Introduction

Opportunistic mycoses caused by soil saprophytes that are only rarely pathogenic to immunocompetent hosts have recently emerged as important causes of disease in immunosuppressed dogs and cats. Unlike the endemic mycoses such as histoplasmosis, blastomycosis, and sporotrichosis, the morphologic features of most opportunistic fungi are not sufficiently unique to allow genus- or species-level classification based on morphology alone, necessitating culture or molecular testing for definitive identification. As a result, many practitioners first encounter these fungi when histologic evaluation of a skin biopsy or cytologic evaluation of a lymph node aspirate yields a diagnosis of fungal infection without further identification or classification. Fortunately, with a complete histologic or cytologic description, these organisms can often be placed into clinically useful categories based on morphologic features such as pigmentation, hyphal diameter, frequency of septation, and distribution in tissue. These categories include phaeohyphomycosis (pigmented hyphal or yeast forms), hyalohyphomycosis (nonpigmented hyphal forms), and eumycotic mycetoma (characterized by the formation of fungal colonies in tissue that can be visualized grossly as black or white tissue grains).

Traditionally, infections caused by opportunistic mycoses have been rare, most often manifested as cutaneous or nasal phaeohyphomycosis in cats or as disseminated hyalohyphomycosis in young adult dogs. Over the past 20 yr, however, the incidence of opportunistic fungal infection has increased substantially in dogs receiving immunosuppressive therapy, with a recent study showing an incidence of 6.5% in dogs being treated for immune-mediated disease.¹ This increased incidence is especially true for protocols that include cyclosporine, which was approved for use in dogs in 2003.^1,2 A search of the small animal literature for cases of systemic mycoses in animals receiving immunosuppressive medication showed that the vast majority of cases have been described since 2006.^3–20 This change mimics a similar epidemiologic pattern that arose in human patients over the past 2 decades following the development of potent new immunosuppressive drugs and aggressive immunosuppression protocols associated with solid organ and bone marrow transplantation.²¹ Anecdotally, the risk for development of opportunistic mycoses appears to increase when multiagent protocols are used. This has been documented in human patients with inflammatory bowel disease, with one investigation finding that the addition of each immunosuppressive medication increased the patients’ overall risk of infection.²² In the authors’ experiences, animals who are at highest risk for developing opportunistic fungal infection are those on two immunosuppressant drugs if one of those is cyclosporine, or dogs who are on any three immunosuppressant medications.

In addition to drug therapy, naturally occurring causes of immunocompromise may also predispose to opportunistic mycoses. Examples include Candida spp. urinary tract infection in diabetic dogs and cats; disseminated aspergillosis and hyalohyphomycosis in German shepherd dogs with suspected familial multifactorial immunodeficiency; Pneumocystis pneumonia in miniature dachshunds with combined variable immunodeficiency; systemic mold infection in young dogs with hereditary cobalamin deficiency (Imerslund-Grasbeck syndrome); and phaeohyphomycosis in cats with diabetes mellitus, retrovirus infection, or lymphoid neoplasia.^23–30 In general, opportunistic mycoses associated with naturally occurring causes of immunocompromise are rare in dogs and cats, but their clinical presentations have been fairly well described, and several have morphologic characteristics that allow them to be identified to genus level without culture. In contrast, opportunistic fungal infections that occur in animals on immunosuppressive therapy are currently more likely to be encountered by the small animal practitioner and are also more likely to be caused by saprophytic organisms with which practitioners and pathologists are unfamiliar. Therefore, the purpose of this review is to facilitate identification and management of infection caused by fungal saprophytes that are not typical pathogens, with a focus on those that are most often associated with immunosuppressive drug therapy.

Mechanisms of Immunosuppression

For many years, single-agent glucocorticoid therapy was the mainstay of treatment for immune-mediated disease in dogs and cats. Glucocorticoids alter the host immune response nonselectively through numerous mechanisms, including inhibition of cytokine production and release, impaired expression and function of Fc receptors on macrophages, decreased phagocytosis, decreased macrophage antigen processing and presentation, and reduction of lymphocyte numbers, among others.^31,32 Although these effects ultimately impact all arms of the immune response, the cell-mediated Th1 response (critical for defense against fungal organisms) is particularly inhibited. Unfortunately, glucocorticoids alone are often not sufficient to resolve more severe manifestations of immune-mediated disease, and the deleterious effects associated with their long-term use in dogs often prompt practitioners to add other immunosuppressants so that glucocorticoids can be tapered more quickly.³¹ Of these additional immunosuppressants, drugs most often used 20 yr ago included alkylating agents (e.g., cyclophosphamide, chlorambucil), folic acid antagonists (e.g., methotrexate), and nucleotide analogues (e.g., azathioprine). Although azathioprine and chlorambucil continue to be commonly used in dogs and cats, respectively, many of the other older drugs have been replaced with newer, more selective agents (e.g., cyclosporine, mycophenolate mofetil, and leflunomide), which may have greater efficacy or fewer side effects than their older counterparts.³¹ Mycophenolate and leflunomide inhibit proliferation and clonal expansion of both B and T cells by interfering with purine and pyrimidine synthesis, respectively.³¹

Cyclosporine, the first immunosuppressant drug licensed by the FDA for use in dogs, is currently considered by many practitioners to be the first choice of an immunosuppressant to be added to glucocorticoid therapy for the treatment of immune-mediated disease. Based on published case reports, the authors’ own observations, and a recent large retrospective study of dogs treated for immune-mediated disease, administration of cyclosporine appears to be a risk factor for the development of opportunistic fungal infections in dogs.^{1,4,6,9–13,16,33} This may be partially because of the fact that it targets T helper cell functions that are critical for host defense against fungal infection. Cyclosporine enters the T cell and forms a physical complex with cyclophilin that inhibits calcineurin, an intracellular phosphatase without which the cell cannot activate the gene transcription factor nuclear factor of activated T cells, leading to decreased production of IL-2 and IFN-γ.² This results in decreased activation and proliferation of T cells and, in particular, a blunted Th1 response. In dogs, the degree of immunosuppression associated with cyclosporine administration has been shown to be dose-dependent.³⁴

Diagnosis of Opportunistic Fungal Infections

Unlike the more common endemic mycoses such as cryptococcosis, histoplasmosis, and blastomycosis, specific serologic assays are generally not available for the diagnosis of opportunistic fungal infections, making culture or molecular identification necessary for definitive diagnosis. Unfortunately, fungal culture is often not performed at the time of initial biopsy because mycotic infection is not suspected until results of histopathology are available. From a practical standpoint, categorization of opportunistic fungal infections based on morphologic features in histologic or cytologic samples is often adequate for predicting prognosis and determining general treatment options; therefore, rebiopsy to obtain tissue for fungal culture is often not performed, especially when clients cannot afford the newer triazoles that may be indicated for treatment of some specific pathogens. This approach is perhaps most detrimental for immunocompetent dogs with hyalohyphomycosis, in which several of the potential pathogens are inherently resistant to amphotericin B and itraconazole, making genus- and species-level identification important.³⁵

When culture is performed, fresh tissue (rather than swabs of exudate) should be submitted as soon as possible after collection to a laboratory with expertise in fungal culture. These authors prefer to submit to a laboratory in which molecular identification of cultured isolates is routinely performed when morphologic features alone are insufficient for definitive identification¹. Best results are typically achieved by contacting the laboratory for shipping instructions prior to sending a sample. An alternative to culture that has become more routinely available in recent years is extraction, amplification, and sequencing of fungal DNA from paraffin-embedded tissues, which is appealing because it does not require rebiopsy. Reported sensitivities for these assays have ranged from 54 to 94%.^36–38 Although several commercial laboratories now offer molecular diagnostics based on panfungal amplification and sequencing, these authors prefer to use a laboratory for which assay sensitivity and specificity have been described^b.³⁶

Because the organisms that cause opportunistic mycoses are common contaminants, it is important for the clinician to recognize that isolation of a saprophytic fungus from a nonsterile site (such as skin or nasal cavity) should not be considered evidence of infection unless there is also cytologic or histologic evidence of pyogranulomatous inflammation accompanied by tissue invasion by a morphologically compatible organism. Likewise, results of polymerase chain reaction amplification and sequencing from DNA extracted from paraffin sections should always be interpreted in the context of clinical and histologic or cytologic findings, as the relatively high sensitivity of this technique may produce false-positive results due to amplification of contaminant DNA on paraffin blocks or commensal fungal DNA in tissues.³⁶

Phaeohyphomycosis

Phaeohyphomycosis refers to infection caused by saprophytic fungi that produce pigmented (or dematiaceous) yeast or hyphae in tissue. Infections caused by the genera Alternaria, Aureobasidium, Bipolaris, Cladophialophora (previously Xylohypha or Cladosporium), Curvularia, Exophiala, Fonsecaea, Lecytophora, Microsphaeropsis, Moniliella, Mycoleptodiscus, Phialophora, Ramichloridium, Scolecobasidium, Scytalidium, and Ulocladium have been reported in veterinary species.^{12,14,29,38–52} These infections generally arise from cutaneous inoculation, with sites of infection including skin, subcutaneous tissue, and central nervous system, as well as disseminated disease.

The traditional clinical presentation associated with phaeohyphomycosis is characterized by the development of cutaneous lesions in areas that may contact soil, such as the nose, ears, and digits, in immunocompetent cats (Figure 1A).^{44,48,50,53–55} Infection occurs by direct inoculation, which leads to chronic granulomatous inflammation. Affected areas may be grossly pigmented and thus can be mistaken for melanoma. Lesions in these patients tend to be locally invasive but do not typically disseminate beyond regional lymph nodes. Systemic dissemination, as reported in two dogs with mycotic meningoencephalitis and nephritis and in a cat with multiorgan involvement, is rare in the absence of immunosuppression and carries a poor prognosis.^36,46,56 Historically, animals with central nervous system infection had a poor to grave prognosis, but recent evidence suggests that the combination of surgery with medical therapy may lead to better outcomes.⁴⁰

View Figure

• Download as Powerpoint
• Download Figure

A diagnosis of phaeohyphomycosis can be made cytologically or histologically if pigmented fungal organisms are visualized. It should be noted that pigment may not be visible cytologically (or may appear blue-green rather than brown on Wright-Giemsa stained preparations) but can often be documented histologically as brown coloration in the hyphal or yeast wall (Figure 1B). When the pathogen is not heavily pigmented, it may be necessary to examine an unstained slide or to use a Fontana-Masson stain to better visualize the melanin pigment.

In immunocompromised patients, the most common presentation for phaeohyphomycosis appears to be that of focal or multifocal ulcerative or nodular cutaneous lesions in dogs receiving multiple immunosuppressant medications (Figure 2).^11,12,17,57 Dissemination of disease to involve multiple cutaneous lesions on distant parts of the body, multiple lymph nodes, or even other organ systems seems to occur more often in immunocompromised than immunocompetent patients.^9,14,15

View Figure

• Download as Powerpoint
• Download Figure

Medical treatment of phaeohyphomycosis is often difficult, in part because melanin is a potent virulence factor; therefore, surgical resection with wide margins is the treatment of choice for focal lesions that are resectable.^58,59 For immunocompetent cats with digit or pinna involvement, full digit resection (even if the lesion involves only the distal phalanx) or partial pinnectomy should be performed. If a full digit or more is involved, limb amputation should be considered. Medical therapy may be prescribed for several months following lesion resection, especially if there is concern about the extent of the surgical margin.⁵² When surgical resection is not possible, medical therapy will often result in initial clinical improvement, but recurrence of lesions is very common.⁵² Therefore, these authors recommend treatment for 9–12 mo or more in immunocompetent animals with nonresectable phaeohyphomycosis, and even then, some cases either fail to respond or develop lesion recurrence after antifungal medication is discontinued.

Data regarding in vitro susceptibilities for human and veterinary isolates of pigmented fungi is sparse, but based on this limited information, itraconazole, voriconazole, posaconazole, and amphotericin B have the most consistent activity against pigmented fungi, whereas reported activities of fluconazole and ketoconazole have been poor.⁶⁰ Itraconazole (Table 1) is the drug most commonly used to treat phaeohyphomycosis in veterinary patients because it has reasonable efficacy without the cost of the newer triazoles, and because many veterinarians are familiar with its use. Posaconazole, although more expensive, is a triazole with a broader spectrum that may have better efficacy than itraconazole, and is typically easier to administer in cats because of its availability as a well-tolerated oral suspension.^29,61 Voriconazole, also very expensive, is a triazole that reaches high concentrations in the central nervous system, making it the drug of choice for treating intracranial phaeohyphomycosis in dogs.⁴⁰ Although use of voriconazole has been mostly limited to dogs because neurotoxicity has been observed in treated cats, the use of a lower dose in conjunction with therapeutic drug monitoring may allow use of this medication in cats in the future.^62–64

TABLE 1 Drugs Most Often Used for the Treatment of Opportunistic Mycoses

View Fullscreen

In patients receiving immunosuppressive medications, cutaneous phaeohyphomycosis often responds more readily to medical therapy than does comparable disease that develops in an immunocompetent patient, but only if cyclosporine can be discontinued and other immunosuppressant medications can be rapidly tapered.¹⁶ The authors have observed two cases in which lesions caused by cutaneous phaeohyphomycosis resolved even without antifungal therapy after immunosuppressant medications were discontinued. However, dissemination of disease despite therapy has also been observed, so the authors’ recommendation is always to discontinue immunosuppressant medications as quickly as possible and also treat medically with itraconazole, posaconazole, or voriconazole for at least 2 mo beyond lesion resolution. If lesions are focal or otherwise amenable to wide surgical excision, a combination of medical and surgical therapy may be effective.¹⁶

Hyalohyphomycosis

Hyalohyphomycosis refers to infection caused by fungi that form hyphal elements with nonpigmented (or hyaline) walls in tissue. These hyphae are septate, typically branch at acute angles, and may contain areas of globose swelling, especially at the end of the hyphae or between septations.⁶⁵ Most of these organisms do not have other distinct morphologic characteristics, precluding further identification. Genera reported to cause hyalohyphomycosis in veterinary patients include Acremonium, Chrysosporium, Fusarium, Geosmithia, Oxyporus, Paecilomyces, Purpureocillium, Pseudallescheria and its asexual form (anamorph), Scedosporium, Sagenomella, Geomyces, Schizophyllum, Scopulariopsis, Lecythophora, and Westerdykella, among others.^66–72 By convention, this category does not include the manifestations of aspergillosis that can usually be identified as such based on their typical clinical presentations in association with morphologic features that are fairly distinct (narrow, straight-walled septate hyphae that branch at acute angles, occasionally with conidia and conidiophores when lesions are in the nasal cavity).⁶⁵ Examples include sinonasal aspergillosis caused by Aspergillus fumigatus, and disseminated aspergillosis caused by Aspergillus terreus and Aspergillus deflectus. Other types of aspergillosis may be indistinguishable from hyalohyphomycosis without culture, and therefore may be grouped into the category of hyalohyphomycosis when culture or sequencing has not been performed. Examples include Aspergillus caninus (formerly Phialosimplex caninus), which produces scant irregular hyphae and numerous round to ovoid fungal structures in tissue, and Aspergillus alabamensis, both of which have been described to cause infection in dogs on immunosuppressants.^7,73

Hyalohyphomycosis occurs much more often in dogs than in cats, with young adult, large-breed dogs most commonly affected. Clinical signs in immunocompetent animals are similar to those associated with systemic aspergillosis and include poor body condition, fever, lameness, and lymphadenomegaly, as well as central nervous system and ocular signs.⁷⁴ Osteomyelitis, diskospondylitis, pneumonia, and chorioretinitis are common.^66,69,70,75 Other reported sites of infection include kidney, liver, spleen, heart, bone marrow, skin, bladder, pancreas, thyroid, and adrenal gland.^5,66,67,76 Prognosis is guarded to poor, with high mortality rates reported.⁷⁷ Although survival for more than 1 yr has been described in dogs treated with amphotericin B or triazoles, recurrent disease develops despite therapy in the vast majority of cases.^73,77

When hyalohyphomycosis develops in dogs receiving immunosuppressive medications, the clinical course may differ from that typically seen in immunocompetent dogs in two ways. First, the disease may be confined to the skin without evidence of systemic dissemination.⁸ Second, like phaeohyphomycosis, hyalohyphomycosis that develops in animals on immunosuppressive therapy may respond more readily to medical therapy than disease that occurs in immunocompetent patients, but only if cyclosporine can be discontinued and other immunosuppressant medications can be tapered. Still, the prognosis for these patients should be considered guarded because of the potential for dissemination.^3,5

Serology for detection of fungal galactomannan antigen is routinely included in the diagnostic evaluation of human patients suspected to have invasive aspergillosis or hyalohyphomycosis, but has not been well evaluated in veterinary patients.^78,79 In an investigation of galactomannan assay results in dogs with systemic aspergillosis as well as those with other types of systemic fungal disease, the assay was found to be positive in two dogs with typical hyalohyphomycosis and negative in one dog who developed hyalohyphomycosis involving only skin and regional lymph nodes after immunosuppressant therapy.⁸⁰ In addition, positive galactomannan results have been observed in human patients with endemic mycoses (e.g., blastomycosis and coccidioidomycosis) as well as those receiving beta-lactam antibiotics.^81,82 Based on very limited information in dogs, it appears that a positive galactomannan assay may increase the index of suspicion for either hyalohyphomycosis or aspergillosis in animals with supportive clinical signs, but specificity for this purpose has not yet been adequately studied. False-negative results may also occur, especially in dogs without widespread disease.

Options for treatment of hyalohyphomycosis include amphotericin B, itraconazole, voriconazole, posaconazole, and caspofungin. In human patients, Pseudallescheria boydii and its asexual form Scedosporium apiospermum, Fusarium spp., Paecilomyces variotii, and Purpureocillium lilacinum (formerly Paecilomyces lilacinus) typically have highly resistant in vitro susceptibility patterns and are inherently more difficult to treat.³⁵ Specifically, P boydii and S apiospermum are routinely resistant to amphotericin B and have variable susceptibility to the triazoles, P lilacinum is typically resistant to amphotericin B, P variotii is often voriconazole-resistant, and Scopulariopsis isolates are resistant to triazoles with variable susceptibility to amphotericin B.^35,83,84 Voriconazole is considered the first-line treatment of choice for several of these inherently resistant fungi, with posaconazole either as salvage therapy or as first-line therapy when P variotii infection is identified.^35,84 The role of antifungal susceptibility testing for the treatment of non-Aspergillus molds is unclear, as clinical breakpoints have not been established for human or veterinary patients, and the correlation between susceptibility results and clinical outcomes has not been demonstrated. However, one recent retrospective study of human isolates suggested that a lower minimum inhibitory concentration for the first drug used for treatment was associated with a better outcome, especially if that drug was amphotericin B.⁸⁵ In general, susceptibility testing is often of limited value for veterinary cases because it requires culture, is expensive, requires submission of an isolate to an experienced reference laboratory^c, and many clients cannot afford the newer triazoles or echinocandins that the results may indicate are more likely to be effective. However, when an isolate is available and client finances do not limit choice of therapy, susceptibility testing may provide important information about which antifungals have a high minimum inhibitory concentration and thus are unlikely to result in a positive outcome, especially if used as first-line therapy.

In animals with disseminated hyalohyphomycosis when the specific pathogen is unknown, ideal therapy would usually combine a course of lipid-complexed amphotericin B with a triazole, the latter of which is then continued indefinitely. Alternatively, when financial or logistical limitations make the use of amphotericin B difficult, patients who do not require hospitalization may be treated with a triazole alone. The efficacy of voriconazole and posaconazole is generally better than itraconazole for the treatment of hyalohyphomycosis, but the cost is significantly greater.³⁵ Posaconazole has been used successfully to resolve or reduce clinical signs in dogs with systemic aspergillosis, although relapse is common.⁸⁶ Similar results may be achieved in some dogs with hyalohyphomycosis. Although fluconazole is the least expensive triazole, its efficacy for infections caused by molds such as hyalohyphomycosis is generally inferior to the newer triazoles, so its use is not recommended. The addition of terbinafine to treatment regimens using azole antifungals has been recommended but not well studied in dogs. Caspofungin may be used for the treatment of Scedosporium spp. infections in human patients, but no data regarding its use for treating hyalohyphomycosis in animals is available.³⁵

For dogs who develop hyalohyphomycosis while on immunosuppressant medication, treatment with a newer triazole may be sufficient to resolve lesions that are limited to the skin if cyclosporine can be discontinued and other immunosuppressants tapered.⁴ Surgical removal of a focal lesion is rarely indicated and should only be considered if the lesion is confined to the skin and a thorough diagnostic evaluation has failed to find evidence of other organ involvement. If surgery is performed, the authors typically follow with medical therapy for at least 2 mo and council owners regarding the risk of relapse.

Eumycotic Mycetoma

The term “eumycotic mycetoma” refers to chronic pyogranulomatous lesions associated with the formation of fungal colonies that appear histologically as aggregates of fungal hyphae (Figure 3), and grossly as tissue grains. Mycetomas can be further classified based on the type of fungi that cause them, with pigmented fungi causing black-grain mycetoma and nonpigmented fungi causing white-grain mycetoma. Black-grain mycetomas most often result from direct implantation of fungal organisms from soil and typically cause nodular lesions with draining tracts in the skin and subcutaneous tissue. Darkly pigmented tissue grains may be observed in the exudate or within the tissue.⁸⁷ Occasionally, lesions may involve bone. Cladophialophora bantiana, Curvularia lunata, and Madurella spp. have been reported as causative agents of black-grain mycetoma in veterinary patients.^87–89 White-grain mycetomas are most often caused by P boydii or Acremonium species and typically manifest as chronic body wall or intraabdominal granulomas that develop months to years after a contaminated penetrating wound or surgical site, especially after dehiscence.^90–92 Clinical signs may include the presence of a draining tract or mass involving the body wall, an intraabdominal mass, or peritonitis.

View Figure

• Download as Powerpoint
• Download Figure

Because the causative agents of eumycotic mycetoma are inherently resistant to antifungal medication, the treatment of choice is aggressive surgical resection of all infected tissues with wide margins. Unfortunately, lesion location may preclude complete resection. However, because the lesions are slowly growing, debulking the lesions followed by long-term triazole therapy may improve quality of life for a significant period of time.⁹³ Although local extension of disease may be significant, dissemination beyond adjacent tissues is rare.

Example Case Description

A 4 yr old spayed female French bulldog was originally presented for evaluation of multifocal neurologic deficits, including right-sided head tilt and absent menace in the left eye. Brain magnetic resonance imaging findings were consistent with meningoencephalitis of unknown origin. Immunosuppressive therapy was instituted with prednisone (3.75 mg/kg/day per os [PO]) and cytosine arabinoside, 200 mg/m² (administered as 50 mg/m² given subcutaneously q 12 hr for four doses) every 4 wk. Clinical signs improved and prednisone was tapered to 0.5 mg/kg/day over the next 12 wk. At 14 wk postdiagnosis, neurologic signs recurred and a dermatopathy thought to be related to cytosine arabinoside administration developed, prompting discontinuation of that drug. Prednisone was increased to 1.6 mg/kg/day, and cyclosporine was added at a dose of 5.4 mg/kg PO q 12 hr. Two weeks later, skin lesions developed on three paws and the dog was presented for re-evaluation. Physical examination at that time was normal with the exception of several small, round, raised nodules on the lateral surface of the back left paw (Figure 4A) and on the palmar surface of both front paws (Figure 4B).

View Figure

• Download as Powerpoint
• Download Figure

Histologic examination of punch biopsies taken from two of these lesions showed pyogranulomatous dermatitis with intralesional fungal hyphae and yeast, some of which appeared to be lightly pigmented (Figure 4C). The presence of pigment in the fungal organisms was confirmed with a Masson-Fontana stain for melanin (Figure 4D). Hyphae were 5–10 μm wide with frequent irregular septations and nonparallel walls. Culture of a tissue biopsy yielded an isolate identified as Exophiala spp.

The patient was diagnosed with multifocal phaeohyphomycosis associated with immunosuppressive therapy. Treatment with itraconazole (10 mg/kg PO q 24 hr) was started, and cyclosporine was discontinued. Prednisone was decreased to 0.25 mg/kg PO q 24 hr over a period of 3 wk and then discontinued. Re-examination 2 wk later showed no cutaneous lesions. Unfortunately, the patient was euthanized ∼6 mo later because of a progressive neurologic disease.

Summary

The incidence of opportunistic fungal infections in small animal patients has increased with the escalated use of multiagent immunosuppressive therapy, especially when cyclosporine is included. Cytologic or histologic categorization of the mycosis based on organism morphology, pigmentation, hyphal size, and tissue distribution allows sufficient classification to generally predict prognosis and suggest treatment options, although specific pathogen identification based on culture or DNA amplification and sequencing provides additional information that may be important in choosing appropriate therapy in dogs with hyalohyphomycosis. Avoidance of excessive or long-term immunosuppression (especially with cyclosporine) in dogs and cats is recommended to prevent development of these infections.