Veterinary Dermatology 2005, 16, 373384

Alterations of epidermal proliferation and cytokeratin expression in

skin biopsies from heavy draught horses with chronic pastern
dermatitis

Blackwell Publishing Ltd

FLORIAN GEBUREK*, BERNHARD OHNESORGE*, ECKEHARD DEEGEN*, RENATE

DOELEKE and MARION HEWICKER-TRAUTWEIN
*Klinik fr Pferde, Tierrztliche Hochschule Hannover, Bischofsholer Damm 15, D-30173 Hannover, Germany,
Institut fr Pathologie, Tierrztliche Hochschule Hannover, Bnteweg 17, D-30559 Hannover, Germany
(Received 16 August 2004; accepted 17 August 2005)

Abstract We report the historical, clinical and histopathological characteristics of skin lesions in biopsies from
37 heavy draught horses with chronic pastern dermatitis. The skin lesions were divided into four macroscopic
groups: scaling (group I, n = 5), hyperkeratotic and hyperplastic plaque-like lesions (group II, n = 14), nodular
skin masses (group III, n = 16) and verrucous skin lesions (group IV, n = 2). The principal histological findings
were hyperkeratosis and epidermal hyperplasia. There was a gradual increase in epidermal hyperplasia from
groups I to IV, suggesting that the lesions represent different stages of disease. In all cases, there was perivascular
dermatitis dominated by T lymphocytes with an increase in MHC class II-positive dendritic-like cells. Immunohistochemical labelling for cytokeratins CK5/6(4), CK10 and CK14 indicated a change in their expression
pattern. This correlated with the degree of epidermal hyperplasia, indicating abnormal differentiation of
keratinocytes. There was a statistically significant correlation between the severity of skin lesions and several other
factors including increasing age, increasing cannon circumference, prominence of anatomical structures such as
fetlock tufts of hairs, ergots and chestnuts, and bulges in the fetlock region.

IN TRO D U CT I ON
Pastern dermatitis, also known as greasy heel, scratches
or mud fever in the Anglo-American literature, or as
mauke in the German literature, is a progressive
inflammatory skin lesion involving the posterior
pasterns of horses.13 Pastern dermatitis occurs in all
breeds, but is most common in heavy draught horses
with feathering.3 Affected horses initially show oedema
and scaling that progresses to exudation and crusting.
Pastern dermatitis is thought to be multifactorial with
many potential causes, e.g. infection with bacteria
(Staphylococcus spp., Dermatophilus congolensis),
dermatophytosis, chorioptic mange, trombiculosis,
photosensitization, vasculitis or contact with chemical
irritants.3 Recently, the occurrence of papillomatous
pastern dermatitis associated with spirochetes and
nematodes identified as Pelodera strongyloides in a
Tennessee Walking Horse was reported.4 When the
aetiology is indeterminable, the term idiopathic
pastern dermatitis is used.13 In these chronic cases,
the skin of the posterior pasterns becomes hyperkeratotic and lichenified. Finally, hyperplasia is visible as
papillomatous or multifocal circumscribed verrucous
masses referred to as grapes.13 These chronic, verrucous pastern lesions, which almost exclusively occur in
Correspondence: M. Hewicker-Trautwein, Institut fr Pathologie,
Tierrztliche Hochschule Hannover, Bnteweg 17, D-30559 Hannover,
Germany. E-mail: Marion.Hewicker-Trautwein@tiho-hannover.de
2005 European Society of Veterinary Dermatology

heavy draught horses, are known as warzenmauke

(warze = wart in English; muche = German medieval
word for a lameness-inducing disease of horses) in
the German veterinary literature. Although chronic
hyperplastic skin lesions of the pasterns of heavy
draught horses have been described for centuries, detailed
descriptions of the histopathology are not available,
except from older reports or dissertations mainly by
German or French authors.59
The purpose of this study was to describe the clinical
and histopathological features of chronic pastern
dermatitis in biopsies from 37 heavy draught horses of
different breeds. The proliferative activity of epidermal
basal cells and expression of cytokeratins were characterized immunohistochemically with antibodies to the
proliferation marker Ki-67 and cytokeratins CK5/6(4),
CK10 and CK14. The dermal cellular immune response
was assessed with antibodies for leucocyte and MHC
class II antigens.

M AT E R IA L S A N D M E T H O D S
Historical and clinical details
Data from 37 heavy draught horses of different breeds
with chronic pastern dermatitis were collected (Table 1).
For each horse, this included details of housing, feeding
regimes, amount of work, use for breeding, condition
of the hair/hooves, duration of the skin lesions, presence
or absence of white markings or pruritus in the foot
373

374

F Geburek et al.

Table 1. Clinical data, sites and numbers of biopsy specimens and macroscopic lesions on the feet of 37 heavy draught horses with chronic
pastern dermatitis
Case no.

Cold blood breed

Age (years)

Sex

Cannon circumference
(in cm)

Biopsied
foot/feet

Total number
of biopsies

Macroscopic
lesion

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
C1
C2

Mecklenburgisch*
Rhenisch German
Rhenisch German
Westphalian
Lower Saxon
Unknown
Westphalian
Altmrkisch*
Rhenisch German
Westphalian
Rhenisch German
Rhenisch German
Westphalian
Westphalian
Belgian Draught
Mecklenburgisch
Rhenisch German
Mecklenburgisch
Lower Saxon
Belgian Draught
Rhenisch German
Percheron
Percheron
Percheron
Rhenisch German
Westphalian
Westphalian
Saxon
Westphalian
Rhenisch German
Rhenisch German
Polish
Saxon
Altmrkisch
Lower Saxon
Westphalian
Rhenisch German
Altmrkisch
Rhenisch German

8
6
9
5
2
5
3
17
4
3
5
6
4
4
10
7
5
7
3
6
9
15
13
9
19
7
10
13
8
8
7
14
7
8
9
10
12
3
3

MC
MC
F
F
F
MC
M
F
M
M
M
M
M
M
M
F
M
MC
F
M
M
M
M
MC
MC
M
M
M
M
M
M
F
M
M
F
M
F
M
M

27.0
32.0
27.5
28.25
26.0
26.75
28.0
31.0
29.0
27.5
31.0
33.0
28.0
27.5
31.0
29.0
32.0
28.75
26.0
32.0
32.5
34.5
33.5
34.0
31.25
33.0
29.5
37.5
31.5
29.5
31.5
24.5
33.5
31.25
27.0
32.0
37.0
27,5
28.5

LH, RH
LH
RH
RH
LH
LF, RH
RH
RH
LH
RH
LH
LH
LH
LH
LH
RH
RH
RH
LH
RF
RH
LH
LH
RH
RH
LH
RH
RH
LH
LH
LH
LH
LH
RH
RH
LH
RH
LH
RH

3
2
2
2
2
4
2
2
1
1
2
1
2
2
2
2
2
2
2
3
2
2
2
2
2
2
1
2
2
2
2
2
1
2
2
3
1
1
1

Sc
Sc
Sc
Sc
Sc
Hhp
Hhp
Hhp
Hhp
Hhp
Hhp
Hhp
Hhp
Hhp
Hhp
Hhp
Hhp
Hhp
Hhp
Nm
Nm
Nm
Nm
Nm
Nm
Nm
Nm
Nm
Nm
Nm
Nm
Nm
Nm
Nm
Nm
Vm
Vm
None
None

M, male; MC, male castrated; F, female; LH, left hind; RH, right hind; LF, left front; RF, right front; C, control horse; *East German
Coldblood; Sc, scaling; Hhp, hyperkeratotic/hyperplastic plaques; Nm, nodular masses; Vm, verruccous masses.

region and presence or absence of other skin lesions.

The cleanliness of stable facilities and quality of the
air and light were graded subjectively as good = 1,
moderate = 2 or bad = 3. The ground where horses
were kept outdoors was recorded as pasture/soil, sand,
or rubber meadows. Coat condition was graded subjectively as good = 1 (clean, smooth, shiny), moderate = 2
(partly dirty, partly rough, partly dull) or bad = 3 (dirty,
rough, dull). The condition of the hooves was graded
as good = 1 (clean, no trimming necessary, little or no
thrush) or bad = 2 (dirty, trimming necessary, noticeable
thrush).
A scoring system was developed for evaluating the
pastern skin lesions according to their type, extent and
severity. The clinical extent and macroscopic appearance of skin lesions were recorded and documented by
preparing schematic drawings and by colour photography.
Parameters recorded were erythema, exudation, crust
formation, scaling, hyperkeratotic and hyperplastic

plaque-like lesions, erosions/excoriations, greasiness and

malodorous skin surface, nodular masses, or verrucous
masses with rugged surfaces. Each of these nine features
(if observed) was multiplied by 13, depending on the
severity of the lesion (1 = mild, 2 = moderate, 3 = severe).
These were added together to give a total severity score
for each horse. The cannon circumference of all four
feet was measured in centimetres with a tape measure
at the smallest circumference of the metacarpal and
metatarsal bone immediately beneath the carpal or
tarsal joint. The average cannon circumference was
then calculated. Normal anatomical structures, i.e.
fetlock tufts of hairs (feathering), ergots, chestnuts
and bulges in the fetlock region were recorded. Skin
scrapings were examined for Chorioptes sp.

Biopsy specimens
From each horse, at least one punch biopsy of 6 mm in
diameter was taken from the diseased skin. The location

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 373384

Pastern dermatitis in heavy draught horses

375

Table 2. Antibodies used for immunohistochemical staining

Antibody

Clone or identification

Specificity

Dilution

Source

Anti-Ki-67
Anti-CK5/6(4)
Anti-CK10
Anti-CK14
Anti-CD3
Anti-CD79a
Anti-MHC II

MIB-1
D5/16B4
DE-K10
LL002
A0452
HM57
H42A

Human Ki-67
Human CK 5/6(4)
Human CK10
Human CK14
Human CD3
Human CD79a
Equine MHC II

1 : 75
1 : 3000
1 : 80
1 : 1000/1 : 50*
1 : 300
1 : 60
1 : 120

Dako
Dako
Dako
Novocastra, Newcastle upon Tyne, UK
Dako
Dako
VMRD, Pullman, Washington, USA

*In biopsies with moderate and severe epidermal hyperplasia the 1 : 1000 dilution resulted in a very weak staining intensity. Therefore, the
sections were also stained with a lower antibody dilution of 1 : 50.

and total number of biopsies taken from the 37 cases

are given in Table 1. Tissue specimens were fixed in 4%
neutral buffered formaldehyde, embedded in paraffin
wax, cut at 4 m, mounted on coated slides (SuperFrost
Plus, Menzel-Glser, Wiesbaden, Germany) and stained
with haematoxylin and eosin or cresyl echt violet for
demonstration of mast cells. Selected sections were
stained with by the Warthin Starry silver staining
method. For comparison, from all 37 horses one 6-mm
punch biopsy was taken from the macroscopically
normal skin of the lateral neck. Control 6-mm skin
biopsies were included from two heavy draught horses
without clinical dermatological abnormality. These
were taken from the pasterns (C 1 and C 2 in Table 1)
and from the neck of horse C2.

Antibodies and immunohistochemical labelling

procedures
Antibodies used for immunohistochemical labelling
are shown in Table 2.
Paraffin sections were dewaxed and rehydrated. Endogenous peroxidase activity was blocked by incubation
for 30 min in a solution of 0.5% H2O2 in 85% ethanol.
For antigen retrieval, sections stained with antibodies
to Ki-67 or cytokeratins were pretreated by heating in
a microwave oven for 2 5 min in citric acid buffer
(0.01 mol L1, pH 6.0). For detection of CD3 and CD79a
antigens, sections were pretreated with unmasking
fluid G (Biologo, Kronshagen, Germany) for 10 min at
95 C. Sections stained with antibodies to MHC class
II antigens were incubated for 15 min at room temperature (rt) in a 0.25% solution of Triton X-100, rinsed in
phosphate-buffered saline (PBS), and pretreated with
a 0.05% pronase E (Merck, Darmstadt, Germany)
solution for 20 min at 37 C. Following incubation with
inactivated normal goat serum for 20 min at rt and
incubation with the primary antibodies (1820 h at
4 C) biotin-conjugated goat-antimouse IgG (H + L)
or goat-antirabbit IgG (H + L) (both from Vector
Laboratories, Burlingame, USA), diluted 1 : 200 in
PBS containing 10% inactivated normal horse serum
were used as second antibodies for 30 min at rt. After
incubation with the ABC solution (Vectastain Elite
ABC Kit, Vector Laboratories, Burlin-game, CA,
USA) for 30 min at rt, peroxidase activity was detected
by incubation in 3-amino-9-ethylcarbazol solution
(AEC-Plus, Dako, Hamburg, Germany). Sections were

counterstained with Mayers haematoxylin. Negative

control sections, in which the primary antibody was
replaced by appropriately diluted normal mouse or
rabbit serum, were included in all staining runs.
Positive control tissues were equine lymph node for
CD3, CD79a, MHC II and Ki-67 antigens and normal
human and equine skin for CK5/6(4), CK10 and
CK14.

Statistical analysis
SAS software (version 6.04, Statistical Analysis Institute, Cary, NC, USA) was used for statistical analysis.
The clinical and morphological data were either direct
and quantitative (age, cannon circumference) or qualitative, grouped or graded (hygienic condition of the
stable, character of the ground where the horses were
kept on outdoors, feeding, amount of work, use of
stallions for breeding, grooming condition of the hair/
hooves, white markings in the foot region, fetlock tufts
of hairs, ergots, chestnuts, anatomically normal bulges
in the fetlock region, pruritus). The previously mentioned
parameters were correlated using Spearmans rank
correlation.
Differences in clinical severity for paired data were
compared with Wilcoxons signed-rank test, i.e. use of
stallions for breeding, grooming condition of hooves,
anatomically normal bulges in the fetlock region, white
markings in the foot region, pruritus and frequency of
cleaning stables. KruskalWallis test was used when
more than two groups had to be compared, i.e. sex,
amount of work, grooming condition, type of ground
horses were kept on. Means and standard deviations
were calculated. Differences were considered significant at P < 0.05.
Immunolabelled cells were counted in pastern skin
sections and compared to the number in skin sections
from the neck. Immunolabelling for Ki-67 was quantified by counting basal epidermal cells with Ki-67positive nuclei in interfollicular epidermal areas in five
high-power fields (HPF) at 400 magnification. The
field of view was 600 m. Mast cells, CD3-, CD79a- and
MHC class II-positive cells were counted in five HPF
(400) in the upper dermis. The fields were selected to
contain perivascular inflammatory foci. The mean
numbers of positively labelled nuclei or cells, respectively, were documented, standard deviations were
calculated and correlation was tested by Spearmans

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 373384

376

F Geburek et al.

Figure 1. Fetlock region with scaling (group I, case no. 16).

Figure 2. Hyperkeratotic and hyperplastic plaque-like lesions

(arrow head) in the carpus region (group II, case no. 16).

rank correlation. To compare differences between

numbers of cells, Wilcoxons signed-rank test was used.

R ESU LTS
Clinical and macroscopic findings
The data obtained from the owners or animal attendants indicated that in the majority of horses (n = 17),
the skin lesions had been present for at least 2 years. In
10 cases, the lesions had not been noted or recognized
by the owners or animal attendants as pastern dermatitis.
Some horses had additional dermatological diseases
such as facial alopecia (case no. 21), sweet itch (case
nos 3, 23, 25, 35) or inflammation of elbows and knees
(case no. 15). Hoof disease (case no. 6) and ulcerating
subsolar abscesses (case no. 1) were each found in one
horse. All other horses were clinically healthy apart
from the pastern dermatitis.
Based on macroscopic findings, the pastern skin
lesions were divided into four different groups (Table 1).
In horses of group I (n = 5), scaling was the predominant
sign (Fig. 1). Lesions in horses of group II (n = 14)
were characterized by hyperkeratotic and hyperplastic
plaque-like lesions (Fig. 2) as well as areas with scaling.
In horses of group III (n = 16), nodular skin masses
were present (Figs 3 and 4), and horses of group IV
(n = 2) were affected by verrucous skin lesions with

rugged surfaces (Figs 5 and 6). Horses of groups III

and IV also had scaling and plaque-like lesions and
greasy, malodorous skin with focal excoriations/erosions
(Fig. 6).
Statistical analysis revealed a significant correlation
between the severity of skin lesions and age (P = 0.0005),
increasing mean cannon circumference (P < 0.0001)
and the prominence of anatomical structures such as
fetlock tufts of hairs (P = 0.0012), ergots (P = 0.0002)
and chestnuts (P = 0.0002). Clinical disease was worse
in horses with anatomically normal bulges in the fetlock
region (Fig. 3) (P < 0.001). Clinical disease was also
worse in horses with poor hoof condition (P = 0.0091).
Horses kept in outside pens on rubber meadows were
less affected than those kept on sand or soil (P = 0.037).
As stable hygiene and quality decreased, clinical
severity increased (P = 0.0203). Sex, breed, amount
of work, use for breeding, grooming and white markings did not have any significant correlation with the
dermatitis.
Twenty-seven of 37 (73%) horses with pastern
dermatitis were pruritic. In 22 of 37 horses (60%),
Chorioptes sp. skin mites were found in scrapings but
there was no statistically significant correlation
between the incidence of mites and severity of pastern
dermatitis.

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 373384

Pastern dermatitis in heavy draught horses

Figure 3. Hind limb with bulge (arrow) in the fetlock region and
multiple nodular skin masses (group III, case no. 34).

Histopathology
In all cases, the main histopathological finding was
epidermal hyperplasia and hyperkeratosis. The degree
of hyperplasia increased from group I to IV and was
most pronounced in groups III and IV (Fig. 7).
Group I. In horses of this group, there was slight-tomoderate epidermal hyperplasia with variation between
samples obtained from different feet. In all five horses
of this group, slight-to-moderate orthokeratotic hyperkeratosis was present, which, in three cases, was associated with focal areas of parakeratosis. In three horses
of this group, serocellular crusts contained bacterial
colonies, Chorioptes sp. skin mites and numerous, mostly
degenerate neutrophilic granulocytes. There was slight
focal spongiosis of the epidermis in three cases and in
one of them intraepithelial pustules containing degenerate neutrophilic granulocytes were seen.
Group II. In horses of this group, slight to moderate
epidermal hyperplasia with formation of rete pegs and
slight to moderate spongiosis were present. The degree
of epidermal hyperplasia varied between different horses
and between samples obtained from different feet. In
all cases, epidermal hyperplasia was accompanied by
orthokeratotic hyperkeratosis with foci of parakeratosis.

377

Figure 4. Nodular skin masses with focal erosions (group III, case
no. 28).

Keratohyalin granules were reduced in number or

were absent.
Serocellular crusts with numerous, mostly degenerate,
neutrophilic granulocytes and bacterial colonies were
seen in four horses. In eight cases, Chorioptes sp. skin
mites were present within the crusts.
Group III. In 13/16 horses, moderate to marked epidermal hyperplasia was present, whereas in the remaining
three horses, this was slight or slight to moderate with
prominent rete pegs. All had slight to moderate focal
spongiosis and variable orthokeratotic hyperkeratosis,
which was accompanied by focal parakeratosis in nine.
There was focal to total absence of keratohyalin granules.
Serocellular crust contained bacterial colonies and
numerous, mostly degenerate, neutrophilic granulocytes
in six horses. In three horses, Chorioptes sp. skin mites
were seen. In three cases, epidermal erosions or ulcerations
were present. Intraepithelial pustules containing degenerate neutrophilic granulocytes were found in five cases.
Group IV. Both horses of this group showed marked
epidermal hyperplasia with formation of prominent
rete pegs and prominent dermal papillae. Keratohyalin
granules were reduced or absent and there was slight to
moderate spongiosis. Moderate orthokeratotic hyperkeratosis was accompanied by focal parakeratosis in

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 373384

378

F Geburek et al.

Figure 5. Early verrucous skin lesions with greasy, malodorous

coating (group IV, case no. 37).

Figure 6. Verrucous skin lesions (group IV, case no. 21), 2.5 years
after first examination.

one horse. Serocellular crusts with bacterial colonies

and numerous, mostly degenerate, neutrophilic granulocytes were seen in one horse.
In the superficial dermis of all cases, there was a
slight (n = 20), slight to moderate (n = 7) moderate to
severe (n = 9) or severe (n = 1) perivascular lymphocytic
dermatitis with smaller numbers of plasma cells,
macrophages, mast cells and eosinophilic granulocytes.
Fibrosis in the superficial dermis was characterized
by an increase in broad collagen fibre bundles. In comparison to skin samples from the neck and from normal
pasterns, dermal fibrosis was slight to moderate in
groups I to III and moderate to marked in the two
horses of group IV. There was accompanying dilatation
of lymphatics, few to several in groups I to III and
numerous in group IV. No vasculitis or granulation
tissue was seen in any horse nor were spirochaete-like
organisms seen in silver-stained sections.
In the dermis of the pasterns of the two healthy heavy
draught horses, there was a mild perivascular infiltrate
of lymphocytes. The macroscopically normal skin from
the neck of horses with pastern dermatitis had mild
lymphocytic exocytosis in the epithelium in 15 cases. In
21/37, the lymphocytes were located within the basal
epidermal layers and only single lymphocytes were found
in the stratum spinosum. In 35 cases, lymphocytic
exocytosis was accompanied by a slight and in two cases

by moderate perivascular dermatitis with lymphocytes

and, in some cases, single or few plasma cells, macrophages or eosinophilic granulocytes. The number of
cresyl echt violet stained mast cells in diseased pasterns
and macroscopically normal neck skin were not significantly different (Fig. 8).

Immunohistochemistry
Ki-67. Ki-67-positive cells were located predominantly
in the basal epidermis but in several horses, occasional
Ki-67-positive nuclei were located in the suprabasal
layers. Ki-67-positive proliferating cells from diseased
pasterns were significantly increased (P = 0.001) (Ki67-positive cells 53.54 48.88) compared with those
from the neck (9.77 9.12). The number of Ki-67positive cells correlated with the clinical severity (P =
0.009) but there was no significant correlation between
the disease group and number of Ki-67-positive cells.
CD3- and CD79a-positive cells. The mean numbers of
CD3-positive T lymphocytes and of CD79a-positive
cells are shown in Fig. 8. In the superficial dermis, significantly higher numbers of CD3-positive T lymphocytes
than CD79a-positive cells were present (P < 0.0001). The
majority of CD79a-positive cells were lymphocytes (95%)
and only few (5%) plasma cells. Single lymphocytes

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 373384

Pastern dermatitis in heavy draught horses

379

Figure 7. Histopathological findings in skin sections from heavy draught horses with chronic pastern dermatitis. (a) Skin section from a horse
with scaling (group I, case no. 4). There is orthokeratotic hyperkeratosis, slight to moderate epidermal hyperplasia and a Chorioptes sp. mite
(arrow). Haematoxylin and eosin, bar = 200 m. (b) Skin section from a horse with hyperplastic/hyperkeratotic plaques (group II, case no. 16).
There is marked ortho- and parakeratotic hyperkeratosis and acanthosis of the epithelium. In the dermal tissue, several dilated lymph vessels
(arrows) are present. Haematoxylin and eosin, bar = 200 m. (c) Skin section from a horse with nodular skin masses (group III, case no. 22).
Marked orthokeratotic hyperkeratosis and acanthosis of the epidermis. Haematoxylin and eosin, bar = 200 m. (d) Skin section from a horse
with verrucous skin lesions (group IV, case no. 36). Prominent epidermal hyperplasia with acanthosis and formation of rete pegs, and numerous
dilated lymph vessels (arrows) within the fibrotic dermal tissue. Haematoxylin and eosin, bar = 800 m.
2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 373384

380

F Geburek et al.
However, there were significantly more MHC class IIpositive keratinocytes in diseased epidermis than in
normal neck epidermis (P = 0.0004). In the superficial
dermis, MHC class II-positive inflammatory cells or
with dendritic morphology were predominantly perivascular. More were seen in diseased pasterns than in the
neck (P = 0.0013).

Figure 8. Quantification of CD3-positive T cells, CD79a-positive B

cells and mast cells counted in five high-power fields. The arithmetic
mean number of cells is reported for skin biopsies from the neck and
feet of horses with pastern dermatitis and from biopsies from the feet
of two healthy draught horses. Error bars = standard deviation.

Cytokeratins. In normal human skin, CK5/6(4) and

CK14 labelling is confined to the basal epidermis. In
equine neck and normal pastern skin, weak staining was
also seen in the cytoplasm of suprabasal keratinocytes.
CK10 was seen exclusively in suprabasilar epithelium
in normal human and equine skin. In the pasterns of
horses with mild epidermal hyperplasia, the labelling
pattern of CK5/6(4), CK14 and CK10 was similar to
that of normal skin. With moderate or marked epidermal
hyperplasia, CK5/6(4) and CK14 labelling was decreased
in the basal layer and increased in suprabasal keratinocytes. In contrast, CK10 labelling decreased in suprabasal keratinocytes as epidermal hyperplasia increased
(Fig. 10).

D ISC U S S IO N

Figure 9. Quantification of epidermal and dermal cells expressing

MHC class II antigen counted in five high-power fields. The
arithmetic mean number of cells is reported for skin biopsies from
the neck and feet of horses with pastern dermatitis and for biopsies
from the feet of two healthy draught horses. Error bars = standard
deviation.

infiltrating the epidermis were CD3-positive. There was

no significant correlation between the number of CD3and CD79a-positive lymphocytes and the group or
clinical severity of skin lesions.
MHC class II staining. MHC class II antigen stainingcells were seen in the epidermis and dermis of almost
all horses with pastern dermatitis (Fig. 9). In the
epidermis of the feet and neck, granular cytoplasmic
staining occurred in numerous keratinocytes of the
spinous and cornified layers. No membrane staining of
keratinocytes was seen and few labelled cells resembling
Langerhans cells were seen in the epidermis. Counting
of MHC class II-positive Langerhans cells in the
epidermis was difficult because of the presence of
numerous MHC class II-positive keratinocytes, although
the number of Langerhans cells in diseased epidermis
did not appear to differ from those in normal skin.

Chronic pastern dermatitis was divided into four groups

based on macroscopic appearance. Histological
examination of skin biopsies showed that the main
histopathological feature was epidermal hyperplasia
with predominantly orthokeratotic hyperkeratosis.
There was a gradual increase in epidermal hyperplasia
with the severity of macroscopic appearance, which
suggests that the lesions represent different stages of
the disease. To confirm this, we plan longer-term follow
up studies with repeat biopsies from individual patients.
Immunohistochemistry for Ki-67 antigen indicated
increased germinative cells in the basal epidermis,
correlating with the clinical severity of the lesion. This
suggests that epidermal hyperplasia and hyperkeratosis in
these horses is the result of increased epidermal proliferation. The positive correlation between the severity or
stage of skin lesions and the circumference of the
cannon is in agreement with earlier studies.10,11 Additionally, the present study indicates a correlation with
the prominence of other anatomical structures such as
fetlock tufts of hairs, chestnuts, ergots and, if present,
bulges in the fetlock region. These findings may indicate that heavy draught horses with these anatomical
structures are predisposed to epithelial hyperproliferation and ultimately, nodular or verrucous lesions.
Bulges in the fetlock region are anatomically normal
skin folds that seem to be prominent especially in
heavy horses.12 Following the NAV (Nomina anatomica
veterinaria), they can be denoted as plicae cutis. These
structures might be confounded with calluses occurring
as a symptom of chronic idiopathic pastern dermatitis
(German maukeschwielen).13
In the epidermis, basal cells undergo a defined
process of differentiation and maturation. During this

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 373384

Pastern dermatitis in heavy draught horses

381

Figure 10. Immunohistochemical staining pattern for cytokeratins

in skin sections from heavy draught horses with chronic pastern
dermatitis. ABC method, AEC substrate, Mayers haematoxylin
counterstain. (a) Staining for cytokeratin 5/6(4) (case no. 17). In
comparison to normal skin from the neck, there is a complete shift of
cytoplasmic staining of keratinocytes with negativity of the stratum
basale (Sb) and increased cytoplasmic staining of the stratum
granulosum (Sg). Bar = 200 m. (b) Staining for cytokeratin 14 (case
no. 36). As for cytokeratin 5/6(4), a shift of cytoplasmic keratinocyte
staining is noted. There is marked decrease of staining in the stratum
basale (Sb) and stratum spinosum (Ss) and increased
immunoreactivity of the stratum granulosum (Sg). Bar = 100 m.
(c) Staining for cytokeratin 10 (case no. 15). There is decreased
cytoplasmic keratinocyte immunoreactivity in the stratum spinosum
(Ss). Bar = 50 m.
2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 373384

382

F Geburek et al.

differentiation process, cytokeratins are synthesized

and they are a major component of the intracellular
cytoskeleton of epidermal cells. In the skin, CK5 and
CK14 are exclusively synthesized by basal cells, whereas
differentiation-specific cytokeratins, i.e. CK1 and CK10,
are limited to suprabasal keratinocytes.14 The changes
of cytokeratin immunoreactivity observed with antibodies to cytokeratins CK5/6(4), CK10 and CK14 in
this study indicate that abnormal differentiation of
keratinocytes occurs with hyperplasia of the epidermis.
In human diseases such as psoriasis with hyperplastic
epidermal changes,15,16 there is expression of proliferationassociated cytokeratins (e.g. CK6, CK16 and CK17).
Future studies in horses should therefore examine for
these cytokeratins.
The aetiology of the chronic pastern dermatitis
remains undetermined. Several authors suggest that
genetic factors may play a role.9,10,1719 Our study indicates that anatomical features, which are genetically
determined, correlate with disease severity.
Other potential causes include moisture and
dampness9,12,17 confirmed by our finding that horses
kept on rubber meadows had less severe clinical disease
than those kept on sand or soil. In contrast to some
authors,2,6 we did not find that infrequent and poor
general grooming increased clinical severity although
quality of hoof care correlated with lesion severity.
Hoof canker is reported in heavy draught horses with
poorly and infrequently groomed hooves and has
clinical and histopathological parallelisms to chronic
idiopathic pastern dermatitis.20,21
In contrast with some authors,6,18,22 we could not find
a significant correlation between the clinical severity
and white markings in the region. Photodynamic
mechanisms did not seem to be involved because white
markings on the heads were unaffected.
Horses in our study were pruritic and were automutilated by biting and rubbing, as found by other
authors.1,9 Direct or indirect irritation by Chorioptes
mites is the most probable cause. In 60% of study horses,
Chorioptes mites and, in several cases, bacterial colonies were present. Infestation with mites might be a
consequence of proliferative epidermal changes and
production of excess keratin providing nutrition and a
suitable living environment for the parasites. Mite
infestation might lead to further epidermal proliferation and the lesions may have been induced by early or
continuous stimulus by the mites. Bacteria or other
unknown infectious agent or its antigens may also be
implicated. Such stimuli may cause inflammation with
T-lymphocyte secretion of cytokines IL-1 and IL-6,
which cause epidermal hyperplasia.14 Epidermal growth
factor (EGF) and transforming growth factor-
(TGF-), have significant roles in the regulation of
keratinocyte proliferation14 and up-regulation is
induced by secretion of IFN- and INF- by inflammatory cells. There was mononuclear perivascular
dermatitis in all horses with chronic pastern dermatitis,
so up-regulation of growth factors may be involved in
the epidermal hyperproliferation. In future investiga-

tions of the pathogenesis of this disease, expression of

cytokines and growth factors should be examined.
In all cases, there was perivascular dermatitis
dominated by T lymphocytes. No increase in mast cells
or eosinophils was noted but this does not exclude
hypersensitivity involving Langerhans and T cells.
MHC class II-positive keratinocytes were present in
the pastern and neck epidermis of all horses. Staining
was cytoplasmic and granular not surface expression
of MHC class II. Cytoplasmic granular immunoreactivity for MHC class II antigen has been described in
equine squamous cell carcinomas where it is suggested
it may facilitate local antitumour response.27 It has also
been noted in acanthotic and neoplastic keratinocytes
of sheep exposed to high levels of sunlight for several
years.27,28 The relevance of MHC class II immunoreactivity in keratinocytes of equine hyperplastic and normal
skin in our study is unclear. Immunoreactivity of our
antibody in control equine lymph node closely resembled the staining pattern described by other authors,27
who used a cross-reactive antibody to human MHC
class II. It therefore seems unlikely that the keratinocyte
labelling represents a nonspecific immunohistochemical
reaction. Surprisingly, fewer keratinocytes stained in
hyperplastic than normal skin. The reason is unknown.
More MHC class II-positive dendritic-like cells were
seen in the superficial dermis in diseased skin, which
may indicate that these cells are presenting antigens to
T lymphocytes.
Spirochaetes were found in a Tennessee Walking
Horse with papillomatous pastern dermatitis4 and
spirochaetes of the genus Treponema play a role in
papillomatous digital dermatitis in cattle.23 These
organisms could also play a role in the pathogenesis of
chronic proliferative pastern dermatitis. Although no
spirochaete-like organisms were found on silver-stained
sections in this investigation, further studies with
immunohistochemistry or polymerase chain reactions
are indicated.
Recently, the clinical signs and pathological skin
changes in draught horses have been compared with a
condition in humans known as chronic progressive
lymphoedema or elephantiasis nostras verrucosa. This
is characterized by dermal fibrosis and dilation of
lymphatic vessels.24 Dermal fibrotic changes and dilated
lymphatic spaces seen in the majority of the horses of
the present investigation may indicate that insufficient
lymphatic drainage could be involved in the development of the disease. However, horses in groups I, II and
III had few dilated lymphatics, in contrast to human
patients with elephantiasis nostras verruccosa.25,26
In conclusion, this study indicates that heavy draught
horses of different breeds are predisposed to develop
alterations in proliferation and differentiation of keratinocytes in their pastern skin, resulting in progressive
epidermal hyperplasia and production of excess
keratin. Future studies should focus on the possible
role of cytokines and growth factors in the development
of this disease and the possible relevance of genetic
factors and infectious organisms such as spirochaetes.

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 373384

Pastern dermatitis in heavy draught horses

ACKN OWLEDGE ME NT S
The authors are very grateful to Dr H.W. Meyer, Dr R.
Scharnhlz, Dipl. ing. agr. Susanne Rimkus and to
all horse owners. The assistance of Dr K. Rohn in
performing the statistical analyses is gratefully
acknowledged.

REFEREN CE S
1. Pascoe RRR, Knottenbelt DC. Iatrogenic and idiopathic
disorders. In: Pascoe RRR, Knottenbelt DC eds. Manual
of Equine Dermatology, 1st edn. London, Edinburgh:
W.B. Saunders, 1999: 2003.
2. Stannard T. Stannards illustrated equine dermatology
notes. Miscellaneous. Veterinary Journal of Dermatology
2000; 11: 2213.
3. Scott DW, Miller WH. Miscellaneous skin diseases. In:
Scott DW, Miller WH eds. Equine Dermatology. St.
Louis, Missouri: Saunders, 2003: 6879.
4. Rashmir-Raven AM, Black SS, Rickard LG et al.
Papillomatous pastern dermatitis with spirochetes and
Pelodera strongyloides in a Tennessee Walking Horse.
Journal of Veterinary Diagnostic Investigation 2000; 12:
28791.
5. Schutzer E. Ein Beitrag zur Kenntnis der Dermatitis
Chronica Verrucosa in der Fesselbeuge des Pferdes
(Sogenannter Straubfu). Dissertation, Germany: University of Leipzig, 1909: 46.
6. Gustine G. Die sogenannte warzenmauke des pferdes
(Dermatitis chronica verrucosa). Monatshefte fr Praktische Tierheilkunde 1910; 21: 51037.
7. Hieronymi E. Haut. In: Joest E ed. Spezielle Pathologische Anatomie der Haustiere. Berlin: Verlag Schoetz,
1924: 44250.
8. Hll A. Erfahrungen ber Dermatitis chronica verrucosa
(Warzenmauke). Berliner und Mnchener Tierrztliche
Wochenschrift 1956; 69: 46971.
9. Duclos P. Les Eau-aux-jambes du Cheval: Dermatose
Hyperplastique des Membres. Dissertation, France:
National Veterinary School of Lyon, 1972.
10. Schper W. Konstitutionelle hauterkrankungen beim
pferd. Zeitschrift fr Tierzchtung und Zchtungsbiologie
1937; 37: 295330.
11. Weischer F. Zur klrung der erb- und umweltbeziehungen
der warzenmauke des pferdes. Tierrztliche Umschau
1949; 4: 31820.
12. Rottlnder W. Beitrge zur Aetiologie des Ekzems in
der Fesselbeuge beim Pferde (sogenannte Mauke).
Dissertation, Germany: Faculty of Medicine, University
of Leipzig, 1908.
13. Kumpf J. Beitrag zur Behandlung der Warzenmauke des
Pferdes. Dissertation, Germany: Humboldt University
Berlin, 1953.

383

14. Suter MM, Crameri FM, Olivry T et al. Keratinocyte

biology and pathology. Veterinary Dermatology 1997; 8:
67100.
15. Weiss RA, Eichner R, Sun TT. Monoclonal antibody
analysis of keratin expression in epidermal diseases: a
48 kD and a 56 kD keratin as molecular markers for
hyperproliferative keratinocytes. Journal of Cell Biology
1984; 98: 1397406.
16. Leigh IM, Navsaria H, Purkis PE et al. Keratins (K16
and K17) as markers of keratinocyte hyperproliferation
in psoriasis in vivo and in vitro. British Journal of Dermatology 1995; 133: 50111.
17. Zarrouk K, Huard R, Krichen R. Les eaux-aux jambes
des equids: a propos des premiers cas tunisiens. Archs
Inst. Pasteur Tunis 1979; 56: 35670.
18. Silbersiepe E, Berge E, Mller H. Krankheiten an
fesselgelenk und fessel. In: Silbersiepe E, Berge E, Mller H,
eds. Lehrbuch der Speziellen Chirurgie fr Tierrzte und
Studierende. Stuttgart: Ferdinand Enke Verlag, 1986:
3424.
19. Wallraf A, Hamann H, Deegen E et al. Untersuchung
zur prvalenz von mauke bei deutschen kaltblutpferderassen. Berliner und Mnchener Tierrztliche
Wochenschrift 2004; 117: 14852.
20. English M, Pollen S. Pastern dermatitis and unguilysis in
two draft horses. Equine Practice 1995; 17: 2530.
21. Brgi O. Untersuchungen ber den sog. Hufkrebs und
Straubfuss des Pferdes. Schweizer Archiv fr Tierheilkunde
1910; 52: 13761.
22. Ferraro GL. Pastern dermatitis in Shires and Clydesdales. Journal of Equine Veterinary Science 2001; 21:
5247.
23. Rijpkema SGT, David GP, Hughes SL et al. Partial identification of spirochetes from two diary cows with digital
dermatitis by polymerase chain reaction analysis of the
16S ribosomal RNA gene. The Veterinary Record 1997;
140: 2579.
24. Ferraro G. Chronic progressive lymphedema in draft
horses. Journal of Equine Veterinary Science 2003; 23:
18990.
25. Schissel DJ, Hivnor C, Elston DM. Elephantiasis nostras
verrucosa. Cutis 1998; 62: 779.
26. Vaccaro M, Borgia F, Guarneri F et al. Elephantiasis
nostras verrucosa. International Journal of Dermatology
2000; 39: 76073.
27. Prez J, Mozos E, Martin MP et al. Immunohistochemical study of the inflammatory infiltrate associated with
equine squamous cell carcinoma. Journal of Comparative Pathology 1999; 121: 38597.
28. Townsend WL, Gorrell MD, Ladds PW. Major histocompatibility complex antigens in normal, acanthotic
and neoplastic ovine skin: an association between
tumour invasiveness and low level MHC class I expression. Veterinary Immunology and Immunopathology
1995; 45: 23752.

Rsum Nous rapportons les donnes anamnestiques, cliniques et histopathologiques des lsions cutanes
observes sur les biopsies de 37 chevaux prsentant une dermatite chronique des paturons. Les lsions cutanes
taient de 4 types: squames (groupe I, n = 5), hyperkratose et lsions en plaques hyperplasiques (groupe II,
n = 14), masses nodulaires (groupe III, n = 16), et lsions verruqueuses (groupe IV, n = 2). Les modifications
histopathologiques principales taient une hyperkratose et une hyperplasie pidermique. Il existait une augmentation graduelle de lhyperplasie pidermique du groupe I au IV suggrant que les lsions reprsentent des stades
diffrents de la maladie. Dans tous les cas, on notait une dermatite privasculaire majorit de lymphocytes T
2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 373384

384

F Geburek et al.
avec une augmentation des cellules dendritiques CMH de classe II. Limmunohistochimie pour les cytokratines
CK5/6(4), CK10 et CK14 a montr une modification de leur expression. Ceci tait corrl avec lhyperplasie
pidermique, indiquant une diffrenciation anormale des keratinocytes. Une corrlation statistiquement significative a t observe entre la svrit des lsions cutanes et plusieurs facteurs, comme lge, la circonfrence du
cannon, la prominence de structures anatomiques comme les touffes de poils, les ergots et des fissures dans la
rgion du pied.
Resumen A continuacin describimos la historia y las caractersticas clnicas e histopatolgicas de las lesiones
de piel en 37 caballos de tiro con dermatitis crnica de las cuartillas. Las lesiones de la piel se dividieron en cuatro
grupos macroscpicos: descamacin (grupo I, n = 5); lesiones hiperqueratticas e hiperplsticas en placa (grupo
II, n = 14); lesiones nodulares (grupo III, n = 16), y lesiones verrucosas (grupo IV, n = 2). Los principales hallazgos histolgicos fueron hiperqueratosis e hiperplasia de la epidermis. Encontramos un incremento gradual en la
hiperplasia de la epidermis del grupo I al grupo IV, lo cual sugiere que las lesiones representan diferentes estados
del proceso. En todos los casos se observ dermatitis perivascular dominada por linfocitos T, con aumento de
las clulas dendrticas positivas para antgenos de clase II del complejo mayor de histocompatibilidad (MHC II).
La tincin mediante inmunohistoqumica para citoqueratinas CK5/6 (4), CK10 y CK14 indic un cambio en
las caractersticas de expresin. Este cambio estaba asociado con el grado de hiperplasia de la epidermis,
indicando una diferenciacin anmala de los queratinocitos. Se observ una correlacin estadsticamente significativa entre la magnitud de las lesiones de la piel y otros factores, incluidos mayor edad, mayor circunferencia
del tercer metatarsiano, y prominencia de estructuras anatmicas como las cernejas, espolones, espejuelos y
abultamientos en la regin del menudillo.
Zusammenfassung Wir beschreiben die anamnestischen, klinischen und histopathologischen Charakteristika
von Hautvernderungen in Biopsien von 37 schweren Zugpferden mit chronischem Fessel-Ekzem. Die
Hautvernderungen wurden in vier makroskopische Gruppen eingeteilt: Schuppenbildung (Gruppe I, n = 5),
hyperkeratotische und hyperplastische Plaque-hnliche Vernderungen (Gruppe II, n = 14), knotenfrmige
Hautmassen (Gruppe III, n = 16), und warzenartige Hautvernderungen (Gruppe IV, n = 2). Die hauptschlichen histologischen Befunde bestanden aus Hyperkeratose und epidermaler Hyperplasie. Es gab eine graduelle
Zunahme der epidermalen Hyperplasie von Gruppe I bis IV, was darauf hindeutet, dass die Vernderungen
verschiedene Erkrankungsstadien reprsentieren. In allen Fllen bestand eine perivaskulre Dermatitis, die
dominiert war von T Lymphozyten mit einer Zunahme an MHC Class II positiven Dendriten-hnlichen Zellen.
Immunhistochemische Frbung fr Cytokeratine CK5/6(4), CK 10 und CK 14 deutete eine Vernderung ihres
expression patterns (Suchmusters) an. Dieses korrelierte mit dem Ausma der epidermalen Hyperplasie, was
auf eine abnormale Differenzierung der Keratinozyten hindeutete. Es bestand eine statistisch signifikante
Korrelation zwischen dem Ausma der Hautvernderungen und verschiedenen anderen Faktoren wie
zunehmendem Lebensalter, zunehmendem Umfang des Rohrbeins, auffallenden anatomischen Strukturen wie
Haarbschel am Fesselgelenk, am Sporn und an den Kastanien, und Umfangsvermehrungen in der Fesselregion.

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 373384