrane were normal. No epithelial cystic lesions were found by confocal microscopy.
Discussion We examined the cornea1epithelium of patients with cystic disorders of the cornea by confocal microscopy. Cystic lesions were found in 69.2% of the patients examined. The usefulnessof the confocal microscope in the diagnosis of Fuchs’ dystrophy through an opaque cornea has
633
Ophthalmology
Volume 105, Number 4, April 1998
Figure 2. Confocal microscopic images of the cornea1 eplthelmm of patients with eplthehal basement membrane dystrophy. A, tangential sectlon through the subeplthehal space shows highly reflective material (arrows). B, sectlon at the basal-eplthelial level showmg globular cytoplasm and cell nuclei (arrow). C, confocal image of a cyst wxh a diameter of 200 ,um (arrows) and reflective material wlthm the lesion (asterlqk). D, mtraeptthehal basement membrane (arrows), which appeared separated from normal basal eplthehal cells. E, obhque sectlon showmg an abnormal subeplthehal layer. Multiple fibrdlar structures are present (arrows). F, cell nuclei arranged in lmear pattern (arrows) and others formmg a mxrocyst (asterisk). Normal keratocyte nuclei seen m the antenor stroma (K). Horizontal field width ~610 Wm.
been documented.” These authors found irregularly shaped endothelial cells that were three times their normal size, with a hazy stroma and a disrupted epithelium. The current confocal microscopic examination of the cornea1 epithelium of patients with Fuchs’ dystrophy showed four patterns: (1) cystic lesions with well-delineated borders and low internal reflectivity (Fig lA, arrow); (2) cystic lesions with irregular internal reflectivity surrounded by the cytoplasm of epithelial cells (Fig lA, arrowhead); (3) elongated bullae with tubular shapes and multiple round reflective structures within the lesion (Fig 1B); and (4) abnormal epithelial cells with increased intercellular space (Fig lB, asterisk). Figure 1E is provided
634
to compare the confocal microscope image and the histopathologic picture. Figures 1A and 1B probably correspond to optical sections obtained between arrows in Figure 1E. Two factors must be taken into account before interpretation. First, most confocal microscopic sections were taken at a tangential axis. Second, the micrograph shown is not from any of the patients examined in this study, so that no direct correlation is possible. Nevertheless, it seems clear that confocal microscopy tends to show the superficial cystic lesions. The epithelial cells forming cyst walls are not consistently seen by confocal microscopy. Whenever such cells are seen, they appear flattened and arranged concentrically (Fig lA, arrow-
Herndndez-Quintela
et al * Confocal Microscopy of the Cornea to the epithelium may be abnormal epithelial cells adjacent to an aberrant basement membrane.’ One patient with Meesmann’s dystrophy had intraepithelial cysts. The basement membrane appeared normal, and no other abnormal structures were seen (Fig 3). Pathologic studies have described disordered and thickened epithelium, multilaminar basement membrane, intracystic periodic acid-Schiff- and adenosine monophosphate-positive fibrillogranular material, and glycogen within cell vacuoles 9,22,23
Figure 3. Tangenttal confocal mtcroscoptc tmage of the cornea1 eptthehum of a pattent wtth Meesmann’s dystrophy. The cyst 1s surrounded by the plasma membrane of an epnhelial cell. There are two reflecttve, small (10 pm) round tmages wtthm the cytoplasm (arrows) probably corresponding to cell nuclei. Normal epithehal cells nuclet (arrowheads). Horizontal field wtdth -610 pm.
head). The elongated horizontal profile of some lesions (Fig lB, arrows) probably is because of separation of epithelial layers, as shown in histopathologic studies.’ The internal reflectivity within lesions is caused by degenerated epithelial cells and cell debris (Fig lB, arrowheads).2,9 Figures 1D and 1E can be compared directly because both are sagittal sections. The three clinicopathologic basic elements of epithelial basement membrane dystrophy’ may explain our confocal microscopic findings: 1. The basement membrane is thickened and extends into the epithelium as a dense reflective linear structure lying over normal basal epithelial cells (Fig 2D, arrows). 2. The abnormal epithelial cells and intraepithelial microcysts appear as highly reflective cell nuclei and globular lesions, respectively (Figs 2B and 2C). 3. There is fibrillar material between the epithelial basement membrane and Bowman’s layer (Fig 2E, arrows). As in Fuchs’ dystrophy, images of the cystic lesions probably show degenerating epithelial cells and cell debriszS9The highly reflective material below the epithelium (Fig 2A, arrows) probably is caused by intracystic cell debris, which reaches the subepithelial space via a continuity of the intraepithelial cysts with the subepithelial .zone.6The abnormal appearance of the cornea1 epithelial cell in Figure 2B may be caused by intracellular edema described as “pale” cells by Tripathi and Bror? in 1973. The case with map-like cornea1 lesions had round reflective structures. They differed from normal epithelial cell nuclei in that they were smaller in size and had more uniform appearance. They were arranged in two patterns, one wavy-linear (Fig 2F, arrows) and the other encircling a cyst (Fig 2F, asterisk). These structures probably are the nuclei of multinucleate epithelial giant cells. These cells were found in a histopathologic study of a patient with fingerprint dystrophy.3 The structures extending up
davanagh et al” and Tomii and Kinoshitar’ have described the appearance of the normal human epithelium under the confocal microscope. They found great variations in the reflection of the cells from the objective. This was confirmed by Wiegand et al,” who interpreted the high reflectance as being caused by cells that were about to desquamate. In our study, superficial epithelial cells of normal control eyes were observed as polygonal structures, with reflectance variability, and cell nuclei were not seen consistently. We found this same normal pattern in the confocal microscopic images of patients with recurrent erosion syndrome (Fig 4). Confocal microscopy provided an in vivo evaluation of cystic epithelial cornea1 lesions. The images were compatible with the clinical and histologic picture of the disease. The best images were obtained during the first 40 seconds of examination. After this, eye movement increased and the video recorder did not obtain high-quality images. This instrument can be used to monitor disease progression over time. Evaluation of these changes may increase our understanding of its pathogenesis. New technology requires improved image analysis. For example, the objective analysis of desquamating cells in recurrent erosion syndrome or the observation of cyst development in epithelial basement membrane dystrophy may be a clinical application of the confocal microscope that needs to be developed. Confocal microscopy is a new and evolving technology. As such, there still is much to be learned. We believe it is a useful tool for under-
Figure 4. Confocal microscopic image of the cornea1 superficial epithelium of a patient with recurrent erosion syndrome. A mucus band appears as a bright line over the epithelial mosaic. Horizontal field width -610 Pm.
635
Ophthalmology standing cornea.
how the pathologic
processes
develop
Volume 10.5, Number 4, April 1998 in the
References 1. Bron AJ, Tripathi RC. Cystic disorders of the cornea1 epithelium. I Clinical aspects. Br J Ophthalmol 1973;57:36175. 2. Tripathi RC, Bron AJ. Cystic disorders of the cornea1 epithelium. II Pathogenesis. Br J Ophthalmol 1973;57:37690. 3. Brodrick JD, Dark AJ, Peace GW. Fingerprint Dystrophy of the Cornea. A Histologic Study. Arch Ophthalmol 1974;92:483-9. 4. Cogan DG, Kuwabara T, Donaldson DD, Collins E. Microcystic dystrophy of the cornea. A partial explanation for its pathogenesis. Arch Ophthalmol 1974;92:470-4. 5. Rodrigues MM, Fine BS, Laibson PR, Zimmerman LE. Disorders of the cornea1 epithelium. A clinicopathologic study of dot, geographic, and fingerprint patterns. Arch Ophthalmol 1974;92:475-82. 6. Fogle JA, Kenyon KR, Stark WJ, Green RW. Defective epithelial adhesion in anterior cornea1 dystrophies. Am J Ophthalmol 1975; 79:925 -40. cornea1 dystrophy. Trans Am 7. Laibson PR. Microcystic Ophthalmol Sot 1976; 74:488-53 1. 8. Dark AJ. Cogan’s microcystic dystrophy of the cornea: ultrastructure and photomicroscopy. Br J Ophthalmol 1978;62:821-30. 9. Waring III GO, Rodrigues MM, Laibson PR. Cornea1 dystrophies. I. Dystrophies of the epithelium, Bowman’s layer and stroma. Surv Ophthalmol 1978;23:71- 122. 10. Cavanagh HD, Jester JV, Essepian, et al. Confocal microscopy of the living eye. CLAO J 1990; 16:65-73.
636
11. Tomii S, Kinoshita S. Observation of the cornea1 epithelium by scanning confocal microscope. Scanning 1994; 16:3056. 12. Wiegand W, Thaer AA, Kroll P, et al. Optical sectioning of the cornea with a new confocal in vivo slit-scanning videomicroscope. Ophthalmology 1995; 102:568-75. 13. Prydal JI, Dilly PN. In vivo confocal microscopy of the cornea and tear film. Scanning 1995; 17:133-5. 14. CavanaghHD, Petrol1WM, Alizadeh H, et al. Clinical and diagnostic use of in vivo confocal microscopy in patients with cornea1 disease. Ophthalmology 1993; 100:1444-54. 15. Essepain JP, Rajpal RK, Azar DT, et al. The use of confocal microscopy in evaluating cornea1 wound healing after excimer laser keratectomy. Scanning 1994; 16:300-4. 16. Essepian JP, Rajpal R, O’Brien TP. Tandem scanning confocal microscopic analysis of ciprofloxacin cornea1 deposits in vivo. Cornea 1995; 14:402-7. 17. Corbett MC, Prydal JI, Verma S, et al. An in vivo investigation of the structures responsible for cornea1 haze after photorefractive keratectomy and their effect on visual function. Ophthalmology 1996; 103: 1366-80. 18. Winchester K, Mathers WD, Stuphin JE, Daley TF. Diagnosis of Acanthamoeba keratitis in vivo with confocal microscopy. Cornea 1995; 14:10-7. 19. Kaufman SC, Beuerman RW, Kaufman HE. Diagnosis of advanced Fuchs’ endothelial dystrophy with the confocal microscope [letter]. Am J Ophthalmol 1993; 116:652-3. 20. Cohen RA, Chew SJ, Gebhardt BM, et al. Confocal microscopy of cornea1 graft rejection. Cornea 1995; 14:467-72. 21. Mathers WD, Stuphin JE, Folberg, et al. Outbreak of keratitis presumed to be caused by Acanthamoeba. Am J Ophthalmol 1996; 121:129-42. 22. Kuwabara T, Ciccarelli EC. Meesmann’s dystrophy. A pathological study. Arch Ophthalmol 1964;71:676-82. 23. Fine BS, Yannoff M, Pitts E, Slaughter FD. Meesmann’s epithelial dystrophy of the cornea. Am J Ophthalmol 1977;83:633-42.