Histoplasmosis and Blastomycosis

S102 Histoplasmosis and Blastomycosis Robert W. Bradsher From the Division ofInfectious Diseases, University ofArkansas for Medical Sciences, and the Veterans Affairs Medical Center, Little Rock, Arkansas Histoplasmosis and blastomycosis are caused by dimorphic fungi, can be epidemic or endemic, and can produce a spectrum of illness, from subclinical infection to progressive disseminated disease. Diagnosis of both is best made by visualization of yeast in tissue or by culture. Itraconazole is the drug of choice for treatment of both histoplasmosis and blastomycosis, except in cases of life-threatening infection, for which amphotericin B is indicated. A heavy inoculum of Histoplasma capsulatum may cause acute pulmonary infection in an otherwise healthy host, resulting in fever, hypoxia, and pulmonary infiltrates. Opportunistic histoplasmosis develops as chronic pulmonary histoplasmosis in those with a structural defect in the lung (emphysema) or as disseminated histoplasmosis in patients with cellular immune deficiency (due to immunosuppressants or AIDS). Blastomyces dermatitidis causes both pulmonary and extrapulmonary disease. Lung involvement may mimic bacterial pneumonia, while chronic presentations mimic lung cancer or tuberculosis. Skin is the most common extrapulmonary site of disease, followed by bone, prostate, and central nervous system. Endemic mycoses in the middle portion of the United States include histoplasmosis and blastomycosis. Both are caused by dimorphic fungi, which exist in nature in the mycelial phase and convert to the yeast phase at body temperature. Both may cause epidemics following a point-source of infection or sporadic endemic infection. Histoplasmosis, and to a lesser degree blastomycosis, can cause subclinical illness, after which subsequent protection against progressive infection is afforded by cellular immune mechanisms. In contrast, both can cause progressive disseminated disease. Diagnosis of both is best made by visualization of yeast in smears or tissue specimens or by culture. Since colonization does not occur, as with Candida or Aspergillus species, identification of the responsible fungus ensures the correct diagnosis. Itraconazole has been shown to be the drug of choice for treatment of both infections, except in life-threatening cases, for which amphotericin B should be used. Histoplasma capsulatum may cause acute pulmonary infection in the normal (otherwise healthy) host, resulting in fever, hypoxia, and pulmonary infiltrates. Opportunistic histoplasmosis develops in those with a structural defect in the lung (emphysema) and presents as chronic pulmonary histoplasmosis; in patients with cellular immune deficiency (due to immunosuppressant medications or HIV infection), it presents as progressive disseminated histoplasmosis. Untreated, this latter form of infection may be rapidly fatal. Blastomyces dermatitidis causes both pulmonary and extrapulmonary disease. Lung involvement may mimic bacterial pneumonia, while chronic presentations mimic lung cancer or Reprints or correspondence: Dr. Robert W. Bradsher, Director, Division of Infectious Diseases, University of Arkansas for Medical Sciences, Slot 639, 4301 West Markham Street, Little Rock, Arkansas 72205. Clinical Infectious Diseases 1996; 22(8uppl 2):8102-11 This article is in the public domain. tuberculosis. Skin is the most common extrapulmonary site of disease (manifested as either verrucous or ulcerative lesions), followed by bone, prostate, and CNS. Histoplasmosis H capsulatum was first described as a cause of disease in humans by Darling in 1906, following an autopsy of a Martinique native who lived in Panama [1]. Subsequently, he described two more patients with the same infection: a Cantonese and another Martinique native, both of whom had resided in Panama for an extended time. The fact that the organism looked like an encapsulated parasite inside a histiocyte prompted the name. Twenty years later, a case of histoplasmosis in Minnesota was reported [2]. As recorded in an articulate and entertaining report by Sell [3], the next reported case of histoplasmosis occurred in Nashville, Tennessee, and resulted in a landmark article by Christie and Peterson in 1945 that changed the understanding of the interactions of fungi with human hosts [4]. In 1934, a third-year medical student at Vanderbilt University (Nashville) was puzzled by the blood smear of an infant with a strange anemia. Drs. Katherine Dodd, Edna Tompkins, Henry Meleny, and Ernest Goodpasture recalled the smears from Darling's original case and confirmed that the findings were indicative of histoplasmosis, thus enabling the first antemortem diagnosis of this infection [5]. Cultures performed by Dr. William DeMonbreun of specimens of bone marrow and blood from an autopsy performed within 2 hours of the child's death revealed not a parasite but a fungus, which displayed thermal dimorphism of mycelial growth at room temperature and yeast-phase growth during incubation at 37°C [6]. Later, he fulfilled Koch's postulates for the organism by experimental infection in mice and monkeys. Five years after that case, DeMonbreun described a case of naturally acquired histoplasmosis following an autopsy of a em 1996; 22 (Supp1 2) Histoplasmosis and Blastomycosis dog with hepatomegaly [7]. Two years later, a lO-month-old child died of complications of a neurosurgical procedure for a subdural process. Surprisingly, nodules were discovered at autopsy in the lung, liver, and spleen and H. capsulatum was subsequently cultured from hilar lymph nodes by Emmons at the National Institutes of Health. Although their association was not recognized at the time, this case and that of the dog with histoplasmosis were the first examples of asymptomatic infection. In 1940 Meleney reviewed the world's literature and reported on 45 cases of histoplasmosis, noting that the disease was uniformly fatal [8]. During this same period of time (again, as carefully recorded by Sell [3]), Gass and others from the Tennessee Department of Health were studying radiographic evidence of tuberculosis in relation to tuberculin sensitivity in school children [9]. From a county neighboring Nashville, 1,291 children with presumed pulmonary tuberculous calcifications underwent tuberculin skin testing, but only 39% of the tests were positive. This finding almost caused physicians to abandon the tuberculin skin test because of its unreliability. In 1944, a 5-month-old infant was admitted to Vanderbilt Hospital (Nashville) with a presumed acute leukemia, but laboratory tests revealed the causative organism of histoplasmosis. Dr. J. C. Peterson developed a filtrate of the original cultures from DeMonbreun's laboratory and named it histoplasmin. The antigen was injected intradermally into the child, resulting in a large amount of erythema and induration. Both parents' histoplasmin skin tests also were positive, as were those of many other children who had no history of illness. Drs. Christie and Peterson were convinced that a benign form of histoplasmosis had occurred. Children with pulmonary calcifications thought to be due to tuberculosis were skin-tested with histoplasmin and tuberculin; 49% were histoplasmin-positive and tuberculinnegative, while only 33% were histoplasmin-negative and tuberculin-positive [4]. These results confirmed that a large number of normal persons had been asymptomatically infected with H. capsulatum early in life, resulting in pulmonary calcifications. Studies by Palmer [10] and Edwards et al. [11] of military recruits subsequently confirmed these findings. Histoplasmosis was not a rare infection associated with a very high mortality rate but usually produced few or no symptoms and resolved with normal immune functions. In certain geographic areas of the Mississippi and Ohio River valleys, histoplasmosis is now known to affect up to 90% of the residents. Further elucidation regarding histoplasmosis continued to come from investigators at Vanderbilt [12 -16]. Goodwin and Des Prez [13] described the various manifestations ofhistoplasmosis and developed a schema of the interactions of the fungus with the human host (table 1). In most normal hosts with a history of moderate to minimal exposure, histoplasmosis usually was subclinical or asymptomatic. Infection could occur either with the primary inoculation or as reinfection following subsequent exposure to the fungus, particularly if continuous S103 Table 1. Classification of histoplasmosis, according to the method of Goodwin and Des Prez [13]. Type of host Normal With mild exposure With heavy exposure Abnormal With opportunistic infection With excessive fibrotic response Type or characteristic(s) of histoplasmosis Usually asymptomatic but occasionally symptomatic primary infection; asymptomatic reinfection Acute primary pulmonary infection and/or acute pulmonary reinfection Chronic pulmonary (due to structural abnormalities) or progressive disseminated (due to cellular immune deficiency) Histoplasmoma or mediastinal fibrosis exposure did not occur in the area of endemicity. For example, those who left the area of endemicity for histoplasmosis for a number of years could have a clinical or subclinical infection again upon return [15]. The occurrence in normal hosts of acute symptomatic pulmonary histoplasmosis manifested as fever, chills, myalgia, dyspnea, and hypoxia was described, but usually these patients had a history of significant exposure. The means of exposure included destroying chicken coops that had remained unused for long periods of time, cutting down trees that had been known to be bird roosts, and spelunking in caves known to have large bat populations. Such epidemiologic information made evident the fact that bird droppings acted as a growth nutrient for H. capsulatum, although birds themselves are not colonized or infected because of their high body temperatures [7]. Bat guano was also recognized as a growth factor for the organism; however, in contradistinction to birds, bats (as mammals) could be infected with the fungus. As with the asymptomatic form of infection, the clinical picture of acute pulmonary histoplasmosis could be seen in both primary infections and reinfections [15]. Histoplasmosis was also described as occurring in the abnormal host. In those with excessive fibrotic response to the fungus, mediastinal fibrosis and retroperitoneal fibrosis were noted [17]. A solitary pulmonary nodule called a histoplasmoma can also be caused by excessive fibrosis, but it is associated with less severe manifestations than the excessive scarring and collagen production associated with mediastinal fibrosis [18]. In these cases, growth of the fungus is not considered to be the pathophysiology; the assumption is that there is simply too much immune response to the fungal antigens. In other abnormal hosts, H. capsulatum can act as an opportunistic pathogen. In those with preexisting structural abnormalities of the lung due to centrilobular or bullous emphysema, chronic pulmonary histoplasmosis may occur [16]. Usually found in middle-aged or older cigarette smokers, this condition S104 Bradsher involves cavitary lesions evident on chest radiographs and symptoms of fever and copious sputum production. Antifungal treatment can reduce symptoms and signs, but the relapse rate is substantial [19, 20]. It is thought that growth of the fungus is not as much of a problem as is the inability of host defenses to clear the organism because of the structural defects due to emphysema. Finally, in the abnormal host with cellular immune deficiency, progressive disseminated histoplasmosis may occur. Currently, the usual clinical setting for this severe and often fatal condition is HIV infection, associated with depression of the CD4 lymphocyte count (to < 150-200/mm3) [21-23]. The clinical manifestations ofhistoplasmosis in patients with AIDS have been extensively reviewed by several groups [2126]. Prior to the AIDS epidemic, patients with lymphoreticular neoplasms (e.g., Hodgkins disease), patients undergoing immunosuppressant chemotherapy (e.g., for organ transplantation or rheumatic diseases), patients with sarcoidosis, or patients receiving steroids were those who most commonly had progressive disseminated histoplasmosis [14]. The manifestations of this form of infection are primarily caused by the focal destructive granulomatous lesions that develop in reaction to the hematogenously disseminated fungi. Pulmonary manifestations are usually expected with histoplasmosis, but in the disseminated form of the disease lung involvement is not frequently encountered (except in the most severe cases, in which interstitial or miliary involvement is evident on chest radiographs). Infection of the reticuloendothelial system accounts for the majority of signs in cases of progressive disseminated histoplasmosis [14]. Bone marrow involvement is common, and thrombocytopenia, anemia, and/or leukopenia frequently occur. Because of this localization, culture and histologic examination of bone marrow are routine methods of diagnosis of this condition [24]. Fever, hepatosplenomegaly, and gastrointestinal ulceration (with particular frequency of ulcers in the oropharyngeal area) are commonly seen. Gastrointestinal bleeding in patients with AIDS who reside or have resided in an area in which Histoplasma species are endemic should prompt a search for this fungus [25]. Less commonly, Addison's disease, meningitis, or endocarditis may be diagnosed secondarily to histoplasmosis [14]. The diagnosis of histoplasmosis has been reviewed extensively by Wheat [26]. For cases of acute pulmonary histoplasmosis or mild clinical cases of histoplasmosis, a history of exposure sets the clinician on the trail of this diagnosis. Some epidemiologic exposure is usually discovered in these cases [13]. Without such a history, the diagnosis of acute pulmonary histoplasmosis may be difficult. With chronic pulmonary histoplasmosis or progressive disseminated histoplasmosis, culture of clinical specimens will yield the organism, generally within 2-3 weeks. Most laboratories prefer to isolate the fungus in the mycelial phase and confirm the identification of H capsulatum by converting it to the mold phase during incubation at 35°C to 37°C or by performing antibody studies on the culture material. Histopathology also allows a firm diagnosis, with use of either Wright's staining of bone marrow em 1996;22 (Suppl 2) or peripheral blood smears or Grocott-Gomori methenaminesilver nitrate staining of tissue specimens. Wheat has reported success in diagnosis of primary histoplasmosis as well as relapse with detection of Histoplasma polysaccharide antigen in the urine ofinfected individuals [27,28]; commercial testing for this antigen is now available. Even though skin testing with histoplasmin allowed understanding of the epidemiology of this infection, the usefulness of skin testing for diagnosis is low [26]. In areas of endemicity, skin-test positivity simply proves prior infection and does not indicate whether histoplasmosis is responsible for the clinical signs and symptoms of the patient. Similar comments could be made with regard to Histoplasma serology. Both CF and immunodiffusion precipitin bands for antibodies are associated with sufficiently high rates of false-positivity and false-negativity as to limit their usefulness [19, 26]. Positive serological studies should prompt even greater efforts to obtain tissue material for culture and histopathology. Another reason not to use histoplasmin skin testing is that the test may stimulate a rise in antibody titer in previously infected persons, thus giving the false impression of recent infection [13]. Antifungal treatment of mild cases of histoplasmosis is not necessary. In cases of acute pulmonary histoplasmosis, supportive care with supplemental oxygen and even ventilatory support may be more important than antifungal therapy. Typically, for the severely ill adult patient with this condition, amphotericin B is administered (up to a total dose of 500 mg over 2-3 weeks); some patients are given corticosteroids in addition to the antifungal agent. Whether itraconazole might be effective is unknown. In cases of progressive disseminated histoplasmosis, amphotericin B is considered the drug of choice for life-threatening infection as well as for histoplasma meningitis. Table 2 shows the success rates of therapy with various agents for disseminated histoplasmosis in patients with AIDS (who received primary therapy and/or suppressive therapy to prevent relapse) as well as in persons without AIDS [21, 2933]. The first oral agent used, ketoconazole, was associated with successful outcomes for patients without AIDS [29], but this agent was found to be associated with an unacceptably high failure rate for both initial therapy and chronic suppressive therapy given to patients with AIDS [21]. Amphotericin B was effective for suppressive therapy for histoplasmosis in patients with AIDS [34]. Itraconazole has been effective in primary [30] as well as chronic suppressive [31] treatment of histoplasmosis in patients with AIDS. In addition, in the treatment of non-AIDS-associated progressive disseminated histoplasmosis, itraconazole has been shown to be effective [32]. In cases of chronic pulmonary histoplasmosis, ketoconazole and itraconazole have been shown to be effective, although a relatively high relapse rate can be expected [20, 32]; similar rates of relapse have been associated with amphotericin B therapy as well [19]. The use of fluconazole against chronic pulmonary histoplasmosis has not been well studied, but this drug does not appear to be as effective as itraconazole in patients em Histoplasmosis and Blastomycosis 1996;22 (Supp1 2) Table 2. sis. The success rates of therapy for disseminated histoplasmo- No. of treatment successes per no. of patients treated Patients with AIDS Treatment agent administered [reference] Ketoconazo1e [21] [29] Itraconazole [30] [31] [32] Fluconazole [33] Amphotericin B [21] Primary therapy Suppressive therapy to prevent relapse 11/22 20/40 Patients without AIDS Primary therapy 13/17 50/59 39/42 10/10 66/76 69/90 21/26 with AIDS; the agent might be considered for a patient who is unable to tolerate itraconazole, however [33]. Blastomycosis The first reported case of blastomycosis occurred in Baltimore and was described by Gilchrist in 1894 as a skin infection caused by a protozoan organism [35]; the illness became known as Gilchrist's disease. There were errors in the initial descriptions of the illness since it is rare in the areas surrounding Baltimore and the causative organism is a fungus that infects the skin secondarily rather than primarily. Gilchrist was the first to refute portions of his own descriptions when he isolated and named the fungus Blastomyces dermatitidis [36]. The initial cases were perceived as mainly dermatologic, according to historical reviews [37,38]. The concept of primary pulmonary blastomycosis was not adopted until the pathological descriptions were delineated [38]. Although there have been cases involving cutaneous inoculation of the fungus in laboratory workers, almost all cases of blastomycosis are recognized to begin from a pulmonary portal of entry [37]. B. dermatitidis is a thick-walled yeast cell with broad-based budding daughter cells. Like H. capsulatum, the fungus is dimorphic. The mechanisms for transformation from mycelia to yeast have been described as a result of heat-shock, proteininduced uncoupling of oxidative phosphorylation [39]. Growth is observed as a fluffy white mold on appropriate media at room temperature and as brown, wrinkled colonies of yeast at 37°C. Because this infection is sometimes difficult to diagnose and need not be reported to public health authorities, the S105 epidemiology of blastomycosis is not as well understood as that of histoplasmosis. The number of cases reported is widely variable and dependent on the incidence of the disease and the interest of physicians in reporting these infections. From 1896 to 1968 [40], 1,476 cases were reported; prevalence studies by Furcolow and colleagues indicated that rates in some geographic areas were as high as 0.5-4 cases per 100,000 population per year by 1967 [41]. As with histoplasmosis, most clinical cases of blastomycosis occur in states surrounding the Mississippi and Ohio Rivers: Arkansas, Kentucky, Mississippi, North Carolina, Tennessee, Louisiana, Illinois, and Wisconsin [37]. Endemic or isolated infections account for the majority of cases of blastomycosis, but a few epidemics of infection from point sources have also been described [42-50]. Association of epidemic cases with outdoor exposure points to a common-source outbreak. The obvious mechanism of infection is inhalation of spores from the soil. However, B. dermatitidis has not been isolated from soil nearly as commonly as H. capsulatum [13, 37]. Klein et al. [48] added substantially to the understanding of blastomycosis with the report of an outbreak of blastomycosis at Eagle River, Wisconsin, in 1985. A total of 89 fifth-grade children went to a camp to examine natural habitats of beavers and other environmental sites. Approximately one-half of the children were subsequently found to have blastomycosis; the majority of cases were subclinical. For the first time, isolation of B. dermatitidis from soil was associated with an epidemic. Klein et al. subsequently reported another, smaller point-source epidemic of infection, again associated with isolation of B. dermatitidis from soil [49]. In each instance of isolation, the specimens were of wet earth containing animal droppings, proving that the fungus exists in microfoci in soil. In the epidemics, the isolation of the fungus was from soil near bodies of water. It has yet to be determined whether water is the primary transmission factor or simply poses a greater exposure risk because of the recreational opportunities in areas with wildlife or water [51]. Infections with this fungus begin with inhalation of spores into the lung. If the organism evades nonspecific host defense mechanisms [52], the fungus undergoes a phase transition to yeast cells; the cells increase in number in the parenchyma of the lung and spread to other organs via the bloodstream. With the development of immunity, inflammatory pyogranulomatous reactions occur at the initial pulmonary site and at the widespread foci of infection. The initial response to the fungus is suppurative and is followed by formation of granulomas. This mixed neutrophilic and mononuclear cell response is distinctive of blastomycosis, although necrosis or fibrosis may also occur [37]. Typically, the granuloma of blastomycosis does not caseate, as it does in tuberculosis. Despite spontaneous resolution of the pneumonia in some cases, endogenous reactivation may occur at either pulmonary or extrapulmonary sites, with or without previous therapy [53, 54]. S106 Bradsher Table 3. Disease processes mimicked by blastomycosis. Type of blastomycosis Acute pulmonary Chronic pulmonary Cutaneous CNS Lymphadenopathy Miliary Hypercalcemia Disease process mimicked Bacterial pneumonia Tuberculosis, malignancy Basal cell carcinoma, squamous cell carcinoma, pyoderma gangrenosum, keratoacanthoma Malignancy, meningioma Lymphoma, amyloidosis Tuberculosis, carcinomatosis Sarcoidosis The clinical manifestations of blastomycosis are highly variable. Weight loss, fever, malaise, fatigue, and other nonspecific complaints are fairly common but offer little diagnostic help. The typical patient is a young to middle-aged male who works or takes recreation outdoors. In a specific outbreak ofbiastomycosis, children and women are as likely as males to be infected. Aside from an epidemic, it is rare for children to have blastomycosis [55]. The male-to-female ratio has been reported to range from 4: 1 to 15: 1 in series of endemic infections [56]. However, some of these series cases were in Veterans Administration medical centers, a fact which conspicuously adds bias to the ratio. Over the past 13 years, 78 male patients and 57 female patients have been referred to me for treatment of blastomycosis. Of these, 47% had extrapulmonary manifestations and 53% had only lung involvement. Women accounted for only 30% of the extrapu1monary cases, while 47% of the pneumonia cases involved women. The diagnosis of many patients' blastomycosis will be delayed because the illness is uncommon and can mimic many other disease processes (table 3). A pulmonary infiltrate is the most common presentation of clinical blastomycosis; the majority of patients have an alveolar or mass-like infiltrate [56]. In one report [57], such an infiltrate was radiographically apparent in 16 of 17 patients whose films were abnormal. In one series of 46 patients with blastomycosis in Arkansas, 32% had a mass and 48% had an alveolar infiltrate evident on chest radiograph [58]. Miliary or reticulonodular patterns are the next most frequent presentation. Although cavitary disease may occur, it is not as common as in cases of chronic pulmonary histoplasmosis or tuberculosis. Because of the mass lesions evidenced on their chest roentgenograms, many patients with blastomycosis are initially thought to have lung cancer. The different types of clinical presentations of pulmonary blastomycosis include acute pneumonia and chronic pneumonia. As mentioned, blastomycosis mimics other diseases. Patients with acute pneumonia due to blastomycosis may appear to have acute bacterial pneumonia because of the occurrence of fever, chills, and a productive purulent cough, with or without hemoptysis. Chronic pneumonia due to blastomycosis presents 2-6 months prior to diagnosis with symptoms including weight loss, night sweats, fever, cough with sputum production, and em 1996; 22 (Suppl 2) chest pain. Many patients with these symptoms are thought to have a malignancy or tuberculosis. Patients with a blastomycosis-related pulmonary infiltrate may have no pulmonary symptoms at all and may deny pulmonary complaints even after extensive questioning and diagnosis following routine roentgenography. In the series of Arkansas patients, 2 were asymptomatic, 16 had a clinical picture of chronic pneumonia, and 8 initially had acute pneumonia [58]. Skin lesions are the most common manifestation of extrapulmonary blastomycosis and are either verrucous or ulcerative. The verrucous, or fungating, form is raised and has a sharp but irregular border. Biopsy of these lesions shows papillomatosis and downward proliferation of the epidermis with intraepidermal abscesses [59]. The hyperplasia and acanthosis may suggest a diagnosis of cancer unless specific examinations are undertaken, such as with Grocott-Gomori methenaminesilver nitrate stain. Ulcers due to blastomycosis undergo the same histologic changes as verrucous forms but are different in that the subcutaneous abscess drains to the surface. The borders are heaped up, and the base usually contains exudate. Osteomyelitis due to B. dermatitidis infection occurs in up to 25% of extrapulmonary cases [56]. The vertebrae, pelvis, sacrum, skull, ribs, and long bones are most commonly affected, but involvements of almost every bone have been reported [38, 60]. The radiographic appearance of blastomycosis is not specific and cannot be differentiated from that of other fungal, bacterial, or neoplastic diseases. The genitourinary system is the next most frequently involved site, and because males are more likely to have extrapulmonary manifestations than are females, prostatitis and epididymo-orchitis have been reported most commonly in cases with such involvement [37, 38, 61]. Collection of urine after prostatic massage improves the rate of detection of genitourinary involvement. Typically, as with skin or bone infection, the organism will cause concurrent presentations in the lung as well as the prostate or testicle; chest radiography should be performed in every case of this infection to aid the diagnosis. Endometrial infection acquired by sexual contact with a man with blastomycosis of the penis [62] and tubo-ovarian abscess following hematogenous dissemination [63] are examples of female genital tract infection, a less frequently diagnosed complication of blastomycosis. Meningitis and-even more commonly-epidural or cranial abscesses are manifestations of the neurological involvement of blastomycosis [64, 65]. These conditions may be difficult to diagnose. Evaluation of lumbar spinal fluid is rarely definitive. Ventricular fluid has been associated with higher rates of culture positivity; in one series of cases, culture of fluid obtained by lumbar puncture led to the diagnosis in only 2 of 22 cases, whereas ventricular fluid specimens were culture-positive in 6 of 7 cases [64]. Lesions ofblastomycosis have been reported to cause disease in virtually every organ. Widely disseminated or miliary blastomycosis may occur with adult respiratory distress syndrome em 1996; 22 (Suppl 2) Histoplasmosis and Blastomycosis (ARDS) as the presenting feature [66, 67]. The majority of patients, but not all, with this pattern of diffuse infiltrates, noncardiac pulmonary edema, and refractory hypoxemia die very quickly [68]; three patients in one series died with ARDS despite intensive medical and antifungal therapy [58]. A recent report noted a higher survival rate among patients with ARDS and blastomycosis, a rate which was thought to be due to more rapid diagnosis and more prompt escalation of the dosage of amphotericin B [67]. Abscesses are most common in the subcutaneous tissue, as already noted, but they may be found in the brain, skeletal system, prostate, myocardium, pericardium, orbit, sinuses, pituitary gland, adrenal gland, or any organ [38]. The reticuloendothelial system can be involved; several cases of lymph node or hepatic involvement have been reported [37, 58, 69], and a few cases of splenic abscesses have been documented [70]. Lesions in the mouth and oropharyngeal area occur but are not found with the same frequency as lesions due to H. capsulatum in cases of disseminated histoplasmosis. One exception with regard to blastomycosis is the relatively high frequency of localization in the larynx [71]. Although blastomycosis may cause infections in immunocompromised patients, other fungal infections such as progressive disseminated histoplasmosis or cryptococcal meningitis are more likely to be opportunistic. Infection in immunosuppressed patients is thought to develop following environmental exposure or through subsequent reactivation, just as in immunocompetent patients. Unlike similar fungi, B. dermatitidis has been reported as a significant pathogen following infection with HIV in only a relatively small number of cases [72, 73]. The diagnosis of blastomycosis is made by either visualization ofthe fungus in tissue or exudate or growth ofthe organism in culture. As with H. capsulatum, B. dermatitidis is relatively easy to detect in both smears and cultures, and since colonization does not occur, that discovery is reliable for establishing a firm diagnosis [56]. In addition to wet preparations of sputum with potassium hydroxide, cytological specimens can be examined for a dependable diagnosis; since blastomycosis often looks radiographically like carcinoma of the lung, cytological analysis is commonly performed. In areas where this infection occurs infrequently, many cases will be diagnosed only following one or more invasive procedures such as bronchoscopy or open biopsy. Other available diagnostic tests include those for CF antibodies, immunodiffusion precipitin bands, and delayed hypersensitivity of skin to blastomycin. Unfortunately, these studies are unreliable for diagnosis. Cross-reactivity to antigens of various fungi, particularly between B. dermatitidis and H. capsulatum, is severely limiting to specificity of the assays [74, 75]. For example, persons with blastomycosis are just as apt to have CF antibodies to histoplasmin as those to blastomycin [37]. For not a single patient in one series of bIastomycosis cases was the CF titer of antibodies to blastomycin greater than 1:8 [58]. S107 Better results have been reported with enzyme-linked immunosorbent assay (EIA) techniques using a yeast-phase antigen [76]. However, similar results with immunodiffusion and EIA were obtained when sera from patients with disseminated disease were tested. In a report of the largest epidemic to date, Klein et al. noted that antibody was detected by EIA, immunodiffusion, and CF techniques in 77%, 28%, and 9% of patients, respectively [48]. Therefore, serodiagnosis of blastomycosis is problematic owing to potentially low rates of sensitivity and specificity. New antibody-detection systems (as described by Klein et aL [76], Lo and Notenboom [77], and Bradsher and Pappas [78]) may be useful, as might new antigens such as a surface protein of B. dermatitidis isolated by Klein and Jones, which may allow more consistent detection of antibody in patients [79]. Skin testing with blastomycin is, unfortunately, no better as a diagnostic procedure. This mycelial-phase antigen does not provide suitable specificity or sensitivity for reliable assessment of a patient's condition and is no longer obtainable clinically. It may be useful epidemiologically; in the Eagle River outbreak, 15 of 48 patients tested had a positive skin reaction [48]. Unlike the detection of long-lasting immunity with in vitro lymphocyte reactivity to yeast antigen in cases ofblastomycosis [80], blastomycin skin test reactions rapidly diminish and become negative on repeated testing over time [44]. Until antigen testing is available for B. dermatitidis, as has been described for H. capsulatum [27], the diagnosis of blastomycosis will depend on visualization of the fungus on a smear, in tissue, or in culture. The first consideration in treating a patient with blastomycosis is whether or not to use an antifungal agent. Subclinical disease occurs with this infection just as with H. capsulatum and Coccidioides immitis infections. In the 1970s Sarosi and King [44] and Tosh et al. [45] reported on patients infected in an outbreak who did not receive amphotericin B but recovered. In describing the largest epidemic of blastomycosis reported, Klein et al. noted that only nine of the 48 patients infected with the fungus received antifungal therapy [48]. Cellular immunity was considered to be the protective response in patients who were found to have blastomycosis [81, 82]. In a group of forestry workers from an area endemic for blastomycosis, in vitro evidence of prior subclinical infection has been detected [83]. In addition, there have been reports of endogenous reactivation of blastomycosis after as long a latency period as 40 years [84]. Therefore, it appears that some patients with blastomycosis recover spontaneously, without specific antifungal therapy, just as occurs in the majority ofcases of histoplasmosis and coccidioidomycosis. The clinical presentation of the patient and the toxicity of antifungal agents are the major determinants of whether to simply keep the patient under observation or to administer an antifungal agent. The observation approach should be limited to cases of mild pulmonary blastomycosis. If relapse of infection does occur, it may be in the lung again or in a site of Sl08 em Bradsher dissemination (skin, bone, CNS, or other locations). Relapse has been discovered after the initial pulmonary infection either with or without treatment [53, 54], so if therapy is not given the patient should be monitored for a prolonged period. If the patient's condition deteriorates or the pneumonia progresses, antifungal therapy should begin. Observation alone may also be indicated if a patient's condition improves before the culture is reported to be positive for B. dermatitidis. The presence of pleural disease or of any extrapulmonary manifestations during the course of illness requires antifungal treatment. Effective therapy for blastomycosis has been available since the introduction of amphotericin B in 1956 [85]. Intravenous amphotericin B at a total dosage of at least 1.0 g over weeks resulted in cure without relapse in 84%, 77%, 91 %, 87%, and 90% of the patients treated in five large series of these fungal infections [27, 86-89]. A total dosage of 2 g has been associated with cure rates of up to 97% [90]. Relapse of blastomycosis following therapy with amphotericin B occurs only rarely and appears to be dose-dependent. In one retrospective series, relapse occurred in only 2 of 30 patients who received a total dose of > 1.5 g of the drug, vs. 5 of 19 patients who received a smaller dose [88]. Most of the relapses of blastomycosis following therapy with amphotericin B occurred shortly after its completion. However, patients have had relapses of disease after as long as 9 years following treatment [53]. The high degree of antimycotic activity of amphotericin B is associated with a relatively great toxicity. In a group of patients treated with the drug for blastomycosis and described by Abernathy [90], almost three-fourths had a decline in renal function, and a number of other toxicities were also reported (including anemia, anorexia and nausea, fever, hypokalemia, and thrombophlebitis). Interruption of therapy during some point in the course was required for 41 % ofpatients, and termination of therapy before the desired amount had been given was mandatory because of toxicity in 14% of the patients [90]. Therefore, the primary reason for withholding or postponing treatment of pulmonary blastomycosis has been the toxicity of amphotericin B. Azole antifungal agents work by disrupting the cell membranes following interactions with cytochrome P-450 enzymes; this disruption ultimately reduces the concentration of ergosterol, the major sterol of the membrane of the fungus [91]. Orally absorbed forms of ketoconazole, itraconazole, and fluconazole are generally well tolerated. Ketoconazole has been found to cause sporadic cases ofsevere hepatocellular damage [91], but this appears to be a very rare occurrence. At high dosages, hormonal abnormalities with resultant gynecomastia, dysfunctional uterine bleeding, and oligospermia may occur [29, 58, 92]. Compared to the effects of amphotericin B, ketoconazole causes common but less severe adverse effects, including nausea and vomiting (when the drug is taken without a meal), dizziness, pruritus, and headache [29, 58]. The absorption of both ketoconazole and itraconazole requires an acidic gastric content; thus, antacids or H2 blocking agents such as cimetidine or ranitidine are contraindicated dur- 1996; 22 (Suppl 2) Table 4. Results of therapy with itraconazole for blastomycosis in two studies. No. of patients [reference] Result of therapy Cure Per no. who received adequate therapy Relapse Failure n = 48* [32] n = 42* [94] 43 38 38/40 36/39 1 4 4 1 NOTE. Results of therapy with amphotericin B are discussed in the text. * Nine patients in [32] and 16 patients in [94] were receiving treatment with itraconazole after failure of therapy with ketonazole or fluconazole or after relapse following therapy with either agent. ing therapy with them. Agents such as rifampin, oral contraceptives, phenobarbital, and phenytoin may be associated with reduced efficacy of itraconazole on account ofincreased metabolism of the antifungal drug. Itraconazole may cause substantial increases in the blood levels of cyclosporine or digoxin; careful review of other medications must be made when itraconazole therapy is planned. Ketoconazole has been effective as therapy for blastomycosis, but its side effects have prompted a search for other agents. The mechanism ofaction and the pharmacokinetics of itraconazole are similar to those ofketoconazole [91]. The major theoretical advantage of itraconazole over ketoconazole is a lower rate of toxicity in investigational trials [93], particularly with regard to endocrine abnormalities. Beginning in 1985, the drug was used at a dosage of 200-400 mg per day in the Mycoses Study Group of the National Institutes of Health [32] (table 4). Of the 40 patients with blastomycosis treated for an adequate duration with the agent, only two were not considered to be cured. I used the agent at a dosage of 200 mg/d to treat 42 patients with blastomycosis [94] and noted a good response (table 4). Although the initial response to itraconazole was rapid in all cases, in five the response was less than satisfactory: two patients had relapses that were successfully treated with another course of itraconazole, one had CNS lesions and was switched to therapy with amphotericin B, and two required treatment with amphotericin B after relapse. The experience with fluconazole in treatment of B. dermatitidis infections has been limited; a pilot study (with dosages of 200-800 mg/d) was recently described [95]. This agent, which is approved by the Federal Drug Administration only for use against cryptococcal and candidal infections, should not be considered equivalent to itraconazole for treatment of blastomycosis. At least 6 months of oral therapy with azole antifungal agents has been recommended by most authorities, but there have been no trials ofa shorter span of therapy. Comparative lengths of therapy need to be studied to determine the optimal duration. Difficulties in the design of suitable trials comparing amphoter- cm 1996;22 (Supp1 2) Histoplasmosis and Blastomycosis icin Band imidazoles have been due to the intravenous administration of amphotericin B and its relative toxicity vs. that of the oral medicines. Comparisons may be possible in the future with severely ill patients who have blastomycosis if itraconazole or other agents continue to be associated with adequate efficacy rates. Because relatively few cases of this infection occur in a single location, multicenter trials similar to those perfonned by the Mycoses Study Group will likely be required. The results of therapy with itraconazole for blastomycosis (table 4) are encouraging [32, 91, 94]. However, for the very ill patient, amphotericin B remains the treatment of choice. Perhaps when improvement is noted following administration of --500 mg of amphotericin B, a switch to itraconazole for the remainder of the treatment regimen may be appropriate. One major disadvantage of oral therapy is the potential of noncompliance. The majority of patients whose treatment with ketoconazole failed had not taken the drug as they were instructed, a problem which is usually not encountered with intravenous therapy with amphotericin B. In addition, CNS infection with blastomycosis may occur during therapy with itraconazole or ketoconazole since penetration into the brain is minimal. Cases have been reported in which, following the clearing of cutaneous or pulmonary infection with ketoconazole, CNS infection was subsequently diagnosed [58, 96, 97]. On the basis of available infonnation, itraconazole (at a dosage of200 mg/d for 6 months) should replace amphotericin B in therapy for compliant patients who do not have overwhelming or life-threatening blastomycosis. Because of less toxicity and despite its higher cost, itraconazole has replaced ketoconazole for first-line therapy. However, for a patient with a life-threatening manifestation of infection (such as ARDS) or whose blastomycosis is associated with CNS involvement, amphotericin B remains the treatment of choice. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. References 1. Darling ST. A protozoon general infection producing pseudotubercles in the lungs and focal necroses in the liver, spleen and lymphnodes. JAMA 1906;46:1283-5. 2. Riley WA, Watson Cl Histoplasmosis of Darling with report of a case originating in Minnesota. Am J Trop Med 1926;6:271-82. 3. Sell SH. Appreciation of histoplasmosis: the Vanderbilt story. South Med J 1989; 82:238-42. 4. Christie A, Peterson JC. Pulmonary calcification in negative reactors to tuberculin. Am J Public Health 1945;35: 1131-47. 5. Dodd K, Tompkins EH. A case of histoplasmosis of Darling in an infant. Am J Trop Med 1934; 14:127-37. 6. DeMonbreun WA. The cultivation and cultural characteristics of Darling's Histoplasma capsulatum. Am J Trop Med 1934; 14:93-125. 7. De Monbreun W A. The dog as a natural host for Histoplasma capsulatum: report of a case of histoplasmosis in this animal. Am J Trop Med 1939; 19:565-87. 8. Meleney HE. Histoplasmosis (reticulo-endothelial cytomycosis): a review with mention of thirteen unpublished cases. Am J Trop Med 1940; 20:603 -16. 9. Gass RS, Gauld RL, Harrison EF, Stewart HC, Williams WC. Tuberculosis studies in Tennessee: roentgenological evidence oftuberculous infection 26. 27. 28. 29. 30. 31. 32. 33. S109 in relation to tuberculin sensitivity in school children. Am Rev Tuberc 1938;38:441-7. Palmer CEo Nontuberculous pulmonary calcification and sensitivity to histoplasmin. Public Health Rep 1945;60:513-20. Edwards LB, Acquaviva FA, Livesay VT, Cross FW, Palmer CEo An atlas of sensitivity of tuberculin, PPD-B, and histoplasmin in the United States. Am Rev Respir Dis 1969; 99(4)(suppl): 1-132. Goodwin RA Jr, Des Prez RM. Pathogenesis and clinical spectrum of histoplasmosis. South Med 11973;66:13-25. Goodwin RA lr, Des Prez RM. Histoplasmosis. Am Rev Respir Dis 1978; 117:929-56. Goodwin RA Jr, Shapiro JL, Thurman GH, Thurman SS, Des Prez RM. Disseminated histoplasmosis: clinical and pathologic correlations. Medicine (Baltimore) 1980;59:1-33. Goodwin RA, Loyd JE, Des Prez RM. Histoplasmosis in normal hosts. Medicine (Baltimore) 1981;60:231-66. Goodwin RA Jr, Snell JD Jr. The enlarging histoplasmoma: concept of a tumor-like phenomenon encompassing the tuberculoma and coccidioidoma. Am Rev Respir Dis 1969; 100:1-12. Goodwin RA, Nickell JA, Des Prez RM. Mediastinal fibrosis complicating healed primary histoplasmosis and tuberculosis. Medicine (Baltimore) 1972; 51 :227 -46. Ajello L. Relationship of Histoplasma capsulatum to avian habitats. Public Health Rep 1964; 79:266- 70. Goodwin RA Jr, Owens FT, Snell JD, et al. Chronic pulmonary histoplasmosis. Medicine (Baltimore) 1976; 55 :413 - 52. Slama TG. Treatment of disseminated and progressive cavitary histoplasmosis with ketoconazole. Am J Med 1983;74(suppl IB):70-3. Wheat LJ, Connolly-Stringfield PA, Baker RL, et al. Disseminated histoplasmosis in the acquired immune deficiency syndrome: clinical findings, diagnosis and treatment, and review of the literature. Medicine (Baltimore) 1990;69:361-74. Wheat LJ, Slama TG, Zeckel ML. Histoplasmosis in the acquired immune deficiency syndrome. Am J Med 1985; 78:203-10. Bonner JR, Alexander WJ, Dismukes WE, et al. Disseminated histoplasmosis in patients with the acquired immune deficiency syndrome. Arch Intern Med 1984; 144:2178-81. Kurtin PJ, McKinsey DS, Gupta MR, Driks M. Histoplasmosis in patients with acquired immunodeficiency syndrome: hematologic and bone marrow manifestations. Am J Clin Patho11990;93:367-72. Graham BD, McKinsey DS, Driks MR, Smith DL. Colonic histoplasmosis in acquired immunodeficiency syndrome: report of two cases. Dis Colon Rectum 1991;34:185-90. Wheat LJ. Diagnosis and management of histoplasmosis. Eur J Clin Microbioi Infect Dis 1989; 8:480-90. Wheat LJ, Kohler RB, Tewari RP. Diagnosis of disseminated histoplasmosis by detection of Histoplasma capsulatum antigen in serum and urine specimens. N Engl J Med 1986;314:83-8. Wheat LJ, Connolly-Stringfield P, Blair R, Connolly K, Garringer T, Katz BP. Histoplasmosis relapse in patients with AIDS: detection using Histoplasma capsulatum variety capsulatum antigen levels. Ann Intern Med 1991; 115:936-41. National Institute of Allergy and Infectious Diseases Mycoses Study Group. Treatment of blastomycosis and histoplasmosis with ketoconazole. Ann Intern Med 1985; 103:861-72. Wheat J, Hafner R, Korzun AH, et al. Itraconazole treatment of disseminated histoplasmosis in patients with the acquired immunodeficiency syndrome. Am J Med 1995; 98:336-42. Wheat J, Hafuer R, Wulfsohn M, et al. Prevention of relapse of histoplasmosis with itraconazole in patients with the acquired immunodeficiency syndrome. Ann Intern Med 1993; 118:610-6. Dismukes WE, Bradsher RW Jr, Cloud GC, et al. Itraconazole therapy of blastomycosis and histoplasmosis. Am 1 Med 1992;93:489-97. Norris S, Wheat J, McKinsey D, et al. Prevention of relapse ofhistoplasmosis with fluconazole in patients with the acquired immunodeficiency syndrome. Am J Med 1994; 96:504-8. SIlO Bradsher 34. McKinsey DS, Gupta MR, Driks MR, Smith DL, O'Connor M. Histoplasmosis in patients with AIDS: efficacy of maintenance amphotericin B therapy. Am J Med 1992;92:225-7. 35. Gilchrist TC. Protozoan dermatitis. J Cutan Gen Dis 1894; 12:496-9. 36. Gilchrist TC, Stokes WR. A case of pseudo-lupus vulgaris caused by a blastomyces. J Exp Med 1898;3:53-78. 37. Sarosi GA, Davies SF. Blastomycosis. Am Rev Respir Dis 1979; 120:91138. 38. Witorsch P, Utz JP. North American blastomycosis: a study of 40 patients. Medicine (Baltimore) 1968;47:169-200. 39. Medoff G, Painter A, Kobayashi GS. Mycelial- to yeast-phase transitions of the dimorphic fungi Blastomyces dermatitidis and Paracoccidioides brasiliensis. J Bacteriol 1987; 169:4055-60. 40. Furcolow ML, Chick EW, Busey JF, Menges RW. Prevalence and incidence studies of human and canine blastomycosis. I. Cases in the United States, 1881-1968. Am Rev Respir Dis 1970; 102:60-7. 41. Furcolow ML, Busey JF, Menges RW, Chick EW. Prevalence and incidence studies of human and canine blastomycosis. II. Yearly incidence studies in three selected states, 1960-1967. Am J Epidemiol 1970; 92: 121-31. 42. Centers for Disease Control. Blastomycosis-North Carolina. MMWR 1976;25:205-6. 43. Smith JG Jr, Harris JS, Conant NF, Smith DT. An epidemic of North American blastomycosis. JAMA 1955; 158:641-6. 44. Sarosi GA, King RA. Apparent diminution of the blastomycin skin test: follow-up of an epidemic of blastomycosis. Am Rev Respir Dis 1977; 116:785-8. 45. Tosh FE, Hammerman KJ, Weeks RJ, Sarosi GA. A common source epidemic of North American blastomycosis. Am Rev Respir Dis 1974; 109:525-9. 46. Kitchen MS, Reiber CD, Eastin GB. An urban epidemic ofNorth American blastomycosis. Am Rev Respir Dis 1977; 115:1063-6. 47. Cockerill FR III, Roberts GD, Rosenblatt JE, Utz JP, Utz DC. Epidemic of pulmonary blastomycosis (Namekagon fever) in Wisconsin canoeists. Chest 1984;86:688-92. 48. Klein BS, Vergeront JM, Weeks RJ, et al. Isolation of Blastomyces dermatitidis in soil associated with a large outbreak of blastomycosis in Wisconsin. N Engl J Med 1986;314:529-34. 49. Klein BS, Vergeront JM, DiSalvo AF, Kaufman L, Davis JP. Two outbreaks of blastomycosis along rivers in Wisconsin: isolation of Blastomyces dermatitidis from riverbank soil and evidence of its transmission along waterways. Am Rev Respir Dis 1987; 136:1333-8. 50. Armstrong CW, Jenkins SR, Kaufman L, Kerkering TM, Rouse BS, Miller GB Jr. Common-source outbreak of blastomycosis in hunters and their dogs. J Infect Dis 1987; 155:568-70. 51. Bradsher RW. Water and blastomycosis: don't blame beaver [editorial]. Am Rev Respir Dis 1987; 136: 1324-6. 52. Sugar AM, Picard M. Macrophage- and oxidant-mediated inhibition of the ability of live Blastomyces dermatitidis conidia to transform to the pathogenic yeast phase: implications for the pathogenesis of dimorphic fungal infections. J Infect Dis 1991; 163:371-5. 53. Landis FB, Varkey B. Late relapse of pulmonary blastomycosis after adequate treatment with amphotericin B: case report. Am Rev Respir Dis 1976; 113:77-81. 54. Recht LD, Philips JR, Eckman MR, Sarosi GA. Self-limited blastomycosis: a report of thirteen cases. Am Rev Respir Dis 1979; 120: 1109-12. 55. Steele RW, Abernathy RS. Systemic blastomycosis in children. Pediatr Infect Dis 1983;2:304-7. 56. Bradsher RW. Blastomycosis. Infect Dis Clin North Am 1988;2:877-98. 57. Halvorsen RA, Duncan JD, Merten DF, Gallis HA, Putman CEo Pulmonary blastomycosis: radiologic manifestations. Radiology 1984; 150: 1-5. 58. Bradsher RW, Rice DC, Abernathy RS. Ketoconazole therapy of endemic blastomycosis. Ann Intern Med 1985; 103:872-9. 59. Barr CC, Gamel JW. Blastomycosis of the eyelid. Arch Ophthalmol 1986; 104:96-7. cm 1996;22 (Suppl 2) 60. Moore RM, Green NE. Blastomycosis of bone: a report of six cases. J Bone Joint Surg 1982; 64A: 1097-110 1. 61. Inoshita T, Youngberg GA, Boelen LJ, Langston J. Blastomycosis presenting with prostatic involvement: report of 2 cases and review of the literature. J Uro11983; 130:160-2. 62. Farber ER, Leahy MS, Meadows TR. Endometrial blastomycosis acquired by sexual contact. Obstet GynecoI1968;32:195-9. 63. Murray JJ, Clark CA, Lands RH, Heim CR, Burnett LS. Reactivation blastomycosis presenting as a tuboovarian abscess. Obstet Gynecol 1984;64:828-30. 64. Kravitz GR, Davies SF, Eckman MR, Sarosi GA. Chronic blastomycotic meningitis. Am J Med 1981;71:501-5. 65. Roos KL, Bryan JP, Maggio WW, Jane JA, Scheld WM. Intracranial blastomycoma. Medicine (Baltimore) 1987;66:224-35. 66. Evans ME, Haynes JB, Atkinson JB, Delvaux TC Jr, Kaiser AB. Blastomyces dermatitidis and the adult respiratory distress syndrome: case reports and review of the literature. Am Rev Respir Dis 1982; 126: 1099102. 67. Meyer KC, McManus EJ, Maki DG. Overwhelming pulmonary blastomycosis associated with the adult respiratory distress syndrome. N Engl J Med 1993;329:1231-6. 68. Skillrud DM, Douglas WW. Survival in adult respiratory distress syndrome caused by blastomycosis infection. Mayo Clin Proc 1985;60:266-9. 69. Bradsher, RW, Martin MR, Wilkes TD, Waltman C, Bolyard K. Unusual presentations of blastomycosis: ten case summaries. Infect Med 1990;7:10-9. 70. Dubuisson RL, Jones TB. Splenic abscess due to blastomycosis: scintigraphic, sonographic, and CT evaluation. AJR Am J Roentgenol 1983; 140:66-8. 71. Suen JY, Wetmore SJ, Wetzel WJ, Craig RD. Blastomycosis of the larynx. Ann Otol Rhinol Laryngol 1980;89:563-6. 72. Pappas PG, Pottage JC, Powderly WG, et al. Blastomycosis in patients with the acquired immunodeficiency syndrome. Ann Intern Med 1992; 116:847-53. 73. Witzig RS, Hoadley DJ, Greer DL, Abriola KP, Hernandez RL. Blastomycosis and human immunodeficiency virus: three new cases and review. South Med J 1994;87:715-9. 74. Cox RA. Cross-reactivity between antigens of Coccidioides immitis, Histoplasma capsulatum. and Blastomyces dermatitidis in lymphocyte transformation assays. Infect Immun 1979;25:932-8. 75. Davies SF, Sarosi GA. Serodiagnosis of histoplasmosis and blastomycosis. Am Rev Respir Dis 1987; 136:254-5. 76. Klein BS, Kuritsky IN, Chappell WA, et al. Comparison of the enzyme immunoassay, immunodiffusion, and complement fixation tests in detecting antibody in human serum to the A antigen of Blastomyces dermatitidis. Am Rev Respir Dis 1986; 133:144-8. 77. Lo CY, Notenboom RH. A new enzyme immunoassay specific for blastomycosis. Am Rev Respir Dis 1990; 141:84-8. 78. Bradsher RW, Pappas PG. Detection of specific antibodies in human blastomycosis by enzyme immunoassay. South Med J 1995;88:1256-9. 79. Klein BS, Jones JM. Isolation, purification, and radiolabeling of a novel 120-kD surface protein on Blastomyces dermatitidis yeasts to detect antibody in infected patients. J Clin Invest 1990; 85: 152-61. 80. Bradsher RW, Alford RH. Blastomyces dermatitidis antigen-induced lymphocyte reactivity in human blastomycosis. Infect Immun 1981;33: 485-90. 81. Bradsher RW. Development of specific immunity in patients with pulmonary or extrapulmonary blastomycosis. Am Rev Respir Dis 1984; 129:430-4. 82. Bradsher RW, Ulmer WC, Marmer DJ, Townsend JW, Jacobs RF. Intracellular growth and phagocytosis of Blastomyces dermatitidis by monocyte-derived macrophages from previously infected and normal subjects. J Infect Dis 1985; 151:57-64. 83. Vaaler AK, Bradsher RW, Davies SF. Evidence of subclinical blastomycosis in forestry workers in northern Minnesota and northern Wisconsin. Am J Med 1990; 89:470-6. em 1996; 22 (Suppl 2) Histoplasmosis and Blastomycosis 84. Ehni W. Endogenous reactivation in blastomycosis. Am J Med 1989;86:831-2. 85. Bennett JE. Antifungal agents. In: Mandell GL, Douglas RG Jr, Bennett JE, eds. Principles and practice of infectious diseases. 3rd ed. New York: Churchill Livingstone, 1990:361-70. 86. Busey JF, Baker R, Birch L, et al. Blastomycosis Cooperative Study of the Veterans Administration. Blastomycosis. I. A review of 198 collected cases in Veterans Administration hospitals. Am Rev Respir Dis 1964; 89:659- 72. 87. Lockwood WR, Allison F Jr, Batson BE, Busey JF. The treatment of North American blastomycosis: ten years' experience. Am Rev Respir Dis 1969; 100:314-20. 88. Parker JD, Doto IL, Tosh FE. A decade of experience with blastomycosis and its treatment with amphotericin B: A National Communicable Disease Center cooperative mycoses study. Am Rev Respir Dis 1969; 99:895-902. 89. Seabury JH, Dascomb HE. Results of the treatment of systemic mycoses. JAMA 1964; 188:509-13. SIll 90. Abernathy RS. Amphotericin therapy of North American blastomycosis. Antimicrobial Agents and Chemotherapy-1966; 1967:208-11. 91. Como JA, Dismukes WE. Oral azole drugs as systemic antifungal therapy. N Engl J Med 1994;330:263-72. 92. Pont A, Graybill JR, Craven PC, et al. High-dose ketoconazole therapy and adrenal and testicular function in humans. Arch Intern Med 1984; 144:2150-3. 93. Tucker RM, Haq Y, Denning DW, Stevens DA. Adverse events associated with itraconazole in 189 patients on chronic therapy. J Antimicrob Chemother 1990; 26:561-6. 94. Bradsher RW. Blastomycosis. Clin Infect Dis 1992; 14(suppl 1):S82-90. 95. Pappas PG, Bradsher RW, Chapman SW, et al. Treatment of blastomycosis with fluconazole: a pilot study. Clin Infect Dis 1995; 20:267 - 71. 96. Hebert CA, King JW, George RB. Late dissemination of pulmonary blastomycosis during ketoconazole therapy. Chest 1989; 95:240-2. 97. Pitrak DL, Andersen BR. Cerebral blastomycoma after ketoconazole therapy for respiratory tract blastomycosis. Am J Med 1989; 86: 713-4.