The Veterinary Journal 200 (2014) 440–445
Contents lists available at ScienceDirect
The Veterinary Journal j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / t v j l
Canine aural cholesteatoma: A histological and immunohistochemical study Barbara Banco a, Valeria Grieco a, Mauro Di Giancamillo b, Valentina Greci a, Olga Travetti b, Pieranna Martino a, Carlo M. Mortellaro a, Chiara Giudice a,* a
Dipartimento di Scienze Veterinarie e Sanità Pubblica, Università degli Studi di Milano, via Celoria 10, 20133 Milan, Italy Dipartimento di Scienze Veterinarie per la Salute, la Produzione Animale e la Sicurezza Alimentare, Università degli Studi di Milano, via Celoria 10, 20133 Milan, Italy
b
A R T I C L E
I N F O
Article history: Accepted 19 March 2014 Keywords: Cholesteatoma Cytokeratins Dog Ear Epidermoid Ki67
A B S T R A C T
Canine aural cholesteatoma is an epidermoid cyst that forms in the middle ear cavity as a rare complication of otitis media but the aetiopathogenesis remains controversial. In the present study, 13 cases of canine aural cholesteatoma were investigated histologically and immunohistochemically and compared with cases of chronic otitis. The immunohistochemical investigation was performed using the following monoclonal antibodies: anti-cytokeratins (CK) 14, 16, 8/18, and 19, and anti-Ki67. The proliferative indexes (PIs) of cholesteatomata and otitis epithelium were calculated as the percentage of Ki67 positive nuclei/ total nuclei. Histologically, the cholesteatomata were composed of a hyperplastic, hyperkeratotic epithelium (matrix) resting on a fibrous perimatrix, infiltrated by inflammatory cells and devoid of cutaneous adnexa. Immunohistochemically, the cholesteatoma epithelium was CK14- and CK16-positive, and CK8/ 18- and CK19-negative. A similar pattern of CK expression was found in otitis externa. In otitis media, ciliated epithelium stained CK8/18- and CK19-positive in all layers, CK14-positive in the basal layers, and CK16-negative. The mean PIs in cholesteatomata and otitides were 18.8 and 17.8, respectively. The immunohistochemical pattern of CK expression in cholesteatomata, when compared with chronic otitis, was suggestive of hyperproliferative epithelium, but its origin could not be demonstrated. Comparable PI values were obtained in cholesteatoma and in chronic otitis, which confirmed that Ki67 is a valuable indicator of a hyperproliferative state, but not a predictor of aggressiveness. © 2014 Elsevier Ltd. All rights reserved.
Introduction Aural cholesteatoma is an epidermoid cyst that forms in the middle ear and is composed of hyperplastic, hyperkeratotic squamous epithelium surrounding a core of keratin debris. It has been described as ‘skin growing in the wrong place’ (Strunk, 1993) and the term ‘cholesteatoma’ itself, coined by Müller in 1838, is currently considered a misnomer because it is neither a granulomatous lesion or a neoplasm (Friedmann and Arnold, 1993; Strunk, 1993). In humans, aural cholesteatoma is classified as congenital or acquired (Persaud et al., 2007). Congenital cholesteatoma develops beneath an intact tympanic membrane in patients whose clinical history does not include an obvious ear infection or auditory tube dysfunction (Strunk, 1993). It is hypothesised to originate from embryonic epidermoid remnants that fail to involute (Koltai et al., 2002). Acquired cholesteatoma is a complication of chronic otitis and it is
* Corresponding author. Tel.: +39 0250 318105. E-mail address: [email protected] (C. Giudice). http://dx.doi.org/10.1016/j.tvjl.2014.03.018 1090-0233/© 2014 Elsevier Ltd. All rights reserved.
thought to originate either as a result of squamous metaplasia of the middle ear epithelium or from the migration of external ear canal epithelium through a tympanic perforation (Sadé, 1980; Semaan and Megerian, 2006). Other workers have suggested that it could originate from an invagination of a portion of the tympanic membrane into the epitympanum, secondary to a Eustachian tube dysfunction (retraction pocket theory; Sadé, 1980; Olszewska et al., 2003). To date, the origin of the epithelium of cholesteatoma remains uncertain. In dogs, cholesteatoma is mostly considered an acquired disease and a severe complication of otitis media, with or without concurrent rupture of the tympanic membrane (Little et al., 1991a, 1991b; Venker-Van Haagen, 2005; Harran et al., 2012). To our knowledge, congenital cholesteatoma has not been reported in dogs. In human medicine, several immunohistochemical studies have been performed in an attempt to identify the origin of cholesteatoma epithelium (Bujía et al., 1993; Kuijpers et al., 1996; Olszewska et al., 2005). More specifically, the expression of cytokeratins, the intermediate filaments of epithelia, has been extensively investigated, since the pattern of cytokeratin expression in a particular epithelium varies with its anatomical location, developmental stage,
B. Banco et al./The Veterinary Journal 200 (2014) 440–445
441
Table 1 Details of antibodies employed in immunohistochemistry (antigen retrieval, dilution and sources). Antibodies
Clone
Antigenic retrieval
Dilution
Source
CK14 CK16 CK8/18 CK19 Ki67
LL002 LL025 5D3 B170 MIB-1
Pressure cooker, 10 min, citrate buffer (pH 6) Pressure cooker, 10 min, citrate buffer (pH 6) Pepsina, 37 °C, 14 min Pepsina, 37 °C, 14 min Pressure cooker, 18 min, citrate buffer (pH 6)
1 : 2000 1 : 200 1 : 200 1 : 200 1 : 600
NeoMarkers Novocastra Zymed Novocastra Dakocytomation
a
Zymed.
and state of differentiation. However, conflicting results have been reported. To elucidate the mechanisms of the progressive growth that characterise cholesteatomata, numerous studies in human medicine focused on the proliferative capability of lining epithelium (Strunk, 1993; Gersdorff et al., 2006). Investigations used MIB 1 (Ki67) immunostaining, in an attempt to characterise the aggressiveness of the cholesteatoma and the likelihood of recurrence (Bujía et al., 1996; Mallet et al., 2003). To the best of our knowledge, detailed histological and immunohistochemical characterisations of canine middle ear cholesteatoma are not currently available in the veterinary literature. Therefore, the aim of the current study was to characterise canine aural cholesteatoma, comparing epithelial immunophenotypes (cytokeratins) and proliferative activity (Ki67) in cholesteatomata, in normal auditory canals, and in otitis externa and otitis media.
otitis) was observed on video-otoscopy in all cases. Eight cases had rupture of the tympanic membrane, and amorphous, pearly material in the middle ear cavity was evident in nine cases. The middle ear content was cultured in each of the 13 cases. Staphylococcus intermedius was isolated in three cases, Proteus mirabilis in two cases, E. coli in one case, Pseudomonas aeruginosa and P. mirabilis in one case and Pseudomonas aeruginosa and Klebsiella pneumoniae in another case. No bacteria were recovered in five cases. Histopathology Normal external auditory meatus The normal meatus was lined by a thin squamous epithelium, one to two cell layers thick, resting on a derma containing hair follicles, sebaceous and ceruminous glands, and supported by annular cartilage.
Materials and methods Thirteen aural cholesteatomata from 11 dogs were reviewed; two cases were recurrences. Specimens were obtained by total ear canal ablation and lateral bulla osteotomy (TECALBO) or ventral bulla osteotomy (VBO), fixed in 10% neutral buffered formalin, and submitted to the Department of Veterinary Science and Public Health (DIVET) of Milan for histopathology. In all cases, clinical, radiological (computed tomography scan, CT), and video-otoscopy findings were consistent with the diagnosis of aural cholesteatoma (Little et al., 1991b; Travetti et al., 2010; Greci et al., 2011). Signalment, clinical signs, and results of middle ear swabs for bacteriological culture were recorded for each dog. Follow-up consisted of a clinical examination. Ten histological specimens of chronic otitis externa or media from 10 dogs and 1 normal canine external meatus auditory canal specimen were retrieved from DIVET archives and were included in the study. All histological specimens were routinely processed, paraffin embedded, and stained with haematoxylin and eosin (HE). Serial sections were obtained and immunolabelled using the standard avidin–biotin–peroxidase complex (ABC) procedure (Hsu et al., 1981) and the following antibodies: anti-cytokeratin 8/18, 14, 16, and 19, and anti- Ki67 protein. Primary antibodies and antigen retrieval methods used are listed in Table 1. A section of normal dog skin served as a positive control. For negative controls, the primary antibody was replaced by serum from a clinically healthy horse. Cytokeratin immunostaining was quantitatively interpreted by light microscopy (positive and negative). When a positive result was obtained, immunostaining was scored as faint, moderate, or intense. The expression of Ki67 (MIB-1) was quantitatively assessed using an automatic image analysis system (Image Pro Plus 4.5, Media Cybernetics). The relative percentage of immunostained nuclei in 10 high power fields was calculated and recorded as a proliferative index (PI).
Results
Cholesteatoma Ten cases of aural cholesteatoma were composed of segments of the wall of a cystic lesion that contained abundant amorphous lamellar keratin debris. The cysts were lined by a multi-layered keratinising squamous epithelium (matrix) resting on a stroma (perimatrix) composed of dense fibrovascular connective tissue devoid of adnexa (Fig. 1). The epithelium was intensely hyperplastic (5–25 layers thick), with severe orthokeratotic hyperkeratosis, hypergranulosis and mild parakeratosis. Keratinocytes showed progressive maturation; no mitotic figures were observed. Mild multifocal intracellular oedema and minimal neutrophilic exocytosis were present. The perimatrix immediately beneath the epithelium was composed of loosely arranged collagen fibres expanded by moderate oedema; the deepest layers of the perimatrix were densely cellular, composed of fibroblasts and fibrocytes embedded in abundant collagenous stroma. There was moderate to severe stromal inflammation, consisting of numerous perivascular lymphocytes and plasma cells, scattered haemosiderin-laden macrophages and a few
Table 2 Signalment data of dogs with cholesteatoma and mean values of proliferative index (PI) of the epithelium as obtained by immunolabelling with monoclonal antibody MIB1.
Signalment and clinical features of dogs with aural cholesteatoma
Case no.
Signalment and follow-up data for dogs with cholesteatoma are summarised in Table 2. Thirteen cases of aural cholesteatoma were diagnosed in 11 dogs. Two cases (numbers 3 and 10) were recurrences. The mean age at diagnosis was 7 years (range, 5–10 years); eight dogs were males (one neutered) and there were three females (one spayed). A variety of breeds were represented. All dogs had a history of chronic recurrent otitis externa that was unresponsive to topical or systemic therapy over the previous 3–30 weeks. The main clinical signs recorded were head shaking, otodynia, and head tilt. The main findings on computed tomography (CT) were severe expansion and osteolysis of the tympanic bulla with loss of air contrast. Partial or total occlusion of the horizontal canal (end-stage
1 2 3 4 5 6 7 8 9 10 11 12 13 a
Signalment
PI
Pug, male, 8 years Mixed breed, male, 5 years Mixed breed, male, 6 years Flat coated retriever, male, 10 years Poodle, castrated male, 5 years Afghan hound, male, 8 years Weimaraner, female, 9 years American cocker spaniel, male, 6 years Schnauzer , male, 5 years Poodle, castrated male, 5 years Mixed breed, spayed female, 9 years Labrador, male, 9 years Golden retriever, female, 5 years
–a 12.45 14.12 20.51 –a 17.35 8.42 –a 31.79 15.90 34.81 15.74 16.93
Recurrence
Recurrence of case 2
Recurrence of case 5
Specimen could not be processed as it constituted only keratin debris.
442
B. Banco et al./The Veterinary Journal 200 (2014) 440–445
Fig. 1. Canine middle ear, case 3 (Table 2). Section of the cholesteatoma wall composed of an intensely hyperplastic and hyperkeratotic keratinising squamous epithelium (matrix) lining a cystic cavity filled with keratin debris and resting on a dense fibrous stroma (HE). Bar 50 μm.
neutrophils. Numerous small calibre hyperaemic vessels, often surrounded by microhaemorrhages and oedema, were also present. In four cases (numbers 7, 11, 12, and 13), multiple cholesterol granulomas were also found expanding the perimatrix. Granulomas were composed of epithelioid macrophages, multinucleated foreign-body giant cells, and scattered lymphocytes, concentrated around acicular cholesterol crystals. In two cases (numbers 9 and 12) a small segment of monolayered columnar ciliated respiratory epithelium (normal middle ear epithelium) was identified. In three cases (numbers 1, 5, and 8), the specimens surgically obtained from the middle ear were composed exclusively of amorphous keratin debris. The presence of keratin debris alone, as flakes in an expansive lesion of the middle ear, is highly suggestive of cholesteatoma (Ferlito et al., 1997). However, these three specimens were excluded from further immunohistochemical investigation. Otitides Ten cases of canine chronic otitis externa were retrieved from the archives. Of these, three cases had associated otitis media. The main signalment data are summarised in Table 3. Histologically, chronic otitis externa was characterised by severe hyperplasia of the epidermis (6–15 layers thick) with moderate to severe orthokeratotic hyperkeratosis, hypergranulosis, mild intracellular oedema, and spongiosis (Fig. 2). Scattered intraepithelial pustules and multifocal ulcerations covered with serocellular crusts, occasionally containing
Table 3 Signalment data of dogs with chronic otitis and mean values of proliferative index (PI) of the epithelium, as obtained by immunolabelling with monoclonal antibody MIB1. Case no. 1 2 3 4 5 6 7 8 9 10
Signalment
PI
German shepherd, male, 4 years Mixed breed, female, 13 years Maltese terrier, female, 3 years German shepherd, female, 7 years Cocker spaniel, male, 8 years Yorkshire terrier, female, 3 years Schnauzer, male, 6 years English bulldog, male 9 years German shepherd, male, 4 years Mixed breed, male, 10 years
24.55 23.35 17.23 9.98 9.70 13.53 22.98 25.91 12.89 18.37
Fig. 2. Canine auditory canal, case 8 (Table 3). Severe otitis externa with intense hyperplasia and hyperkeratosis of the epidermis, fibrosis, and intense dermal lymphoplasmacytic infiltrate (HE). Bar 200 μm. Inset: section of a normal auditory canal; lining epithelium is composed of one to two layers of cells. Sebaceous and apocrine glands are present in the derma (HE). Bar 200 μm.
bacterial aggregates, were also common. The derma was characterised by severe fibrosis, moderate numbers of perivascular aggregates of lymphocytes, plasma cells and haemosiderinladen macrophages, dilation of ceruminous glands with stasis of secretum, and numerous hyperaemic vessels with perivascular oedema. Otitis media, associated with otitis externa in three cases (numbers 6, 7 and 9), was characterised by multifocal areas of ulceration and erosion of the respiratory epithelium. When recognisable, columnar epithelial cells were either shrunken and detached, or swollen with loss of cilia. Rarely, areas lined by squamous epithelium with neutrophilic exocytosis were also evident. The underlying corion was fibrotic, hyperaemic, infiltrated by numerous lymphocytes and plasma cells, with fewer neutrophils and haemosiderin-laden macrophages. In one case, a cholesterol granuloma was also present. Immunohistochemistry The immunohistochemical labelling results are summarised in Table 4. Epithelium lining the normal external auditory meatus was moderately, diffusely CK14-positive and CK16-, CK8/18-, and CK19negative. When cornified epithelial cells were recognisable, they were mostly CK14-negative. The epithelium of ceruminous glands was immunostained with CK8/18 and CK19. Only scattered epithelial cells of the external canal epithelium stained Ki67-positive (PI, 1.4). The cholesteatoma epithelium (matrix) was in all cases CK14and CK16-positive and CK8/18- and CK19-negative. CK14 staining was intense in keratinocytes throughout all matrix layers (Fig. 3). CK16 staining was also intense, but limited to the suprabasal layers of the epithelium (Fig. 4). The PI in the cholesteatoma epithelium varied from 8.4 to 34.8 (mean, 18.8; median, 16.4; standard deviation, SD 7.8; Table 2) among cases. Positive nuclei were mainly detected in epithelial cells of the basal layer with only rare, scattered positive nuclei within the suprabasal layers. The epithelium in otitis externa was strongly CK14- and CK16positive and CK8/18- and CK19-negative in all cases. CK14 labelling was diffuse in all layers of the epithelium (Fig. 5), while CK16 immunolabelling was limited to the suprabasal layers (Fig. 6). In the
B. Banco et al./The Veterinary Journal 200 (2014) 440–445
443
Table 4 Summary of immunohistochemical results of cytokeratin (CK) immunolabelling in normal ear, cholesteatoma and otitides. Normal aural meatus
CK14 CK16 CK8/18 CK19 a b c
Aural cholesteatoma
Otitis externa
Otitis media
Suprabasal layers
Basal layers
Suprabasal layers
Basal layers
Suprabasal layers
Basal layers
Ciliated epithelium
Squamous epithelium
− − − −
+ − − −
+ + − −
+ − − −
+ + − −
+ − − −
+ − + +
+b +c − −
a
Ciliated epithelium in otitis media: CK14 immunolabelling was limited to basal epithelial layer. Squamous epithelium in otitis media: CK14 immunolabelling was diffuse to basal and suprabasal epithelial layers. Squamous epithelium in otitis media: CK16 immunolabelling was limited to subrabasal epithelial layers.
few cases of otitis media, columnar epithelium was diffusely CK8/ 18 and CK19-positive, while CK14 stained only basal layers and CK16 was always negative. Squamous epithelium was diffusely CK14positive, while CK16 staining was limited to the suprabasal layers
Fig. 3. Canine middle ear, case 4 (Table 2). Cholesteatoma: lining epithelium immunostained intensely and diffusely cytokeratin 14 positive. Immunohistochemistry: avidin–biotin–peroxidase complex method, 3-amino-9-ethylcarbazole chromogen, Mayer’s haematoxylin counterstain. Bar 50 μm.
Fig. 4. Canine middle ear, case 3 (Table 2). Cholesteatoma: suprabasal epithelial layers were intensely cytokeratin 16-positive. Immunohistochemistry: avidin–biotin– peroxidase complex method, 3-amino-9-ethylcarbazole chromogen, Mayer’s haematoxylin counterstain. Bar 50 μm.
and CK8/18 and CK19 were always negative. The PIs ranged from 9.7 to 25.9 (mean, 17.8; median, 17.8; SD 5.8; Table 3). Positive nuclei were observed mainly in epithelial cells of the basal layer, with only occasional, scattered positive nuclei in the suprabasal layers.
Fig. 5. Canine auditory canal, case 3 (Table 3). Otitis externa: lining epithelium immunostained intensely and diffusely cytokeratin 14 positive. Immunohistochemistry: avidin–biotin–peroxidase complex method, 3-amino-9-ethylcarbazole chromogen, Mayer’s haematoxylin counterstain. Bar 50 μm.
Fig. 6. Canine auditory canal, case 3 (Table 3). Otitis externa: suprabasal epithelial layers were intensely cytokeratin 16-positive. Immunohistochemistry: avidin–biotin– peroxidase complex method, 3-amino-9-ethylcarbazole chromogen, Mayer’s haematoxylin counterstain. Bar 50 μm.
444
B. Banco et al./The Veterinary Journal 200 (2014) 440–445
Discussion This study investigated the histological and immunohistochemical features of canine aural cholesteatoma. Cases were characterised by otoscopic and tomographic findings indicating the presence of a cystic, expanding mass in the middle ear with loss of air contrast on CT. Histologically, canine cholesteatoma was characterised by an intensely hyperplastic hyperkeratotic epithelium (matrix) resting on a fibrous perimatrix, variably infiltrated by inflammatory cells and always devoid of cutaneous adnexa. Epithelial hyperplasia and hyperkeratosis were also common features in chronic otitis externa, although markedly less severe than in cholesteatoma. Moreover, the cutaneous adnexa, which were absent in cholesteatomata, were always detectable in otitis specimens. Thus, epithelial hyperplasia and the absence of adnexa are histological features highly suggestive of cholesteatomata. However, we believe that the histological diagnosis of cholesteatoma in biopsy specimens should always be supported by consistent clinical and tomographic findings. Aural cholesteatoma is, by definition, a cystic lesion composed of squamous epithelium filling and eventually expanding the middle ear. Since the middle ear cavity is normally lined by ciliated epithelium, different hypotheses have been formulated to explain the origin of the squamous epithelium. Some authors have interpreted squamous epithelium as a metaplastic change of ciliated epithelium elicited by severe chronic inflammation of the middle ear (Peck, 1977; Little et al., 1991b). Other workers have hypothesised that the squamous epithelium migrated from the external ear through a defect of the tympanic membrane (Palva et al., 1972). The outer epithelium of the tympanic membrane and the epithelium of the adjacent ear are capable of migration in severe inflammation (Makino and Amatsu, 1986). More recently, Lanz and Wood (2004) and Blutke et al. (2010) reported that the ventral portion of the tympanic cavity in dogs is normally covered by a thin layer of simple squamous epithelium. This epithelium could, during inflammation, undergo uncontrolled proliferation, giving rise to cholesteatoma. In our study, limited portions of squamous epithelium, occasionally contiguous with ciliated epithelium, were observed in a few cases of otitis media. This squamous epithelium could be interpreted either as focal metaplasia of ciliated epithelium or as a normal anatomical feature, as suggested by Lanz and Wood (2004) and Blutke et al. (2010). However, the exact origin of these segments of epithelium could not be definitively identified, based on our morphological observations. The results of immunophenotyping in this study were mostly consistent with previous reports for human cholesteatomata (Kim et al., 2002; Olszewska et al., 2005; Olszewska and Sudhoff, 2007) and reflected a hyperproliferative state and altered differentiation of keratinocytes. More specifically, in humans, CK16 is considered a marker of epithelial hyperproliferation, as it is expressed in cholesteatomata and in chronic inflammatory or neoplastic epidermal diseases (Bujía et al., 1993). CK16 is absent in normal epidermis, except in the tympanic membrane and in zones submitted to intense pressure or stress (e.g. the soles of the feet; Weiss et al., 1984; Castelijns et al., 1999; Ishiko et al., 2007; Caliman e Gurgel et al., 2010). In dogs, CK16 immunostaining has been demonstrated in epidermal neoplasia (Yasuno et al., 2009) as well as in normal skin, and it was proposed that this reflected an adaptation of canine skin to higher loads and stresses (Walter, 2001). In the present study, CK16 was expressed in the cholesteatoma matrix and hyperplastic epithelium of chronic otitides, but not in normal aural epithelium. Thus, while CK16 expression confirmed the hyperproliferative nature of canine cholesteatoma, it failed to elucidate the origin of the keratinising epithelium comprising the cholesteatoma matrix. CK14 is known as a marker of terminal differentiation in humans, typical of stratified epithelia and synthesised in the basal epithe-
lial cells. In non-pathological conditions, the basal epithelial layers of the human external aural meatus are positively labelled (Broekaert et al., 1992; Olszewska et al., 2005). In aural cholesteatomata, CK14 immunohistochemical staining also extends into the suprabasal epithelial layers of the matrix, suggesting an altered epithelial differentiation pattern (Broekaert et al., 1988; Olszewska et al., 2005). In addition, CK14 expression in the cholesteatoma matrix has been interpreted as an indication of the epidermal origin of cholesteatoma (Vennix et al., 1996; Olszewska et al., 2005), supporting the hypothesis of migration rather than epithelial metaplasia. In veterinary medicine, CK14 is a well-known marker of the basal cell layers of normal canine skin (Walter, 2001). In the present study, CK14 expression extended to all layers of the matrix of canine cholesteatoma, consistent with findings in human cholesteatoma. The same pattern of CK14 expression was observed in the squamous epithelium in all cases of otitis, while CK14 expression in the normal external meatus was mostly observed in basal epithelial cells. Our results suggest that altered epithelial differentiation occurs, but they do not confirm that the cholesteatoma matrix originates from epidermis rather than epithelial metaplasia. CK19 and CK8/18 are markers of simple epithelia in both humans and dogs (Moll et al., 1982; Griffey et al., 1993; Lee et al., 1994). Consistent with previous studies in humans (Olszewska et al., 2005; Olszewska and Sudhoff, 2007), these markers in our cases were expressed in middle ear ciliated columnar epithelium but not in cholesteatoma, otitis externa, or the normal auditory meatus. Taken together, the immunohistochemistry results in our study were consistent with cholesteatoma having a hyperproliferative epithelium, but gave no definitive clues concerning the origin of the epithelial cells. MIB 1 (Ki67) immunostaining has been extensively employed in human medicine to evaluate the proliferative activity of the epithelium of cholesteatoma and it has been proposed as a tool to estimate the aggressiveness of cholesteatoma and its likelihood of recurrence (Bujía et al., 1996; Mallet et al., 2003). An increase in the PI of cholesteatoma keratinocytes compared with normal external auditory meatal keratinocytes has been reported (Sudhoff et al., 1995; Bujía et al., 1996; Olszewska et al., 2003); however, the relationship between the PI and clinical signs of aggressiveness was not statistically significant. In the present study, the PI of the cholesteatoma matrix was higher than that of normal meatal skin (1.4) but similar to that of chronic otitides. Moreover, the PI in the single case of cholesteatoma in this study that recurred after 5 months (case 2) was 12.4, among the lowest obtained in our cholesteatomata. These observations confirmed that Ki67 is a valuable indicator of the hyperproliferative state of the cholesteatoma matrix, but in the cases reported here, it was not a predictor of aggressiveness. Conclusions Similar cytoskeletal immunophenotypes were demonstrated in the cholesteatoma matrix, the epithelium of an external ear canal with otitis, and in the squamous epithelium of the middle ear. This suggests that the cytokeratin pattern of expression could not be used, at least in canine species, as a valid indicator of cholesteatoma origin. Ki67 immunostaining confirmed the hyperproliferative features of the cholesteatoma matrix, which is consistent with the progressive expansive growth that defines the lesion. However, Ki67 did not seem to be related to specific features of aggressiveness in cholesteatomata, since similar values were obtained in chronic otitides. It is possible that mechanisms other than progressive growth play a decisive role in the pathogenesis of bone lysis and tissue destruction associated with cholesteatoma. Future investigations into the role of inflammation in the development of cholesteatomata and the inflammatory mediators released in the perimatrix might help elucidate the mechanisms of the clinical aggressiveness in choles-
B. Banco et al./The Veterinary Journal 200 (2014) 440–445
teatomata. However, the histological evaluation of cholesteatoma, even using specific tools such as immunohistochemical assays, cannot assess the multifaceted problem of cholesteatoma recurrence. Other factors, not least the adequacy of surgical resection, can contribute significantly to recurrence of the lesion, independent of the intrinsic proliferative capability of the epithelium composing the cholesteatoma. In our opinion, aural cholesteatoma is a complex lesion and management must integrate clinical, radiological and histological data. Conflict of interest statement None of the authors has any financial or personal relationships that could inappropriately influence or bias the content of the paper. Acknowledgements The authors received no financial support for the research, authorship, and/or publication of this article. References Blutke, A., Parzefall, B., Steger, A., Goedde, T., Hermanns, W., 2010. Inflammatory polyp in the middle ear of a dog: A case report. Veterinarni Medicina 55, 289–293. Broekaert, D., Cornille, A., Eto, H., Leigh, I., Ramaekers, F., Van Muijen, G., Coucke, P., De Bersaques, J., Kluyskens, P., Gillis, E., 1988. A comparative immunohistochemical study of cytokeratin and vimentin expression in middle ear mucosa and cholesteatoma, and in epidermis. Virchows Archiv. A, Pathological Anatomy and Histopathology 413, 39–51. Broekaert, D., Coucke, P., Leperque, S., Ramaekers, F., Van Muijen, G., Boedts, D., Leigh, I., Lane, B., 1992. Immunohistochemical analysis of the cytokeratin expression in middle ear cholesteatoma and related epithelial tissues. The Annals of Otology, Rhinology, and Laryngology 101, 931–938. Bujía, J., Schilling, V., Holly, A., Stammberger, M., Kastenbauer, E., 1993. Hyperproliferation-associated keratin expression in human middle ear cholesteatoma. Brazilian Journal of Otorhinolaryngology 113, 364–368. Bujía, J., Kim, C., Ostos, P., Kastenbauer, E., Hültner, L., 1996. Role of interleukin 6 in epithelial hyperproliferation and bone resorption in middle ear cholesteatomas. European archives of Otorhinolaryngology 253, 152–157. Caliman e Gurgel, J.D., Pereira, S.B., Alves, A.L., Ribeiro, F.Q., 2010. Hyperproliferation markers in ear canal epidermis. Brazilian Journal of Otorhinolaryngology 76, 667–671. Castelijns, F.A., Gerritsen, M.J., van Erp, P.E., van de Kerkhof, P.C., 1999. Immunohistochemical assessment of keratin 16 on paraffin-embedded sections of normal and hyperproliferative skin: Monoclonal antibodies Ks 8.12 and LL025 in a comparative study. Archives of Dermatological Research 291, 59–63. Ferlito, A., Devaney, K.O., Rinaldo, A., Milroy, C.M., Wenig, B.M., Iurato, S., McCabe, B.F., 1997. Clinicopathological consultation. Ear cholesteatoma versus cholesterol granuloma. The Annals of Otology, Rhinology, and Laryngology 106, 79–85. Friedmann, I., Arnold, W., 1993. Inflammatory diseases of the middle ear and their sequelae. In: Friedmann, I., Arnold, W. (Eds.), Pathology of the Ear, Churchill Livingstone, London, UK, pp. 88–92. Gersdorff, M.C., Debaty, M.E., Tomasi, J.P., 2006. Pathophysiology of cholesteatoma. Revue de Laryngologie Otologie Rhinologie (Bord) 127, 115–119. Greci, V., Travetti, O., Di Giancamillo, M., Lombardo, R., Giudice, C., Banco, B., Mortellaro, C.M., 2011. Middle ear cholesteatoma in 11 dogs. The Canadian Veterinary Journal 52, 631–636. Griffey, S.M., Madewell, B.R., Dairkee, S.H., Hunt, J.E., Naydan, D.K., Higgins, R.J., 1993. Immunohistochemical reactivity of basal and luminal epithelium-specific cytokeratin antibodies within normal and neoplastic canine mammary glands. Veterinary Pathology 30, 155–161. Harran, N.X., Bradley, K.J., Hetzel, N., Bowlt, K.L., Day, M.J., Barr, F., 2012. MRI findings of a middle ear cholesteatoma in a dog. Journal of the American Animal Hospital Association 48, 339–343. Hsu, S.M., Raine, L., Fanger, H., 1981. Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: A comparison between ABC and unlabeled antibody (PAP) procedures. Journal of Histochemistry and Cytochemistry 29, 577–580.
445
Ishiko, A., Dekio, I., Fujimoto, A., Kameyama, K., Sakamoto, M., Benno, Y., Amagai, M., Nishikawa, T., 2007. Abnormal keratin expression in circumscribed palmar hypokeratosis. Journal of the American Academy of Dermatology 57, 285–291. Kim, H.J., Tinling, S.P., Chole, R.A., 2002. Expression patterns of cytokeratins in cholesteatomas: Evidence of increased migration and proliferation. Journal of Korean Medical Science 17, 381–388. Koltai, P.J., Nelson, M., Castellon, R.J., Garabedian, E.N., Triglia, J.M., Roman, S., Roger, G., 2002. The natural history of congenital cholesteatoma. Archives of Otolaryngology – Head and Neck Surgery 128, 804–809. Kuijpers, W., Vennix, P.P., Peters, T.A., Ramaekers, F.C., 1996. Squamous metaplasia of the middle ear epithelium. Acta Oto-Laryngologica 116, 293–298. Lanz, O.I., Wood, B.C., 2004. Surgery of the ear and pinna. Veterinary Clinics of North America Small Animal Practice 34, 567–599. Lee, R.J., Sidey, C., Narula, A.A., James, R.F., 1994. The nature of the epithelium in acquired cholesteatoma: Part 3-cytokeratin patterns in aural epithelial and cholesteatoma cells grown in cell culture. Clinical Otolaryngology and Allied Sciences 19, 516–520. Little, C.J., Lane, J.G., Pearson, G.R., 1991a. Inflammatory middle ear disease of the dog: The pathology of otitis media. Veterinary Record 128, 293–296. Little, C.J., Lane, J.G., Gibbs, C., Pearson, G.R., 1991b. Inflammatory middle ear disease of the dog: The clinical and pathological features of cholesteatoma, a complication of otitis media. Veterinary Record 128, 319–322. Makino, K., Amatsu, M., 1986. Epithelial migration on the tympanic membrane and external canal. Archives of Otolaryngology 243, 39–42. Mallet, Y., Nouwen, J., Lecomte-Houcke, M., Desaulty, A., 2003. Aggressiveness and quantification of epithelial proliferation of middle ear cholesteatoma by MIB1. The Laryngoscope 113, 328–331. Moll, R., Franke, W.W., Schiller, D.L., Geiger, B., Krepler, R., 1982. The catalog of human cytokeratins: Patterns of expression in normal epithelia, tumors and cultured cells. Cell 31, 11–24. Müller, J., 1838. Von der geschichteten perlmutter glanzenden Fettgeschwulst. In: Reimer, G. (Ed.), Ueber den feineren Bau und die formen der krankhaften Geschwulste, Berlin, Germany, pp. 49–54. Olszewska, E., Chodynicki, S., Chyczewski, L., 2003. Some markers of proliferative activity in cholesteatoma epithelium in adults. Otolaryngologia Polska. the Polish Otolaryngology 57, 85–89. Olszewska, E., Lautermann, J., Koc, C., Schwaab, M., Dazert, S., Hildmann, H., Sudhoff, H., 2005. Cytokeratin expression pattern in congenital and acquired pediatric cholesteatoma. European Archives of Otorhinolaryngology 262, 731–736. Olszewska, E., Sudhoff, H., 2007. Comparative cytokeratin distribution patterns in cholesteatoma epithelium. Histology and Histopathology 22, 37–42. Palva, T., Palva, A., Dammert, K., 1972. Keratin in middle ear squamous epithelium. Archives of Otolaryngology 95, 214–220. Peck, G.L., 1977. Cholesteatoma. In: McCabe, B.F., Sade, J., Abramson, M. (Eds), 1st International Conference. Birmingham, AL, 45. Persaud, R., Hajioff, D., Trinidade, A., Khemani, S., Bhattacharyya, M.N., Papadimitriou, N., Kalan, A., Bhattacharyya, A.K., 2007. Evidence-based review of aetiopathogenic theories of congenital and acquired cholesteatoma. The Journal of Laryngology and Otology 121, 1013–1109. Sadé, J., 1980. Retraction pockets and attic cholesteatomas. Acta Oto-RhinoLaryngologica Belgica 34, 62–84. Semaan, M.T., Megerian, C.A., 2006. The pathophysiology of cholesteatoma. Otolaryngologic Clinics of North America 39, 1143–1159. Strunk, C.L., 1993. Cholesteatoma. In: Bailey, B.J., Johnson, J.T., Newlands, S.D. (Eds.), Head and Neck Surgery-Otolaryngology, vol. 2. Lippincott Williams and Wilkins, Philadelphia, PA, USA, pp. 1635–1646. Sudhoff, H., Bujía, J., Fisseler-Eckhoff, A., Holly, A., Schulz-Flake, C., Hildmann, H., 1995. Expression of a cell-cycle-associated nuclear antigen (MIB 1) in cholesteatoma and auditory meatal skin. The Laryngoscope 105, 1227–1231. Travetti, O., Giudice, C., Greci, V., Lombardo, R., Mortellaro, C.M., Di Giancamillo, M., 2010. Computed tomography features of middle ear cholesteatoma in dogs. Veterinary Radiology and Ultrasound 51, 374–379. Venker-Van Haagen, A.J., 2005. The ear. In: Ear, Nose, Throat and Tracheobronchial Diseases in Dogs and Cats, 1st ed. Schlütersche, Verlagsgesellschaft mbH, Hannover, Germany, pp. 1–50. Vennix, P.P., Kuijpers, W., Peters, T.A., Tonnaer, E.L., Ramaekers, F.C., 1996. Keratinocyte differentiation in acquired cholesteatoma and perforated tympanic membranes. Archives of Otolaryngology – Head and Neck Surgery 122, 825–832. Walter, J.H., 2001. Cytokeratins in the canine epidermis. Veterinary Dermatology 12, 81–87. Weiss, R.A., Eichner, R., Sun, T.T., 1984. Monoclonal antibody analysis of keratin expression in epidermal diseases: A 48- and 56-kdalton keratin as molecular markers for hyperproliferative keratinocytes. The Journal of Cell Biology 98, 1397–1406. Yasuno, K., Nishiyama, S., Suetsugu, F., Ogihara, K., Madarame, H., Shirota, K., 2009. Cutaneous clear cell adnexal carcinoma in a dog: Special reference to cytokeratin expression. Journal of Veterinary Medical Science 71, 1513–1517.
Contents lists available at ScienceDirect
The Veterinary Journal j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / t v j l
Canine aural cholesteatoma: A histological and immunohistochemical study Barbara Banco a, Valeria Grieco a, Mauro Di Giancamillo b, Valentina Greci a, Olga Travetti b, Pieranna Martino a, Carlo M. Mortellaro a, Chiara Giudice a,* a
Dipartimento di Scienze Veterinarie e Sanità Pubblica, Università degli Studi di Milano, via Celoria 10, 20133 Milan, Italy Dipartimento di Scienze Veterinarie per la Salute, la Produzione Animale e la Sicurezza Alimentare, Università degli Studi di Milano, via Celoria 10, 20133 Milan, Italy
b
A R T I C L E
I N F O
Article history: Accepted 19 March 2014 Keywords: Cholesteatoma Cytokeratins Dog Ear Epidermoid Ki67
A B S T R A C T
Canine aural cholesteatoma is an epidermoid cyst that forms in the middle ear cavity as a rare complication of otitis media but the aetiopathogenesis remains controversial. In the present study, 13 cases of canine aural cholesteatoma were investigated histologically and immunohistochemically and compared with cases of chronic otitis. The immunohistochemical investigation was performed using the following monoclonal antibodies: anti-cytokeratins (CK) 14, 16, 8/18, and 19, and anti-Ki67. The proliferative indexes (PIs) of cholesteatomata and otitis epithelium were calculated as the percentage of Ki67 positive nuclei/ total nuclei. Histologically, the cholesteatomata were composed of a hyperplastic, hyperkeratotic epithelium (matrix) resting on a fibrous perimatrix, infiltrated by inflammatory cells and devoid of cutaneous adnexa. Immunohistochemically, the cholesteatoma epithelium was CK14- and CK16-positive, and CK8/ 18- and CK19-negative. A similar pattern of CK expression was found in otitis externa. In otitis media, ciliated epithelium stained CK8/18- and CK19-positive in all layers, CK14-positive in the basal layers, and CK16-negative. The mean PIs in cholesteatomata and otitides were 18.8 and 17.8, respectively. The immunohistochemical pattern of CK expression in cholesteatomata, when compared with chronic otitis, was suggestive of hyperproliferative epithelium, but its origin could not be demonstrated. Comparable PI values were obtained in cholesteatoma and in chronic otitis, which confirmed that Ki67 is a valuable indicator of a hyperproliferative state, but not a predictor of aggressiveness. © 2014 Elsevier Ltd. All rights reserved.
Introduction Aural cholesteatoma is an epidermoid cyst that forms in the middle ear and is composed of hyperplastic, hyperkeratotic squamous epithelium surrounding a core of keratin debris. It has been described as ‘skin growing in the wrong place’ (Strunk, 1993) and the term ‘cholesteatoma’ itself, coined by Müller in 1838, is currently considered a misnomer because it is neither a granulomatous lesion or a neoplasm (Friedmann and Arnold, 1993; Strunk, 1993). In humans, aural cholesteatoma is classified as congenital or acquired (Persaud et al., 2007). Congenital cholesteatoma develops beneath an intact tympanic membrane in patients whose clinical history does not include an obvious ear infection or auditory tube dysfunction (Strunk, 1993). It is hypothesised to originate from embryonic epidermoid remnants that fail to involute (Koltai et al., 2002). Acquired cholesteatoma is a complication of chronic otitis and it is
* Corresponding author. Tel.: +39 0250 318105. E-mail address: [email protected] (C. Giudice). http://dx.doi.org/10.1016/j.tvjl.2014.03.018 1090-0233/© 2014 Elsevier Ltd. All rights reserved.
thought to originate either as a result of squamous metaplasia of the middle ear epithelium or from the migration of external ear canal epithelium through a tympanic perforation (Sadé, 1980; Semaan and Megerian, 2006). Other workers have suggested that it could originate from an invagination of a portion of the tympanic membrane into the epitympanum, secondary to a Eustachian tube dysfunction (retraction pocket theory; Sadé, 1980; Olszewska et al., 2003). To date, the origin of the epithelium of cholesteatoma remains uncertain. In dogs, cholesteatoma is mostly considered an acquired disease and a severe complication of otitis media, with or without concurrent rupture of the tympanic membrane (Little et al., 1991a, 1991b; Venker-Van Haagen, 2005; Harran et al., 2012). To our knowledge, congenital cholesteatoma has not been reported in dogs. In human medicine, several immunohistochemical studies have been performed in an attempt to identify the origin of cholesteatoma epithelium (Bujía et al., 1993; Kuijpers et al., 1996; Olszewska et al., 2005). More specifically, the expression of cytokeratins, the intermediate filaments of epithelia, has been extensively investigated, since the pattern of cytokeratin expression in a particular epithelium varies with its anatomical location, developmental stage,
B. Banco et al./The Veterinary Journal 200 (2014) 440–445
441
Table 1 Details of antibodies employed in immunohistochemistry (antigen retrieval, dilution and sources). Antibodies
Clone
Antigenic retrieval
Dilution
Source
CK14 CK16 CK8/18 CK19 Ki67
LL002 LL025 5D3 B170 MIB-1
Pressure cooker, 10 min, citrate buffer (pH 6) Pressure cooker, 10 min, citrate buffer (pH 6) Pepsina, 37 °C, 14 min Pepsina, 37 °C, 14 min Pressure cooker, 18 min, citrate buffer (pH 6)
1 : 2000 1 : 200 1 : 200 1 : 200 1 : 600
NeoMarkers Novocastra Zymed Novocastra Dakocytomation
a
Zymed.
and state of differentiation. However, conflicting results have been reported. To elucidate the mechanisms of the progressive growth that characterise cholesteatomata, numerous studies in human medicine focused on the proliferative capability of lining epithelium (Strunk, 1993; Gersdorff et al., 2006). Investigations used MIB 1 (Ki67) immunostaining, in an attempt to characterise the aggressiveness of the cholesteatoma and the likelihood of recurrence (Bujía et al., 1996; Mallet et al., 2003). To the best of our knowledge, detailed histological and immunohistochemical characterisations of canine middle ear cholesteatoma are not currently available in the veterinary literature. Therefore, the aim of the current study was to characterise canine aural cholesteatoma, comparing epithelial immunophenotypes (cytokeratins) and proliferative activity (Ki67) in cholesteatomata, in normal auditory canals, and in otitis externa and otitis media.
otitis) was observed on video-otoscopy in all cases. Eight cases had rupture of the tympanic membrane, and amorphous, pearly material in the middle ear cavity was evident in nine cases. The middle ear content was cultured in each of the 13 cases. Staphylococcus intermedius was isolated in three cases, Proteus mirabilis in two cases, E. coli in one case, Pseudomonas aeruginosa and P. mirabilis in one case and Pseudomonas aeruginosa and Klebsiella pneumoniae in another case. No bacteria were recovered in five cases. Histopathology Normal external auditory meatus The normal meatus was lined by a thin squamous epithelium, one to two cell layers thick, resting on a derma containing hair follicles, sebaceous and ceruminous glands, and supported by annular cartilage.
Materials and methods Thirteen aural cholesteatomata from 11 dogs were reviewed; two cases were recurrences. Specimens were obtained by total ear canal ablation and lateral bulla osteotomy (TECALBO) or ventral bulla osteotomy (VBO), fixed in 10% neutral buffered formalin, and submitted to the Department of Veterinary Science and Public Health (DIVET) of Milan for histopathology. In all cases, clinical, radiological (computed tomography scan, CT), and video-otoscopy findings were consistent with the diagnosis of aural cholesteatoma (Little et al., 1991b; Travetti et al., 2010; Greci et al., 2011). Signalment, clinical signs, and results of middle ear swabs for bacteriological culture were recorded for each dog. Follow-up consisted of a clinical examination. Ten histological specimens of chronic otitis externa or media from 10 dogs and 1 normal canine external meatus auditory canal specimen were retrieved from DIVET archives and were included in the study. All histological specimens were routinely processed, paraffin embedded, and stained with haematoxylin and eosin (HE). Serial sections were obtained and immunolabelled using the standard avidin–biotin–peroxidase complex (ABC) procedure (Hsu et al., 1981) and the following antibodies: anti-cytokeratin 8/18, 14, 16, and 19, and anti- Ki67 protein. Primary antibodies and antigen retrieval methods used are listed in Table 1. A section of normal dog skin served as a positive control. For negative controls, the primary antibody was replaced by serum from a clinically healthy horse. Cytokeratin immunostaining was quantitatively interpreted by light microscopy (positive and negative). When a positive result was obtained, immunostaining was scored as faint, moderate, or intense. The expression of Ki67 (MIB-1) was quantitatively assessed using an automatic image analysis system (Image Pro Plus 4.5, Media Cybernetics). The relative percentage of immunostained nuclei in 10 high power fields was calculated and recorded as a proliferative index (PI).
Results
Cholesteatoma Ten cases of aural cholesteatoma were composed of segments of the wall of a cystic lesion that contained abundant amorphous lamellar keratin debris. The cysts were lined by a multi-layered keratinising squamous epithelium (matrix) resting on a stroma (perimatrix) composed of dense fibrovascular connective tissue devoid of adnexa (Fig. 1). The epithelium was intensely hyperplastic (5–25 layers thick), with severe orthokeratotic hyperkeratosis, hypergranulosis and mild parakeratosis. Keratinocytes showed progressive maturation; no mitotic figures were observed. Mild multifocal intracellular oedema and minimal neutrophilic exocytosis were present. The perimatrix immediately beneath the epithelium was composed of loosely arranged collagen fibres expanded by moderate oedema; the deepest layers of the perimatrix were densely cellular, composed of fibroblasts and fibrocytes embedded in abundant collagenous stroma. There was moderate to severe stromal inflammation, consisting of numerous perivascular lymphocytes and plasma cells, scattered haemosiderin-laden macrophages and a few
Table 2 Signalment data of dogs with cholesteatoma and mean values of proliferative index (PI) of the epithelium as obtained by immunolabelling with monoclonal antibody MIB1.
Signalment and clinical features of dogs with aural cholesteatoma
Case no.
Signalment and follow-up data for dogs with cholesteatoma are summarised in Table 2. Thirteen cases of aural cholesteatoma were diagnosed in 11 dogs. Two cases (numbers 3 and 10) were recurrences. The mean age at diagnosis was 7 years (range, 5–10 years); eight dogs were males (one neutered) and there were three females (one spayed). A variety of breeds were represented. All dogs had a history of chronic recurrent otitis externa that was unresponsive to topical or systemic therapy over the previous 3–30 weeks. The main clinical signs recorded were head shaking, otodynia, and head tilt. The main findings on computed tomography (CT) were severe expansion and osteolysis of the tympanic bulla with loss of air contrast. Partial or total occlusion of the horizontal canal (end-stage
1 2 3 4 5 6 7 8 9 10 11 12 13 a
Signalment
PI
Pug, male, 8 years Mixed breed, male, 5 years Mixed breed, male, 6 years Flat coated retriever, male, 10 years Poodle, castrated male, 5 years Afghan hound, male, 8 years Weimaraner, female, 9 years American cocker spaniel, male, 6 years Schnauzer , male, 5 years Poodle, castrated male, 5 years Mixed breed, spayed female, 9 years Labrador, male, 9 years Golden retriever, female, 5 years
–a 12.45 14.12 20.51 –a 17.35 8.42 –a 31.79 15.90 34.81 15.74 16.93
Recurrence
Recurrence of case 2
Recurrence of case 5
Specimen could not be processed as it constituted only keratin debris.
442
B. Banco et al./The Veterinary Journal 200 (2014) 440–445
Fig. 1. Canine middle ear, case 3 (Table 2). Section of the cholesteatoma wall composed of an intensely hyperplastic and hyperkeratotic keratinising squamous epithelium (matrix) lining a cystic cavity filled with keratin debris and resting on a dense fibrous stroma (HE). Bar 50 μm.
neutrophils. Numerous small calibre hyperaemic vessels, often surrounded by microhaemorrhages and oedema, were also present. In four cases (numbers 7, 11, 12, and 13), multiple cholesterol granulomas were also found expanding the perimatrix. Granulomas were composed of epithelioid macrophages, multinucleated foreign-body giant cells, and scattered lymphocytes, concentrated around acicular cholesterol crystals. In two cases (numbers 9 and 12) a small segment of monolayered columnar ciliated respiratory epithelium (normal middle ear epithelium) was identified. In three cases (numbers 1, 5, and 8), the specimens surgically obtained from the middle ear were composed exclusively of amorphous keratin debris. The presence of keratin debris alone, as flakes in an expansive lesion of the middle ear, is highly suggestive of cholesteatoma (Ferlito et al., 1997). However, these three specimens were excluded from further immunohistochemical investigation. Otitides Ten cases of canine chronic otitis externa were retrieved from the archives. Of these, three cases had associated otitis media. The main signalment data are summarised in Table 3. Histologically, chronic otitis externa was characterised by severe hyperplasia of the epidermis (6–15 layers thick) with moderate to severe orthokeratotic hyperkeratosis, hypergranulosis, mild intracellular oedema, and spongiosis (Fig. 2). Scattered intraepithelial pustules and multifocal ulcerations covered with serocellular crusts, occasionally containing
Table 3 Signalment data of dogs with chronic otitis and mean values of proliferative index (PI) of the epithelium, as obtained by immunolabelling with monoclonal antibody MIB1. Case no. 1 2 3 4 5 6 7 8 9 10
Signalment
PI
German shepherd, male, 4 years Mixed breed, female, 13 years Maltese terrier, female, 3 years German shepherd, female, 7 years Cocker spaniel, male, 8 years Yorkshire terrier, female, 3 years Schnauzer, male, 6 years English bulldog, male 9 years German shepherd, male, 4 years Mixed breed, male, 10 years
24.55 23.35 17.23 9.98 9.70 13.53 22.98 25.91 12.89 18.37
Fig. 2. Canine auditory canal, case 8 (Table 3). Severe otitis externa with intense hyperplasia and hyperkeratosis of the epidermis, fibrosis, and intense dermal lymphoplasmacytic infiltrate (HE). Bar 200 μm. Inset: section of a normal auditory canal; lining epithelium is composed of one to two layers of cells. Sebaceous and apocrine glands are present in the derma (HE). Bar 200 μm.
bacterial aggregates, were also common. The derma was characterised by severe fibrosis, moderate numbers of perivascular aggregates of lymphocytes, plasma cells and haemosiderinladen macrophages, dilation of ceruminous glands with stasis of secretum, and numerous hyperaemic vessels with perivascular oedema. Otitis media, associated with otitis externa in three cases (numbers 6, 7 and 9), was characterised by multifocal areas of ulceration and erosion of the respiratory epithelium. When recognisable, columnar epithelial cells were either shrunken and detached, or swollen with loss of cilia. Rarely, areas lined by squamous epithelium with neutrophilic exocytosis were also evident. The underlying corion was fibrotic, hyperaemic, infiltrated by numerous lymphocytes and plasma cells, with fewer neutrophils and haemosiderin-laden macrophages. In one case, a cholesterol granuloma was also present. Immunohistochemistry The immunohistochemical labelling results are summarised in Table 4. Epithelium lining the normal external auditory meatus was moderately, diffusely CK14-positive and CK16-, CK8/18-, and CK19negative. When cornified epithelial cells were recognisable, they were mostly CK14-negative. The epithelium of ceruminous glands was immunostained with CK8/18 and CK19. Only scattered epithelial cells of the external canal epithelium stained Ki67-positive (PI, 1.4). The cholesteatoma epithelium (matrix) was in all cases CK14and CK16-positive and CK8/18- and CK19-negative. CK14 staining was intense in keratinocytes throughout all matrix layers (Fig. 3). CK16 staining was also intense, but limited to the suprabasal layers of the epithelium (Fig. 4). The PI in the cholesteatoma epithelium varied from 8.4 to 34.8 (mean, 18.8; median, 16.4; standard deviation, SD 7.8; Table 2) among cases. Positive nuclei were mainly detected in epithelial cells of the basal layer with only rare, scattered positive nuclei within the suprabasal layers. The epithelium in otitis externa was strongly CK14- and CK16positive and CK8/18- and CK19-negative in all cases. CK14 labelling was diffuse in all layers of the epithelium (Fig. 5), while CK16 immunolabelling was limited to the suprabasal layers (Fig. 6). In the
B. Banco et al./The Veterinary Journal 200 (2014) 440–445
443
Table 4 Summary of immunohistochemical results of cytokeratin (CK) immunolabelling in normal ear, cholesteatoma and otitides. Normal aural meatus
CK14 CK16 CK8/18 CK19 a b c
Aural cholesteatoma
Otitis externa
Otitis media
Suprabasal layers
Basal layers
Suprabasal layers
Basal layers
Suprabasal layers
Basal layers
Ciliated epithelium
Squamous epithelium
− − − −
+ − − −
+ + − −
+ − − −
+ + − −
+ − − −
+ − + +
+b +c − −
a
Ciliated epithelium in otitis media: CK14 immunolabelling was limited to basal epithelial layer. Squamous epithelium in otitis media: CK14 immunolabelling was diffuse to basal and suprabasal epithelial layers. Squamous epithelium in otitis media: CK16 immunolabelling was limited to subrabasal epithelial layers.
few cases of otitis media, columnar epithelium was diffusely CK8/ 18 and CK19-positive, while CK14 stained only basal layers and CK16 was always negative. Squamous epithelium was diffusely CK14positive, while CK16 staining was limited to the suprabasal layers
Fig. 3. Canine middle ear, case 4 (Table 2). Cholesteatoma: lining epithelium immunostained intensely and diffusely cytokeratin 14 positive. Immunohistochemistry: avidin–biotin–peroxidase complex method, 3-amino-9-ethylcarbazole chromogen, Mayer’s haematoxylin counterstain. Bar 50 μm.
Fig. 4. Canine middle ear, case 3 (Table 2). Cholesteatoma: suprabasal epithelial layers were intensely cytokeratin 16-positive. Immunohistochemistry: avidin–biotin– peroxidase complex method, 3-amino-9-ethylcarbazole chromogen, Mayer’s haematoxylin counterstain. Bar 50 μm.
and CK8/18 and CK19 were always negative. The PIs ranged from 9.7 to 25.9 (mean, 17.8; median, 17.8; SD 5.8; Table 3). Positive nuclei were observed mainly in epithelial cells of the basal layer, with only occasional, scattered positive nuclei in the suprabasal layers.
Fig. 5. Canine auditory canal, case 3 (Table 3). Otitis externa: lining epithelium immunostained intensely and diffusely cytokeratin 14 positive. Immunohistochemistry: avidin–biotin–peroxidase complex method, 3-amino-9-ethylcarbazole chromogen, Mayer’s haematoxylin counterstain. Bar 50 μm.
Fig. 6. Canine auditory canal, case 3 (Table 3). Otitis externa: suprabasal epithelial layers were intensely cytokeratin 16-positive. Immunohistochemistry: avidin–biotin– peroxidase complex method, 3-amino-9-ethylcarbazole chromogen, Mayer’s haematoxylin counterstain. Bar 50 μm.
444
B. Banco et al./The Veterinary Journal 200 (2014) 440–445
Discussion This study investigated the histological and immunohistochemical features of canine aural cholesteatoma. Cases were characterised by otoscopic and tomographic findings indicating the presence of a cystic, expanding mass in the middle ear with loss of air contrast on CT. Histologically, canine cholesteatoma was characterised by an intensely hyperplastic hyperkeratotic epithelium (matrix) resting on a fibrous perimatrix, variably infiltrated by inflammatory cells and always devoid of cutaneous adnexa. Epithelial hyperplasia and hyperkeratosis were also common features in chronic otitis externa, although markedly less severe than in cholesteatoma. Moreover, the cutaneous adnexa, which were absent in cholesteatomata, were always detectable in otitis specimens. Thus, epithelial hyperplasia and the absence of adnexa are histological features highly suggestive of cholesteatomata. However, we believe that the histological diagnosis of cholesteatoma in biopsy specimens should always be supported by consistent clinical and tomographic findings. Aural cholesteatoma is, by definition, a cystic lesion composed of squamous epithelium filling and eventually expanding the middle ear. Since the middle ear cavity is normally lined by ciliated epithelium, different hypotheses have been formulated to explain the origin of the squamous epithelium. Some authors have interpreted squamous epithelium as a metaplastic change of ciliated epithelium elicited by severe chronic inflammation of the middle ear (Peck, 1977; Little et al., 1991b). Other workers have hypothesised that the squamous epithelium migrated from the external ear through a defect of the tympanic membrane (Palva et al., 1972). The outer epithelium of the tympanic membrane and the epithelium of the adjacent ear are capable of migration in severe inflammation (Makino and Amatsu, 1986). More recently, Lanz and Wood (2004) and Blutke et al. (2010) reported that the ventral portion of the tympanic cavity in dogs is normally covered by a thin layer of simple squamous epithelium. This epithelium could, during inflammation, undergo uncontrolled proliferation, giving rise to cholesteatoma. In our study, limited portions of squamous epithelium, occasionally contiguous with ciliated epithelium, were observed in a few cases of otitis media. This squamous epithelium could be interpreted either as focal metaplasia of ciliated epithelium or as a normal anatomical feature, as suggested by Lanz and Wood (2004) and Blutke et al. (2010). However, the exact origin of these segments of epithelium could not be definitively identified, based on our morphological observations. The results of immunophenotyping in this study were mostly consistent with previous reports for human cholesteatomata (Kim et al., 2002; Olszewska et al., 2005; Olszewska and Sudhoff, 2007) and reflected a hyperproliferative state and altered differentiation of keratinocytes. More specifically, in humans, CK16 is considered a marker of epithelial hyperproliferation, as it is expressed in cholesteatomata and in chronic inflammatory or neoplastic epidermal diseases (Bujía et al., 1993). CK16 is absent in normal epidermis, except in the tympanic membrane and in zones submitted to intense pressure or stress (e.g. the soles of the feet; Weiss et al., 1984; Castelijns et al., 1999; Ishiko et al., 2007; Caliman e Gurgel et al., 2010). In dogs, CK16 immunostaining has been demonstrated in epidermal neoplasia (Yasuno et al., 2009) as well as in normal skin, and it was proposed that this reflected an adaptation of canine skin to higher loads and stresses (Walter, 2001). In the present study, CK16 was expressed in the cholesteatoma matrix and hyperplastic epithelium of chronic otitides, but not in normal aural epithelium. Thus, while CK16 expression confirmed the hyperproliferative nature of canine cholesteatoma, it failed to elucidate the origin of the keratinising epithelium comprising the cholesteatoma matrix. CK14 is known as a marker of terminal differentiation in humans, typical of stratified epithelia and synthesised in the basal epithe-
lial cells. In non-pathological conditions, the basal epithelial layers of the human external aural meatus are positively labelled (Broekaert et al., 1992; Olszewska et al., 2005). In aural cholesteatomata, CK14 immunohistochemical staining also extends into the suprabasal epithelial layers of the matrix, suggesting an altered epithelial differentiation pattern (Broekaert et al., 1988; Olszewska et al., 2005). In addition, CK14 expression in the cholesteatoma matrix has been interpreted as an indication of the epidermal origin of cholesteatoma (Vennix et al., 1996; Olszewska et al., 2005), supporting the hypothesis of migration rather than epithelial metaplasia. In veterinary medicine, CK14 is a well-known marker of the basal cell layers of normal canine skin (Walter, 2001). In the present study, CK14 expression extended to all layers of the matrix of canine cholesteatoma, consistent with findings in human cholesteatoma. The same pattern of CK14 expression was observed in the squamous epithelium in all cases of otitis, while CK14 expression in the normal external meatus was mostly observed in basal epithelial cells. Our results suggest that altered epithelial differentiation occurs, but they do not confirm that the cholesteatoma matrix originates from epidermis rather than epithelial metaplasia. CK19 and CK8/18 are markers of simple epithelia in both humans and dogs (Moll et al., 1982; Griffey et al., 1993; Lee et al., 1994). Consistent with previous studies in humans (Olszewska et al., 2005; Olszewska and Sudhoff, 2007), these markers in our cases were expressed in middle ear ciliated columnar epithelium but not in cholesteatoma, otitis externa, or the normal auditory meatus. Taken together, the immunohistochemistry results in our study were consistent with cholesteatoma having a hyperproliferative epithelium, but gave no definitive clues concerning the origin of the epithelial cells. MIB 1 (Ki67) immunostaining has been extensively employed in human medicine to evaluate the proliferative activity of the epithelium of cholesteatoma and it has been proposed as a tool to estimate the aggressiveness of cholesteatoma and its likelihood of recurrence (Bujía et al., 1996; Mallet et al., 2003). An increase in the PI of cholesteatoma keratinocytes compared with normal external auditory meatal keratinocytes has been reported (Sudhoff et al., 1995; Bujía et al., 1996; Olszewska et al., 2003); however, the relationship between the PI and clinical signs of aggressiveness was not statistically significant. In the present study, the PI of the cholesteatoma matrix was higher than that of normal meatal skin (1.4) but similar to that of chronic otitides. Moreover, the PI in the single case of cholesteatoma in this study that recurred after 5 months (case 2) was 12.4, among the lowest obtained in our cholesteatomata. These observations confirmed that Ki67 is a valuable indicator of the hyperproliferative state of the cholesteatoma matrix, but in the cases reported here, it was not a predictor of aggressiveness. Conclusions Similar cytoskeletal immunophenotypes were demonstrated in the cholesteatoma matrix, the epithelium of an external ear canal with otitis, and in the squamous epithelium of the middle ear. This suggests that the cytokeratin pattern of expression could not be used, at least in canine species, as a valid indicator of cholesteatoma origin. Ki67 immunostaining confirmed the hyperproliferative features of the cholesteatoma matrix, which is consistent with the progressive expansive growth that defines the lesion. However, Ki67 did not seem to be related to specific features of aggressiveness in cholesteatomata, since similar values were obtained in chronic otitides. It is possible that mechanisms other than progressive growth play a decisive role in the pathogenesis of bone lysis and tissue destruction associated with cholesteatoma. Future investigations into the role of inflammation in the development of cholesteatomata and the inflammatory mediators released in the perimatrix might help elucidate the mechanisms of the clinical aggressiveness in choles-
B. Banco et al./The Veterinary Journal 200 (2014) 440–445
teatomata. However, the histological evaluation of cholesteatoma, even using specific tools such as immunohistochemical assays, cannot assess the multifaceted problem of cholesteatoma recurrence. Other factors, not least the adequacy of surgical resection, can contribute significantly to recurrence of the lesion, independent of the intrinsic proliferative capability of the epithelium composing the cholesteatoma. In our opinion, aural cholesteatoma is a complex lesion and management must integrate clinical, radiological and histological data. Conflict of interest statement None of the authors has any financial or personal relationships that could inappropriately influence or bias the content of the paper. Acknowledgements The authors received no financial support for the research, authorship, and/or publication of this article. References Blutke, A., Parzefall, B., Steger, A., Goedde, T., Hermanns, W., 2010. Inflammatory polyp in the middle ear of a dog: A case report. Veterinarni Medicina 55, 289–293. Broekaert, D., Cornille, A., Eto, H., Leigh, I., Ramaekers, F., Van Muijen, G., Coucke, P., De Bersaques, J., Kluyskens, P., Gillis, E., 1988. A comparative immunohistochemical study of cytokeratin and vimentin expression in middle ear mucosa and cholesteatoma, and in epidermis. Virchows Archiv. A, Pathological Anatomy and Histopathology 413, 39–51. Broekaert, D., Coucke, P., Leperque, S., Ramaekers, F., Van Muijen, G., Boedts, D., Leigh, I., Lane, B., 1992. Immunohistochemical analysis of the cytokeratin expression in middle ear cholesteatoma and related epithelial tissues. The Annals of Otology, Rhinology, and Laryngology 101, 931–938. Bujía, J., Schilling, V., Holly, A., Stammberger, M., Kastenbauer, E., 1993. Hyperproliferation-associated keratin expression in human middle ear cholesteatoma. Brazilian Journal of Otorhinolaryngology 113, 364–368. Bujía, J., Kim, C., Ostos, P., Kastenbauer, E., Hültner, L., 1996. Role of interleukin 6 in epithelial hyperproliferation and bone resorption in middle ear cholesteatomas. European archives of Otorhinolaryngology 253, 152–157. Caliman e Gurgel, J.D., Pereira, S.B., Alves, A.L., Ribeiro, F.Q., 2010. Hyperproliferation markers in ear canal epidermis. Brazilian Journal of Otorhinolaryngology 76, 667–671. Castelijns, F.A., Gerritsen, M.J., van Erp, P.E., van de Kerkhof, P.C., 1999. Immunohistochemical assessment of keratin 16 on paraffin-embedded sections of normal and hyperproliferative skin: Monoclonal antibodies Ks 8.12 and LL025 in a comparative study. Archives of Dermatological Research 291, 59–63. Ferlito, A., Devaney, K.O., Rinaldo, A., Milroy, C.M., Wenig, B.M., Iurato, S., McCabe, B.F., 1997. Clinicopathological consultation. Ear cholesteatoma versus cholesterol granuloma. The Annals of Otology, Rhinology, and Laryngology 106, 79–85. Friedmann, I., Arnold, W., 1993. Inflammatory diseases of the middle ear and their sequelae. In: Friedmann, I., Arnold, W. (Eds.), Pathology of the Ear, Churchill Livingstone, London, UK, pp. 88–92. Gersdorff, M.C., Debaty, M.E., Tomasi, J.P., 2006. Pathophysiology of cholesteatoma. Revue de Laryngologie Otologie Rhinologie (Bord) 127, 115–119. Greci, V., Travetti, O., Di Giancamillo, M., Lombardo, R., Giudice, C., Banco, B., Mortellaro, C.M., 2011. Middle ear cholesteatoma in 11 dogs. The Canadian Veterinary Journal 52, 631–636. Griffey, S.M., Madewell, B.R., Dairkee, S.H., Hunt, J.E., Naydan, D.K., Higgins, R.J., 1993. Immunohistochemical reactivity of basal and luminal epithelium-specific cytokeratin antibodies within normal and neoplastic canine mammary glands. Veterinary Pathology 30, 155–161. Harran, N.X., Bradley, K.J., Hetzel, N., Bowlt, K.L., Day, M.J., Barr, F., 2012. MRI findings of a middle ear cholesteatoma in a dog. Journal of the American Animal Hospital Association 48, 339–343. Hsu, S.M., Raine, L., Fanger, H., 1981. Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: A comparison between ABC and unlabeled antibody (PAP) procedures. Journal of Histochemistry and Cytochemistry 29, 577–580.
445
Ishiko, A., Dekio, I., Fujimoto, A., Kameyama, K., Sakamoto, M., Benno, Y., Amagai, M., Nishikawa, T., 2007. Abnormal keratin expression in circumscribed palmar hypokeratosis. Journal of the American Academy of Dermatology 57, 285–291. Kim, H.J., Tinling, S.P., Chole, R.A., 2002. Expression patterns of cytokeratins in cholesteatomas: Evidence of increased migration and proliferation. Journal of Korean Medical Science 17, 381–388. Koltai, P.J., Nelson, M., Castellon, R.J., Garabedian, E.N., Triglia, J.M., Roman, S., Roger, G., 2002. The natural history of congenital cholesteatoma. Archives of Otolaryngology – Head and Neck Surgery 128, 804–809. Kuijpers, W., Vennix, P.P., Peters, T.A., Ramaekers, F.C., 1996. Squamous metaplasia of the middle ear epithelium. Acta Oto-Laryngologica 116, 293–298. Lanz, O.I., Wood, B.C., 2004. Surgery of the ear and pinna. Veterinary Clinics of North America Small Animal Practice 34, 567–599. Lee, R.J., Sidey, C., Narula, A.A., James, R.F., 1994. The nature of the epithelium in acquired cholesteatoma: Part 3-cytokeratin patterns in aural epithelial and cholesteatoma cells grown in cell culture. Clinical Otolaryngology and Allied Sciences 19, 516–520. Little, C.J., Lane, J.G., Pearson, G.R., 1991a. Inflammatory middle ear disease of the dog: The pathology of otitis media. Veterinary Record 128, 293–296. Little, C.J., Lane, J.G., Gibbs, C., Pearson, G.R., 1991b. Inflammatory middle ear disease of the dog: The clinical and pathological features of cholesteatoma, a complication of otitis media. Veterinary Record 128, 319–322. Makino, K., Amatsu, M., 1986. Epithelial migration on the tympanic membrane and external canal. Archives of Otolaryngology 243, 39–42. Mallet, Y., Nouwen, J., Lecomte-Houcke, M., Desaulty, A., 2003. Aggressiveness and quantification of epithelial proliferation of middle ear cholesteatoma by MIB1. The Laryngoscope 113, 328–331. Moll, R., Franke, W.W., Schiller, D.L., Geiger, B., Krepler, R., 1982. The catalog of human cytokeratins: Patterns of expression in normal epithelia, tumors and cultured cells. Cell 31, 11–24. Müller, J., 1838. Von der geschichteten perlmutter glanzenden Fettgeschwulst. In: Reimer, G. (Ed.), Ueber den feineren Bau und die formen der krankhaften Geschwulste, Berlin, Germany, pp. 49–54. Olszewska, E., Chodynicki, S., Chyczewski, L., 2003. Some markers of proliferative activity in cholesteatoma epithelium in adults. Otolaryngologia Polska. the Polish Otolaryngology 57, 85–89. Olszewska, E., Lautermann, J., Koc, C., Schwaab, M., Dazert, S., Hildmann, H., Sudhoff, H., 2005. Cytokeratin expression pattern in congenital and acquired pediatric cholesteatoma. European Archives of Otorhinolaryngology 262, 731–736. Olszewska, E., Sudhoff, H., 2007. Comparative cytokeratin distribution patterns in cholesteatoma epithelium. Histology and Histopathology 22, 37–42. Palva, T., Palva, A., Dammert, K., 1972. Keratin in middle ear squamous epithelium. Archives of Otolaryngology 95, 214–220. Peck, G.L., 1977. Cholesteatoma. In: McCabe, B.F., Sade, J., Abramson, M. (Eds), 1st International Conference. Birmingham, AL, 45. Persaud, R., Hajioff, D., Trinidade, A., Khemani, S., Bhattacharyya, M.N., Papadimitriou, N., Kalan, A., Bhattacharyya, A.K., 2007. Evidence-based review of aetiopathogenic theories of congenital and acquired cholesteatoma. The Journal of Laryngology and Otology 121, 1013–1109. Sadé, J., 1980. Retraction pockets and attic cholesteatomas. Acta Oto-RhinoLaryngologica Belgica 34, 62–84. Semaan, M.T., Megerian, C.A., 2006. The pathophysiology of cholesteatoma. Otolaryngologic Clinics of North America 39, 1143–1159. Strunk, C.L., 1993. Cholesteatoma. In: Bailey, B.J., Johnson, J.T., Newlands, S.D. (Eds.), Head and Neck Surgery-Otolaryngology, vol. 2. Lippincott Williams and Wilkins, Philadelphia, PA, USA, pp. 1635–1646. Sudhoff, H., Bujía, J., Fisseler-Eckhoff, A., Holly, A., Schulz-Flake, C., Hildmann, H., 1995. Expression of a cell-cycle-associated nuclear antigen (MIB 1) in cholesteatoma and auditory meatal skin. The Laryngoscope 105, 1227–1231. Travetti, O., Giudice, C., Greci, V., Lombardo, R., Mortellaro, C.M., Di Giancamillo, M., 2010. Computed tomography features of middle ear cholesteatoma in dogs. Veterinary Radiology and Ultrasound 51, 374–379. Venker-Van Haagen, A.J., 2005. The ear. In: Ear, Nose, Throat and Tracheobronchial Diseases in Dogs and Cats, 1st ed. Schlütersche, Verlagsgesellschaft mbH, Hannover, Germany, pp. 1–50. Vennix, P.P., Kuijpers, W., Peters, T.A., Tonnaer, E.L., Ramaekers, F.C., 1996. Keratinocyte differentiation in acquired cholesteatoma and perforated tympanic membranes. Archives of Otolaryngology – Head and Neck Surgery 122, 825–832. Walter, J.H., 2001. Cytokeratins in the canine epidermis. Veterinary Dermatology 12, 81–87. Weiss, R.A., Eichner, R., Sun, T.T., 1984. Monoclonal antibody analysis of keratin expression in epidermal diseases: A 48- and 56-kdalton keratin as molecular markers for hyperproliferative keratinocytes. The Journal of Cell Biology 98, 1397–1406. Yasuno, K., Nishiyama, S., Suetsugu, F., Ogihara, K., Madarame, H., Shirota, K., 2009. Cutaneous clear cell adnexal carcinoma in a dog: Special reference to cytokeratin expression. Journal of Veterinary Medical Science 71, 1513–1517.
