Itraconazole maintenance treatment for histoplasmosis in AIDS: a prospective, multicenter trial

PURPOSE

To study the efficacy and safety of maintenance treatment with itraconazole for disseminated histoplasmosis in patients with AIDS.

PATIENTS AND METHODS

This was a prospective, multicenter, open-label study conducted at university-based hospitals participating in the AIDS Clinical Trial Group (ACTG). Forty-six AIDS patients with mild to moderate disseminated histoplasmosis who had successfully completed 12 weeks of induction treatment with itraconazole were treated with itraconazole, 200 mg once daily (42 patients) or 400 mg once daily (4 patients). Patients were followed at monthly intervals with clinical and laboratory assessment for relapse or toxicity. Primary outcome measures were relapse of histoplasmosis and survival. Secondary outcome measures included drug-limiting toxicity and changes in serum and urine Histoplasma polysaccharide antigen (HPA) levels.

RESULTS

Two patients relapsed during a median follow-up period of 87 weeks. The 1-year relapse-free rate was estimated to be 95.3% (95% CI, 85.3%-99.7%). One relapse may have been related to poor adherence to treatment and the second to concurrent administration of rifampin. From the start of maintenance treatment, the estimated 1-year survival rate was 73.0% (95% CI, 67.5%-77.9%). Five patients discontinued treatment because of suspected drug toxicity, three of whom had possible or probable hepatotoxicity. Median serum and urine HPA levels declined significantly during treatment. The only patient in whom antigen levels rose >2 U developed clinical relapse 1 week later; antigen levels were unavailable in the other relapsing patient.

CONCLUSIONS

Itraconazole, 200 mg daily, is effective in preventing relapse of disseminated histoplasmosis in patients with AIDS. It is generally well tolerated, but clinicians should be alert for drug interactions and possible hepatotoxicity.

Histoplasmosis. Experience during outbreaks in Indianapolis and review of the literature

Histoplasmosis remains a common infection in endemic regions of North America and Latin America, causing a broad spectrum of clinical findings. Experience during recurrent outbreaks in Indianapolis has shown the importance of immunosuppressive conditions including the acquired immunodeficiency syndrome (AIDS) as a risk factor for disseminated disease and expanded our knowledge of the common clinical manifestations. Pericarditis, rheumatologic manifestations, esophageal compression, and sarcoidlike manifestations were found to be relatively common findings in histoplasmosis. These studies have established the useful role of serologic testing and have led to the discovery of antigen testing for diagnosis of histoplasmosis. This experience also has offered the opportunity to examine the outcome of treatment in persons with AIDS, contributing to studies that have found itraconazole to be an excellent alternative to amphotericin B in persons with mild or moderately severe infection.

Treatment of histoplasmosis with fluconazole in patients with acquired immunodeficiency syndrome. National Institute of Allergy and Infectious Diseases Acquired Immunodeficiency Syndrome Clinical Trials Group and Mycoses Study Group

PURPOSE

This study assesses the efficacy and safety of fluconazole therapy in patients with acquired immunodeficiency syndrome (AIDS) and mild to moderately severe manifestations of disseminated histoplasmosis.

PATIENTS AND METHODS

This was a multicenter, open-label, nonrandomized prospective trial. All patients had AIDS and disseminated histoplasmosis. Patients were treated with 1,200 mg of fluconazole given by mouth once on the first day, then 600 mg once daily for 8 weeks, and those patients who improved clinically were then assigned fluconazole maintenance therapy 200 mg once daily for at least 1 year. Interim analysis revealed a high failure rate (10 of 20, 50%), causing revision of the protocol to increase the fluconazole dose to 1,600 mg given once on the first day, then 800 mg once daily, and the duration to 12 weeks for induction therapy and then 400 mg daily for 1 year for maintenance therapy.

MEASUREMENTS AND MAIN RESULTS

Thirty-six of 49 patients (74%; 95% confidence interval [CI]: 59% to 85%) with mild to moderately severe clinical manifestations who entered into the revised study responded to 800 mg of fluconazole daily for 12 weeks as induction therapy. Of the seven patients who failed induction therapy because of progression of histoplasmosis, one died of the infection. Of 36 patients who entered into the maintenance phase of the study receiving 400 mg of fluconazole daily for 1 year, 11 (30.5%) relapsed, including one who died (2.8%). Two of the 49 patients (4.1%) were removed because of grade 4 adverse events, alkaline phosphatase elevation for one and aspartate aminotransferase elevation in the other. The relapse-free rate at 1 year was 53% (95% CI: 32% to 89%), prompting closure of the study.

CONCLUSIONS

Fluconazole 800 mg daily is a safe and moderately effective induction therapy for mild or moderately severe disseminated histoplasmosis in patients with AIDS. On the basis of historic comparison, fluconazole 400 mg daily is less effective than itraconazole 200 to 400 mg daily or amphotericin B 50 mg given weekly as maintenance therapy to prevent relapse.

Cross-reactivity in Histoplasma capsulatum variety capsulatum antigen assays of urine samples from patients with endemic mycoses

We evaluated cross-reactivity in the antigen assay used for the diagnosis of histoplasmosis by testing urine samples from patients with disseminated fungal infections. The mycoses chosen for this study were selected on the basis of the observation that during clinical testing, cross-reactions may occur between Histoplasma capsulatum var. capsulatum, Paracoccidioides brasiliensis, Blastomyces dermatitidis, Coccidioides immitis, and Penicillium marneffei. We detected antigen in 12 of 19 patients with blastomycosis, 8 of 9 with paracoccidioidomycois, in 17 of 18 with P. marneffei infection, and in one with disseminated H. capsulatum var. duboisii infection. Cross-reactions were not observed in the assays for six patients with disseminated coccidioidomycosis. Cross-reactivity between the agents of other endemic mycoses should be considered in interpreting a positive H. capsulatum var. capsulatum antigen assay. Antigen detection may provide a rapid, provisional diagnosis for patients with serious infections caused by one of these organisms.

Hypothesis on the mechanism of resistance to fluconazole in Histoplasma capsulatum

An AIDS patient with disseminated histoplasmosis who improved during treatment with fluconazole but remained fungemic and subsequently relapsed is described. Isolates obtained from blood during therapy showed a progressive increase in fluconazole MIC from 0.625 to 20 micrograms/ml. The pretreatment, or parent, isolate and the posttreatment, or relapse, isolate demonstrated identical genetic patterns by PCR fingerprinting with three different primers. Fluconazole was less potent inhibitor of the growth of the relapse isolate than of the pretreatment isolate (50% inhibitory concentration [IC50] = 11.7 microM), while itraconazole was more potent (relapse isolate IC50 = 0.0011 microM versus pretreatment isolate IC50 = 0.0064 microM). Neither the increased sensitivity to itraconazole nor the decreased activity of fluconazole on the growth of the relapse isolate results from changes in the intracellular content of these agents. To reach 50% inhibition of ergosterol synthesis in both the parent and relapse isolates, about 2 nM itraconazole was needed; with fluconazole, 50% inhibition was achieved at 20.9 microM and 55.5 microM, respectively. Resistance to fluconazole may develop during treatment and results from decreased sensitivity of ergosterol synthesis.

Histoplasmosis in the acquired immunodeficiency syndrome

Opportunistic infections are common and major causes of morbidity in patients with AIDS. Endemic mycoses pose serious risks for patients in certain parts of the world. Histoplasmosis occurs in 2-5% of patients with AIDS in the Ohio and Mississippi River valleys of the United States and in over 25% of patients from a few cities. Antigen testing has become a highly useful method for diagnosing histoplasmosis rapidly, evaluating the response to treatment and diagnosing relapse. Treatment with amphotericin B or itraconazole is effective (90% or higher) if the patient is not seriously ill at the time of diagnosis but the mortality approaches 50% for those with multiorgan failure. Itraconazole blood levels should be monitored and drugs that impair the absorption or accelerate the metabolism of itraconazole should be avoided. Prophylaxis with itraconazole may be appropriate in areas with an incidence of histoplasmosis. A recently completed study in cities which have unusually high rates of histoplasmosis will provide greater insight into the role of prophylactic antifungal therapy.

Murine laminin binds to Histoplasma capsulatum. A possible mechanism of dissemination

Histoplasmosis, an increasingly important opportunistic infection in immunosuppressed subjects, is characterized by hematogenous dissemination of the yeast from the lung. The mechanism of this dissemination is not fully understood. Laminin, the major glycoprotein of the extracellular matrix, is known to mediate the attachment of various invasive pathogens to host tissues. In the current study, laminin is demonstrated to bind to Histoplasma capsulatum in a rapid, specific, and saturable manner. Scatchard analysis with 125I-labeled laminin revealed an estimated 3.0 x 10(4) binding sites per yeast with an apparent Kd for laminin binding of 1.6 x 10(-9) M. Laminin binding to H. capsulatum was decreased from 62 +/- 1 to 17 +/- 1 ng (P < 0.001) in the presence of 3,000 nM of Ile-Lys-Val-Ala-Val, a pentapeptide within one major cell attachment site of laminin. A 50-kD H. capsulatum laminin-binding protein was demonstrated using an 125I-Ln blot of H. capsulatum cell wall proteins. The 50-kD protein is also recognized by antibodies directed at the 67-kD laminin receptor, suggesting they are related. This study proposes a possible mechanism for H. capsulatum attachment to laminin, an important first step required for the yeast to recognize and traverse the basement membrane.

Itraconazole treatment of disseminated histoplasmosis in patients with the acquired immunodeficiency syndrome. AIDS Clinical Trial Group

PURPOSE

Amphotericin B has been the treatment of choice for disseminated histoplasmosis in patients with acquired immunodeficiency syndrome (AIDS). Oral antifungal agents would be welcome alternatives to standard treatment of disseminated histoplasmosis in less severe cases. The purpose of this study was to assess the efficacy and safety of itraconazole therapy in patients with AIDS and disseminated histoplasmosis.

PATIENTS AND METHODS

This was a multicenter, open-label, nonrandomized prospective trial conducted in university hospitals of the AIDS Clinical Trial Group. All patients had AIDS and first episodes of disseminated histoplasmosis. Patients with central nervous system involvement or with severe clinical manifestations were excluded. Patients were treated with itraconazole BID by mouth 300 mg for 3 days and then 200 mg BID for 12 weeks. Resolution of clinical findings, clearance of positive cultures, and drug tolerance were the main outcome measurements. A secondary objective was effect of therapy on Histoplasma capsulatum var capsulatum antigen levels.

RESULTS

Of 59 evaluable patients, 50 (85%) responded to therapy. Five patients withdrew because of progressive infection, 1 died of a presumed pulmonary embolus within the first week of therapy without improvement, 2 withdrew because of toxicity, and 1 was lost to follow-up after week 2 of therapy. Patients with moderately severe clinical (fever > 39.5 degrees C or Karnofsky score < 60) or laboratory abnormalities (alkaline phosphatase > 5 times normal or albumin < 3 g/dL) at baseline tended to respond more poorly than did other patients. Resolution of complaints of fever and improvement in fatigue occurred after a median of 3 and 6 weeks, respectively, and weight gain after 2 weeks. Fungemia cleared after a median of 1 week. H capsulatum var capsulatum antigen cleared from the urine and serum at rates of 0.2 and 0.3 units per week, respectively.

CONCLUSIONS

Itraconazole is safe and effective induction therapy for mild disseminated histoplasmosis in patients with AIDS, offering an alternative to amphotericin B in such cases. Patients with moderately severe or severe histoplasmosis should first be treated with amphotericin B and then may be switched to itraconazole after achieving clinical improvement.

Endemic mycoses in AIDS: a clinical review

Histoplasmosis and coccidioidomycosis are serious opportunistic infections in patients with AIDS who reside in areas of endemicity of the United States and Central and South America. Blastomycosis, although less common, also must be recognized as an opportunistic infection in patients with AIDS. Prompt diagnosis requires knowledge of the clinical syndromes and diagnostic tests as well as a high index of suspicion. Histoplasmosis and blastomycosis respond well to antifungal treatment, but relapse is common without chronic suppressive therapy. Improvements in treatment are needed in coccidioidomycosis. Research is needed to identify preventive strategies for patients at risk. These strategies may include use of prophylactic antifungal therapy or vaccination.

Diagnosis of histoplasmosis by antigen detection during an outbreak in Indianapolis, Ind

In this study we examine the sensitivity of Histoplasma capsulatum var capsulatum antigen detection for the diagnosis of histoplasmosis. This was a retrospective review of the sensitivity of antigen detection in patients who were diagnosed as having self-limited, chronic pulmonary, or disseminated histoplasmosis during an outbreak in Indianapolis, Ind. All patients had clinical and laboratory evidence of histoplasmosis, and specimens of urine or serum that were obtained from the patients were tested for H capsulatum var capsulatum antigen. Of the 195 patients who were studied, the following forms of the infection were found: disseminated (n = 108), self-limited (n = 70), chronic pulmonary (n = 14), and asymptomatic (n = 3). Antigen was detected in 92%, 21%, and 39% of the patients with the disseminated, chronic pulmonary, and self-limited forms of histoplasmosis, respectively. Tests for the antigen are most useful in patients with clinical findings of disseminated infection. Antigen detection also may be useful in those patients with more severe pulmonary involvement, especially during the first month of illness when serologic tests for antibodies may be negative.

Histoplasmosis: recognition and treatment

Histoplasmosis has become an important mycosis in regions of endemicity in North and Central America. Traditionally, treatment has been reserved for patients with disseminated or chronic pulmonary histoplasmosis. The availability of safe and effective oral regimens, however, has offered alternatives to amphotericin B. Administration of amphotericin B is highly effective as therapy and produces a rapid clinical response; it remains the treatment of choice for patients with severe or moderately severe manifestations of histoplasmosis. Ketoconazole and itraconazole are well tolerated and are effective alternatives to amphotericin B for treatment of patients with milder illnesses or for use following response to amphotericin B. The determination of fluconazole's role in therapy for histoplasmosis awaits completion of ongoing trials. Continued research is needed to develop better-tolerated fungicidal alternatives to amphotericin B and oral agents with better absorption and drug interaction profiles than those of itraconazole and ketoconazole. Preventive strategies should be explored to reduce the frequency of histoplasmosis among individuals from regions of endemicity who are at high risk for more severe manifestations of histoplasmosis.

Histoplasmosis and coccidioidomycosis in individuals with AIDS. A clinical review

Histoplasmosis and coccidioidomycosis are serious opportunistic infections in patients with AIDS residing in endemic areas of the United States and Central and South America. Prompt diagnosis requires knowledge of the clinical syndromes and diagnostic tests as well as a high index of clinical suspicion. Histoplasmosis responds well to a variety of treatments but relapses without chronic suppressive antifungal therapy. Improvements in treatment are needed in coccidioidomycosis. Research is needed to identify preventive strategies in at risk patients. These may include use of prophylactic antifungal therapy or vaccination.

Prevention of relapse of histoplasmosis with fluconazole in patients with the acquired immunodeficiency syndrome

OBJECTIVE

To assess the effectiveness of fluconazole for suppression of relapse of histoplasmosis in patients with acquired immunodeficiency syndrome (AIDS).

DESIGN

Retrospective, nonrandomized, open trial.

SETTING

Multicenter at two university referral centers and in five private practices.

PATIENTS

Seventy-six patients with AIDS and disseminated histoplasmosis who completed induction treatment with amphotericin B, itraconazole, or fluconazole and maintained on treatment with fluconazole to prevent relapse.

INTERVENTIONS

Fluconazole was given at dosages of 100 to 400 mg per day. Patients were followed by their primary physicians, who completed questionnaires collecting information about treatment and relapse status. Blood and urine specimens were submitted periodically for Histoplasma capsulatum var. capsulatum antigen determination.

MEASUREMENTS AND MAIN RESULTS

Nine of the 76 patients relapsed during fluconazole therapy and another was removed from the study because of allergic rash. Survival after initiation of therapy for histoplasmosis was 94 weeks, ranging from 74 weeks for those who received less than 1 g of amphotericin B for induction and none for maintenance therapy to 156 weeks for those who received greater than 1 g for induction and additional amphotericin B for maintenance therapy before beginning fluconazole (P < 0.02). Antigen levels fell at rates of 0.05 units/week in urine and 0.02 units/week in serum in patients who were successfully maintained in remission and increased by > or = 2 units/week in 4 of 6 patients who relapsed.

CONCLUSIONS

Fluconazole > or = 200 mg daily is a reasonable choice for chronic suppressive therapy of histoplasmosis in patients who cannot take itraconazole because of drug interactions, malabsorption, or side effects.

Clearance of Histoplasma capsulatum variety capsulatum antigen is useful for monitoring treatment of experimental histoplasmosis

We sought to determine whether measurement of Histoplasma capsulatum var. capsulatum antigen concentration in tissues and blood provided a marker for antifungal effect of itraconazole in a nonlethal murine model of histoplasmosis. Treatment with itraconazole (Sporanox), in cyclodextrin, was evaluated in a pulmonary model of histoplasmosis. Mice infected with 4.0 x 10(7) yeast-phase organisms by endotracheal inoculation were treated with itraconazole, 1.5 mg twice daily by gavage, for 10 consecutive days, beginning on day 4 of infection. All mice were sacrificed on day 15 of infection. Blood, spleen, and lung tissues were removed for culture and quantification of antigen. Numbers of organisms were significantly lower in spleens from the treated group: 20.8 +/- 41.8 vs. 65.8 +/- 39.1 in the control group, P = 0.017. Numbers of organisms in lung were 9.6 +/- 27.3 colony forming units in treated versus 24.2 +/- 36.3 in control animals, P = 0.267. Antigen concentrations in spleen tissue and serum were lower in treated versus control mice: spleen, 1.8 +/- .6 units in treated versus 11.0 +/- 2.3 in controls, P < 0.001; serum, 0.8 +/- 0.2 units in treated versus 2.2 +/- 1.0 units in controls, P < 0.001. Lung antigen concentrations were similar in the two groups, 19.2 +/- 1.4 units in treated compared to 17.9 +/- 3.0 units in control mice, P = 0.142. The cyclodextrin formulation of itraconazole (Sporanox) demonstrated antifungal activity in experimental histoplasmosis. Antigen detection was a useful marker for antifungal effect.

Objective decision-making following a portal film: the results of a pilot study

PURPOSE

To discriminate between random and systematic treatment setup errors using portal films.

METHODS AND MATERIALS

A bi-dimensional analytic techniques using multiple analyses of variance based on Hotelling's T2 statistics to derive numerical and graphical measures of daily portal film accuracy and precision has been trialed using 88 daily portal films from seven patients' treatment.

RESULTS

A demonstration is provided of how a reasonable approximation of random variation from the intended (Simulator) field center, and systematic displacement of the mean position of the portal film centers may be derived from a minimum number of portal films. If a random error as great as 10 mm exists, at least six or seven portal films are considered necessary to reliably detect and quantify the size of a systematic error.

CONCLUSION

Our results suggest that a modest systematic error could go undetected until the end of a 5 or 6 week course of treatment if only one portal film is obtained each week. A greater number of portal films should be performed during the first week of treatment to reduce the frequency of such errors. Efforts to separate and quantify both random and systematic errors in setup are worthwhile and will lead to improvements in outcome at the individual patient level and at a departmental level in the development of quality assurance programs.

Prevention of relapse of histoplasmosis with itraconazole in patients with the acquired immunodeficiency syndrome

OBJECTIVE

To assess the efficacy and safety of itraconazole in preventing relapse of histoplasmosis after induction therapy with amphotericin B in patients with the acquired immunodeficiency syndrome (AIDS) and disseminated histoplasmosis.

DESIGN

A prospective, multicenter, open-label clinical trial, with follow-up for at least 52 weeks.

SETTING

Tertiary care hospitals participating in a clinical investigation sponsored by the National Institutes of Allergy and Infectious Diseases (AIDS Clinical Trial Group and Mycoses Study Group).

PATIENTS

Forty-two patients with AIDS who had successfully completed induction therapy for disseminated histoplasmosis amphotericin B, at least 15 mg/kg body weight given over 4 to 12 weeks.

INTERVENTIONS

Itraconazole, 200 mg given orally twice daily.

MAIN OUTCOME MEASURES

Response to therapy, specifically prevention of histoplasmosis relapse, was the main outcome measure. Secondary end points were survival and the effect of therapy on Histoplasma capsulatum variety capsulatum antigen levels in urine and serum. Plasma itraconazole concentrations were measured to document drug absorption and compliance with therapy.

RESULTS

The median follow-up was 109 weeks, and median survival was 98 weeks. Two relapses occurred (5%; 95% CI, 0.5% to 16%), one in a patient withdrawn from the study 18 weeks earlier and one in a patient who did not comply with the study therapy. Patients with elevated antigen levels at study entry showed clearance of antigen from urine and serum; urine specimens became negative in 43% of patients (CI, 26% to 59%), and serum specimens became negative in 75% of patients (CI, 56% to 94%). Only one patient discontinued treatment because of itraconazole toxicity (hypokalemia).

CONCLUSIONS

Itraconazole, 200 mg twice daily, is safe and effective in preventing relapse of disseminated histoplasmosis in patients with AIDS. Antigen clearance from blood and urine correlates with clinical efficacy.

Chronic osteomyelitis caused by Staphylococcus aureus: controlled clinical trial of nafcillin therapy and nafcillin-rifampin therapy

A controlled trial of treatment of chronic osteomyelitis caused by Staphylococcus aureus compared nafcillin alone with nafcillin plus rifampin for a six-week period. Treatment was well tolerated, the only adverse effect being mild neutropenia in four of 18 patients; no toxicity was observed from rifampin. Eight of ten patients in the combined treatment group had a favorable clinical response (with follow-up of two to four years) as compared to four of eight in the nafcillin group (P = .2). Despite the failure to show a statistically significant advantage of rifampin plus nafcillin, we conclude that the combination, along with appropriate surgery, should be considered for patients with chronic staphylococcal osteomyelitis.

Humoral and cellular immune response in ocular histoplasmosis

We have compared several humoral and cellular immune responses in ocular and in acute histoplasmosis syndromes. T-lymphocyte subset analysis revealed elevated ratios of T-helper/T-suppressor lymphocytes in ten of 22 patients with ocular histoplasmosis, whereas ratios were depressed in 18 of 22 with acute histoplasmosis syndromes. Lymphocyte transformation responses to mitogens and to histoplasma antigens were similar in both acute and ocular histoplasmosis. However, histoplasma antibody levels as measured by immunodiffusion, complement fixation, and radioimmunoassay were elevated in only three of 22 patients with ocular histoplasmosis, compared with 18 of 22 with acute histoplasmosis. Natural killer cell cytotoxicity was depressed in ten of 20 patients with ocular histoplasmosis compared with none of 18 with acute histoplasmosis. These findings support the clinical observations that ocular histoplasmosis and acute histoplasmosis syndrome are seldom, if ever, seen in the same patient.

Cerebrospinal fluid Histoplasma antibodies in central nervous system histoplasmosis

We have evaluated the Histoplasma antibody response in the cerebrospinal fluid (CSF) in nine patients with central nervous system histoplasmosis and 98 controls. While the CSF Histoplasma antibody response identified eight of the nine patients, CSF cultures were positive in only two. Of controls with histoplasmosis but without meningitis (13 patients), or without histoplasmosis (85 patients), elevated CSF antibodies were detected by complement fixation in seven, by IgG radioimmunoassay in 17, and by IgM radioimmunoassay in five. Measurement of the CSF Histoplasma antibody response appears useful for identifying meningitis in patients with histoplasmosis, although cross-reactions occur in half of patients with other forms of chronic fungal meningitis. Patients with these other infections can usually be identified by tests for CSF Coccidioides antibodies, or cryptococcal antigens.

Commercially available (ENDO-STAPH) assay for teichoic acid antibodies. Evaluation in patients with serious Staphylococcus aureus infections and in controls

We have evaluated the clinical usefulness of a commercially available teichoic acid antibody assay (ENDO-STAPH). Teichoic acid antibody titers up to a 1:2 serum dilution were observed in 20% of normal subjects, thus, titers of 1:4 or more were considered positive. Of patients with Staphylococcus aureus infections, 16 of 23 with endocarditis, 12 of 20 with complicated bacteremia, three of 17 with uncomplicated bacteremia, and ten of 20 with nonbacteremic infections had positive titers. Only four of 70 controls had positive titers. Results agreed with those using our standard assay in 130 of 151 specimens. Results were reproducibly positive or negative in 17 of 18 specimens that were retested. Results were also reproducible in ten specimens retested using a different lot of standardized antigen. The ENDO-STAPH assay should broaden the clinical applications of assays for TAA.

Clinical and laboratory features of disseminated histoplasmosis during two large urban outbreaks

Clinical and laboratory features have been reviewed in 66 episodes of disseminated histoplasmosis that occurred during two large urban outbreaks in Indianapolis. Immunosuppression, age greater than 54 years, and presence of other serious underlying illnesses predisposed to the disseminated form of the disease; only 21% of patients lacked one of these risk factors. Central nervous system findings, splenomegaly, hepatomegaly, and lymphopenia suggested disseminated disease but were present in only about one-third of patients. Miliary or diffuse pulmonary infiltrates also suggested dissemination and were noted in about one-third of patients, while mediastinal lymphadenopathy was present in only 17%. Histoplasmal serologic tests, positive in 90% of patients, provided useful diagnostic clues. The diagnosis could be confirmed by culture in 88% of patients, and special stains were positive in about two-thirds. Although 10% of patients recovered without treatment, 11 patients (17%) died because of failure to suspect the diagnosis and initiate therapy promptly. Amphotericin B was effective in all patients receiving at least 500 mg, but relapse occurred if the total dose was less than 30 mg/kg. Ketoconazole appeared effective in non-immunosuppressed patients but not in those with underlying immunosuppression; however, a controlled trial comparing ketoconazole and amphotericin B is required to establish the role of this new fungistatic oral agent.