Coccidioidomycosis in Solid Organ Transplantation

Coccidioidomycosis is an endemic fungal infection of the southwestern United States. Normally a self-limited infection in healthy hosts, coccidioidomycosis can become a serious complication in patients who have had solid organ transplantation. Among patients whose solid organ transplantation was complicated by coccidi-oidomycosis, the infection has a variety of clinical presentations. Disseminated disease is common and has substantial morbidity. Patients at risk for coccidioidal infection should be identified so that antifungal pro-phylactic therapy can be initiated. Treatment options include amphotericin B or azoles. Secondary prophylaxis is recommended because relapse is frequent. Coccidioidomycosis is an endemic fungal disease caused by Coccidioides immitis. The region of endem-icity is restricted to the Western Hemisphere; in North America, the region of endemicity includes the southwestern United States, northern Mexico, and parts of Central America. In the United States, desert regions of California and Arizona are the major areas of en-demicity [1, 2]. In healthy persons, coccidioidomycosis usually manifests with influenza-like symptoms or, less commonly, with pneumonia [2, 3]. Rarely, the disease can disseminate to extrapulmonary sites. Risk factors for disseminated disease include diabetes mellitus, pregnancy , male sex, race with dark pigmentation (African American, Filipino, Hispanic, and others), and im-munodeficiency (such as HIV infection or organ trans-plantation) [1, 2]. In 1971, the death of a child due to disseminated coccidioidomycosis after renal transplantation prompted Murphey et al. [4] to recommend that potential trans-plantation candidates with any evidence of prior cocci-dioidal infection should be excluded. Other authors have reiterated similar ideas [5]. Despite these concerns, solid organ transplantation has flourished in locations of high endemicity for coc-cidioidomycosis. We review the current literature from a few transplant centers on coccidioidomycosis and solid organ transplantation. METHODS A search of the medical literature for all papers published from January 1960 through September 2000 was conducted with use of the MEDLINE, EMBASE, and PubMed (National Library of Medicine) databases. Key words included " coccidioidomycosis " and " organ transplantation. " Also, the references from these papers were used to identify other relevant references. The search was limited to English-language sources. Papers relating to bone marrow transplantation were not included. RESULTS The MEDLINE search identified many individual case reports and 6 case series documenting coccidioido-mycosis in organ transplant recipients. Seventy-three cases were identified, and clinical information was provided for 68 patients in 26 articles [4, 6–30]. Abstracted

information included the following: age; sex; race; year of transplantation; year report was published; organ transplanted; antirejection therapy (if any); characteristics of pretransplantation coccidioidomycosis infection, if known, including time to transplantation; treatment; use of prophylactic therapy after transplantation; characteristics of the posttransplantation clinical presentation of coccidioidal infection, including time after transplantation; organ(s) involved; sites cultured; serological findings; treatment; outcome; follow-up; and autopsy findings. This collection of cases was reviewed and is referred to herein as the "Coccidioidomycosis in Transplantation" (CIT) data set. Table 1 summarizes the features of patients presenting with pulmonary coccidioidomycosis, table 2 summarizes the features of patients presenting with extrapulmonary coccidioidomycosis, and table 3 summarizes the group of patients with evidence of coccidioidal infection before transplantation.

Demographics
Reports of coccidioidomycosis complicating solid organ transplantation have appeared in the medical literature since 1967.
Race was specified for 45 of 69 patients in the CIT data set: 26 patients were white, 2 were black, 13 were Hispanic, 2 were Native American, 1 was Filipino, 1 was Korean, and 1 was Samoan. Except for 2 children who were 6 and 10 years old, the patients in the CIT data set were adults (range, 16-69 years).

Incidence of Coccidioidomycosis after Transplantation
The incidence of coccidioidomycosis after transplantation in areas of high endemicity is summarized in table 4. The earliest of these studies was a review of all cases of coccidioidomycosis

306;
; 95% CI for the difference in incidence, Ϫ0.5 P p .085 to 6.6). If a true decrease in the incidence exists, it may be because efforts have been made to identify patients at risk for coccidioidal disease and administer prophylactic therapy.

Clinical Characteristics
Risk factors for coccidioidomycosis. Immunosuppression has long been recognized as a risk factor in the presentation of coccidioidomycosis after transplantation. In 1977, Smithline et al. [14] reported increased mortality among renal transplant recipients with disseminated coccidioidomycosis who received high-dose corticosteroid treatment. Since that report, immu-nosuppressive regimens have changed, and prednisone doses have decreased substantially. All patients in the CIT data set were receiving corticosteroids as part of their immunosuppressive regimen, although details of dose and duration were not consistently reported. It is not known how other components of immunosuppressive regimens affect the presence or clinical presentation of active coccidioidal infections.
Treatment of acute rejection is also a risk factor for coccidioidal infection. In the CIT data set, 13 patients with coccidioidomycosis had antirejection therapy of some kind (high-dose corticosteroids, polyclonal antilymphocyte or antithymocyte preparations, and monoclonal therapy such as OKT3) [4, 6-10, 12, 14, 20, 25, 31]; of these, 9 had fatal disseminated disease. In 1996, Serota [9] reviewed his experience with a cohort of 160 renal transplant recipients over a 5-year period and concluded that among other factors, patients requiring у2 courses of antirejection therapy had an increased risk of coccidioidomycosis.
Concurrent immunosuppressing illnesses may be present in transplant recipients; these include diabetes mellitus, malignant neoplasms, uremia, HIV infection, and collagen-vascular disease [2]. Although these illnesses may increase the likelihood of disease dissemination in patients who have not received a transplant [10,20,27,29,35], there is insufficient information from our CIT data set to conclude that these concurrent illnesses influence the presentation of coccidioidomycosis in transplant recipients [9,22]. In 1 series, renal transplant recipients with diabetes who had coccidioidomycosis did not differ significantly from a control group of transplant recipients who had diabetes but who did not have coccidioidomycosis [8].
The risk of coccidioidomycosis after transplantation is increased by either a prior history of coccidioidomycosis or any positive serological findings obtained just before transplantation [6,22,33,34]. The increased risk of active disease in patients with a positive history or serological test may be attenuated postoperatively by initiating antifungal prophylactic therapy. The clinical course of patients with evidence of coccidioidomycosis before transplantation is summarized in table 3. Of the 21 patients listed, 11 received antifungal prophylactic therapy after transplantation. Seven of the 10 patients who did not receive this treatment died of disseminated coccidioidomycosis. Nine of the 11 patients who received antifungal prophylactic therapy survived without complications; the other 2 patients had prophylactic therapy discontinued, resulting in 1 patient dying [28] and the other having septic arthritis [9]. At the University of Arizona, 9 of 21 patients who had a history of coccidioidomycosis before transplantation were given antifungal prophylactic therapy, and none of them had postoperative recurrence of coccidioidomycosis. However, 10 of the other 12 patients who did not receive antifungal prophylactic therapy developed active coccidioidomycosis [6,33].
Risk factors for disseminated disease in the noncompromised host include race, male sex, pregnancy, blood type, and HLA type [2]. Disseminated coccidioidomycosis has an increased likelihood of developing in the following racial groups: blacks, Filipinos, Hispanics, and Native Americans [2]. In the CIT data set, the distribution of transplant recipients with coccidioidomycosis does not clearly have an overrepresentation of the highrisk racial groups, but the absence of an association cannot be excluded because the racial distribution in the denominator is not known. Race was not a significant factor for the presence of coccidioidal infection in renal transplant recipients [8], although a follow-up report from the same institution showed disseminated coccidioidomycosis in 3 Hispanics and 1 Native American [22]. The latter report did not indicate whether this racial distribution was atypical of the area's patient population.
Blood type B and male sex (but not HLA type) are additional risk factors for disseminated coccidioidomycosis but not for the mere presence of coccidioidomycosis [8]. In the CIT data set, 12 of 18 patients presenting with pulmonary coccidioidomycosis (table 1) and 24 of 29 patients with disseminated coccidioidomycosis (table 2) were male.
Clinical presentation. The first reports of coccidioidomycosis after transplantation described disseminated disease with catastrophic outcome. Retrospective case series from areas of endemicity identified less dramatic cases of coccidioidomycosis, often limited to the lungs. Patients with pulmonary infection presented variably, some with an acute illness consisting of fever, productive cough, shortness of breath, altered sensorium, or fulminant respiratory failure with septic shock [30]. Other patients presented more insidiously, with anorexia, weight loss, fatigue, and the absence of fever or pulmonary symptoms. Asymptomatic cavitary lung lesions were discovered in 2 liver transplant recipients during a radiographic evaluation for an unrelated reason [7].
In noncompromised hosts, extrapulmonary dissemination is rare. The most common sites are skin, bone and joints, and meninges [2]. In transplant recipients with coccidioidal infection, dissemination to multiple sites is common, up to 75% in 1 series [8], with or without concurrent pulmonary involvement [36]. Table 5 summarizes the sites of dissemination in patients from the CIT data set. Multiple sites of dissemination were common. The transplant graft was infected in 10 of the cases.
Timing of infection. Coccidioidomycosis after solid organ transplantation can occur at any time; the highest risk is during the first year [8,22]. In the CIT data set, the time of infection was reported for 47 patients [4, 7, 8, 10, 13-15, 19-26, 28]. Of these, 24 (51%) presented with infection within 3 months after transplantation and 33 (70%) presented within 1 year. After the first year, there was no particular distribution in the time of presentation.
Postinfection time to transplantation. In areas of endemicity for coccidioidomycosis, primary coccidioidomycosis may develop in a patient awaiting transplantation. How long should a patient wait before transplantation can be performed safely? The interval (range, 2 months to 4 years) between active coccidioidal infection and organ transplantation was known for 13 patients in the CIT data set (table 3) [6,10,22,[27][28][29]. Eleven of these 13 patients had transplantation у1 year after the episode of coccidioidomycosis. One patient had a heart transplantation performed successfully 5 months after an episode of pulmonary coccidioidomycosis [22]; this patient was given ketoconazole preoperatively and postoperatively. For pa- tients who contract an acute coccidioidal infection, transplantation should be delayed as long as possible (ideally, 1 year) and done when the infection is clinically controlled. These patients should receive uninterrupted antifungal prophylactic therapy throughout the transplantation process, continuing long into the posttransplantation period (perhaps indefinitely). A registry of organ transplant recipients who have had antecedent coccidioidal infections has been established [6] to facilitate a better understanding of the risks and benefits of transplantation in this setting. Mortality. Early reports of transplant recipients with coccidioidomycosis indicated that mortality was high [14]; overall mortality was 63% in the first case series of kidney transplant recipients with coccidioidomycosis and 72% in patients with disseminated infection [8]. Recent reports from Arizona kidney and heart transplantation programs indicate that the mortality is lower than in earlier years, ranging from 0 [34] to 25% [22]. However, in a recent retrospective review of mortality among liver transplant recipients with coccidioidomycosis at the University of California Los Angeles during 1984-1994, overall mortality was 50% (4 of 8 cases); all of the patients in this series who died had disseminated infection [7]. The decrease in coccidioidomycosis-related mortality in the Arizona transplant programs may be the result of a greater appreciation of the problem and more rapid diagnosis and initiation of therapy [33]. The use of less toxic antifungal therapy may also be a factor.

Laboratory Studies
Skin tests. Tests for delayed hypersensitivity to coccidioidin or spherulin antigens have been done to determine whether a patient has had previous exposure to C. immitis. Immunocompromised hosts, including transplant recipients, react poorly to skin tests; this type of evaluation is not sensitive for patients with active disease [14].
In a retrospective review of coccidioidomycosis in the heart transplant service at the University of Arizona, Hall et al. [6] reported 13 asymptomatic cardiac transplant recipients whose skin tests for coccidioidomycosis were positive. Recurrent coccidioidomycosis did not develop in any patient after transplantation, and only 2 patients received antifungal therapy after transplantation. Thus, a positive skin test result was not predictive of the development of coccidioidomycosis after transplantation and should not exclude an otherwise acceptable candidate for transplantation. Commercially produced skin test antigens are not currently available and may not be reintroduced into the market.
Serological evaluation. The serological evaluation of coccidioidomycosis has been recently reviewed [33]. The tests primarily use 2 antigens. The first is the tube-precipitin antigen, which detects antibodies relatively early in the course of primary infection. The second is the complement-fixing antigen, which detects antibodies that arise later in the course of infection; the concentration of the latter roughly correlates with the extent of infection and can be monitored serially to follow the course of infection. A commercially available ELISA uses a proprietary antigen to detect antibodies in patients with antibodies to both the tube-precipitin and complement-fixing antigens [33].
Diminished immunocompetence can make the diagnosis of coccidioidomycosis difficult. In an early series [14], all 5 renal transplant recipients with disseminated coccidioidal infection had no response to coccidioidin skin testing, and 3 of the 5 had no detectable serological reaction. This lack of serological reactivity has also been noted at our institution (unpublished data) and others [8,10,22,25].
False-positive coccidioidal serological results have been documented in candidates for lung transplantation who had underlying cystic fibrosis. This finding was likely from high circulating levels of nonspecific or cross-reacting serum proteins that interfered with the test [37].
Cultures. The definitive method of diagnosis is to isolate the fungus in a clinical specimen. C. immitis is not a fastidious organism; it grows well on most media used in clinical microbiology laboratories and can be detected within 5 days.
C. immitis has been isolated from cultures of various body fluids. Because coccidioidomycosis typically localizes to the lungs, expectorated sputum samples are frequently sent to the laboratory for culture but may be negative in the presence of active infection. The yield can be improved with the use of bronchoscopy [8].
The ability to establish the diagnosis of disseminated coccidioidomycosis is enhanced by the use of a urine fungal culture  [4, 6-16, 18-23, 26-28, 31]. Total exceeds 38 because coccidioidomycosis disseminated to 11 site for some patients. [8,14]. The principal diagnostic tool in 6 of 10 renal transplant recipients with disseminated coccidioidomycosis was a positive result from urine fungal culture [8]. Blood cultures occasionally yield positive results for patients presenting with overwhelming infection [7,8,12,13].
Radiology. No predominant pattern of pulmonary disease was seen in a retrospective study of radiographic findings of renal transplant recipients presenting with coccidioidomycosis [36]. Lobar, nodular, alveolar, or miliary infiltrates noted in у1 lobe were all seen. Cavitary disease was unusual. Pulmonary patterns of disease were similar whether infection was isolated to the lungs or part of disseminated disease. The time to radiographic resolution after antifungal therapy among survivors ranged from 14 to 120 days.
Leukopenia. Neutropenia has not been associated with coccidioidomycosis in transplant recipients. However, many transplant recipients have lymphopenia as a result of their immunosuppressive regimens. A lymphocyte count of !1000 cells/mm 3 occurred in 4 of 5 kidney recipients with pulmonary coccidioidomycosis and in 10 of 12 recipients with disseminated disease [8].
Miscellaneous laboratory findings. Thyrotoxicosis and recurrent hypercalcemia occurred in a heart transplant recipient with disseminated coccidioidomycosis involving the thyroid [13]. The mechanism of the hypercalcemia was consistent with extrarenal production of 1,25-dihydroxyvitamin D.
Pathology. The presence of a mature spherule of C. immitis in any tissue is pathognomonic of coccidioidal infection. A single specimen may show numerous spherules in all stages of maturation [12]. The surrounding tissue may show a granulomatous [38] or suppurative inflammation with prominent polymorphonuclear cells [10,19,21]. Well-defined granulomas with giant cells accompanied by either caseating [30,31] or noncaseating [12] granulomas have been reported.

Prophylaxis and Treatment
Most patients reported in the CIT data set received amphotericin B (tables 1 and 2) before azoles were available. After initial control of the infection, many patients had recurrent infection that required retreatment. Azoles were used in the more recent cases to suppress recurrent infection. Medical treatment alone was sometimes inadequate to control the infection, and reduction of immunosuppression was required [28].
The treatment of coccidioidomycosis after transplantation is similar in many respects to the treatment of coccidioidomycosis in the noncompromised host. After coccidioidal infection is diagnosed, it is important to determine the extent of infection. Treatment recommendations for specific manifestations of coccidioidomycosis have been reviewed recently [39].
Specific antifungal therapy for the treatment of coccidioidomycosis includes amphotericin B (0.5-0.7 mg/kg/day) and azoles (ketoconazole, 400 mg/day; fluconazole, 400-800 mg/ day; or itraconazole, 200 mg b.i.d.) [39]. Generally, for rapidly progressing nonmeningeal infections, many physicians familiar with coccidioidomycosis select amphotericin B as initial therapy because its antifungal effect appears to have a more rapid onset than that of azoles [33,39]. After improvement has been seen with amphotericin B, therapy can be changed to a less toxic azole. Subacute or chronic infections are more likely to be treated initially with an azole [39]. Meningitis is also an indication for either fluconazole or itraconazole [39].
Two patients in the CIT data set received ketoconazole alone (table 2). Both patients had widely disseminated infection and both survived. No other published data were found for the use of azoles as initial therapy for transplant recipients with coccidioidomycosis. In other immunocompromised patients such as those infected with HIV, available data agree with the statements that azoles can be used initially in the treatment of less severe infections [40] and that fluconazole is effective in treating HIV-infected patients with coccidioidal meningitis [41].
Azoles have been used for both primary prophylaxis for highrisk patients and secondary prophylaxis for patients who had initial therapy with amphotericin B. Typically, doses of 100-200 mg/day have been used for primary prophylaxis and 200-400 mg/day for secondary prophylaxis [9]. The optimal duration of prophylactic therapy is unknown and tends to be prolonged (6-12 months or longer) and individualized [6,7]. The azoles bind and inhibit the hepatic cytochrome P-450 system regulating cyclosporine metabolism, frequently causing increased cyclosporine levels [42], and increase the risk of nephrotoxicity. A similar phenomenon has been observed in patients given tacrolimus [43]. Cyclosporine or tacrolimus levels should be monitored when fluconazole doses are 1100 mg daily [43]. The cost of long-term fluconazole for fungal prophylaxis is partially offset by the savings from a decreased cyclosporine dose [9]. Azole-induced hepatotoxicity was not reported in the CIT data set [9,32].
Lipid formulations of amphotericin became available after the majority of patients in the CIT data set were treated. When compared with conventional amphotericin B, lipid formulations of amphotericin B cause less nephrotoxicity in cyclosporine-treated transplant recipients [44] and have equal efficacy in immunosuppressed patients with systemic fungal infections [45]. The efficacy of lipid amphotericin B preparations in treating transplant recipients with coccidioidomycosis has not been reported but may warrant future study.
Cumulative doses of amphotericin B were reported for 33 patients in the CIT data set and ranged from 10 mg to 4.0 g. Of the 13 patients who received !1 g of amphotericin B, 3 (23%) survived. Twenty patients received у1 g of amphotericin B and 11 (55%) survived. Twenty-two patients received treatment with azoles; 2 patients received primary azole therapy, and 20 received azole therapy concurrent with or after a course of amphotericin B. Of these patients given azoles, 11 (50%) survived. Survival was associated with higher cumulative doses of amphotericin B and azole suppression.

SUMMARY
A review of the medical literature was conducted to summarize the reports of transplant recipients whose posttransplantation courses were complicated by coccidioidomycosis. Generally, the incidence of posttransplantation coccidioidomycosis is low, but its morbidity and mortality may be high, especially among patients with disseminated infection. Patients at risk for coccidioidal infection should be identified so that antifungal prophylactic therapy can be initiated. Transplant recipients with coccidioidomycosis can have various clinical presentations, and dissemination is common. Treatment with amphotericin B is commonly required. Secondary prophylactic therapy is recommended because relapse is frequent. Many questions persist regarding coccidioidomycosis in transplantation; future studies should refine screening, prophylactic therapy, and treatment for these patients.