Russell E. Lewis and P. David Rogers
LEARNING OBJECTIVES
Upon completion of the chapter, the reader will be able to:
1. Differentiate epidemiologic differences and host risk factors for acquisition of primary and opportunistic invasive fungal pathogens.
2. Recommend appropriate empiric or targeted antifungal therapy for the treatment of invasive fungal infections.
3. Describe the components of a monitoring plan to assess effectiveness and adverse effects of pharmacotherapy for invasive fungal infections.
4. Evaluate the role of antifungal prophylaxis in the prevention of opportunistic fungal pathogens.
KEY CONCEPTS
The diagnosis of endemic fungal infections is often prompted by a patient history of prolonged infectious symptoms, travel or residence in an endemic area, and/or participation in activities that result in exposures to soil contaminated by endemic fungi.
The approach to antifungal therapy in patients with endemic fungal infections is determined by the severity of clinical presentation, the patient’s underlying immunosuppression, and potential toxicities and drug interactions associated with antifungal treatment.
Secondary prophylaxis or suppressive therapy is recommended for endemic mycoses in immunocompromised patients, especially in hosts with pronounced defects in T-cell–mediated immunity (i.e., AIDS).
Commensal or environmental fungi that are typically harmless can become invasive mycoses when the host immune defenses are impaired. Host immune suppression and risk for opportunistic mycoses can be broadly classified into three categories: (a) quantitative or qualitative deficits in neutrophil function, (b) deficits in cell-mediated immunity, and (c) disruption of integument and/or microbiologic barriers.
Familiarity with the epidemiology and frequency of nonalbicans Candida species in the institution is essential before selecting empiric antifungal therapy for invasive candidiasis.
Laboratory identification of Candida in clinical samples must be performed to the species level whenever possible, as Candida species differ considerably in their susceptibility to antifungal agents.
If a patient is non-neutropenic, clinically stable, and has never received prior azole therapy, fluconazole 800 mg/day (12 mg/kg/day) is an appropriate first-line therapy for invasive candidiasis until identification of the Candidaisolate. Liposomal amphotericin B 3 mg/kg, an echinocandin (i.e., caspofungin 70 mg on day 1, then 50 mg/day), or voriconazole, are suitable options for empiric therapy in patients with neutropenic fever.
Clinical trials performed by the National Institute of Allergy and Infectious Diseases (NIAID) Mycoses Study Group showed that 2 weeks of induction antifungal therapy with combination amphotericin B (0.7 mg/kg/day) plus flucytosine (100 mg/kg/day) for cryptococcal meningitis, followed by consolidation therapy with fluconazole (400 mg daily) for 8 weeks was as effective as 4 weeks of combination therapy, and had fewer toxicities.
Nodular or halo-like lesions detected by high-resolution computed tomography (HRCT) scans are often the first indication of invasive pulmonary aspergillosis.
Immunocompromised patients on fluconazole with progressive sinus or pulmonary disease by radiography should be evaluated for possible mold infection.
Many experts now consider voriconazole as the first-line treatment for invasive aspergillosis (IA) in patients without significant contraindications (e.g., drug interactions or pre-existing liver dysfunction) to azole therapy.
Invasive fungal infection or invasive mycoses are general terms for diseases caused by invasion of living tissue by fungi. Unlike superficial mycoses (see Chap. 80), invasive mycoses invade internal organs, can disseminate throughout the body, and are associated with high rates of morbidity and mortality, particularly in the immunocompromised host. Invasive fungal infections are broadly categorized as either primary or opportunistic invasive mycoses. Primary invasive fungal infections are caused by fungal spores or conidia in the soil that, when disturbed, can become aerosolized and inhaled leading to infection, even in an immunocompetent patient. Because these fungi are often endemic to certain soil types and hence geographically restricted, primary invasive fungal pathogens are also known as endemic fungi. In the United States, three species (Histoplasma capsulatum, Blastomyces dermatitidis, and Coccidioides immitis) account for most of these infections (Table 84–1). In contrast, opportunistic fungal infections occur only in the setting of compromised host immune defenses and are caused by a wider spectrum of less virulent fungal species that are generally incapable of causing infection in healthy patients (see Table 84–1). Hence, the spectrum, severity, and outcome of opportunistic fungal infections are heavily influenced by the degree, type, and severity of host immunosuppression. As a general rule, opportunistic fungal infections are difficult to diagnose, uniformly fatal if not treated early and aggressively, and associated with high rates of morbidity and mortality.
Table 84–1 Invasive Mycoses
Primary (Endemic) Invasive Fungi
Histoplasma capsulatuma
Coccidioides immitisa
Blastomyces dermatitidisa
Opportunistic Invasive Fungi
Yeast
Candida species (C. albicans, C. glabrata, C. parapsilosis, C. tropicalis, C. krusei, and others)
Cryptococcus neoformansa
Trichosporon spp. and others
Mold
Hyalohyphomycetes
Aspergillus fumigatus and other speciesa
Fusarium solani and Fusarium oxysporum
Zygomycoses (Mucor, Absidia, Rhizopus, Cunninghamella, and Rhizomucor)
Penicillium
Phaeohyphomycetes
Pseudallescheria boydii (Scedosporium spp.)
Bipolaris
Alternaria
Other
Pneumocystis jiroveci (formerly P. carinii)a,b
aMost common.
bRecently reclassified as a fungus.
ENDEMIC MYCOSES
EPIDEMIOLOGY
Endemic mycoses are true primary fungal pathogens capable of causing infection in otherwise healthy individuals. In immunocompromised patients, endemic fungal infections often present with a more fulminant course (in the case of primary infections) or reactivate to cause life-threatening infection. Because initial symptoms of an endemic fungal infection are often nonspecific and overlap with other slowly-progressing infections (e.g., tuberculosis), a careful patient history concerning travel and activities associated with potential exposure to soil contaminated with endemic fungi is essential for the diagnosis and early treatment of infection.
Two of the most common endemic fungal infections (histoplasmosis and North American blastomycosis) are found in overlapping regions in the eastern and central river basins of the United States (Fig. 84–1).1 Histoplasma capsulatum var. capsulatum, the causative fungus of histoplasmosis, grows heavily in soil contaminated with bird or bat excreta, which serve to enhance sporulation of the fungus.2 Activities in endemic regions that are classically associated with high exposures to Histoplasma capsulatum include cave exploration (spelunking), working in or demolishing chicken coops, demolition of older buildings, woodcutting in forests with large bird roosts, or spreading avian excreta as fertilizer. For blastomycosis, decaying organic matter, warm humid conditions, and proximity to water or frequent rainfall seems to support growth of this fungus.3 Occupational or recreational activities that disturb soil heavily contaminated with Blastomyces dermatitidis are common risk factors for development of blastomycosis. Coccidioidomycosis differs from histoplasmosis and blastomycosis, as the fungus is associated with arid to semiarid climates, hot summers, low altitude, alkaline soil, and sparse flora. Hence the fungus is found in the southwestern regions of the United States stretching from western Texas to southern California (see Fig. 84–1).4Epidemics of coccidioidomycosis have been reported in California following dust storms and earth quakes, including cycles of intense drought and rain, which favor the growth cycle of the fungus and enhance dispersion of its specialized spore forms called arthroconidia.
PATHOPHYSIOLOGY
Endemic fungi share several key biologic and ecological characteristics that contribute to their pathogenicity in humans. All endemic fungi exhibit temperature-dependent dimorphism, meaning they can propagate as either yeast(single cells that reproduce by budding into daughter cells) or molds (multicellular filamentous fungi that reproduce through production of conidia or spores). At environmental temperatures (25–30°C [77–86°F]), H. capsulatum, B. dermatitidis, and C. immitis grow in the mold form producing 2- to 10-μm round to oval shaped (Histoplasma and Blastomyces) or barrel-shaped (Coccidioides) conidia that are dispersed throughout the environment and in air currents. At physiologic temperatures, the conidia germinate into yeast (Histoplasma and Blastomyces) or in specialized cell forms called spherules (Coccidioides) that are resistant to killing by resident alveolar macrophages and neutrophils in the lung. Control of infection is mediated by the development of antigen-specific T-lymphocyte response that enhances macrophage fungicidal activity and formation of a granuloma to contain the fungus.5 Not surprisingly, patients with T-cell-mediated immune deficiency (e.g., AIDS patients and transplant recipients) or suppressed cellular immunity due to drug therapy (e.g., chemotherapy, high-dose corticosteroids, or tumor necrosis-α-blockers) are especially prone to severe reactivation of fungal disease.
FIGURE 84–1. Geographic localization of primary (endemic) fungi in the United States.
The most common route of infection for endemic fungi is the respiratory tract, where conidia aerosolized from contaminated soil are inhaled into the lung. Once in the lung, conidia are phagocytosed but not destroyed by resident macrophages and neutrophils in the alveoli and bronchioles. Within 2 to 3 days, conidia germinate into facultative yeast resistant to phagocytosis and killing by macrophages and neutrophils (Fig. 84–2A and B). For C. immitis,germination of the arthroconidia results in the formation of a sac-like structure called a spherule filled with endospores (Fig. 84–2C). Spherules then rupture to release large numbers of endospores, which are the propagating form of the infection. Control of infection in the lungs is typically accomplished through formation of granulomas. However, in patients exposed to an overwhelming inoculum, or lower inocula in the setting of suppressed T-cell–mediated immunity, dissemination outside the lung to the skin and oral mucosa (especially blastomycoses), adrenal glands, bone, spleen, thyroid, GI tract, heart, and CNS is possible and uniformly fatal if left untreated.
FIGURE 84–2. Histopathology of endemic mycoses in tissue. A. Histoplasmosis (yeast). B. Blastomycoses (broad-based budding yeast). C. Coccidioidomycoses (spherules with endospores)
CLINICAL PRESENTATION AND DIAGNOSIS
Histoplasmosis and coccidioidomycosis are frequently asymptomatic infections in immunocompetent patients or present as a self-limiting, influenza-like illness 1 to 3 weeks after inhalation of conidia. The clinical presentation of blastomycosis can range from asymptomatic infection, to acute or chronic pneumonia that develops 30 to 40 days after exposure, to full-blown disseminated disease.
General
• Symptomatic endemic fungal infections generally present as persistent and sometimes progressive pneumonia accompanied by fever, chills, cough, arthralgias, night sweats, and weight loss that is indistinguishable from other chronic infections such as pulmonary tuberculosis.
• The diagnosis of endemic fungal infection is often prompted by a patient history of a prolonged infectious symptoms, travel or residence in an endemic area, and/or participation in activities that result in exposures to soil contaminated by endemic fungi.
Radiographs
• Chest radiographs often reveal either diffuse or nodular infiltrates in the lung, accompanied by enlargement of the hilar and/or mediastinal lymph nodes.
• Fulminant pneumonia may be seen with high inoculum exposures, resulting in diffuse lung infiltrates that proceed to acute respiratory distress syndrome (ARDS) and respiratory failure.
SIGNS AND SYMPTOMS
• Rheumatologic symptoms such as severe arthritis, pericarditis, and erythema nodosum may be seen in 10% to 30% of patients with endemic fungi.2,6
• Dissemination outside the lung is common in patients with suppressed cellular immunity and frequently produces signs of progressing infection.
• Ulcerative oral and cutaneous lesions may also arise with any endemic fungal infections.
• Verrucose skin lesions on sun-exposed areas on the face are particularly suggestive of progressing blastomycosis and are frequently mistaken for cutaneous malignancy.6
• Dissemination of the fungi to bone marrow may result in anemia or thrombocytopenia.
• Hepatomegaly, splenomegaly, and adrenal insufficiency can also occur with dissemination of the endemic fungi to these internal organs.
• Seizures, meningeal signs, and hydrocephalus are common findings with dissemination to the CNS and portend an especially poor prognosis in the setting of disseminated coccidioidomycosis.
Definitive diagnosis of an endemic fungal infection requires growth of the fungus from body fluids or tissue, or evidence of cellular or tissue invasion in clinical samples by histopathologic staining. However, cultures may only be positive in the setting of high inoculum exposures, pneumonia, or disseminated disease.7 Serologic testing is helpful in the diagnosis and management of patients with histoplasmosis or coccidioidomycosis but lacks sufficient specificity for diagnoses of B. dermatitidis.6 In general, a fourfold rise in antibody titers of Histoplasma or Coccidioides, or any titer greater than 1:16 suggests active infection. However, many clinicians still consider titers as low as 1:8 as evidence of active disease because undetectable titers may be present in one-third of all active infections.4 Enzyme-linked immunosorbent assays (ELISAs) have been developed for detection of Histoplasma antigen in serum and urine, and a new radioimmunoassay directed against surface proteins of B. dermatitidis has shown promising sensitivity and specificity. Serial antigen testing can also provide a means for assessing response to antifungal therapy and early detection of relapse in patients with histoplasmosis or coccidioidomycosis.
Patient Encounter 1, Part 1
A 39-year-old male with chronic steroid-dependent asthma who recently relocated to Phoenix, Arizona presents with a 4-week history of increasing fever, dry cough, and pain upon deep inspiration. He also reports arthralgias and night sweats over the last 3 weeks. A chest radiograph reveals a small area of consolidation in the left lower lobe and some hilar adenopathy. Otherwise, all other routine tests and cultures appear negative.
What are this patient’s risk factors for developing an endemic fungal infection?
What is the most likely endemic fungal pathogen based on this patient’s history?
What additional information is needed to select antifungal therapy?
The clinical presentation of blastomycosis covers a wide spectrum ranging from asymptomatic infections to flu-like illness resembling other upper respiratory tract illnesses; to infections resembling bacterial pneumonia with acute onset, high fever, lobar infiltrates, and cough; to subacute or chronic respiratory illness with complex symptoms resembling tuberculosis or lung cancer or fulminant lung infections with high fever, diffuse infiltrates, and an ARDS-like presentation.6 As mentioned previously, the skin is the most common site of dissemination typically involving sunlight exposed body areas (i.e., nose, face, and arms) and mucous membranes.
TREATMENT
General Approach
The approach to antifungal therapy in patients with endemic fungal infections is determined by the severity of clinical presentation, the patient’s underlying immunosuppression, and potential toxicities and drug interactions associated with antifungal treatment. Immunocompetent patients with mild disease following exposure to H. capsulatum or C. immitis often experience a benign course of infection and rarely require antifungal therapy. Typically, these patients are followed in the outpatient setting with serial antigen testing to confirm resolving infection. Patients without clinical improvement within the first month are typically treated with oral itraconazole for 6 to 12 weeks (Table 84–2).1,4,6 Other newer azoles such as voriconazole and posaconazole appear to have good activity against endemic fungi; however, there are currently insufficient data to recommend their routine first-line use. Fluconazole (400–800 mg/day) is somewhat less effective than itraconazole but may have fewer GI adverse effects and drug interactions than itraconazole in patients who require prolonged therapy.1,4,6 Patients with progressive symptoms longer than 2 weeks or titers greater than 1:8 of histoplasmosis or coccidioidomycosis antigen are candidates for immediate antifungal therapy.1,4 Any patient with underlying immunosuppression should also receive immediate antifungal therapy. The following signs and symptoms are considered to be indicators of severe disease that requires hospitalization and initial treatment with systemic amphotericin B (see Table 84–2).1,4,6
Table 84–2 Therapeutic Approach to Endemic Fungal Infections
• Hypoxia indicated by a partial pressure of oxygen less than 80 mm Hg
• Hypotension (systolic blood pressure less than 90 mm Hg)
• Impaired mental status
• Anemia (hemoglobin less than 10 g/dL [100 g/L or 6.2 mmol/L])
• Leukopenia (less than 1 × 103/mm3 [1 × 109/L])
• Elevated hepatic transaminases (greater than five times upper limit of normal) or bilirubin (greater than 2.5 times upper limit of normal)
• Coagulopathy
• More than 10% loss in body weight
• Evidence of dissemination including cutaneous manifestations
• Meningitis
The treatment of blastomycoses is heavily dependent on the severity of clinical manifestations. Generally, patients with mild disease can be managed as outpatients with oral itraconazole.6 Patients with evidence of severe pulmonary disease or dissemination require initial treatment as inpatients with amphotericin B-based regimens until they are clinically stable, whereupon they can complete a 6- to 12-month treatment course as outpatients with oral azoles.6 Methylpredni-solone (0.5-1 mg/kg daily IV) during the first 1 to 2 weeks of antifungal therapy is often considered for patients who develop respiratory complications during initial treatment, including hypoxemia or significant respiratory distress.
Patient Encounter 1, Part 2
Selecting Antifungal Therapy
The patient’s serum titer for coccidioidomycosis returns as greater than 1:32. Based on the information presented, select an appropriate treatment plan for the patient’s coccidioidomycosis.
Does the patient require antifungal treatment at this time?
If the patient is considered to have moderately severe disease, what are the recommended treatment options?
PATIENT MONITORING AND SIDE EFFECTS
Response to antifungal therapy may be slow in patients with a prolonged history of infection or severe manifestations. However, gradual improvements in symptoms and reduction in fever are indicators of response to antifungal therapy. For histoplasmosis and coccidioidomycosis, decreasing antigen titers are also indicative of response to antifungal therapy.1,4
Antifungals used for the treatment of endemic mycoses can be associated with clinically-significant drug interactions and toxicities, especially with the prolonged treatment courses that are often required in the management of endemic mycoses. Itraconazole is available as a capsule formulation and as a liquid. The liquid formulation of itraconazole has several advantages over the capsule: it has a better oral bioavailability and does not require the low gastric pH that is required for dissolution and absorption of the capsule. However, the oral solution is somewhat dilute, has an unpalatable aftertaste (an issue when taking months of therapy), and has a much higher rate of GI side effects. Therefore, the capsule formulation is often preferred provided patients are not on acid-suppression therapy (i.e., proton pump inhibitors, histamine antagonists, or antacids).
Drug interactions are an important concern in patients taking long-term azole therapy. Itraconazole is a substrate and inhibitor of the cytochrome P450 (CYP)3A4 enzyme and the drug transporter, P-glycoprotein. Coadministration of itraconazole with inducers of this enzyme system (e.g., rifampin, phenytoin, and phenobarbital) can dramatically increase the clearance of itraconazole (and to a lesser extent fluconazole), resulting in ineffective plasma and tissue concentrations of the drug.8-11 In general, coadministration of itraconazole with these inducers should be avoided. In some cases, plasma trough levels can be drawn once the patient reaches steady state (greater than 7 days of therapy) to ensure adequate drug absorption. Concentrations less than 0.25 mcg/mL (0.25 mg/L) should be considered evidence of insufficient itraconazole absorption11 as trough concentrations should ideally approach 1 mcg/mL (1 mg/L) by the time the patient is in steady state.12
As a potent inhibitor of CYP450 enzymes including CYP3A4, itraconazole can dramatically decrease the clearance of many important medications metabolized through this enzyme, leading to potentially dangerous drug interactions. Patients receiving anticoagulation therapy with warfarin, immunosuppressive therapy with cyclosporine or tacrolimus, those taking midazolam, HMG-CoA reductase inhibitors (statins), rifabutin, chemotherapy agents (e.g., vinca alkaloids, busulfan, and cyclophosphamide), and digoxin will require dosage adjustment and careful monitoring while receiving itraconazole therapy.13Although fluconazole is not as potent an inhibitor of CYP3A4 as itraconazole, drug interactions can still be severe, especially at higher fluconazole dosages (i.e., 800 mg/day).13
All azole antifungals carry the potential for rash, photosensitivity, and hepatotoxicity. In general, hepatotoxicity is mild and reversible, presenting as asymptomatic increases in liver transaminases or less commonly, an increase in total bilirubin. Fulminant hepatic failure has been reported with itraconazole. Therefore, serial monitoring of liver function is recommended in all patients on long-term azole therapy. Long-term therapy with itraconazole has also been associated with reversible adrenal suppression and cardiomyopathy associated with the drug’s negative inotropic effects. These adverse effects can be prevented or managed with close monitoring and follow-up of patients on long-term therapy.
Amphotericin B is the mainstay of treatment of patients with severe endemic fungal infections. The conventional deoxycholate formulation of the drug can be associated with substantial infusion-related adverse effects (e.g., chills, fever, nausea, rigors, and in rare cases hypotension, flushing, respiratory difficulty, and arrhythmias). Premedication with low doses of hydrocortisone, acetaminophen, nonsteroidal anti-inflammatory agents, and meperidine is common to reduce acute infusion-related reactions. Venous irritation associated with the drug can also lead to thrombophlebitis; hence, central venous catheters are the preferred route of administration in patients receiving more than a week of therapy.
The most severe adverse effect associated with amphotericin B therapy is nephrotoxicity, which occurs through the renal vascular effects of the drug (constriction of the afferent arterioles in the kidney tubule), and direct toxicity to the kidney tubular membrane. Generally, nephrotoxicity with amphotericin B is reversible provided the drug is stopped. However, treatment interruptions can be problematic in patients with severe infections. Precipitous decreases in glomerular filtration occasionally are seen with the initiation of amphotericin B therapy, especially in patients with marked dehydration. Infusion of normal saline before and after amphotericin B, a practice known as “sodium loading” can blunt precipitous decreases in renal perfusion pressure and slow that rate of decline in the glomerular filtration rate. Tubular toxicity can be delayed by avoiding the use of other drugs with known tubular toxicity such as aminoglycosides, cyclosporine, cisplatin, or foscarnet. Generally, tubular toxicity manifests in patients with severe wasting of potassium and magnesium in the urine. Therefore, patient electrolytes must be carefully monitored and potassium and magnesium supplementation is often required. Hypokalemia and hypomagnesemia frequently precede decreases in glomerular filtration (increased serum creatinine) especially in patients who are adequately hydrated.14 Continued tubular damage, however, eventually results in decreases in renal blood flow and glomerular filtration through tubuloglomerular feedback mechanisms that further constrict the afferent arteriole.
During the 1990s, amphotericin B was reformulated into three different lipid-based formulations (Abelcet, Ambisome, and Amphotec) that have reduced rates of nephrotoxicity compared to the conventional deoxycholate formulation (Fungizone). Two of the formulations (Abelcet and Ambisome) have also shown reductions in the rates of infusion-related reactions. Although these lipid formulations are generally considered to be as effective as conventional amphotericin B deoxycholate, they are not dosed equivalently to the standard formulation. Unlike conventional amphotericin B, which is administered at dosages ranging from 0.6 to 1.5 mg/kg/day, lipid formulation doses are threefold to fivefold higher on a milligram-per-milligram basis, ranging from 3 to 5 mg/kg/day. Only one prospective study has directly compared the efficacy of a lipid amphotericin formulation to the conventional formulation.15 In a small study of AIDS patients with moderate to severe histoplasmosis, liposomal amphotericin B (Ambisome) was more effective than amphotericin B, with response rates of 84% and 64%, respectively. Ambisome may also be the preferred agent in patients with CNS infections over other lipid formulations, due to its improved CNS penetration.16
PROPHYLAXIS
Primary prophylaxis, before development of infection, is generally not recommended for endemic fungi but may be considered for patients who are severely immunocompromised. Patients with HIV infection with CD4 cell counts less than 150 cells/mm3 (histoplasmosis) or less than 250 cells/mm3 (coccidioidomycosis) living in endemic areas with high endemic case rates (greater than 10 cases per 100 patient-years) or with positive IgM or IgG antibodies to the fungal pathogen serology should receive itraconazole 200 mg daily.12 Secondary prophylaxis or suppressive therapy with itraconazole 200 mg daily, to prevent recurrence of infection is recommended for blastomycosis in immunosuppressed patients if immunosuppression cannot be reversed.6,12 In patients with prior CNS disease, fluconazole or voriconazole are the preferred drug due to the limited penetration of itraconazole into the CNS.
Patient Encounter 2, Part 1
A 43-year-old male in the surgical ICU after exploratory laparotomy following a motor vehicle accident develops fever that is unresponsive to broad-spectrum antibacterial therapy (piperacillin-tazobactam 3.75 g every 6 hours, gentamicin 120 mg every 8 hours, and vancomycin 1 g every 12 hours). The patient has a central venous catheter and a Foley catheter. Blood cultures are negative at the time, but the patient has yeast growing in the sputum and urine. Laboratory studies reveal a white blood cell count of 11.3 × 103/mm3 (11.3 × 109/L)
What are this patient’s risk factors for developing an invasive fungal infection?
What current evidence suggests that this patient has an invasive fungal infection?
If antifungal therapy is empirically started in this patient, which species need to be treated?
OPPORTUNISTIC MYCOSES
Commensal or environmental fungi that are typically harmless can become invasive mycoses when the host immune defenses are impaired. Host immune suppression and risk for opportunistic mycoses can be broadly classified into three categories:
• Quantitative or qualitative deficits in neutrophil function
• Deficits in cell-mediated immunity
• Disruption of the integument and/or microbiologic barriers
Quantitative defects in neutrophils (neutropenia) resulting from neoplastic diseases, cytotoxic chemotherapy, marrow transplantation, or aplastic anemia are among the most common risk factors for opportunistic mycoses. Qualitative defects may be seen in certain disease states (e.g., advanced diabetes mellitus and chronic granulomatous disease) or with high-dose corticosteroid therapy. Deficits in T-cell–mediated immunity secondary to AIDS, high-dose corticosteroid therapy, cyclosporine or other immunosuppressive drugs, chemotherapy, transplantation, bone marrow failure, and various other disorders have become increasingly common with the prolonged survival of transplant patients on chronic immunosuppressive therapy.
Immune deficits arising from disruption of the integument or GI/genitourinary barriers can also predispose patients to fungal infections. The most common types of integument/barrier disruptions are surgery, use of central venous and urinary catheters, hyperalimentation, and mucositis secondary to cytotoxic chemotherapy. Broad-spectrum antibacterial therapy can also predispose patients to fungal infections through disruption of the microbiologic flora in the gut, which allows overgrowth of Candida species. Successful management of opportunistic fungal pathogens, therefore, requires the reversal or reduction of underlying deficits in the host immune system.
All of the opportunistic mycoses are difficult to diagnose and often must be treated empirically before diagnosis is proven. Deciding when to initiate antifungal therapy and what opportunistic pathogens to cover is a decision governed largely by the cumulative immune deficits and clinical status of the host.
INVASIVE CANDIDIASIS
EPIDEMIOLOGY
Candida species are the most common opportunistic fungal pathogens encountered in hospitals, ranking as the third to fourth most common cause of nosocomial bloodstream infections in United States.17 The incidence of nosocomial candidiasis has increased steadily since the early 1980s, with the widespread use of central venous catheters, broad-spectrum antimicrobials, and other advancements in the supportive care of critically ill patients. In the 1980s, C. albicans accounted for over 80% of all bloodstream yeast isolates cultured from patients. By the late 1990s, this relative frequency of C. albicans had decreased to 50% in national surveys of bloodstream infections without a corresponding decrease in infections caused by nonalbicans species. Because of the inherent resistance (e.g., C. krusei) or diminished susceptibility (e.g., C. glabrata) of many of the nonalbicans species, the introduction of fluconazole in the early 1990s is often cited as the key element driving the shift in the microbiology of invasive candidiasis. However, it is likely that other institution-specific factors (e.g., increasing use of central venous catheters and increasing intensity of cytotoxic/mucotoxic chemotherapy) and use of broad-spectrum antibiotic therapy have contributed equally to this trend.18,19 It is important to be familiar with the relative epidemiology and frequency of nonalbicans Candida species in the institution or intensive care unit (ICU) before selecting empiric antifungal therapy for invasive candidiasis.
PATHOGENESIS AND CLINICAL PRESENTATION
Invasive candidiasis is not a single syndrome, rather a spectrum of infections that differ in terms of clinical presentation and course depending on the type of host immune immunosuppression. Many forms of invasive candidiasis are potentially severe, however, with high (30–60%) rates of crude morbidity and mortality.19 The most common form of invasive candidiasis is seen in non-neutropenic patients with disruption of the GI, skin or microbiologic barriers giving rise to a bloodstream infection (fungemia) from, or seeding to, a central venous catheter. Catheter-related candidemia carries a good prognosis if appropriate antifungal therapy is instituted early with catheter removal.19Fungemia can be of high density, however, leading to metastatic sites of infection and invasion of deep organs with increased morbidity. Therefore, the infection must be taken seriously, especially in patients with poor performance status (i.e., a high Acute Physiology, Age, and Chronic Health Evaluation II score) in the ICU.
Patients with acute disseminated candidiasis share many similar features as patients with catheter-related candidemia, except infection generally arises from the gut following mucotoxic chemotherapy and the patients are often pro foundly ill. Hematogenous spread to noncontiguous organs is common in patients with acute disseminated candidiasis, and outcome is heavily dependent upon recovery from neutropenia.19 Fluconazole prophylaxis has markedly decreased the incidence of acute disseminated candidiasis among high-risk patient groups such as bone marrow transplant and acute leukemia patients.19However, breakthrough infections with fluconazole-resistant C. glabrata and C. krusei are still a concern.
Some forms of invasive candidiasis are dominated by deep-organ infection and may never be detected by blood cultures. Chronic disseminated candidiasis or hepatosplenic candidiasis is a unique presentation of candidemia seen after recovery from neutropenia. Candidemia during the period of neutropenia may be initially localized to the portal circulation with dissemination to contiguous organs. After recovery of neutrophils, an inflammatory response is seen against areas of focal infection in the liver and spleen. This inflammatory response produces abdominal pain that is associated with increases in alkaline phosphatase levels and hepatocellular enzymes.19 Diagnosis is typically confirmed by patient history (recent neutropenia), and multiple areas of lucency in the liver and spleen on CT.
Focal invasive candidiasis has been reported for virtually every organ, even following apparently uncomplicated catheter-related fungemia. The most common sites of infection are the kidney, eye, and bone. Candida in the urine can be an indication of renal candidiasis or an obstructing fungus ball; however, it must be distinguished from more benign colonization of the urinary tract, especially in patients with chronic indwelling urinary catheters.19 All patients with candidemia should undergo an eye exam to rule out Candida endophthlamitis, which can be sight-threatening if not recognized early.19
FIGURE 84–3. Opportunistic mycoses in clinical samples. A. Candidiasis (tissue). B. Cryptococcosis (India ink stain of CSF). C. Aspergillosis (tissue). D. Zygomycosis (tissue).
Laboratory diagnosis of invasive candidiasis is established by detection of the yeast in blood cultures or another sterile site (Fig. 84–3A)19. Growth of Candida from urine, sputum, or respiratory secretions (including bronchoalveolar lavage) is not considered to be evidence of invasive infection, as these areas frequently become colonized with Candida species in patients receiving broad-spectrum antibiotics.19Colonization at multiple distinct body sites or with high density of Candida species, however, may precede invasive infection. Therefore, preemptive antifungal therapy may be indicated in colonized high-risk populations such as those with neutropenic fever, transplant recipients, or following major abdominal surgery.19 Although Candida are not particularly fastidious organisms, the sensitivity of blood cultures is relatively poor (less than 60%) and a negative culture does not rule out infection.19 The poor sensitivity of blood cultures for detecting invasive disease has led to the study of novel serodiagnostic tests to detect antibodies, fungal metabolites, fungal cell wall antigens, or nucleic acids of Candida species. Of the four approaches, antigen testing based on the detection of β-glucan polymers in the cell wall of Candida have appeared most promising; however, none of these diagnostic tests have achieved routine clinical use.
Laboratory identification of Candida in clinical samples must be performed to the species level whenever possible, as Candida species differ considerably in their susceptibility to antifungal agents.19,20Rapid discrimination of C. albicans from common nonalbicans Candida species can be accomplished by the germ-tube test, which presumptively identifies C. albicans by the early formation (less than 4 hours) of a hyphae-like structure when the yeast in incubated in serum at 37°C (98.6°F). Definitive species identification, however, may require an additional 48 to 72 hours after the organism is isolated on agar. Fluorescent in situ hybridization (FISH) of Candida species-specific DNA sequences can reduce the time needed for definitive species identification, but is not available at most hospitals.
C. albicans remains the most common cause of invasive candidiasis, is the most virulent of Candida species, but is the most susceptible to commonly used antifungals including fluconazole.19,20 Like C. albicans, C. tropicalis is a relatively virulent species that has a tropism for causing deep tissue invasion. C. tropicalis is generally sensitive to antifungals including fluconazole.19,20 C. parapsilosis is a less virulent species seen frequently in neonates and in adults with central venous catheters. Although C. parapsilosis is less virulent, many isolates form thick biofilms on prosthetic materials and catheters that make the organism difficult to eradicate.19,20 C. parapsilosis is generally susceptible to most antifungals including fluconazole. However, higher dosages of echinocandins (e.g., 70–100 mg/day of caspofungin) have been suggested due to the decreased potency of the echinocandin class against this species. C. krusei is a less-common species associated with breakthrough infections in heavily immunocompromised patients and should always be considered resistant to fluconazole.19,20 Interestingly, most fluconazole-resistant isolates of C. krusei retain susceptibility to itraconazole and voriconazole, based on laboratory analysis.
C. glabrata has become a common cause of both de novo candidemia in heavily immunocompromised hosts and breakthrough infection in patients on fluconazole prophylaxis. Although C. glabrata is less virulent than other Candida species, infections with this organism are typically seen in patients with poor performance status, therefore mortality remains high. The marginal susceptibility of C. glabrata to fluconazole dictates that other agents such as amphotericin B or the echinocandins be considered as first-line therapy until susceptibility to fluconazole can be documented.19,20 The effectiveness of voriconazole or posaconazole for fully fluconazole-resistant C. glabrata fungemia is not well established, and cross-resistance among these triazole antifungals has been documented in laboratory studies.21,22
TREATMENT
Six antifungals (amphotericin B, fluconazole, voriconazole, caspofungin, micafungin, and anidulafungin) have been studied as montherapy in prospective, randomized comparative clinical trials for the treatment of invasive candidiasis.19,23–27 While these treatment options are considered to have relatively equivalent efficacy, they differ somewhat in toxicity and associated drug-drug interaction profiles. Lipid amphotericin B formulations are probably as effective as the aforementioned agents; however, evidence supporting their use is derived principally from open-label observational studies and empiric therapy trials of febrile neutropenia.19 Moreover, the acquisition cost of lipid amphotericin B formulations is relatively higher compared to fluconazole and the echinocandins. As a result, their first-line use is not prominently recommended in evidence-based guidelines for proven infection.20 No prospective randomized, controlled clinical trials have been published comparing antifungal therapies for proven acute disseminated candidiasis in neutropenic patients, chronic disseminated candidiasis, or other forms of deep-organ candidiasis.19
Patient Encounter 2, Part 2
Selecting Antifungal Therapy
The patient is started on fluconazole 400 mg/day, but 3 days later has persistent fever and develops hypotension and decreased urine output. Blood cultures reveal a germ tube-negative yeast growing in the blood. Laboratory studies revealed a WBC of 12.3 × 103/mm3 (12.3 × 109/L), aspartate aminotransferase 68 IU/L (1.13 μKat/L), alanine aminotransferase 75 IU/L (1.25 μKat/L), alkaline phosphatase 168 IU/L (2.8 μKat/L), and normal bilirubin. Serum creatinine is 1.8 mg/dL (159 μmol/L).
What factors suggest empiric antifungal therapy should be changed in this patient?
What are the most likely fungal species growing from the blood?
What other procedures should be recommended in this patient to improve response to antifungal therapy?
A majority of patients are treated empirically for invasive candidiasis before conclusive evidence of infection is available to direct therapy. Empiric therapy for invasive candidiasis should be considered in any patient with persistent, unexplained fever and host deficits that predispose patients to candidemia, including broad-spectrum antibacterial therapy, presence of a central venous catheter, patients with severe organ dysfunction or on dialysis, patients with neutropenia or qualitative deficiencies in host immunity (e.g., due to high-dose corticosteroid therapy), or colonization with Candida at one or more body sites. If a patient is non-neutropenic, clinically stable (i.e., normotensive with relatively normal organ function), and has never received prior azole therapy, fluconazole 800 mg/day (12 mg/kg/day) is an appropriate first-line therapy for invasive candidiasis until speciation of the Candida isolate is confirmed.19 Echinocandins (caspofungin, micafungin, anidulafungin) are preferred as first-line agents in more critically ill patients with compromised renal function, hypotension/sepsis, or in institutions/ICUs with relatively high rates (greater than 10%) of C. glabrata or C. krusei (Table 84–3).19 Treatment is continued for at least 2 weeks or longer for complicated infections (endovascular source, metastatic seeding).
One caveat is that cryptococcoisis or endemic fungi occasionally produce fungemia in lymphopenic patients that initially misidentified as Candida. Therefore, initial treatment with a lipid amphotericin B formulation may be judicious in profoundly lymphopenic patients (i.e., CD4+ less than 250/mm3) with yeast in blood cultures until fungal identification is confirmed, as echinocandins have poor activity against non-Candida yeast. Timely initiation of appropriate antifungal therapy is critical as any delay in the initiation of antifungal therapy once a patient has a positive blood culture significantly increases mortality and the potential for metastatic infections.28
Amphotericin B deoxycholate 0.7 mg/kg/day or an echinocandin (i.e., caspofungin, micafungin, or anidula-fungin); voriconazole; or a lipid amphotericin B formulation are recommended as empiric therapy in patients with neutropenic (i.e., absolute neutrophil count less than 500 PMN/mm3) fever. Other newer echinocandins (micafungin and anidulafungin) are effective alternatives to caspofungin. If the neutropenia is of shorter duration and the patient is at lower risk for mold infections (e.g., solid tumor patients with 2 weeks or less of neutropenia), higher-dose fluconazole (i.e., 800 mg/day or 12 mg/kg) could be considered.19 Lipid amphotericin B formulations, an echinocandin, or voriconazole are often the preferred agents in febrile patients with 3 weeks or more of neutropenia to expand coverage against molds.19,29If the yeast is identified as C. glabrata, published treatment guidelines recommend the use of either caspofungin or possibly amphotericin B formulations.19 Voriconazole is not recommended until further clinical data support its use for C. glabrata, as laboratory studies have suggested potential cross-resistance with fluconazole-resistant isolates of this species. Fluconazole can also be considered for C. glabratainfections if the isolate is documented to be susceptible or susceptible-dose-dependent to fluconazole.19 C. krusei infections can be treated with an echinocandin, amphotericin B, or voriconazole. Patients who respond to therapy are medically stable, are not neutropenic, and are taking oral medications can be transitioned to oral fluconazole provided isolate susceptibility is documented by minimum inhibitory concentration testing.
For uncomplicated catheter-related candidemia (no evidence of organ involvement), therapy should be continued for at least 2 weeks from the last positive blood culture.19 In neutropenic patients, therapy should be continued until resolution of neutropenia.19 Whenever possible, central venous catheters should be removed to decrease the duration of fungemia and risk of recurrent infections.19
Treatment recommendations for other forms of invasive candidiasis are based primarily on anecdotal evidence and expert opinion. Deep-organ candidiasis requires prolonged therapy to achieve a cure; therefore, importance is placed on the use of convenient and nontoxic long-term treatment regimens. Fluconazole (400 mg/day or 6 mg/kg/day) is the preferred regimen in clinically-stable patients. Amphotericin B, lipid amphotericin B formulations, and possibly caspofungin can be considered for refractory cases or clinically unstable patients. Infections of the eye, bone, pancreas, or gallbladder are typically treated with either amphotericin B or fluconazole; however, there are few data to support the use of echinocandins. Urinary candidiasis is an ill-defined group of syndromes that can range from benign colonization (candiduria) to invasive disease of the renal parenchyma. Asymptomatic non-neutropenic patients with candiduria do not require antifungal therapy, as no study has demonstrated the value of transiently clearing Candida from the urine. Patients should receive 7 to 14 days of antifungal therapy for urinary candidiasis if they are: (a) symptomatic; (b) have clinical or laboratory evidence of infection; (c) are neutropenic; (d) are low-birth-weight infants; (e) will undergo urologic manipulations; or (f) have renal allografts. Removal of urinary tract instruments, including Foley catheters and stents, is recommended whenever possible. The preferred therapy is fluconazole 200 mg daily, although IV amphotericin B deoxycholate 0.3 to 1 mg/kg/day is also effective. Other antifungal agents do not achieve appreciable concentrations in the urine and therefore should not be considered for urinary candidiasis. Irrigation with amphotericin B is not effective for infections above the bladder and should not be used in higher-risk patients with the exception of its use as a diagnostic tool for confirming a localized infection of the bladder. Candida infections of the renal parenchyma secondary to metastatic seeding from the bloodstream are treated in a similar fashion to candidemia.
Mucocutaneous candidiasis is generally not life threatening nor invasive and can be treated with topical azoles (clotrimazole troches), oral azoles (fluconazole, ketoconazole, or itraconazole), or oral polyenes (such as nystatin or oral amphotericin B). Orally administered and absorbed azoles (ketoconazole, fluconazole, or itraconazole solution), amphotericin B suspension, IV echinocandins, or IV amphotericin B are recommended for refractory or recurrent infections.19
Table 84–3 Therapeutic Approach to Opportunistic Fungal Infections in Adults
Although more invasive, esophageal candidiasis does not typically evolve into a life-threatening infection. However, topical therapy is ineffective. Azoles (fluconazole, itraconazole solution, or voriconazole), echinocandins, or IV amphotericin B (in cases of unresponsive infections) are effective treatment options. Parenteral therapy should be used in patients who are unable to take oral medications.19
PATIENT MONITORING AND SIDE EFFECTS
Response to antifungal therapy in invasive candidiasis is often more rapid than for endemic fungal infections. Resolution of fever and sterilization of blood cultures are indications of response to antifungal therapy. Toxicity associated with antifungal therapy is similar in these patients as described earlier with the caveat that some toxicities may be more pronounced in critically-ill patients with invasive candidiasis. Nephrotoxicity and electrolyte disturbances, with amphotericin B in particular, are problematic and may not be avoidable even with lipid amphotericin B formulations. Therefore, there is a growing emphasis on the first-line use of fluconazole in lower-risk patients and echinocandins in higher-risk patients to reduce the potential for patient adverse effects. Decisions to use one class of antifungal agents over the other are principally driven by concerns of nonalbicans species, patient tolerability, or history of prior fluconazole exposure (risk factor for nonalbicans species.).
PROPHYLAXIS
Fluconazole (400 mg/day) has been extensively studied as a prophylactic regimen to prevent invasive candidiasis in patients with prolonged (greater than 2 weeks) neutropenia.29–34 Placebo-controlled, prospective randomized trials performed in the 1990s demonstrated that fluconazole was effective in reducing the frequency, morbidity, and in some trials mortality, due to invasive candidiasis when administered until marrow recovery. However, the major limitation with fluconazole is its lack of mold coverage needed for high-risk patients with persistent neutropenia. Studies have examined the use of itraconazole, voriconazole, posaconazole, or the echinocandin micafungin as prophylaxis in hematopoietic cell transplant recipients until engraftment to provide protection against both Candida and Aspergillus species.34–37 Although all antifungals studies have demonstrated a benefit in reducing fungal infections, all of the drugs have limitations with respect to prolonged administration in high-risk patients. Therefore, the approach toward antifungal prophylaxis is highly institution-specific depending on the patient population, epidemiology of invasive fungal infections, and options for outpatient IV drug therapy.
Use of antifungal prophylaxis for invasive candidiasis in the non-neutropenic patients remains an area of controversy. Prophylaxis should be targeted toward clearly-defined high-risk transplant populations (e.g., liver, pancreatic, or small-bowel transplantation) or ICU patients (i.e., neonatal intensive care) with rates of invasive candidiasis exceeding 10%, despite aggressive infection-control procedures.38,39Fluconazole prophylaxis (400 mg/day) reduces the rate of Candida peritonitis in patients with refractory GI perforation and trended toward decreased rates of invasive candidiasis in select adult patients admitted to a surgical ICU for more than 3 days.38,39 Fluconazole is also effective in reducing the rate of invasive candidiasis in neonates.36 However, prophylaxis can result in excessive antifungal use in lower-risk patients; therefore, many experts have advocated preemptive (i.e., starting therapy based on biomarkers of infection) or empirical (symptoms of infection) treatment approaches in this population in lieu of prophylaxis. A recently completed multi-institutional prospective randomized trial of administering empirical fluconazole (800 mg/day vs. placebo) in ICU patients with persistent fever, however, did not demonstrate a benefit for in the non-neutropenic population.40 It is hoped that ongoing studies will better define the risks versus benefits of routine antifungal prophylaxis or preemptive treatment approaches in the ICU setting.
CRYPTOCOCCOSIS
EPIDEMIOLOGY
Cryptococcus neoformans is an encapsulated yeast that can infect apparently normal hosts but is more frequently associated with severe infections in immunocompromised patients. C. neoformans is divided into two varieties based on serotype: C. neoformans var. neoformans (serotypes a and d) and C. neoformans var. gatti (serotypes b and c). C. neoformans var. gatti is found predominantly in tropical and subtropical climates associated with eucalyptus trees, whereas C. neoformans var. neoformans is found worldwide and is associated with pigeon droppings and other avian excreta. Before the AIDS pandemic, cryptococcosis was a relatively uncommon disease but became a leading cause of meningitis among HIV-infected patients. Although the incidence of this infection has declined somewhat with the widespread use of highly active antiretroviral therapy (HAART), C. neoformans remains an important pathogen in immunocompromised patients, including cancer patients who often present with the pulmonary form of the infection.
PATHOGENESIS AND CLINICAL PRESENTATION
C. neoformans is acquired primarily through inhalation of the desiccated yeast particles found in the environment. Inhaled cells reach distal alveolar spaces where they gradually rehydrate and form their characteristic polysaccharide capsules that enable resistance to phagocytosis. Defects in cellular immunity allow reconstitution of the protective capsule and multiplication of yeast in the lungs. Although alveolar macrophages phagocytose the yeast, containment and killing requires a coordinated response between innate and adaptive humoral (complement and anticryptococcal antibodies) and T-cell-mediated host responses.5 Deficiencies in cell-mediated immunity allow the yeast to survive as a facultative intracellular pathogen in macrophages as they migrate from the lung to draining lymph nodes, leading to dissemination via the bloodstream to the meninges.
Patient Encounter 3
Invasive Mold Infection
A 40-year-old female with acute myelogenous leukemia at day 115 post matched-allogeneic donor hematopoietic stem cell transplantation presents to the clinic with increasing complaints of nausea, stomach cramping, and rash on the hands spreading up her arms. She also complains of pain upon deep inspiration. By laboratory examination, she is noted to have an alanine aminotransferase of 85 IU/L (1.42 μKat/L), aspartate aminotransferase 75 IU/L (1.25 μKat/L) and total bilirubin of 2.1 mg/dL (36 μmol/L). Her current medications include tacrolimus 5 mg twice daily (most recent level: 8 ng/mL [8 mcg/L]), levofloxacin 500 mg daily, fluconazole 200 mg/day, valacyclovir 500 mg twice daily, metoprolol 25 mg twice daily, and benzonatate (tessalon) pearls. She is admitted to the hospital for suspected graft-versus-host disease exacerbation. CT scan of the chest reveals three to four dense pleural base nodules in both lung fields. The primary service wishes to start voriconazole.
What are the patient’s risk factors for developing an invasive mold infection?
Is voriconazole an acceptable option in this patient? Are there any drug interaction concerns?
Unlike most opportunistic fungi, true virulence factors have been identified for C. neoformans. The capsules, including the soluble polysaccharides released from the yeast cells during infection, impair phagocytosis and binding of anticryptococcal antibodies. Primary cryptococcal infection begins in the lung, presenting as a mildly symptomatic or asymptomatic infection that resolves spontaneously or results in an encapsulated, usually noncalcified lung nodule. It is common for these isolated nodules to be detected on chest x-rays during routine workup. Diagnosis of primary cryptococcosis is only made if the nodule is aspirated or removed because of concerns of primary lung cancer.
In the immunocompromised host, infection of the lung may present with more diffuse, bilateral, and interstitial disease that mimics the presentation of Pneumocystis jiroveci (carinii) pneumonia (PCP). Dissemination to other organs, particularly the CNS, eye, and possibly the skin, is more likely to occur in patients with severe deficits in cell-mediated immunity. Fever, cough, dyspnea, and pleural pain are common at presentation with accompanying hypoxemia that can rapidly evolve to acute respiratory failure. Because of the features of diffuse pulmonary cryptococcosis overlap with other opportunistic pathogens, early diagnosis requires bronchoalveolar lavage or transbronchial biopsy, which can effectively diagnose 80% to 100% of cases.41 The clinical course of diffuse cryptococcal pneumonia can be as severe as PCP, with mortality rates approaching 100% in untreated patients by 48 hours.
C. neoformans is strongly neurotropic and readily disseminates from the lung to the CNS, specifically the leptomeninges, and occasionally the parenchyma of the brain. The clinical characteristics of cryptococcal meningitis differ somewhat, however, between patients with and without underlying AIDS. In patients without AIDS, disease presentation is more insidious and symptoms such as dizziness, irritability, decreased comprehension, and unstable gait may present many weeks to months before the diagnosis is established.41 Patients with AIDS generally present much later in the course of disease with severe meningoencephalitis.37 The most common signs and symptoms on presentation are fever, headache, meningismus, photophobia, mental status changes, and seizures. CT or more sensitive MRI may reveal cerebral edema, multiple areas of enhanced nodules, or a single mass lesion (cryptococcoma). Examination of the cerebrospinal fluid (CSF) often reveals increased opening pressure upon lumbar puncture, but glucose, protein, and leukocyte levels can be normal.41
LABORATORY DIAGNOSIS
Clinical diagnosis is confirmed by cultures from the blood, CSF, or other clinically relevant fluids or tissue. However, early diagnosis is suggested by direct observation of C. neoformans in the CSF by India ink staining (Fig. 84–3B).41 Similarly, detection of cryptococcal antigen in either serum or CSF can provide a rapid diagnosis with greater than 95% sensitivity and specificity and appears to correlate with fungal burden.41 A positive serum antigen test of greater than 1:4 strongly suggests cryptococcal infection, and greater than or equal to 1:8 is indicative of active disease. Antigen titers in serum are positive in 99% of patients with cryptococcal meningitis and typically exceed titers of 1:2,048 in patients with AIDS.41 However, the time course of cryptococcal antigen elimination is unknown, and a positive test result can persist for many years. Changes in the CSF cryptococcal antigen titers have limited value in the monitoring of drug therapy for cryptococcal meningitis, although it is expected that a decrease should be seen after two or more weeks of antifungal therapy.41
TREATMENT
Cryptococcal meningitis is fatal if left untreated. Because pneumonia frequently precedes dissemination of disease and subsequent meningitis, all patients with culture-, histopathology-, or serology-proven disease should receive antifungal therapy. In patients with isolated pulmonary cryptococcosis, fluconazole is generally considered to be the therapy of choice (see Table 84–2).41 Alternatively, itraconazole, voriconazole, or combination therapy (fluconazole plus flucytosine) has also been used with some success but these regimens are generally considered inferior to amphotericin B and are recommended only for persons unable to tolerate or unresponsive to standard treatment. Echinocandins do not have clinically useful activity against C. neoformans.
Disseminated or CNS cryptococcosis requires a more aggressive treatment approach. Pretreatment predictors of poor outcome with antifungal therapy include:
• Progressive underlying disease or immunodysfunction
• Abnormal mental status at the time of presentation
• Increased opening pressure on lumbar puncture (greater than 260 mm H2O)
• High fungal burden as reflected by a CSF antigen titer (in AIDS patients) of greater than 1:2,048
Prospective randomized trials completed prior to the recognition of AIDS demonstrated high response rates (approximately 80%) with the combined use of amphotericin B and flucytosine for 4 to 6 weeks. Although sterilization of the CSF could be achieved in most patients within 2 weeks with this regimen, a substantial number of patients (30-40%) developed dose-limiting toxicities and relapse was seen in roughly 50% of patients. Therefore, a treatment approach was devised that consisting of distinct treatment phases to minimize toxicity and reduce the risk of relapse. Clinical trials performed by the National Institute of Allergy and Infectious Diseases (NIAID) Mycoses Study Group after the recognition of AIDS showed that 2 weeks of induction antifungal therapy with combination amphotericin B (0.7 mg/kg/day) plus flucytosine (100 mg/kg/day) for cryptococcal meningitis, followed by consolidation therapy with fluconazole (400 mg daily) for 8 weeks was as effective as 4 weeks of combination therapy, with fewer toxicities (see Table 84–2).12,41 Other studies have suggested fluconazole plus amphotericin may be an acceptable option in patients who cannot tolerate therapy with flucytosine (see Table 84–2).12
PROPHYLAXIS
Fluconazole (200 mg/day) is recommended as maintenance therapy for life in patients with persistent underlying immune dysfunction to prevent recurrent cryptococcal meningitis.12 Available data have demonstrated that it is safe to discontinue maintenance therapy in AIDS patients who have had a sustained immunologic response on effective antiretroviral therapy.12 Occasionally, initiation of HAART can result in the reactivation of a subclinical, immunologic manifestation of cryptococcal infection (or other opportunistic infections). Manifestations of this so-called immune reconstitution syndrome (IRIS) may include exacerbations of meningitis or necrotizing pneumonia. Antifungal therapy plus a nonsteroidal anti-inflammatory agent or prednisone have been used successfully in patients with cryptococcal-associated immune reconstitution syndrome.12 However, the optimal management of this recently defined clinical entity remains unknown.
INVASIVE ASPERGILLOSIS
EPIDEMIOLOGY
Invasive molds, particularly Aspergillus, have become an increasingly important complication of cancer therapy and organ transplantation. Patients with acute leukemia and recipients of allogeneic hematopoietic cell transplants are at especially high risk for invasive aspergillosis (IA) due to prolonged neutropenia and deficiencies in cell-mediated immunity associated with graft-versus-host disease and its treatment. More than 180 species within the genus Aspergillus have been described, but only four species are commonly associated with invasive infection: Aspergillus fumigatus, Aspergillus flavus, Aspergillus terreus, and Aspergillus niger. Of these four species, A. fumi gatusaccounts for most of human infections. However, identification of Aspergillus mold in culture to the species level is still essential because the incidence of amphotericin B–resistant Aspergillus terreus and Aspergillus flavus have increased over the last 10 years among high-risk patients. Early and accurate diagnosis of IA remains the most important barrier to the effective management of this infection, which is associated with crude mortality rates ranging from 60% to 100%.42
PATHOGENESIS AND CLINICAL PRESENTATION
The pathogenesis of IA is defined largely by the underlying immune dysfunction of the host. The most common route of acquisition for Aspergillus is through the respiratory tract. Conidia dispersed in air currents are continuously inhaled through the sinuses and mouth and penetrate down to distal alveolar spaces (see Fig. 84–3C). Most conidia are rapidly phagocytosed and removed by resident macrophages and neutrophils in the upper and lower respiratory tract.5 However, macrophage function may be suppressed following transplantation, cytotoxic chemotherapy, or in patients who have received high-dose corticosteroid therapy. Conidia that escaped phagocytosis begin to germinate into hyphal forms that are too large for ingestion by macrophages. Hyphal forms of Aspergillus then invade blood vessels or contiguous tissues or bone (in sinuses) resulting in hemorrhage and/or infarction, and coagulative necrosis. Once in the bloodstream, viable hyphal fragments can break off and disseminate to distal organs including the brain. Control of the infection at this stage requires development of an adaptive Th-1 type response to enhance the fungicidal activity of professional effector cells (i.e., neutrophils) against hyphal elements.5 Patients with dysregulated, suppressed T-cell-mediated immunity, or prolonged neutropenia are unable to control the infection and are at high risk for dissemination of the infection. Without antifungal therapy, IA in immunosuppressed patients is uniformly fatal.
Signs and symptoms of IA are predictably muted in the immunocompromised host. Fever is common but nonspecific for infection and may be accompanied by pleuritic chest pain, cough, hemoptysis, and/or friction rub.40Neurologic signs including seizures, hemiparesis, and stupor may be present in patients with dissemination to the brain. Cutaneous plaques or papules characterized by a central necrotic ulcer or eschar occur in up to 10% of patients with disseminated disease; however, concomitant blood cultures are often negative. Chest radiographs cannot detect early forms of disease and may remain negative in up to 10% of patients within 1 week of death.43,44 Nodular lesions detected by high-resolution computed tomography (HRCT) scans are often the first indication of invasive pulmonary aspergillosis along with fever, and reveal small wedge-shaped or nodular lesions, typically surrounded by intermediate attenuation called the “halo sign.”43,44 These early lesions on CT scans represent hemorrhage and edema surrounding an infarcted blood vessel. Despite “effective” antifungal therapy, lesions on CT scan may continue to increase in size in neutropenic patients until neutrophil counts recover, at which time they begin to cavitate, forming the “air-crescent sign” on chest radiographs, indicative of resolving infection. Immunocompromised patients on fluconazole with progressive sinus or pulmonary disease by radiography should be considered to have a possible mold infection and receive empiric antifungal therapy directed (at minimum) against Aspergillus species.42
LABORATORY DIAGNOSIS
Like other invasive mycoses, definitive diagnosis of aspergillosis requires histopathologic evidence of hyphal invasion in tissue (Fig. 84–4). However, procedures needed to establish a definitive diagnosis by sampling of suspicious lesions (e.g., fine-needle aspiration or thoracoscopic lung biopsy) are not feasible in many patients with underlying thrombocytopenia secondary to hematologic malignancies or chemotherapy. Even if hyphae are observed in tissue, histopathology alone cannot distinguish Aspergillus from other angioinvasive septate molds such as Fusarium, which have different patterns of antifungal susceptibility.42 Therefore, respiratory and/or wound cultures (if cutaneous or sinus/hard palate lesions are present) are important factors in the modification of empiric antifungal therapy.
Respiratory cultures including sputum, bronchial washings, or bronchoalveolar lavage have a low sensitivity for diagnosis of IA but a high positive predictive value in immunocompromised patients.42Therefore, a negative bronchoalveolar lavage culture does not rule out invasive pulmonary aspergillosis, but a positive culture in a high-risk patient (e.g., allogeneic hematopoietic cell transplant patients) indicates pulmonary aspergillosis in at least 60% of such patients. Blood cultures have little diagnostic value for IA but may reflect true disease with A. terreus. Patients with limited lung involvement or on prophylactic or empiric antifungal therapy may continue to be culture-negative for Aspergillus species, despite the appearance of progressing disease.42 Therefore, clinical specialists should never consider negative cultures as an indication for stopping antifungal therapy in patients with suspected or proven aspergillosis.
Considerable effort has been focused in the last decade to develop nonculture based laboratory methods (antigen detection, polymerase chain reaction [PCR], and metabolite detection) for the diagnosis of IA. The hope is that these surrogate tests could detect early evidence of Aspergillus infection before significant target organ damage eventually detected by CT scans occurs. The FDA has approved an ELISA-based assay for the detection of a polysaccharide component of the Aspergillus cell wall called galactomannan. Although several large prospective studies have found that the sensitivity and specificity of the assay exceeded 90% in neutropenic patients with hematologic malignancies, the median time span between galactomannan detection and clinical signs and symptoms of IA averages less than 6 days.42 Other factors such as patient immune status (neutropenia versus graft-versus-host disease), antifungal therapy, and diet may affect the interpretation of the galactomannan test.42 For example, false-positive results have been reported in pediatric patients, patients receiving piperacillin-tazobactam for neutropenic fever, and following the ingestion of certain cereals, pastas, nutritional supplements, or soy sauce.42 Hence, there are numerous opportunities for false-positive tests. Although some animal studies and clinical data suggest that rising galactomannan levels are a harbinger of breakthrough infection, there are still limited data supporting the use of this test to guide and monitor antifungal therapy. At this time it appears that the galactomannan test (and other nonculture based strategies) will serve as complementary methods to confirm results from microbiologic, histopathologic, and radiographic investigations directed toward diagnosing IA.
TREATMENT
To date, only two comparative randomized controlled clinical trials have evaluated antifungal therapies for the treatment of diagnosis-proven IA and only one study was sufficiently powered to measure differences in response to antifungal therapy.45 In that study, unblinded investigators compared patients initially randomized to the newer triazole, voriconazole, to patients initially treated with amphotericin B deoxycholate.45 The study design was unique in that it allowed a change from the randomized drug to any other licensed antifungal therapy, without requiring that the patient be classified as a treatment failure on randomized therapy. Nearly 80% of patients randomized to receive amphotericin B deoxycholate were switched to other licensed antifungal therapies (mean duration 10 days) versus 36% of patients in the voriconazole arm (mean duration 77 days). The poor tolerability of amphotericin B deoxycholate was not surprising given the relatively higher dosages (1 mg/kg/day IV) and prolonged treatment courses required in the treatment of IA. At the end of the study (12 weeks), a higher proportion of patients in the voriconazole arm remained alive (70.8%) compared to amphotericin B deoxycholate-treated patients (57.9%). On the basis of these results, many experts now consider voriconazole as the initial drug of choice for IA in patients without significant contraindications (e.g., drug interactions or pre-existing liver dysfunction) to azole therapy (see Table 84–2).42 Voriconazole also appears to have some efficacy in CNS aspergillosis, a form of IA with historical mortality rates approaching 100%. Itraconazole has activity against Aspergillus and is frequently used as prophylaxis, but is not considered a particularly effective treatment option for invasive disease.42
FIGURE 84–4. Pathogenesis of invasive aspergillosis (IA).
Lipid formulations of amphotericin B, echinocandins, or posaconazole can be considered as possible alternatives to voriconazole therapy and may be preferred agents in patients with breakthrough infection on azole antifungals (including itraconazole or fluconazole). Several recent open-label case series have suggested that combination therapy, with an echinocandin and mold-active triazole such as voriconazole, may be more effective than voriconazole alone for IA that has failed amphotericin B–based therapy.42 Once antifungal therapy has begun, the duration and intensity of antifungal therapy is directed by host-specific factors including clinical response, underlying immunosuppression, tolerability, and plans for future chemotherapy/immunosuppression. In the heavily immunocompromised patient, complete eradication of the fungus is unlikely and suppressive therapy may be required until well after recovery of cellular immune function. Reactivation from residual infarcts or devitalized tissue in the sinus or lung harboring aspergillosis is a concern if the patient will receive further immunosuppressive therapy. Therefore, surgical débridement of the sinus or excision of large lung lesions is often pursued if the patient is not profoundly thrombocytopenic. Relapsing or breakthrough Aspergillus infections respond less favorably to antifungal therapy than de novo IA and may require more aggressive measures (combination therapy, immunotherapy, or surgery) to stabilize the infection.
PROPHYLAXIS
Although recently published guidelines for preventing opportunistic infections in hematopoietic cell transplant recipients do not provide concrete recommendations for antifungal prophylaxis against Aspergillus, prophylaxis should be considered in certain high-risk subgroups with rates of IA exceeding 10%. These groups include: (a) patients with prolonged pre-engraftment periods (e.g., cord-blood transplant recipients), (b) patients with a history of IA prior to transplantation, (c) patients receiving transplants with a high risk of graft-versus-host disease (e.g., haploidentical allogeneic transplant) or infection (e.g., T-cell–depleted transplant), any patient with graft-versus-host disease on high-dose corticosteroid therapy (greater than 1 mg/kg prednisone equivalent) with or without antithymocyte globulin or tumor necrosis factor blockade (i.e., infliximab), and (d) any patient transplanted with active cytomegalovirus disease, which is associated with an increased risk of subsequent mold infections due to the immunosuppressive effect of the virus. Posaconazole was shown in two prospective randomized trials to reduce Aspergillus-associated death in patients with acute high-risk leukemia and reduce mold infections in patients with graft-versus-host disease following hematopoetic stem cell transplantation.35,36 Similar data are available for voriconazole,37 but less benefit was observed versus standard fluconazole prophylaxis in the hematopoetic stem cell transplant patients. Prophylactic approaches, however, are often highly institution and patient specific.
PATIENT MONITORING AND SIDE EFFECTS
Response to antifungal therapy in invasive molds is slow and difficult to judge by clinical signs alone. Resolution of fever and eventual clearing of CT scans (in the case of lung infections) are indications of response to antifungal therapy. Toxicity associated with antifungal therapy is similar in these patients as in those described earlier. In addition to the adverse effects mentioned earlier, voriconazole may cause transient visual changes (photopsia) in approximately one-third of patients with the first few doses of therapy. Occasionally, these visual disturbances are accompanied by hallucinations and may require discontinuation of therapy. Voriconazole and posaconazole exhibit wide intrapatient and interpatient pharmacokinetic variability due to variable absorption and metabolism, respectively.42 Therefore, many experts advocate therapeutic drug monitoring in patients with documented disease receiving the drugs as monotherapy, or in patients with suspected progression (voriconazole, posaconazole) or toxicities (voriconazole) while on therapy. Although the therapeutic ranges are not well established, improved responses in patients receiving posaconazole as salvage therapy have been noted when plasma concentrations 3 to 4 hours after the oral dose approached 1 mcg/mL (1 mg/L).46 For voriconazole, patient response rates are improve if steady-state trough concentrations surpass 1 mcg/mL (1 mg/L); however, the risk of CNS toxicity increases as plasma concentrations increase above 5.5 mcg/mL (5.5 mg/L).47 Patients often require prolonged therapy, particularly if they remain immunosuppressed. In many cases, antifungal therapy may be continued indefinitely until complete resolution of underlying immunosuppression.
Abbreviations Introduced in This Chapter
Self-assessment questions and answers are available at http://www.mhpharmacotherapy.com/pp.html.
REFERENCES
1. Wheat LJ, Freifeld AG, Kleiman MB, et al. Clinical practice guidelines for the management of patients with histoplasmosis: 2007 update by the Infectious Diseases Society of America. Clin Infect Dis 2007;45: 807–825.
2. Wheat LJ, Kauffman CA. Histoplasmosis. Infect Dis Clin North Am 2003;17:1–19.
3. Pappas PG. Blastomycosis. Semin Respir Crit Care Med 2004;25:113–121.
4. Galgiani JN, Ampel NM, Blair JE, et al. Coccidioidomycosis. Clin Infect Dis 2005;41:1217–1223.
5. Romani L. Immunity to fungal infections. Nat Rev Immunol 2004; 4:11–23.
6. Chapman SW, Dismukes WE, Proia LA, et al. Clinical practice guidelines for the management of blastomycosis: 2008 update by the Infectious Diseases Society of America. Clin Infect Dis 2008;46:1801–1812.
7. Lortholary O, Denning DW, Dupont B. Endemic mycoses: A treatment update. J Antimicrob Chemother 1999;43:321–331.
8. Albengres E, Le Louet H, Tillement JP. Systemic antifungal agents. Drug interactions of clinical significance. Drug Saf 1998;18:83–97.
9. Todd JR, Arigala MR, Penn RL, King J W. Possible clinically significant interaction of itraconazole plus rifampin. AIDS Patient Care STDS 2001;15:505–510.
10. Nicolau DP, Crowe HM, Nightingale CH, Quintiliani R. Rifampin-fluconazole interaction in critically ill patients. Ann Pharmacother 1995;29:994–996.
• Glasmacher A, Hahn C, Leutner C, et al. Breakthrough invasive fungal infections in neutropenic patients after prophylaxis with itraconazole. Mycoses 1999;42:443–451.
• Guidelines for Prevention and Treatment of Opportunistic Infections in HIV-Infected Adults and Adolescents - June 18, 2008. AIDSinfo June 18, 2008.
11. Gubbins PO, McConnell SA, Penzak SR. Antifungal agents. In: Piscitelli SC, Rodvold KA, eds. Drug Interactions in Infectious Diseases. Totowa, NJ: Humana Press, 2001.
12. Groll AH, Piscitelli SC, Walsh TJ. Clinical pharmacology of systemic antifungal agents: A comprehensive review of agents in clinical use, current investigational compounds, and putative targets for antifungal drug development. Adv Pharmacol 1998;44:343–499.
13. Johnson PC, Wheat LJ, Cloud GA, et al. Safety and efficacy of liposomal amphotericin B compared with conventional amphotericin B for induction therapy of histoplasmosis in patients with AIDS. Ann Intern Med 2002;137:105–109.
14. Groll AH, Giri N, Petraitis V, et al. Comparative efficacy and distribution of lipid formulations of amphotericin B in experimental Candida albicans infection of the central nervous system. J Infect Dis 2000;182:274–282.
15. Wisplinghoff H, Seifert H, Wenzel RP, Edmond MB. Current trends in the epidemiology of nosocomial bloodstream infections in patients with hematological malignancies and solid neoplasms in hospitals in the United States. Clin Infect Dis 2003;36:1103–1110.
16. Rex JH, Bennett JE, Sugar AM, et al. A randomized trial comparing fluconazole with amphotericin B for the treatment of candidemia in patients without neutropenia. N Engl J Med 1994;331:1325–1330.
17. Pappas PG, Rex JH, Sobel JD, et al. Guidelines for treatment of candidiasis. Clin Infect Dis 2004;38:161–189.
18. Rex JH, Pfaller MA, Walsh TJ, et al. Antifungal susceptibility testing: Practical aspects and current challenges. Clin Microbiol Rev 2001;14:643–658.
19. Pfaller MA, Diekema DJ, Messer SA, et al. Activities of fluconazole and voriconazole against 1,586 recent clinical isolates of Candida species determined by broth microdilution, disk diffusion, and Etest methods: Report from the ARTEMIS global antifungal susceptibility program, 2001. J Clin Microbiol 2003;41:1440–1446.
20. Pfaller MA, Messer SA, Boyken L, et al. In vitro activities of voriconazole, posaconazole, and fluconazole against 4,169 clinical isolates of Candida spp. and Cryptococcus neoformans collected during 2001 and 2002 in the ARTEMIS global antifungal surveillance program. Diagn Microbiol Infect Dis 2004;48:201–205.
21. Mora-Duarte J, Betts R, Rotstein R, et al. Comparison of caspofungin and amphotericin B for invasive candidiasis. N Engl J Med 2002;347: 2020–2029.
22. Kullberg BJ, Sobel JD, Ruhnke M, et al. Voriconazole versus a regimen of amphotericin B followed by fluconazole for candidaemia in non-neutropenic patients: A randomised non-inferiority trial. Lancet 2005;366:1435–1442.
23. Kuse E, Chetchotisakd P, da Cunha C, et al. Micafungin versus liposomal amphotericin B for candidemia and invasive candidiasis: A phase III randomized double-blind trial. Lancet 2007;369: 1519–1527.
24. Pappas P, Rotstein C, Betts R, et al. Micafungin versus caspofungin for the treatment of candidemia and other forms of invasive candidiasis. Clin Infect Dis 2007;45:883–893.
25. Reboli A, Rotstein C, Pappas P, et al. Anidulafungin versus fluconazole for invasive candidiasis. New Eng J Med 2007;356:2472–2482.
26. Morrell M, Fraser V, Kollef M. Delaying the empiric treatment of Candida bloodstrem infection until positive blood culture results are obtained: A potential risk factor for hospital mortality. Antimicrob Agent Chemother 2005;49:3640–3645.
27. Hughes WT, Armstrong D, Bodey GP, et al. 2002 guidelines for the use of antimicrobial agents in neutropenic patients with cancer. Clin Infect Dis 2002;34:730–751.
28. Goodman JL, Winston DJ, Greenfield RA, et al. A controlled trial of fluconazole to prevent fungal infections in patients undergoing bone marrow transplantation. N Engl J Med 1992;326:845–851.
29. Winston DJ, Chandrasekar PH, Lazarus HM, et al. Fluconazole prophylaxis of fungal infections in patients with acute leukemia. Results of a randomized placebo-controlled, double-blind, multicenter trial. Ann Intern Med 1993;118:495–503.
30. Slavin MA, Osborne B, Adams R, et al. Efficacy and safety of fluconazole prophylaxis for fungal infections after bone marrow transplantation—a prospective, randomized, double-blind study. J Infect Dis 1995;171:1545–1552.
31. Winston DJ, Maziarz RT, Chandrasekar PH, et al. Intravenous and oral itraconazole versus intravenous and oral fluconazole for long-term antifungal prophylaxis in allogeneic hematopoietic stem-cell transplant recipients—a multicenter, randomized trial. Ann Intern Med 2003;138:705–713.
32. Van Burik J, Ratanatharathorn V, Lipton J, et al. Randomized, double-blind trial of micafungin versus fluconazole for prophylaxis of invasive fungal infections in patients undergoing hematopoietic stem cell transplant. Clin Infect Dis 2004;39:1407–1416.
33. Cornely O, Maertens J, Winston D, et al. Posaconazole versus fluyconazole or itraconazole prophylaxis in patients with neutropenia. N Eng J Med 1007;356:348–359.
34. Ullman A, Lipton J, Vesole D, et al. Posaconazole or fluconazole for prophylaxis in severe graft versus host disease. N Eng J Med 2007; 356:335–347.
35. Wingard J, Carter S, Walsh T, et al. Results of a randomized, double-blind trial of Fluconazole (FLU) vs. Voriconazole (VORI) for the prevention of invasive fungal infections (IFI) in 600 allogeneic blood and marrow transplant (BMT) patients. Blood 2007;110: Abstract #163.
36. Pelz RK, Hendrix CW, Swoboda SM, et al. Double-blind placebo-controlled trial of fluconazole to prevent candidal infections in critically ill surgical patients. Ann Surg 2001;233:542–548.
37. Rex JH, Sobel JD. Prophylactic antifungal therapy in the intensive care unit. Clin Infect Dis 2001;32:1191–1200.
38. Shuster M, Edward J, Sobel J, et al. Empirical fluconazole versus placebo for intensive care unit patients: A randomized trial. Ann Intern Med 2008;149:83–90.
39. Saag MS, Graybill RJ, Larsen RA, et al. Practice guidelines for the management of cryptococcal disease. Infectious Diseases Society of America. Clin Infect Dis 2000;30:710–718.
40. Walsh TJ, Anaissie EJ, Denning DW, et al. Treatment of aspergillosis: Clinical Practice Gudelines of the Infectious Diseases Society of America. Clin Infect Dis 2008;46:327–360.
41. Caillot D, Mannone L, Cuisenier B, Couaillier JF. Role of early diagnosis and aggressive surgery in the management of invasive pulmonary aspergillosis in neutropenic patients. Clin Microbiol Infect 2001;7:54–61.
42. Caillot D, Casasnovas O, Bernard A, et al. Improved management of invasive pulmonary aspergillosis in neutropenic patients using early thoracic computed tomographic scan and surgery. J Clin Oncol 1997;15:139–147.
43. Herbrecht R, Denning DW, Patterson TF, et al. Voriconazole versus amphotericin B for the primary treatment of aspergillosis. New Eng J Med 2002;347:408–415.
44. Walsh T, Raad I, Patterson T, et al. Treatment of invasive aspergillosis with posaconazole in patients who are refractory to or intolerant of conventional therapy: An externally controlled trial. Clin Infect Dis 2007;44:2–12.
45. Pascual A, Calandra T, Bolay S. Voriconazole therapeutic drug monitoring in patients with invasive mycoses improves safety and efficayc outcomes. Clin Infect Dis 2007;46:201–211.
46. Lionakis MS, Kontoyiannis DP. Fusarium infections in critically ill patients. Semin Respir Crit Care Med 2004;25:159–169.49.
47. Kontoyiannis DP, Lionakis MS, Lewis RE, et al. Zygomycosis in a tertiary-care cancer center in the era of Aspergillus-active antifungal therapy: A case-control observational study of 27 recent cases. J Infect Dis 2005;191:1350–1360.