Veterinary Dermatology 2005, 16, 47 51

Bacteria and fungi on the surface and within noninflamed hair

follicles of skin biopsy specimens from horses with healthy skin or
inflammatory dermatoses

Blackwell Publishing, Ltd.

CHRISTOPHER P. COOK*, DANNY W. SCOTT*, HOLLIS N. ERB and

WILLIAM H. MILLER Jr*
Departments of *Clinical Sciences,
Biomedical Sciences and Population Medicine and Diagnostic Services, College of Veterinary Medicine,
Cornell University, Ithaca, NY 14853, USA
(Received 7 April 2004; accepted 10 June 2004)

Abstract A retrospective study using light microscopy was performed to assess the prevalence of surface and
follicular bacteria and fungi in skin biopsy specimens from 247 horses with inflammatory dermatoses and from
27 horses with healthy skin. Cocci were found on the surface of specimens from 23% (95% confidence interval
18%, 29%) and 7% (95% confidence interval, 0%, 19%), respectively, of horses with skin disease and horses with
healthy skin. Of the nine dermatoses with at least 10 cases in our series of horses, bacterial folliculitis had a higher
prevalence of surface bacteria (57%; 95% confidence interval 34%, 81%) than the other eight (which all had a
prevalence < 30%). There was a significant association between the prevalence of surface cocci and the extent of
epidermal hyperkeratosis. Cocci were found in the keratin of noninflamed hair follicles in only 2% of the horses
with skin disease, and in none of the horses with healthy skin. Fungal poroconidia were found on the surface of
4% of the horses with skin disease, and on none of the horses with healthy skin. Yeasts were not found.

IN TRO D U CT I ON
It is generally accepted that bacteria and bacterial
microcolonies are commonly found in the surface
debris of skin biopsy specimens from animals with
dermatological disease. These microcolonies are often
of no diagnostic or therapeutic significance.13 Yeasts
(presumably Malassezia spp.) can also be found on the
surface of skin biopsy specimens from dogs and cats
with various skin disorders, and can be of no diagnostic
or therapeutic importance.2 A few studies have specifically addressed these issues in dogs4 and cats,5,6 but the
authors were unable to find any published studies that
examined the prevalence of micro-organisms by light
microscopy on the surface or in the hair follicles of
horses with skin disease, or in healthy skin of any species.
It is also generally accepted that animals with seborrheic skin (various abnormalities of keratinization)
have increased numbers of bacteria and yeast on their
skin, and that they may be predisposed to infections
with these micro-organisms.2,3 However, we could find
no published studies in any species, including humans,
wherein the light-microscopic prevalence of surface and
follicular bacteria and/or yeast in skin biopsy specimens
was evaluated in terms of the extent of hyperkeratosis.
Our aims in this study were: (1) to determine the
prevalence of bacteria and fungi on the surface and

Correspondence: D. W. Scott, Department of Clinical Sciences,

College of Veterinary Medicine, Cornell University, Ithaca, NY
14853, USA. E-mail: shb3@cornell.edu
2005 European Society of Veterinary Dermatology

within noninflamed hair follicles in skin biopsy specimens from horses with healthy skin or inflammatory
dermatoses; (2) to determine if there was a statistically
significant association between the prevalence of bacteria
and/or fungi and the degree of epidermal hyperkeratosis;
and (3) to determine if there was an association between
the prevalence of bacteria and/or fungi and particular
dermatoses.

M AT E R IA L S A N D M E T H O D S
A retrospective study of skin biopsy specimens from
247 horses with inflammatory skin disease was performed. All biopsy specimens had been submitted to
the Pathology Service at the College of Veterinary
Medicine at Cornell University between 1972 and
1987, processed routinely for histopathological evaluation, and stained with haematoxylin and eosin. All
diagnoses were confirmed by standard clinical and
pathological criteria (Table 1).3 In addition, 27 skin
biopsy specimens were taken from horses with healthy
skin that had been submitted to our necropsy service.
We wanted to mimic routine procedures in diagnostic pathology, so special stains and serial sections were
not examined. One section of tissue was examined per
slide. If multiple specimens were present on a slide, the
section with the least amount of artefact was chosen
for evaluation. One author (CC), who was blinded to
the diagnoses, examined all specimens for bacteria and
fungi on the epidermal surface and within the lumina
of noninflamed hair follicles. Specimens were also
47

48

CP Cook et al.

Table 1. Diagnosis-specific prevalences and confidence intervals for 247 horses with inflammatory dermatoses
Surface bacteria
Diagnosis
Eosinophilic granuloma
Insect hypersensitivity
Onchocerciasis
Bacterial folliculitis
Pemphigus foliaceus
Dermatophilosis
Atopy
Unilateral papular dermatosis
Vasculitis
Dermatophytosis
Bacterial pseudomycetoma
Erythema multiforme
Urticaria
Bacterial cellulitis
Adverse cutaneous drug reaction
Ventral midline dermatitis
Photodermatitis
Sterile granuloma
Contact dermatitis
Systemic lupus erythematosus
Discoid lupus erythematosus
Alopecia areata
Sterile panniculitis
Multisystemic eosinophilic
epitheliotropic disease
Demodicosis
Eumycotic mycetoma
Sporotrichosis
Habronemiasis
Primary seborrhea
Normal

No. of
horses

Yes

95%
CI (%)

29
23
21
21
18
16
13
11
10
8
7
6
7
5
5
5
5
5
4
4
4
4
3
3

8
5
3
12
4
4
2
1
3
1
2
0
1
1
0
0
2
1
1
1
1
1
1
0

28
22
14
57
22
25
15
9
30
12
29
0
14
20
0
0
40
20
25
25
25
25
33
0

2
2
2
2
2
27

1
0
0
0
1
2

50
0
0
0
50
7

Table 2. Grading of epidermal hyperkeratosis

Grade

Definition

0 (normal)

Thickness of stratum corneum thickness of

underlying epidermis
Thickness of stratum corneum > 1 to < 2 times
thickness of underlying epidermis
Thickness of stratum corneum 23 times thickness
of underlying epidermis
Thickness of stratum corneum > 3 times thickness
of underlying epidermis

1 (mild)
2 (moderate)
3 (severe)

Follicular bacteria

Prevalence
(%)

examined for the extent of epidermal hyperkeratosis.

Hyperkeratosis was graded on a scale of 03 (normal,
mild, moderate, severe, Table 2) by approximation.

Statistical analysis7
All statistical methods were performed using Statistix7
(Analytical Software, Tallahassee, FL, USA, 32317,
2000). Prevalences and 95% (corrected) confidence
intervals (CI) were calculated for the three yes/no variables (surface bacteria, follicular bacteria and surface
fungi) for each diagnosis. For all diagnoses combined
(including healthy), the hyperkeratosis score was
tested using Wilcoxons rank-sum test for difference
between horses having surface bacteria and horses not
having surface bacteria. This was repeated for follicu-

Surface fungi

Yes

Prevalence
(%)

95%
CI (%)

Yes

Prevalence
(%)

95%
CI (%)

10,46
3,41
0,32
34,81
0,44
1,49
0,39
0,31
0,63
0,42
0,69
0,8
0,47
0,61
0,10
0,10
0,93
0,61
0,80
0,80
0,80
0,80
0,100
0,17

0
0
0
3
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0

0
0
0
14
0
6
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0

0,2
0,2
0,2
0,32
0,3
0,21
0,4
0,5
0,5
0,6
0,7
0,8
0,7
0,10
0,10
0,10
0,10
0,10
0,12
0,12
0,12
0,12
0,17
0,17

1
0
1
0
1
1
0
1
1
0
0
0
0
0
1
0
1
0
1
0
0
0
0
0

3
0
5
0
6
6
0
9
10
0
0
0
0
0
20
0
20
0
25
0
0
0
0
0

0,12
0,2
0,16
0,2
0,19
0,21
0,4
0,31
0,34
0,6
0,7
0,8
0,7
0,10
0,61
0,10
0,61
0,10
0,80
0,12
0,12
0,12
0,17
0,17

0,100
0,25
0,25
0,25
0,100
0,19

0
0
0
0
0
0

0
0
0
0
0
0

0,25
0,25
0,25
0,25
0,25
0,2

0
0
0
0
0
0

0
0
0
0
0
0

0,25
0,25
0,25
0,25
0,25
0,2

lar bacteria and surface fungi. Tests were two sided,

and significance was inferred at P 0.05.

R E S U LT S
Cocci (about 1 m in diameter) were found on the skin
surface in 57 of 247 (23%; 95% CI: 18, 29%) specimens
from horses with inflammatory dermatoses (Table 1).
Cocci were present singly or in microcolonies (Fig. 1).
Cocci were seen in the keratin of noninflamed hair
follicles in only four of 247 (2%) specimens (three
horses with staphylococcal folliculitis; one horse with
dermato-philosis). Cocci were also found on the skin
surface of these four specimens. Yeasts were not found.
Among the nine dermatoses with at least 10 cases in
our series, all but bacterial folliculitis had similar
prevalences of cocci on the skin surface (Table 1). This
was demonstrated by the inclusion of each of the eight
prevalences within the 95% CI of the other common
dermatoses.
Fungal poroconidia (brown in colour; up to about
40 m long 8 m wide) were found on the surface
of nine of 247 (4%) specimens (one case each of
eosinophilic granuloma, onchocerciasis, dermatophilosis,
pemphigus foliaceus, unilateral papular dermatosis,
vasculitis, contact dermatitis, adverse cutaneous drug

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 47 5 1

Bacteria and fungi in equine skin biopsies

49

reaction, and photodermatitis) (Fig. 2). Cocci were

also present on the surface of three of these nine cases.
Of the 27 healthy skin specimens, bacteria were seen
on the surface in only two specimens (7%; 95% CI: 0,
19%). There were no follicular bacteria or surface or
follicular fungi (95% CI: 0%, 2%).
The diagnosis-specific prevalences and confidence
intervals are shown in Table 1. The hyperkeratosis
scores were higher for horses with surface bacteria
than for those without (Table 3); there was no association between the hyperkeratosis scores and the few
observed cases of follicular bacteria or surface fungi.
Figure 1. Photomicrograph: equine epidermal stratum corneum;
numerous cocci are present singly (white arrow) and in clusters
(black arrow). H&E. Bar = 10 m.

Figure 2. Photomicrograph: equine epidermis; fungal poroconidia

are present in the stratum corneum (arrow). H&E. Bar = 24 m.

Table 3. Hyperkeratosis scores of horses

(out of 274 total) with surface bacteria,
follicular bacteria or surface fungi

D IS C U S S IO N
In our study, cocci were found on the skin surface and
in the keratin of noninflamed hair follicles in 23% and
2%, respectively, of the skin biopsy specimens from
horses with various dermatoses. Similar studies revealed
prevalences of 3.6% and 1.6%, and 16% and 1%, in
dogs and cats, respectively.4,5 In the canine study,4 it was
noted that when cocci were found in noninflamed hair
follicles, every dog had a clinically relevant bacterial
skin infection requiring systemic antibiotic therapy.
In our study, there was no association between the
prevalence of surface or follicular bacteria and the
most common specific equine dermatoses except for
higher prevalence in (not surprisingly) bacterial folliculitis. There was a significant association between the
prevalence of surface (but not follicular) bacteria and
the extent of epidermal hyperkeratosis.
Yeasts were not seen on the skin surface or in the
hair follicles of the skin biopsy specimens from horses
with various dermatoses. This is consistent with the
paucity of reports of yeast (Malassezia) dermatitis in
horses.3 In dogs and cats, yeasts (presumably Malassezia
spp.) were seen on the skin surface or in hair follicles in
0.7% or 0.1% and in 1% or 0%, respectively, of the specimens.4,5 Malassezia dermatitis is more commonly
recognized in dogs and cats.2 In a retrospective study of
skin biopsy specimens from cats with dermatological
diseases,6 yeasts (presumably Malassezia spp.) were
present on the surface in 2.7% of the cats studied. The
authors of that paper indicated that the only histopathological change that was common to all yeastcontaining specimens was hyperkeratosis. However,
although 36% of the biopsy specimens from our horses
with various dermatoses had variable degrees of epidermal hyperkeratosis, no yeasts were observed.

Hyperkeratosis score

Micro-organism

No. of
horses

Minimum

Median

Maximum

P (ranksum test)

Surface bacteria*
Follicular bacteria
Surface fungi

59
4
9

0
0
0

1
0
0

3
3
3

< 0.0001
0.97
0.94

*The 75th percentiles for horses with and without surface bacteria were scores of 2 and 1,
respectively.
2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 4751

50

CP Cook et al.

Fungal poroconidia (presumably Alternaria spp.)

were found on the surface in 4% of our specimens from
horses with skin disease. The presence of poroconidia
was not associated (by inspection) with a particular
dermatosis or the extent of surface hyperkeratosis.
Fungal poroconidia have been seen in cytological preparations from horses with Trichophyton equinum dermatophytosis.8 Anecdotal reports indicate that these
poroconidia are seen fairly commonly in skin scrapings and cytological preparations from horses with
skin disease.8 Alternaria spp. and other saprophytic
fungi are routinely isolated from the skin and hair of
healthy horses.3
We were unable to find published reports in any species, wherein the light-microscopic prevalence of bacteria and/or yeast on the surface and/or in hair follicles
of healthy skin was determined. In our study, cocci were
seen on the surface of only 7% of specimens from horses
with healthy skin. Fungi were not seen on the surface.
Neither bacteria nor fungi were seen in hair follicles.
In conclusion, surface bacteria were seen in skin
biopsy specimens most often (but not always;
prevalence = 57%) in horses with bacterial folliculitis.
However, surface bacteria were also seen in many other
dermatoses and even in two of 27 healthy skin specimens. Surface bacteria presence was associated with
slightly higher hyperkeratosis scores (although the full
range in scores was observed in horses both with and
without surface bacteria). Follicular bacteria and surface
fungi were rare in horses with inflammatory dermatoses,

were not seen in the healthy skin specimens, and were

not associated with hyperkeratosis.

REFERENCES
1. Yager JA, Scott DW. The skin and appendages. In: Jubb
KVF, Kennedy PC, Palmer N eds. Pathology of Domestic
Animals, 4th edn, Vol. I. New York: Academic Press,
1993: 531738.
2. Scott DW, Miller WH Jr, Griffin CE. Muller and Kirks
Small Animal Dermatology, 6th edn. Philadelphia: W.B.
Saunders Co, 2001.
3. Scott DW, Miller WH Jr. Equine Dermatology. Philadelphia: W.B. Saunders Co, 2003.
4. Scott DW. Bacteria and yeast on the surface and within
noninflamed hair follicles of skin biopsies from dogs with
non-neoplastic dermatoses. Cornell Veterinarian 1992;
82: 37986.
5. Scott DW. Bacteria and yeast on the surface and within
noninflamed hair follicles of skin biopsies from cats with
non-neoplastic dermatoses. Cornell Veterinarian 1992;
82: 3717.
6. Mauldin EA, Morris DO, Goldschmidt MH. Retrospective study: the presence of Malassezia in feline skin
biopsies. A clinicopathological study. Veterinary Dermatology 2002; 13: 713.
7. Zar JH. Biostatistical Analysis, 3rd edn. Princeton: Prentice Hall, 1996.
8. Scott DW. Marked acantholysis associated with dermatophytosis due to Trichophyton equinum in two horses.
Veterinary Dermatology 1994; 5: 10510.

Rsum Une tude rtrospective histopathologique a t ralise pour dterminer la prvalence des bactries
et des champignons la surface de la peau et dans les follicules pileux, chez 247 cehavux souffrant de dermatose
inflammatoire et chez 27 chevaux sains. Des cocci ont t retrouvs en surface chez 23% (intervalle de confiance
de 95%, 18%, 29%) et 7% (intervalle de confiance de 95%, 0%, 19%), respectivement des chevaux avec et sans
dermatose. Pour les 9 dermatoses frquentes (plus de 10 cas), seules les folliculites bactriennes prsentaient des
taux plus levs de bactries en surface (57%; intervalle de confiance de 95% 34%, 81%) que les huit autres. Il
existait une association significative entre la prvalence des bactries en surface et ltendue de lhyperkratose
parakratosique. Des cocci ont t retrouvs dans la kratine des follicules pileux non inflamms dans seulement
2% des cas souffrant de dermatose, et jamais chez les chevaux sains. Des poroconidies fongiques ont t retrouves
la surface pour 4% des chevaux dermatose et jamais pur les chevaux sains. Des levures nont pas t retrouves.
Resumen Se realiz un estudio retrospectivo con microscopa ptica para valorar la prevalencia de bacterias
y hongos superficiales y foliculares en muestras de biopsia cutnea de 247 caballos con dermatosis inflamatorias
y de 27 caballos con piel sana. Se encontraron cocos en la superficie de muestras en el 23% (95% de intervalo de
confianza, 18%, 29%) y 7% (95% de intervalo de confianza, 0%, 19%), respectivamente, de caballos con enfermedad cutnea y caballos con piel sana. De las 9 dermatosis con al menos 10 casos en nuestras series de caballos,
slo las foliculitis bacterianas tenan una prevalencia superior de bacterias superficiales (57%; intervalo de confianza 95%, 34%, 81%) que los 8 restantes (los cuales no diferan con prevalencias <30%). Exista una asociacin
significativa entre la prevalencia de los cocos superficiales y la extensin de la hiperqueratosis epidrmica. Los
cocos se encontraron en la queratina de folculos no inflamados slo en 2% de los caballos con enfermedad
cutnea, y ninguno de los caballos con piel normal. Se encontraron poroconidios fngicos en la superficie del 4%
de los caballos con enfermedad cutnea, y en ninguno de los caballos con piel sana. No se encontraron levaduras.

Zusammenfassung Um die Prvalenz von oberflchlichen und follikulren Bakterien und Pilzen in Hautproben von 247 Pferden mit entzndlichen Hauterkrankungen und von 27 Pferden mit gesunder Haut
abzuschtzen, wurde eine retrospektive lichtmikroskopische Studie durchgefhrt. Kokken wurden auf der Probenoberflche von 23% (95% Konfidenzintervall 18%, 29%) und 7% (95% Konfidenzintervall, 0%, 19%) der Proben
von Pferden mit Hauterkrankungen, beziehungsweise von Pferden mit gesunder Haut gefunden. In unserer Serie
von Pferden hatte von den 9 Dermatosen mit wenigstens 10 Erkrankungsfllen nur die bakterielle Follikulitis
2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 47 5 1

Bacteria and fungi in equine skin biopsies

51

eine hhere Prvalenz von Oberflchenbakterien (57%; 95% Konfidenzintervall 34%, 81%) als die anderen 8 (die
sich nicht voneinander unterschieden alle Prvalenzen waren <30%). Es gab eine signifikante Beziehung
zwischen der Prvalenz von Oberflchenbakterien und dem Ausma epidermaler Hyperkeratose. Kokken wurden im Keratin nicht entzndeter Haarfollikel bei nur 2% der Pferde mit Hautkrankheiten und bei keinem der
Pferde mit gesunder Haut gefunden. Pilz-Porokonidia wurden auf der Oberflche bei 4% der Pferde mit Hauterkrankungen und bei keinem der Pferde mit gesunder Haut gefunden. Hefepilze wurden nicht gefunden.

2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 4751