Introduction
Clinically and radiographically aggressive bone lesions are most commonly caused by osteosarcoma (OSA) in dogs; however, osteomyelitis is an important differential diagnosis and may be fungal or bacterial in nature.1,2 In areas where the organisms are endemic, an important cause of fungal osteomyelitis is Coccidioides spp. Coccidioides immitis and Coccidioides posadasii are dimorphic fungal organisms that thrive in the dusty, dry soil of the southwestern United States as well as Central and South America.3,4 Infection typically occurs via inhalation of the organism, but may progress to dissemination throughout the body, affecting the skin, CNS, and bone, among other tissues.3,4
Coccidioides osteomyelitis is rarely described in the veterinary literature, with most relevant reports having been published in the 1950s and an additional case report published in 1970.5–8 The condition was first described by Hage and Moulton8 in 1954, who reported fulminant disease identified both radiographically and at necropsy in 4 dogs. That report, like the others from the mid-20th century, describes the clinical findings obtained with the diagnostic tools available at that time, but does not provide much useful information for present-day veterinary practitioners. In a 2003 article, Johnson et al9 describe 24 dogs with coccidioidomycosis. Twelve dogs underwent radiography, and 6 of the 12 had bony changes. However, no information about treatment or outcome was provided.
Given the lack of information on Coccidioides osteomyelitis (COM) in dogs and the importance of differentiating COM from OSA, the objectives of the study reported here were to describe the signalment, clinical signs, serologic test results, treatment, and outcome of dogs with COM and to compare those findings with findings for dogs with OSA. We hypothesized that dogs with COM would require prolonged (> 1 year) treatment with an azole antifungal and that they would differ from dogs with OSA in that they would be younger, weigh less, and have lesions that were more commonly proliferative, axial, and polyostotic.
Materials and Methods
Electronic medical records of the Midwestern University Companion Animal Clinic were searched to identify dogs examined between January 2015 and August 2020 for which diagnosis codes related to coccidioidomycosis, osteomyelitis, aggressive bone lesions, or OSA had been assigned. Dogs were included in the COM group if coccidioidomycosis had been diagnosed by means of serologic testing within 60 days of identification of a bone lesion and if bone lesions compatible with fungal osteomyelitis had been documented on radiographs examined by a board-certified radiologist. In addition, dogs included in the COM group were required to satisfy at least one of the following criteria: confirmation of the diagnosis of coccidioidomycosis via histology or cytology, improvement of bone lesions on radiographs after fluconazole treatment was initiated, or improvement of clinical signs while undergoing fluconazole treatment with at least 90 days of follow-up. Dogs were included in the OSA group if they had a bone lesion compatible with OSA on radiographs examined by a board-certified radiologist and histologic confirmation of the lesion as OSA.
Data gathered from the medical records of dogs with COM consisted of signalment; comorbidities; clinical signs; results of serologic testing for coccidioidomycosis, diagnostic imaging, and cytologic or histologic examination; dosage of fluconazole and duration of treatment; adjunctive medications; and recheck intervals. Data gathered for dogs with OSA consisted of signalment, comorbidities, clinical signs, and results of serologic testing for coccidioidomycosis, diagnostic imaging, and histologic examination.
Age and body weight were compared between groups with a Welch t test after data were determined to be normally distributed with the Shapiro Wilk test. The Fisher exact test was used to determine whether group (COM vs OSA) was significantly associated with specific clinical signs (present vs absent), extent of bone involvement (monostotic vs polyostotic), or disease distribution (appendicular vs axial). Bone involvement was considered polyostotic if lesions involved > 1 discrete bone; the bones of each hemipelvis (os coxae) were considered a single bone for the purposes of statistical analysis. Bony lesions were considered nonadjacent if there was radiographically normal bone between affected areas. A χ2 test was used to determine whether group (COM vs OSA) was significantly associated with radiographic appearance (lytic, proliferative, or mixed). Changes in IgG titers over time were evaluated with a Wilcoxon matched-pairs signed rank test. All analyses were performed with standard software (GraphPad Prism version 9.0.0; GraphPad Software). Values of P ≤ 0.05 were considered significant.
Results
The search of the electronic medical records database identified 140 cases of coccidioidomycosis, of which 11 satisfied the inclusion criteria. An additional evaluation of 56 records coded for nonspecific terms related to osteomyelitis identified 9 additional cases that satisfied the inclusion criteria. Of these 20 cases, 6 were excluded because the duration of follow-up was inadequate (n = 3), the diagnosis of coccidioidomycosis had been made > 60 days prior to the identification of bone lesions (2), or radiographic bone lesions were not typical for fungal osteomyelitis (1). The remaining 14 cases were included in the COM group.
The electronic medical records search also identified 113 cases of OSA, of which 25 satisfied the inclusion criteria. Of these 25 cases, 9 were excluded because of a lack of histologic confirmation (n = 5) or because radiographic images had not been evaluated by a board-certified radiologist (4). The remaining 16 cases were included in the OSA cohort.
Dogs with COM
Mean weight of the 14 dogs with COM was 19.1 kg (SD, 6.7 kg), and mean age was 4.8 years (SD, 3.9 years). Two were sexually intact females, 9 were spayed females, and 3 were neutered males. Mixed-breed dogs were most common (n = 3), followed by American Pit Bull Terriers (2). Clinical signs at presentation included pain (n = 12), lameness (10), a visible swelling (4; Figure 1), respiratory signs (2), and paraplegia (1). Clinically relevant comorbidities included atopy (n = 2) and a mammary mass, elbow osteoarthritis, and an oral spindle cell tumor (1 each). All dogs were positive for coccidioidomycosis on serologic testing, with a median IgG titer of 1:48 (range, 1:8 to 1:256). Six of 13 dogs had positive IgM serologic test results; IgM was not measured in the remaining dog. All dogs began treatment with fluconazole at a mean dosage of 18.1 mg/kg/d (SD, 3.0 mg/kg/d). Adjunctive treatments initiated at diagnosis included prednisone (n = 7), tramadol (7), carprofen (4), and gabapentin (3).
Limb swelling associated with a distal humeral lesion in an approximately 6-month-old female mixed-breed dog with Coccidioides osteomyelitis (COM).
Citation: Journal of the American Veterinary Medical Association 260, 1; 10.2460/javma.20.12.0696
Bone lesions were appendicular in 6 dogs and axial in 5 dogs; in the remaining 3 dogs, lesions affected both the appendicular and axial skeleton. Thirteen lesions were documented in the appendicular skeleton; they were located in the humerus (n = 6), scapula (2), radius (2), tibia (2), and ulna (1). Twelve lesions were documented in the axial skeleton involving the ribs (n = 4), pelvis (2), and thoracic (2), cervical (2), and lumbar (2) vertebrae. The median number of sites affected was 1, with 9 dogs having monostotic disease and 5 dogs having polyostotic disease; the greatest number of sites affected in a single dog was 7, with lesions in the scapula, humeri, hemipelvis, a rib, and 2 nonadjacent vertebrae. Four dogs with polyostotic disease had nonadjacent lesions, and 1 dog had osteomyelitis involving T6, T7, and T8. Of the 22 radiographically documented lesions, 15 had a mixed appearance, 4 were predominantly proliferative, and 3 were predominantly lytic.
Three dogs had a definitive diagnosis of COM on the basis of cytologic examination (n = 1), histologic examination of a biopsy specimen (1), or necropsy (1). The dog in which the diagnosis was confirmed by means of cytologic examination had lesions of the right humerus and left radius, and fungal spherules were identified on examination of a fine-needle aspirate of the humeral lesion (Figure 2); this dog did not return for further evaluation after the initial diagnosis. The dog in which the diagnosis was confirmed by histologic examination had osteomyelitis of T12, and a biopsy sample was obtained during a T11-T13 hemilaminectomy. The dog’s condition improved from paraplegic to nonambulatory paraparetic following surgery, but the dog was lost to follow-up 17 days postoperatively (31 days after starting fluconazole treatment). The dog in which the diagnosis was confirmed at necropsy was a 14-year-old German Shorthair Pointer that had a concurrent oral spindle cell tumor. This dog was euthanized because of cachexia, decreased appetite, lethargy, and impaired mobility 78 days after diagnoses of coccidioidomycosis and left humeral osteomyelitis were made. At necropsy, pyogranulomatous inflammation was identified in the meninges surrounding the brain and myocardium, and pyogranulomatous inflammation with intralesional Coccidioides organisms was identified in a mass surrounding the carina, the tracheobronchial lymph nodes, the left humerus, the right radius, and the left fifth and right sixth ribs.
Photograph of an immature Coccidioides spherule and pyogranulomatous inflammation in a fine-needle aspirate of the humeral lesion in the dog in Figure 1. Wright-Giemsa stain; bar = 10 μm. Image courtesy Dr. Kathryn Wycislo, DVM, PhD, DACVP.
Citation: Journal of the American Veterinary Medical Association 260, 1; 10.2460/javma.20.12.0696
The remaining 11 dogs were all followed up for > 90 days after the initiation of fluconazole treatment, and clinical improvement was documented in all 11. Follow-up time ranged from 99 to 1,041 days (median, 356 days). All dogs had a minimum of 2 recheck evaluations after the initial diagnosis, with a maximum of 10 recheck evaluations (median, 3 recheck evaluations).
Dogs were reexamined approximately 1, 3, 6, 12, 18, and 24 months after the initiation of fluconazole treatment (Table 1). Compared with the percentage of dogs with clinical signs prior to treatment (14/14), percentages of dogs with clinical signs were significantly decreased 1 (P = 0.002), 3 (P = 0.002), 6 (P < 0.001), and 12 (P = 0.001) months after the initiation of fluconazole treatment. Similarly, IgG titers were significantly lower 6 (P = 0.02) and 12 (P = 0.008) months after initiation of fluconazole treatment, compared with titers prior to treatment (Figure 3).
Clinical signs and serum anti-Coccidioides IgG titers over time in 14 dogs with Coccidioides osteomyelitis.
| Time interval | No. examined | Median (range) time after diagnosis (d) | No. with clinical signs of osteomyelitis | IgG titer | |
|---|---|---|---|---|---|
| No. tested | Median (range) titer | ||||
| Baseline | 14 | NA | 14 | 14 | 1:48 (1:8–1:256) |
| 1 mo | 10 | 35.5 (29–51) | 4a | 1 | 1:8 |
| 3 mo | 8 | 89 (73–100) | 3a | 7 | 1:8 (1:8–1:64) |
| 6 mo | 8 | 174.5 (140–233) | 0a | 7 | 1:16 (1:4–1:64)a |
| 12 mo | 9 | 336 (253–386) | 2a | 9 | 1:8 (undetectable to 1:32)a |
| 18 mo | 4 | 553 (509–596) | 0 | 4 | 1:20 (1:4–1:32) |
| 24 mo | 3 | 694 (678–702) | 0 | 2 | 1:12 (1:8–1:16) |
NA = Not applicable.
aSignificantly (P < 0.05) different from baseline value (ie, value obtained at the time of diagnosis of Coccidioides osteomyelitis).
Median serum IgG titers in 14 dogs with COM before (0 months; baseline) and 3 (n = 7), 6 (7), and 12 (9) months after initiation of treatment with fluconazole. Error bars represent the upper limit of the range. *Significantly (P < 0.05) different from baseline titer.
Citation: Journal of the American Veterinary Medical Association 260, 1; 10.2460/javma.20.12.0696
Four dogs had IgG titers that increased after falling initially. However, in all 4 dogs, the titer increased by only 1 dilution, and none of the dogs had clinical signs of disease at the time these values were obtained. Three of these dogs were lost to follow-up after the increase in titer; however, the fourth dog had extensive follow-up (1,041 days with clinical remission achieved) documenting that this increase in IgG titer was transient and the titer was lower or unchanged at later recheck examinations.
Two dogs were found to have an IgG titer < 1:4 after the initiation of treatment and were considered to be in clinical remission. One of these dogs did not have a detectable IgG titer 356 days after initiation of treatment for a solitary lesion affecting C2. This dog’s initial titer was 1:16, which decreased to 1:4 after 73 days of treatment and was undetectable when next evaluated on day 356; no recheck examinations performed after fluconazole was discontinued were available for review. The other dog had an IgG titer of 1:2 after 978 days of treatment for a scapular lesion, at which time fluconazole was discontinued; 63 days after fluconazole was discontinued (1,041 days after initiation of treatment), the IgG titer remained 1:2. This dog had had an initial IgG titer of 1:64, which decreased to 1:8 by the time of the first recheck examination at 51 days; however, no further decrease in titer was noted until 880 days, when the IgG titer was 1:4. After 1 year of treatment, 8 of 9 dogs with follow-up data had persistently high IgG titers (Figure 3).
Ten dogs were reexamined approximately 1 month after the initiation of treatment (Table 1). At that time, clinical signs had resolved in 6 dogs and markedly improved in another 2, including a dog with swelling of the elbow region secondary to a COM lesion of the olecranon and a dog with residual neurologic deficits related to a T12 lesion treated via hemilaminectomy. Signs of pain and marked clinical signs were still present in the remaining 2 dogs. This consisted of the dog with lesions in 7 sites and a dog in which coccidioidomycosis had been diagnosed 24 days earlier, but osteomyelitis of C2 had not been appreciated and chiropractic therapy had been administered by the referring veterinarian. At the time of the 1-month recheck examination, 4 dogs were finishing tapering courses of prednisone, but no other analgesic medications were being administered. Both dogs with persistent clinical signs were prescribed tapering courses of prednisone at an initial anti-inflammatory dosage.
Eight dogs were reexamined approximately 3 months after the initiation of treatment (Table 1). At this time, clinical signs had resolved in 5 of the dogs, improved in 2, and remained unchanged in 1. The dog in which clinical signs remained unchanged was the 14-year-old dog, and the dog was euthanized at this visit. One of the dogs that was improved still experienced occasional (monthly) lameness, and the other had a recurrence of neck pain after discontinuing gabapentin.
Eight dogs were reexamined 6 months after the initiation of treatment, and clinical signs had resolved in all 8 dogs. Nine dogs were reexamined 12 months after the initiation of treatment, and owners of 2 of these dogs reported that the dogs had subtle, nonspecific signs of discomfort, such as difficulty rising or lying down. No apparent pain or weakness was noted on examination of either of these dogs. One dog had lesions affecting T6-T8; the other dog had a lesion affecting the proximal ulna. None of the dogs examined 18 (n = 4) and 24 (3) months after the initiation of treatment had clinical signs referable to COM. One year after diagnosis, 8 dogs were still receiving fluconazole because results of coccidioidomycosis titer testing were positive, 1 dog had negative coccidioidomycosis titer testing results but was still receiving fluconazole, 1 dog was dead, and 4 dogs had been lost to follow-up.
Follow-up diagnostic imaging was performed in 5 dogs. Most commonly, follow-up radiography was performed approximately 3 months (range, 73 to 99 days) after the diagnosis of coccidioidomycosis. One dog had imaging performed on days 38, 99, and 178, and another dog had radiography performed on days 73 and 140. Three dogs showed improvement in the radiographic appearance of the lesions, as indicated by sclerosis, increased opacity in areas of lysis, and a smoother, more mature periosteal reaction (Figure 4). In 1 dog, the radiographic appearance of the lesion was static, and in another dog, mixed changes in a C1 lesion were noted. This lesion had a reduction in the irregular periosteal reaction on the dorsolateral portion of the vertebra, but the contralateral aspect had a more prominent periosteal reaction, suggesting persistently active osteomyelitis.
Dorsoventral radiographic images of the pelvis of a 4-year-old spayed female Boston Terrier with COM involving the left hemipelvis. Images were obtained at the time of diagnosis (A) and 3 (B), 6 (C), and 12 (D) months after initiation of treatment with fluconazole. Notice the increased sclerosis, smoother margins, and contraction of the lesion affecting the ilium, acetabulum, and ischium over time.
Citation: Journal of the American Veterinary Medical Association 260, 1; 10.2460/javma.20.12.0696
Dogs with OSA
The 16 dogs with OSA had a mean weight of 35.9 kg (SD, 11.3 kg) and mean age of 8.5 years (SD, 3.6 years). Nine were spayed females, and 7 were neutered males. Mixed-breed dogs were most common (n = 5), followed by Labrador Retrievers (4). Clinical signs at presentation included pain (n = 13), lameness (11), and a visible swelling (4). Serologic testing for coccidioidomycosis was performed in 15 of the 16 dogs. In the remaining dog, serologic testing was not performed, but the diagnosis of OSA was confirmed via cytology prior to amputation. A single dog with OSA had a positive serologic test result (IgG titer, 1:8); this dog presented for lameness and swelling of a thoracic limb, and radiography revealed a lytic lesion of the proximal humerus. The dog was treated with fluconazole but re-presented 20 days later with worsened swelling and pain; recheck radiography revealed worsening of the humeral lesion with progressive cortical thinning as well as evidence of soft tissue nodules in the lungs. Euthanasia was elected and OSA was confirmed in the lungs and humerus at necropsy.
Bone lesions were appendicular in 15 of the 16 dogs with OSA and axial in 1. Locations of appendicular lesions included the distal femur (n = 4), distal radius (3), proximal tibia (3), proximal humerus (2), distal tibia (2), and metatarsus (1); the single dog with an axial lesion had nasal OSA. Lesions were monostotic in 15; the dog with metatarsal OSA was unique in having polyostotic lesions affecting the second and third metatarsal bones with a periosteal reaction on the distal row of tarsal bones. Four OSA lesions were predominantly lytic, and 12 were mixed in appearance. All OSA lesions were confirmed histologically, 10 via biopsy (9 amputations and 1 nasal biopsy), 3 via amputation and necropsy, and 3 via necropsy alone.
Comparisons of dogs with COM versus OSA
Dogs with COM were significantly (P = 0.01) younger and weighed significantly (P < 0.001) less than dogs with OSA, and dogs with COM were significantly (P = 0.006) more likely to have axial lesions than were dogs with OSA. Extent of bone involvement (monostotic vs polyostotic) was not significantly different between groups; however, polyostotic disease involving nonadjacent bones was significantly (P = 0.04) more common in dogs with COM. Radiographic appearance (lytic vs proliferative vs mixed) was not significantly different between dogs with COM and dogs with OSA (Figure 5).
Radiographic appearance of lesions in a dog with COM (A) and a dog with osteosarcoma (B). Both lesions have marked periosteal proliferation, relatively well-defined margins along the periosteal reaction, and areas of cortical thinning and loss. For both dogs, the diagnosis was made on the basis of cytologic testing; the diagnosis in the dog with osteosarcoma was confirmed by means of histopathology.
Citation: Journal of the American Veterinary Medical Association 260, 1; 10.2460/javma.20.12.0696
Discussion
In the present study, dogs with COM typically had a rapid improvement in clinical signs after initiating treatment with fluconazole but required long-term antifungal treatment, with IgG titers remaining high in most dogs 1 year after the diagnosis was made. Dogs with COM differed from dogs with OSA in being younger, weighing less, and having more axial and nonadjacent polyostotic lesions. However, radiographic features had a great degree of overlap between groups, confounding the ability to make a diagnosis on the basis of diagnostic imaging alone.
Dogs with COM in our study were significantly younger and weighed significantly less than dogs with OSA, confirming our hypothesis. This finding is consistent with the current understanding that OSA affects predominantly older dogs, with a reported median age of 7 years.1 However, OSA can also occur in younger dogs,1 and 1- and 3-year-old dogs with OSA were documented in the present study. Also, the oldest dog in either group was a 14-year-old German Shorthair Pointer that had polyostotic COM, emphasizing that this is not a disease of exclusively young or middle-aged dogs. Findings for the OSA group were also consistent with previous studies1 documenting a tendency for the disease to affect large- and giant-breed dogs, whereas COM was noted in small, medium, and large dogs in our study.
Presenting clinical signs of pain, lameness, and visible swelling were similar between groups in the present study. Although 1 dog with COM affecting the vertebral column presented for neurologic deficits, vertebral COM lesions were most commonly discovered incidentally or noted in dogs presented for nonspecific discomfort. Two dogs in the COM group also presented with respiratory signs, which is consistent with the inhalation route of transmission of Coccidioides spp.3,10 Respiratory signs in dogs with OSA could occur as a result of pulmonary metastatic disease, but respiratory signs were not observed at presentation in any of the dogs with OSA in this study. One dog with OSA did develop hemoptysis 176 days after amputation, and euthanasia was elected; bronchial invasion of a large pulmonary metastatic lesion was identified at necropsy.
Previous reports4,11 have discussed a propensity for COM lesions in dogs to be predominantly appendicular in location. In contrast, COM lesions in humans are predominantly axial, affecting the vertebral column, ribs, and pelvis most commonly.11–13 Dogs with COM described in the present study did not demonstrate a marked predilection for appendicular lesions, with 6 dogs having appendicular lesions, 5 having axial lesions, and 3 having both appendicular and axial lesions. Axial lesions were also significantly more common among dogs with COM than among dogs with OSA. In 1 dog with COM, only an appendicular lesion was identified antemortem on radiographs, but at necropsy, additional lesions of the ribs and radius were identified. This case raises the possibility that some COM lesions may go undiagnosed on the basis of clinical signs alone. If COM is diagnosed or highly suspected, a thorough orthopedic examination followed by nuclear scintigraphy or survey radiography, particularly of the axial skeleton, should be considered in an effort to determine the extent of disease. In humans, diagnosis of lytic axial COM lesions via radiography is notoriously difficult, and nuclear scintigraphy is employed as a sensitive method for the detection of affected areas.12,13
Despite the possibility of undiagnosed polyostotic disease, most lesions in both groups in the present study were apparently monostotic. A single dog with OSA had polyostotic disease, whereas 5 of the 14 dogs with COM had polyostotic disease. Only dogs with COM had polyostotic lesions at distant (nonadjacent) bony locations, which was significantly different from the case for dogs with OSA. Although metastasis to skeletal sites in dogs with OSA is described, this was not documented in any of the dogs in this study. Reported rates of osseous metastasis in dogs with OSA detected with various modalities vary1; however, the 4 of 14 dogs with COM in the present study with nonadjacent osseous lesions appeared to exceed the percentage of dogs with OSA that have nonadjacent lesions documented by nuclear scintigraphy. Authors of a study14 evaluating 399 dogs with OSA identified suspected nonadjacent metastatic bony lesions via nuclear scintigraphy in only 7.8% of the dogs. Results of our study, combined with results of these previous studies, suggest that the index of suspicion for COM over OSA should be increased in cases in which multiple distant aggressive bone lesions are identified; however, this finding does not exclude OSA or other neoplastic diseases from consideration.
As has been described previously, OSA has a variety of radiographic appearances and may vary from predominantly lytic to predominantly productive or osteoblastic.1,2 When broadly classified as proliferative, mixed, or lytic, radiographic appearance was not significantly different between the OSA and COM groups in our study. Thus, we rejected our hypothesis that COM lesions would appear more proliferative, compared with OSA lesions. Most lesions for both groups had both proliferative and lytic components. This is consistent with previous descriptions of overlap in the radiographic appearance of neoplasia and fungal osteomyelitis.2
Serologic IgG titers ranged from 1:8 to 1:256 at the time of diagnosis for dogs with COM. Serologic testing in human patients with coccidioidomycosis has a demonstrated relationship between IgG titer and disease severity.11 Although a similar relationship has not been clearly described in dogs, it has been suggested that higher titers may be associated with disseminated or more severe infection.4 In a longitudinal study of dogs in Arizona, Shubitz et al15 found that only 5 of 28 (18%) dogs with IgG titers ranging from < 1:2 to 1:16 had clinical signs associated with infection, and Greene4 has suggested that titers ≤ 1:8 are only suggestive of infection in dogs. In a previous study9 that evaluated results of serologic testing for 20 dogs with coccidioidomycosis, no obvious correlation was identified between IgG titer and clinical sign duration or severity. Similarly, in the dogs of this study, no clear trend between disease severity and titer was noted.
In the present study, there was a decrease in IgG titers over time, with the median titer significantly lower, compared with the baseline titer, after 6 and 12 months of treatment. In general, the IgG titer decreased after initiation of treatment with fluconazole, except in 4 dogs that had a slight increase after initiating treatment. However, whether this was due to inadequate dosing or absorption of fluconazole or expected mild fluctuations in titers between laboratories was not known. Although fluconazole has a recommended dosage range of 10 to 20 mg/kg/d,4 anecdotally, clinicians in endemic areas managing disseminated cases typically prescribe fluconazole at dosages closer to the high end of the range, as evidenced by the mean dosage of 18.1 mg/kg/d in our study. Inadequate bioavailability of the drug may play a role in clinical efficacy and changes in titer; however, the bioavailability of various fluconazole formulations has not been evaluated, and none of these dogs had blood fluconazole concentrations measured. Itraconazole has been proposed as the more effective azole drug for skeletal osteomyelitis in humans12; however, fluconazole is typically the mainstay of treatment in veterinary medicine owing to cost, formulation, and the apparent efficacy of a compounded product, whereas compounded itraconazole is not recommended for use in dogs.16,17
The prognosis for dogs with COM has not been well described, with some references reporting a 0% likelihood of complete recovery for dogs with multiple bone involvement.4 On the basis of clinical signs and results of serologic testing, 2 of the 14 dogs in this study achieved clinical remission at days 356 and 978 after initiating treatment. However, 8 of 9 dogs for which follow-up information was available 1 year after diagnosis still had IgG titers ≥ 1:8, indicating the need for prolonged azole treatment. One dog had no clinical signs 62 days after fluconazole treatment was ended; however, the likelihood of recrudescence, even after prolonged treatment in dogs, is unknown. Response to treatment of skeletal infections in humans ranges from 23% to 100%, but relapse is common and treatment with antifungal medication is often lifelong.13 Close monitoring of dogs with COM even after apparent clinical resolution with low or undetectable IgG titers is warranted owing to the possibility of recrudescence.
In most instances, clinical signs improved rapidly in dogs with COM in the present study. Intermittent mild signs such as stiffness after rest were still noted by a few owners after 1 year, but dogs had an apparently good quality of life. Dogs that did not have marked, rapid clinical improvement included the 14-year-old dog with multisystemic involvement affecting the lymph nodes, lungs, heart, brain, and multiple bones. This dog may have had some degree of waning immunocompetence associated with advanced age in that multisystemic dissemination has been associated with immunosuppression in humans.18 A dog that underwent hemilaminectomy for decompression of the spinal cord also did not have a rapid resolution of clinical signs owing to the associated myelopathy. This dog had an improvement in its neurologic status following surgery, but remained nonambulatory 17 days after surgery, after which time it was lost to follow-up. Other authors have described dogs with CNS coccidioidomycosis treated with surgery that had persistent neurologic abnormalities but improved clinical signs at recheck intervals of 1 to 2 months.19
Follow-up radiography was performed inconsistently among dogs of this study. In most dogs that underwent radiography at 3 months, there were improvements in the radiographic appearance of lesions. However, the duration of treatment or overall likelihood for COM lesions to return to a normal radiographic appearance remains unknown, as complete radiographic resolution was not documented for any dog in this study. Bone serving as a persistent reservoir for organisms and potentially increasing the likelihood of recrudescence is an area for further longitudinal investigation.
The present study was limited by its retrospective nature, in particular the number of dogs lost to follow-up and the inconsistent recheck intervals and diagnostic testing. Most bone lesions were not confirmed as COM with cytology or histopathology, but the authors tried to exclude other likely etiologies by evaluating results of serologic testing, documenting clinical improvement with fluconazole treatment, and including only dogs with ≥ 90 days of follow-up. In the single dog with OSA that had a positive coccidioidomycosis serologic test result, deterioration of clinical status was evident 20 days after diagnosis, and OSA was confirmed on necropsy.
The present study documented the signalment, clinical signs, results of diagnostic imaging and serologic testing, and response to treatment of a cohort of dogs with COM, which have not been described in the current veterinary literature. Similarities in presentation and clinical and radiographic findings existed between dogs with COM and dogs with OSA. Thus, clinicians should use all available data to counsel owners and recommend appropriate diagnostic and therapeutic strategies for dogs suspected to have these conditions.
Acknowledgments
No third-party funding or support was received in connection with this study or the writing or publication of the manuscript. The authors declare that there were no conflicts of interest.
References
- 1. ↑
Ehrhart NP, Ryan SD, Fan TM. Tumors of the skeletal system. In: Withrow SJ, Vail DM, Page RL, eds. Withrow and MacEwen's Small Animal Clinical Oncology. 5th ed. WB Saunders; 2013:463–503.
- 2. ↑
Wrigley RH. Malignant versus nonmalignant bone disease. Vet Clin North Am Small Anim Pract. 2000;30(2):315–347, vi–vii.
- 3. ↑
Graupmann-Kuzma A, Valentine BA, Shubitz LF, Dial SM, Watrous B, Tornquist SJ. Coccidioidomycosis in dogs and cats: a review. J Am Anim Hosp Assoc. 2008;44(5):226–235.
- 5. ↑
Reed RE. Diagnosis of disseminated canine coccidioidomycosis. J Am Vet Med Assoc. 1956;128(4):196–201.
- 6.
Maddy KT. Disseminated coccidioidomycosis of the dog. J Am Vet Med Assoc. 1958;132(11):483–489.
- 7.
Brodey RS, Roszel JF, Rhodes WH, Bohn FK, Enck J. Disseminated coccidioidomycosis in a dog. J Am Vet Med Assoc. 1970;157(7):926–933.
- 9. ↑
Johnson LR, Herrgesell EJ, Davidson AP, Pappagianis D. Clinical, clinicopathologic, and radiographic findings in dogs with coccidioidomycosis: 24 cases (1995–2000). J Am Vet Med Assoc. 2003;222(4):461–466.
- 10. ↑
Davidson AP, Shubitz LF, Alcott CJ, Sykes JE. Selected clinical features of coccidioidomycosis in dogs. Med Mycol. 2019;57(suppl 1):S67–S75.
- 11. ↑
Shubitz LF. Comparative aspects of coccidioidomycosis in animals and humans. Ann N Y Acad Sci. 2007;1111:395–403.
- 12. ↑
Taljanovic MS, Adam RD. Musculoskeletal coccidioidomycosis. Semin Musculoskelet Radiol. 2011;15(5):511–526.
- 13. ↑
Blair JE. State-of-the-art treatment of coccidioidomycosis skeletal infections. Ann N Y Acad Sci. 2007;1111:422–433.
- 14. ↑
Jankowski MK, Steyn PF, Lana SE, et al. Nuclear scanning with 99mTc-HDP for the initial evaluation of osseous metastasis in canine osteosarcoma. Vet Comp Oncol. 2003;1(3):152–158.
- 15. ↑
Shubitz LE, Butkiewicz CD, Dial SM, Lindan CP. Incidence of coccidioides infection among dogs residing in a region in which the organism is endemic. J Am Vet Med Assoc. 2005;226(11):1846–1850.
- 16. ↑
Renschler J, Albers A, Sinclair-Mackling H, Wheat LJ. Comparison of compounded, generic, and innovator-formulated itraconazole in dogs and cats. J Am Anim Hosp Assoc. 2018;54(4):195–200.
- 17. ↑
Mawby DI, Whittemore JC, Genger S, Papich MG. Bioequivalence of orally administered generic, compounded, and innovator-formulated itraconazole in healthy dogs. J Vet Intern Med. 2014;28(1):72–77.
- 18. ↑
Adam RD, Elliott SP, Taljanovic MS. The spectrum and presentation of disseminated coccidioidomycosis. Am J Med. 2009;122((8):770–777.
- 19. ↑
Bentley RT, Heng HG, Thompson C, et al. Magnetic resonance imaging features and outcome for solitary central nervous system Coccidioides granulomas in 11 dogs and cats. Veterinary Radiol Ultrasound. 2015;56(5):520–530.
