Sources of Funding: This study was self-funded.
Calcinosis cutis in dogs: histopathological and clinical analysis of 46 cases
Article first published online: 8 APR 2013
© 2013 The Authors. Veterinary Dermatology © 2013 ESVD and ACVD
Volume 24, Issue 3, pages 355–e79, June 2013
How to Cite
Doerr, K. A., Outerbridge, C. A., White, S. D., Kass, P. H., Shiraki, R., Lam, A. T. and Affolter, V. K. (2013), Calcinosis cutis in dogs: histopathological and clinical analysis of 46 cases. Veterinary Dermatology, 24: 355–e79. doi: 10.1111/vde.12026
Conflict of Interest: No conflicts of interest have been declared.
- Issue published online: 14 MAY 2013
- Article first published online: 8 APR 2013
- Manuscript Accepted: 24 FEB 2013
Calcinosis cutis is well recognized in dogs with endogenous hyperglucocorticism and iatrogenic hyperglucocorticism, but the pathogenesis is still unclear.
The objectives of the study were to identify possible correlations between histopathological patterns of dermal mineralization in skin biopsies and underlying causes for calcinosis cutis in dogs, as well as to determine breed predilection and age of onset for dogs within a hospital population. In addition, mineral analysis was performed on four biopsy samples.
Forty-six dogs with histopathologically confirmed calcinosis cutis were evaluated.
Medical records and histological sections of dogs with calcinosis cutis diagnosed by histopathology over a 21 year period were reviewed. Infrared spectrometry was used to identify the mineral in the paraffin blocks. Exact chi-squared test was used to identify breed predispositions, while a Mann–Whitney U-test was used to identify age correlations.
Labrador retrievers, Rottweilers, boxers and Staffordshire terriers were the breeds most commonly affected in this study. Most dogs had either an exogenous or an endogenous source of corticosteroids, with the exception of five dogs with renal insufficiency. In the majority of cases, mineralization was found throughout the entire dermis. The average age of onset of calcinosis cutis for dogs with endogenous hyperglucocorticism was older than that of dogs with iatrogenic hyperglucocorticism. Using infrared spectrometry, apatite crystals were found to be the source of mineral.
Conclusions and clinical importance
There was no observable difference in the histopathological pattern of calcinosis cutis from dogs with endogenous hyperglucocorticism versus iatrogenic hyperglucocorticism. While glucocorticoid therapy appears to predispose dogs to developing calcinosis cutis, it remains unclear whether there is a specific dose or combination of factors that initiates the mineral deposition. Furthermore, the mineral deposition in dogs with calcinosis cutis was found to be apatite.
La calcinose cutanée est bien reconnue chez les chiens atteints d'hypercorticisme endogène et iatrogénique mais sa pathogénie reste obscure.
Les objectifs de cette étude étaient d'identifier les possibles corrélations entre les patrons histopathologiques de minéralisation dermique des biopsies cutanées et les causes sous-jacentes de la calcinose cutanée chez le chien, ainsi que de déterminer les prédilections de race et d'âge d'apparition pour les chiens issus d'une population hospitalière. En outre, l'analyse minérale a été réalisée sur quatre échantillons de biopsies.
Quarante-six chiens atteints de calcinose cutanée confirmée par histopathologie ont été inclus.
Les données médicales et les sections histopathologiques de chiens atteints de calcinose cutanée diagnostiquée par histopathologie sur une période de 21 ans ont été revues. La spectrométrie infrarouge a été utilisée pour détecter les minéraux au sein des blocs de paraffine. Un test de chi-deux exact a été utilisé pour identifier les prédispositions raciales tandis qu'un test de Mann–Whitney U a été appliqué pour identifier les corrélations d'âge.
Les Labradors retrievers, Rottweilers, boxers et Staffordshire terriers étaient les races les plus fréquemment reconcentrées dans cette étude. La plupart des chiens avaient une source de corticostéroïdes soit exogène soit endogène à l'exception de cinq chiens atteints d'insuffisance rénale. Dans la majorité des cas, la minéralisation étaient localisée à travers tout le derme. L'âge moyen d'apparition de la calcinose cutanée pour les chiens atteints d'hypercorticisme endogène était plus élevé que celui des chiens atteints d'hypercorticisme iatrogène. La spectrométrie infrarouge a permis de révéler les cristaux d'apatite comme source minérale.
Conclusions et importance clinique
Aucune différence n'était visible entre les patrons histopathologiques de calcinose cutanée de chiens avec hypercorticisme endogène versus hypercorticisme iatrogène. Alors que les traitements glucocorticoïdes semblent prédisposer les chiens à développer une calcinose cutanée, s'il y a une dose spécifique ou une combinaison de facteurs qui initie les dépôts minéraux reste obscur. En outre, le dépôt minéral chez les chiens atteints de calcinose cutanée s'est révélé être des cristaux d'apatite.
Calcinosis cutis es un fenómeno reconocido en perros con hiperglucocorticismo endógeno y iatrogénico, pero la patogenia no es bien conocida.
Los objetivos de este estudio fueron identificar una posible correlación entre los patrones histopatológicos de mineralización dermal en biopsias de piel y las causas subyacentes de calcicosis cutis en perros, así como determinar si existe una predilección de raza y edad de aparición en una población de un hospital veterinario. Además, se realizó un análisis de los minerales en cuatro biopsias de piel.
Se evaluaron cuarenta y seis perros con calcicosis cutis confirmada mediante histopatología.
Se revisaron los historiales clínicos y las secciones histológicas de perros con calcinosis cutis diagnosticada mediante histopatología durante un periodo de 21 años. Se utilizó espectrofotometría de infrarrojos para identificar el mineral en los bloques de parafina. Se usó la prueba de chi-cuadrado para identificar la predisposición de raza, mientras que se escogió la prueba U de Mann-Whitney para identificar la correlación con la edad.
Las razas mas afectadas en este estudio fueron Labrador Retriever, Rottweiler, Boxer y Staffordshire Terrrier. La mayoría de los perros tenían una fuente de corticosteroides endógena o exógena, con la excepción de cinco perros que tenían insuficiencia renal. En la mayoría de los casos la mineralización se observó en toda la dermis. La edad media de aparición en perros con hiperglucorticismo endógeno fue mayor que en perros con un proceso iatrogénico. Utilizando espectrofotometría de infrarrojos la fuente de mineral se identificó como cristales de apatita.
Conclusiones e importancia clínica
No hubo una diferencia apreciable en el patrón histopatológico entre animales con calcicosis cutis endógena y exógena. Si bien el tratamiento con glucocorticoides parece predisponer a la aparición de calcicosis cutis, aún no esta claro si existe una dosis especifica o una combinación de factores que inicia la deposición de mineral. Además el mineral depositado en perros se identificó como apatita.
Calcinosis cutis ist bei Hunden mit endogenem Hyperglukokortizismus und bei iatrogenem Hyperglukokortizismus gut bekannt, die Pathogenese ist jedoch immer noch unklar.
Die Ziele dieser Studie waren es, mögliche Korrelationen zwischen den histopathologischen Merkmalen der dermalen Mineralisierung in Hautbiopsien und den zugrunde liegenden Ursachen für die Calcinosis cutis bei Hunden herauszufinden, sowie eine Rassenprädisposition und das Alter des Auftretens bei Hunden innerhalb einer Klinikpopulation zu bestimmen. Zusätzlich wurde eine Mineralanalyse an vier Hautbiopsien durchgeführt.
Sechsundvierzig Hunde mit histopathologisch bestätigter Calcinosis cutis wurden evaluiert.
Die Krankengeschichten und die histologischen Schnitte der Hunde mit Calcinosis cutis, die mittels Histopathologie über einen Zeitraum von 21 Jahren diagnostiziert worden waren, wurden durchgesehen. Es wurde die Infrarotspektrometrie verwendet, um das Mineral in den Paraffinblöcken zu identifizieren. Der Exakte Chi-Quadrat Test wurde eingesetzt, um die Rassenprädispositionen zu identifizieren, während der Mann-Whitney U Test verwendet wurde, um die Korrelationen mit dem Alter zu bestimmen.
Labrador Retriever, Rottweiler, Boxer und Staffordshire Terrier waren die Rassen, die in dieser Studie am meisten betroffen waren. Die meisten Hunde hatten entweder eine exogene oder eine endogene Quelle an Kortikosteroiden, mit Ausnahme von fünf Hunden mit Niereninsuffizienz. In der Mehrheit der Fälle wurde die Mineralisierung durch die gesamte Dermis gefunden. Das durchschnittliche Alter, in dem die Calcinosis cutis bei Hunden mit endogenem Hyperglukokortizismus auftrat, war höher als bei Hunden mit iatrogenem Hyperglukokortizismus. Mittels Infrarotspektrometrie wurden Apatitkristalle als Ursprung der Mineralien gefunden.
Schlussfolgerungen und klinische Bedeutung
Es bestand kein deutlicher Unterschied zwischen den histopathologischen Mustern der Calcinosis cutis der Hunde mit endogenem Hyperglukokortizismus versus jenen mit iatrogenem Hyperglukokortizismus. Während die Glukokortikoidtherapie die Hunde zur Entwicklung von Calcinosis cutis zu prädisponieren scheint, bleibt es unklar, ob eine bestimmte Dosis oder eine Faktorenkombination zur Ablagerung der Minerale führen. Weiters bestand die Mineralablagerung bei Hunden mit Calcinosis cutis aus Apatit.
Calcinosis cutis (CC) is an uncommon occurrence, in which inorganic, insoluble mineral salts are deposited in the dermis, subcutis or, rarely, the epidermis. Most commonly, the mineral precipitation involves the collagen and elastin fibres in the dermis. This process can be seen secondary to, or in association with, a number of disorders. It is usually divided into the following four separate categories: metastatic, iatrogenic, dystrophic and idiopathic.[1, 3] Metastatic calcification describes calcium salts that precipitate within normal tissues as a consequence of underlying systemic calcium or phosphorous imbalance. This is most often seen in dogs in the context of impaired renal function, but has also been reported with systemic blastomycosis and paecilomycosis.[5, 6] Iatrogenic CC occurs secondary to percutaneous absorption or injection of calcium-containing products into the skin.[7, 8] Dystrophic calcification usually occurs as a result of local tissue injury and is seen most commonly in dogs with iatrogenic hyperglucocorticism or endogenous hyperadrenocorticism, but has also been reported in association with systemic disease (e.g. leptospirosis). Furthermore, conditions causing inflammation in the skin, such as follicular cysts, foreign body granulomas, interdigital pyoderma, demodicosis or pilomatrixomas, have also been reported to be associated with dystrophic calcification and ossification.[2, 9, 10]
Dogs and chinchillas are the only known species of animals that develop CC in response to hyperadrenocorticism. Idiopathic CC has a histological appearance similar to glucocorticoid-induced CC. However, it occurs in dogs less than 1 year of age with no history of glucocorticoid excess and spontaneously resolves within a year.[2, 10] Idiopathic CC must be differentiated from calcinosis circumscripta, which also commonly occurs in young dogs but does not resolve spontaneously. Calcinosis circumscripta most commonly occurs on pressure points or areas of trauma, especially in young, growing dogs with enhanced calcium and phosphate metabolism.[2, 10] Histologically, calcinosis circumscripta is distinctive, with large, irregular foci of basophilic granular deposits that elicit a marked foreign body response and fibrosis.
Metaplastic ossification in the skin, or osseous metaplasia, is characterized by deposition of calcium and phosphorous as hydroxyapatite crystals in a proteinaceous matrix, with formation of lamellar bone. While rare, it is typically seen in conjunction with multifocal dystrophic CC caused by iatrogenic hyperglucocorticism or endogenous hyperadrenocortism.
To the authors' knowledge, no reports exist that review a large cohort of dogs with CC. As such, the aims of this study were to characterize patterns of mineralization and their association with other histological features and to identify possible correlations with underlying aetiologies. Potential breed, age and sex distributions were also investigated as possible predisposing factors of CC in dogs. In addition, analysis of the mineral composition of the calcified deposits was performed.
Materials and methods
The computer records of dogs seen at the Veterinary Medical Teaching Hospital, (VMTH) from 1 January 1989 to 1 August 2011 were searched using the keywords ‘calcinosis cutis’ and ‘canine’. Inclusion criteria were the diagnosis of CC via histopathology (diffuse mineral deposition in the dermal collagen and adnexal appendages). Dogs with calcinosis circumscripta, characterized by focal mineralization forming a mass, were excluded.
The signalment, body weight, age of onset of skin lesions, lesion distribution, serum calcium levels, concurrent disease and/or medical treatments and their duration were recorded from medical records.
Dogs were organized into the following four groups based upon the reported associated cause for CC: endogenous hyperadrenocorticism, suspect iatrogenic hyperglucocorticism, renal disease and unknown. For dogs with suspect iatrogenic hyperglucocorticism, the dose and length of the course of corticosteroids and/or administration of other immunosuppressive agents, as well as the disease for which corticosteroids had been prescribed, were recorded if this information was available.
Haematoxylin and eosin stained sections of skin biopsies or postmortem skin samples from all dogs included in the study were reviewed. Two authors (V.K.A. and C.A.O.) were blinded as to the clinical history of the dog, while one author (K.A.D.) was not blinded. The following features were evaluated in each sample: (i) presence of epidermal atrophy; (ii) distribution of dermal mineralization, i.e. superficial (the dermo-epidermal junction to the sebaceous gland), mid-dermis (the sebaceous gland to the hair follicle bulb) or deep dermis (the hair follicle bulb to the subcutaneous tissue); (iii) pattern of mineralization, i.e. ‘multinodular’ to describe multiple nodular mineral deposits, ‘multifocal’ for small, scattered mineral deposits and ‘diffuse’ for an even distribution of mineral throughout the dermis; and (iv) the presence and distribution of ossification.
Additional histological findings recorded were as follows: (i) type of inflammatory cells; (ii) amount of fibrosis surrounding any mineral or bone, (iii) folliculitis and/or furunculosis; and (iv) transfollicular or transepidermal mineral elimination. The degree of inflammation, mineralization and fibrosis was assessed subjectively by three investigators as mild, moderate or severe and scored collectively.
The breed distribution within the hospital population was evaluated using an exact chi-squared test, while the age distribution among iatrogenic hyperglucocorticism versus endogenous hyperadrenocorticism was evaluated using a Mann–Whitney U-test.
A total of four samples were examined for mineral extraction and evaluation, as follows: one from a dog with iatrogenic hyperglucocorticism, two from dogs with endogenous hyperadrenocorticism, and one from an unknown disease aetiology. Portions of the paraffin blocks of each case were melted, and the formalin-fixed tissue was collected in Eppendorff tubes.
A razor blade was used to isolate tissue areas with large mineral deposits from the paraffin blocks. The processed samples were subsequently examined by the oil immersion method using a polarized-light microscope (Universal Polarizing Microscope, Carl Zeiss Inc., Thornwood, NY, USA) followed by infrared spectrometry to identify mineral components.
The potassium bromide disc method was adopted in infrared spectrometry. The processed sample and KBr powder were mixed together (approximately 1:150 w/w) and ground manually in an agate mortar with a pestle before the disc was pressed. The infrared spectra were collected using a Nicolet™iS10 FT-IR spectrometer (Thermo Fisher Scientific Inc., Madison, WI, USA) at room temperature immediately after the preparation of the discs. Each spectrum was acquired by the accumulation of 36 scans at a resolution of 4 cm.
The database search revealed 46 dogs that had been diagnosed with CC confirmed by histopathology with slides available for review.
Eight breeds of dogs represented over half of the cases in this study. The most common breeds were Labradors and Labrador mixed breeds (5 of 46; 10.8%), boxers and boxer mixed breeds (5 of 46; 10.8%) and Rottweilers (4 of 46; 8.6%). Akitas, Staffordshire terriers, golden retrievers, pit bull terriers, Pomeranians and German shepherd dogs each represented two of 46 or 4.3% of the cases in this study. Using an exact chi-squared test, there was a significant difference in the breed distribution of dogs affected with calcinosis cutis versus the rest of the VMTH population (P < 0.0001), with Rottweiler/Labrador retriever mixed breeds, Rottweilers, Staffordshire terriers, boxers, boxer mixed breeds, Akitas and Pomeranians over-represented. However, based on the small number within each breed, this finding should be regarded as tentative.
Male dogs were the most common gender group represented (26 of 46), with castrated male dogs representing 36.9% (17 of 46) and intact male dogs 19.5% (9 of 46). Female spayed dogs represented 32.6% (15 of 46) and intact female dogs 10.8% (5 and 46). The median age at the time of diagnosis was 7.6 years (range 1–14 years).
The median age of the dogs with endogenous hyperadrenocorticism with CC was significantly higher (median 11 years; range 1–14 years) than the median age of dogs with iatrogenic hyperglucocorticism with CC (median 6 years; range1–12 years; P = 0.0005).
The mean body weight of all dogs was 26.7 kg (range 2.4–59 kg). Dermatological findings varied from firm, palpable, erythematous papules to well-demarcated plaques. Most dogs had lesions in multiple regions of the body. The distribution of the lesions included the dorsum, flank, axilla, ventral thorax, extremities, inguinal region, head and perigenital region (see Table 1). The distribution of skin lesions had no apparent association with disease aetiology. The subjective nature of these observations in the medical records precluded the ability to perform statistical analysis.
|Location||Endogenous (11)||Iatrogenic (25)||Unknown (4)||Renal (6)|
|Dorsum||100% (11)||64% (16)||50% (2)||50% (3)|
|Flank||9.1% (1)||12% (3)||0||0|
|Axilla||0||12% (3)||0||16.6% (1)|
|Ventral thorax||0||16% (4)||25% (1)||16.6% (1)|
|Extremities||18.1% (2)||20% (5)||50% (2)||33.3% (2)|
|Inguinal region||9.1% (1)||24% (6)||25% (1)||33.3% (2)|
|Head||27.2% (3)||20% (5)||0||50% (3)|
|Perigenital region||9.1% (1)||12% (3)||25% (1)||0|
Upon review of medical records, potential underlying causes of the CC included iatrogenic hyperglucocorticism (25 of 46; 54.3%), endogenous hyperadrenocorticism (11 of 46; 23.9%), and renal insufficiency (6 of 46; 13.1%). Four dogs had unknown underlying disease. Hypercalcaemia was noted in three of 46 (6.5%) of the dogs for which biochemistry data were available for review. One hypercalcaemic dog was receiving calcitriol postparathyroidectomy and had concurrent chronic renal insufficiency [serum calcium 12.9 mg/dL (normal 9.6–11.2 mg/dL) and serum phosphorus 9.3 mg/dL (normal 2.6–5.2 mg/dL)], one hypercalcaemic dog had chronic renal insufficiency only (serum calcium11.8 mg/dL and serum phosphorus 13.5 mg/dL), and one hypercalcaemic dog was diagnosed with meningoencephalitis and was also treated with corticosteroids (serum calcium 13.5 mg/dL and serum phosphorus 13.1 mg/dL).
Known dosages of corticosteroids were recorded in 22 of 46 (47.8%) dogs, whereas in 12 of 46 (26.0%) the therapeutic history or dosages of corticosteroids were unknown; 12 of 46 (26.0%) dogs did not receive corticosteroids. Corticosteroids were the sole drug utilized in 11 of 46 (23.9%) dogs diagnosed with CC; azathioprine and corticosteroids were utilized in 10 of 46 (21.7%) dogs; ciclosporin and corticosteroids were used in five of 46 (10.8%) dogs; cyclophosphamide and corticosteroids were used in two of 46 (4.3%) dogs; and a combination of ciclosporin, azathioprine, corticosteroids and cyclophosphamide was utilized in one of 46 (2.1%) of dogs. Five dogs (10%) had other immunosuppressive/immunomodulatory medications along with oral corticosteroids, including vincristine, luflenomide, tetracycline/niacinamide, ophthalmic cyclosporine, and topical hydrocortisone. Diseases for which treatment with corticosteroids resulted in iatrogenic hyperglucocorticism were as follows: immune-mediated polyarthritis (4 of 22; 18.1%), immune-mediated haemolytic anaemia (3 of 22; 13.6%), myelopathy (3 of 22; 13.6%), meningoencephalitis (3 of 22; 13.6%), allergic dermatitis (2 of 22; 9.1%), pemphigus foliaceus (2 of 22; 9.1%), immune-mediated thrombocytopenia (2 of 22; 9.1%), aplastic anaemia (1 of 22; 4.5%), suspected cutaneous adverse drug eruption (1 of 22; 4.5%) and multifocal brain disease (1 of 22; 4.5%). Types of corticosteroids utilized were prednisone only (16 of 22; 72.7%), methylprednisolone only (2 of 22; 9%) and dexamethasone sodium phosphate (DSP) injectable and prednisone (4 of 22; 18.1%). Dosages of prednisone ranged from 0.214 to 2.12 mg/kg/day with an average of 1.06 mg/kg/day. Methylprednisolone doses ranged from 0.9 to 1.2 mg/kg/day with an average of 1.05 mg/kg/day. Of the four dogs receiving DSP, each received one subcutaneous injection followed by a maintenance dosage of prednisone (range 0.69–1.25 mg/kg/day; average 0.98 mg/kg/day). The average dose of DSP was 0.26 mg/kg/day once with a range from 0.25 to 0.294 mg/kg. The median duration of corticosteroid therapy before onset of CC was 7.5 months (range 6–9 months).
Thirty-seven of 46 (80%) of the histopathological samples were from antemortem excisional or punch biopsies, while nine of 46 (19.5%) were obtained at the time of postmortem examination.
A multinodular pattern of mineralization, including osseous metaplasia, was seen in 13 of 46 (28.2%) cases (Figure 1). A diffuse pattern was seen in 22 of 46 (47.8%; Figure 2), while a multifocal pattern was noted in 11 of 46 (23.9%; Figure 3).
Mineralization was present throughout all layers of the dermis in 27 of 46 (58.6%) cases, limited to the mid-dermis or deep dermis in six of 46 (13%) cases and extended from the superficial to mid-dermis in 12 of 46 (26%) cases. In four of 46 (8.6%) cases, mineralization was found in the subcutaneous tissue as well as the dermis, except for one dog where it was only in the subcutaneous tissue. Osseous metaplasia was observed in nine of 46 (19.5%) cases.
Epidermal atrophy was seen in 14 of 46 (30.4%) cases. Mild to moderate fibrosis was present in 29 of 46 (63%) cases, with scar tissue formation in nine of 46 (19.5%) cases (see Figure 1). Folliculitis and/or furunculosis were observed in 11 of 46 (23.9%) cases.
Histological features in association with clinical risk factors
The incidence of histological features with each disease aetiology is summarized in Table 2. Due to confounding factors of multiple clinicians, dosages and diagnostic tests, and the difficulty of retrospectively assessing the clinical severity of the disease in individual dogs, a statistically supported correlation of disease aetiology and histological findings could not be made. Likewise, there was no evident correlation between breed with subjectively evaluated mineralization pattern, fibrosis and inflammation.
|Endogenous (11)||Iatrogenic (25)||Unknown (4)||Renal (6)|
|Diffuse||27.2% (3)||48% (12)||100% (4)||50% (3)|
|Multinodular||45.4% (5)||28% (7)||0||16.6% (1)|
|Multifocal||27.2% (3)||24% (6)||0||33.3% (2)|
|Epidermal atrophy||63.6% (7)||16% (4)||25% (1)||16.6% (1)|
|Osteoma cutis||45.4% (5)||16% (4)||0||0|
|Transepithelial elimination||45.4% (5)||68% (17)||75% (3)||50% (3)|
|Scarring fibrosis||27.2% (3)||16% (4)||0||16.6% (1)|
Mid-dermal or deep dermal mineralization was seen in two dogs diagnosed with acute renal failure. Endogenous hyperadrenocorticism was diagnosed in seven of 14 (50%) dogs with epidermal atrophy evident on histopathology. Osteoma cutis was found in five of nine (55.5%) dogs with confirmed pituitary-dependent hyperadrenocorticism. No statistically significant correlations could be determined.
Light microscopy revealed grains with very low birefringency (i.e. almost isotropic) among remnants of paraffin in three samples. The grains had a slightly higher refractive index than 1.4960. These optical characteristics are consistent with those of apatite [Ca5(PO4)3(OH,F,Cl)]. Similar grains were not observed in the fourth sample.
The identification of those grains was undertaken by infrared spectrometry. Figure 4 compares infrared spectra of four samples with those of apatite and paraffin standards provided by commercial software installed in the spectrometer. Apatite has the primary and secondary peaks in wavenumber regions of 900–1200 and 500–700/cm, respectively (see Figure 4b). The primary peak is characterized by a shoulder at the higher wavenumber side, and the secondary peak is a composite of two. As seen in Figure 4, the spectrum of the dog in Figure 4c showed two distinct peaks of apatite, suggestive of the highest content of apatite among the four samples examined. Apatite peaks were also observed for the dogs in Figure 4d and e, although they were less distinct than those for the dog in Figure 4c. However, unlike those three samples, the spectrum of the dog in Figure 4f did not show distinct apatite peaks, which was consistent with the light microscopic observations.
The present study documents iatrogenic hyperglucocorticism as the major cause for CC in a cohort of 46 dogs, followed by endogenous hyperadrenocorticism and underlying renal disease. The underlying diseases were not known in four dogs. Male dogs were over-represented and large breeds are predisposed, with Pomeranians (2 of 46 dogs) representing the only small breed in this group. Rottweilers and Rottweiler/Labrador retriever mixes, Staffordshire terriers, boxers, boxer mixes, Akitas and Pomeranians are over-represented when compared with the hospital populations. Statistical analysis of the breed predisposition of CC has not been reported previously, except for documentation of increased risk for endogenous hyperadrenocorticism in boxers, Boston terriers, dachshunds and miniature poodles. While these were statistically significant findings, there were still small numbers of dogs with calcinosis cutis for each breed.
The exact pathomechanism of CC is not clear. It is defined as deposition of insoluble, inorganic mineral salts in the dermis, epidermis or subcutis. The main factors necessary for tissue mineralization to occur include cation and anion deposition in tissues such as collagen or elastin. In normal skin, calcium and phosphate ions are in metastable equilibrium in the extracellular fluid. Changes in collagen fibrils may induce phase transformation of the ions, resulting in formation of crystal nuclei, which are deposited on organic matrices. Furthermore, with abnormally elevated mitochondrial calcium phosphate levels, crystal deposition and cell death may result.
The majority of the dogs included in this study developed CC due to iatrogenic hyperglucocorticism (54.3%) and endogenous hyperadrenocorticism (23.9%). The role of cortisol in the development of CC is considered a result of its gluconeogenic and protein catabolic activity.[14, 15] The rearrangement of the molecular structure of proteins leads to formation of an organic matrix that attracts and binds calcium.[14, 15] Furthermore, tissue damage may lead to dystrophic calcification and further increase of intracellular calcium influx. In humans, it is hypothesized that the ectopic soft tissue calcification in ‘idiopathic’ CC is due to increased amounts of γ-carboxyglutamic acid, which contribute to the calcium and phospholipid binding.[16, 17]
Metastatic calcification occurs most commonly with chronic renal failure, as seen in six dogs (13.1%) in this study. Typically, the calcium-to-phosphorous (Ca:P) product exceeds 70 in order to result in metastatic calcification. Hypocalcaemia, with a resultant Ca:P product >70 was documented in only three of six dogs. There were insufficient numbers of serum biochemical profiles available to evaluate any further trends in the Ca:P product.
Dogs with calcinosis cutis due to endogenous hyperadrenocorticism had a significantly higher median age when compared with dogs with iatrogenic hyperglucocorticism. The median age of dogs with endogenous hyperadrenocorticism was consistent with the previously reported age of onset for Cushing's disease, as 9–11 years. Many of the dogs receiving corticosteroids had an underlying disease that may have originally occurred at a younger age. However, due to the small numbers of each disease entity, age of onset could not be correlated with a specific disease. Due to the small number of dogs with renal disease and the wide age distribution, there did not appear to be any correlation with age of onset of renal disease and onset of calcinosis cutis.
The distribution of skin lesions does not appear to be an indicator for a particular underlying cause for CC. In accordance with previous reports, the dogs had firm, palpable, erythematous papules to well-demarcated plaques, with multiple regions of the body affected. The subjective nature of the dermatological descriptions in the medical record precluded the ability to perform statistical analysis, but the dorsum appeared to be affected in all dogs with endogenous hyperadrenocorticism and was the most common body region affected with calcinosis cutis, while the inguinal and head lesions were the second most prevalent areas of calcinosis cutis (see Table 1).
Inconsistency in documentation of medications administered did not allow documentation of a direct correlation of duration of administration of exogenous corticosteroids or other medication with the development of CC. The retrospective nature of the study precludes the ability to control for the exact dose of corticosteroids used in each dog, as well as to be able to account for any administration error or follow-up. Iatrogenic hyperglucocorticism with CC has been documented to develop after only 7 weeks of corticosteroid therapy or as late as 3 years of dexamethasone therapy every other day.
No correlation was noted between the histopathological findings and underlying disease in dogs with CC. Neither pattern nor distribution of mineral deposits had distinct associations with underlying diseases or medication regimens. Moreover, the patterns and composition of inflammatory infiltrates, which were mostly composed of histiocytes and lymphocytes, as well as the amount of fibrosis, did not indicate particular underlying diseases.
Epidermal atrophy, which is a cutaneous change often seen in dogs with hyperadrenocorticism, was seen in only 30.4% of the dogs in this study and included dogs with all patterns of mineralization. Although not as common as expected, this is probably due to the fact that inflammation elicited by mineral deposition resulted in epidermal hyperplasia. Subjectively, in most biopsies, the follicles appeared to be in telogen.
Osteoma cutis was seen in the majority of dogs with pituitary-dependent hyperglucocorticism and a smaller percentage of dogs with iatrogenic hyperglucocorticism in this study, which has been recognized previously.[2, 20]
Polarizing microscopy and infrared spectrometry identified the deposited minerals as apatite in formalin-fixed tissues of three dogs with CC associated with different underlying diseases. The lack of identification of apatite in the fourth sample, from a dog with an unknown underlying disease, may be due to the small size of the available sample, because no mineral was found. In idiopathic CC in a human, the chemical composition of the mineral was determined qualitatively and quantitatively using both Fourier transform infrared and Raman microscopic techniques.[12, 21, 22] The combination of techniques has revealed the most specific information on the composition of the deposits.[12, 21, 22] The two major components in human idiopathic CC of subcutaneous tissues are type B carbonated apatite and β-carotene. In future studies, it may be prudent to perform Raman spectroscopy to determine whether β-carotene is also a component of CC in canine skin.
In conclusion, this study was unable to determine a correlation between underlying disease processes and the histopathological pattern of mineral deposition in the skin of dogs with CC. It was also evident that corticosteroid therapy plays a role in the development of CC, particularly in younger dogs, but it was uncertain whether dose ranges or a combination of medications influence the risk of mineral deposition. Certain breeds were over-represented that had not been previously documented. Owing to the confounding factor of multiple clinicians, dosages and diagnostic tests, and the difficulty of retrospectively assessing the clinical severity of the disease in individual dogs, a statistically supported comparison of disease aetiology and histological findings could not be made.
- 1Cutaneous mineralization and ossification. In: Goldsmith L, Katz S, Gilchrest B et al., eds. Fitzpatrick's Dermatology in General Medicine. 5th edn. New York: McGraw-Hill, 1999; 1829–1835., .
- 10Neoplastic and non-neoplastic tumors. In: Muller and Kirk's Small Animal Dermatology. 6th edn. Philadelphia: W.B. Saunders, 2001; 1398–1402., , .
- 18Hyperadrenocorticism. In: Ettinger S, Feldman E, eds. Textbook of Veterinary Internal Medicine. 6th edn. St Louis: Elsevier Saunders, 2005; 1592–1600..
- 19Four case studies in the use and misuse of corticosteroid therapy. Vet Med 1991; 86: 1186–1198., .