Vet Dermatol 2015; 26: 451e105

DOI: 10.1111/vde.12247

Canine sterile nodular panniculitis: a retrospective study

of 39 dogs
Caitlin L. Contreary*, Catherine A. Outerbridge, Verena K. Affolter, Philip H. Kass and
Stephen D. White
*William R. Pritchard Veterinary Medical Teaching Hospital, Departments of Veterinary Medicine and Epidemiology, Veterinary Pathology
Microbiology and Immunology and Population Health and Reproduction, School of Veterinary Medicine, University of California Davis, 1 Garrod
Drive, Davis, CA 95616, USA
Correspondence: Caitlin L. Contreary, William R. Pritchard Veterinary Medical Teaching Hospital, University of California Davis, 1 Garrod Drive,
Davis, CA 95616, USA. E-mail:ccontr@ucdavis.edu

Background Canine sterile nodular panniculitis (SNP) is an inflammatory disease of the panniculus that is typically managed with immunomodulatory or immunosuppressive treatments. It has been reported to be a cutaneous marker of an underlying systemic disease.
Hypothesis/Objectives To assess the presence or absence of concurrent systemic diseases associated with
canine SNP and to document breed predispositions.
Animals Thirty nine dogs presented to a veterinary teaching hospital from 1990 to 2012 which met inclusion
criteria.
Methods Inclusion in this retrospective study required a diagnosis of SNP via histopathological analysis and
negative special stains for infectious organisms. Breed distributions of affected dogs were compared to all other
dogs examined at this hospital during the study period. Correlations between the histological pattern of panniculitis and the histological presence of dermatitis, clinical presentation of lesions, dog breed and therapeutic outcomes were assessed.
Results Australian shepherd dogs, Brittany spaniels, Dalmatians, Pomeranians and Chihuahuas were significantly over-represented, but correlations between inflammatory patterns of panniculitis and other histological
and clinical factors were not identified. Based on the information available in medical records, 32 dogs (82.1%)
had no concurrent systemic diseases identified. Four dogs had concurrent polyarthritis, which may be related to
SNP through unknown mechanisms.
Conclusions/Clinical Importance This study identified several novel breed predilections for SNP; it failed to
find any clear correlations with associated systemic diseases other than polyarthritis. The histological inflammatory pattern of SNP does not predict therapeutic outcome.

Introduction
Panniculitis is a disease in which subcutaneous fat (panniculus adiposus) becomes inflamed. Gross lesions
resulting from inflammation of this tissue can be single or
multiple, and may be focal or generalized in their distribution.1 Lesions often present as deep nodules with or without clinically observed draining tracts, fistulae and ulcers.
These lesions can be confused with deep pyoderma,
cutaneous neoplasia or a cutaneous cyst.1,2 Infectious
aetiologies for panniculitis include bacterial, fungal, viral,
parasitic and protozoal organisms. Noninfectious aetiologies include foreign body reactions, post-vaccination/injection reactions, thermal burns, trauma, vitamin E

Accepted 6 June 2015

Presented as an abstract of the North American Veterinary Dermatology Forum meeting in Nashville, TN, USA and published:
Vet Dermatol 2015; 26: 146.
Sources of Funding: This study was self-funded.
Conflicts of Interest: There are no conflicts of interest to
declare.
2015 ESVD and ACVD, Veterinary Dermatology, 26, 451e105.

deficiency, pancreatic disease, immune-mediated disease and idiopathic sterile nodular panniculitis (SNP).1,3
The diagnosis of SNP is made after infectious or other
noninfectious causes have been ruled out. Ideally, diagnostic tests for SNP include histopathological analysis of
full-thickness skin biopsy samples with special stains to
rule out aetiological agents, and bacterial, fungal and
mycobacterial cultures of deep tissues. Cytological examination of fine needle aspirate samples is considered to
be insufficient because results are often inconsistent with
the histopathological findings.1
The current standard treatment for SNP involves
immunosuppression with oral glucocorticoids, with or
without adjunctive therapies such as oral ciclosporin and/
or a combination of a tetracycline antibiotic with niacinamide.1,4 The pathogenesis of SNP is poorly understood, although previous studies and case reports have
proposed that it should be considered a cutaneous marker of systemic diseases, including, but not limited to,
pancreatic and hepatic disease.1,512
The aims of this retrospective study were to evaluate
case records for evidence of concurrent systemic
451

Contreary et al.

diseases and breed predilections for the development of

SNP. Histological review of biopsy materials focused on
identification of potential correlations between the inflammatory patterns of panniculitis and the presence of concurrent dermatitis, with the clinical presence of draining
tracts or ulcers. Histological patterns of inflammation
were also evaluated for correlations with specific breeds
and with attainment of clinical remission.

quently, special stains were sometimes performed retrospectively in

order to ensure that uniform methods had been applied to all biopsy
materials. Those evaluated included Fites and Ziehl Neelson stains
for acid fast bacteria, and Gomori methenamine silver and periodicacid Schiff stains for fungal organisms. Immunohistochemistry with
an anti-Bacille Calmette Guerin antibody (Dako Corporation; Carpinteria, CA, USA) was also performed as an additional screening tool for
mycobacteria and other infectious organisms.13

Medical record review

Materials and methods

A computerized medical record search was performed for dogs diagnosed with SNP at the Veterinary Medical Teaching Hospital, University of California Davis (VMTH-UCD) from 1990 to 2012. The
diagnosis of SNP was based upon clinical presentation, haematoxylin
and eosin (H&E) stained biopsy samples and negative special stains
for infectious agents. Availability of medical records and histological
sections (both H&E and special stains) for review was required for
inclusion of the case in the study. Due to the retrospective nature of
the study, deep bacterial, fungal and mycobacterial culture results
were available for only 30, 19 and 5 dogs, respectively. Conse-

The following information was obtained from medical records of dogs

meeting the inclusion criteria: signalment; duration of skin disease at
time of biopsy; number, type and location of skin lesions; histological
presence/absence of dermatitis; and presence/absence of concurrent systemic diseases identified either at the time of diagnosis or
during subsequent visits. Evidence of a concurrent systemic disease
was determined by the presence of supportive clinical signs, including heart murmurs, anorexia, lethargy, vomiting, diarrhoea, weight
loss/gain and pyrexia. These clinical signs were recorded along with
results of complete blood counts (CBC), serum biochemistry panels
(SBP), urinalyses (UA), abdominal ultrasound (AUS), abdominal radiographs and thoracic radiographs when available (Table 1). The results

Table 1. Historical Information, presence of concurrent disease and laboratory diagnostic tests

Dog number
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
38
39

Febrile

Reported
anorexia

Y
Y

Reported
lethargy

Y
Y
Y

Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y

Y
Y
Y

Y
Y
Y

Y
Y
Y

Y
Y
Y

Y
Y
Y
Y

Y
Y
Y

Y
Y

Y
Y

Y
Y

Y
Y

Y
Y

UA

SBP

Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y

CBC

Y
Y
Y
Y
Y
Y
Y
Y
Y

AUS

AXR

TXR
Y
Y

Diagnosed
concurrent
disease
IMPA
SD
SD

Y
IMPA
Y
Y
Y
Y
Y
Y

Y
Y
Y
Y
Y
Y

Y
Y
Y

IMPA
Y
Y
Y

DM

Y
Y

Y
Y

Y
Y
Y
Y
Y
Y
Y
Y
Y

Y
Y
Y
Y
Y
Y
Y
Y

IMPA

CBC, complete blood count; SBP, serum biochemistry profile; UA, urinalysis; AUS, abdominal ultrasound; AXR, abdominal radiographs; TXR,
thoracic radiographs; Y, test performed; IMPA, immune-mediated polyarthritis; SD, seizure disorder; DM, diabetes mellitus.

452

2015 ESVD and ACVD, Veterinary Dermatology, 26, 451e105.

Canine sterile nodular panniculitis

of bacterial, mycobacterial, and fungal cultures from lesional skin

were also recorded if available (Table 2). Initial and tapering doses of
prescribed medications were noted. Attainment of remission (defined as complete absence of lesions), time to remission, occurrence
of relapse and time to relapse were recorded. Any reported adverse
effects of medications were also noted.

Histopathology
Evaluation of H&E stained sections was carried out for the pattern of
inflammation, the predominant cell type(s) present, and the presence
and extent of concurrent dermatitis. The patterns of subcutaneous
inflammation were categorized as: (i) septal; (ii) nodular; (iii) diffuse;
(iv) intralobular; or (v) a combination of different patterns. Involvement of the dermis was categorized as either: (i) superficial; (ii) middermal; or (iii) deep dermal.

Statistical analysis
Breed distributions were compared between dogs with SNP and the
remainder of the canine hospital population seen at the VMTH-UCD
from 1990 to 2012 using an exact chi-square test of homogeneity.
Specific breeds with more than one case of panniculitis that con-

Table 2. Biopsied samples special stains and culture results

Dog
number

Negative
special
stains

Negative
bacterial
culture

Negative
fungal
culture

Negative
mycobacterial
culture

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
38
39

Y
Y
Y
Y
Y
y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y

Y
P*
TNP
P*
Y
Y
Y
Y
TNP
Y
Y
TNP
Y
Y
Y
P*
Y
Y
Y
Y
TNP
TNP
P*
Y
P*
TNP
Y
P*
TNP
Y
Y
TNP
Y
Y
TNP
Y
Y
Y
Y

TNP
TNP
TNP
TNP
Y
Y
Y
Y
TNP
Y
Y
TNP
Y
Y
Y
Y
Y
TNP
Y
TNP
TNP
TNP
Y
TNP
TNP
TNP
TNP
TNP
TNP
TNP
Y
TNP
Y
Y
TNP
TNP
Y
Y
Y

TNP
TNP
TNP
Y
TNP
TNP
TNP
TNP
TNP
TNP
Y
TNP
TNP
TNP
TNP
Y
TNP
TNP
Y
TNP
TNP
TNP
TNP
TNP
Y
TNP
TNP
TNP
TNP
TNP
TNP
TNP
TNP
TNP
TNP
TNP
TNP
TNP
TNP

Y, negative test obtained; P*, positive culture; TNP, test not

performed.

2015 ESVD and ACVD, Veterinary Dermatology, 26, 451e105.

tributed to overall significant differences were identified using breedspecific contributions to the chi-square statistic that were significant
(P < 0.05). Significant results are reported as observed to expected
(O:E) ratios the number of dogs belonging to a particular breed with
panniculitis relative to the number of dogs expected with panniculitis
based on the total hospital distribution of dog breeds seen.

Results
Medical record review
Signalment: Thirty nine dogs met inclusion criteria, of
which eight (20.5%) were Australian shepherd dogs, four
each (10.3%) were Chihuahuas, Labrador retrievers or
mixed breed dogs, and two each (5%) were Rottweilers,
Dalmatians, Pomeranians or Brittany spaniels. One dog
represented each of the following breeds: pug, Cardigan
Welsh corgi, American cocker spaniel, Jack Russell terrier, dachshund, Chesapeake Bay retriever, golden retriever, Airedale terrier, toy poodle, boxer and beagle.
Nineteen (48.7%) of the dogs were spayed females, 15
(38.5%) were castrated males, two (5.1%) were intact
females and three (7.7%) were intact males. The mean
age at presentation was 6.5 yr (range 113 yr).
A total of 131,281 dogs were admitted to the VMTHUCD between 1990 and 2012. There was a significant difference in the breed distribution between dogs with SNP
and the overall hospital population (P < 0.0001). Compared to the expected distribution, the following breeds
were significantly overrepresented with diagnoses of
SNP: Australian shepherd dogs (n = 8; O:E = 13.3), Brittany spaniels (n = 2; O:E = 6.7), Dalmatians (n = 2; O:
E = 6.7), Pomeranians (n = 2; O:E = 6.7) and Chihuahuas
(n = 4; O:E = 5.0).
History and physical examination findings
Mean duration of lesions prior to initial presentation to the
VMTH-UCD was 3 months (range 0.2512 months). Nine
dogs had been treated previously with systemic corticosteroids, one dog had been treated with a systemic antihistamine (diphenhydramine), 12 dogs had been treated
with one systemic antibiotic and 19 dogs had been treated with more than one systemic antibiotic. The dosages
of corticosteroids administered previously were not
recorded in the majority of these cases.
Locations of lesions according to body region were
trunk (30), neck (9), face/head (7), shoulder (6), limb (4)
and hip (2). A solitary lesion was present in 6 of 39 dogs
(15.4%); the remaining 33 dogs had multiple lesions (Figure 1). The most commonly affected areas were the
trunk (30 of 39 dogs; 76.9%) and the neck (9 of 39 dogs;
23.1%). The most common skin lesions were dermal or
subcutaneous nodules, documented in 35 of 39 dogs
(89.7%). The nodules were draining at the time of initial
presentation in 13 dogs and two dogs had ulcerated nodules. The remaining four dogs presented with ulcers (1 of
39), haemorrhagic bullae (1 of 39), crusts and alopecia (1
of 39) and a plaque (1 of 39).
Eight dogs (20.5%) were pyrexic at the time of diagnosis (mean 40C; range 39.645C). In all cases, these
dogs temperatures were within normal limits during subsequent visits. Six dogs (15.4%) were reported to be
anorexic at the time of diagnosis and four dogs (10.3%)
453

Contreary et al.

Figure 1. Multiple subcutaneous ulcerated nodules over the trunk

and both shoulders of a Chihuahua.

were noted to be lethargic by their owners. All dogs presenting with clinical signs of systemic illness (fever, anorexia or lethargy) had a CBC, SBP and AUS performed
with unremarkable findings. These results are summarized in Table 1. In all dogs, clinical signs of systemic illness quickly resolved after SNP treatment was instituted.
No dog in this study presented with vomiting, diarrhoea,
weight loss/gain, auscultable heart murmur or palpable
hepatomegaly.

Clinicopathologic findings
Results for complete blood counts were available for 28
of 39 dogs (71.8%) and revealed a nonregenerative, normocytic, normochromic anaemia in 12 of 28 (median
haematocrit: 31%; mean 28.3  2.8%; range 24.5
39.2%; normal 4055%), and a leukocytosis with a
mature neutrophilia in 9 of 28 dogs (median: 19.7 9 103/
lL; mean 27.1 9 103/lL  12,499.44 9 103/lL; range
1648.8x103/lL; normal 310.5 9 103/lL). Other abnormalities noted included monocytosis in five dogs and lymphopenia in three dogs.
Urinalysis results were available in 28 of 39 dogs
(71.8%); none had significant proteinuria. Serum biochemistry panels were available for review from 25 of 39
dogs (64.1%). Alkaline phosphatase values were elevated
in 13 of 25, (median 238.0 IU/L; mean 374.9 IU/
L  227.14 IU/L; range 173773 IU/L; normal 21170 IU/
L), aspartate aminotransferase values were elevated in 6
of 25 (median: 79 IU/L; mean: 81 IU/L  12.09 IU/L;
range 6899 IU/L; normal 1942 IU/L) and alanine aminotransferase values were elevated in 2 of 25 dogs (median:
140.5 IU/L; mean: 141 IU/L  14.85 IU/L; range 130
151 IU/L; normal 1970 IU/L). One dog had elevations of
all three enzymes; an AUS at the time of diagnosis did not
indicate evidence of hepatic disease. Five dogs had elevations in two of these enzymes; four of which had AUS
performed at the time of diagnosis. There was no evidence of hepatic or pancreatic disease in three of these
four dogs, whereas the fourth had ultrasonographic
changes interpreted to be consistent with a steroid hepatopathy. This dog had not received corticosteroids
according to the medical record, but due to the retrospective nature of the study, the possibility of corticosteroid
use cannot not be excluded entirely. This dog had skin
lesions present for 1 week prior to presentation and neither repeated AUS nor cortisol function tests were performed. However, the dog developed no clinical signs of
systemic disease during the follow-up period of
16 months.

Concurrent diseases
Four dogs (10.3%) were diagnosed with concurrent polyarthritis via arthrocentesis; in all four cases the signs of
polyarthritis developed within days of the skin lesions.
One of the four dogs was pyrexic at the time of diagnosis.
All dogs had clinical signs referable to joint disease; two
of the four with shifting leg lameness, one nonweight
bearing on the left hind limb and one with joint effusion of
the left tarsus.
Other concurrent systemic disease identified in the
study group included one dog with a 13-month history of
well-controlled diabetes mellitus prior to development of
SNP and two dogs with long-standing histories (3 and
6 yr) of seizure disorders that had been well-managed
with potassium bromide. These conditions were not
thought to be related to the SNP. The remaining 32 dogs
(82.1%) had no identifiable evidence of concurrent systemic disease based on available clinical findings or
results of diagnostic tests recorded in the medical
records (see below), either at the time of their initial presentation, or during the follow-up period available for
evaluation.

Imaging studies
Abdominal ultrasound was performed at the time of diagnosis in 20 of 39 (51.3%) of the dogs in this study. As
mentioned above, one had evidence of a mild steroid
hepatopathy. All other ultrasonographic examinations
were within normal limits. Abdominal radiographs were
performed in three other dogs (7.7%) and were within
normal limits. Thoracic radiographs were performed in 23
of 39 dogs (60.0%) and were also within normal limits.
Overall, 13 dogs had CBC, SBP, UA and AUS performed
to evaluate for concurrent diseases and an additional two
dogs had these tests with the exception of an UA.

454

2015 ESVD and ACVD, Veterinary Dermatology, 26, 451e105.

Histopathological analysis and tissue cultures of skin

biopsies
Histological patterns of inflammation identified in biopsy
materials included: a nodular pattern in 14 of 39 cases
(35.9%), Figure 2); a septal pattern in 5 of 39 cases
(12.8%); a diffuse pattern in 5 of 39 cases (12.8%); and
an intralobular pattern in one case. A combination of two
patterns was identified in 14 of 39 cases (35.9%). The
combination patterns identified were: nodular and diffuse

Canine sterile nodular panniculitis

Figure 2. Nodular panniculitis with multifocal nodules composed of

inflammatory cells (409 magnification; haematoxylin and eosin staining).

(n = 7), septal and diffuse (n = 3), septal and intralobular

(n = 2), and septal and nodular (n = 2). There were never
more than two inflammatory patterns in the same sample. The lesions were predominated by histiocytes and
neutrophils in 34 of 39 cases (87.2%; Figure 3). Lymphocytes were the predominant inflammatory cell in 3 of 39
cases, and plasma cells predominated in 2 of 39 cases.
There was no discernible association between histological patterns or predominant cell type(s) identified with
lesion localization, breed of dog or reported clinical
response to therapy. Therefore, statistical tests were not
applied to these comparisons.
Concurrent dermatitis was noted in 24 of 39 cases
(61.5%). The inflammation was limited to the deep dermis in 6 of 24 cases, extended to the mid dermis in 4 of
24 cases and to the superficial dermis in 14 of 24 cases.
The same predominant cell types found in the panniculus
were also found in the dermis in all 24 of these cases.
Seven of 14 cases (50%) with inflammation extending to
the superficial dermis had draining tracts and/or ulcers on
physical examination, whereas 5 of 15 (33.3%) cases
without histological evidence of dermatitis had draining

tracts and/or ulcers on physical examination. There

appeared to be no correlation between the presence of
draining tracts or ulcers and the histologic presence of
dermatitis. These results are summarized in Table S1 in
Supporting information.
Deep tissue bacterial cultures were performed via
punch biopsy from the lesions of 30 of 39 dogs (77%).
Fungal cultures were performed from the lesions of 19
dogs and mycobacterial cultures from the lesions of five
dogs. All fungal and mycobacterial cultures were negative. Bacterial cultures revealed rare numbers of Staphylococcus intermedius/pseudintermedius in 4 of 30 dogs
and rare numbers of coagulase-negative Staphylococcus
spp. in 2 of 30 dogs. The tissue samples for culture were
not always collected from lesions with obvious draining
tracts. Staphylococcus intermedius/pseudintermedius
was cultured from nodules with associated draining tracts
in 3 of 4 dogs and from a single plaque in one dog. Coagulase-negative Staphylococcus spp. was cultured from
nodular lesions in both dogs, one of which was associated with a draining tract. Culture results are summarized
in Table 2.

Figure 3. Nodule primarily composed of histiocytes and neutrophils

(1009 magnification, haematoxylin and eosin staining).

Therapy and clinical response

Treatment, attainment of remission and occurrence of
relapse in all dogs are summarized in Table S2 in Supporting information. One dogs lesions resolved without any
treatment; the other 38 were prescribed various oral
medications. Thirty two dogs in the study were prescribed corticosteroids (with or without concurrent nonsteroidal immunomodulatory or immunosuppressive
medications). The most commonly prescribed corticosteroids were prednisone (Roxane Laboratories Inc.;
Columbus, OH, USA) or prednisolone (Lloyd Inc.; Shenandoah, IA, USA) at 0.82.5 mg/kg once daily or 0.5
1.3 mg/kg twice daily. Prednisone was used in 21 cases
and prednisolone in seven cases. Nineteen dogs started
treatment with twice daily corticosteroids and 13 with
once daily corticosteroids. The remaining six dogs either
died soon after diagnosis (one dog), were lost to followup with no corticosteroids prescribed (three dogs) or
were managed without corticosteroids (two dogs). All
losses to follow-up occurred after the diagnosis of SNP
had been made.
Medications used in conjunction with corticosteroids
included a combination of a tetracycline antibiotic, usually
doxycycline (ParPharmaceutical Companies Inc.; Spring
Valley, NY, USA) and niacinamide (Rugby Laboratories;
Lavonia, MI, USA) in 14 dogs. Dosages ranged from 250
to 500 mg/dog twice to thrice daily for tetracycline, 4.2
6.5 mg/kg twice daily for doxycycline and 250500 mg/
dog thrice daily for niacinamide. Immunosuppressive
drugs used included ciclosporin (Atopica, Novartis Animal
Health; Greensboro, NC, USA) in nine dogs at a dose
ranging from 3 to 5.5 mg/kg once daily); azathioprine
(Imuran, Excella; Feucht, Germany) in six dogs at a dose
ranging from 1 to 2.2 mg/kg once daily); and chlorambucil
(Leukeran; Excella, Feucht, Germany) in one dog at a
dose of 0.18 mg/kg once daily. Thirteen dogs initially
managed with corticosteroids alone needed at least one
of the nonsteroidal immunomodulatory or immunosuppressive medications mentioned above added to control

2015 ESVD and ACVD, Veterinary Dermatology, 26, 451e105.

455

Contreary et al.

the SNP. Seven dogs were prescribed more than one

nonsteroidal medication during management of the SNP.
Doxycycline and niacinamide were used as the sole therapy from the time of diagnosis in one dog, which was lost
to follow-up after diagnosis of SNP. Chlorambucil was
used as the sole therapy from the time of diagnosis for
one dog which was lost to follow-up after 4 months but
was clinically improved at the time of its last examination.
All six of the dogs from which Staphylococcus was isolated ultimately received an immunosuppressive medication. Only one of these six dogs was prescribed systemic
antimicrobial therapy (oral cefalexin: 23 mg/kg twice daily
for 1 month), and ultimately achieved remission of SNP
5 months post-diagnosis. Three of the remaining five
dogs achieved clinical remission with immunosuppressive treatment. One of the five had regression of initial
lesions and no new lesion developed, but it was subsequently lost to follow-up. The remaining dog was lost to
follow-up after initial diagnosis.
Three dogs in this study were euthanized. One was
euthanized 4 months after diagnosis due to concurrent
immune-mediated polyarthritis and consequent quality of
life concerns. This dog had not achieved remission while
being managed with prednisone 1 mg/kg once daily and
ciclosporin 5.2 mg/kg once daily, but the initial lesions
were resolving and no new lesions had been noted since
the time of diagnosis. The second dog was euthanized
13 months after diagnosis due to a haemoabdomen.
Postmortem examination was not performed, but trauma
was suspected. This dog had normal abdominal and thoracic radiographs at the time of SNP diagnosis, but had
not yet achieved clinical remission while being managed
with 17.8 mg/kg thrice daily of tetracycline and niacinamide, and prednisone 0.36 mg/kg once daily. However,
the initial lesions were resolving and no new lesions had
been noted since the time of diagnosis. The third dog
was euthanized 2 months after diagnosis of SNP due to a
diagnosis of Nocardia spp. from a superficial bacterial culture of a crusted lesion on the dorsum. This dog had
achieved remission 1 month after diagnosis of SNP with
the use of prednisolone 0.5 mg/kg twice daily, and doxycycline and niacinamide both 21.7 mg/kg thrice daily, but
had relapsed 1 month later after discontinuation of the
prednisolone. This dog had a negative deep tissue culture
at the time of diagnosis. The final dog died 2 months after
diagnosis of SNP for unknown reasons, but had a normal
CBC and SBP at the time of SNP diagnosis. This dog had
not achieved remission while being managed with prednisolone 0.5 mg/kg twice daily and ciclosporin 3.0 mg/kg
once daily, although initial lesions were resolving at the
time of death.

required medical management. Of those 20, only four

(20.0%) were able to achieve remission with corticosteroids alone. The other 16 dogs required a combination
of a corticosteroid and another medication. These medications included a tetracycline antibiotic/niacinamide
combination (eight dogs), azathioprine (five dogs) and
ciclosporin (three dogs). Seven of the eight dogs that did
not achieve remission showed regression of initial lesions
and no new lesions at subsequent visits but were lost to
follow-up before remission was achieved. The eighth dog
was lost to follow-up after 1 month, before any improvement was noted.
The mean time to remission for the 21 dogs that
achieved it was 2.9 months (range 111 months). Five of
these dogs relapsed after medications were tapered. At
the time of relapse, three dogs were receiving prednisone
or prednisolone alone, one was receiving azathioprine
alone after complete taper and withdrawal of corticosteroids, and one was receiving prednisone, tetracycline
and niacinamide. Four of these five dogs were able to
achieve remission a second time with increased doses of
medications and did not have their medications tapered
again. One dog was lost to follow-up after relapse. The
mean time from remission to relapse was 10 months
(range 330 months). There appeared to be no correlation
between breed and attainment of remission or subsequent relapse. Only in two dogs were all medications discontinued and they remained in remission. One of these
dogs was managed with prednisone, the other with doxycycline and niacinamide.
Adverse effects of treatment
Twenty dogs had at least one adverse event during the
course of treatment. Those most commonly reported
included polyuria (14 dogs), polydipsia (14 dogs) and
polyphagia (nine dogs). Less commonly reported adverse
effects were weight gain (three dogs), panting (three
dogs), muscle atrophy (three dogs) and haematemesis
(one dog). All of these adverse effects were attributed to
the administration of corticosteroids and diminished in
severity or resolved as corticosteroid therapy was
tapered or discontinued. One dog became leukopenic
4 months after starting azathioprine and another became
thrombocytopenic 22 months after starting this medication. The leukopenia and thrombocytopenia resolved with
cessation of azathioprine use. One dog that achieved
remission with prednisolone, tetracycline and niacinamide was diagnosed with nocardiosis via culture
1 month later and was euthanized as noted above.

Discussion

Follow-up
Follow-up ranged from 0 to 2490 days (mean 251 days).
There was no available follow-up after initial diagnosis for
nine dogs (23.1%). Of the 30 dogs for which follow-up
information was available, one died within 2 months as
noted above. Of the remaining 29 dogs, 21 (72.4%)
achieved clinical remission of SNP, defined as complete
absence of any skin lesions. One dog was able to achieve
remission without treatment 2 weeks after diagnosis and
was followed for two additional weeks; the other 20

This study has documented a breed predisposition to

SNP in Australian shepherd dogs, Brittany spaniels, Chihuahuas, Dalmatians and Pomeranians, as compared to
the general hospital population. Previously reported breed
predilections include dachshunds and poodles.2,6,11,14,15
Evaluation of SNP breed distributions in other hospital
populations are needed to confirm these findings. The
most common gross lesions reported in this case series
were cutaneous and subcutaneous nodules. Most cases
(84.6%) had multiple lesions, the majority of which were

456

2015 ESVD and ACVD, Veterinary Dermatology, 26, 451e105.

Canine sterile nodular panniculitis

found on the trunk and neck. In previous studies, the

majority of dogs with SNP had multiple lesions.2,5 In a
previous study, 83% of dogs with SNP had lesions on the
trunk and 33.3% had lesions on the neck and ventral cervical area.5
It is critical to determine that an infectious aetiology is
not the cause of panniculitis when establishing the diagnosis of SNP. Hence, it is important to evaluate the samples for the presence of pathogens by using special
stains and cultures. Due to the retrospective nature of the
present study, the exact location of the biopsy was not
always known and tissue culture results were not always
available. Special stains were negative for all cases,
including those that were not cultured at the time of submission. Moreover, follow-up information was available
for every dog which did not have an initial bacterial culture
at the time of diagnosis. Deep tissue bacterial cultures
taken from lesional skin yielded low numbers of Staphylococcus species in six dogs. The small numbers of organisms isolated and the clinical response of all six dogs to
immunosuppressive therapy without the concurrent use
of systemic antibiotics likely reflected surface contamination rather than the Staphylococcus being a causative
agent in these dogs.
Six dogs had no investigation for systemic disease performed. Two of these dogs had a CBC and SBP performed at another veterinary clinic within 1 month of SNP
diagnosis. All six of these dogs were seen through the
dermatology service and were not showing any clinical
signs of systemic illness at the time of initial presentation.
The lack of standardization of a systemic investigation
may reflect a bias by the varying services in the VMTHUCD. A more thorough systemic workup was often
available for the patients admitted through the internal
medicine or emergency medicine services as opposed to
the dermatology service, because dogs that presented to
the former two services were more likely to be considered systemically ill.
In the present study, SNP occurred concurrently with
immune-mediated polyarthritis in 4 of the 39 dogs
(10.3%). This is a smaller proportion than reported by
OKell et al.5 in which 4 of 14 dogs (28.6%) had concurrent arthropathies. Mechanisms by which panniculitis and
arthropathies may be aetiologically associated have been
reviewed elsewhere.5 In addition to arthropathies, dogs
with SNP were identified with a variety of conditions such
as pancreatic, hepatic, cardiac, renal or adrenal disease,
and some dogs had more than one concurrent systemic
disease.5 By contrast (but similar to a previous report3),
the majority of the dogs in our study were largely free of
identifiable concurrent diseases at the time of initial presentation and during follow-up. In particular, no pancreatic
or adrenal pathology was visualized in those dogs that
had AUSs performed, and there was no clinical evidence
of primary liver disease in any case studied. However,
due to the retrospective nature of this study and lack of
standardization for medical investigations, it is conceivable that a subclinical concurrent disease could have been
missed in some cases.
The clinic-pathological changes noted in our cases
were most likely associated with the inflammatory nature of SNP or prior use of corticosteroids. Of the seven

dogs that did have identifiable concurrent systemic disease, only four (10.2%) showed evidence of a concurrent inflammatory disease in another location
(polyarthritis with associated lameness or joint effusion). Three dogs had histories of long-standing diseases that were well-managed at the time of SNP
development (either diabetes mellitus or a seizure disorder), and were not thought to be related to the development of SNP. The two dogs with a long-standing
seizure disorder were managed with potassium bromide (KBr). A prior case report has linked KBr to panniculitis in two dogs.16 However, both dogs in the
present study continued to receive KBr along with
immunosuppressive therapy for SNP and both showed
improvement of their clinical signs. Therefore, an
adverse reaction to KBr was considered unlikely but
could not be ruled out.
Neither the histological pattern of inflammation, the
types of inflammatory cells present nor involvement of
the dermis, correlated with locations of the lesions, specific breeds and presence of draining tracts or presence/absence of concurrent systemic diseases. Nodular
inflammatory aggregates were the most common inflammatory pattern (35%) seen in this study, followed by septal, diffuse and intralobular patterns. Neutrophils and
histiocytes predominated in most lesions, but differences
seen in cellular infiltrates likely change over time and
could result from inherent differences in locations where
skin biopsies were taken. Moreover, involvement of the
dermis was not necessarily associated with formation of
draining tracts and draining tracts were seen without evidence of extensive dermatitis.
Oral immunosuppressive/immunomodulatory medical
management is required in most cases of SNP and in this
study there was no correlation between time to remission and the initial medications prescribed. Only one case
for which follow-up was available resolved without treatment. A combination of a corticosteroid and nonsteroidal
immunosuppressive/immunomodulatory medication(s)
was the most successful method of treatment. This combination was necessary for 85% of the dogs that
achieved remission, with the other 15% achieving remission with corticosteroid therapy alone. Relapse was
noted in five dogs after treatment was either discontinued or tapered, which supports prior reports of the need
for long-term or even lifelong treatment in some dogs.16
Due to the retrospective nature of this study follow-up
was not available in all dogs, but from available data only
two dogs maintained remission after all medications were
discontinued.
Adverse effects of medications were primarily limited
to corticosteroid-induced polyuria, polydipsia and polyphagia. Two dogs experienced either leukopenia or thrombocytopenia while receiving azathioprine, which highlighted
the importance of monitoring CBCs while administering
azathioprine. One dog developed nocardiosis while
receiving 0.5 mg/kg prednisolone twice daily and 500 mg
of both tetracycline and niacinamide thrice daily. Whether
this was acquired during therapy as an effect of immunosuppression is unknown.
In conclusion, this study reports new breed predispositions for SNP and shows that the disease often occurs

2015 ESVD and ACVD, Veterinary Dermatology, 26, 451e105.

457

Contreary et al.

independently of underlying systemic diseases. It is

noteworthy that concurrent polyarthritis was seen in
approximately 10% of the dogs in this study, a
clinicopathological correlation that has been reported
previously.5 No correlations could be made between the
histological pattern of panniculitis and the histological
presence of dermatitis, clinical presentation of lesions,
dog breed or therapeutic outcomes.

References
1. Miller WH Jr, Griffin DE, Campbell KL. Miscellaneous skin diseases. In: Muller and Kirks Small Animal Dermatology, 7th edition. St. Louis, MO: Elsevier Health 2013; 701704.
2. Yamagishi C, Momoi Y, Kobayahi T et al. A retrospective study
and gene analysis of canine sterile panniculitis. J Vet Med Sci
2007; 69: 915924.
3. Scott DW, Anderson WI. Panniculitis in dogs and cats: a retrospective analysis of 78 cases. J Am Anim Hosp Assoc 1988; 24:
551559.
4. Kim H, Kang M, Kim J. Sterile panniculitis in dogs: new diagnostic findings and alternative treatments. Vet Dermatol 2011; 22:
352359.
5. OKell AL, Inteeworn N, Diaz SF et al. Canine sterile nodular panniculitis: a retrospective study of 14 cases. J Vet Intern Med
2010; 24: 278284.
6. Moreau PM, Fiske RA, Lees GE et al. Disseminated necrotizing
panniculitis and pancreatic nodular hyperplasia in a dog. J Am
Vet Med Assoc 1982; 180: 422425.
7. Gear RNA, Bacon NJ, Langley-Hobbs S et al. Panniculitis,
polyarthritis, and osteomyelitis associated with pancreatic
neoplasia in two dogs. J Small Anim Pract 2006; 47: 400
404.

8. Brown PJ, Mason KV, Merrett DJ et al. Multifocal necrotizing

steatitis associated with pancreatic carcinoma in three dogs. J
Small Anim Pract 1994; 35: 129132.
9. Dennis MM, OBrien TD, Wayne T et al. Hyalinizing pancreatic
adenocarcinoma in six dogs. Vet Pathol 2008; 45: 475483.
10. Mellanby RJ, Stell A, Baines E et al. Panniculitis associated with
pancreatitis in a Cocker Spaniel. J Small Anim Pract 2003; 44:
2428.
11. Hughes D, Goldschmidt MH, Washabau RJ et al. Serum a1-antitrypsin concentration in dogs with panniculitis. J Am Vet Med
Assoc 1996; 209: 15821584.
12. German AJ, Foster AP, Holden D et al. Sterile nodular panniculitis and pansteatitis in three weimaraners. J Small Anim Pract
2003; 44: 449455.
13. Bonenberger TE, Ihrke PJ, Naydan DK et al. Rapid identification
of tissue micro-organisms in skin biopsy specimens from
domestic animals using polyclonal BCG antibody. Vet Dermatol
2001; 12: 4147.
14. Aikawa T, Yoshigae Y, Kanazono S. Epidural idiopathic sterile
pyogranulomatous inflammation causing spinal cord compressive injury in five Miniature Dachshunds. Vet Surg 2008; 37:
594601.
15. Post K. Nodular panniculitis in a female Toy Poodle. Can Vet J
1983; 24: 152153.
16. Boynosky NA, Stokking LB. Potassium-bromide associated panniculitis. J Small Anim Pract 2014; 55: 640642.

Supporting Information
Additional Supporting Information may be found in the
online version of this article.
Table S1. Breed and histopathological analysis.
Table S2. Treatment, remission and relapse.

sume


Re

rile canine (SNP) est une maladie inflammatoire de lhypoderme qui
Contexte La panniculite nodulaire ste
re par des immunomodulateurs ou des immunosuppresseurs. Elle a e

te

de

crite comme un marqueur
se ge

de maladie syste

mique sous-jacente.
cutane
ses/Objectifs De

terminer la pre

sence ou labsence dune maladie syste

mique concourante
Hypothe

e a la SNP canine et documenter les pre

dispositions raciales.
associe

sente

s a un ho
^pital ve

te

rinaire denseignement de 1990 
Sujets Trente et un chiens pre
a 2012 remplissant les conditions dinclusion.

thodes Linclusion dans cette e

tude re

trospective ne

cessitait le diagnostic de SNP par histopatholoMe




gie et des colorations speciales negatives pour les organismes infectieux. Les distributions de race des

te

compare

es a tous les autres chiens examine

s au sein de lho
^pital au cours de la pe

richiens atteints ont e

tude. Les corre

lations entre les patrons histopathologiques de la panniculite et la pre

sence hisode de

sentation clinique des le

sions, la race des chiens et la re

ponse au traitement
tologique de dermatite, la pre

te

e

value

es.
ont e

sultats Les bergers australiens, dalmatiens, e

pagneuls bretons, loulous de Pome

ranie et chihuahuas
Re

taient significativement plus repre

sente

s mais la corre

lation entre les patrons inflammatoires de pannie

te

identifie

. A partir des informations disponculite et autre facteurs histologiques et cliniques nont pas e

dicaux, 32 chiens (82.1%) navaient pas de maladie syste

mique identifie

e.
ibles dans les dossiers me
^tre relie

e 

canismes inconQuatre chiens avaient une polyarthrite concurrente pouvant e
a la SNP via des me
nus.

tude identifie plusieurs nouvelles pre

dilections raciales pour
Conclusions et importance clinque Cette e

lation claire na pu e
^tre identifie

e avec des maladies syste

miques autres que la polla SNP; aucune corre

dire une re

ponse the

rapeuyarthrite. Le patron inflammatoire histologique de la SNP ne permet pas de pre
tique.
Resumen

n la paniculitis este

ril nodular canina (SNP) es una enfermedad inflamatoria del pan
culo que
Introduccio
t
picamente se maneja con tratamientos inmunomoduladores o inmunosupresores. Se ha indicado que

mica subyacente.
puede ser un marcador cut
aneo de enfermedad siste

458

2015 ESVD and ACVD, Veterinary Dermatology, 26, 451e105.

Canine sterile nodular panniculitis

tesis/Objetivos evaluar la presencia o ausencia de enfermedades siste

micas concurrentes asociHipo

n de razas
adas con casos de SNP en perros y documentar la disposicio
~o 2012 que
Animales 39 perros se presentaron a un hospital veterinario universitario desde 1990 al an

n.
cumpl
an los criterios de inclusio

todos la inclusio

n en este estudio retrospectivo requer
a diagno

stico de SNP mediante an

Me
alisis

gico, resultados negativos para las tinciones especiales para organismos infecciosos. La dishistopatolo

n de las razas en los perros afectados se comparo

con otros perros examinados en esta hospital
tribucio

n histolo

gico de paniculitis y la
durante el per
odo de estudio. Se evaluaron las correlaciones entre el patro

gica de dermatitis, la presentacio

n cl
nica de las lesiones, la raza y resultados terape

utipresencia histolo
cos.
Resultados perros Pastores Australianos, Spaniel Bretones, D

almatas, Pomeranian y Chihuahuas estu
n inflamatorio de paniculitis y
vieron m

as representados. No se identificaron correlaciones entre el patro

gicos y cl
nicos. Basado en la informacio

n disponible en los historiales 32 perros
otros factores histolo
(82%) no presentaron otras enfermedades concurrentes. Cuatro perros ten
an poliartritis, que podr
a estar
relacionado con la SNP por mecanismos desconocidos.

n para
Conclusiones e importancia cl
nica este estudio identifica varias razas novedosas con predileccio

n clara con ninguna enfermedad siste

mica excepto con poliartritis. El
SNP; falla en identificar una correlacio

n histolo

gico inflamatorio de SNP no predice el resultado terape

utico.
patro
Zusammenfassung
ndliche Erkrankung des PannikuHintergrund Die canine sterile nodulare Pannikulitis (SNP) ist eine entzu
lus, die typischerweise mit immunmodulatorischen oder immunsuppressiven Behandlungen gehandhabt
werden kann. Sie ist als kutaner Marker einer zugrundeliegenden systemischen Erkrankung beschrieben.
Hypothese/Ziele Das Vorliegen oder das Nicht-Vorliegen von begleitenden systemischen Erkrankungen,
die mit der caninen SNP in Zusammenhang gebracht werden und eine Dokumentation von Rassenpr
adispositionen.
Tiere Neununddreiig Hunde, die einem veterin
armedizinischen Schulklinik zwischen 1990 und 2012 vorllten.
gestellt wurden, die die Inklusionskriterien erfu
Methoden Der Aufnahme in diese retrospektive Studie ging eine Diagnose von SNP mittels
se Organismen voraus. Die Rassenverteilunr infektio
Histopathologie und negativen Spezialfarbungen fu
gen der betroffenen Hunde wurden mit allen anderen w
ahrend der Studiendauer in diesem Tierspital untersuchten Hunde verglichen. Die Korrelationen zwischen den histologischen Mustern der Pannikulitis und
der histologische Nachweis einer Dermatitis, die klinische Pr
asentation der Ver
anderungen, die Hunderasse und der therapeutische Erfolg wurden beurteilt.
Ergebnisse Australian Sheperds, Brittany Spaniels, Dalmatiner, Zwerspitze und Chihuahuas waren sigberreprasentiert, aber es konnten keine Korrelationen zwischen den Entzu
ndungsmustern der
nifikant u
Pannikulitis und anderen histologischen und klinischen Faktoren identifiziert werden. Aufgrund der Information, die in den Patientenkarteien vermerkt war, hatte 32 Hunde (82,1%) keine begleitenden systemischen
Erkrankungen. Vier Hunde hatten eine gleichzeitige Polyarthritis, welche mit der SNP verwandt sein
nnte, ohne dass die Mechanismen dafu
r bekannt sind.
ko
Schlussfolgerungen/Klinische Bedeutung Diese Studie identifizierte einige neue Rassenpr
adispositior SNP; es konnten keine klaren Korrelationen mit zusammenh
nen fu
angenden systemischen Erkrankungen
ndungsmuster der SNP l
auer mit Polyarthritis hergestellt werden. Das histologische Entzu
asst keine
Prognose auf den therapeutischen Erfolg zu.

(SNP)

/ SNP
1990201239

SNP

32(82.1%)4
SNP
/ SNP
SNP

2015 ESVD and ACVD, Veterinary Dermatology, 26, 451e105.

e104

Contreary et al.

(SNP),
,
/ SNP,,
19902012,39,
SNP,
,,
,
,,
,32(82.1%)
,SNP
,SNP;,
SNP

e105

2015 ESVD and ACVD, Veterinary Dermatology, 26, 451e105.