Cutaneous nodules and erythematous lesions in a dog

Maciej Guzera1,2 DVM PhD DipECVCP MRCVS
Agnieszka Cekiera3 DVM PhD
Cynthia de Vries4 DVM DipECVP

1 Laboklin, Bad Kissingen, Germany
2 Laboklin Polska, Warsaw, Poland
3 Wrocław University of Environmental and Life Sciences, Wrocław, Poland
4 Laboklin, Hoensbroek, the Netherlands

Dog, 6 years old, male neutered, Dogo Argentino.

A dog was referred for a dermatology consultation due to the presence of cutaneous lesions located on the flanks for the last month. The owner had an impression that the dog is not bothered by them. The patient had a medical history of multiple comorbidities which included immune-mediated thrombocytopenia (IMTP), hypothyroidism, dermatophytosis (Trichophyton sp. infection) and recurrent urinary tract infection. Presence of anaemia and episodes of pyrexia were also reported. The dog was treated with amoxicillin/clavulanic acid, itraconazole, topical enilconazole, prednisone, azathioprine, gastroprotectants, levothyroxine, multivitamin, mineral, omega acids and iron supplementation, as well as shampoo therapy. Previous attempts to cease immunosuppressive therapy were unsuccessful due to recurrence of thrombocytopenia. There was no significant travel history.

Scaly, erythematous lesions located on the flanks along the spine were observed; many were covered with dried exudate. In the deeper portions of the skin multiple nodules were identified. The nodules were of variable size but not greater than 1 cm in diameter. Patches of alopecia scatter around the body were also detected (Figure 1A-B). The physical findings were otherwise unremarkable.

Figure 1A-B. Scaly, erythematous lesions in the flank area in a dog.

Haematology was performed and showed mild to moderate anaemia (haematocrit = 0.30 L/L, reference interval, RI: 0.37-0.61 L/L). Absolute reticulocyte count was 26x10^9/L, RI: 10-110x10^9/L. MCV and MCHC were within the reference ranges. Mild lymphopenia (0.93x10^9/L, RI: 1.05-5.1x10^9/L) was also observed.

Biochemistry profile revealed moderate elevation of ALP (654 U/L, RI: 23-212 U/L) and mild elevation of ALT (277 U/L, RI: 10-125 U/L). AST was borderline high (51 U/L, RI: 0-50 U/L). CK and lipase were within limits. Hypothyroidism treatment was monitored – T4 concentration was moderately decreased (0.5 µg/dL, RI: 1-4 µg/dL).

Urinary dipstick was unremarkable and there was no active sediment.

No pathogens were observed on trichogram examination and skin scrapings.

Impression smears were taken from the superficial skin lesions (Figure 2A-B). Material from the nodules was obtained by fine-needle biopsy (Figure 3A-D).

Figure 2. Impression smear cytology from the skin lesion from a dog; 50x objective (A); 100x objective (B) (Hemacolor stain).
Figure 3. Fine-needle biopsy from the skin nodule from a dog; unstained slide (A); 10x objective, Hemacolor stain (B); 50x objective, Hemacolor stain (C and D).

What is your interpretation of the clinicopathological findings?

What further tests would you perform?

What is your diagnosis?



Mild to moderate anaemia was classified as nonregenerative based on the low reticulocyte count. Red blood cell indices were within limits suggesting a normocytic normochromic anaemia. Blood smear evaluation would be required to confirm this. This type of anaemia is frequently associated with chronic disease – inflammation / infection and hypothyroidism could have contributed to anaemia in this patient. Based on the history of anaemia and its mild severity, a haemolytic disorder (leading to a pre-regenerative anaemia) was unlikely. There was no evidence of bleeding. Mild lymphopenia was likely secondary to the action of glucocorticoids – drug treatment +/- stress.

Moderate elevation of ALP was likely secondary to glucocorticoid treatment – increase in steroid-induced isoenzyme of ALP. Mild elevations of ALT and AST were indicative of hepatocellular injury. Drug hepatopathy was a consideration. Primary or secondary hepatobiliary diseases were also possible and could not have been excluded without further diagnostic work-up. Muscle injury was unlikely given normal CK activity.
T4 concentration in dogs on treatment for hypothyroidism should be measured 4 to 6 hours after administration of levothyroxine. T4 level should be in the upper half of the reference range or slightly increased. When the concentration is lower dose modification should be considered. In this patient, T4 level could have been suppressed by exogenous glucocorticoids. Therapeutic range for dogs on drugs which influence T4 concentration has not been established, thus decisions on dose adjustments should be primarily guided by the response to therapy.

Impression smear cytology showed high numbers of neutrophils (many of which were degenerated), as well as low numbers of lymphocytes and macrophages. Occasional pale blue intracellular coccoid-structures were also identified. The background was pale pink, granular, mildly haemodiluted and contained nuclear remnants. These findings were consistent with marked neutrophilic bacterial inflammation. The cocci observed stained less intensively than usually which could have reflected their degeneration due to antibiotic therapy. Superficial infection / inflammation was likely secondary to an underlying disease process.
The material collected from the skin nodules by fine-needle biopsy was oily on macroscopic evaluation. Cytologic evaluation showed moderate numbers of vacuolated macrophages (with occasional bi- and multinucleated cells) and low numbers of nondegenerated neutrophils. The background was heavily vacuolated; mild haemodilution was present. These findings indicated the presence of moderate mixed inflammation and most likely reflected pyogranulomatous inflammation of subcutaneous adipose tissue / panniculitis. Although no bacteria were seen infection could not have been totally ruled out.


Histopathology was recommended in order to confirm cytological diagnosis and exclude neoplasia or infection. One of the nodules was excised and submitted for evaluation. A moderately demarcated, non-encapsulated, highly cellular nodule was effacing and replacing the subcutaneous adipose tissue architecture (Figure 4). The nodule consisted of a multifocal areas of adipocyte necrosis, surrounded by a multifocal to coalescing, moderate to marked, mixed inflammatory infiltrate consisting of high numbers of macrophages and, to a lesser extent, non-degenerate neutrophils, lymphocytes and plasma cells. Macrophages showed a moderate vacuolation of the cytoplasm. Multifocally, small groups of intact and well-differentiated adipocytes were intermingled within the inflammatory infiltrate. Multifocal mild proliferation of vascular-rich and well-differentiated granulation tissue was seen. Infectious agents, foreign body material or malignancy criteria could not be confirmed within the examined sections. These histopathological findings were consistent nodular panniculitis.

Figure 4. Histology of panniculitis in a dog (20x objective, haematoxylin and eosin stain).

Abdominal ultrasound would be required to investigate possible liver and/or biliary pathology but was not performed at this stage.
Given that the patient had diseases with suspected immune-mediated aetiology (panniculitis, IMTP) testing for systemic autoimmune disorder / systemic lupus erythematous by checking for antinuclear antibodies (ANA) and rheumatoid factor (RF) was considered.
Culture and sensitivity of the skin lesions/nodules would be helpful in guiding further therapy.
Serology and microbiology testing were omitted because of the risk of obtaining false negative results – the patient was on a long-term treatment with systemic corticosteroids and was receiving antibiotics.
Bile acid stimulation test would be necessary to better evaluate liver function – not performed.


All clinical and laboratory findings were most consistent with the diagnosis of canine sterile nodular panniculitis. Evidence of superficial bacterial infection of the skin was also present (most likely secondary to the underlying disease). Hepatobiliary disease could not be ruled out without additional diagnostic tests.

Panniculitis developed when the patient was receiving prednisolone. Given the above treatment with a more potent glucocorticoid, methylprednisolone was initiated. Other medications (except prednisone) were continued as before.
On the follow-up 10 days later, an improvement was observed – the skin lesions started to heal and their extent decreased. Although an increase in red blood cell parameters was also noted, the patient remained anaemic.
One month after the initial visit, he was generally well in himself but the skin condition deteriorated. He was pruritic and new exudative lesions appeared. Skin calcifications of variable size were also observed (confirmed by cytological examination). They were located all over the skin of the animal but were mainly grouped around the back and the flanks (Figure 5A-D). The skin was thin and dry, and his abdomen was mildly distended. A further increase in the biliary and hepatocellular enzymes occurred. Iatrogenic hyperadrenocorticism was suspected based on the clinical picture. Abdominal ultrasound was performed and showed a reduced size of both adrenal glands which further supported a chronic excess of exogenous glucocorticoids. There were also evidence of urolithiasis, cholelithiasis, moderate intrahepatic cholestasis and hepatomegaly with parenchymal degeneration (suspicion of steroid hepatopathy). Given imagining findings, hepatobiliary disease, aside from glucocorticoid treatment, likely contributed to the increases in ALP, ALT and AST. A diagnosis of iatrogenic hyperadrenocorticism can be confirmed by an ACTH stimulation test – not performed in this case due to financial reasons. In order to reverse the signs of iatrogenic hyperadrenocorticism, protect liver, normalize thyroid hormone levels and prevent a drop in haematological parameters, it was planned to gradually replace high-dose glucocorticoid therapy with cyclosporine. Hepatoprotective dietary supplement was also recommended. The prognosis was guarded to poor as treatments of concurrent diseases were mutually exclusive.

Figure 5A-D. Skin calcifications and exudative lesions in the back and flank area in a dog.

Panniculitis is the inflammation of the subcutaneous adipose tissue. The condition may appear as generalized or localised firm to fluctuant sharply-demarcated nodules usually located on the trunk, neck or limbs. These lesions may ulcerate and ooze oily or exudative material. Systemic clinical signs are usually associated with generalized disease and may include pyrexia, anorexia and lethargy. Etiological factors may include trauma, foreign-body / injection-site reaction, infection (bacterial but also fungal), immune-mediated process, pancreatic disease, adverse drug reaction, vasculopathy and vitamin E deficiency. Identification of the exact cause can be challenging in clinical settings.
Canine sterile nodular panniculitis is a disease which may be idiopathic or secondary to an underlying systemic disease. Breeds predisposed include Dachshunds and Poodles. Its diagnosis is usually based on the signalment (dog breed, young age), distribution of lesions (typically trunk), cytological picture, negative bacterial and fungal cultures and histopathological findings – including negative Periodic acid–Schiff (PAS) staining for fungi. Histopathology is particularly important in cases with firm, fibrotic lesions in which cytology may reveal the presence of atypical mesenchymal cells risking misinterpretation. Standard treatment of sterile panniculitis includes antibiotics and immunosuppressive drugs (e.g. corticosteroids, cyclosporine).
Iatrogenic hyperadrenocorticism is a consequence of exposure to exogenous glucocorticoids in anti-inflammatory or immunosuppressive doses. Clinical and laboratory signs are identical to the spontaneous form of the disease. In dogs with signs of hyperadrenocorticism gradual discontinuation of steroid therapy is recommended. This case demonstrates partial failure of the treatment in a patient with multiple comorbidities which require opposing therapies.