- Email: [email protected]
Comparative Histopathological Study of Sporotrichosis and American Tegumentary Leishmaniosis in Dogs from Rio de Janeiro
J. Comp. Path. 2010, Vol. 143, 1e7
Available online at www.sciencedirect.com
www.elsevier.com/locate/jcpa
Comparative Histopathological Study of Sporotrichosis and American Tegumentary Leishmaniosis in Dogs from Rio de Janeiro L. H. M. Miranda*, L. P. Quintella†, I. B. Santos*, R. V. C. Oliveira‡, R. C. Menezes*, F. B. Figueiredo* and T. M. P. Schubach* *Laborato´rio de Pesquisa Clı´nica em Dermatozoonoses em Animais Dome´sticos, † Servic¸o de Anatomia Patolo´gica and ‡ Laborato´rio de Epidemiologia Clı´nica, Instituto de Pesquisa Clı´nica Evandro Chagas/Fundac¸a˜o Oswaldo Cruz (IPEC/FIOCRUZ), Av. Brasil 4365, Manguinhos, Rio de Janeiro, Brazil
Summary Pyogranulomatous lesions from 80 dogs with sporotrichosis and 26 dogs with American tegumentary leishmaniosis (ATL) were compared microscopically in order to identify features that would support the diagnostic suspicion and direct the subsequent search for the aetiological agent of either infection. Odds ratios and their respective 95% confidence intervals were calculated in order to evaluate the impact of the microscopical findings on the diagnosis of either disease. Lesions with well-formed granulomata were 14 times more likely to be due to sporotrichosis than ATL. Marked neutrophil infiltration into granulomata was 12.26 times more likely to be associated with sporotrichosis when compared with lesions having mild neutrophilic infiltration. Absence of lymphocytes and macrophages in the peripheral infiltrate was associated with a 9.71 and 4.93 higher chance, respectively, of being sporotrichosis rather than ATL compared with lesions where these cells were present. Lesions with a perivascular, perifollicular and interstitial peripheral inflammatory infiltrate were 5.48 times more likely to be due to sporotrichosis than ATL when compared with lesions with a diffuse peripheral infiltrate. Histopathological analysis may therefore contribute to the diagnosis of sporotrichosis or ATL skin lesions in dogs since this method permits the identification of features that direct the diagnostic suspicion, thus facilitating the search for the aetiological agent in histological sections, permitting the precise request of subsequent tests and thereby reducing costs and time taken to achieve a definitive diagnosis and the initiation of appropriate therapy. Ó 2009 Elsevier Ltd. All rights reserved. Keywords: American tegumentary leishmaniosis; dog; histopathology; sporotrichosis
Introduction Sporotrichosis is a disease of man and various animal species that is caused by infection with the dimorphic fungus Sporothrix schenckii (Schell, 1998). Canine sporotrichosis is considered to be unusual, but an increase in the number of cases has been observed in Rio de Janeiro since 1998. This increase is related to the epidemic situation of sporotrichosis in the region, with the disease involving people as well as dogs and cats (Schubach et al., 2006). Correspondence to: L. H. M. Miranda (e-mail: luisahmiranda@gmail. com). 0021-9975/$ - see front matter doi:10.1016/j.jcpa.2009.12.010
American tegumentary leishmaniosis (ATL) is also endemic in dogs in Rio de Janeiro and is currently the main differential diagnosis for canine sporotrichosis. Besides being clinically similar, the two diseases overlap in their geographical distribution and present cross-reactivity in serological tests (Santos et al., 2007). Demonstration of the aetiological agent is necessary to establish a definitive diagnosis of sporotrichosis or ATL, with culture isolation of S. schenckii and Leishmania spp., respectively, being the standard method (Santos et al., 2007). Histopathological analysis of biopsies of lesional skin is a complementary diagnostic tool. The cutaneous lesions of canine sporotrichosis and ATL are similar, with both diseases being Ó 2009 Elsevier Ltd. All rights reserved.
2
L.H.M. Miranda et al.
characterized by a pyogranulomatous inflammatory process (Schubach et al., 2006; Santos et al., 2007). Histochemical staining techniques such as Grocott’s silver staining and periodic acid-Schiff (PAS) permit the visualization of fungal structures in histological sections (Kwon-Chung and Bennett, 1992), but this is not always possible because S. schenckii yeast are sparse in the lesions of canine sporotrichosis (Schubach et al., 2006; Santos et al., 2007). The diagnosis of leishmaniosis can be made by observation of the amastigote forms of Leishmania spp. in sections of lesional skin. However, due to their scarcity, these parasitic forms are not always detected and their identification requires experience (Sotto et al., 1989). The application of immunological methods such as immunohistochemistry (IHC) enhances the histological diagnosis of sporotrichosis and ATL (Sells and Burton, 1981; Salinas et al., 1989; Marques et al., 1992; Quintella et al., 2009); however, these methods are costly and are only indicated after the establishment of a diagnostic suspicion. The aim of the present study was to compare the histopathological findings of pyogranulomatous lesions from dogs with sporotrichosis and ATL in order to identify features that might support the diagnostic suspicion and direct the subsequent search for the aetiological agent.
Materials and Methods Samples
Paraffin wax-embedded tissue specimens of skin lesions from dogs with sporotrichosis or ATL, stored in the archives of the Pathological Anatomy Service Servic¸o de Anatomia Patolo´gica (SEAP), were studied. The dogs were from endemic areas in the State of Rio de Janeiro and were seen at the Laborato´rio de Pesquisa Clı´ nica em Dermatozoonoses em Animais Dome´sticos (LAPCLIN-DERMZOO), Instituto de Pesquisa Clı´ nica Evandro Chagas (IPEC), Fundac¸a˜o Oswaldo Cruz (Fiocruz), between 2001 and 2008. The skin lesions were divided into two groups. Group I consisted of lesions from 90 dogs with sporotrichosis in which diagnosis was confirmed by the isolation of S. schenckii in mycological culture. Group II consisted of lesions from 35 dogs with ATL in which diagnosis was confirmed by the isolation and characterization of Leishmania (Viannia) braziliensis. On microscopical examination, 80 (88.9%) cases of sporotrichosis had pyogranulomatous inflammatory lesions, nine cases (10.0%) had lesions with nonspecific inflammation and in one case (1.1%) the lesions were of non-suppurative granulomatous inflammation. In dogs with ATL, 26 cases (74.3%)
had pyogranulomatous inflammatory lesions and nine cases (25.7%) had lesions with non-specific inflammation. Only those lesions with pyogranulomatous inflammation were selected for further analysis. Data from the dogs with sporotrichosis have been partially reported in a previous descriptive histopathological investigation (Miranda et al., 2009). Histopathological Analysis
Sections (5 mm) were stained with haematoxylin and eosin (HE) for microscopical analysis. Specific features of the pyogranulomatous inflammatory infiltrates were examined and graded (Adams, 1974). The features of any non-granulomatous ‘peripheral’ inflammatory infiltrate were also classified (Ackerman et al., 1997). These features and their grading are summarized in Table 1. Data Analysis
Data were analyzed using the Statistical Package for the Social Sciences (SPSS) software (version 16.0). The frequencies of the following variables were determined for each group: granuloma differentiation, granuloma organization, intensity of the neutrophil infiltration into the granuloma, distribution of granulomatous and non-granulomatous peripheral infiltrates and cell types present in the infiltrate. Odds ratios (ORs) and their respective 95% confidence intervals (CIs) were calculated to evaluate the impact of the microscopical findings observed on the diagnosis of either disease. A level of significance of 5% was considered for analysis.
Results A summary of the microscopical analysis of the lesions of sporotrichosis and ATL is presented in Table 2. The impact of the microscopical findings on the diagnosis of either disease, reported as ORs, their respective 95% CIs and the P value is shown in Table 3. Well-formed granulomata (Fig. 1) were present in the lesional skin of 37.5% of cases of sporotrichosis and poorly formed granulomata (Figs. 2, 3) were present in the lesional skin of 96.2% of cases of ATL. Therefore, by OR, the presence of well-formed granulomata indicates a 14 times greater risk of sporotrichosis than ATL when compared with lesions containing poorly formed granulomata. Epithelioid cells dominated in the granulomata of dogs with sporotrichosis (73.8%) and ATL (69.2%). Multinucleate giant cells were seen in only one case of sporotrichosis. Granulomata were diffusely distributed in the lesional skin of dogs with both sporotrichosis (71.3%) and ATL (61.5%). An interstitial granulomatous
3
Pathology of Sporotrichosis and Leishmaniosis Table 1 Definition of histopathological terms Term
Categories
Definition
Mild
Inflammatory infiltrates contain activated cells of the MPS forming compact aggregates or interstitial cords Granuloma with prominent neutrophil infiltration Infiltration of neutrophils into the granuloma is scattered or moderately dense Infiltration of neutrophils into the granuloma is dense and more cellular than that of mild intensity infiltrates
Granuloma/granulomatous infiltrate Pyogranuloma Intensity of neutrophil infiltration of a granuloma
Marked Activated cells of the MPS in granulomas
Macrophage
Large polygonal cells with plump eccentric nuclei and pale eosinophilic, slightly granular, cytoplasm with distinct borders Large elongate cells with eccentric reniform nuclei and abundant pale eosinophilic cytoplasm with indistinct borders
Epithelioid cells Granuloma organization Distribution of the granulomatous infiltrate
Well formed Poorly formed
Well delineated with a predominance of epithelioid cells Poorly delineated and/or a predominance of macrophages
Interstitial Nodular Diffuse
Cells of the MPS scattered between collagen bundles Infiltration of cells of the MPS forming nodules Dense dermal infiltrate that destroys or obscures existing structures
Peripheral infiltrate Distribution of the peripheral infiltrate
Non-granulomatous infiltrate found outside the granulomata Perivascular, perifollicular and interstitial Diffuse
Infiltration around venules, hair follicles and scattered between bundles of collagen Dense dermal infiltrate that destroys or obscures existing structures
MPS, mononuclear phagocyte system. Table 2 Summary of frequencies of the main microscopical features in skin lesions of sporotrichosis and ATL Feature
Category
Sporotrichosis (% of cases)
ATL (% of cases)
Well formed Poorly formed
37.5 62.5
3.8 96.2
Macrophages Epithelioid cells
26.2 73.8
30.8 69.2
Interstitial Nodular Diffuse
5.0 23.7 71.3
0 38.5 61.5
Multinucleated giant cells
Absent Present
99.2 0.8
100.0 0
Intensity of neutrophil infiltration into granulomata
Mild Marked
47.5 52.5
92.3 7.7
Plasma cells around granulomata
Absent Present
10.0 90.0
11.5 88.5
Absent Perivascular, perifollicular and interstitial Diffuse
16.2 70.0
3.8 42.3
13.8
53.9
Plasma cells in the peripheral infiltrate
Absent Present
16.2 83.8
3.8 96.2
Lymphocytes in the peripheral infiltrate
Absent Present
30.0 70.0
3.8 96.2
Neutrophils in the peripheral infiltrate
Absent Present
53.7 46.3
34.6 65.4
Macrophages in the peripheral infiltrate
Absent Present
72.5 27.5
30.8 69.2
Granuloma organization Predominant differentiation of the activated cells of the MPS in granulomata Distribution of the granulomatous infiltrate
Peripheral infiltrate
MPS, mononuclear phagocyte system.
4
L.H.M. Miranda et al. Table 3 Statistical analysis of the main microscopical features in skin lesions of sporotrichosis and ATL Categories
OR*
CIeOR (95%)
P value‡
Granuloma organization
Well organized† Poorly organized
15.00 1.00
1.93e116.45
0.001
Predominant differentiation of the mononuclear phagocytes in granulomata
Macrophages† Epithelioid cells
0.80 e
0.30e2.11 e
0.800
Intensity of neutrophil infiltration into granulomata
Marked† Mild
13.26 1.00
2.94e59.90
<0.001
Plasma cells around granulomata
Absent† Present
0.85 e
0.21e3.50 e
1.000
Plasma cells in the peripheral infiltrate
Absent† Present
4.85 1.00
0.60e39.03 e
0.180
Lymphocytes in the peripheral infiltrate
Absent† Present
10.71 1.00
1.37e83.66 e
0.007
Neutrophils in the peripheral infiltrate
Absent† Present
2.20 1.00
0.88e5.51 e
0.115
Macrophages in the peripheral infiltrate
Absent† Present
5.93 1.00
2.26e15.60 e
<0.001
Perivascular, perifollicular and interstitial† Diffuse
6.48
2.33e17.98
<0.001
1.00
e
Variable
Distribution of the peripheral infiltrate
*
Reference category: ATL. Reference category of each variable. ‡ P < 0.05 indicates a significant association. †
infiltrate (Fig. 4) was observed in 5.0% of skin biopsies from dogs with sporotrichosis, but was not observed in lesional skin from dogs with ATL. The intensity of neutrophil infiltration into pyogranulomata was marked in 52.5% of cases of sporotrichosis. In contrast, mild neutrophilic infiltration into pyogranulomata was observed in 92.3% of cases of ATL. The presence of marked neutrophilic infiltra-
tion was 12.26 fold more likely to be found in the lesions of sporotrichosis than ATL. The distribution of the peripheral inflammatory infiltration was perivascular, perifollicular and interstitial (Fig. 5) in 70.0% of cases of sporotrichosis. In contrast, this infiltration was diffuse (Fig. 6) in 53.9% of cases of ATL. Lesions with a perivascular, perifollicular and interstitial peripheral inflammatory
Fig. 1. Skin from a dog with sporotrichosis in which there are wellformed granulomata that contain neutrophils (arrows). There is an outer zone of plasma cells (dashed arrows). HE. 200.
Fig. 2. Skin from a dog with ATL in which there is a diffuse granulomatous infiltrate. A poorly formed granuloma is outlined. HE. 100.
Pathology of Sporotrichosis and Leishmaniosis
Fig. 3. Skin from a dog with ATL showing a granuloma predominantly composed of macrophages. HE. 200.
infiltrate therefore had a 5.48 higher chance of being sporotrichosis than ATL when compared with lesions with a diffuse peripheral infiltrate. Lymphocytes were present in the peripheral infiltrate in 70.0% of cases of sporotrichosis cases and in 96.2% of cases of ATL. Macrophages were present in the peripheral infiltrate in 27.5% of cases of sporotrichosis and in 69.2% of cases of ATL. The absence of lymphocytes and macrophages in the peripheral infiltrate was associated with a 9.71 and 4.93 fold greater risk of the lesion being due to sporotrichosis, respectively, when compared with lesions where these cells were present.
5
Fig. 5. Skin from a dog with sporotrichosis. There is a perifollicular infiltrate with plasma cells surrounding hair follicles. HE. 200.
Systematic microscopical analysis of the skin lesions of the dogs studied here showed that, although the defin-
itive diagnosis depends on demonstration of the aetiological agent, the presence of histopathological characteristics that favour the diagnosis of either sporotrichosis or ATL might be used to support the diagnostic suspicion. In the present study, histopathological analysis showed a predominance of poorly formed granulomata in both diseases. Nevertheless, the presence of well-formed granulomata is more likely to be associated with sporotrichosis in dogs (OR ¼ 15.00) when compared with ATL. However, the lack of this feature does not exclude a diagnosis of sporotrichosis. A marked intensity of neutrophil infiltration into the granulomata was also associated with the lesions of sporotrichosis and a mild intensity of such infiltration was more common in cases of ATL. Therefore,
Fig. 4. Skin from a dog with sporotrichosis showing interstitial granulomatous inflammation. Activated cells of the mononuclear phagocyte system (MPS) are scattered between collagen bundles (arrows). HE. 200.
Fig. 6. Skin from a dog with ATL. There is a poorly formed granuloma (arrows) and a diffuse non-granulomatous infiltrate found at the periphery (dashed arrows). HE. 200. Inset: activated cells of the MPS forming granulomata. HE. 400.
Discussion
6
L.H.M. Miranda et al.
the observation of pyogranulomata rich in neutrophils is more likely to be associated with sporotrichosis than ATL in dogs (OR ¼ 13.26). The distribution of the peripheral inflammatory infiltrate was predominantly perivascular, perifollicular and interstitial in the lesions of dogs with sporotrichosis, whereas it was predominantly diffuse in cases of ATL. Statistical analysis suggested that the diagnosis of canine ATL may be favoured when the peripheral inflammatory infiltrate is diffuse (OR ¼ 6.48). Moreover, analysis suggested that canine ATL is more likely where lesions have lymphocytes (OR ¼ 10.71) and macrophages (OR ¼ 5.93) in peripheral infiltrate. It should be noted that poor delineation of granulomata and the presence of macrophages in the diffuse peripheral infiltrates of canine ATL resulted in imprecise distinction between granulomatous and non-granulomatous infiltrates. Therefore, because of the subjectivity of morphological analysis, further investigation is necessary to confirm the reliability of these findings (macrophages in the peripheral infiltrate and diffuse distribution of the peripheral infiltrate) in distinguishing the two diseases. Although culture of the aetiological agent is the method of choice for the diagnosis of sporotrichosis or ATL, this test may not be performed or may not be available and, even if performed, no growth of the microorganism might be observed, generally because of methodological limitations (Moore and Ackerman, 1946; Schwarz, 1982). In addition, maintenance of the culture for a long period of time might be necessary (Schwarz, 1982). Thus, the differential diagnosis between sporotrichosis and ATL is frequently made based on clinicaleepidemiological evidence and serological tests (Barros et al., 2005; Santos et al., 2007). Clinically, the skin lesions of sporotrichosis and ATL have similar characteristics in dogs (Santos et al., 2007). Higher frequencies of multiple lesions and respiratory signs have been reported for animals with sporotrichosis (Schubach et al., 2006; Santos et al., 2007). Furthermore, the duration of the lesions seems to be shorter in canine sporotrichosis (Schubach et al., 2006) compared with dogs experimentally infected with ATL (Pirmez et al., 1988). With respect to serological tests, a significant rate of positive serology for leishmaniosis has been observed in dogs with sporotrichosis (Santos et al., 2007). Histopathological examination is a complementary diagnostic tool and might be the only alternative available for the identification of the aetiological agent, particularly in situations in which no specimens are collected for culture of the microorganism (Moskowitz et al., 1986; Werner and Werner, 1994; Rodrı´ guez and Sarmiento, 1998; Barros et al., 2005). The results of the present study suggest that
histopathological analysis may significantly contribute to the differential diagnosis between canine cutaneous sporotrichosis and ATL. This method detects features that directs the diagnostic suspicion and guides the search for the aetiological agent in histological sections, permitting the request of appropriate subsequent tests, reducing costs and the time taken to achieve a definitive diagnosis and initiate appropriate therapy.
Acknowledgments The authors thank I. D. F. Gremia˜o, S. A. Pereira, T. Okamoto, M. F. L. Almeida, M. D. P. E. Espinoza, F. C. C. Rodrigues, A. Correia, A. C. Silva, R. S. Reis, P. C. F. Monteiro, M. F. Madeira and R. Me´xas for technical support. This study was partially financed by the Programa de Apoio a Pesquisa Estrate´gica em Sau´de (PAPES IV), FIOCRUZ, by the Fundac¸a˜o Carlos Chagas Filho de Amparo a` Pesquisa do Estado do Rio de Janeiro (FAPERJ), and by the Conselho Nacional de Desenvolvimento Cientı´ fico Tecnolo´gico (CNPq), Brazil. TMPS is a CNPq researcher. The study was approved by the Ethics Committee on the Use of Animals, FIOCRUZ.
References Ackerman AB, Chongchitnant N, Sanchez J, Guo Y, Bennin B et al. (1997) Proceeding to specific diagnosis. In: Histologic Diagnosis of Inflammatory Skin Diseases: an Algorithmic Method Based on Pattern Analysis, 2nd Edit., G Stamathis, Ed., Williams & Wilkins, Baltimore, pp. 107e167. Adams DO (1974) The structure of mononuclear phagocytes differentiation in vivo e I. Sequential fine histological studies of the effect of Bacillus Calmette-Guerin (BCG). American Journal of Pathology, 76, 17e39. Barros MBL, Schubach AO, do Valle ACF, Galhardo MCG, Schubach TMP et al. (2005) Positive Montenegro skin test among patients with sporotrichosis in Rio de Janeiro. Acta Tropica, 93, 41e47. Kwon-Chung K, Bennett J (1992) Sporotrichosis. In: Medical Mycology, 2nd Edit., K Kwon-Chung, J Bennett, Eds, Lea & Febiger, Philadelphia, pp. 707e729. Marques MEA, Coelho KIR, Sotto MN, Bacchi CE (1992) Comparison between histochemical and immunohistochemical methods for diagnosis of sporotrichosis. Journal of Clinical Pathology, 45, 1089e1093. Miranda LHM, Quintella LP, Santos IB, Menezes RC, Figueiredo FB et al. (2009) Histopathology of canine sporotrichosis: a morphological study of 86 cases from Rio de Janeiro (2001e2007). Mycopathologia, 168, 79e87. Moore M, Ackerman LV (1946) Sporotrichosis with radiate formation in tissue: report of a case. Archives of Dermatology and Syphilology, 53, 253e264. Moskowitz LB, Ganjei P, Ziegels-Weissman J, Clearly TJ, Penneys NS et al. (1986) Immunohistological
Pathology of Sporotrichosis and Leishmaniosis
identification of fungi in systemic and cutaneous mycoses. Archives of Pathology & Laboratory Medicine, 110, 433e436. Pirmez C, Marzochi MC, Coutinho SG (1988) Experimental canine mucocutaneous leishmaniasis (Leishmania braziliensis braziliensis). Memo´rias do Instituto Oswaldo Cruz, 83, 145e151. Quintella LP, Cuzzi T, Madeira MF, Okamoto T, Schubach AO (2009) Immunoperoxidase technique using an anti-Leishmania (L.) chagasi hyperimmune serum in the diagnosis of culture-confirmed American tegumentary leishmaniasis. Revista do Instituto de Medicina Tropical de Sa˜o Paulo, 51, 83e86. Rodrı´ guez G, Sarmiento L (1998) The asteroid bodies of sporotrichosis. American Journal of Dermatopathology, 20, 246e249. Salinas G, Valderrama L, Palma G, Montes G, Saravia NG (1989) Deteccion de amastigotas en leishmaniasis cutanea y mucocutanea por el metodo de inmunoperoxidasa, usando anticuerpo policlonal: sensibilidad y especificidad comparadas con metodos convencionales de diagnostico. Memo´rias do Instituto Oswaldo Cruz, 84, 53e60. Santos IB, Schubach TMP, Leme LRP, Okamoto T, Figueiredo FB et al. (2007) Sporotrichosis e the main differential diagnosis with tegumentary leishmaniosis in dogs from Rio de Janeiro, Brazil. Veterinary Parasitology, 143, 1e6.
7
Schell WA (1998) Agents of chromoblastomycosis and sporotrichosis. In: Topley & Wilson’s Microbiology and Microbial Infections, 9th Edit., L Allejo, RJ Hay, Eds, Arnold, London, pp. 315e336. Schubach TMP, Schubach AO, Okamoto T, Barros MBL, Figueiredo FB et al. (2006) Canine sporotrichosis in Rio de Janeiro, Brazil: clinical presentation, laboratory diagnosis and therapeutic response in 44 cases (1998e2003). Medical Mycology, 44, 87e92. Schwarz J (1982) The diagnosis of deep mycoses by morphologic methods. Human Pathology, 13, 519e533. Sells PG, Burton M (1981) Identification of Leishmania amastigotes and their antigens in formalin fixed tissue by immunoperoxidase staining. Transactions of the Royal Society of Tropical Medicine and Hygiene, 75, 461e468. Sotto MN, Yamashiro-Kanashiro EH, Da Matta VLR, Brito T (1989) Cutaneous leishmaniasis of the New World: diagnostic immunopathology and antigen pathway in skin and mucosa. Acta Tropica, 46, 121e130. Werner AH, Werner BE (1994) Sporotrichosis in man and animal. International Journal of Dermatology, 33, 692e700.
July 15th, 2009 ½ Received, Accepted, December 22nd, 2009
Available online at www.sciencedirect.com
www.elsevier.com/locate/jcpa
Comparative Histopathological Study of Sporotrichosis and American Tegumentary Leishmaniosis in Dogs from Rio de Janeiro L. H. M. Miranda*, L. P. Quintella†, I. B. Santos*, R. V. C. Oliveira‡, R. C. Menezes*, F. B. Figueiredo* and T. M. P. Schubach* *Laborato´rio de Pesquisa Clı´nica em Dermatozoonoses em Animais Dome´sticos, † Servic¸o de Anatomia Patolo´gica and ‡ Laborato´rio de Epidemiologia Clı´nica, Instituto de Pesquisa Clı´nica Evandro Chagas/Fundac¸a˜o Oswaldo Cruz (IPEC/FIOCRUZ), Av. Brasil 4365, Manguinhos, Rio de Janeiro, Brazil
Summary Pyogranulomatous lesions from 80 dogs with sporotrichosis and 26 dogs with American tegumentary leishmaniosis (ATL) were compared microscopically in order to identify features that would support the diagnostic suspicion and direct the subsequent search for the aetiological agent of either infection. Odds ratios and their respective 95% confidence intervals were calculated in order to evaluate the impact of the microscopical findings on the diagnosis of either disease. Lesions with well-formed granulomata were 14 times more likely to be due to sporotrichosis than ATL. Marked neutrophil infiltration into granulomata was 12.26 times more likely to be associated with sporotrichosis when compared with lesions having mild neutrophilic infiltration. Absence of lymphocytes and macrophages in the peripheral infiltrate was associated with a 9.71 and 4.93 higher chance, respectively, of being sporotrichosis rather than ATL compared with lesions where these cells were present. Lesions with a perivascular, perifollicular and interstitial peripheral inflammatory infiltrate were 5.48 times more likely to be due to sporotrichosis than ATL when compared with lesions with a diffuse peripheral infiltrate. Histopathological analysis may therefore contribute to the diagnosis of sporotrichosis or ATL skin lesions in dogs since this method permits the identification of features that direct the diagnostic suspicion, thus facilitating the search for the aetiological agent in histological sections, permitting the precise request of subsequent tests and thereby reducing costs and time taken to achieve a definitive diagnosis and the initiation of appropriate therapy. Ó 2009 Elsevier Ltd. All rights reserved. Keywords: American tegumentary leishmaniosis; dog; histopathology; sporotrichosis
Introduction Sporotrichosis is a disease of man and various animal species that is caused by infection with the dimorphic fungus Sporothrix schenckii (Schell, 1998). Canine sporotrichosis is considered to be unusual, but an increase in the number of cases has been observed in Rio de Janeiro since 1998. This increase is related to the epidemic situation of sporotrichosis in the region, with the disease involving people as well as dogs and cats (Schubach et al., 2006). Correspondence to: L. H. M. Miranda (e-mail: luisahmiranda@gmail. com). 0021-9975/$ - see front matter doi:10.1016/j.jcpa.2009.12.010
American tegumentary leishmaniosis (ATL) is also endemic in dogs in Rio de Janeiro and is currently the main differential diagnosis for canine sporotrichosis. Besides being clinically similar, the two diseases overlap in their geographical distribution and present cross-reactivity in serological tests (Santos et al., 2007). Demonstration of the aetiological agent is necessary to establish a definitive diagnosis of sporotrichosis or ATL, with culture isolation of S. schenckii and Leishmania spp., respectively, being the standard method (Santos et al., 2007). Histopathological analysis of biopsies of lesional skin is a complementary diagnostic tool. The cutaneous lesions of canine sporotrichosis and ATL are similar, with both diseases being Ó 2009 Elsevier Ltd. All rights reserved.
2
L.H.M. Miranda et al.
characterized by a pyogranulomatous inflammatory process (Schubach et al., 2006; Santos et al., 2007). Histochemical staining techniques such as Grocott’s silver staining and periodic acid-Schiff (PAS) permit the visualization of fungal structures in histological sections (Kwon-Chung and Bennett, 1992), but this is not always possible because S. schenckii yeast are sparse in the lesions of canine sporotrichosis (Schubach et al., 2006; Santos et al., 2007). The diagnosis of leishmaniosis can be made by observation of the amastigote forms of Leishmania spp. in sections of lesional skin. However, due to their scarcity, these parasitic forms are not always detected and their identification requires experience (Sotto et al., 1989). The application of immunological methods such as immunohistochemistry (IHC) enhances the histological diagnosis of sporotrichosis and ATL (Sells and Burton, 1981; Salinas et al., 1989; Marques et al., 1992; Quintella et al., 2009); however, these methods are costly and are only indicated after the establishment of a diagnostic suspicion. The aim of the present study was to compare the histopathological findings of pyogranulomatous lesions from dogs with sporotrichosis and ATL in order to identify features that might support the diagnostic suspicion and direct the subsequent search for the aetiological agent.
Materials and Methods Samples
Paraffin wax-embedded tissue specimens of skin lesions from dogs with sporotrichosis or ATL, stored in the archives of the Pathological Anatomy Service Servic¸o de Anatomia Patolo´gica (SEAP), were studied. The dogs were from endemic areas in the State of Rio de Janeiro and were seen at the Laborato´rio de Pesquisa Clı´ nica em Dermatozoonoses em Animais Dome´sticos (LAPCLIN-DERMZOO), Instituto de Pesquisa Clı´ nica Evandro Chagas (IPEC), Fundac¸a˜o Oswaldo Cruz (Fiocruz), between 2001 and 2008. The skin lesions were divided into two groups. Group I consisted of lesions from 90 dogs with sporotrichosis in which diagnosis was confirmed by the isolation of S. schenckii in mycological culture. Group II consisted of lesions from 35 dogs with ATL in which diagnosis was confirmed by the isolation and characterization of Leishmania (Viannia) braziliensis. On microscopical examination, 80 (88.9%) cases of sporotrichosis had pyogranulomatous inflammatory lesions, nine cases (10.0%) had lesions with nonspecific inflammation and in one case (1.1%) the lesions were of non-suppurative granulomatous inflammation. In dogs with ATL, 26 cases (74.3%)
had pyogranulomatous inflammatory lesions and nine cases (25.7%) had lesions with non-specific inflammation. Only those lesions with pyogranulomatous inflammation were selected for further analysis. Data from the dogs with sporotrichosis have been partially reported in a previous descriptive histopathological investigation (Miranda et al., 2009). Histopathological Analysis
Sections (5 mm) were stained with haematoxylin and eosin (HE) for microscopical analysis. Specific features of the pyogranulomatous inflammatory infiltrates were examined and graded (Adams, 1974). The features of any non-granulomatous ‘peripheral’ inflammatory infiltrate were also classified (Ackerman et al., 1997). These features and their grading are summarized in Table 1. Data Analysis
Data were analyzed using the Statistical Package for the Social Sciences (SPSS) software (version 16.0). The frequencies of the following variables were determined for each group: granuloma differentiation, granuloma organization, intensity of the neutrophil infiltration into the granuloma, distribution of granulomatous and non-granulomatous peripheral infiltrates and cell types present in the infiltrate. Odds ratios (ORs) and their respective 95% confidence intervals (CIs) were calculated to evaluate the impact of the microscopical findings observed on the diagnosis of either disease. A level of significance of 5% was considered for analysis.
Results A summary of the microscopical analysis of the lesions of sporotrichosis and ATL is presented in Table 2. The impact of the microscopical findings on the diagnosis of either disease, reported as ORs, their respective 95% CIs and the P value is shown in Table 3. Well-formed granulomata (Fig. 1) were present in the lesional skin of 37.5% of cases of sporotrichosis and poorly formed granulomata (Figs. 2, 3) were present in the lesional skin of 96.2% of cases of ATL. Therefore, by OR, the presence of well-formed granulomata indicates a 14 times greater risk of sporotrichosis than ATL when compared with lesions containing poorly formed granulomata. Epithelioid cells dominated in the granulomata of dogs with sporotrichosis (73.8%) and ATL (69.2%). Multinucleate giant cells were seen in only one case of sporotrichosis. Granulomata were diffusely distributed in the lesional skin of dogs with both sporotrichosis (71.3%) and ATL (61.5%). An interstitial granulomatous
3
Pathology of Sporotrichosis and Leishmaniosis Table 1 Definition of histopathological terms Term
Categories
Definition
Mild
Inflammatory infiltrates contain activated cells of the MPS forming compact aggregates or interstitial cords Granuloma with prominent neutrophil infiltration Infiltration of neutrophils into the granuloma is scattered or moderately dense Infiltration of neutrophils into the granuloma is dense and more cellular than that of mild intensity infiltrates
Granuloma/granulomatous infiltrate Pyogranuloma Intensity of neutrophil infiltration of a granuloma
Marked Activated cells of the MPS in granulomas
Macrophage
Large polygonal cells with plump eccentric nuclei and pale eosinophilic, slightly granular, cytoplasm with distinct borders Large elongate cells with eccentric reniform nuclei and abundant pale eosinophilic cytoplasm with indistinct borders
Epithelioid cells Granuloma organization Distribution of the granulomatous infiltrate
Well formed Poorly formed
Well delineated with a predominance of epithelioid cells Poorly delineated and/or a predominance of macrophages
Interstitial Nodular Diffuse
Cells of the MPS scattered between collagen bundles Infiltration of cells of the MPS forming nodules Dense dermal infiltrate that destroys or obscures existing structures
Peripheral infiltrate Distribution of the peripheral infiltrate
Non-granulomatous infiltrate found outside the granulomata Perivascular, perifollicular and interstitial Diffuse
Infiltration around venules, hair follicles and scattered between bundles of collagen Dense dermal infiltrate that destroys or obscures existing structures
MPS, mononuclear phagocyte system. Table 2 Summary of frequencies of the main microscopical features in skin lesions of sporotrichosis and ATL Feature
Category
Sporotrichosis (% of cases)
ATL (% of cases)
Well formed Poorly formed
37.5 62.5
3.8 96.2
Macrophages Epithelioid cells
26.2 73.8
30.8 69.2
Interstitial Nodular Diffuse
5.0 23.7 71.3
0 38.5 61.5
Multinucleated giant cells
Absent Present
99.2 0.8
100.0 0
Intensity of neutrophil infiltration into granulomata
Mild Marked
47.5 52.5
92.3 7.7
Plasma cells around granulomata
Absent Present
10.0 90.0
11.5 88.5
Absent Perivascular, perifollicular and interstitial Diffuse
16.2 70.0
3.8 42.3
13.8
53.9
Plasma cells in the peripheral infiltrate
Absent Present
16.2 83.8
3.8 96.2
Lymphocytes in the peripheral infiltrate
Absent Present
30.0 70.0
3.8 96.2
Neutrophils in the peripheral infiltrate
Absent Present
53.7 46.3
34.6 65.4
Macrophages in the peripheral infiltrate
Absent Present
72.5 27.5
30.8 69.2
Granuloma organization Predominant differentiation of the activated cells of the MPS in granulomata Distribution of the granulomatous infiltrate
Peripheral infiltrate
MPS, mononuclear phagocyte system.
4
L.H.M. Miranda et al. Table 3 Statistical analysis of the main microscopical features in skin lesions of sporotrichosis and ATL Categories
OR*
CIeOR (95%)
P value‡
Granuloma organization
Well organized† Poorly organized
15.00 1.00
1.93e116.45
0.001
Predominant differentiation of the mononuclear phagocytes in granulomata
Macrophages† Epithelioid cells
0.80 e
0.30e2.11 e
0.800
Intensity of neutrophil infiltration into granulomata
Marked† Mild
13.26 1.00
2.94e59.90
<0.001
Plasma cells around granulomata
Absent† Present
0.85 e
0.21e3.50 e
1.000
Plasma cells in the peripheral infiltrate
Absent† Present
4.85 1.00
0.60e39.03 e
0.180
Lymphocytes in the peripheral infiltrate
Absent† Present
10.71 1.00
1.37e83.66 e
0.007
Neutrophils in the peripheral infiltrate
Absent† Present
2.20 1.00
0.88e5.51 e
0.115
Macrophages in the peripheral infiltrate
Absent† Present
5.93 1.00
2.26e15.60 e
<0.001
Perivascular, perifollicular and interstitial† Diffuse
6.48
2.33e17.98
<0.001
1.00
e
Variable
Distribution of the peripheral infiltrate
*
Reference category: ATL. Reference category of each variable. ‡ P < 0.05 indicates a significant association. †
infiltrate (Fig. 4) was observed in 5.0% of skin biopsies from dogs with sporotrichosis, but was not observed in lesional skin from dogs with ATL. The intensity of neutrophil infiltration into pyogranulomata was marked in 52.5% of cases of sporotrichosis. In contrast, mild neutrophilic infiltration into pyogranulomata was observed in 92.3% of cases of ATL. The presence of marked neutrophilic infiltra-
tion was 12.26 fold more likely to be found in the lesions of sporotrichosis than ATL. The distribution of the peripheral inflammatory infiltration was perivascular, perifollicular and interstitial (Fig. 5) in 70.0% of cases of sporotrichosis. In contrast, this infiltration was diffuse (Fig. 6) in 53.9% of cases of ATL. Lesions with a perivascular, perifollicular and interstitial peripheral inflammatory
Fig. 1. Skin from a dog with sporotrichosis in which there are wellformed granulomata that contain neutrophils (arrows). There is an outer zone of plasma cells (dashed arrows). HE. 200.
Fig. 2. Skin from a dog with ATL in which there is a diffuse granulomatous infiltrate. A poorly formed granuloma is outlined. HE. 100.
Pathology of Sporotrichosis and Leishmaniosis
Fig. 3. Skin from a dog with ATL showing a granuloma predominantly composed of macrophages. HE. 200.
infiltrate therefore had a 5.48 higher chance of being sporotrichosis than ATL when compared with lesions with a diffuse peripheral infiltrate. Lymphocytes were present in the peripheral infiltrate in 70.0% of cases of sporotrichosis cases and in 96.2% of cases of ATL. Macrophages were present in the peripheral infiltrate in 27.5% of cases of sporotrichosis and in 69.2% of cases of ATL. The absence of lymphocytes and macrophages in the peripheral infiltrate was associated with a 9.71 and 4.93 fold greater risk of the lesion being due to sporotrichosis, respectively, when compared with lesions where these cells were present.
5
Fig. 5. Skin from a dog with sporotrichosis. There is a perifollicular infiltrate with plasma cells surrounding hair follicles. HE. 200.
Systematic microscopical analysis of the skin lesions of the dogs studied here showed that, although the defin-
itive diagnosis depends on demonstration of the aetiological agent, the presence of histopathological characteristics that favour the diagnosis of either sporotrichosis or ATL might be used to support the diagnostic suspicion. In the present study, histopathological analysis showed a predominance of poorly formed granulomata in both diseases. Nevertheless, the presence of well-formed granulomata is more likely to be associated with sporotrichosis in dogs (OR ¼ 15.00) when compared with ATL. However, the lack of this feature does not exclude a diagnosis of sporotrichosis. A marked intensity of neutrophil infiltration into the granulomata was also associated with the lesions of sporotrichosis and a mild intensity of such infiltration was more common in cases of ATL. Therefore,
Fig. 4. Skin from a dog with sporotrichosis showing interstitial granulomatous inflammation. Activated cells of the mononuclear phagocyte system (MPS) are scattered between collagen bundles (arrows). HE. 200.
Fig. 6. Skin from a dog with ATL. There is a poorly formed granuloma (arrows) and a diffuse non-granulomatous infiltrate found at the periphery (dashed arrows). HE. 200. Inset: activated cells of the MPS forming granulomata. HE. 400.
Discussion
6
L.H.M. Miranda et al.
the observation of pyogranulomata rich in neutrophils is more likely to be associated with sporotrichosis than ATL in dogs (OR ¼ 13.26). The distribution of the peripheral inflammatory infiltrate was predominantly perivascular, perifollicular and interstitial in the lesions of dogs with sporotrichosis, whereas it was predominantly diffuse in cases of ATL. Statistical analysis suggested that the diagnosis of canine ATL may be favoured when the peripheral inflammatory infiltrate is diffuse (OR ¼ 6.48). Moreover, analysis suggested that canine ATL is more likely where lesions have lymphocytes (OR ¼ 10.71) and macrophages (OR ¼ 5.93) in peripheral infiltrate. It should be noted that poor delineation of granulomata and the presence of macrophages in the diffuse peripheral infiltrates of canine ATL resulted in imprecise distinction between granulomatous and non-granulomatous infiltrates. Therefore, because of the subjectivity of morphological analysis, further investigation is necessary to confirm the reliability of these findings (macrophages in the peripheral infiltrate and diffuse distribution of the peripheral infiltrate) in distinguishing the two diseases. Although culture of the aetiological agent is the method of choice for the diagnosis of sporotrichosis or ATL, this test may not be performed or may not be available and, even if performed, no growth of the microorganism might be observed, generally because of methodological limitations (Moore and Ackerman, 1946; Schwarz, 1982). In addition, maintenance of the culture for a long period of time might be necessary (Schwarz, 1982). Thus, the differential diagnosis between sporotrichosis and ATL is frequently made based on clinicaleepidemiological evidence and serological tests (Barros et al., 2005; Santos et al., 2007). Clinically, the skin lesions of sporotrichosis and ATL have similar characteristics in dogs (Santos et al., 2007). Higher frequencies of multiple lesions and respiratory signs have been reported for animals with sporotrichosis (Schubach et al., 2006; Santos et al., 2007). Furthermore, the duration of the lesions seems to be shorter in canine sporotrichosis (Schubach et al., 2006) compared with dogs experimentally infected with ATL (Pirmez et al., 1988). With respect to serological tests, a significant rate of positive serology for leishmaniosis has been observed in dogs with sporotrichosis (Santos et al., 2007). Histopathological examination is a complementary diagnostic tool and might be the only alternative available for the identification of the aetiological agent, particularly in situations in which no specimens are collected for culture of the microorganism (Moskowitz et al., 1986; Werner and Werner, 1994; Rodrı´ guez and Sarmiento, 1998; Barros et al., 2005). The results of the present study suggest that
histopathological analysis may significantly contribute to the differential diagnosis between canine cutaneous sporotrichosis and ATL. This method detects features that directs the diagnostic suspicion and guides the search for the aetiological agent in histological sections, permitting the request of appropriate subsequent tests, reducing costs and the time taken to achieve a definitive diagnosis and initiate appropriate therapy.
Acknowledgments The authors thank I. D. F. Gremia˜o, S. A. Pereira, T. Okamoto, M. F. L. Almeida, M. D. P. E. Espinoza, F. C. C. Rodrigues, A. Correia, A. C. Silva, R. S. Reis, P. C. F. Monteiro, M. F. Madeira and R. Me´xas for technical support. This study was partially financed by the Programa de Apoio a Pesquisa Estrate´gica em Sau´de (PAPES IV), FIOCRUZ, by the Fundac¸a˜o Carlos Chagas Filho de Amparo a` Pesquisa do Estado do Rio de Janeiro (FAPERJ), and by the Conselho Nacional de Desenvolvimento Cientı´ fico Tecnolo´gico (CNPq), Brazil. TMPS is a CNPq researcher. The study was approved by the Ethics Committee on the Use of Animals, FIOCRUZ.
References Ackerman AB, Chongchitnant N, Sanchez J, Guo Y, Bennin B et al. (1997) Proceeding to specific diagnosis. In: Histologic Diagnosis of Inflammatory Skin Diseases: an Algorithmic Method Based on Pattern Analysis, 2nd Edit., G Stamathis, Ed., Williams & Wilkins, Baltimore, pp. 107e167. Adams DO (1974) The structure of mononuclear phagocytes differentiation in vivo e I. Sequential fine histological studies of the effect of Bacillus Calmette-Guerin (BCG). American Journal of Pathology, 76, 17e39. Barros MBL, Schubach AO, do Valle ACF, Galhardo MCG, Schubach TMP et al. (2005) Positive Montenegro skin test among patients with sporotrichosis in Rio de Janeiro. Acta Tropica, 93, 41e47. Kwon-Chung K, Bennett J (1992) Sporotrichosis. In: Medical Mycology, 2nd Edit., K Kwon-Chung, J Bennett, Eds, Lea & Febiger, Philadelphia, pp. 707e729. Marques MEA, Coelho KIR, Sotto MN, Bacchi CE (1992) Comparison between histochemical and immunohistochemical methods for diagnosis of sporotrichosis. Journal of Clinical Pathology, 45, 1089e1093. Miranda LHM, Quintella LP, Santos IB, Menezes RC, Figueiredo FB et al. (2009) Histopathology of canine sporotrichosis: a morphological study of 86 cases from Rio de Janeiro (2001e2007). Mycopathologia, 168, 79e87. Moore M, Ackerman LV (1946) Sporotrichosis with radiate formation in tissue: report of a case. Archives of Dermatology and Syphilology, 53, 253e264. Moskowitz LB, Ganjei P, Ziegels-Weissman J, Clearly TJ, Penneys NS et al. (1986) Immunohistological
Pathology of Sporotrichosis and Leishmaniosis
identification of fungi in systemic and cutaneous mycoses. Archives of Pathology & Laboratory Medicine, 110, 433e436. Pirmez C, Marzochi MC, Coutinho SG (1988) Experimental canine mucocutaneous leishmaniasis (Leishmania braziliensis braziliensis). Memo´rias do Instituto Oswaldo Cruz, 83, 145e151. Quintella LP, Cuzzi T, Madeira MF, Okamoto T, Schubach AO (2009) Immunoperoxidase technique using an anti-Leishmania (L.) chagasi hyperimmune serum in the diagnosis of culture-confirmed American tegumentary leishmaniasis. Revista do Instituto de Medicina Tropical de Sa˜o Paulo, 51, 83e86. Rodrı´ guez G, Sarmiento L (1998) The asteroid bodies of sporotrichosis. American Journal of Dermatopathology, 20, 246e249. Salinas G, Valderrama L, Palma G, Montes G, Saravia NG (1989) Deteccion de amastigotas en leishmaniasis cutanea y mucocutanea por el metodo de inmunoperoxidasa, usando anticuerpo policlonal: sensibilidad y especificidad comparadas con metodos convencionales de diagnostico. Memo´rias do Instituto Oswaldo Cruz, 84, 53e60. Santos IB, Schubach TMP, Leme LRP, Okamoto T, Figueiredo FB et al. (2007) Sporotrichosis e the main differential diagnosis with tegumentary leishmaniosis in dogs from Rio de Janeiro, Brazil. Veterinary Parasitology, 143, 1e6.
7
Schell WA (1998) Agents of chromoblastomycosis and sporotrichosis. In: Topley & Wilson’s Microbiology and Microbial Infections, 9th Edit., L Allejo, RJ Hay, Eds, Arnold, London, pp. 315e336. Schubach TMP, Schubach AO, Okamoto T, Barros MBL, Figueiredo FB et al. (2006) Canine sporotrichosis in Rio de Janeiro, Brazil: clinical presentation, laboratory diagnosis and therapeutic response in 44 cases (1998e2003). Medical Mycology, 44, 87e92. Schwarz J (1982) The diagnosis of deep mycoses by morphologic methods. Human Pathology, 13, 519e533. Sells PG, Burton M (1981) Identification of Leishmania amastigotes and their antigens in formalin fixed tissue by immunoperoxidase staining. Transactions of the Royal Society of Tropical Medicine and Hygiene, 75, 461e468. Sotto MN, Yamashiro-Kanashiro EH, Da Matta VLR, Brito T (1989) Cutaneous leishmaniasis of the New World: diagnostic immunopathology and antigen pathway in skin and mucosa. Acta Tropica, 46, 121e130. Werner AH, Werner BE (1994) Sporotrichosis in man and animal. International Journal of Dermatology, 33, 692e700.
July 15th, 2009 ½ Received, Accepted, December 22nd, 2009