Handheld Reflectance Confocal Microscopy for the Diagnosis of Conjunctival Tumors ELISA CINOTTI, JEAN-LUC PERROT, BRUNO LABEILLE, NELLY CAMPOLMI, MARINE ESPINASSE, DAMIEN GRIVET, GILLES THURET, PHILIPPE GAIN, CATHERINE DOUCHET, CAROLINE ANDREA, MAHER HAOUAS, AND FRE´DE´RIC CAMBAZARD PURPOSE:
To evaluate whether the handheld in vivo reflectance confocal microscopy that has been recently developed for the study of skin tumors is suitable for the diagnosis of conjunctival tumors. DESIGN: Prospective study, observational case series. METHODS: We prospectively evaluated the reflectance confocal microscopy features of 53 conjunctival lesions clinically suspicious for tumors of 46 patients referred to the University Hospital of Saint-Etienne (France) by using the handheld device. Twenty-three lesions were excised (3 nevi, 10 melanomas, 5 squamous cell carcinoma, 2 lymphomas, and 3 pinguecula/pterygium) while the other 30, presenting no reflectance confocal microscopy malignant features, were under follow-up for at least 1 year. Clinical reflectance confocal microscopy and histologic diagnosis were compared. RESULTS: In vivo reflectance confocal microscopy diagnosis was in agreement with the histologic diagnosis in all cases and none of the lesions that were not excised show any clinical progression under follow-up. CONCLUSION: In vivo reflectance confocal microscopy with a handheld dermatology-dedicated microscope can play a role in the noninvasive diagnosis of conjunctival lesions. Further studies should be performed to better define the diagnostic ability of this technique. (Am J Ophthalmol 2015;159:324–333. Ó 2015 by Elsevier Inc. All rights reserved.)
T
HE CLINICAL DIAGNOSIS OF CONJUNCTIVAL TUMORS
is often challenging,1–4 and in case of clinical doubt of malignant tumor excision is mandatory. However, a surgical excision in this area may have functional or aesthetic consequences for the patient and may be technically complex for the surgeon. New imaging techniques are of high interest in the diagnosis of
Accepted for publication Oct 29, 2014. From the Departments of Dermatology (E.C., J.-L.P., B.L., C.A., F.C.), Ophthalmology (N.C., M.E., D.G., G.T., P.G., M.H.), and Pathology (C.D.), University Hospital of Saint-Etienne, Saint-Etienne, France; Biology, Engineering and Imaging of Corneal Graft Laboratory, EA2521, Jean Monnet University (N.C., M.E., D.G., G.T., P.G.), Saint-Etienne, France; and Institut Universitaire de France (G.T.), Paris, France. Inquiries to Elisa Cinotti, Dermatology Department, University Hospital of Saint-Etienne, 42055 Saint-Etienne Cedex 2, France; e-mail: [email protected]
324
Ó
2015 BY
conjunctival tumors to reduce the number of unnecessary surgical excisions. In vivo reflectance confocal microscopy is a noninvasive, high-resolution imaging technique that has been demonstrated to be useful for the diagnosis of conjunctival tumors.2,5–9 However, this technique is not extensively used, mainly owing to the difficulty of acquiring the images, especially because reflectance confocal microscopes dedicated to ophthalmology have limited numbers of degrees of freedom and limited possibilities of mechanical displacement. We investigated whether the handheld reflectance confocal microscope that has been developed for the cutaneous tumors, and which has been demonstrated to be applicable to the study of the genital10–13 and the oral13,14 mucosa with easy and quick examinations, could play a role in the diagnosis of conjunctival tumors. We present a series of conjunctival tumors described from a clinical standpoint and the first that has been investigated by the handheld in vivo reflectance confocal microscope.
METHODS FORTY-SIX PATIENTS (24 FEMALE, 22 MALE, AVERAGE AGE 53
years, range 13-94 years) presenting with a total of 53 conjunctival lesions suggestive of tumors were recruited at the Dermatology Department of the University Hospital of Saint-Etienne, France, between January 1, 2011 and June 30, 2013. Clinical and in vivo reflectance confocal microscopy diagnosis were prospectively established together by 2 ophthalmologists (D.G., E.M.) and 3 dermatologists (E.C., J.-L.P., B.L.). Tumors confined to the eyelid margin were excluded owing to the particular features of this area of transition between the tarsal conjunctiva and the skin. A slit-lamp examination was performed to establish the clinical diagnosis. In vivo reflectance confocal microscopy examination was carried out with the handheld microscope dedicated to skin (VivaScope 3000; Caliber, Rochester, New York, USA, distributed in Europe by MAVIG GmbH, Mu¨nchen, Germany) (Figure 1), which uses an 830-nm class 1B (classification CDRH—Center for Devices and Radiological Health) diode laser not harmful to eyes, with a wavelength that does not induce ocular glare. This system has a high optical resolution (horizontal and vertical axis: 1.25 mm and 5 mm, respectively) and allows
ELSEVIER INC. ALL
RIGHTS RESERVED.
0002-9394/$36.00 http://dx.doi.org/10.1016/j.ajo.2014.10.031
visualization of tumors up to 250 mm in depth from the epithelial surface of the conjunctiva. Each image corresponds to a horizontal section of 1 mm 3 1 mm. Before the ocular examination, topical anesthesia was performed with oxybuprocaine hydrochloride 1.6 mg/0.4 mL and tetracaine hydrochloride 1% applied into the lower conjunctival sac of the eye and a transparent ophthalmic gel of Carbomer 974P (Gel larme; The´a, Clermont-Ferrand, France) was placed on the ocular region to be examined. A disposable sterile transparent film (Visulin; Paul Hartmann AG, Heidenheim, Germany) was applied on the reflectance confocal microscope tip. In some cases an eyelid retractor was used to prevent blinking. Examinations were performed on patients in supine position. Reflectance confocal microscopy examination time per lesion was around 5-10 minutes. An institutional review board approval was obtained (institutional review board at the University Hospital of Saint-Etienne, France, number IORG0007394; study filed under reference number IRBN332014/CHUSTE). A patient informed consent was always obtained orally during the first consultation and before this examination. Before imaging the tumors, the normal conjunctiva of the contralateral eye was also observed under in vivo reflectance confocal microscopy and features of normal conjunctiva were assessed. We examined both benign and malignant conjunctival tumors, including nevi, primary acquired melanosis (PAM), melanomas, extrascleral growth of uveal melanomas, squamous cell carcinomas (SCC), and lymphomas. Moreover, some cases of pinguecula and pterygium were also observed when in doubt of SCC. Two cases of nevus of Ota were also examined to exclude the presence of areas of malignant transformation. The in vivo reflectance confocal microscopy diagnosis of nevus was determined in the presence of junctional and/or stromal hyperreflective homogeneous medium-sized (1020 mm), roundish cells organized in nests, with the absence of (1) pagetoid cells, (2) atypical cells at the epitheliumstromal junction, and (3) disarrangement of the epithelial layers.15 The presence of stromal pseudocyst-like structures partly filled with monomorphous material allowed diagnosis of the epithelial-cystic nevus. Lesions with hyperreflective cells confined to the basal layer of the epithelium16 and/or small (20 mm) were diagnosed as PAM with atypia.5 The in vivo reflectance confocal microscopy diagnosis of conjunctival melanoma and extrascleral growths of uveal melanomas were established in the presence of large (>20 mm) dendritic or roundish hyperreflective cells at the epithelial-stromal junction and/or in the stroma associated with the possible presence of large pagetoid cells. The in vivo reflectance confocal microscopy diagnosis of SCC was made in the presence of large (>20 mm) epithelial cells17 and/or a disarranged pattern of the epithelium. VOL. 159, NO. 2
FIGURE 1. In vivo reflectance confocal microscopy examination of the conjunctiva using the handheld microscope dedicated to the skin.
The presence of horizontal and dilated blood vessels and the presence of abundant dendritic cells was an additional criterion that confirmed the SCC diagnosis. Pinguecula and pterygium were diagnosed in the absence of epithelial atypia and the subepitehlial presence of degenerated stromal collagen that presented with a coiled shape and a fibrovascular proliferation, respectively. Increased leukocytes (small hyperrefractive roundish homogeneous cells) in the stroma were an additional finding. Mucosa-associated lymphoid tissue (MALT) lymphoma was diagnosed in the presence of a normal epithelium and abundant small hyperrefractive roundish cells corresponding to lymphocytes in the stroma. A surgical excision and a histopathologic examination were performed in 20 cases clinically and/or in vivo reflectance confocal microscopy suspicious for malignant tumors (melanoma, SCC, MALT lymphoma). In addition, the first 3 cases of nevi were excised to perform a histopathologic examination considering that the in vivo reflectance confocal microscopy aspect of the conjunctival nevi had never been described using the handheld camera. The correlation between in vivo reflectance confocal microscopy and the histopathologic examination was evaluated. Lesions that were not biopsied because considered benign at in vivo reflectance confocal microscopy examination were then monitored for at least 12 months.
RESULTS IN VIVO REFLECTANCE CONFOCAL MICROSCOPY FEATURES
of the normal conjunctiva were the presence of (1) 3-6
HANDHELD CONFOCAL MICROSCOPY FOR CONJUNCTIVAL TUMORS
325
FIGURE 2. In vivo reflectance confocal microscopy features of the normal conjunctiva. Thanks to the convexity of the ocular surface it is possible to obtain a vertical section of the conjunctiva (Left), which show the epithelium (white bracket), the stroma (black bracket), and a flat epithelial-stromal junction (arrow). The epithelium (Center) is composed by polygonal medium size cells with hyperreflective membranes, hyporeflective cytoplasm, and medium reflective round nuclei. In the stroma (Right) elongated hyperreflective structures are organized in a dense meshwork (corresponding to collagen fibers) and prominent capillaries (arrow) are visible.
layers of polygonal medium-size (10 mm) cells with hyperreflective membranes, hyporeflective cytoplasm, and medium reflective round nuclei in the epithelium, (2) a flat epithelial-stromal junction and (3) elongated hyperreflective structures organized in a dense meshwork (corresponding to collagen fibers) and prominent capillaries in the stroma (Figure 2). The movement of blood cell flow could be observed in real time inside blood capillaries. The clinical features, the in vivo reflectance confocal microscopy, and the histopathologic diagnosis of the lesions are reported in the Table. Considering all the 23 excised lesions (Figures 3 and 4), 12 lesions were clinically challenging, and 3 cases of melanomas were clinically diagnosed as benign lesions, whereas in vivo reflectance confocal microscopy diagnosis was in agreement with the histologic diagnosis in all cases, including 10 melanomas, 5 SCC, and 2 MALT lymphomas. The 30 lesions that were not biopsied were diagnosed by in vivo reflectance confocal microscopy as 20 nevi (including 7 nevi with epithelial-cystic inclusions), 2 nevus of Ota (Figure 5), 2 pterygium, and 6 PAM without atypia (Figure 6), and showed no changes following consecutive clinical and in vivo reflectance confocal microscopy monitoring for at least 12 months. No complications such as ocular inflammation and/or infections and/or mechanical trauma were found using this imaging technique.
DISCUSSION OUR STUDY SHOWED THAT HANDHELD IN VIVO REFLEC-
tance confocal microscope dedicated to skin can be a new tool for a noninvasive diagnosis of conjunctival tumors. Two in vivo reflectance confocal microscopes are currently available to explore the ocular surface: the 326
Confoscan 4 slit-scanning confocal microscope (Nidek, Gamagori, Japan) and the in vivo reflectance confocal microscope Heidelberg Retina Tomograph (HRT),18,19 in association with the Rostock Cornea Module (Heidelberg Engineering GmbH, Heidelberg, Germany). Differently from the in vivo reflectance confocal microscope dedicated to skin, these devices do not have a handheld probe but present a hinged support for the camera that renders more difficult the precise alignment with the region of interest. Because of this limitation in the handiness, they are mainly used to visualize cornea and, more rarely, the conjunctiva near the axis.20 There are only few studies2,5–9 about their use for conjunctival tumors. One study5 evaluated 28 melanocytic tumors, 1 article described a case of lymphoma,9 and all the other publications are about squamous neoplasia.2,6–8,21 For the first time we used the handheld in vivo reflectance confocal microscope dedicated to skin (VivaScope 3000) to explore the conjunctiva. This device, thanks to its ease of handling (handheld probe of small size, light weight, equipped with manual control buttons to move in the vertical axis of the tissue) and its small tip, has been recently applied to the imaging of the mucosa10,12,13 and was suitable for the examination of the whole ocular area. VivaScope 3000 is expensive, but the same device can be shared between dermatologists and ophthalmologists, as we demonstrated that it can be used for both skin and eye. Most of the largest dermatologic centers already have this device. Examination is best performed on a cooperative patient who should not excessively move his eyes, while the operator continuously adapts the position of the camera. During our study we also explored the normal conjunctiva that presented with similar features to those described for the Heidelberg device.22–24
AMERICAN JOURNAL OF OPHTHALMOLOGY
FEBRUARY 2015
VOL. 159, NO. 2
TABLE. Clinical Features and Handheld Reflectance Confocal Microscopy and Histologic Diagnosis of the Conjunctival Lesions
Patient Number Sex Age
HANDHELD CONFOCAL MICROSCOPY FOR CONJUNCTIVAL TUMORS
Location
Size (mm)
1 2 3 4 4 5 5 5 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
M F F F F M M M M F M F M M F F M M F F M M F M M F M F M F
78 73 94 42 42 64 64 64 64 66 60 19 59 81 55 47 73 77 59 69 71 73 64 51 56 71 17 23 58 33
Bulbar Bulbar/tarsal Bulbar Bulbar Tarsal Bulbar Tarsal Bulbar Tarsal Limbal Caruncle Limbal Bulbar Limbal Limbal/bulbar Limbal Limbal Limbal Limbal Limbal Limbal Tarsal Tarsal Bulbar Limbal Caruncle Limbal/bulbar Caruncle Bulbar Bulbar
232 533 433 332 432 333 334 2 336 638 433 131 10 3 6 332 9 3 10 535 535 535 333 433 334 835 433 332 131 232 835 331 736 434
27 28 29 30 31
F F F M M
28 32 84 61 31
Bulbar Bulbar Bulbar Caruncle Limbal
Clinical Feature
Dark brown flat papule Multiple macules, light brown Multiple macules, dark brown Light brown macule Light brown macule Red macule Brown macule Brown macule Brown macule Black flat papule Pink papule Brown macule Brown macule Flat rose papule Red macule Pink macule Reddish papule Pink nodule Whitish nodule Whitish nodule Pink papule Pink patch Pink papules Brown macule Light brown macule Light brown papule Brown-yellow patch Brown macule Flat brown papule Macule with a light and dark brown pigmentation 5 3 5 Brown patch 1.5 Brown papule 2 3 4 Light brown macule 15 3 7 Brown macule Brown macule
Evolution (Years)
0.5 10 10 2 2 9 months 9 months 3 months 10 months 7 NA 1 10 0.5 1 NA 1 month 0.5 10 2 1 1 month 0.5 3 months Congenital 5 Congenital NA 48 Congenital 10 2 1 45 5
Clinical Diagnosis
Melanoma? Melanoma Melanoma Melanoma Melanoma Pinguecula? Nevus? Melanoma? Melanoma Nevus? Nevus Nevus Nevus SCC? SCC SCC SCC SCC? SCC? SCC? SCC? MALT lymphoma? MALT lymphoma? Melanoma? Nevus Nevus Nevus Nevus Nevus Nevus
Confocal Microscopy Diagnosis
Histologic Diagnosis
Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Compound nevus Subepithelial nevus Epithelial-cystic nevus SCC SCC SCC SCC SCC Pinguecula Pterygium Pinguecula MALT lymphoma MALT lymphoma Junctional nevus Dermal nevus Subepithelial nevus Compound nevus Junctional nevus Epithelial-cystic nevus Junctional nevus
Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Compound nevus Subepithelial nevus Epithelial-cystic nevus SCC in situ SCC in situ SCC SCC SCC Pinguecula Pinguecula Pinguecula MALT lymphoma MALT lymphoma NA NA NA NA NA NA NA
Epithelial-cystic nevus Epithelial-cystic nevus NA Nevus Epithelial-cystic nevus NA Nevus Epithelial-cystic nevus NA Nevus Epithelial-cystic nevus NA Nevus Epithelial-cystic nevus NA
Confocal Microscopy/Histology Correlation
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes NA NA NA NA NA NA NA NA NA NA NA NA
327
Continued on next page
328
TABLE. Clinical Features and Handheld Reflectance Confocal Microscopy and Histologic Diagnosis of the Conjunctival Lesions (Continued )
Patient Number Sex Age
Location
Size (mm)
AMERICAN JOURNAL OF OPHTHALMOLOGY
32 33 34 35 35 35 36
M M M M M M F
Limbal Bulbar Bulbar Bulbar Bulbar Bulbar Bulbar
735 533 332 3 2 2 634
37
M 27 Caruncule, conjunctive F 59 Bulbar M 48 Bulbar F 48 Bulbar F 62 Limbal F 66 Tarsal F 33 Limbal F 40 Bulbar F 29 Bulbar M 72 Limbal F 68 Limbal
1
15 38 39 40 41 42 43 44 45 46
17 68 51 41 41 41 13
131 334 231 5 4 10 334 532 232
Clinical Diagnosis
10 10 10 10 10 10 Congenital, but recently enlarged 3
Nevus Epithelial-cystic nevus Epithelial-cystic nevus Racial melanosis Racial melanosis Racial melanosis Nevus
Subepithelial nevus Epithelial-cystic nevus Epithelial-cystic nevus Compound nevus Compound nevus Compound nevus Subepithelial nevus
NA NA NA NA NA NA NA
NA NA NA NA NA NA NA
Nevus
Subepithelial nevus
NA
NA
NA Congenital
Nevus of Ota Nevus of Ota PAM PAM PAM PAM Racial melanosis Racial melanosis Pterygium vs SCC? Pterygium vs SCC?
Nevus of Ota Nevus of Ota PAM PAM PAM PAM PAM PAM Pterygium Pterygium
NA NA NA NA NA NA NA NA NA NA
NA NA No NA NA NA NA NA NA NA
Clinical Feature
Erythematous macule Flat light brown papule Brown papule Light brown macule Light brown macule Light brown macule Erythematous macule Brown macule Brown and gray macules Brown and gray macules Multifocal light brown macule Light brown macule Light brown macule Light brown macule Brown macule Light brown macule Grayish nodule Red nodule
Confocal Microscopy/Histology Correlation
Confocal Microscopy Diagnosis
Evolution (Years)
10 10 0.5 10 6 4 NA 10
MALT ¼ mucosa-associated lymphoid tissue; NA ¼ not available; PAM ¼ primary acquired melanosis; SCC ¼ squamous cell carcinoma.
Histologic Diagnosis
FEBRUARY 2015
FIGURE 3. Clinical (Left), histologic (Center) and handheld reflectance confocal microscopy (Right) aspect of a subepithelial nevus (First line), an epithelial-cystic nevus (Second line), and a melanoma (Third line). Handheld reflectance confocal microscopy shows in the subepithelial nevus homogeneous nests of cells (red asterisk) in the stroma; in the epithelial-cystic nevus stromal pseudocystlike structures (red asterisk) partly filled with monomorphous material surrounded by hyperreflective homogeneous cells (red circles); in the melanoma uneven large cells (red arrow) of different shape (roundish, dendritic and polygonal cells) in the epithelium. (Histologic stains: hematoxylin-eosin, first and third lines 203, second line 43.)
Given the lack of in vivo reflectance confocal microscopy validated criteria to diagnose tumors with the handheld VivaScope 3000 camera, in particular in the conjunctival area, we defined new criteria based on those used in the 2 studies of conjunctival tumors explored by the Heildelberg in vivo reflectance confocal microscope,9,22 on in vivo reflectance confocal microscopy diagnostic criteria used for skin tumors15,17 and for genital and oral mucosal10,13 tumors, on histology of conjunctival tumors,4,25 and on our personal experience. Using our diagnostic criteria, in vivo reflectance confocal microscopy diagnosis was in agreement with histology in all cases, and in particular for SCC, melanoma, and MALT lymphoma. In vivo reflectance confocal microscopy was mostly used to confirm the clinical diagnosis of malignant tumors, but in 5 cases of SCC, 4 cases of melanoma, and 2 cases of lymphoma the clinical diagnosis was completely uncertain. Interestingly, it also found VOL. 159, NO. 2
melanomas in 3 lesions that were not clinically suspicious for malignant tumors. In all SCC we found large epithelial cells (Figure 4) and an abundant infiltrate of dendritic cells. Enlarged epithelial cells corresponding to neoplastic epitheliocytes could also be found by Falke and associates2 in their in vivo reflectance confocal microscopy observation on a case of in situ SCC. Nguena and associates21 recently reported a series of 26 SCC observed by in vivo reflectance confocal microscopy in a black population and could not find a statistically significant difference for the parameters that have been so far identified as characteristic of SCC (presence of irregular cell size, hyperreflectivity, prominent bright nucleoli, and mitotic figures) compared to 18 benign conjunctival lesions. Based on our experience we agree that hyperreflectivity, prominent bright nucleoli, and mitotic figures are not in vivo reflectance confocal microscopy valid criteria for SCC. Irregular cell size is a difficult criterion to be evaluated
HANDHELD CONFOCAL MICROSCOPY FOR CONJUNCTIVAL TUMORS
329
FIGURE 4. Clinical and handheld reflectance confocal microscopy appearance of a mucosa-associated lymphoid tissue lymphoma (First line) of the superior (Left) and inferior (Right) tarsal conjunctiva; clinical (Left), histologic (Center), and reflectance confocal microscopy (Right) appearance of a pinguecula (Second line); and a squamous cell carcinoma (Third line). Reflectance confocal microscopy shows in the lymphoma (First line) widespread small hyperrefractive roundish cells corresponding to tumoral lymphocytes in the stroma (red arrow and circle); in the pinguecula (Second line) degenerated stromal collagen with a coiled shape (yellow circle) that corresponds to homogenized areas of collagen in the histologic examination (black asterisk); in the squamous cell carcinoma (Third line) a disarranged pattern of the epithelium (red circle) with some large epithelial cells (red arrow) and abundant dendritic cells (yellow arrow). (Histologic stains: hematoxylin-eosin, second line 103, third line 403.)
because epitheliocytes are not completely regular in form and size in physiologic conditions; moreover, the extent of irregularity that is considered pathologic should be defined. For these reasons in our series we only look for the presence of large epithelial cells (>20 mm). In case of advanced SCC, the irregularity of the cells is more evident and a completely disarranged pattern can be observed, corresponding to epitheliocytes that are completely distorted in their shape and organization. However, we should notice that pterygium can sometimes present dysplastic and pleomorphic cells at histology.26,27 Therefore, although in our small series epithelial cells of pterygium were not large in size and pleomorphic, and pterygium could be differentiated from SCC, large cells 330
could be eventually found in these lesions, and this aspect could point out lesions with epithelial atypia that could be associated with ocular surface squamous neoplasia27 and need surgical excision. We should also notice that in our study in vivo reflectance confocal microscopy images were interpreted in real time during the examination, whereas in the study of Nguena and associates21 in vivo reflectance confocal microscopy images were read retrospectively masked to the clinical appearance. In all cases of melanoma we found enlarged hyperreflective cells, roundish or dendritic, in the epithelium and/or the stroma, corresponding to malignant melanocytes (Figure 3). These cells were uneven in size and sometimes enlarged nuclei were visible as central hyporeflective areas.
AMERICAN JOURNAL OF OPHTHALMOLOGY
FEBRUARY 2015
FIGURE 5. Clinical (Left) and handheld reflectance confocal microscopy (Center and Right) appearance of a nevus of Ota. Handheld reflectance confocal microscopy of the conjunctival part of the nevus showed large homogeneous dendritic bright cells (red arrow) between the collagen fibers.
FIGURE 6. Clinical (Left) and handheld reflectance confocal microscopy (Right) aspect of primary acquired melanosis without atypia. Handheld reflectance confocal microscopy shows homogeneous hyperreflective cells confined to the basal layer of the epithelium.
Messmer and associates5 have already demonstrated in a study on 28 lesions that in vivo reflectance confocal microscopy had a high sensitivity and specificity for the melanocytic lesions of the conjunctiva, finding a sensitivity of 89% and a specificity of 100% for melanoma. They misdiagnosed only 1 lesion over 28, diagnosing PAM with atypia in a challenging lesion that histology classified as melanoma/PAM with atypia. With our study we confirmed the diagnostic ability of in vivo reflectance confocal microscopy for melanocytic lesions and we confirmed that in vivo reflectance confocal microscopy can also identify melanocytes as hyperreflective cells in amelanotic melanocytic lesions (3 in our series, 1 melanoma and 2 nevi). Concerning the 6 patients with conjunctival melanomas, 3 had a previous history of ocular melanoma (2 uveal, 1 of the eyelid), and 4 had a multifocal conjunctival involvement, indicating that in vivo reflectance confocal microscopy can be a useful imaging technique to monitor patients with a history of melanoma and multifocal lesions. VOL. 159, NO. 2
The in vivo reflectance confocal microscopy aspect of MALT lymphoma has already been reported by Pichierri and associates,9 consisting of highly reflective small-size roundish cells arranged diffusely or in nests at epithelial and subepithelial conjunctival levels. In our 2 cases we found the same kind of cells in the stroma (Figure 4), but in vivo reflectance confocal microscopy could not identify cytologic atypia in order to distinguish tumoral from inflammatory cells. Therefore, in vivo reflectance confocal microscopy could point out the necessity of a histologic examination in case of an evocative aspect of lymphoma, but the final diagnosis should be left to histology. Considering the excellent diagnostic ability with the use of in vivo reflectance confocal microscopy for conjunctival malignant tumors and the fact that conjunctival surgery could be disturbing for the patient, we decided not to excise 30 lesions that were considered benign by in vivo reflectance confocal microscopy. These lesions were diagnosed as nevus in 20 cases, a nevus of Ota in 2 cases, pterygium
HANDHELD CONFOCAL MICROSCOPY FOR CONJUNCTIVAL TUMORS
331
in 2 cases, and PAM without atypia in 6 cases. The followup during at least 12 months of all these lesions did not show any clinical progression. In the 2 cases of nevus of Ota the exploration of the entire nevus was possible, differently from what is done performing focal biopsies of clinically suspicious areas. The aspect of nevus of Ota (Figure 5) has never been described using in vivo reflectance confocal microscopy, and we found large homogeneous dendritic bright cells between the collagen fibers that could be either isolated or grouped in clusters and could also be found in apparently uninvolved conjunctiva. The absence of malignant transformation was assessed by the absence of polymorphic hyperreflective cells, uneven in size and shape, suggestive of malignant melanocytes. In conclusion, our pilot study strongly suggests that handheld in vivo reflectance confocal microscopy can be efficiently employed for a noninvasive and rapid investigation of the conjunctiva, including melanoma, SCC, and MALT lymphoma. To date, the diagnosis of conjunctival lesions has been based on the slit-lamp appearance and surgery was performed in all suspicious cases. In our series most
lesions could be accurately diagnosed clinically by their slit-lamp appearance, but some were either under- or over-diagnosed. Handheld in vivo reflectance confocal microscopy is an additional tool that can add information to the clinical examination in order to proceed with more evidence to the surgical excision. Although in vivo reflectance confocal microscopy does not replace histologic examination for definitive diagnosis, it can avoid unnecessary surgery in patients who cannot undergo surgery, such as elderly or debilitated patients. Moreover, in cases of large or multifocal lesions it allows exploration of an entire lesion differently from a biopsy. Larger multicenter studies should now be performed to validate our data and better define the diagnostic capability of handheld in vivo reflectance confocal microscopy. In particular, diagnosis performances should be compared with ophthalmologic in vivo reflectance confocal microscopy, and management using this advanced handheld in vivo reflectance confocal microscopy should be compared to the conventional management using slitlamp examination and biopsies.
ALL AUTHORS HAVE COMPLETED AND SUBMITTED THE ICMJE FORM FOR DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST and the following were reported. G.T. and P.G. are consultants for The´a Laboratories and Quantel Medical. J.-L.P., B.L., and F.C. are consultants for Leo Pharma, Pierre Fabre, LVMH, MSD, Amgen, Pfizer, and Roche Laboratories. Funding was provided through the Groupement Interregional pour la recherche Clinique Rhoˆne-Alpes Auvergne (GIRCI RAA) (Lyon, France) 2012 Young Investigator award. Contributions of authors: design and conduct of the study (E.C., J.-L.P., B.L., N.C., M.E., D.G., G.T., P.G., C.D., C.A., F.C.); collection of data, management (E.C., J.-L.P., B.L., N.C., M.E., D.G., G.T., P.G., C.D., C.A., M.H.); analysis and interpretation of the data (E.C., J.-L.P., B.L.); preparation of the manuscript (E.C.); review and approval of the manuscript (E.C., J.-L.P., B.L., N.C., M.E., D.G., G.T., P.G., C.D., C.A., M.H., F.C.).
REFERENCES 1. Maschi C, Caujolle J-P, Liolios I, Costet C. Benign conjunctival tumors. J Fr Ophtalmol 2013;36(9):796–802. 2. Falke K, Zhivov A, Zimpfer A, Stachs O, Guthoff RF. Diagnosis of conjunctival neoplastic lesions by confocal in-vivo microscopy. Klin Monatsbla¨tter Fu¨r Augenheilkd 2012; 229(7):724–727. 3. Folberg R, Jakobiec FA, Bernardino VB, Iwamoto T. Benign conjunctival melanocytic lesions. Clinicopathologic features. Ophthalmology 1989;96(4):436–461. 4. Ranty M-L, Quintyn J-C, Uro-Coste E, Delisle M-B. Ocular conjunctival pathology. A ten-year retrospective study in Toulouse-Rangueil University Hospital and literature review. Ann Pathol 2012;32(3):170–176. 5. Messmer EM, Mackert MJ, Zapp DM, Kampik A. In vivo confocal microscopy of pigmented conjunctival tumors. Graefes Arch Clin Exp Ophthalmol 2006;244(11):1437–1445. 6. Duchateau N, Hugol D, D’Hermies F, et al. Contribution of in vivo confocal microscopy to limbal tumor evaluation. J Fr Ophtalmol 2005;28(8):810–816. 7. Xu Y, Zhou Z, Xu Y, et al. The clinical value of in vivo confocal microscopy for diagnosis of ocular surface squamous neoplasia. Eye (Lond) 2012;26(6):781–787. 8. Parrozzani R, Lazzarini D, Dario A, Midena E. In vivo confocal microscopy of ocular surface squamous neoplasia. Eye (Lond) 2011;25(4):455–460.
332
9. Pichierri P, Martone G, Loffredo A, Traversi C, Polito E. In vivo confocal microscopy in a patient with conjunctival lymphoma. Clin Experiment Ophthalmol 2008;36(1):67–69. 10. Cinotti E, Perrot JL, Labeille B, Adegbidi H, Cambazard F. Reflectance confocal microscopy for the diagnosis of vulvar melanoma and melanosis: preliminary results. Dermatol Surg 2012;38(12):1962–1967. 11. Cinotti E, Chol C, Perrot JL, Labeille B, Forest F, Cambazard F. Anal melanosis diagnosed by reflectance confocal microscopy. Australas J Dermatol 2013; http://dx. doi.org/10.1111/ajd.12096. 12. Cinotti E, Perrot JL, Labeille B, Douchet C, Mottet N, Cambazard F. Laser photodynamic treatment for in situ squamous cell carcinoma of the glans monitored by reflectance confocal microscopy. Australas J Dermatol 2014; 55(1):72–74. 13. Debarbieux S, Perrot JL, Erfan N, et al. Reflectance confocal microscopy of mucosal pigmented macules: a review of 56 cases including 10 macular melanoma. Br J Dermatol 2014; 170(6):1276–1284. 14. Erfan N, Hofman V, Desruelles F, et al. Labial melanotic macule: a potential pitfall on reflectance confocal microscopy. Dermatol Basel Switz 2012;224(3):209–211. 15. Guitera P, Menzies SW, Longo C, Cesinaro AM, Scolyer RA, Pellacani G. In vivo confocal microscopy for diagnosis of melanoma and basal cell carcinoma using a two-step method: analysis of 710 consecutive
AMERICAN JOURNAL OF OPHTHALMOLOGY
FEBRUARY 2015
16.
17.
18.
19.
20.
clinically equivocal cases. J Invest Dermatol 2012;132(10): 2386–2394. Langley RGB, Burton E, Walsh N, Propperova I, Murray SJ. In vivo confocal scanning laser microscopy of benign lentigines: comparison to conventional histology and in vivo characteristics of lentigo maligna. J Am Acad Dermatol 2006;55(1):88–97. Rishpon A, Kim N, Scope A, et al. Reflectance confocal microscopy criteria for squamous cell carcinomas and actinic keratoses. Arch Dermatol 2009;145(7):766–772. Niederer RL, McGhee CNJ. Clinical in vivo confocal microscopy of the human cornea in health and disease. Prog Retin Eye Res 2010;29(1):30–58. Wang Y, Le Q, Zhao F, Hong J, et al. Application of in vivo laser scanning confocal microscopy for evaluation of ocular surface diseases: lessons learned from pterygium, meibomian gland disease, and chemical burns. Cornea 2011;30(Suppl 1):S25–S28. Zhivov A, Stachs O, Kraak R, Stave J, Guthoff RF. In vivo confocal microscopy of the ocular surface. Ocul Surf 2006; 4(2):81–93.
VOL. 159, NO. 2
21. Nguena MB, van den Tweel JG, Makupa W, et al. Diagnosing ocular surface squamous neoplasia in East Africa: case-control study of clinical and in vivo confocal microscopy assessment. Ophthalmology 2014;121(2):484–491. 22. Messmer EM, Mackert MJ, Zapp DM, Kampik A. In vivo confocal microscopy of normal conjunctiva and conjunctivitis. Cornea 2006;25(7):781–788. 23. Efron N, Al-Dossari M, Pritchard N. In vivo confocal microscopy of the palpebral conjunctiva and tarsal plate. Optom Vis Sci 2009;86(11):E1303–E1308. 24. Efron N, Al-Dossari M, Pritchard N. In vivo confocal microscopy of the bulbar conjunctiva. Clin Experiment Ophthalmol 2009;37(4):335–344. 25. Shields CL, Shields JA. Tumors of the conjunctiva and cornea. Surv Ophthalmol 2004;49(1):3–24. 26. Golu T, Mogoanta L, Streba CT, et al. Pterygium: histological and immunohistochemical aspects. Romanian J Morphol Embryol Rev 2011;52(1):153–158. 27. Chui J, Coroneo MT, Tat LT, Crouch R, Wakefield D, Di Girolamo N. Ophthalmic pterygium. Am J Pathol 2011; 178(2):817–827.
HANDHELD CONFOCAL MICROSCOPY FOR CONJUNCTIVAL TUMORS
333
Biosketch Elisa Cinotti, MD, graduated from University of Genoa (Italy) cum laude in 2007. She did her Dermatology residency at the University of Genoa. Since 2012 she is working in the University Hospital of Saint-Etienne (France) where she is conducting several researches on skin and eye noninvasive imaging, in particular on Reflectance Confocal Microscopy in cooperation with ophthalmologists. She has published over 40 peer-reviewed articles, most of them about confocal microscopy.
333.e1
AMERICAN JOURNAL OF OPHTHALMOLOGY
FEBRUARY 2015
To evaluate whether the handheld in vivo reflectance confocal microscopy that has been recently developed for the study of skin tumors is suitable for the diagnosis of conjunctival tumors. DESIGN: Prospective study, observational case series. METHODS: We prospectively evaluated the reflectance confocal microscopy features of 53 conjunctival lesions clinically suspicious for tumors of 46 patients referred to the University Hospital of Saint-Etienne (France) by using the handheld device. Twenty-three lesions were excised (3 nevi, 10 melanomas, 5 squamous cell carcinoma, 2 lymphomas, and 3 pinguecula/pterygium) while the other 30, presenting no reflectance confocal microscopy malignant features, were under follow-up for at least 1 year. Clinical reflectance confocal microscopy and histologic diagnosis were compared. RESULTS: In vivo reflectance confocal microscopy diagnosis was in agreement with the histologic diagnosis in all cases and none of the lesions that were not excised show any clinical progression under follow-up. CONCLUSION: In vivo reflectance confocal microscopy with a handheld dermatology-dedicated microscope can play a role in the noninvasive diagnosis of conjunctival lesions. Further studies should be performed to better define the diagnostic ability of this technique. (Am J Ophthalmol 2015;159:324–333. Ó 2015 by Elsevier Inc. All rights reserved.)
T
HE CLINICAL DIAGNOSIS OF CONJUNCTIVAL TUMORS
is often challenging,1–4 and in case of clinical doubt of malignant tumor excision is mandatory. However, a surgical excision in this area may have functional or aesthetic consequences for the patient and may be technically complex for the surgeon. New imaging techniques are of high interest in the diagnosis of
Accepted for publication Oct 29, 2014. From the Departments of Dermatology (E.C., J.-L.P., B.L., C.A., F.C.), Ophthalmology (N.C., M.E., D.G., G.T., P.G., M.H.), and Pathology (C.D.), University Hospital of Saint-Etienne, Saint-Etienne, France; Biology, Engineering and Imaging of Corneal Graft Laboratory, EA2521, Jean Monnet University (N.C., M.E., D.G., G.T., P.G.), Saint-Etienne, France; and Institut Universitaire de France (G.T.), Paris, France. Inquiries to Elisa Cinotti, Dermatology Department, University Hospital of Saint-Etienne, 42055 Saint-Etienne Cedex 2, France; e-mail: [email protected]
324
Ó
2015 BY
conjunctival tumors to reduce the number of unnecessary surgical excisions. In vivo reflectance confocal microscopy is a noninvasive, high-resolution imaging technique that has been demonstrated to be useful for the diagnosis of conjunctival tumors.2,5–9 However, this technique is not extensively used, mainly owing to the difficulty of acquiring the images, especially because reflectance confocal microscopes dedicated to ophthalmology have limited numbers of degrees of freedom and limited possibilities of mechanical displacement. We investigated whether the handheld reflectance confocal microscope that has been developed for the cutaneous tumors, and which has been demonstrated to be applicable to the study of the genital10–13 and the oral13,14 mucosa with easy and quick examinations, could play a role in the diagnosis of conjunctival tumors. We present a series of conjunctival tumors described from a clinical standpoint and the first that has been investigated by the handheld in vivo reflectance confocal microscope.
METHODS FORTY-SIX PATIENTS (24 FEMALE, 22 MALE, AVERAGE AGE 53
years, range 13-94 years) presenting with a total of 53 conjunctival lesions suggestive of tumors were recruited at the Dermatology Department of the University Hospital of Saint-Etienne, France, between January 1, 2011 and June 30, 2013. Clinical and in vivo reflectance confocal microscopy diagnosis were prospectively established together by 2 ophthalmologists (D.G., E.M.) and 3 dermatologists (E.C., J.-L.P., B.L.). Tumors confined to the eyelid margin were excluded owing to the particular features of this area of transition between the tarsal conjunctiva and the skin. A slit-lamp examination was performed to establish the clinical diagnosis. In vivo reflectance confocal microscopy examination was carried out with the handheld microscope dedicated to skin (VivaScope 3000; Caliber, Rochester, New York, USA, distributed in Europe by MAVIG GmbH, Mu¨nchen, Germany) (Figure 1), which uses an 830-nm class 1B (classification CDRH—Center for Devices and Radiological Health) diode laser not harmful to eyes, with a wavelength that does not induce ocular glare. This system has a high optical resolution (horizontal and vertical axis: 1.25 mm and 5 mm, respectively) and allows
ELSEVIER INC. ALL
RIGHTS RESERVED.
0002-9394/$36.00 http://dx.doi.org/10.1016/j.ajo.2014.10.031
visualization of tumors up to 250 mm in depth from the epithelial surface of the conjunctiva. Each image corresponds to a horizontal section of 1 mm 3 1 mm. Before the ocular examination, topical anesthesia was performed with oxybuprocaine hydrochloride 1.6 mg/0.4 mL and tetracaine hydrochloride 1% applied into the lower conjunctival sac of the eye and a transparent ophthalmic gel of Carbomer 974P (Gel larme; The´a, Clermont-Ferrand, France) was placed on the ocular region to be examined. A disposable sterile transparent film (Visulin; Paul Hartmann AG, Heidenheim, Germany) was applied on the reflectance confocal microscope tip. In some cases an eyelid retractor was used to prevent blinking. Examinations were performed on patients in supine position. Reflectance confocal microscopy examination time per lesion was around 5-10 minutes. An institutional review board approval was obtained (institutional review board at the University Hospital of Saint-Etienne, France, number IORG0007394; study filed under reference number IRBN332014/CHUSTE). A patient informed consent was always obtained orally during the first consultation and before this examination. Before imaging the tumors, the normal conjunctiva of the contralateral eye was also observed under in vivo reflectance confocal microscopy and features of normal conjunctiva were assessed. We examined both benign and malignant conjunctival tumors, including nevi, primary acquired melanosis (PAM), melanomas, extrascleral growth of uveal melanomas, squamous cell carcinomas (SCC), and lymphomas. Moreover, some cases of pinguecula and pterygium were also observed when in doubt of SCC. Two cases of nevus of Ota were also examined to exclude the presence of areas of malignant transformation. The in vivo reflectance confocal microscopy diagnosis of nevus was determined in the presence of junctional and/or stromal hyperreflective homogeneous medium-sized (1020 mm), roundish cells organized in nests, with the absence of (1) pagetoid cells, (2) atypical cells at the epitheliumstromal junction, and (3) disarrangement of the epithelial layers.15 The presence of stromal pseudocyst-like structures partly filled with monomorphous material allowed diagnosis of the epithelial-cystic nevus. Lesions with hyperreflective cells confined to the basal layer of the epithelium16 and/or small (20 mm) were diagnosed as PAM with atypia.5 The in vivo reflectance confocal microscopy diagnosis of conjunctival melanoma and extrascleral growths of uveal melanomas were established in the presence of large (>20 mm) dendritic or roundish hyperreflective cells at the epithelial-stromal junction and/or in the stroma associated with the possible presence of large pagetoid cells. The in vivo reflectance confocal microscopy diagnosis of SCC was made in the presence of large (>20 mm) epithelial cells17 and/or a disarranged pattern of the epithelium. VOL. 159, NO. 2
FIGURE 1. In vivo reflectance confocal microscopy examination of the conjunctiva using the handheld microscope dedicated to the skin.
The presence of horizontal and dilated blood vessels and the presence of abundant dendritic cells was an additional criterion that confirmed the SCC diagnosis. Pinguecula and pterygium were diagnosed in the absence of epithelial atypia and the subepitehlial presence of degenerated stromal collagen that presented with a coiled shape and a fibrovascular proliferation, respectively. Increased leukocytes (small hyperrefractive roundish homogeneous cells) in the stroma were an additional finding. Mucosa-associated lymphoid tissue (MALT) lymphoma was diagnosed in the presence of a normal epithelium and abundant small hyperrefractive roundish cells corresponding to lymphocytes in the stroma. A surgical excision and a histopathologic examination were performed in 20 cases clinically and/or in vivo reflectance confocal microscopy suspicious for malignant tumors (melanoma, SCC, MALT lymphoma). In addition, the first 3 cases of nevi were excised to perform a histopathologic examination considering that the in vivo reflectance confocal microscopy aspect of the conjunctival nevi had never been described using the handheld camera. The correlation between in vivo reflectance confocal microscopy and the histopathologic examination was evaluated. Lesions that were not biopsied because considered benign at in vivo reflectance confocal microscopy examination were then monitored for at least 12 months.
RESULTS IN VIVO REFLECTANCE CONFOCAL MICROSCOPY FEATURES
of the normal conjunctiva were the presence of (1) 3-6
HANDHELD CONFOCAL MICROSCOPY FOR CONJUNCTIVAL TUMORS
325
FIGURE 2. In vivo reflectance confocal microscopy features of the normal conjunctiva. Thanks to the convexity of the ocular surface it is possible to obtain a vertical section of the conjunctiva (Left), which show the epithelium (white bracket), the stroma (black bracket), and a flat epithelial-stromal junction (arrow). The epithelium (Center) is composed by polygonal medium size cells with hyperreflective membranes, hyporeflective cytoplasm, and medium reflective round nuclei. In the stroma (Right) elongated hyperreflective structures are organized in a dense meshwork (corresponding to collagen fibers) and prominent capillaries (arrow) are visible.
layers of polygonal medium-size (10 mm) cells with hyperreflective membranes, hyporeflective cytoplasm, and medium reflective round nuclei in the epithelium, (2) a flat epithelial-stromal junction and (3) elongated hyperreflective structures organized in a dense meshwork (corresponding to collagen fibers) and prominent capillaries in the stroma (Figure 2). The movement of blood cell flow could be observed in real time inside blood capillaries. The clinical features, the in vivo reflectance confocal microscopy, and the histopathologic diagnosis of the lesions are reported in the Table. Considering all the 23 excised lesions (Figures 3 and 4), 12 lesions were clinically challenging, and 3 cases of melanomas were clinically diagnosed as benign lesions, whereas in vivo reflectance confocal microscopy diagnosis was in agreement with the histologic diagnosis in all cases, including 10 melanomas, 5 SCC, and 2 MALT lymphomas. The 30 lesions that were not biopsied were diagnosed by in vivo reflectance confocal microscopy as 20 nevi (including 7 nevi with epithelial-cystic inclusions), 2 nevus of Ota (Figure 5), 2 pterygium, and 6 PAM without atypia (Figure 6), and showed no changes following consecutive clinical and in vivo reflectance confocal microscopy monitoring for at least 12 months. No complications such as ocular inflammation and/or infections and/or mechanical trauma were found using this imaging technique.
DISCUSSION OUR STUDY SHOWED THAT HANDHELD IN VIVO REFLEC-
tance confocal microscope dedicated to skin can be a new tool for a noninvasive diagnosis of conjunctival tumors. Two in vivo reflectance confocal microscopes are currently available to explore the ocular surface: the 326
Confoscan 4 slit-scanning confocal microscope (Nidek, Gamagori, Japan) and the in vivo reflectance confocal microscope Heidelberg Retina Tomograph (HRT),18,19 in association with the Rostock Cornea Module (Heidelberg Engineering GmbH, Heidelberg, Germany). Differently from the in vivo reflectance confocal microscope dedicated to skin, these devices do not have a handheld probe but present a hinged support for the camera that renders more difficult the precise alignment with the region of interest. Because of this limitation in the handiness, they are mainly used to visualize cornea and, more rarely, the conjunctiva near the axis.20 There are only few studies2,5–9 about their use for conjunctival tumors. One study5 evaluated 28 melanocytic tumors, 1 article described a case of lymphoma,9 and all the other publications are about squamous neoplasia.2,6–8,21 For the first time we used the handheld in vivo reflectance confocal microscope dedicated to skin (VivaScope 3000) to explore the conjunctiva. This device, thanks to its ease of handling (handheld probe of small size, light weight, equipped with manual control buttons to move in the vertical axis of the tissue) and its small tip, has been recently applied to the imaging of the mucosa10,12,13 and was suitable for the examination of the whole ocular area. VivaScope 3000 is expensive, but the same device can be shared between dermatologists and ophthalmologists, as we demonstrated that it can be used for both skin and eye. Most of the largest dermatologic centers already have this device. Examination is best performed on a cooperative patient who should not excessively move his eyes, while the operator continuously adapts the position of the camera. During our study we also explored the normal conjunctiva that presented with similar features to those described for the Heidelberg device.22–24
AMERICAN JOURNAL OF OPHTHALMOLOGY
FEBRUARY 2015
VOL. 159, NO. 2
TABLE. Clinical Features and Handheld Reflectance Confocal Microscopy and Histologic Diagnosis of the Conjunctival Lesions
Patient Number Sex Age
HANDHELD CONFOCAL MICROSCOPY FOR CONJUNCTIVAL TUMORS
Location
Size (mm)
1 2 3 4 4 5 5 5 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
M F F F F M M M M F M F M M F F M M F F M M F M M F M F M F
78 73 94 42 42 64 64 64 64 66 60 19 59 81 55 47 73 77 59 69 71 73 64 51 56 71 17 23 58 33
Bulbar Bulbar/tarsal Bulbar Bulbar Tarsal Bulbar Tarsal Bulbar Tarsal Limbal Caruncle Limbal Bulbar Limbal Limbal/bulbar Limbal Limbal Limbal Limbal Limbal Limbal Tarsal Tarsal Bulbar Limbal Caruncle Limbal/bulbar Caruncle Bulbar Bulbar
232 533 433 332 432 333 334 2 336 638 433 131 10 3 6 332 9 3 10 535 535 535 333 433 334 835 433 332 131 232 835 331 736 434
27 28 29 30 31
F F F M M
28 32 84 61 31
Bulbar Bulbar Bulbar Caruncle Limbal
Clinical Feature
Dark brown flat papule Multiple macules, light brown Multiple macules, dark brown Light brown macule Light brown macule Red macule Brown macule Brown macule Brown macule Black flat papule Pink papule Brown macule Brown macule Flat rose papule Red macule Pink macule Reddish papule Pink nodule Whitish nodule Whitish nodule Pink papule Pink patch Pink papules Brown macule Light brown macule Light brown papule Brown-yellow patch Brown macule Flat brown papule Macule with a light and dark brown pigmentation 5 3 5 Brown patch 1.5 Brown papule 2 3 4 Light brown macule 15 3 7 Brown macule Brown macule
Evolution (Years)
0.5 10 10 2 2 9 months 9 months 3 months 10 months 7 NA 1 10 0.5 1 NA 1 month 0.5 10 2 1 1 month 0.5 3 months Congenital 5 Congenital NA 48 Congenital 10 2 1 45 5
Clinical Diagnosis
Melanoma? Melanoma Melanoma Melanoma Melanoma Pinguecula? Nevus? Melanoma? Melanoma Nevus? Nevus Nevus Nevus SCC? SCC SCC SCC SCC? SCC? SCC? SCC? MALT lymphoma? MALT lymphoma? Melanoma? Nevus Nevus Nevus Nevus Nevus Nevus
Confocal Microscopy Diagnosis
Histologic Diagnosis
Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Compound nevus Subepithelial nevus Epithelial-cystic nevus SCC SCC SCC SCC SCC Pinguecula Pterygium Pinguecula MALT lymphoma MALT lymphoma Junctional nevus Dermal nevus Subepithelial nevus Compound nevus Junctional nevus Epithelial-cystic nevus Junctional nevus
Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Melanoma Compound nevus Subepithelial nevus Epithelial-cystic nevus SCC in situ SCC in situ SCC SCC SCC Pinguecula Pinguecula Pinguecula MALT lymphoma MALT lymphoma NA NA NA NA NA NA NA
Epithelial-cystic nevus Epithelial-cystic nevus NA Nevus Epithelial-cystic nevus NA Nevus Epithelial-cystic nevus NA Nevus Epithelial-cystic nevus NA Nevus Epithelial-cystic nevus NA
Confocal Microscopy/Histology Correlation
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes NA NA NA NA NA NA NA NA NA NA NA NA
327
Continued on next page
328
TABLE. Clinical Features and Handheld Reflectance Confocal Microscopy and Histologic Diagnosis of the Conjunctival Lesions (Continued )
Patient Number Sex Age
Location
Size (mm)
AMERICAN JOURNAL OF OPHTHALMOLOGY
32 33 34 35 35 35 36
M M M M M M F
Limbal Bulbar Bulbar Bulbar Bulbar Bulbar Bulbar
735 533 332 3 2 2 634
37
M 27 Caruncule, conjunctive F 59 Bulbar M 48 Bulbar F 48 Bulbar F 62 Limbal F 66 Tarsal F 33 Limbal F 40 Bulbar F 29 Bulbar M 72 Limbal F 68 Limbal
1
15 38 39 40 41 42 43 44 45 46
17 68 51 41 41 41 13
131 334 231 5 4 10 334 532 232
Clinical Diagnosis
10 10 10 10 10 10 Congenital, but recently enlarged 3
Nevus Epithelial-cystic nevus Epithelial-cystic nevus Racial melanosis Racial melanosis Racial melanosis Nevus
Subepithelial nevus Epithelial-cystic nevus Epithelial-cystic nevus Compound nevus Compound nevus Compound nevus Subepithelial nevus
NA NA NA NA NA NA NA
NA NA NA NA NA NA NA
Nevus
Subepithelial nevus
NA
NA
NA Congenital
Nevus of Ota Nevus of Ota PAM PAM PAM PAM Racial melanosis Racial melanosis Pterygium vs SCC? Pterygium vs SCC?
Nevus of Ota Nevus of Ota PAM PAM PAM PAM PAM PAM Pterygium Pterygium
NA NA NA NA NA NA NA NA NA NA
NA NA No NA NA NA NA NA NA NA
Clinical Feature
Erythematous macule Flat light brown papule Brown papule Light brown macule Light brown macule Light brown macule Erythematous macule Brown macule Brown and gray macules Brown and gray macules Multifocal light brown macule Light brown macule Light brown macule Light brown macule Brown macule Light brown macule Grayish nodule Red nodule
Confocal Microscopy/Histology Correlation
Confocal Microscopy Diagnosis
Evolution (Years)
10 10 0.5 10 6 4 NA 10
MALT ¼ mucosa-associated lymphoid tissue; NA ¼ not available; PAM ¼ primary acquired melanosis; SCC ¼ squamous cell carcinoma.
Histologic Diagnosis
FEBRUARY 2015
FIGURE 3. Clinical (Left), histologic (Center) and handheld reflectance confocal microscopy (Right) aspect of a subepithelial nevus (First line), an epithelial-cystic nevus (Second line), and a melanoma (Third line). Handheld reflectance confocal microscopy shows in the subepithelial nevus homogeneous nests of cells (red asterisk) in the stroma; in the epithelial-cystic nevus stromal pseudocystlike structures (red asterisk) partly filled with monomorphous material surrounded by hyperreflective homogeneous cells (red circles); in the melanoma uneven large cells (red arrow) of different shape (roundish, dendritic and polygonal cells) in the epithelium. (Histologic stains: hematoxylin-eosin, first and third lines 203, second line 43.)
Given the lack of in vivo reflectance confocal microscopy validated criteria to diagnose tumors with the handheld VivaScope 3000 camera, in particular in the conjunctival area, we defined new criteria based on those used in the 2 studies of conjunctival tumors explored by the Heildelberg in vivo reflectance confocal microscope,9,22 on in vivo reflectance confocal microscopy diagnostic criteria used for skin tumors15,17 and for genital and oral mucosal10,13 tumors, on histology of conjunctival tumors,4,25 and on our personal experience. Using our diagnostic criteria, in vivo reflectance confocal microscopy diagnosis was in agreement with histology in all cases, and in particular for SCC, melanoma, and MALT lymphoma. In vivo reflectance confocal microscopy was mostly used to confirm the clinical diagnosis of malignant tumors, but in 5 cases of SCC, 4 cases of melanoma, and 2 cases of lymphoma the clinical diagnosis was completely uncertain. Interestingly, it also found VOL. 159, NO. 2
melanomas in 3 lesions that were not clinically suspicious for malignant tumors. In all SCC we found large epithelial cells (Figure 4) and an abundant infiltrate of dendritic cells. Enlarged epithelial cells corresponding to neoplastic epitheliocytes could also be found by Falke and associates2 in their in vivo reflectance confocal microscopy observation on a case of in situ SCC. Nguena and associates21 recently reported a series of 26 SCC observed by in vivo reflectance confocal microscopy in a black population and could not find a statistically significant difference for the parameters that have been so far identified as characteristic of SCC (presence of irregular cell size, hyperreflectivity, prominent bright nucleoli, and mitotic figures) compared to 18 benign conjunctival lesions. Based on our experience we agree that hyperreflectivity, prominent bright nucleoli, and mitotic figures are not in vivo reflectance confocal microscopy valid criteria for SCC. Irregular cell size is a difficult criterion to be evaluated
HANDHELD CONFOCAL MICROSCOPY FOR CONJUNCTIVAL TUMORS
329
FIGURE 4. Clinical and handheld reflectance confocal microscopy appearance of a mucosa-associated lymphoid tissue lymphoma (First line) of the superior (Left) and inferior (Right) tarsal conjunctiva; clinical (Left), histologic (Center), and reflectance confocal microscopy (Right) appearance of a pinguecula (Second line); and a squamous cell carcinoma (Third line). Reflectance confocal microscopy shows in the lymphoma (First line) widespread small hyperrefractive roundish cells corresponding to tumoral lymphocytes in the stroma (red arrow and circle); in the pinguecula (Second line) degenerated stromal collagen with a coiled shape (yellow circle) that corresponds to homogenized areas of collagen in the histologic examination (black asterisk); in the squamous cell carcinoma (Third line) a disarranged pattern of the epithelium (red circle) with some large epithelial cells (red arrow) and abundant dendritic cells (yellow arrow). (Histologic stains: hematoxylin-eosin, second line 103, third line 403.)
because epitheliocytes are not completely regular in form and size in physiologic conditions; moreover, the extent of irregularity that is considered pathologic should be defined. For these reasons in our series we only look for the presence of large epithelial cells (>20 mm). In case of advanced SCC, the irregularity of the cells is more evident and a completely disarranged pattern can be observed, corresponding to epitheliocytes that are completely distorted in their shape and organization. However, we should notice that pterygium can sometimes present dysplastic and pleomorphic cells at histology.26,27 Therefore, although in our small series epithelial cells of pterygium were not large in size and pleomorphic, and pterygium could be differentiated from SCC, large cells 330
could be eventually found in these lesions, and this aspect could point out lesions with epithelial atypia that could be associated with ocular surface squamous neoplasia27 and need surgical excision. We should also notice that in our study in vivo reflectance confocal microscopy images were interpreted in real time during the examination, whereas in the study of Nguena and associates21 in vivo reflectance confocal microscopy images were read retrospectively masked to the clinical appearance. In all cases of melanoma we found enlarged hyperreflective cells, roundish or dendritic, in the epithelium and/or the stroma, corresponding to malignant melanocytes (Figure 3). These cells were uneven in size and sometimes enlarged nuclei were visible as central hyporeflective areas.
AMERICAN JOURNAL OF OPHTHALMOLOGY
FEBRUARY 2015
FIGURE 5. Clinical (Left) and handheld reflectance confocal microscopy (Center and Right) appearance of a nevus of Ota. Handheld reflectance confocal microscopy of the conjunctival part of the nevus showed large homogeneous dendritic bright cells (red arrow) between the collagen fibers.
FIGURE 6. Clinical (Left) and handheld reflectance confocal microscopy (Right) aspect of primary acquired melanosis without atypia. Handheld reflectance confocal microscopy shows homogeneous hyperreflective cells confined to the basal layer of the epithelium.
Messmer and associates5 have already demonstrated in a study on 28 lesions that in vivo reflectance confocal microscopy had a high sensitivity and specificity for the melanocytic lesions of the conjunctiva, finding a sensitivity of 89% and a specificity of 100% for melanoma. They misdiagnosed only 1 lesion over 28, diagnosing PAM with atypia in a challenging lesion that histology classified as melanoma/PAM with atypia. With our study we confirmed the diagnostic ability of in vivo reflectance confocal microscopy for melanocytic lesions and we confirmed that in vivo reflectance confocal microscopy can also identify melanocytes as hyperreflective cells in amelanotic melanocytic lesions (3 in our series, 1 melanoma and 2 nevi). Concerning the 6 patients with conjunctival melanomas, 3 had a previous history of ocular melanoma (2 uveal, 1 of the eyelid), and 4 had a multifocal conjunctival involvement, indicating that in vivo reflectance confocal microscopy can be a useful imaging technique to monitor patients with a history of melanoma and multifocal lesions. VOL. 159, NO. 2
The in vivo reflectance confocal microscopy aspect of MALT lymphoma has already been reported by Pichierri and associates,9 consisting of highly reflective small-size roundish cells arranged diffusely or in nests at epithelial and subepithelial conjunctival levels. In our 2 cases we found the same kind of cells in the stroma (Figure 4), but in vivo reflectance confocal microscopy could not identify cytologic atypia in order to distinguish tumoral from inflammatory cells. Therefore, in vivo reflectance confocal microscopy could point out the necessity of a histologic examination in case of an evocative aspect of lymphoma, but the final diagnosis should be left to histology. Considering the excellent diagnostic ability with the use of in vivo reflectance confocal microscopy for conjunctival malignant tumors and the fact that conjunctival surgery could be disturbing for the patient, we decided not to excise 30 lesions that were considered benign by in vivo reflectance confocal microscopy. These lesions were diagnosed as nevus in 20 cases, a nevus of Ota in 2 cases, pterygium
HANDHELD CONFOCAL MICROSCOPY FOR CONJUNCTIVAL TUMORS
331
in 2 cases, and PAM without atypia in 6 cases. The followup during at least 12 months of all these lesions did not show any clinical progression. In the 2 cases of nevus of Ota the exploration of the entire nevus was possible, differently from what is done performing focal biopsies of clinically suspicious areas. The aspect of nevus of Ota (Figure 5) has never been described using in vivo reflectance confocal microscopy, and we found large homogeneous dendritic bright cells between the collagen fibers that could be either isolated or grouped in clusters and could also be found in apparently uninvolved conjunctiva. The absence of malignant transformation was assessed by the absence of polymorphic hyperreflective cells, uneven in size and shape, suggestive of malignant melanocytes. In conclusion, our pilot study strongly suggests that handheld in vivo reflectance confocal microscopy can be efficiently employed for a noninvasive and rapid investigation of the conjunctiva, including melanoma, SCC, and MALT lymphoma. To date, the diagnosis of conjunctival lesions has been based on the slit-lamp appearance and surgery was performed in all suspicious cases. In our series most
lesions could be accurately diagnosed clinically by their slit-lamp appearance, but some were either under- or over-diagnosed. Handheld in vivo reflectance confocal microscopy is an additional tool that can add information to the clinical examination in order to proceed with more evidence to the surgical excision. Although in vivo reflectance confocal microscopy does not replace histologic examination for definitive diagnosis, it can avoid unnecessary surgery in patients who cannot undergo surgery, such as elderly or debilitated patients. Moreover, in cases of large or multifocal lesions it allows exploration of an entire lesion differently from a biopsy. Larger multicenter studies should now be performed to validate our data and better define the diagnostic capability of handheld in vivo reflectance confocal microscopy. In particular, diagnosis performances should be compared with ophthalmologic in vivo reflectance confocal microscopy, and management using this advanced handheld in vivo reflectance confocal microscopy should be compared to the conventional management using slitlamp examination and biopsies.
ALL AUTHORS HAVE COMPLETED AND SUBMITTED THE ICMJE FORM FOR DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST and the following were reported. G.T. and P.G. are consultants for The´a Laboratories and Quantel Medical. J.-L.P., B.L., and F.C. are consultants for Leo Pharma, Pierre Fabre, LVMH, MSD, Amgen, Pfizer, and Roche Laboratories. Funding was provided through the Groupement Interregional pour la recherche Clinique Rhoˆne-Alpes Auvergne (GIRCI RAA) (Lyon, France) 2012 Young Investigator award. Contributions of authors: design and conduct of the study (E.C., J.-L.P., B.L., N.C., M.E., D.G., G.T., P.G., C.D., C.A., F.C.); collection of data, management (E.C., J.-L.P., B.L., N.C., M.E., D.G., G.T., P.G., C.D., C.A., M.H.); analysis and interpretation of the data (E.C., J.-L.P., B.L.); preparation of the manuscript (E.C.); review and approval of the manuscript (E.C., J.-L.P., B.L., N.C., M.E., D.G., G.T., P.G., C.D., C.A., M.H., F.C.).
REFERENCES 1. Maschi C, Caujolle J-P, Liolios I, Costet C. Benign conjunctival tumors. J Fr Ophtalmol 2013;36(9):796–802. 2. Falke K, Zhivov A, Zimpfer A, Stachs O, Guthoff RF. Diagnosis of conjunctival neoplastic lesions by confocal in-vivo microscopy. Klin Monatsbla¨tter Fu¨r Augenheilkd 2012; 229(7):724–727. 3. Folberg R, Jakobiec FA, Bernardino VB, Iwamoto T. Benign conjunctival melanocytic lesions. Clinicopathologic features. Ophthalmology 1989;96(4):436–461. 4. Ranty M-L, Quintyn J-C, Uro-Coste E, Delisle M-B. Ocular conjunctival pathology. A ten-year retrospective study in Toulouse-Rangueil University Hospital and literature review. Ann Pathol 2012;32(3):170–176. 5. Messmer EM, Mackert MJ, Zapp DM, Kampik A. In vivo confocal microscopy of pigmented conjunctival tumors. Graefes Arch Clin Exp Ophthalmol 2006;244(11):1437–1445. 6. Duchateau N, Hugol D, D’Hermies F, et al. Contribution of in vivo confocal microscopy to limbal tumor evaluation. J Fr Ophtalmol 2005;28(8):810–816. 7. Xu Y, Zhou Z, Xu Y, et al. The clinical value of in vivo confocal microscopy for diagnosis of ocular surface squamous neoplasia. Eye (Lond) 2012;26(6):781–787. 8. Parrozzani R, Lazzarini D, Dario A, Midena E. In vivo confocal microscopy of ocular surface squamous neoplasia. Eye (Lond) 2011;25(4):455–460.
332
9. Pichierri P, Martone G, Loffredo A, Traversi C, Polito E. In vivo confocal microscopy in a patient with conjunctival lymphoma. Clin Experiment Ophthalmol 2008;36(1):67–69. 10. Cinotti E, Perrot JL, Labeille B, Adegbidi H, Cambazard F. Reflectance confocal microscopy for the diagnosis of vulvar melanoma and melanosis: preliminary results. Dermatol Surg 2012;38(12):1962–1967. 11. Cinotti E, Chol C, Perrot JL, Labeille B, Forest F, Cambazard F. Anal melanosis diagnosed by reflectance confocal microscopy. Australas J Dermatol 2013; http://dx. doi.org/10.1111/ajd.12096. 12. Cinotti E, Perrot JL, Labeille B, Douchet C, Mottet N, Cambazard F. Laser photodynamic treatment for in situ squamous cell carcinoma of the glans monitored by reflectance confocal microscopy. Australas J Dermatol 2014; 55(1):72–74. 13. Debarbieux S, Perrot JL, Erfan N, et al. Reflectance confocal microscopy of mucosal pigmented macules: a review of 56 cases including 10 macular melanoma. Br J Dermatol 2014; 170(6):1276–1284. 14. Erfan N, Hofman V, Desruelles F, et al. Labial melanotic macule: a potential pitfall on reflectance confocal microscopy. Dermatol Basel Switz 2012;224(3):209–211. 15. Guitera P, Menzies SW, Longo C, Cesinaro AM, Scolyer RA, Pellacani G. In vivo confocal microscopy for diagnosis of melanoma and basal cell carcinoma using a two-step method: analysis of 710 consecutive
AMERICAN JOURNAL OF OPHTHALMOLOGY
FEBRUARY 2015
16.
17.
18.
19.
20.
clinically equivocal cases. J Invest Dermatol 2012;132(10): 2386–2394. Langley RGB, Burton E, Walsh N, Propperova I, Murray SJ. In vivo confocal scanning laser microscopy of benign lentigines: comparison to conventional histology and in vivo characteristics of lentigo maligna. J Am Acad Dermatol 2006;55(1):88–97. Rishpon A, Kim N, Scope A, et al. Reflectance confocal microscopy criteria for squamous cell carcinomas and actinic keratoses. Arch Dermatol 2009;145(7):766–772. Niederer RL, McGhee CNJ. Clinical in vivo confocal microscopy of the human cornea in health and disease. Prog Retin Eye Res 2010;29(1):30–58. Wang Y, Le Q, Zhao F, Hong J, et al. Application of in vivo laser scanning confocal microscopy for evaluation of ocular surface diseases: lessons learned from pterygium, meibomian gland disease, and chemical burns. Cornea 2011;30(Suppl 1):S25–S28. Zhivov A, Stachs O, Kraak R, Stave J, Guthoff RF. In vivo confocal microscopy of the ocular surface. Ocul Surf 2006; 4(2):81–93.
VOL. 159, NO. 2
21. Nguena MB, van den Tweel JG, Makupa W, et al. Diagnosing ocular surface squamous neoplasia in East Africa: case-control study of clinical and in vivo confocal microscopy assessment. Ophthalmology 2014;121(2):484–491. 22. Messmer EM, Mackert MJ, Zapp DM, Kampik A. In vivo confocal microscopy of normal conjunctiva and conjunctivitis. Cornea 2006;25(7):781–788. 23. Efron N, Al-Dossari M, Pritchard N. In vivo confocal microscopy of the palpebral conjunctiva and tarsal plate. Optom Vis Sci 2009;86(11):E1303–E1308. 24. Efron N, Al-Dossari M, Pritchard N. In vivo confocal microscopy of the bulbar conjunctiva. Clin Experiment Ophthalmol 2009;37(4):335–344. 25. Shields CL, Shields JA. Tumors of the conjunctiva and cornea. Surv Ophthalmol 2004;49(1):3–24. 26. Golu T, Mogoanta L, Streba CT, et al. Pterygium: histological and immunohistochemical aspects. Romanian J Morphol Embryol Rev 2011;52(1):153–158. 27. Chui J, Coroneo MT, Tat LT, Crouch R, Wakefield D, Di Girolamo N. Ophthalmic pterygium. Am J Pathol 2011; 178(2):817–827.
HANDHELD CONFOCAL MICROSCOPY FOR CONJUNCTIVAL TUMORS
333
Biosketch Elisa Cinotti, MD, graduated from University of Genoa (Italy) cum laude in 2007. She did her Dermatology residency at the University of Genoa. Since 2012 she is working in the University Hospital of Saint-Etienne (France) where she is conducting several researches on skin and eye noninvasive imaging, in particular on Reflectance Confocal Microscopy in cooperation with ophthalmologists. She has published over 40 peer-reviewed articles, most of them about confocal microscopy.
333.e1
AMERICAN JOURNAL OF OPHTHALMOLOGY
FEBRUARY 2015
