Successful Treatment of Intracardiac and Intraocular Blastomycosis in a Dog with Combination Azole Therapy

Introduction

Cardiac infections are uncommon in the dog. Infectious myocarditis is usually the result of either a protozoal or viral infection (e.g., Trypanosoma, Leishmania, Toxoplasma, parvovirus, West Nile virus), while infectious endocarditis is usually the result of bacterial infection.^1–4 Fungal disease involving the heart is even less common than other cardiac infections. One fungal disease known to affect the heart is coccidiomycosis.⁵ It can result in life threatening effusive and constrictive pericarditis. Successful therapy for this form of coccidiomycosis has been described.⁵ Other cardiac fungal infections are exceedingly uncommon. Several dogs suffering from blastomycosis with cardiovascular involvement were described in one retrospective case series.⁶ All reported animals with either pericardial or intracardiac disease either died or were euthanized. The following report describes a case of disseminated blastomycosis in a dog with severe ocular and life-threatening cardiac involvement. Successful treatment of intracardiac blastomycosis without any residual cardiac disease detectable by electrocardiography or echocardiography is documented.

Case Report

A 4 yr old spayed female Labrador retriever weighing 35.8 kg was admitted to the Michigan State University Veterinary Teaching Hospital with a 1 wk history of draining cutaneous lesions, visual deficits, coughing, lethargy, and inappetence. The dog had access to a wooded rural area in north central Michigan, but also spent time in the southern region of the Upper Peninsula (Michigan’s major northern land mass surrounded by northeastern Wisconsin and Lakes Superior, Huron, and Michigan) where a soil excavation had recently occurred. There was no travel history outside of Michigan. Three days prior to presentation, the dog had been empirically treated with oral itraconazole^a (200 mg q 12 hr) and enrofloxacin^b (204 mg q 24 hr). Lack of response to empirical therapy and progressive visual deficits prompted referral.

On physical examination, the dog was depressed, mildly dehydrated, but in good body condition. Rectal temperature was elevated at 39.7°C. Thoracic auscultation did not reveal any cardiac murmur or arrhythmia; however, harsh, increased bronchovesicular sounds were noted diffusely. Approximately 20–30 nodular, ulcerative, draining cutaneous lesions were present on the ears, limbs, and interdigital areas, but were most numerous and severe along the trunk. Peripheral lymph nodes were moderately enlarged.

On ophthalmic examination, direct and consensual pupillary light reflexes were absent in the right eye (oculus dexter [OD]) and markedly reduced in the left eye (oculus sinister [OS]). Menace response was also absent OD and markedly reduced OS. The reflexes/responses OS progressed to absent over the following 12 hr. Tear production was normal and fluorescein staining for corneal ulceration was negative. Slit-lamp biomicroscopic examination revealed episcleral injection, diffuse corneal edema, keratitic precipitates, severe aqueous flare (4+ OD, 3+ OS based on a 0–4+ scale where 0 was normal), and posterior synechiae oculus uterque (OU). Intraocular pressures (IOPs) were 42 mm Hg and 22 mm Hg (reference range, 15–25 mm Hg) OD and OS, respectively, using applanation tonometry. Fundic examination via indirect ophthalmoscopy revealed bilateral vitreal haze and exudative retinal detachments (Figures 1A, B). Approximately 20% of the retina remained attached in the superior fundus OS; however, complete detachment was present in the right eye. The left optic nerve head was hyperemic and elevated. The optic nerve could not be visualized OD.

View Figure

• Download as Powerpoint
• Download Figure

The dog was hospitalized for additional diagnostics, IV fluid therapy, and supportive care. Dorzolamide hydrochloride 2%^c (1 drop OU q 8 hr) and timolol maleate 0.5%^d (1 drop OU q 12 hr) were initiated to decrease aqueous humor production and lower IOPs.^7,8 Prednisolone acetate 1%^e (1 drop OU q 4 hr) and flurbiprofen sodium 0.03%^f (1 drop OU q 6 hr) were administered topically to decrease intraocular inflammation and pain associated with the ophthalmic lesions.⁹ Tropicamide 1%^g (1 drop, q 12 hr) was used to decrease the risk of further posterior synechiae formation and to lessen pain associated with uveitis. IOPs were monitored 2–3 ×/day as there was potential for decreased aqueous drainage with tropicamide therapy in eyes at high risk for secondary glaucoma.⁹ The IOPs were 15 mm Hg and 10 mm Hg OD and OS, respectively, 24 hr after initiating therapy. Oral itraconazole was continued, but at a decreased dose (200 mg q 24 hr). Oral prednisolone therapy^h (0.5 mg/kg q 12 hr for 3 days, then q 24 hr) was also initiated to reduce posterior segment inflammation.

Additional diagnostics included a complete blood count, serum biochemical profile, thoracic radiographs, abdominal ultrasound exam, and cytologic examination of the skin lesions and lymph nodes. Hematologic evaluation revealed a moderate, normocytic normochromic nonregenerative anemia (hematocrit, 25.4%; reference range, 41–55%) and a marked segmented neutrophilia (39.32 × 10³/μL; reference range, 4–8.2 × 10³/μL) with a left shift (band count, 0.41 × 10³/μL; reference range, 0–0.1 × 10³/μL). The serum biochemical profile revealed hypoalbuminemia (2 g/dL; reference range, 2.8–4 g/dL), hypocalcemia (8.7 mg/dL; reference range, 9.4–10.9 mg/dL), and hypoglycemia (66 mg/dL; reference range, 80–120 mg/dL). Ionized calcium concentration from a venous blood gas analysis was normal.

Thoracic radiographs revealed a mild, diffuse interstitial pulmonary pattern. Abdominal ultrasound revealed a mildly enlarged left medial iliac lymph node. Cytologic examination of impression smears from the cutaneous lesions revealed marked neutrophilic and macrophagic inflammation with occasional bacterial cocci and basophilic yeasts consistent with Blastomyces spp. Peripheral lymph node cytology revealed marked neutrophilic and macrophagic lymphadenitis with intralesional Blastomyces spp. Histopathology of a skin lesion biopsy was characterized by marked inflammatory cell infiltrates arranged in discrete to coalescing nodules within the dermis and hypodermis. Those nodules were composed of degenerative neutrophils and epithelioid macrophages. The centers of those pyogranulomas contained variable numbers of fungal forms that were 10–30 μm in diameter and characterized by a thick, refractile, double-contoured wall, and occasional broad-based budding. The histopathologic diagnosis was moderate to marked pyogranulomatous dermatitis with intralesional fungal yeast consistent with Blastomyces dermatitidis (B. dermatitidis).

Based on those findings, a diagnosis of disseminated blastomycosis was established. In addition to the medications mentioned previously, IV ampicillin/sulbactamⁱ (30 mg/kg q 8 hr) was also initiated on the first day of hospitalization for secondary bacterial dermatitis. Although the dog was already on itraconazole, the recommended treatment of canine blastomycosis, dual therapy with oral fluconazole^j (17 mg/kg/day divided into two doses) was initiated as fluconazole may have better ocular penetration.^10,11 There are no studies documenting the success of dual azole therapy in veterinary medicine; however, considering the poor prognosis for restoring vision in eyes with posterior segment disease, the authors hypothesized dual therapy would result in a better clinical outcome by combining the efficacy of itraconazole with the theoretically better ocular pharmacokinetics of fluconazole. To monitor therapy, urine Blastomyces antigen was quantified^k in a sample obtained on the first day of hospitalization. The initial antigen level was 17.14 U (reference range, nondetectable) which was a high positive.

During the second day of hospitalization, the dog became restless and tachypneic. Cardiothoracic auscultation revealed tachycardia (heart rate, 250 beats/min) with pulse deficits and weak pulses. An electrocardiogram (EKG) revealed monomorphic ventricular tachycardia with a rate of 250–280 beats/min. Another venous blood gas analysis was performed and did not reveal any electrolyte or acid-base disturbances. IV lidocaine^l (2 mg/kg loading dose, then 50–60 µg/kg/min infusion with repeated 2 mg/kg boluses) was ineffective in converting the arrhythmia. High doses of IV procainamide^m (8 mg/kg loading dose, then a 50 µg/kg/min infusion with intermittent 5 mg/kg boluses) were used to convert the arrhythmia and maintain a stable rhythm and rate. Even after conversion to a predominantly sinus rhythm, a Mobitz type II second degree atrioventricular block, intermittent premature ventricular complexes, and short runs of ventricular tachycardia were present for several hours.

An echocardiogram revealed a 1 cm × 1 cm hyperechoic nodule in the left ventricle that was adhered to the papillary muscle. Additionally, a right atrial mural lesion and a small amount of pericardial effusion were present (Figures 2A, B). The small volume of effusion did not allow for pericardiocentesis. There was moderate mitral regurgitation with mild left atrial enlargement (left atrial/aortic ratio, 1.68; reference range, < 1.3).¹² The morphology of the mitral valve was normal. Systolic function was subjectively decreased with low normal fractional shortening (31%; reference range, 27–48%) on M-mode.¹² Ejection fraction was low normal at 56%. Left ventricular internal diameters at end systole (LVIDs) and diastole were 3.7 cm (reference range, 1.8–3.5 cm) and 5.1 cm (reference range, 3.7–5.1), respectively.¹² The mildly increased LVIDs further supported a decrease in systolic function. Although not severe, the mitral valve insufficiency, increased left atrial size, and the borderline myocardial performance prompted pharmacologic intervention to support systolic function and produce vasodilation. Enalaprilⁿ (0.25 mg/kg per os [PO] q 12 hr) and pimobendan^o (0.21 mg/kg PO q 12 hr) were initiated.

View Figure

• Download as Powerpoint
• Download Figure

The dog gradually improved over the next 7 days of hospitalization. During that period, the dog was transitioned from procainamide to oral sotalol^p (1.1 mg/kg PO q 12 hr) as the electrocardiographic abnormalities improved. Ophthalmic examination at that time revealed minimal changes OD compared with the initial retinal detachment and inflammation; however, the pressure was well controlled (15–18 mm Hg) with the antiglaucoma medications. Inflammation OS was improved (grade 2+ aqueous flare), and the retina had partially reattached. Retinal detachment was estimated at 50%, compared with 80% initially. Keratitic precipitates, posterior synechiae, optic nerve head hyperemia and elevation, subretinal infiltrates, and vitreal haze were improved, but present. Both eyes were still blind based on absent menace responses and negative vision testing (i.e., cotton ball, maze, and visual placing). An oral antibiotic^q (17.5 mg/kg q 12 hr) was used in place of IV antibiotics. Clinically, the dog was brighter, more active, and had an improved appetite. Hematologic and biochemical analyses revealed improvement in all previously mentioned abnormalities. After 10 days of hospitalization, the dog was discharged with oral itraconazole, fluconazole, sotalol, enalapril, pimobendan, and amoxicillin trihydrate/clavulanate potassium in addition to the topical ophthalmic medications mentioned previously.

Over the next few weeks, the dog improved clinically. Sotalol was gradually tapered starting 14 days after hospital discharge and eventually discontinued 28 days after discharge, at which time an EKG revealed a normal sinus rhythm. The urine antigen level 28 days after discharge was 15.02 U, which was a substantial decrease. Partial vision was regained OS, but OD was unchanged. An echocardiographic examination performed at the follow-up appointment 28 days after discharge revealed resolution of the ventricular nodule, right atrial mural lesion, and pericardial effusion. The LVID was 2.9 cm, which was supportive of normal systolic function. The left atrial/aortic ratio was also normal at 1.2. Mild mitral regurgitation was still present. Amoxicillin trihydrate/clavulanate potassium, pimobendan, and enalapril were discontinued on 28 days after discharge.

Six weeks following discharge, the dog returned to the Michigan State University Veterinary Teaching Hospital for enucleation OD due to the development of glaucoma that was refractory to topical therapy. A profound sinus bradycardia was appreciated on physical exam (heart rate, 40 beats/min). An EKG did not reveal evidence of heart block or other conduction abnormalities. In the absence of information gained from an atropine response test, the authors chose to place a temporary pacemaker to support cardiac rhythm prior to anesthesia. A standard transconjunctival enucleation was performed without complication. Histopathology revealed pyogranulomatous chorioretinitis with retinal detachment and yeast consistent with B. dermatitidis (Figure 3). The bradycardia resolved after enucleation and was not documented at any subsequent follow-up appointments.

View Figure

• Download as Powerpoint
• Download Figure

After 11 wk of combination azole therapy, itraconazole was discontinued. The urine antigen level was 2.55 U. The decision to discontinue itraconazole was based on the potential increased risk of side effects associated with using both drugs, the cost of using two drugs, and the better ocular pharmacokinetics of fluconazole. After the first 2 mo of fluconazole monotherapy, the dog’s urine antigen level was 0.53 U. The continued decrease in antigen level supported the use of fluconazole as the sole antifungal agent.

Overall, systemic prednisolone and topical prednisolone acetate were used for a total of 14 wk. At a follow-up appointment performed at that time, ophthalmic examination did not reveal evidence of inflammation or active disease OS. Other than persistent, mild, mitral regurgitation, echocardiographic findings were normal and unchanged from previous examination. Treatment with oral fluconazole was continued based on urine antigen levels. At the time of cessation of therapy 7 mo after initial evaluation, the urine antigen level was nondetectable. At the time this manuscript was prepared, the dog was normal according to the owner, vision was maintained OS, and there was no evidence of cardiovascular disease.

Discussion

Blastomycosis results from infection with the thermally dimorphic fungus, B. dermatitidis, which is considered endemic in the Ohio, Mississippi, and Missouri River Valleys, the Great Lakes region, and other parts of the central and southeastern United States.¹³ The dog in this report is of a typical signalment and environment for dogs infected with Blastomyces spp. The close proximity to water and exposure to a recent soil excavation are both risk factors for the disease. The dog in this report also had very typical multisystem involvement of the lungs, lymphatics, eyes, and skin, with many expected clinical and laboratory findings associated with disseminated blastomycosis; however, cardiac involvement as observed in this dog is extremely rare.^13–15 In one retrospective report of eight cardiovascular blastomycosis cases spanning over 20 yr, six dogs had cardiac infections and two dogs had extracardiac granulomas compressing the left atrium.⁶ The majority of dogs in the previous case series presented for signs related to other organ systems and most had evidence of disseminated disease. Similar to the current report, many of the dogs had rhythm and conduction disturbances, and echocardiographic lesions were seen in all animals in which the procedure was performed. None of the six dogs with myocardial, pericardial, or endocardial blastomycosis survived, which is in contrast to the present report in which the dog survived with complete resolution of cardiac disease.⁶

One limitation in the current report is lack of cytologic confirmation that the cardiac lesions were due to Blastomyces organisms. Blastomyces was documented histopathologically as a cause of pericarditis in several of the dogs from the previously mentioned case series,⁶ but evaluation of pericardial fluid was not reported. Pericardial fluid has been useful in diagnosing coccidiomycosis-induced pericarditis, but it is unknown if this would be true for blastomycosis.¹⁶ Pericardiocentesis was not performed in the present case due to the relatively small amount of effusion and potential harm it might have caused. Cardiac troponin I levels are elevated in cases of myocardial inflammation and could have value in supporting a diagnosis of infectious myocarditis, but this was not evaluated as the authors did not feel it would provide any new/discriminatory information. Given the location of lesions, confirmed blastomycosis diagnosis, and response to therapy, other infectious diseases were considered highly unlikely and not pursued. Other differentials such as neoplasia and thrombus were also considered unlikely given the entire case scenario.

The sinus bradycardia that resolved after enucleation of the glaucomatous eye was another interesting finding. The bradycardia was present weeks after sotalol discontinuation and was more profound than what is expected with topical ophthalmic glaucoma medications.¹⁷ One possible explanation was glaucoma and endophthalmitis causing chronic vagal stimulation. There are reports documenting chronic oculocardiac reflex secondary to both maxillary fractures and intraorbital foreign bodies.^18,19 Autonomic dysfunction is also recognized in humans with glaucoma.²⁰ An atropine response test may have provided more information by determining if the bradycardia was a result of increased vagal tone. That test was not performed because of the possibility of decreased aqueous drainage OS leading to a pressure increase and glaucoma.

This report is also unique in documenting the use of dual azole therapy for blastomycosis. The recommended treatment of nonlife-threatening blastomycosis is itraconazole.^10,13,15 It is as effective and has fewer side effects compared with combination amphotericin B and ketoconazole, and is more effective than ketoconazole alone.^10,21 In vitro and human studies also support itraconazole as being more effective than other azole options, including fluconazole.^22–24 In humans, however, high-dose fluconazole has efficacy comparable to itraconazole.²⁵ Fluconazole is also effective in treating canine blastomycosis, but there are no prospective veterinary studies evaluating its effectiveness compared with itraconazole. A recent retrospective study comparing itraconazole and fluconazole in canine blastomycosis did raise some concerns regarding fluconazole efficacy as noninferiority testing could not prove the drugs equivalent, and the duration of treatment was significantly longer for dogs receiving fluconazole.²⁶ The water solubility, low protein binding, and other pharmacokinetics of fluconazole make it a desirable choice in specific infection sites because it more readily penetrates the blood-ocular, blood-brain, and blood-prostatic barriers than itraconazole.¹¹ The decision to use both azoles was based on the efficacy of itraconazole for disseminated disease and the favorable pharmacokinetics of fluconazole for treatment of ocular disease. Multimodal antifungal therapy using different drug classes is known to have synergistic effects in treating various mycoses.^21,27 It is unknown whether combination azole therapy has either any inhibitory/additive effects or whether this combination increases the risk of adverse side effects. Regardless, both clinical and laboratory improvement were noted in this dog during treatment. The authors of this report, however, emphasize that conclusions regarding the efficacy of combination azole therapy cannot be determined from a single case report. Further studies investigating the safety and efficacy of combination azole therapy in cases of disseminated disease with ocular involvement are needed.

The use of systemic corticosteroids during treatment remains controversial because of potential immunosuppression.^10,13 Most dogs with ocular involvement have posterior segment disease, and topical ophthalmic steroid preparations penetrate poorly into that region.^28,29 Dogs with endophthalmitis and/or glaucoma also have a poor prognosis for vision compared with dogs with less severe ocular involvement.²⁹ In one report, no adverse effects were seen in association with systemic steroids in conjunction with antifungals, and their administration may have aided in preserving and restoring vision.³⁰ Oral corticosteroids were prescribed because vision was unlikely to return without systemic intervention. Due to the presence of infectious disease, systemic nonsteroidal anti-inflammatory drugs were considered; however, steroids are likely to be more effective at reducing posterior segment inflammation, and the dose used was anti-inflammatory as opposed to immunosuppressive.^9,31

The prognosis for full recovery from blastomycosis is good, with an estimated 70–75% survival rate with appropriate treatment; however, involvement of three or more organ systems warrants a poor prognosis.^10,15 Central nervous system and/or cardiac involvement warrants a grave prognosis.^6,10,13 The dog in this report survived despite multiple organ system involvement and intracardiac lesions.

Conclusion

The overall incidence of cardiac infections secondary to Blastomyces is unknown. Although seldom reported, it is possible that cardiac lesions have been overlooked in some cases, especially where either access to advanced diagnostics, such as echocardiography, may not be readily available or where necropsies are not routinely performed. Blastomycosis should be considered a differential for any dog living in an endemic area that has either a cardiac arrhythmia or evidence of myocarditis in which another cause is not readily apparent. In addition, dogs with diagnosed Blastomyces infections warrant complete cardiac evaluation, including both an EKG and echocardiogram, if any evidence of an arrhythmia, murmur, or cardiac disease is present. There are no previously published reports of survival in dogs with intracardiac lesions, and the prognosis remains grave; however, this case highlights that successful treatment is possible with comprehensive cardiac care and prolonged antifungal therapy. In addition to documenting survival from a Blastomyces cardiac infection, this report is also unique in the use of combination azole therapy to combine optimal effectiveness and ocular drug penetration.

The authors would like to thank Drs. Mathew Annear, Ingeborg Langohr, Kenneth Pierce, and Freya Mowat, as well as Judy Eastman, LVT, and Donna Letavish, LVT, for their contributions.