Computed tomographic findings in canine leproid granuloma syndrome: a case series

Abstract

Objective

To describe the computed tomographic (CT) features of confirmed cutaneous and subcutaneous mycobacterial lesions in dogs.

Animals

CT images from three dogs with confirmed cutaneous and subcutaneous mycobacterial disease were retrospectively reviewed. Dogs presenting to a single hospital between September 2022 and February 2026 for chronic cutaneous and subcutaneous lesions were included if they underwent CT imaging of the lesion(s), surgical biopsy of an active lesion and Mycobacterium Polymerase Chain Reaction (PCR) testing on the tissue sample. Dogs were excluded if Mycobacterium PCR testing was negative.

Clinical Presentation

Dogs presented for chronic multifocal cutaneous and subcutaneous lesions which were unresponsive to initial antimicrobial therapies. Three dogs met the inclusion criteria: a 3.5‐year‐old female spayed Dogue de Bordeaux; a 5‐year 8‐month‐old female spayed whippet cross; and a 4‐year‐old English springer spaniel.

Results

All included cases had a positive mycobacterial PCR of a cutaneous or subcutaneous lesion for which a CT study was available at the time of tissue biopsy. All three cases displayed multifocal cutaneous and subcutaneous soft tissue nodules with peripheral rim enhancement. Two cases had visible cutaneous defects. Two cases had regional lymphadenomegaly. One case had multifocal pulmonary nodules presumed to be mycobacterial granulomas. In two cases, lesions clinically resolved within 5 months of appropriate antibiotic therapy.

Clinical relevance

Cutaneous mycobacterial infection appears similar to other diseases affecting the cutaneous or subcutaneous tissues and may be underdiagnosed due to limited awareness of the disease. CT may be a useful modality for assessing the extent of cutaneous or subcutaneous mycobacterial disease.

Introduction

Mycobacteria are Gram‐positive rod‐shaped bacilli; at the time of writing, there are 188 identified species. ¹ Mycobacteria are a human public health risk, and include the tuberculosis complex, most notably M. tuberculosis, M. africanum, M. avium and M. bovis. ² Many other nontuberculous mycobacterial infections cause cutaneous lesions in humans, including M. leprae (leprosy), M. fortuitum and M. ulcerans, which are thought to be the result of direct inoculation. ³ Cutaneous mycobacterial diseases have been described in dogs, ⁴ , ⁵ , ⁶ , ⁷ , ⁸ and as is for humans, subcutaneous mycobacterial infections in animals are thought to occur through trauma to the cutis or direct inoculation through biting insects. ⁵ In 1998, the term canine leproid granuloma syndrome was coined, and it describes cutaneous and subcutaneous pyogranulomatous lesions in dogs as a result of mycobacterial infection. ⁴ To the authors' knowledge, the computed tomographic (CT) features of nontuberculous mycobacterial disease in dogs have not been described. CT provides rapid, detailed cross‐sectional imaging, allowing visualisation of internal structures without superimposition of other structures. The extent of lesions can be characterised with CT, and in some instances, characteristic lesions allow for differentiation of disease processes. Medical and surgical management of cases may vary between cases, and the lesion distribution identified on CT may inform clinical decisions. The purpose of this case series was to describe the CT findings of three dogs with confirmed cutaneous and subcutaneous mycobacterial infections.

Methods

This retrospective, single‐centre, descriptive case series was conducted using data from the medical records of dogs from the private referral practice Geelong Animal Referral Services. The hospital database was searched to identify those suspected of cutaneous or subcutaneous mycobacterial infection between September 2022 and February 2026. Owner consent had been obtained for all diagnostic procedures and treatments. Database search keywords included “Mycobacteria”, “Mycobacterium”, “Mycobacterium PCR” and ‘PCR’. Dogs were eligible for inclusion if a CT study including the cutaneous or subcutaneous lesion(s) was acquired, a surgical biopsy of an active lesion was collected on the same day, and mycobacterial PCR testing on the tissue sample was performed. Information collected from the medical records included signalment, treatment before referral, clinical findings, CT findings, surgery report, pathology reports including cytology, culture and sensitivity, histopathology and mycobacterial PCR. A diagnosis of Mycobacterium infection was achieved by mycobacterial PCR. Dogs were excluded if they had a negative mycobacterial PCR result.

Cross‐sectional imaging was performed under general anaesthesia. All dogs underwent CT using a helical 16‐slice GE Optima CT520, using the following acquisition parameters: slice thickness 0.56–1.25 mm, pitch 0.56–1.37, kV 120–140, with the patients positioned in sternal recumbency. Images were obtained pre‐ and postcontrast; after administration of 2 mL/kg Iohexol 300 milligrams of iodine per milliliter (mgI/mL), a total dose of 600mgI/kg. Postcontrast images were obtained at 40 and 90–120 s. All images were reviewed by a radiology resident (BC) and a board‐certified radiologist (MM) who were blinded to the laboratory histopathology results. Images were reviewed on a dedicated diagnostic imaging workstation using Keystone Omni (Asteris, Inc. Version 143). Transverse plane acquisitions were reconstructed with high‐ and medium‐frequency algorithms and reviewed in bone and soft tissue windows. Images were reviewed as multiplanar reformats with adjustment of window width, window level and magnification as required.

Owner consent for publication of data was not obtained due to the retrospective nature of the case series and maintenance of patient and client anonymity.

Results

Nine dogs were identified with cutaneous and subcutaneous lesions suspicious of mycobacterial infection and underwent a CT study, surgical biopsy of an active lesion and laboratory testing. Three dogs had confirmation of mycobacterial infection on mycobacterial PCR testing. Six dogs had a negative mycobacterial PCR result.

The six excluded cases were diagnosed with the following: high‐grade mast cell tumor of the left oral commissure with pulmonary metastasis and pyogranulomatous dermatitis (Staphylococcus pseudintermedius); pyogranulomatous dermatitis (Staphylococcus pseudintermedius) of the hock; pyogranulomatous dermatitis (methicillin‐resistant Staphylococcus pseudintermedius) and demodicosis of the face; pyogranulomatous dermatitis (Staphylococcus pseudintermedius) of the right hock; pyogranulomatous and lymphoplasmacytic dermatitis of the right periorbital region with Epicoccum species cultured; and intermediate‐ to high‐grade lymphoma of the subcutaneous tissue of the head and neck with regional lymph node involvement.

Case 1

A 3.5‐year‐old neutered male 51‐kg Dogue de Bordeaux was referred for a 2.5‐year history of a nonhealing, discharging wound on the left flank involving the skin and subcutaneous tissue, which was unresponsive to antibiotic therapy, including amoxicillin/clavulanic acid and cephalexin. Surgical debridement had been attempted but was unsuccessful, and the lesion returned. On presentation to the referral hospital, the dog exhibited a good body condition, and his demeanor was bright, alert and responsive. His heart rate, respiratory rate and rectal temperature were within normal limits. He was ambulatory with a normal gait. There was a 3‐4 mm full‐thickness cutaneous wound at the midlumbar spine to the left of midline, which was associated with the previous surgical scar. One month later, the patient was returned and underwent a CT study of the abdomen and surgical biopsy of the lesion, which was submitted for culture and sensitivity, histopathology and mycobacterial PCR testing.

The CT study of the abdomen revealed multifocal soft tissue subcutaneous nodules ranging in size between 4 mm (spherical shape) to 25 mm × 60 mm (oval shape). Many nodules contained and a fluid attenuating centre, whilst others were soft tissue‐attenuating. Nodules were distributed in collections and chains over a large area of the left ventrolateral abdomen and dorsum, with extension to the right side of the body (Figure 1). Majority of nodules with fluid attenuating centre displayed marked peripheral contrast enhancement. Other nodules (predominantly smaller nodules) showed marked homogenous contrast enhancement. Surrounding fat stranding and thickening of the subcutis were also present. Lesions did not communicate with the peritoneal cavity. Occasional small skin defects with subcutaneous gas were identified on the left abdominal wall. Lymphadenomegaly was not identified. Bacterial culture identified a light growth of E. coli. Histopathology identified marked pyogranulomatous cellulitis with fibrosis. PCR testing revealed the presence of Mycobacterium but was unable to identify the species. The patient was prescribed enrofloxacin (5 mg/kg, PO, once daily), doxycycline (10 mg/kg PO, twice daily) and prednisolone (0.5 mg/kg, PO, every other day). The lesions had clinically resolved at follow‐up two and four months after initiation of medical therapy. At the time of writing, no further follow‐up was available.

Figure 1.

Transverse postcontrast computed tomography (CT) image reconstructed in a soft tissue window of the abdomen at the level of L2 from a dog (case 1) with confirmed subcutaneous mycobacterial infection. Multifocal subcutaneous nodules with peripheral enhancement (*). Window width 375HU; window level 39HU, slice thickness 1.25 mm. D = dorsal. L = Left. R = Right. V = Ventral.

Case 2

A 5‐year 8‐month‐old, spayed female 20.3‐kg whippet cross was presented with a history of severe facial lymphadenomegaly, stiffness and pain on opening her mouth of approximately 12 months' duration, and development of cutaneous nodules on her right antebrachium of 1.5 weeks' duration. Medical therapy of intermittent treatment with amoxicillin‐clavulanate (15.3‐ 30 mg/kg, PO, twice daily) and prednisolone (0.1–1.6 mg/kg, PO, once daily) resulted in short‐term improvement. At presentation, the dog exhibited a good body condition, and her demeanor was quiet, alert and responsive. Her heart rate, respiratory rate and rectal temperature were within normal limits. She was ambulatory with a normal gait. Her mandibular and superficial cervical lymph nodes were markedly enlarged; her popliteal lymph nodes were of normal size. There were subcutaneous nodules present bilaterally near the ventral aspect of the orbits.

A CT study of the head, neck and thorax was performed, and surgical biopsies of the facial nodules and right antebrachium were collected on the same day and submitted for bacterial culture and sensitivity, histopathology and mycobacterial PCR testing. CT findings revealed multifocal ovoid fluid hypoattenuating subcutaneous facial nodules, with moderate peripheral contrast enhancement. The largest nodule measured approximately 16 mm (lateral to lateral) x 29 mm (dorsoventral) × 24 mm (craniocaudal). The nodules were located abaxial to the left and right maxillary bones, ventral to the right and left zygomatic arches, within the left lip commissure and overlying the right temporalis muscle (Figure 2). Moderate regional lymphadenomegaly of the mandibular and medial retropharyngeal lymph nodes was identified. These lymph nodes had heterogeneous soft tissue attenuation, with predominantly peripheral heterogeneous contrast enhancement. The right mandibular lymph node measured approximately 34 mm (lateral to lateral) × 15 mm (dorsoventral) × 32 mm (craniocaudal). Subcutaneous ovoid nodules were also identified on the cranial aspect of the right and left brachia, measuring up to 7 mm in diameter × 26 mm in length (proximodistal). Adjacent to these nodules, were open cutaneous defects in the right antebrachium and left medial aspect of the elbow.

Figure 2.

Transverse postcontrast computed tomography (CT) images reconstructed in a soft tissue window of the head at the level of the eyes (A) and level of the temporomandibular joints (B) from a dog (case 2) with confirmed subcutaneous mycobacterial infection. Multifocal subcutaneous nodules (*) affecting the region of the right maxilla, left zygomatic bone and left commissure of the lips, (A). Enlarged mandibular lymph nodes (<) and a subcutaneous nodule (*) in the right temporalis region (B). Note the peripheral enhancement of all subcutaneous nodules. Window width, 40HU; window level 370HU; slice thickness 1.25. D = dorsal. L = Left. R = Right. V=Ventral.

Bacterial culture from facial nodule biopsy identified a moderate growth of Enterobacter. Histopathology identified severe perivascular to periadnexal neutrophilic and pyogranulomatous dermatitis and cellulitis with folliculitis and furunculosis. On PCR testing, Mycolicibacterium sediminis (formerly Mycobacterium sediminis) was detected.

Empirical medical therapy with antimicrobials with known sensitivity for mycobacteria was initiated. A combination of doxycycline (5 mg/kg, PO, twice daily), enrofloxacin (5 mg/kg, PO, once daily) and clarithromycin (12.3 mg/kg, PO, twice daily), in addition to discontinuation of her chronic prednisolone therapy, was chosen. All facial nodules had clinically resolved following 1 month of antibiotic therapy. At the time of writing, no further follow‐up was available.

Case 3

A 4‐year‐old spayed female 22.8‐kg English springer spaniel was presented for a 9‐month history of a hard lump on the dorsum proximal to the tail base, which subsequently ruptured. Numerous other subcutaneous nodules on the left and right flanks and dorsum had ruptured or had been surgically lanced within the same time frame. Medical therapy included intermittent treatment with amoxicillin‐clavulanate (11.3–22.6 mg/kg, PO, twice daily), cephalexin (27.2 mg/kg, PO, twice daily) and meloxicam (0.1 mg/kg, PO, once daily), which resulted in short‐term improvement. On presentation, the dog's demeanor was bright, alert and responsive. Her heart rate, respiratory rate and rectal temperature were within normal limits. She was ambulatory with a normal gait. She had a diffuse, irregular, firm thickening of soft tissues in the lumbar region, particularly the right lateral lumbar musculature which was uncomfortable on palpation. She had subjectively enlarged superficial cervical lymph nodes.

The patient underwent a CT study of the thorax and abdomen. Tissue samples of the subcutaneous and cutaneous lesions were collected on the same day as the CT and submitted for bacterial culture and sensitivity, histopathology and mycobacterial PCR testing. The CT study revealed multifocal subcutaneous soft tissue‐attenuating nodules (3–8 mm) in the dorsal lumbar region extending over an area approximately 8 cm (cranial to caudal) (Figure 3). The larger nodules contained and a fluid attenuating centre with marked peripheral contrast enhancement, whilst others smaller nodules had homogenous pre‐ and postcontrast attenuation. In addition, adjacent fat stranding, thickening of fascial planes and marked subcutaneous thickening was also evident. The nodules extended into the deeper fascial plane, including the epaxial musculature. Affected musculature was hypoattenuating with heterogeneous contrast enhancement over a length of approximately 12 cm cranial to caudal. Mild regional lymphadenomegaly of the right axillary lymph node and medial iliac lymph nodes was present. These lymph nodes had homogenous pre‐ and postcontrast attenuation. Multifocal soft tissue‐attenuating pulmonary nodules were also identified, ranging between 5 and 8 mm in diameter. These pulmonary nodules were noted in the periphery of the lung lobes.

Figure 3.

Transverse plane postcontrast CT image reconstructed in a soft tissue window, at the level of L3. Multifocal subcutaneous nodules with peripheral enhancement (*) and extension into the left epaxial musculature (^). Window width, 370HU; window level, 40HU; slice thickness, 1.25 mm. D = dorsal. L = Left. R = Right. V=Ventral.

Surgical biopsy of the subcutaneous nodules in the lumbar area was collected. Bacterial culture identified a very light growth of Pasteurella canis and anaerobes. Histopathology identified marked, multifocal coalescing, subacute suppurative and pyogranulomatous cellulitis/panniculitis with focal immature fibrosis, focal epidermal ulceration and neutrophilic pustules. On PCR testing Mycobacterium was detected, but the species could not be identified. The patient was prescribed clarithromycin (11 mg/kg, PO, twice daily), enrofloxacin (6.5 mg/kg, PO, once daily), meloxicam (0.2 mg/kg, PO, once daily) and paracetamol (11 mg/kg, PO, twice daily). The lesions had clinically improved after 2 weeks of medical therapy, after which the dog was lost to follow‐up.

Discussion

This is the first case series to describe the CT features of confirmed cutaneous and subcutaneous mycobacterial infections in dogs. All cases displayed multifocal cutaneous and subcutaneous soft tissue nodules with peripheral contrast enhancement. All lesions were diffuse over a wide area and had varying degrees of subcutaneous thickening and fat stranding. Two of three cases had visible cutaneous defects. The cutaneous and subcutaneous lesions were identified on the head, forelimbs and trunk. No lesions were identified on the ears. Two of three cases had regional lymphadenomegaly; however, the lymph nodes were not sampled, and as such, it could not be determined if this was due to extension of disease or reactive hyperplasia. One of the cases had multifocal pulmonary nodules, for which the dog had no reported respiratory clinical signs. The pulmonary nodules were presumed to be mycobacterial granulomas; however, laboratory samples were not collected from this site. All cases had positive mycobacterial PCR testing. Only one case had the Mycobacterium species identified—case 2, with Mycolicibacterium sediminis (formerly Mycobacterium sediminis). ⁹ All cases had clinical resolution of the cutaneous and subcutaneous lesions within 5 months of appropriate antibiotic therapy. It is unknown how the dogs acquired mycobacterial infections. All cases additionally had concurrent pyogranulomatous infections, which identified E. coli, Enterobacter, Pasteurella canis and anaerobes. In dogs with confirmed cutaneous mycobacterial infections, concurrent pyogranulomatous inflammation involving organisms such as E. coli, Enterobacter spp. or Pasteurella spp. is most likely secondary or opportunistic rather than primary. Mycobacterial pathogens are well recognised for inducing chronic granulomatous to pyogranulomatous dermatitis, as is demonstrated with the presented cases. These chronic lesions disrupt normal skin barriers and create a microenvironment favorable to colonization by environmental or commensal bacteria. In humans, as demonstrated by Gyamfi et al. 2021, secondary microbial infections among Buruli Ulcer patients are common and contribute to delayed wound healing. In contrast, E. coli, Enterobacter and Pasteurella species are typically opportunists in cutaneous contexts, and chronic granulomatous lesions are not typically observed. Their presence within the cutaneous lesions from the presented cases therefore more likely reflects secondary invasion following tissue compromise. This notion is further supported by the lack of clinical resolution of lesions in cases 1 and 3 following initial antibiotic therapies for which the infectious agents were sensitive based on culture and sensitivity testing.

Cutaneous and subcutaneous nontuberculous Mycobacterium infections, now termed canine leproid granuloma syndrome, in dogs were originally described in 1973 and have been documented several times in Australia. ⁴ , ⁵ , ⁸ , ¹⁰ Malik et al. (1998) reported 45 dogs with cutaneous and subcutaneous pyogranulomatous lesions. ⁵ These lesions consisted of single or multiple well‐circumscribed cutaneous and subcutaneous nodules typically on the head and ears, and sporadically on the trunk and distal limbs, ⁵ similar to the lesion distribution reported in this article. Due to the anatomical location of lesions and predominantly short‐coated breeds affected, insect bites were noted as a possible route for disease transmission. ⁵ The cases presented by Malik et al. (1998) and Smits et al. (2012) had no clinical indication of lymph node involvement. ⁵ , ⁸ A case report of a dog with pyogranulomatous cutaneous lesions affecting the face, dorsum, dorsal cervical region and abdomen with generalised lymphadenomegaly identified Mycobacterium avium as the pathogen responsible. ¹⁰ Interestingly, two of the cases presented in this article had regional lymphadenomegaly, one of which had Mycolicibacterium sediminis identified on PCR testing.

Other mycobacterial species may cause similar cutaneous lesions, some of which pose zoonotic risk to humans. Buruli ulcer (M. ulcerans infection) is reported in over 33 countries and causes necrotising cutaneous disease in humans, initially beginning with a papule, nodule or plaque. Buruli ulcer has been reported in dogs ¹¹ and in a horse ¹² in Australia; it is reported to cause ulcerated skin lesions in dogs, typically on the distal limbs. Cutaneous lesions caused by Mycobacterium tuberculosis complex in domestic animals are rarely reported. In a review of 583 cases of canine tuberculosis, 3 cases were reported to have focal cutaneous mass lesions, 2 of which were cause by M. bovis, and 1 was caused by M. microti. ¹³ Recently, canine leproid granuloma was reported to be caused by a member of the Mycobacterium tuberculosis complex isolated from a dog in Uruguay, which had raised painless skin nodules on both pinnae. ¹⁴ Additionally, M. bovis transmission from cats to humans was reported in the United Kingdom. ¹⁵ All cats in the study had severe systemic infections, including discharging lymph nodes, nonhealing discharging wounds and radiographic pulmonary signs. ¹⁵ These reports highlight the necessity for accurate diagnosis due to the public health risk. In humans, M. sediminis (isolated in one case described in this article) has no established clinical significance. ¹⁶

CT features of cutaneous and subcutaneous mycobacterial disease are sporadically reported in the literature. CT features of cervicofacial nontuberculous mycobacterial infections have been reported in children and are described as submandibular adenopathy with single or multiple ring‐enhancing nodules, with minimal fat stranding. ¹⁷ To the authors' knowledge, CT features of cutaneous and subcutaneous lesions caused by mycobacteria have been described in one other nonhuman species. Major et al. (2016) conducted a retrospective study of the CT findings in 20 cats with mycobacterial infections, the majority of which were caused by M. bovis and M. microti. ¹⁸ One of the presented cases displayed widely distributed subcutaneous nodules, two had focal mass lesions over the nasal bridge and one other had an extensive mass lesion over the ventral head and neck. The majority of cats in that study had pulmonary lesions identified on CT. The authors reflected that it is more likely that systemically unwell cats undergo CT rather than those presenting with cutaneous and subcutaneous disease. ¹⁸

Diseases other than mycobacteriosis may cause similar lesions as canine leproid granuloma, including bacterial and fungal infections, immune‐mediated diseases, foreign bodies and neoplasia, all of which require accurate diagnosis for appropriate treatment plans. In 2021, CT features of canine cutaneous lymphoma were described and identified nodular lesions in 80% of cases within the cutaneous, subcutaneous, deep fascial, subfascial and intramuscular layers, and 50% of cases displayed multifocal nodules. ¹⁹ Watton et al. (2021) found that lymphadenopathy was identified on CT imaging in 90% of cases with confirmed cutaneous lymphoma. ¹⁹ In 1999, cutaneous and subcutaneous pythiosis was reported in 15 dogs, describing multifocal ulcerated and nodular cutaneous lesions with draining tracts of the limbs, thoracic wall or perineal regions. ²⁰ Canine sterile nodular panniculitis has been reported to cause solitary or multifocal nodular, ulcerated or nonulcerated cutaneous and subcutaneous lesions commonly identified on the trunk, neck and ventral cervical area. ²¹ , ²² These reports have similar clinical and CT findings to those presented in this article.

This case series documents the CT findings of cutaneous and subcutaneous nontuberculous mycobacterial infections in dogs (canine leproid granuloma) and highlights the necessity for appropriate laboratory testing with biopsy, histology and PCR. The advantage of CT imaging is lesion identification, characterisation and documenting distribution, which may inform surgical and medical treatment plans. CT is also useful for monitoring for disease progression or resolution following medical or surgical therapy.

Future studies should include a larger case series including cases with subcutaneous and subcutaneous nodular lesions of varying etiologies and compare their CT features.

This study has several limitations. Due to the retrospective nature of this case series, the CT acquisition parameters were not standardised. Not all lesions identified on CT imaging were sampled; for example, the pulmonary nodules in case 3 were not sampled to confirm extension of mycobacterial disease. Despite gross resolution of the cutaneous lesions, follow‐up CT was not available for assessment of resolution of deeper lesions, including the epaxial muscles and pulmonary parenchyma for case 3. The small sample size (3 cases) limits concluding on the similarities and variation in disease presentation and CT features for caning leproid granuloma and other nodular skin diseases; it does, however, reflect the rarity of this condition. Finally, the species of Mycobacterium was confirmed in only one case.

Conclusion

Canine leproid granuloma is likely to be underreported due to a lack of awareness of the disease itself and potential subsequent clinical manifestations. This case series demonstrates that CT imaging may be useful in describing the extent of cutaneous and subcutaneous disease and shows it is important to consider mycobacteriosis as a potential cause of chronic cutaneous and subcutaneous lesions. Other disease processes, such as canine cutaneous lymphoma and bacterial pyogranuloma, can have similar clinical and CT features as described for cutaneous and subcutaneous mycobacterium. Accurate diagnosis is crucial for patients presenting with chronic nodular skin disease. In addition, when handling patients with similar clinical signs, the zoonotic potential should be considered in order to protect veterinary staff and pet owners.

Conflicts of interest and sources of funding

The authors declare no conflicts of interest or sources of funding for the work presented here. AI‐assisted technological tools were not enlisted in writing this manuscript or in the production of images.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.