Keywords
septic arthritis - dermatophytosis - shoulder joint - dog
Introduction
Osteomyelitis is an inflammatory process usually secondary to infection which involves
the bone and its medulla. Long-term infection leads to necrosis with lysis and new
bone formation.[1] The adjacent joints may be involved, and it may be difficult to assess whether the
infectious arthritis or the adjacent osteomyelitis is the primary lesion.
Joint and bone infections are usually caused by bacteria inoculated either through
penetrating wounds or hematogenous dissemination.[2] Among the rare cases of mycotic osteoarthritic infections reported in dogs, fungi
such as Blastomyces, Geomyces, Talaromyces, Coccidioides, Histoplasma, and Cryptococcus ssp. have been found.[3]
[4]
[5]
Case History
A 10-year-old Labrador Retriever was referred because of a 3-month history of right
thoracic limb lameness non-responding to various pain managements, including nonsteroidal
anti-inflammatory drugs and corticosteroids. On presentation, non-weight bearing lameness
with severe muscle atrophy of the right thoracic limb was observed. Vital signs were
within normal limits. Extension and flexion of the right shoulder joint were painful,
and the bicipital test was positive. Withdrawal reflex was intact on both front limbs.
Postural reaction was not attempted due to the severe pain. No pain was elicited on
the right elbow and carpus. An articular lesion of the scapulohumeral joint was highly
suspected, although neurological involvement remained possible. Mediolateral radiographs
of the right shoulder joint, taken by the referring veterinarian, revealed an ill-defined
punched-out osteolytic lesion at the articular surface of the caudal glenoid cavity
and the humeral head with new bone formation at the margins of the aforementioned
lysis ([Fig. 1]).
Fig. 1 Radiograph of the shoulder joint, mediolateral view. Ill-defined punched-out osteolysis
and new bone formation at the articular surface of the caudal glenoid cavity and humeral
head with adjacent subchondral bone sclerosis, with mineral opacity in the bicipital
groove. The joint seemed distended.
Computed tomography (CT) revealed a large amount of rough and irregularly outlined
new bone formation at the caudal aspect of the humeral head, caudal border of the
glenoid cavity, at the distal tip of the supraglenoid tubercle and along the bicipital
groove ([Fig. 2A]). The spontaneous attenuation of the joint effusion was high (20–40 Hounsfield unit,
[Fig. 2B]). In the caudal aspect of the humeral head, several small, round-shaped hypoattenuated
defects were visible in the subchondral bone, surrounded by substantial sclerosis.
Fig. 2 Computed tomography of the right shoulder. (A) Large amount of rough and irregularly outlined new bone formation. In the caudal
aspect of the joint, a large amorphous mineral area that was not in contact with any
osseous structure was observed. A thin linear mineral flap is also observed in the
cranial aspect of the joint. The articular surface within the entire joint was slightly
irregularly outlined. (B) The joint was severely distended, and the joint capsule thickened and intensely
enhanced following intravenous contrast administration.
In conclusion, severe right scapulohumeral joint distension, consistent with active
and marked inflammation, was observed. Furthermore, and concomitantly, signs compatible
with osteochondritis dissecans (OCD) of the humeral head, chronic bicipital tendinopathy
and osteoarthrosis were also observed. Based on clinical examination and the imaging
findings, septic arthritis was highly suspected.
A synovial fluid sample was aspirated for cytology; however, due to blood contamination
it was non-diagnostic. Haematologic exams and biochemical profile were unremarkable.
A shoulder joint arthroscopy via lateral approach was performed to explore and to
obtain tissue samples ([Fig. 3]). A focal defect of the cartilage and subchondral bone on the caudal humeral head
was found, compatible with OCD ([Fig. 3A]), as well as severe proliferative synovitis and eburnation on the glenoid cavity.
The cranial compartment of the joint, including the bicipital tendon, could not be
seen due to the severe synovial proliferations in this location. Samples of cartilage
with subchondral bone from the caudal humeral head defect and synovial membrane were
taken for histological analysis and microbiological culture with antibiogram, respectively
([Fig. 3B]). The joint was lavaged with 2L of sterile saline, before standard closure. The
patient was discharged 24 hours postoperatively with standard antimicrobial therapy
of amoxicillin/clavulanate (12.5 mg/kg per os [p.o.] q12h, Kesium, Ceva, France),
pending culture results. Pain was managed with meloxicam (0.1 mg/kg p.o. q24h, Meloxidyl,
Ceva, France) and tramadol (3 mg/kg p.o. q12h, Tralieve, Dechra, United Kingdom) for
7 days.
Fig. 3 Arthroscopy of the right shoulder. (A) Focal osteochondral defect on the caudal surface of the humeral head filled with
fibrocartilaginous tissue. (B) Curettage of this lesion allowed tissues sampling.
Aerobic and anaerobic bacterial cultures were negative. Histologic analysis revealed
synovitis and granulomatous osteomyelitis, with intra-macrophagic fungal components
([Fig. 4B]), compatible with a deep osteoarticular mycosis. The microbiological culture was
continued and at 4 weeks postoperatively, Trichophyton spp. was isolated.
Fig. 4 Histology from tissues withdrawn during the arthroscopy. (A) 100×, Haematoxylin and eosin stain, granulomatous infiltrate adjacent to a scalloped
trabecula of bone; (B) 400×, Periodic acid Schiff, intralesionnal fungal hyphae are visualized (dotted
circles).
Antimicrobial treatment was therefore changed to griseofulvin (10 mg/kg q12h, Fulviderm,
Virbac, France) for 8 weeks. However, as no clinical improvement was observed, the
referring veterinarian decided to interrupt the antifungal treatment, based on a negative
synovial fluid culture obtained by joint aspiration.
Ipsilateral hindlimb lameness with tarsal joint effusion was observed 5 months after
the initial diagnosis. Radiographs of the tarsus showed mild bone and joint surface
remodelling. Although no synovial fluid was sampled, a septic extension to the tarsal
joint was likely. Lameness and pain worsened and the general condition of the dog
deteriorated rapidly and the owner elected euthanasia without consenting to a post-mortem
examination.
Discussion
The most prevalent sylvatic dermatophyte that infects dogs is Trichophyton mentagrophytes which is usually spread to animals and humans by direct contact with contaminated
hair, scales or by fomites.[5] Most Trichophyton infections are suspected to be due to contact with infected rodents or their nests
as they are more adapted to rodents and hedgehogs.[6]
[7] Immunosuppression and endocrine disorders are risk factors for dermatophytosis in
pets, especially spontaneous or iatrogenic hyperadrenocorticism.[6]
[8]
In the reported case and in others, predisposing factors remained undefined.[5] Unfortunately, several circumstances beyond our control resulted in the inability
to further investigate and to explore pathogenetic modalities for this uncommon joint
infection; that is, the remote and distant location the owner was living in, the serious
financial constraints and the unwillingness to consent to further exams such as necropsy.
Also, the immunological status of the animal remained ill-explored; although blood
analyses were unremarkable, the possibility of hyperadrenocorticism had not been investigated.
Immunological suppression secondary to steroids treatment seemed unlikely as it was
prescribed for a short period of time (2 weeks) and at a low dose of 0.5 mg/kg. Direct
or penetrating inoculation into the joint is less likely, given the lack of any trauma
history; nevertheless, contact with rodents and any previously undetected penetrating
wound cannot be completely excluded. Thus, haematogenous spreading to the shoulder
joint seems a possible explanation.
A prevalence of 19.79 to 49.1% of dermatophytosis has been reported in asymptomatic
pets; however, geography, climate and lifestyle are important factors to consider.[9]
[10] It is possible that the patient in our report was asymptomatically and chronically
infected. No culture from the skin was performed to explore this hypothesis; however,
no dermatological abnormality was seen during examination nor reported by the owner.
Deep infection such as septic arthritis or osteomyelitis secondary to Trichophyton's infection is extremely rare.[6] One report in human medicine describes a Trichophyton's proliferation as a complication associated with surgery of a calcaneus fracture
treated with external fixator.[11] In our case, the dog had undergone an osteosynthesis on the contralateral tibia
6 years before. Radiographs from the surgical site were also performed at the time
of diagnosis without radiographic signs of infection. In veterinary patients, there
are few case reports describing uncommon fungal proliferation such as Oxyporus corticola or Rasamsonia piperina causing osteomyelitis, or Blastomyces and Talaromyces georgiensis causing arthritis.[5]
[12]
[13]
[14] To the author's knowledge, this is the first report of Trichophyton's osteoarticular infection in a dog.
A septic arthritis was suspected on presentation based on the involvement of a single
joint, specially a proximal joint, associated with severe pain and the severity of
radiographic signs.[3] The following diagnostic plan was then elected, advanced imaging, cytology (contaminated
with blood), and minimally invasive tissues biopsies and joint exploration, which
allowed this diagnosis.
Computed tomography allowed a better assessment of the osteoarticular remodelling
seen on radiographs. Images of the CT associated with clinical signs made septic arthritis
more likely, even though a tumour remained a possible differential diagnosis.
In septic arthritis, the gold standard for diagnosis includes identification of the
pathogen on a culture. Synovial fluid aspiration has only 30 to 50% likelihood to
identify pathogens with standard culture medium and sensibility can be increased to
53 to 80% using blood enriched media, when considering bacterial infections.[15]
[16]
[17] However, by adding synovial membrane (to synovial fluid) in the culture, it has
been observed that the efficacity to isolate the causal pathogen could reach 100%.[18] In our case, culture on synovial aspiration was not attempted as it was already
planned to sample tissues during arthroscopy. It could have been interesting to compare
results from synovial fluid only and samples harvested during arthroscopy. This case
also illustrates the importance to complete a culture for both bacterial and fungal
culture when a septic arthritis is suspected, assuming that it can take a longer time
to grow fungi than bacteria. Here, the first results regarding aerobic and anaerobic
bacteria were negative but Trichophyton spp. was isolated 10 days later.
In humans, arthroscopic management of septic arthritis is the preferred technique
for most, especially large joints.[19]
[20]
[21] In our case, we elected to perform arthroscopy first, because cytology alone was
not diagnostic and tissue samples from obvious affected area were needed for diagnostic
purpose.[18] Second, copious joint lavage is known to decrease pain associated with joint effusion
and inflammation, and is highly recommended in a patient with septic arthritis.[3] Moreover, joint exploration allowed the visualization of a former OCD lesion seen
on CT; exposure of the subchondral bone of the glenoid cavity was visible and hard
on palpation. A kissing lesion was suspected. Based on the symmetry of the lesion
on the CT and the quality of the sample harvested we did not attempt to sample the
glenoid cavity. This OCD lesion could have been the primary lesion leading to (osteo)arthritis
and thus could be seen as locus minoris resistentiae for secondary infection.
Griseofulvin, a fungistatic antibiotic medication primarily used in dermatophytic
infections, was empirically elected.[22] As no improvement was achieved, its effectiveness in this case remained uncertain
and a possible resistance must be questioned. Mean duration of treatment in small
animals seems to be close to 40 days which means it could take longer to eliminate
Trichophyton spp., and prolonged treatment may have been successful.[23] Based on a recent consensus itraconazole, a fungistatic triazole, seems to be a
more efficient treatment in Trichophytosis and could have been the preferred treatment in this case.[7]
[22] In the unique similar report in human medicine, complete recovery was achieved after
a 3 months' treatment with itraconazole.[11]
Conclusion
In a dog suffering from severe pain in the shoulder, synovial membrane as well as
subchondral bone biopsies from representative areas were harvested arthroscopically
and allowed diagnosis of a dermatophytic septic arthritis and osteomyelitis due to
Trichophyton spp.
Oral griseofulvin for 2 months was an unsuccessful treatment. Itraconazole has been
efficient to treat a similar osteomyelitis in a human. Origin of the infection is
unclear, although dermatophytosis has been reported in healthy animals and can lead
to the suspicion of a haematogenous spread.
