Veterinary Dermatology 2005, 16, 364372

Blackwell Publishing, Ltd.

Clinical, immunological and histopathological ndings in a subpopulation of dogs with pododermatitis

RORY M. BREATHNACH*, KENNETH P. BAKER*, PATRICK J. QUINN, THOMAS A. MCGEADY, CAROL M. AHERNE and BOYD R. JONES* Departments of *Small Animal Clinical Studies; Veterinary Microbiology and Parasitology Veterinary Anatomy and Veterinary Physiology and Biochemistry, Faculty of Veterinary Medicine, University College Dublin, Dublin 4, Ireland
(Received 1 April 2005; accepted 17 August 2005)

Abstract Clinical, immunological and histopathological ndings in 20 adult dogs of varying breeds with chronic ( 6 months) inammation conned to the pedal skin were compared over a 2-year period with those of a group of age-matched controls (n = 20). All affected dogs were pruritic but systemically well. Lesions were present on all four feet in 18/20 cases. Affected feet were characteristically erythematous, swollen, painful and alopecic. Sinus tracts were evident in 4/20 dogs. Despite a methodical series of diagnostic tests, no underlying cause was identied. None of the dogs responded to antimicrobial therapy administered for 8 weeks, none had evidence of ectoparasitism and none satised the criteria for atopic dermatitis. There was no response to a dietary trial using a novel protein source. The condition was characterized histopathologically by epidermal hyperplasia, hyperkeratosis, spongiosis, dermal oedema and perivascular aggregates of lymphocytes and plasma cells. Clinical signs did not correlate with histopathological ndings. Affected dogs had signicantly elevated serum IgG and IgM concentrations. The results of lymphocyte proliferation assays and phenotypic studies to determine the relative percentage of CD3+, CD4+, CD8+ and CD21+ lymphocyte subsets, and the ratio of CD4+/CD8+ cells were not signicantly different between groups. No age, sex or seasonal predilections were noted. All dogs subsequently responded to immunosuppressive doses of prednisolone or cyclosporin. The term immunomodulatory-responsive lymphocyticplasmacytic pododermatitis is proposed to denote what may be a previously unrecognized condition in some dogs with pododermatitis of undetermined aetiology.

I NTRO D U CT I ON
Pododermatitis is a common and potentially debilitating disease of dogs.1 Although pedal involvement is a component of many canine skin diseases, some animals present with lesions conned solely to the foot.2 The clinical history in such cases is often characterized by periods of disease exacerbation and remission. Common causes of pododermatitis in the dog include, but are not limited to, trauma,3 microbial disease,4 ectoparasitism,5 foreign-body reactions6 and immunological skin diseases.1 However, once excluded, there exists a subpopulation of dogs with disease of undetermined aetiology.4 Although poorly dened, animals with idiopathic forms of pododermatitis generally have guarded to poor prognoses.4 Although various predisposing factors have been implicated,68 particular attention has focused on the potential role of bacterial infection4,9 and immunological dysfunction.7,9 Consequently, prolonged use of antibacterial therapy has been advocated to control clinical signs in many affected individuals.4 Although the rationale for such
Correspondence: Rory Breathnach, Department of Small Animal Clinical Studies, Faculty of Veterinary Medicine, University College Dublin, Beleld, Dublin 4, Ireland. Tel.: 0035317166000; Fax: 0035317166022; E-mail: rory.breathnach@ucd.i.e. 364

an approach is understandable in dogs with relapsing interdigital pyoderma,7 some animals with idiopathic pododermatitis do not respond well to antimicrobial therapy. Lesion regression in these latter cases more typically requires the use of anti-inammatory or immunomodulating agents such as glucocorticoids or cyclosporin.10 The aim of this prospective, controlled study was to investigate the clinical, immunological and histopathological ndings in a population of dogs with chronic ( 6 months), nonantibiotic responsive disease conned to the pedal skin. A further aim was to investigate if the poor response to antimicrobial therapy was associated with any underlying cutaneous, metabolic or immunological diseases.

M AT E R IA L S A N D M E T H O D S Case selection criteria

The dogs (n = 20; 10 male and 10 female; mean age = 5.4 years) studied were presented to the University Veterinary Hospital (UVH) of University College Dublin (UCD) for investigation of pedal skin disease over the time period 20032004 (Table 1). Animals were deemed eligible for enrolment once they satised a series of inclusion criteria. Primary criteria included
2005 European Society of Veterinary Dermatology

Cellular inltrates in pododermatitis

Table 1. Subject animal details for dogs with pododermatitis

365
Duration of pedal lesions at presentation (months) 8 12 7 14 8 16 22 7 12 11 15 6 20 8 18 15 20 6 10 12

Case no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Breed Boxer Cocker Spaniel Labrador Bulldog Bernese Mountain Dog West Highland White Terrier Springer Spaniel Terrier cross Labrador cross Rottweiler Boxer Bull Mastiff Springer Spaniel Spaniel cross Labrador cross Bull Terrier Pomeranian Fox-Hound American Bulldog Terrier cross

Sex Female Male Male Female Female Male Female Female Male Female Female Male Male Male Female Female Male Male Male Female

Weight (kg) 22.1 12.8 31 15.9 47.8 9.2 17.4 11.0 36.2 38.7 22.6 51.4 14.8 17.2 23.8 15.1 3.0 27.6 18.8 11.3

Age (years) 5 6 3 5 2 7 12 2 8 6 5 2 10 2 10 5 6 4 4 4

the presence of pedal lesions for 4 months on at least two feet, and the absence of cutaneous lesions at other body sites. The primary criteria were further satised if the results of skin scrapings/hair plucks, cytology, fungal culture, acaricidal trial (four applications of selamectin topically at fortnightly intervals; Stronghold, Pzer Animal Health, Dublin, Ireland), intradermal testing, food trial, haematology, biochemistry, urinalysis, histopathology and immunohistochemical staining (using rabbit antidog IgA, IgG, IgM and complement C3 monoclonal antibodies; ICN Immunobiologicals, IL, USA) of multiple lesional biopsies failed to yield a specic aetiological diagnosis. Animals that responded to an 8-week trial course of systemic antibacterial therapy based on the results of bacterial culture and susceptibility testing, and a 2% miconazole/chlorhexidine-containing shampoo (Malaseb, Leo Animal Health, Dublin, Ireland) itraconazole (2.5 mg kg1 PO q12 h; Sporanox, JanssenCilag, Dublin, Ireland) were excluded. Dogs showing a partial or complete response to a dietary trial using a novel protein source (either Royal Canin sensitivity control diet or Hills z/d ultra fed for a minimum of 8 weeks) were also rejected. Animals with evidence of systemic disease or that satised the criteria for diagnosing atopic dermatitis (AD)11 were also excluded from this study. Prior to inclusion, all oral and injectable glucocorticoid medication was discontinued for at least 4 and 8 weeks, respectively. The secondary inclusion criterion related to histopathological evidence of a perivascular dermatitis reaction pattern with lymphoplasmacytic inltrates within lesional biopsy sections. An age-matched control group (n = 20; 11 male and 9 female; mean age = 5.2 years) consisted of dogs presenting for euthanasia in the UVH for nondermatological disease.

Clinical ndings
A detailed history was obtained and clinical examination (systemic and cutaneous) performed in each case. The clinical ndings were entered into a database (Microsoft Excel) to allow subsequent investigation of any correlation between clinical signs and histopathological ndings. Follow-up clinical examinations were performed at monthly intervals for the rst three months and at three monthly intervals thereafter.

Clinical pathology
Blood samples were obtained by venipuncture for haematology, biochemistry, T4/TSH (thyroid-stimulating hormone) and immunoglobulin assays. Serum total T4 and TSH concentrations were assayed by chemiluminescence (Immulite, EURO/DPC Ltd, Caernarfon, Gwynedd, UK). Serum for immunoglobulin assays was stored at 80 C prior to testing (see succeeding discussions). Urinalysis was performed on free catch samples. Material for cytological examination was collected from lesional sites by skin scraping and ne needle aspiration, and stained with modied Wrights stain (Diff-Quik, Biosciences, Dublin, Ireland).

Parameters of humoral and cell-mediated immunity

Serum immunoglobulins: Immunoglobulin concentrations were quantied with a commercially available enzyme-linked immunosorbent assay (ELISA) kit using afnity-puried heavy-chain specic polyclonal antisera to canine IgA, IgG and IgM (Bethyl Laboratories, Montgomery, TX, USA). The specicities of the antisera used were evaluated by Western blot analysis (data not shown) using puried IgA (Bethyl Laboratories), IgG (Rockland Immunochemicals, Philadelphia, PA, USA) and IgM (Rockland Immunochemicals). There was no cross-reactivity for the anticanine IgM and IgA reagents. The anticanine IgG reagent produced

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 364372

366

R M Breathnach et al. XL-MCL ow cytometer (Beckman Coulter, Buckinghamshire, UK) and associated System II software. A total of 10 000 cells were enumerated in each run. Lymphocytes were selected based on forward and 90 light scatter.

slight cross-reactivity with puried IgA; however, densitometric measurements revealed this reactivity accounted for only 0.26% of that recorded for puried IgG (data not shown). In light of this nding, and the higher concentration of serum IgG compared to serum IgA in the dog,12 this cross-reactivity was considered insignicant. Proliferation assay: Peripheral blood mononuclear cells (PBMCs) were isolated as previously described,13 and resuspended in RPMI-1640 medium (Invitrogen, Paisley, UK) supplemented with 1% glutamine (Invitrogen), 5% heat-inactivated foetal calf serum (Sigma, Dublin, Ireland), 100 IU mL1 penicillin (Invitrogen), 100 g mL1 streptomycin (Invitrogen), 0.25 g mL1 amphotericin B (Sigma), 1% nonessential amino acids (Invitrogen) and 0.1% (50 m) -2-mercaptoethanol (Invitrogen) to obtain a nal cell concentration of 5 106 cells/mL. Triplicate 100 L volumes of PBMCs were pipetted into 96-well round bottom plates and 100 L of the following mitogens were added: pokeweed mitogen (5 g mL1), phorbol myristate (10 ng mL1) and Staphylococcus aureus enterotoxin B (10 g mL1). All mitogens were purchased from Sigma (Dublin, Ireland) and the concentrations selected were based on the results of a prior pilot study (n = 10 dogs). Control cultures were incubated with RPMI-1640 medium alone. The plates were incubated for 72 h at 37 C in a humidied incubator under 5% CO2 atmosphere. Sixteen hours prior to the end of the incubation period, 1.0 mCi 3H-thymidine (Amersham Bioscience, Buckinghamshire, UK) was added to each well. Finally, the cells were harvested and radioactivity quantied by liquid scintillation spectrometry (Trilux 1450 Microbeta, PerkinElmer Life Sciences, UK). The results were expressed as the stimulation index, dened as the ratio of the mean count per minute (CPM) of cells stimulated with mitogen and the mean CPM of cells incubated with medium alone. Lymphocyte phenotyping studies: PBMCs were diluted in FACS buffer (PBS; 5 m EDTA; 0.1% w/v sodium azide) supplemented with 10% heat-inactivated horse serum (Labkem, Dublin, Ireland) to obtain a nal cell concentration of 10 106 cells mL1. Following storage on ice for 20 min, separate 50 L volumes of PBMC suspensions were dual stained using 2 L mouse antidog CD3-FITC (clone CA17.2A12) plus 2 L mouse antidog CD4-PE (clone CA13.1E4), 2 L mouse antidog CD3-biotin plus 2 L mouse antidog CD8()-FITC (clone CA9.JD3) and 2 L mouse antidog CD3-FITC plus 2 L mouse antidog CD21-PE (clone CA2.1D6). The samples were then incubated on ice for a further 30 min. All monoclonal antibodies used were obtained from Prof Peter F. Moore (University of Davis, California, USA). A PE-conjugated streptavidin complex (Labkem) was used to label the biotinylated CD3 monoclonal, while mouse IgG1 and a PE-conjugated rabbit antimouse IgG monoclonal (Serotec, Oxford, UK) were used as negative controls. Following a nal washing step, lymphocyte subsets were enumerated immediately using a Coulter Epics

Histopathological examination
Biopsies of lesional and control skin were obtained by scalpel blade excision from the dorsal digital and palmoplantar surfaces of two feet per dog, placed on card and xed in 10% neutral buffered formalin. Morphologic features were evaluated on 6 m, haematoxylin and eosin (H&E)-stained sections of parafn-embedded tissue. Biopsies were sectioned longitudinally to ensure a consistent plane of orientation.

Statistical analysis
All data were analysed using the statistical package SPSS version 11 for Windows (SPSS Inc., Chicago, IL,USA). Correlations between clinical signs and histopathological ndings were calculated using Spearmans test. The haematological/biochemical results were compared using the independent-samples t-test. For all remaining data sets, the nonparametric MannWhitney U-test was employed as the actual and logarithmically transformed data were not normally distributed. For all analysis performed, signicance was dened as P < 0.05. Whenever possible, the data were reported as the median (95% condence interval).

R E SU LT S Clinical ndings
Males and females were equally represented in this case series. The mean age for onset of clinical signs was 4.3 years (range: 1.310.2 years), and signs were present for 622 months (mean: 12.4 months) prior to sampling. All four feet were affected in 18/20 cases examined. Lesions were present on both the dorsal and the palmoplantar skin surfaces in all cases. Although the region bordering the pads was affected in 15/20 cases, lesions involving the pad surface were not observed. Although not tested statistically, follow-up studies (data not shown) over a 520-month period revealed no evidence of a seasonal component to the disorder or any consistent temporal association with specic therapeutic agents (anthelminthics, antiparasiticides or vaccines). The clinical ndings for each dog are summarized in Table 2. However, it must be emphasized that the data in Table 2 represent one point in time (day of sampling), and that the clinical signs in each case uctuated in line with the dynamic nature of this chronic inammatory disorder. In 15/20 cases, the lesions started on one or two feet; thereafter, the lesions increased in severity and spread to involve the other feet over a gradual period of 16 months. In the remaining ve dogs, the onset was more acute with all four feet exhibiting clinical signs within a period of 1 month. Over the

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 364372

Cellular inltrates in pododermatitis

+++ ++ ++ + 0 + 0 0 0 0 ++ +++ 0 20

367

++ ++ ++ ++ 0 + 0 0 0 0 + ++ ++

12

+++ +++ +++ ++ + +++ 0 0 0 0 0 + +

13

++ +++ ++ ++ + + 0 0 ++ ++ ++ ++ ++

14

+++ ++ ++ + 0 0 0 + 0 ++ + ++ +

15

++ ++ 0 + 0 0 0 0 0 0 0 ++ 0

16

+++ +++ ++ +++ 0 + 0 ++ 0 0 + ++ 0

17

+++ +++ ++ +++ ++ ++ 0 0 0 + ++ + ++

18

+++ ++ ++ +++ 0 + 0 + 0 ++ ++ ++ +

19

Figure 1. Left front foot, case no. 15. Erythema, serous exudation (SE) and epidermal erosion (Er) of the dorsal interdigital skin in a 10-year old Labrador-cross bitch with lymphocyticplasmacytic pododermatitis.

Table 2. Clinical signs of pedal skin disease in dogs with pododermatitis

+++ +++ +++ ++ + + 0 0 ++ ++ + ++ + *0 = absent; + = mild; ++ = moderate; +++ = marked/severe.

+++ +++ ++ +++ ++ 0 0 + 0 + 0 +++ ++

++ ++ ++ + 0 0 0 0 0 ++ + + +

+ +++ +++ ++ ++ +++ 0 +++ 0 + ++ ++ ++

+++ ++ ++ 0 0 0 0 0 + + + ++ 0

+++ ++ 0 ++ 0 + + 0 0 0 + + +

++ +++ ++ ++ 0 + 0 ++ + 0 0 + 0

10

++ +++ ++ ++ 0 + 0 + 0 ++ ++ ++ 0

11

Figure 2. Right front foot, case no. 19. Erythema, alopecia and nodule formation on the palmar surface of a 4-year-old male Bulldog with lymphocyticplasmacytic pododermatitis.

Case #

+++* ++ ++ ++ 0 0 0 0 0 0 + + 0

++ +++ 0 ++ 0 + 0 0 0 + 0 ++ 0

++ ++ ++ ++ + + + 0 0 0 + + ++

course of the study, the dominant clinical signs present in all cases (Figs 1 and 2) were pruritus, erythema, alopecia and localized skin thickening. A web-like appearance to the foot was noted in 15/20 dogs. All dogs had pain/discomfort on pedal examination. At various times, sinus tracts and serosanguinous discharges were evident in 7/20 cases examined. Seropurulent discharges were recorded in 5/20 dogs. Although pustules were evident in 6/20 cases, they were not observed following antimicrobial therapy. Haemorrhagic bullae or nodules were recorded in 5/20 cases.

1 Clinical sign

Pruritus Erythema Alopecia Skin thickening Swelling of toes Hyperpigmentation Seropurulent discharge Serosanguinous discharge Sinus tracts Erosions/ulcers Scarring /web-like appearance Pain/discomfort Regional lymphadenopathy

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 364372

368

R M Breathnach et al.

Serum immunoglobulins
Dogs with pododermatitis (Fig. 3) had a signicant increase in the median serum concentration of both IgG and IgM (P < 0.05; MannWhitney test).

Proliferation assays
The median (95% condence interval (CI)) stimulation index for PBMCs from dogs with pododermatitis was 22.4 (9.547.0), 6.4 (2.215.4) and 11.4 (4.233.8) for pokeweed mitogen, phorbol myristate and S. aureus enterotoxin B, respectively; the corresponding values in the control group were 34.5 (17.482.8), 4.5 (2.2 10.8) and 36.6 (12.588.1), respectively. None of these differences between groups were signicant (P > 0.05; MannWhitney test).

Lymphocyte phenotyping studies

The median (95% CI) relative percentage values for CD3+, CD4+, CD8+ and CD21+ lymphocytes within the PBMC fraction of dogs with pododermatitis were 52.3% (30.174.5%), 31.7% (22.9 45.5%), 23.3% (14.2 34.3%) and 18.7% (8.427.8%), respectively; the corresponding values in the control group were 48.5% (27.764.2%), 28.1% (14.839.6%), 19.4% (12.426.1%) and 18.9% (10.131.8%), respectively. These differences between groups were not signicant (P > 0.05; MannWhitney test). In addition, the ratio of CD3+/ CD4+, CD3+/CD8+, CD3+/CD21+ and CD4+/CD8+ cells did not differ signicantly between groups (P > 0.05; MannWhitney test).

Histopathology
Lesional biopsies were characterized by the presence of a perivascular dermatitis reaction pattern accompanied by dermal oedema, epidermal hyperplasia (15/20 dogs) hyperkeratosis (14/20 dogs) and spongiosis (15/ 20 dogs). In all cases examined, the perivascular inltrates were composed of moderate to marked numbers of lymphocytes and plasma cells (Fig. 4). Although more pronounced in the supercial and mid-dermis, 8/20 dogs had a prominent perivascular response in the deep dermis. Lymphocyte and plasma cell inltrates were also recorded beneath the dermoepidermal junction in 8/20 cases examined (Fig. 5), while focal mononuclear cell exocytosis was evident in the epidermis of 14/20 dogs. A moderate increase in the number of dermal mast cells and macrophages present was observed in 7/20 and 11/20 cases, respectively. Small numbers of eosinophils were recorded in the perivascular region of 7/20 dogs. Although periadnexal involvement was a prominent feature in 6/20 dogs, folliculitis was only observed in two dogs. Keratinized material was present in the dermis of two dogs, neither of which had folliculitis. Dermal (pyo)granulomas were not a feature of this case series. Dermal broplasia was a prominent nding in 6/20 dogs. Correlation studies revealed no statistically signicant associations between the clinical signs listed in Table 2 and any of the following histopathological ndings: perivascular dermatitis, dermal oedema,

Figure 3. Serum concentrations (mg L1) of IgA (a), IgG (b) and IgM (c) in control dogs and dogs with pododermatitis. IgG and IgM concentrations in the pododermatitis population were signicantly higher (P < 0.05; MannWhitney test). The bar represents the median value.

Clinical pathology
There was a mild neutrophilia (without left shift) in 4/20 dogs with pododermatitis. Serum alkaline phosphatase values were moderately elevated (< 460 IU L1; reference range 050 IU L1) in six patients. Signicant increases were recorded in the mean plasma concentrations of potassium and cholesterol in the pododermatitis group (P < 0.05; independent-samples t-test). All dogs with pododermatitis had serum total T4 and TSH values within the reference ranges. Four dogs in the pododermatitis group had 1+ proteinuria. Cytological examination revealed degenerate neutrophils and intracellular bacteria at focal pedal sites in one dog (case no. 3). Although Malassezia organisms were observed in skin scrapings from ve dogs, in each instance the organism count was < 1 yeast per high power eld (400). Low to moderate numbers of lymphocytes and plasma cells were observed in ne needle aspirates (FNAs) from 7/20 dogs. Deep skin scrapings were negative for Demodex spp. mites in all cases.

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 364372

Cellular inltrates in pododermatitis

369

for cyclosporin was 1.54.5 mg kg1 q48 h. Cessation of therapy (n = 5) was associated with a relapse of pedal lesions. In these latter ve cases, the lesions resolved once more following re-institution of immunomodulatory therapy. All dogs remained systemically well at follow-up examinations.

D IS C U S S IO N
The term idiopathic pododermatitis should not be interpreted as a specic disease entity,9,10 as it undoubtedly covers a heterogeneous population of dogs in which pathogenesis may differ substantially between individuals. Nevertheless, because of the close similarity in presenting clinical signs and the consistent tissue reaction pattern at the lesional site, the dogs studied here may represent a denable subpopulation under this umbrella term. Although no signicant correlation was evident between the clinical signs and histopathological ndings, the various causes of pododermatitis in dogs can rarely be differentiated based on clinical signs alone.1,3 In addition, studies relating to other diseases such as demodicosis14,15 and autoimmune dermatoses16 have reported a similar lack of correlation between the clinical and histopathological ndings. In line with previous reports,4,8 the bacteriology performed as part of the primary inclusion criteria isolated Staphylococcus intermedius from at least one lesional site in all dogs with pododermatitis. In addition, cytological examination at this stage revealed degenerate neutrophils and intracellular bacteria at focal pedal sites in 12/20 dogs. Despite this, no benecial response was observed following 8 weeks of systemic and topical antimicrobial therapy. As the cytological ndings post-antimicrobial therapy were not consistent with pyoderma in 19 of the 20 dogs, it was considered unlikely that this failure to respond was related to either drug resistance or inappropriate treatment duration. At rst glance, these data may suggest that contrary to previous reports,7,9 bacterial infection was not a pivotal factor in disease pathogenesis in the population studied. However, the absence of sequential bacterial counts and culture of biopsy material precludes any denitive conclusions on this issue. In addition, interdigital pyoderma may respond poorly to antibacterial therapy if underlying cutaneous, metabolic or immunological diseases are not simultaneously addressed.7,9,17 Based on a combination of history, clinical examination, laboratory ndings and case follow-up, there was no evidence of metabolic disease in the population studied. The higher potassium and cholesterol values in dogs with pododermatitis were considered of no biological signicance as individual values were all within the normal laboratory reference range. The elevated serum alkaline phosphatase values and mild proteinuria were consistent with known previous glucocorticoid medication.

Figure 4. Supercial dermis of dorsal interdigital skin, case no. 9. Perivascular dermatitis containing predominantly lymphocytes (arrows) and plasma cells, and occasional macrophages (arrowheads), within the inammatory inltrates surrounding blood vessels (BV) (H&E, 400; bar = 25 m).

Figure 5. Epidermis (Epi) and supercial dermis (Der) of plantar skin, case no. 11. Diffuse lymphocyticplasmacytic inltrate beneath the dermo-epidermal (arrows) junction and in the supercial dermis. Note the epidermal hyperplasia and mononuclear cell exocytosis (arrowheads) (H&E, 100; bar = 100 m).

epidermal hyperplasia, hyperkeratosis and spongiosis (P > 0.05; Spearmans test).

Response to treatment and clinical follow-up

All 20 dogs exhibited marked clinical improvement over a 28-week period in response to immunosuppressive doses of prednisolone (2 mg kg1 PO q24 h; n = 13) or cyclosporin (5 mg kg1 PO q24 h; Atopica, Novartis Animal Health, Dublin, Ireland; n = 7). In each instance, longer-term control was achieved following gradual tapering of the dose administered once clinical signs had ameliorated. The tapered dose of prednisolone required to maintain longer-term control was 0.4 1.0 mg kg1 q48 h, whereas the corresponding dose

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 364372

370

R M Breathnach et al. immunomodulatory therapy. This nding suggests an underlying immunological basis to the disorder, and furthermore suggests that any bacterial involvement is likely to be secondary to an immune-mediated alteration in skin metabolism. Although complete remission was possible in many of the dogs using immunosuppressive dose rates, the dose of prednisolone or cyclosporin was tapered in each instance to achieve a balance between clinical control and minimizing the side-effects of the immunosuppressive agent used. The tissue reaction pattern in our current study bore some similarities to that reported in canine AD.27,35 However, none of the dogs satised the criteria for diagnosing AD;11 intradermal testing was negative. Furthermore, the mean age for the onset of clinical signs was 4.3 years compared to 1.7 in a recent survey of dogs with AD.36 It could be argued that lymphoplasmacytic inltrates are a common nonspecic nding at mucocutaneous junctions and glabrous skin sites, irrespective of the underlying cause of disease. However, additional results from dogs (n = 17) that satised the primary but not the secondary inclusion criteria suggest that LPP is not a purely stereotypical response observed in all lesional biopsies of inamed pedal skin. Moreover, the mononuclear inltrates observed here were equally present at haired and nonhaired pedal skin sites. In conclusion, our results indicate that the dogs studied were not immunocompromised in terms of systemic health or T- and B-cell function and subset numbers. Instead, the nature of the cellular inltrates, the lack of response to antimicrobial therapy and the profound improvement following immunosuppressive therapy all point to an immunological basis to the disorder. However, although providing valuable indicators, the data (particularly the immunological ndings) are not sufciently robust to characterize a denable entity; thus, the concept of immunomodulatoryresponsive LPP remains a hypothesis. However, to test this hypothesis further, a larger study has now commenced to investigate if dogs with immunomodulatoryresponsive LPP have a consistent pattern of expression of certain key immunological parameters including lesional B- and T-cell responses, dendritic cell populations, MHC class II expression and cytokine proles.

The signicant increase in serum IgG concentrations in the affected population was in agreement with previous ndings in dogs with 18 and deep pyoderma,19 although it is noteworthy that a signicant decrease in IgG levels was reported in one study of dogs with idiopathic deep pyoderma.12 Serum IgM concentrations were also signicantly elevated in dogs with pododermatitis. The low serum IgA concentrations documented in dogs with antibiotic-responsive dermatitis20 and pedal pyoderma21 were not a feature of this case series. Although these results suggest an up-regulated humoral immune response in dogs with pododermatitis, a causal relationship between the two cannot be inferred. Blastogenic responses and relative percentage values for the different lymphocyte subsets were not signicantly different between groups. In addition, there was no alteration in the CD4+/CD8+ lymphocyte ratio as reported in demodicosis,22 German Shepherd pyoderma23 or AD (with Malassezia pachydermatis infection),24 and none of the dogs had laboratory ndings consistent with hypothyroidism. Although the above assays represent relatively crude methods for assessing cell-mediated and humoral immune function,25 the results generated here do not suggest a global B- or T-cell defect in the dogs examined. The tissue reaction pattern observed on histopathology provided further insight into the possible presence of an underlying, nonantibiotic responsive skin disease. However, a perivascular dermatitis pattern often represents a stereotypical response to any inammatory stimulus,26 and does not in itself constitute a specic diagnostic entity.9 Further classication must rely on other component features within the section, including the predominant cell types involved, the plexus of blood vessels participating and the nature or extent of any ancillary morphological changes.27,28 As the predominant cell populations present were lymphocytes and plasma cells, the term lymphocyticplasmacytic pododermatitis (LPP) is proposed to reect the histological appearance of the lesion in keeping with the long-established approach for alimentary tract and respiratory tract lesions such as lymphocyticplasmacytic enteritis29 and lymphocyticplasmacytic rhinitis.30 The presence of other cell types should not detract from the term proposed, as biopsies from dogs with lymphocytic plasmacytic enteritis frequently contain additional neutrophils, eosinophils and macrophages.31 Unfortunately, the data relating to plexus participation and the ancillary morphological changes within the epidermis were nonspecic and common to a large number of skin diseases in the dog.28,32 The presence of large numbers of lymphocytes and plasma cells within a lesion usually indicates immune reactivity33 and strongly implies antigenic drive.27 However, mixed lymphoplasmacytic inltrates have been reported in both antibiotic-responsive and immunomodulatory-responsive dermatoses in the dog, particularly those involving mucocutaneous junctions.34 In this study, all 20 dogs exhibited marked clinical improvement and longer-term control in response to

AC K N OW L E D G E M E N T S
The authors are grateful to Dr Adrian Dunne and Dr Michelle Finlay for help with the statistical analysis, and to the nursing staff within the UVH of UCD who helped with the animal care and management.

REFERENCES
1. White SD. Pododermatitis. Veterinary Dermatology 1989; 1: 118. 2. Ihrke PJ, Stannard AA, Ardans AA et al. Pemphigus

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 364372

Cellular inltrates in pododermatitis

foliaceus of the footpads in three dogs. Journal of the American Veterinary Medical Association 1985; 186: 679. Anderson RK. Canine pododermatitis. The Compendium on Continuing Education for the Practicing Veterinarian 1980; 11: 36171. Mason IS. Canine pyoderma. Journal of Small Animal Practice 1991; 32: 38186. White S. Pododermatitis. In: Locke PH, Harvey RG, Mason IS eds. Manual of Small Animal Dermatology. Gloucestershire: British Small Animal Veterinary Association Publications, 1993: 14656. Shaw S. Pododermatitis. Proceedings of a course in small animal dermatology. Palmerston North: Foundation for Continuing Education of the New Zealand Veterinary Association, 1987: Publication no. 13: 24044. Swaim SF, Lee AH, MacDonald JM et al. Fusion podoplasty for the treatment of chronic brosing interdigital pyoderma in the dog. Journal of the American Animal Hospital Association 1991; 27: 26474. Whitney JC. Some aspects of interdigital cysts in the dog. Journal of Small Animal Practice 1970; 11: 8392. Scott DW, Miller WH, Grifn CE. Small Animal Dermatology, 6th edn. Philadelphia: W.B. Saunders Company, 2001: 304 06 & 667779. Muller GH. Skin diseases of the Chinese Shar-Pei. Veterinary Clinics of North America: Small Animal Practice 1990; 20: 165570. Willemse A. Atopic skin disease: a review and reconsideration of diagnostic criteria. Journal of Small Animal Practice 1986; 27: 77178. Shearer DH, Day MJ. Aspects of the humoral immune response to Staphylococcus intermedius in dogs with supercial pyoderma, deep pyoderma and anal furunculosis. Veterinary Immunology and Immunopathology 1997; 58: 10720. Greeley EH, Kealy RD, Ballam JM et al. The inuence of age on the canine immune system. Veterinary Immunology and Immunopathology 1996; 55: 110. Caswell JF, Yager JA, Ferrer L et al. Canine demodicosis: a re-examination of the histopathologic lesions and description of the immunophenotype of inltrating cells. Veterinary Dermatology 1995; 6: 919. Day M, Power C, Oleshko J et al. Low serum immunoglobulin concentrations in related Weimaraner dogs. Journal of Small Animal Practice 1997; 38: 31115. Werner LL, Brown KA, Halliwell REW. Diagnosis of autoimmune skin disease in the dog: correlation between histopathologic, direct immunouorescent and clinical ndings. Veterinary Immunology and Immunopathology 1983; 5: 4764. Rosser EJ Jr. German shepherd dog pyoderma: a prospective study of 12 dogs. Journal of the American Animal Hospital Association 1997; 33: 35563. Hill PB, Moriello KA, DeBoer DJ. Concentrations of total serum IgE, IgA and IgG in atopic and parasitized dogs. Veterinary Immunology and Immunopathology 1995: 44: 10513. Hall IA, Campbell KL. Serum immunoglobulins in normal dogs and dogs with skin disease. Proceedings of the American Academy of Veterinary Dermatology Annual Meeting. San Diego: American College of Veterinary Dermatology, 1993: 4547.

371

3.

4. 5.

6.

7.

8. 9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20. Felsburg PJ, Glickman LT, Jezyk PF. Selective IgA deciency in the dog. Clinical Immunology and Immunopathology 1985; 36: 297305. 21. Studdert VP, Phillips WA, Studdert MJ et al. Recurrent and persistent infections in related Weimaraner dogs. Australian Veterinary Journal 1984; 61: 26163. 22. Caswell JF, Yager JA, Parker WM et al. A prospective study of the immunophenotype and temporal changes in the histological lesions of canine demodicosis. Veterinary Pathology 1997; 34: 27987. 23. Denerolle P, Bourdoiseau G, Magnol JP et al. German shepherd dog pyoderma: a prospective study of 23 cases. Veterinary Dermatology 1998; 9: 24348. 24. Morris DO, Clayton DJ, Drobatz KJ et al. Response to Malassezia pachydermatis by peripheral blood mononuclear cells from clinically normal and atopic dogs. American Journal of Veterinary Research 2002; 63: 35862. 25. Foster AP. Immunomodulation and immunodeciency. Veterinary Dermatology 2004; 15: 11526. 26. Yager JA, Wilcock BP. Skin biopsy: revelations and limitations. Canadian Veterinary Journal 1988; 29: 969 72. 27. Yager JA, Wilcock BP. Colour Atlas of Surgical Pathology of the Dog and Cat. London: Wolfe Publishing, 1994: 15237. 28. Stannard AA. Perivascular dermatitis. European Society of Veterinary Dermatology Second Workshop on Dermatopathology. London, 1991: 3944. 29. Jergens AE. Inammatory bowel disease: current perspectives. Veterinary Clinics of North America: Small Animal Practice 1999; 29: 50121. 30. Johnson LR, Clarke HE, Bannasch MJ et al. Correlation of rhinoscopic signs of inammation with histologic ndings in nasal biopsy specimens of cats with or without upper respiratory tract disease. Journal of the American Veterinary Medical Association 2004; 225: 395400. 31. Jergens AE, Schreiner CA, Frank DE et al. A scoring index for disease activity in canine inammatory bowel disease. Journal of Veterinary Internal Medicine 2003; 17: 29197. 32. Walder EJ, Conroy JD. Contact dermatitis in dogs and cats: pathogenesis, histopathology, experimental induction and case reports. Veterinary Dermatology 1994; 5: 14962. 33. Streilein JW. Skin-associated lymphoid system (SALT): the next generation. In: Bos JD. ed. Skin Immune System (SIS). Boca Raton: CRC Press, 1989: 2548. 34. Wiemelt SP, Goldschmidt MH, Greek JS et al. A retrospective study comparing the histopathological features and response to treatment in two canine nasal dermatoses, DLE and MCP. Veterinary Dermatology 2004; 15: 34148. 35. Olivry T, Hill PB. The ACVD task force on canine atopic dermatitis (XVIII): histopathology of skin lesions. Veterinary Immunology and Immunopathology 2001; 81: 30509. 36. Zur G, Ihrke PJ, White SD, Kass PH. Canine atopic dermatitis: a retrospective study of 266 cases examined at the University of California, Davis, 19921998. Part I. Clinical features and allergy testing results. Veterinary Dermatology 2002; 13: 89102.

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 364372

372

R M Breathnach et al.
Rsum Les donnes cliniques, immunologiques et histopathologiques observes chez 20 chiens prsentant une inammation podale chronique ( 6 months) ont t compares avec celles dun groupe de chien contrle (n = 20). Un prurit a t not chez tous les chiens atteints, sans atteinte de ltat gnral. Des lsions taient notes sur les 4 pieds dans 18/20 cas. Il sagissait drythme, denure, de douleur et dalopcie. Des stules taient notes pour 4/20 chiens. Aucune cause na t retrouve malgr des examens complmentaires adapts. Aucun chien na rpondu un traitement antibiotique de 8 semaines, aucun parasite na t retrouvn et aucun des chiens ne prsentaient de critres en faveur dune dermatite atopique. Un rgime dviction avec une nouvelle source de protine na pas permis damliorer la dermatose. Lexamen histopathologique a montr une hyperplasie pidermique, une hyperkratose, une spongiose, un oedme dermique et des aggrgats privasculaires de lymphocytes et des plasmocytes. Les signes cliniques ntaient pas corrls avec les modications microscopiques. Les chiens atteints prsentaient une augmentation signicative des IgG et IgM sriques. Des tests de prolifration lymphocytaire et une tude phnotypique nont pas montr de diffrence entre les deux groupes pour les pourcentages de CD3+, CD4+, CD8+ et CD21+, ni dans le rapport CD4+/CD8+. Tous les cbiens ont correctement rpondu ladministration de doses immunosuppressives de corticoide ou de cyclosporine. Lappellation pododermatite lympho-plasmocytaire rpondant limmunomodulation est propsoe pour dcrire cette entit. Resumen Los hallazgos clnicos, inmunolgicos e histopatolgicos en 20 perros adultos de diversas razas afectados con inamacin cutnea pedal crnica ( 6 meses) se compararon durante dos aos con aquellos de un grupo de similar edad (n = 20). Todos los perros afectados presentaron prurito local, pero eran normales en el mbito sistmico. Las lesiones se presentaron en todas las extremidades en 18/20 casos. Los pies afectados presentaron eritema, hinchazn, dolor y alopecia. Tractos sinusales se observaron en 4/20 perros. A pesar de realizar una amplia serie de pruebas diagnosticas, no se encontr ninguna etiologa de fondo. Los perros no respondieron a una dieta con una protena diferente. A nivel histopatolgico los animales presentaron hiperplasia de la epidermis, hiperqueratosis, espongiosis, edema en la dermis e inltracin perivascular de linfocitos y clulas plasmticas. Los signos clnicos no guardaron relacin con los hallazgos histopatolgicos. Los perros afectados presentaron concentraciones elevadas de IgG e IgM en el suero. Los resultados de proliferacin linfocitaria y los estudios fenotpicos para determinar el porcentaje relativo de las subpoblaciones linfocitarias CD3+, CD4+, CD8+ y CD21+, as como la relacin CD4+/CD8+ no fueron signicativamente diferentes entre los dos grupos. Tampoco se observ predileccin en funcin del sexo, edad o estacin del ao. Todos los perros respondieron a dosis inmunosupresoras de prednisolona o ciclosporina. Proponemos el trmino pododermatitis linfoplasmactica con respuesta a inmunomodulacin para indicar lo que podra ser una nueva condicin no reconocida previamente en algunos perros con pododermatitis de etiologa desconocida. Zusammenfassung Klinische, immunologische und histo-pathologische Befunde von 20 adulten Hunden verschiedener Rassen mit chronischer ( 6 Monate) Entzndung, die auf die Haut der Pfoten begrenzt war, wurden ber einen Zeitraum von 2 Jahren mit denen einer Gruppe von Kontrolltieren (n = 20) verglichen, die altersmig bereinstimmten. Alle betroffenen Hunde zeigten Juckreiz, waren aber ansonsten klinisch unauffllig. Lsionen waren in 18/20 Fllen an allen vier Pfoten vorhanden. Betroffene Pfoten waren typischerweise erythemats, geschwollen, schmerzhaft und haarlos. Fistelkanle waren bei 4/20 Hunden vorhanden. Trotz einer systematischen Aufarbeitung mit diagnostischen Tests konnte keine zugrunde liegende Ursache identiziert werden. Keiner der Hunde zeigte eine Verbesserung durch eine antimikrobielle Therapie, die ber acht Wochen verabreicht wurde, keiner hatte Anzeichen von Ektoparasiten und keiner entsprach den Kriterien fr atopische Dermatitis. Eine Verbesserung durch eine Ausschlussdit mit einer neuen Proteinquelle wurde nicht festgestellt. Histo-pathologisch war der Zustand charakterisiert durch epidermale Hyperplasie, Hyperkeratose, Spongiose, dermales dem und perivaskulre Aggregate von Lymphozyten und Plasmazellen. Die klinischen Symptome korrelierten nicht mit den histo-pathologischen Befunden. Die betroffenen Hunde zeigten signikant erhhte Serum IgG und IgM Konzentrationen. Die Ergebnisse der Lymphozyten Proliferationsassays und der phnotypischen Studien zur Bestimmung des relativen Prozentanteils von CD3+, CD4+, CD8+ und CD21+ Lymphozytengruppen sowie des Verhltnisses von CD4+/CD8+ Zellen, zeigten keine signikanten Unterschiede zwischen den Gruppen. Keine Alters-, Geschlechts- oder saisonale Prdilektion wurde festgestellt. In der Folge sprachen alle Hunde auf immunsupprimierende Dosierungen von Prednisolon oder Cyclosporin an. Der Begriff immunomodulatory-responsive lymphocytic-plasmacytic pododermatitis (auf immunmodulatorische Therapie reagierende lympho-plasmazellulre Pododermatitis) wird vorgeschlagen, um einen eventuell bisher nicht erkannten Zustand bei manchen Hunden mit Pododermatitis von unbekannter tiologie zu bezeichnen.

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 364372