Cutaneous and visceral phaeohyphomycosis in an immunocompetent golden retriever

Abstract

A 3-year-old castrated male golden retriever was presented for evaluation of 2 cutaneous masses along the abdominal midline and a 6-month history of progressive lethargy and inappetence. Two years earlier, the dog underwent a gastrotomy to retrieve a foreign object and recovered uneventfully. Fluid aspirated from the lesions was culture-negative for aerobic and anaerobic bacterial growth. Abdominal ultrasound demonstrated a large intra-abdominal mass with apparent communication with the cutaneous lesion, along with gossypiboma from the previous laparotomy. Neoplasia or a sterile abscess were suspected. Exploratory laparotomy was performed and revealed that the intra-abdominal mass was adhered to the abdominal midline and the greater curvature of the stomach. The masses and affected portions of the body wall and stomach were resected and histopathology was consistent with phaeohyphomycosis. Antifungal therapy with voriconazole (6.3 mg/kg, PO, q12h) was initiated. At 1 mo after surgery, all clinical signs had resolved. At 4 and 7 mo after surgery, the dog continued to thrive despite 2 small masses, seen on abdominal ultrasound imaging on the intra-abdominal midline, suggestive of reoccurrence. Continued voriconazole therapy was administered in lieu of further surgical excision. One year after surgery, the masses were no longer present on ultrasonographic evaluation. Phaeohyphomycosis is a rare, opportunistic fungal infection that typically affects the dermis and subcutis of immunocompromised dogs. This is the first report of phaeohyphomycosis in an immunocompetent dog and involving the dermis, subcutis, and abdominal viscera.

Key clinical message:

This case adds to the very limited literature on phaeohyphomycosis in dogs and illustrates that surgery could be a risk factor for infection, even in dogs with no known underlying disease or immunodeficiency.

CASE DESCRIPTION

A 3-year-old castrated male golden retriever was presented for evaluation of 2 cutaneous masses on his abdominal midline. These masses were located along a scar from a previous (2 y earlier) gastrotomy for retrieval of gastric foreign material. The owners reported that the masses had appeared on the dog’s ventral abdomen about 6 mo before presentation and continued to grow. Since their appearance, the dog began showing signs of progressive lethargy and inappetence. Opaque fluid was aspirated from the masses and was submitted for cytology, which revealed suppurative inflammation. Aerobic and anaerobic bacterial cultures of the fluid were negative, and the dog was referred to the VCA Animal Specialty and Emergency Center for further evaluation.

Upon presentation, the dog was febrile with a temperature of 40.2°C (104.3°F) but was alert and responsive. There were 2 subcutaneous mass-like lesions along the abdominal midline scar firmly affixed to the underlying tissues. The more cranially located mass measured 4.39 cm in greatest diameter, whereas the more caudally located mass measured 4.37 cm in greatest diameter. No swelling or erythema were noted, and the dog was not notably in pain on palpation of the masses. An abdominal ultrasound (AUS) was performed and demonstrated that the lesions were avascular, with a large, intra-abdominal component, measuring 7.44 cm in greatest diameter (Figure 1). Differential diagnoses for the lesions included gossypiboma from the previous gastrotomy, sterile abscess, or necrotic/avascular neoplasia. Exploratory laparotomy was pursued.

FIGURE 1.

Abdominal ultrasound images. A — The oval-shaped, hypoechoic mass in the subcutis measures 2.78 × 1.06 cm (indicated by the 4 “+” symbols) and is surrounded by hyperechoic fat. This lesion showed no vascularity when imaged with Doppler. B — The subcutaneous mass communicates through a 1.26-centimeter defect in the ventral abdominal wall, marked by the “+” symbols. C — The heterogeneously hypoechoic intra-abdominal mass, marked by the “+” symbols, measures 5.92 × 7.94 cm and is partially beam-attenuating. No vascularity was identified with Doppler investigation.

A complete blood (cell) count and biochemistry panel were obtained before surgery. The complete blood (cell) count revealed a marked leukocytosis (44.41 × 10⁹/L; reference range: 6 to 17 × 10⁹/L) characterized by neutrophilia (37.56 × 10⁹/L; reference range: 3 to 12 × 10⁹/L) and monocytosis (3.81 × 10⁹/L; reference range: 0.2 to 1.5 × 10⁹/L). The biochemistry panel revealed hypoalbuminemia (1.7 g/dL; reference range: 2.5 to 4.4 g/dL), hyperglobulinemia (6.4 g/dL; reference range: 2.5 to 5.2 g/dL), and an increased ALP level (464 U/L; reference range: 20 to 150 U/L). The dog was premedicated with methadone (0.2 mg/kg, IV) and maropitant (1 mg/kg, IV). The dog was induced under general anesthesia, with midazolam (0.25 mg/kg, IV) and propofol (4 mg/kg, IV), to effect, and was maintained with sevoflurane. The dog was placed in dorsal recumbency and the ventral abdomen was clipped and prepared aseptically for surgery. A midline incision was made along the previous scar and around the 2 masses. Hemostasis was achieved using monopolar electrocautery. When attempting to dissect the masses free from the midline, the masses were penetrated and a large amount of opaque fluid with fine, black stippling was retrieved. The masses could not be freed from the midline. Therefore, a midline laparotomy was performed cranial and caudal to the masses, and the body wall was resected in an elliptical pattern around the midline masses, which allowed for surgical access into the abdomen. Upon exploratory laparotomy, it was seen that the external masses and resected body wall were connected to a large, firm, intra-abdominal mass, which was firmly adhered to the greater curvature of the antrum of the stomach at the level of the angle of incisure (Figure 2). Nearly the entire greater omentum was adhered to the mass. These adhesions were transected using a LigaSure (Covidien, Mansfield, Massachusetts, USA), resulting in near-complete removal of the greater omentum. The spleen, left limb of the pancreas, and their vasculature were spared. To remove the gastric portion of the mass, a partial gastrectomy was completed using a TA 90 stapler (Covidien), with care taken to spare as much of the antrum as possible to allow normal passage of gastric contents. The staple line was oversewn in a continuous pattern using 3-0 polydioxanone suture (PDS II; Ethicon, Somerville, New Jersey, USA). Gloves and instruments were changed and the abdomen was thoroughly lavaged using warm, sterile saline (B. Braum Medical, Irvine, California, USA). The remainder of the abdomen was explored and was normal in appearance. A Jackson-Pratt drain was placed exiting the left caudal abdomen and was secured, using a purse string and Chinese finger trap, with 3-0 nylon suture (Ethilon; Ethicon). The abdomen was closed in routine fashion. Bupivacaine liposome injectable suspension (Nocita; Elanco Animal Health, Greenfield, Indiana, USA) was instilled into each layer throughout closure.

FIGURE 2.

The abdominal mass (marked by the asterisk) is tightly adhered to the greater curvature of the antrum of the stomach (marked by the “X” symbol) at the level of the angle of incisure. The line between the stomach and mass is indicated by arrows.

The dog was discharged 2 d after surgery and prescribed gabapentin (9 mg/kg, PO, q8h), Clavamox (Zoetis; 16 mg/kg, PO, q12h), maropitant (2 mg/kg, PO, q24h), and omeprazole (2 mg/kg, PO, q24h). The Jackson-Pratt drain was removed 7 d post-surgery. Histopathologic analysis of the abdominal mass revealed large mats of fungal hyphae. The slides were stained with hematoxylin and eosin stain, which revealed that the hyphae were septate, parallel-walled, 4 to 6 μm wide, and diffusely brown-pigmented with occasional terminal bulbous projections. These findings were consistent with phaeohyphomycosis (Figure 3). Antifungal therapy with voriconazole (200 mg, PO, q12h) was initiated 14 d after surgery. Intermittent regurgitation was reported for the first week post-surgery, but steadily improved without further intervention. At 1 mo after surgery, the dog’s energy level was gradually increasing, appetite returned to normal, and vomiting or regurgitation were noted to occur approximately once per week. The ultrasound study was unremarkable. At 4 mo after surgery, the dog continued to thrive; however, 2 new, small, intra-abdominal midline masses were seen on AUS. The largest measured 2.77 × 1.13 cm. A cytological examination was completed on aspirates of the masses and findings were consistent with pyogranulomatous inflammation. Given the location of the masses and the dog’s history of phaeohyphomycosis, the cytology was strongly suggestive of recurrence of fungal infection, although no organisms were seen. The owners elected to continue voriconazole therapy in lieu of further surgical excision, and a recheck ultrasound was recommended in 3 mo. At 7 mo after surgery, the dog continued to thrive. The 2 intra-abdominal masses as seen on AUS were smaller than previously recorded, at 0.6 × 1.6 cm and 0.7 × 1.1 cm. Continued voriconazole therapy was pursued, with a plan to monitor via AUS every 6 mo. The owners inadvertently discontinued the voriconazole 4 mo later (11 months after surgery). At 13 mo after surgery, the masses were no longer present on abdominal midline as viewed by AUS and the dog remained free of clinical signs.

FIGURE 3.

Histopathologic images of the abdominal mass stained with hematoxylin and eosin. A — Included are the mucosa (Mc), submucosa (Sm), muscularis (Ms), and serosa (Se) of the stomach. The black box indicates an area of fungal mats. Scale bar: 5 mm. B — Higher-magnification image. Numerous fungal colonies (blue star) are present, surrounded by inflammatory infiltrates (orange star) and fibrosis (black star). Scale bar: 200 μm. C — Parallel-walled hyphae with septation (black arrow) and terminal bulbous projections (orange arrow). Scale bar: 50 μm.

DISCUSSION

This report describes the presentation, diagnosis, treatment, and short-term outcome of an immunocompetent dog with phaeohyphomycosis extending between the abdominal viscera and subcutis. Phaeohyphomycosis is a rare, opportunistic infection by pigmented, saprophytic fungi that are detected worldwide in soil and vegetative matter and typically invade animals via a traumatic injury or contaminated wound (1). They are usually nonpathogenic in immunocompetent animals; however, life-threatening illness has been documented in debilitated or immunocompromised patients, such as those receiving glucocorticoids or cyclosporine (2,3). Although infections in immunocompetent patients have been reported rarely in recent human and veterinary literature, rates of infection in immunocompromised dogs, particularly those requiring cyclosporine, have been on the rise, with incidences as high as 6.5%. The increases in case numbers in both human and veterinary medicine appear to correspond with the increasing use of more aggressive immunosuppressive medications (4). Diagnosis is typically made on histopathology or wet mount where the pigmented organism can be identified. Typically, as in this case, the fungi appear light yellow to dark brown in color (1). Within tissue samples, the fungi are seen as irregularly bulbous or toruloid hyphae with yeast-like projections. Also typically with phaeohyphomycosis, no muriform cells or grains are seen. Phaeohyphomycosis is distinguished from chromoblastomycosis and mycetoma by the absence of specific histopathologic findings such as sclerotic bodies or grains in the tissue (5). Most species involved in phaeohyphomycosis cannot be differentiated based on histopathologic appearance, so a culture is required to specify the inciting organism. At the time of surgery for this case, phaeohyphomycosis was not suspected, and therefore a fungal culture was not done.

Cutaneous and subcutaneous lesions are the most common clinical manifestation of this condition and usually present as cysts or papules. Alternaria spp. are the most common organisms isolated, though Phialophora and Exophiala spp. can cause similar lesions. Phaeohyphomycosis has also been reported to cause pulmonary and cerebral infections, fungal keratitis, and, on rare occasions, disseminated infections (6). Visceral involvement, as reported here, has not previously been documented in a dog. It is not known in this case whether the fungus gained entry through a penetrating wound into the abdomen or whether the fungus originated within the GI tract, though the former is suspected. Because the dog had a history of a previous abdominal surgery with no other known sources of inoculation or trauma, and given the extension of the fungal mass between the stomach and midline incision, the authors suspect that the surgery was the source of inoculation. The foreign body removed during the dog’s previous gastrotomy consisted of rocks, cloth, and various plant materials, including grass. Because these fungal organisms are widespread in the environment and can be found in soil, wood, and decomposing plant debris, it is possible that the dog’s gastric foreign material was the source of the infection. However, this suspicion cannot be proven.

In the human medical literature, surgery has been reported to be curative for patients with lesions that have not already disseminated and when appropriate margins can be resected (7). Because recommended margins have not been established in human medicine and local recurrence has been documented, wide excision is recommended (7). Systemic antifungal agents are often used as adjunctive therapy in patients with known dissemination or in those who are immunocompromised, for whom dissemination poses a higher risk (8). Antifungal agents that have been used to treat phaeohyphomycosis include amphotericin B, posaconazole, voriconazole, and itraconazole, among others (1) Because of the infrequent occurrence of this disease, clinical trials to establish a standard of care have not been feasible. Therefore, it is unknown whether one antifungal agent is associated with a greater chance of survival than others (1). Dissemination of the pathogen, regardless of therapy, carries a mortality rate of >70% (9). Focal or generalized cutaneous lesions may carry a better prognosis in immunocompetent patients if the lesions can be resected and if patients are subsequently treated with long-term antifungal therapy (1). In this case, although complete resection of the fungal mass was achieved and there was no evidence of disseminated disease, systemic antifungal therapy was instituted out of an abundance of caution.

In conclusion, our report documents a rare case of phaeohyphomycosis with intracavitary extension. This case illustrates that, although most cases of phaeohyphomycosis occur in immunocompromised animals, those with no known underlying disease are still at risk of infection. This is the first report of cutaneous phaeohyphomycosis with visceral extension in an immunocompetent dog and adds to the limited literature that exists on the treatment and outcomes for these animals. Surgery could be an indicated risk factor in the development of phaeohyphomycosis even in a healthy animal, though more research is needed to confirm this suspicion. Wide excision with subsequent oral antifungal therapy is a possible curative option for the management of this presentation of phaeohyphomycosis. CVJ