Cryptococcosis is the cause of the most common life-threatening meningitis in AIDS. Early in the epidemic, approximately 5-8% of patients with AIDS developed cryptococcal infection. Where effective antiretroviral treatment (ART) is available, the incidence of cryptococcosis, along with other opportunistic infections, has decreased.(1-3) But the decrease in new cases of cryptococcosis may have started even earlier; other data suggest a decline in incidence associated with the more frequent use of azole antifungals.(4) Between 1996 and 1997, a 56.5% decrease in the incidence of cryptococcal meningitis was reported at San Francisco General Hospital, correlating with a 30% increase in the use of combination antiretroviral regimens that included a protease inhibitor.(3) However, the decrease of cryptococcosis was even more dramatic when comparing incidence in 1995 with incidence in 1996--before the introduction of protease inhibitors--suggesting that patients may have been receiving a benefit from the use of nucleoside analogues alone as well as from azoles. Furthermore, the incidence of cryptococcal meningitis in HIV-infected patients treated at San Francisco General Hospital increased 30% between 1997 and 1998 and has remained constant, reflecting the fact that there are still individuals who are not taking effective ART.(5)
Meningoencephalitis is the most frequent manifestation of cryptococcosis in HIV-infected individuals. Cryptococcal pneumonia is the most frequent fungal pneumonia encountered in persons with AIDS, except in areas hyperendemic for either histoplasmosis or coccidioidomycosis. Before the AIDS era, amphotericin B (0.3 mg/kg/day) plus flucytosine for 4-6 weeks had become standard of care for treatment of cryptococcal meningitis based on clinical trials by Bennett et al (6) and Dismukes et al.(7) This treatment was clearly suboptimal in the management of AIDS, with reported mortality rates ranging up to 30%.(8) Most deaths secondary to cryptococcal meningitis occur within the first 2 weeks of therapy, and many may be related to increased intracranial pressure.(9) A significant decrease in mortality compared with historical controls was observed in the National Institute of Allergy and Infectious Diseases (NIAID) Mycoses Study Group (MSG) and AIDS Clinical Trials Group (ACTG) trial, in which amphotericin B (0.7 mg/kg/day) was used with or without flucytosine for the initial 2 weeks of treatment followed by 8 weeks of consolidation therapy with fluconazole or itraconazole.(9)
This chapter provides a review of the microbiology, epidemiology, clinical syndromes, diagnosis, and treatment of cryptococcosis in AIDS.
| Microbiology and Pathogenesis|
Cryptococcus neoformans is an encapsulated, round-to-oval yeast measuring 4-6 microns with a surrounding polysaccharide capsule ranging in size from 1 to >30 microns when cultivated in the laboratory.(10) In its natural environment, it is smaller and poorly encapsulated. Mycelia are produced bearing basidiospores ranging from 1 to 8 microns in the perfect state, Filobasidiella neoformans. F neoformans has never been isolated either in patients or in nature.(11) Given that particles smaller than 5 microns can enter the lung, but only particles smaller than 2 microns can penetrate the lung deeply, it is postulated that transmission occurs via inhalation of the basidiospores or unencapsulated forms, leading to colonization of the airways and subsequent respiratory infection.(12,13) One study demonstrated that C neoformans could be isolated from the nasopharynx of approximately 50% of AIDS patients with cryptococcosis, whereas C neoformans was not isolated from AIDS patients without cryptococcosis, supporting inhalation as a mode of entry.(14) Although complement-mediated phagocytosis is the primary initial defense against cryptococcal invasion, the absence of an intact cell-mediated response results in ineffective ingestion and killing of the organism, leading to dissemination and increased cryptococcal burden. Given its antiphagocytic properties, the polysaccharide capsule, composed mainly of glucuronoxylomannan, is thought to be the organism's primary virulence factor. The exopolysaccharides of the capsule may contribute to virulence by suppressing the immune response, inhibiting leukocyte migration, and enhancing HIV replication.(15)
C neoformans is distinguished from other yeasts by its ability to assimilate urea, and it possesses membrane-bound phenoloxidase enzymes, which are able to convert phenolic compounds into melanin as demonstrated by certain agars such as birdseed agar. It is postulated that Cryptococcus has a propensity to invade the central nervous system (CNS) because of its ability to synthesize melanin from catecholamines that are present in this tissue in large concentrations.(16) Melanin production is seen rarely in Cryptococcus species other than C neoformans, which allows birdseed agar to be used as a valuable laboratory screening tool.(17)
There are 4 serotypes of C neoformans, designated A, B, C, and D based on antigenic determinants on the polysaccharide capsule.(16) Serotypes A and D (C neoformans var. neoformans) are the most common cause of infection, and 90% of these infections occur in immunocompromised hosts.(18) The majority of AIDS isolates are var. neoformans, even in areas where var. gattii (serotypes B and C) is endemic. Because var. neoformans is ubiquitous and predominantly affects immunocompromised hosts, despite having a high exposure rate to all hosts, it is considered an opportunistic pathogen. In contrast, var. gattii has a limited geographic distribution and affects predominantly immunocompetent male hosts in their second decade of life. The varieties can be distinguished either by serotyping or by growth characteristics on canavanine-glycine-bromthymol blue agar.(19) DNA probes are being evaluated as epidemiologic tools to assist in distinguishing relapse from reinfection.(20,21)
Cryptococcus grows readily from soil contaminated with avian excreta, particularly those of pigeons, possibly because excreta are rich in xanthine, creatinine, urea, and uric acid, all of which Cryptococcus can assimilate. It also has been isolated from nonavian sources, such as vegetables, fruits, and dairy products. In fact, Cryptococcus initially was isolated from peach juice in 1894.(19) There have been no outbreaks attributable to environmental sources and no reports of animal-to-human transmission. Human-to-human transmission is rare. There is 1 case report of a person acquiring cryptococcal endophthalmitis after receiving a transplanted cornea from a donor who had cryptococcosis, and in another case, a health care worker developed localized cutaneous cryptococcosis after autoinoculation with blood from a patient with cryptococcemia.(16,22)
Although the overall incidence of cryptococcosis is unknown, it is higher among patients with AIDS in Africa and Southeast Asia than in the United States, whereas it appears less frequently in Europe.(16) In the developed world, the introduction of potent antiretroviral therapies resulted in a decrease in the incidence of opportunistic infections associated with AIDS. In 1994, the annual prevalence of cryptococcosis was calculated to be between 6.1% and 8.5% among HIV-infected individuals in New York City.(23) Although more men are reported to develop cryptococcal disease, the male-to-female ratio essentially is 1:1, when adjusted for the male predominance in HIV infection. Cryptococcosis in children with AIDS is less common, with a prevalence of approximately 1.4%.(24) More than three fourths of the cases associated with AIDS develop when the CD4 T-lymphocyte count falls below 50 cells/µL.(25) In a 1996 retrospective review of 65 AIDS patients with cryptococcal meningitis in France, the median CD4 count was 46 cells/µL.(26) Cryptococcosis was the initial AIDS-defining illness in 63%.
A population-based active surveillance for cryptococcosis in Houston, Atlanta, San Francisco, and Alabama from 1992 to 1994 found the majority of patients to be male, consistent with the demographics of HIV infection.(4) In an initial analysis of data from Atlanta and San Francisco only, 89% of patients had CD4 counts <100 cells/µL.(4) Univariate analysis identified multiple risks for the development of cryptococcosis, including black race, injection drug use, cigarette smoking, and several environmental exposures (presumed areas where pigeon droppings accumulated). In a multivariate analysis controlling for CD4 counts, factors independently associated with the development of cryptococcosis included being in a warehouse within the previous month and sexual contact with an injection drug user. Of note, patients who had received fluconazole within 3 months of enrollment were significantly less likely to develop cryptococcosis. The analysis of all 4 areas found decreased incidence of cryptococcosis in Atlanta and San Francisco from 1992 to 1994, which was attributed to the increased use of azoles.(4) These findings are consistent with previous studies noting an increase among blacks, cigarette smokers, and individuals residing east of the Mississippi River.(25) A case-control study in a cohort of patients from Baltimore noted no differences in demographic variables, HIV risk factors, or stage of AIDS.(27) Of note, the investigators were unable to detect an increased risk of the development of cryptococcal meningitis in patients who had received short and episodic courses of steroids. A more recent review of the Kaiser Permanente medical record database from 1981 to 2000 reported similar findings, yet also noted a higher incidence of cryptococcosis in men from 1981 to 1985, but a higher incidence in women from 1986 to 1990.(28)
| Clinical Presentation|
The clinical presentation of cryptococcal disease in HIV infection can vary significantly from that in immunocompetent individuals and those with other causes of immunosuppression. Therefore, this section is devoted to the clinical presentations seen in those individuals with HIV infection.
| Pulmonary Cryptococcosis|
Although pulmonary cryptococcosis is diagnosed less frequently than meningitis in patients with AIDS, the lung is most likely the portal of entry.(29) Cryptococcal pneumonia may be either asymptomatic or symptomatic, with or without evidence of dissemination. It is unclear if disseminated disease represents a progression or reactivation of pulmonary disease because many patients have no evidence of pulmonary involvement at the time of diagnosis of disseminated disease. Given the relatively nonspecific clinical signs and symptoms, variable radiographic signs, and increased frequency of other pulmonary opportunistic infections, it is likely that cryptococcal pneumonia is underdiagnosed and not recognized until dissemination.(9) In a retrospective review, Driver et al discovered that 78% of patients with cryptococcal meningitis had evidence of pulmonary disease in the preceding 4 months.(30) In a review by Cameron et al, 11 of 12 AIDS patients with pulmonary cryptococcosis had evidence of extrapulmonary disease.(29)
Patients with pulmonary cryptococcosis may present with cough, fever, malaise, shortness of breath, and pleuritic pain. Physical examination may reveal lymphadenopathy, tachypnea, or rales. Chest radiographs typically reveal focal or diffuse infiltrates similar to those that occur in patients with other opportunistic pathogens, particularly Pneumocystis jiroveci pneumonia (PCP). Less common chest radiographic findings include solitary subpleural nodules, masslike infiltrates with consolidation, hilar and mediastinal adenopathy, or pleural effusions. Rarely, cavitation and empyema have been reported.(31) Cryptococcus is less likely to be isolated from sputum than from a bronchoalveolar lavage; however, this difference may be due to overgrowth of normal respiratory flora and failure to look for Cryptococcus on sputum cultures. All HIV-positive patients in whom C neoformans is isolated from the sputum should receive antifungal therapy, because early treatment of localized pulmonary cryptococcosis can be effective and can prevent dissemination.(32)
As localized pulmonary cryptococcosis without lymph node involvement usually is not associated with a positive serum cryptococcal antigen (CrAg), the finding of a positive CrAg in a symptomatic patient should prompt an evaluation for disseminated disease. Some experts recommend that all patients with cryptococcal pneumonia undergo an evaluation for cryptococcal meningitis (ie, lumbar puncture), even in the absence of neurologic signs or symptoms.(33) More than 10% of patients with disseminated disease may have acute respiratory failure, which carries an extremely high rate of mortality.(34)
| Central Nervous System Cryptococcal Invasion|
The CNS is the most common site of disseminated cryptococcal infection. The seriousness of cryptococcal meningitis is illustrated by a prospective observational study conducted over a 10-month period in Zimbabwe, where antifungal therapy was not available.(35) Eighty-nine patients with cryptococcal meningitis were identified. Cryptococcal meningitis was the AIDS-defining illness in 88%. The median survival without antifungal therapy was 14 days (range 0 to 233 days).
Cryptococcemia often precedes CNS invasion and may persist for an extended period of time (1-16 weeks) despite treatment.(36) CNS invasion may be secondary to hematogenous infection or may represent reactivation disease similar to histoplasmosis or tuberculosis. The time from onset of symptoms to diagnosis ranges from days to months.(33) Infection typically presents as a subacute process characterized by headache, fever, and, less often, altered mental status; however, presentations characteristic of either acute or chronic meningitis can occur. Cranial nerve palsies and papilledema are the most common ocular manifestations seen in patients with cryptococcal CNS invasion. There have been only 3 reported cases of intraocular cryptococcosis in patients with AIDS.(22) Complications of CNS infection include hydrocephalus, motor or sensory deficits, cerebellar dysfunction, seizures, and dementia. Focal disease is rarely described.(37) Intracerebral granulomata, referred to as cryptococcomas, occasionally may be seen on computed tomography (CT) or magnetic resonance imaging (MRI) scans regardless of absence of focal neurologic deficits or increased intracranial pressure, but are not observed commonly in AIDS patients.(38)
Abnormal cerebrospinal fluid (CSF) findings, such as pleocytosis, low glucose concentrations, and high protein concentrations, are seen in approximately 40% of patients with AIDS-related cryptococcal meningitis. The CSF opening pressure is greater than 200 mm H2O in 70% of patients with cryptococcal meningitis.(39) In the 4-arm, 2-step clinical trial conducted by the MSG/ACTG, 93% of deaths occurring within the first 2 weeks of therapy and 40% of deaths occurring between weeks 3 and 10 were associated with increased intracranial pressure.(9) A minimal inflammatory response characterized by <10 lymphocytes/µL CSF is seen in approximately 55% of cases of AIDS-related cryptococcal meningitis.(36) In the review by Darras-Joly et al, 26% of HIV-positive patients presenting with cryptococcal meningitis had normal CSF findings (26); this is consistent with previous reports in the literature. Therefore, findings of an apparently normal CSF should not exclude the possibility of cryptococcal infection.
| Cutaneous Cryptococcal Disease|
Cutaneous cryptococcosis is a sign of dissemination present in approximately 10% of cases and may precede life-threatening disease by several weeks.(40,41) The lesions vary greatly in morphology and mimic many other dermatologic entities. They may be misdiagnosed as molluscum contagiosum.(42) The lesions may appear as papules, tumors, vesicles, plaques, abscesses, cellulitis, purpura, draining sinus, ulcers, bullae, or subcutaneous swelling.
| Other Manifestations|
Adrenal insufficiency may occur secondary to cryptococcal invasion of the adrenal glands. CT scanning should be performed in patients with a diagnosis of adrenal insufficiency to evaluate the size and consistency of the adrenal glands.(19) If no other etiology is found, an adrenal biopsy should be performed to evaluate for fungal or mycobacterial disease. It has been postulated that Cryptococcus may lie dormant in the prostate and serve as the source of systemic relapse after completion of therapy.(38,43) Given that most patients with cryptococcosis are fungemic, it is not surprising that there are reports of disease in many organ systems, including cryptococcal myocarditis, arthritis, and gastroenteritis.
| Immune Reconstitution Inflammatory Syndrome|
(See HIV InSite Knowledge Base chapter Clinical Implications of Immune Reconstitution in AIDS.)
In contrast to Mycobacterium avium complex and cytomegalovirus infections, cryptococcal disease initially was implicated in few reports of unusual manifestations associated with immune recovery on ART. One early report (44) suggested that 3 patients had partial immune reconstitution unmasking latent cryptococcal disease. There also were 2 reports of cryptococcal lymphadenitis in patients receiving potent antiretroviral therapy.(45,46) More recently, however, there have been several larger case series reporting significant complications of cryptococcal-related immune reconstitution inflammatory syndrome (IRIS). Investigators from Dallas, Texas, described 4 cases of IRIS and included a literature review of 21 additional cases.(47) The presentations included 14 cases of lymphadenitis, 10 CNS complications (meningitis in 6 and mass lesions in 4), as well as 1 pulmonary cavitary lesion. The median CD4 count at time of cryptococcal diagnosis and prior to ART was 25 cells/µL. The median CD4 count at time of IRIS was 197 cells/µL and the median times to development of IRIS after cryptococcal diagnosis and after initiation of ART were 11 months (range: 7 weeks to 3 years) and 7 months (range: <2 weeks to 22 months), respectively. In all but 1 patient, the clinical manifestations resolved, and 4 patients required antiinflammatory medications to control their signs and symptoms. A study in Thailand found an IRIS incidence of 19% among 153 symptomatic HIV-infected children, of which 3 cases were attributable to cryptococcal infection.(48) IRIS events occurred at a median of 4 weeks (range: 2-31 weeks) after initiation of ART.
In a retrospective review of 84 patients with HIV-associated cryptococcal disease, authors addressed the difficulty in differentiating IRIS from relapsed disease.(49) For the diagnosis of IRIS in the culture-negative meningitis form, all of the following criteria had to be met: 1) the patient's condition had clinically responded to anticryptococcal treatment; 2) after ART was initiated, the original symptoms returned or new inflammatory symptoms developed; and 3) all culture results in the diagnostic workup for the inflammatory process were negative, and CSF CrAg, if still present, had to show at least a 4-fold decrease from the level recorded when the patient was initially admitted to the hospital. The study also included 3 patients with subclinical cryptococcal disease who developed acute disease within 180 days after initiating ART. Of the 84 patients, 59 initiated ART and 18 (30.5%) of the 59 developed IRIS as defined above after a median time of 30 days (range: 3-330 days) on ART. Patients who developed IRIS were more likely to be antiretroviral naive, have a higher CSF CrAg, have a higher baseline HIV viral load, and have initiated ART within 30 days after cryptococcal diagnosis. At the time of IRIS diagnosis, these patients had higher CSF opening pressures, increased CSF whole blood counts, and higher glucose levels compared with the typical presentation of acute cryptococcal meningitis. Although there was no statistical difference in mortality at 18 months of follow-up between those who had IRIS compared with those who did not, there was a trend toward survival benefit in the IRIS group, in contrast to the findings of a prior retrospective chart review of 120 HIV infected individuals with cryptococcal disease who initiated combination ART.(50) In that study, 10 patients developed IRIS within a median of 8 months (range: 2-37 months) after initiating ART, and 3 of the 10 patients with IRIS died. Having previously undiagnosed HIV, CD4 count <7 cells/µL, and starting ART within 2 months of cryptococcosis diagnosis were independently associated with risk of developing IRIS.
These reports have prompted some clinicians to avoid prescribing ART within the first 2-3 months following the diagnosis of cryptococcosis; however there is no prospective data correlating the timing of ART with the development of and outcomes of IRIS that would support specific recommendations. Studies evaluating the timing of when to start ART in the setting of acute opportunistic infections are under way in an effort to gain better insight into this complication and its management.
| Laboratory Evaluation|
Diagnosis is confirmed by isolation of Cryptococcus from a sterile body site, by histopathologic analysis, or by detection of cryptococcal capsular antigen. The india ink stain that outlines the polysaccharide capsule is positive on direct examination of the CSF approximately 50% of the time in normal hosts with cryptococcal meningitis and in more than 80% of patients with AIDS. Encapsulated yeasts seen on Alcian blue or mucicarmine or Gomori methenamine silver are diagnostic of Cryptococcus. Other stains, such as Fontana-Masson and periodic acid-Schiff stain, reveal yeast cells but are not specific for Cryptococcus.
| Cryptococcal Antigen|
CrAg in the serum usually is indicative of systemic disease and correlates with fungal burden. A localized cryptococcal infection, such as pulmonary cryptococcosis without lymph node involvement, usually is not associated with a positive serum CrAg; a positive result warrants a search for disseminated disease.
CrAg in the CSF is produced locally in the subarachnoid space by the invading yeast and does not represent either active or passive diffusion from the serum into the CNS. Detection of CrAg in either serum or CSF has >95% sensitivity and specificity in the diagnosis of true invasive cryptococcal infection.(51,52) False-positive results can occur secondary to infection with Trichosporon beigelii, which cross-reacts with the antigen. In addition, false-positive results have occurred secondary to residual disinfectant on laboratory test slides and inactivated pronase in a test kit.(53,54) False-negative findings may occur with low CrAg concentrations. False-negative serum CrAg rates as high as 6% were reported prior to the use of pronase, which is used routinely now in all test kits.(52) Although serum CrAg is 99% sensitive in cryptococcal meningitis, a positive result does not indicate that CNS invasion is present. Conversely, a negative serum CrAg result suggests that the patient is unlikely to have CNS disease and may be useful in screening symptomatic patients. The utility of serum cryptococcal antigen in asymptomatic HIV patients as a screening tool has not been studied adequately.(25) Although routine screening for CrAg in asymptomatic patients is not recommended, if a positive titer is noted in an asymptomatic patient, it is recommended that therapy with fluconazole be initiated.(55,56)
The time course of CrAg levels during treatment of cryptococcosis is poorly characterized. C neoformans possesses a capsule of high-molecular-weight polysaccharide, which likely results in its slow clearance from the serum and CSF. Positive CrAg test results can persist for many years. Changes in the CSF CrAg titers have limited value in the management of cryptococcal meningitis, although it is expected that a decrease should be seen after 2 or more weeks of therapy. The MSG investigators followed serial CSF cryptococcal antigens over a 10-week period of acute treatment and every 3 months for 1 year during suppressive treatment.(57) They noted a significant difference in responders compared with nonresponders among those patients whose initial CSF CrAg was >1:8; specifically, 86% of responders had a decrease in titer, compared with 44% of nonresponders. There was no correlation between outcome and changes in serum titers of CrAg during treatment for acute meningitis or during suppressive therapy; however, 43% of patients who relapsed had an elevated CSF CrAg over an initial treatment-suppressed value compared with 5% who had no evidence of relapse. Analysis of a subsequent study by the same investigators showed no correlation between baseline serum or CSF antigen titers and 2- or 10-week outcomes.(58) The CSF antigen decreased (4-fold) on treatment in 88% of patients with a clinical response, compared with 64% of those patients who failed to demonstrate a clinical response. In addition, the CSF antigen increased (4-fold) in 1% of patients who clinically responded, compared with 2% who clinically failed.
A retrospective chart review evaluating the monitoring of serum CrAg titers in patients with AIDS-related cryptococcal disease found that, whereas 87% of 136 patients experienced a decrease in serum CrAg on treatment, no significant difference was found between serum CrAg titers of patients who had a clinical response to treatment and those of patients who experienced persistent disease, probable relapse, or definitive relapse of cryptococcal disease.(59) In fact, 6 of the 10 evaluable patients with evidence of definitive relapse had a decrease in serum CrAg compared with previous titers.
Therefore, the CrAg is a valuable tool for initial diagnosis of cryptococcal disease, but the utility of serial serum or CSF antigens during management is less clear, and there is no evidence to support following serial serum or CSF antigens. There is insufficient data to support definitive conclusions from CrAg tests that become negative and subsequently become positive again; however, as such a result may be consistent with recurrence, a diagnostic workup for recurrent disease would seem advisable. The significance of highly elevated CrAg titers is unknown, but this finding has been associated with a poor prognosis. Management decisions should be made on the basis of an individual clinical assessment, cultures, and histopathology, and they should not rely on CrAg titers alone.
| Susceptibility Testing|
Most C neoformans isolates are sensitive to amphotericin B initially, and those that acquire resistance do so during the course of therapy.(60) Acquired resistance to amphotericin B due to development of a defective sterol delta 8-->7 isomerase has been reported in C neoformans from a patient with cryptococcal meningitis.(61)
There has been concern over the possible emergence of fluconazole-resistant cryptococci, especially among AIDS patients who may require prolonged courses of antifungal therapy. Antifungal susceptibility would be useful in guiding selection and monitoring of antifungal therapy, especially in cases of persistent or recurrent fungal infection. However, in vitro antifungal susceptibility testing is of questionable value because of limited correlation of results with clinical response.(62) Several reports in the literature implying fluconazole resistance on the basis of clinical failure without laboratory evidence to support this conclusion reflect the need for reliable and reproducible susceptibility testing.(63) Alternatively, there are reports of 2 patients receiving fluconazole prophylaxis who developed cryptococcal meningitis with fluconazole-resistant isolates but clinically responded to treatment with fluconazole.(64) A patient also has been reported with multiple clinical relapses of cryptococcal meningitis in which the same strain of C neoformans was isolated, revealing progressive antifungal drug resistance to both amphotericin B and fluconazole.(65)
In a study supporting the use of susceptibility testing, 28 isolates from 25 patients were tested for fluconazole susceptibility and correlated with clinical outcomes.(66) Therapeutic failure was observed in 5 patients who were infected with isolates for which the fluconazole minimum inhibitory concentrations (MICs) were >=16 µg/mL. Of these 5 patients, 4 died from active cryptococcal disease. Of the 20 patients with fluconazole-susceptible isolates, only 2 patients died, both of causes unrelated to cryptococcal disease. These findings suggest that fluconazole susceptibility may be useful in predicting clinical response. In addition, a previous study suggested that fluconazole susceptibility might be useful for determining whether fluconazole can be used as primary therapy.(67)
Most of the available data suggest that resistance is an unusual cause of relapse. An evaluation of 13 isolates from 5 patients with recurrent cryptococcal meningitis found the initial and relapse isolates to be clonally related, confirming that the recurrence was not from a new strain.(68) There were no increases in the MICs among serial isolates tested with either amphotericin B or fluconazole, suggesting that relapse was not due to drug resistance but probably was secondary to host factors. A similar, larger study conducted by investigators from the Centers for Disease Control and Prevention also concluded that recurrent disease typically was associated with relapse with the same strain, which remained susceptible to fluconazole.(69)
Antifungal resistance testing is influenced by the medium used, pH, temperature, and size of inoculum.(70-72) A comparison of 5 commercially available antifungal susceptibility testing systems found good reproducibility among 4 of the 5 systems.(73) The degree of concordance with respect to the MICs of the imidazoles was significantly different among the 5 systems. Even with the current standardization of the macrobroth technique, the optimum length of incubation and definition of endpoints are not known. The Etest, potentially a less complex and more economical method of resistance testing, has been compared with the macrobroth technique.(74) The Etest involves placing a plastic strip containing a defined continuous gradient of an antimicrobial drug onto an inoculated agar. As with the macrobroth technique, the endpoints have not been clinically defined.
Despite advances brought about by the standardization of the macrobroth technique by the National Committee for Clinical Laboratory Standards, routine resistance testing of cryptococcal isolates is not recommended. Such testing may be useful, however, when patients have received prolonged courses of therapy and either relapse or failure is noted.
Also refer to Table
1 and Table
| Acute Infection|
| Antifungal Medication|
Based on the results of the MSG/ACTG clinical trial,(9) amphotericin B (0.7 mg/kg intravenously per day) plus flucytosine (100 mg/kg/day) for 2 weeks, followed by fluconazole (400 mg daily) for 8 weeks of consolidation therapy and 200 mg daily for maintenance therapy, is recommended as first-line therapy for AIDS patients with cryptococcal meningitis. The MSG/ACTG investigators conducted a 4-arm, 2-step clinical trial evaluating the use of amphotericin B (0.7 mg/kg/day) with or without flucytosine (100 mg/kg/day) followed by 8 weeks of either fluconazole (400 mg daily) or itraconazole (400 mg daily). Compared with previous studies reporting mortalities of 14-30%,(8,75,76) the mortality associated with cryptococcal meningitis in this study was markedly reduced: 5.5% at 2 weeks and an overall mortality of approximately 10%.(9) A total of 202 patients received amphotericin B with flucytosine, and 179 received amphotericin B alone. There was no significant difference in clinical outcomes between the 2 groups, although there was a trend toward mycologic improvement at 2 weeks in those patients randomized to receive flucytosine. There was a CSF sterilization rate of 60% at 2 weeks in the amphotericin B plus flucytosine arm compared with 51% in the amphotericin B alone arm. Of 306 patients who continued on to the second phase of the study, 151 received fluconazole and 155 received itraconazole. Again, no differences were noted in clinical outcomes between the 2 groups. At 10 weeks, 72% of the fluconazole group had sterile CSF, compared with 60% of the itraconazole group; however, a higher percentage of patients in the itraconazole group did not have a follow-up lumbar puncture. In the analysis of all 4 treatment arms, there were no significant differences among clinical outcomes, CSF sterilization rates, or mortality. In addition, de Lalla et al conducted a similar study using amphotericin B (1 mg/kg/day) with or without flucytosine for 2 weeks, followed by fluconazole or itraconazole, and noted an overall response rate of 94% with no deaths attributed to cryptococcosis.(77)
Why then, is combination therapy followed by fluconazole selectively recommended? Although it was not statistically significant in the MSG/ACTG study, there was a trend toward better mycologic cure in the amphotericin B plus flucytosine arm.(78) Also, in the follow-up MSG study comparing itraconazole with fluconazole for maintenance therapy, the relapse rate was much higher in the itraconazole arm (23%) than in the fluconazole arm (4%).(78) Not receiving flucytosine during primary therapy was an independent risk factor for relapse. Because of the higher relapse rate seen with itraconazole, the study was discontinued early. As with all other previous studies, fluconazole has proven to be superior to all other antifungals for chronic suppressive therapy. Additionally, in contrast to earlier studies in which subjects demonstrated a high toxicity rate associated with flucytosine, the MSG/ACTG study noted that flucytosine was well tolerated. Given the evidence cited above, it is believed that amphotericin B plus flucytosine followed by fluconazole should be the treatment of choice, provided there are no contraindications to either medication. Clearly, flucytosine may be withheld if the patient is intolerant, and itraconazole can be used as an alternative for fluconazole.
| Amphotericin B Intolerance|
For patients intolerant of amphotericin B, there has been much interest in the use of intralipid or liposomal amphotericin B. There are 3 commercially available lipid-based formulations of amphotericin B: amphotericin B lipid complex, amphotericin B colloidal dispersion, and liposomal amphotericin B.(79) Prior to the availability of these products, individual pharmacies sometimes prepared a "homemade" intralipid preparation by emulsifying amphotericin B into a lipid mixture that was intended for use in total parenteral nutrition. It is beyond the scope of this chapter to discuss in depth the differences among the 3 preparations. Although extensive studies have been conducted in oncology patients, very few data are available on the use of these formulations in HIV patients with fungal infections. There have been conflicting data on which formulation appears to be the least nephrotoxic, although there is a trend favoring liposomal amphotericin B.(80,81)
Unfortunately, the limited clinical data have been disappointing and contradictory. In a randomized comparison of amphotericin B (46 patients) and amphotericin B/intralipid (44 patients) for the treatment of AIDS-associated cryptococcal meningitis, an increased incidence of anemia and nephrotoxicity was noted in the intralipid preparation.(82) The intralipid preparation did reduce the incidence of infusion-related adverse reactions and did show a trend toward an improved rate of mycologic cure; but, as there was no benefit in reducing renal toxicity or improving clinical outcome, preferential use cannot be recommended. The results of a phase II trial of intralipid amphotericin B in 51 neutropenic patients failed to detect any improved toxicity profile.(83) In contrast, in a randomized study comparing amphotericin B lipid complex (ABLC) with amphotericin B in 55 AIDS patients with cryptococcal meningitis, there was less hematologic and nephrotoxicity reported in the ABLC arm.(84) In a randomized study evaluating AmBisome (Gilead, Foster City, CA, USA), a liposomal amphotericin B preparation (15 patients), compared with conventional amphotericin B (13 patients) in AIDS-associated cryptococcal meningitis, an earlier CSF sterilization rate was noted in those randomized to the liposomal preparation.(85) In the conventional amphotericin B arm, 2 patients had to discontinue the drug secondary to nephrotoxicity compared with none in the liposomal arm. The creatinine levels at 10 weeks, however, were not reported. It would be anticipated that patients receiving amphotericin B develop a reversible reduction in glomerular function during the first weeks of therapy that would return to baseline after discontinuation of therapy. The Collaborative Exchange of Antifungal Research (CLEAR) database was queried to identify patients with cryptococcal infection who were treated with ABLC.(86) This database includes both HIV-infected and HIV-uninfected patients, and ABLC may have been used as first- or second-line therapy. Overall, the response rate for patients with HIV infection was 58% (30 of 52 patients), which appears comparable to historical response rates with conventional amphotericin B. Until further studies can demonstrate clear evidence of improved clinical outcomes, the role of lipid preparations of amphotericin B remains uncertain.
Other alternatives for management of patients intolerant of amphotericin B include initial treatment with azole regimens. There have been many anecdotal reports of patients successfully managed with use of azoles initially, but multiple clinical trials have not supported their primary use, including 3 prospective, randomized trials in which failure rates ranged from 40% to 57%.(8,75,87)
Moskovitz et al conducted an open-label, randomized trial comparing itraconazole (200 mg orally twice daily) and fluconazole (400 mg orally once daily) as primary treatment of cryptococcal meningitis in patients with AIDS.(76) There was no statistical difference in the rate of CSF sterilization between the 2 groups, although the study was terminated early, in part because of the clinical response rate of only approximately 40%. The sample size was too small to determine comparative efficacy. Thus, it is difficult to conclude that the effectiveness of azoles is >50% in the initial management of cryptococcal meningitis, and their use should be reserved only for patients that have an absolute contraindication for receiving amphotericin B such as renal dysfunction (excluding end-stage renal disease) or known hypersensitivity to amphotericin B.
There is still much interest in the use of higher doses of fluconazole. Berry et al used fluconazole (800 mg daily) as salvage therapy in 7 patients who previously failed other antifungal therapy and noted responses in 4 patients.(88) Haubrich et al treated 6 patients with AIDS-associated cryptococcal meningitis with fluconazole (800 mg daily) and noted clinical response in 5 patients.(89) The median time to CSF sterilization was 21 days.
The California Collaborative Treatment Group (CCTG) conducted a prospective, open-label trial evaluating fluconazole (400 mg daily) plus flucytosine (150 mg/kg/day) in the treatment of cryptococcal meningitis in AIDS patients.(90) After 10 weeks of therapy, 75% of CSF cultures were negative with a median time of 23 days, which was less time than previously reported with either drug alone. Almost 30% of patients had dose-limiting adverse effects attributable to flucytosine that required drug discontinuation. These results suggest that the combination of fluconazole and flucytosine may enhance the efficacy of fluconazole. CCTG 534 has been completed and the results of a preliminary analysis are available.(91) In this dose-escalating, randomized trial, increasing doses of fluconazole (800 mg to 2,000 mg) alone were compared with fluconazole plus flucytosine (150 mg/kg/day) in divided doses. Flucytosine was given for the first 4 weeks. Patients were randomized into 4 successive cohorts with increasing doses of fluconazole; each cohort required a safety review prior to initiation of the next higher fluconazole dose. Primary therapy with high-dose fluconazole continued for at least 10 weeks or until the CSF was sterilized (and up to 26 weeks). A maintenance regimen of fluconazole (200 mg daily) was continued for 12 months thereafter. Success of primary therapy was defined as survival, clinical remission, and CSF sterilization by 10 weeks. Using this definition, 11%, 37%, 62%, and 62% of patients treated with 800, 1,200, 1,600, or 2,000 mg daily of fluconazole alone, respectively, had therapeutic success. Treatment with fluconazole plus flucytosine was successful in 65-87% of cases treated with various doses of fluconazole. Because the analysis is preliminary, the most effective dose and duration of high-dose fluconazole await further analysis. Further studies are warranted before proposing the use of this combination as part of the primary management of cryptococcal meningitis.
Voriconazole may be an alternative for treating fluconazole-refractory disease but there are insufficient data to recommend its routine use or to suggest that voriconazole may be more effective than fluconazole.
The CCTG also looked at the utility of antifungal susceptibility testing to determine which patients may respond to primary treatment with fluconazole.(67) They retrospectively reviewed cases involving 76 patients from 2 clinical trials who had received fluconazole plus flucytosine. They noted that those patients who failed were more likely to have positive blood and urine cultures, high titers of serum and CSF cryptococcal antigens, and higher MICs of fluconazole, and they were less likely to have received flucytosine. Conversely, logistic regression analysis revealed that a negative blood culture, a low MIC of fluconazole, and treatment with flucytosine were independent factors associated with a good response. Consideration of this combination of fluconazole and flucytosine for use in patients intolerant of or failing amphotericin B therapy is recommended.
| Management of Increased Intracranial Pressure|
Possibly as important as the role of antifungals in the treatment of acute cryptococcal meningitis is the aggressive management of acute cerebral edema and increased intracranial pressure. As stated previously, the majority of deaths associated with cryptococcal meningitis in the MSG/ACTG study were associated with increased intracranial pressure, likely reflecting cerebral edema.(9,92) It is critical to measure intracranial pressure at the time of initial lumbar puncture and again if any clinical deterioration of the patient is noted during therapy. The MSG/ACTG investigators noted that patients who achieved CSF sterilization at 2 weeks had a lower baseline opening pressure compared with those with positive cultures at 2 weeks. Of the 21 patients who died during the initial 2 weeks of therapy, 12 patients had a baseline CSF pressure measured, 6 of whom had a pressure >=350 mm H2O. Two of the other patients who died had opening pressures >550 mm H2O immediately prior to death. Significantly higher survival rates were noted in patients with baseline CSF pressure <190 mm H2O compared with those with CSF pressure 250-349 (p = .05) or >=350 mm H2O (p = .008).
High-volume lumbar puncture, defined as removal of 20-30 mL CSF, was performed 1-2 times daily in a prospective cohort of 10 patients with HIV-associated cryptococcal meningitis complicated by elevated intracranial pressure (range: 260-600 mm H2O).(93) All patients returned to their baseline level of consciousness after normalization of their CSF pressure and all 10 remained alive (range: 7-24 months). Eight patients eventually required placement of lumbar peritoneal shunts, only 1 of which was reported to have been removed subsequently.
The Infectious Diseases Society of America (IDSA) practice guidelines recommend that a repeat lumbar puncture be performed 2 weeks after initiation of therapy in patients with normal baseline opening pressures (<200 mm H2O) to exclude elevated pressures and evaluate culture status.(56) Although the most effective management in patients with elevated baseline opening pressures is unclear, daily lumbar puncture removing approximately 30 mL of CSF until the pressure has decreased 50% is recommended. Patients should have daily lumbar punctures to maintain CSF opening pressures in the normal range; these can be discontinued once the pressure is normal for several days. Lumbar drains, or possibly even ventriculoperitoneal shunts, should be considered in patients whose opening CSF pressure is >400 mm H2O. There are no data to support treatment with acetazolamide to reduce intracranial pressure in cryptococcal meningitis, and an associated hyperchloremic metabolic acidosis may limit its use. The osmotic diuretic mannitol (1.5-2 g per kg intravenously) has been used to treat signs of acutely increased intracranial pressure, but its clinical utility specific to cryptococcal meningitis is unknown. The role of corticosteroids is unknown and remains controversial. In the MSG/ACTG trial, the case records for the ACTG did not include a provision for indicating why steroids were used, so only the 263 MSG participant records could be evaluated for steroid use.(92) Of the 263 patients, 110 received steroids, 69 of whom were given steroids for treatment of amphotericin B infusion-related adverse events. Of the 41 patients given steroids for reasons other than suppression of infusion-related events, 66% had a successful outcome at 2 weeks, compared with 86% of patients who did not receive steroids. The 2-week mortality rate of these 41 patients was 20%, compared with 3% in the group not treated with steroids. Of note, 42% of the steroid-treated patients who had opening pressures measured had opening pressures >=350 mm H2O. Therefore, corticosteroids were strongly associated with mycologic failure, clinical failure, and early death, but this may just reflect the severity of the disease in patients nonrandomly selected to receive steroids, and whether steroids improved or adversely affected outcomes cannot be determined. The number of patients who received either mannitol or acetazolamide was too small to allow any comment regarding benefit.
| Maintenance Therapy|
The relapse rate for cryptococcal meningitis in non-AIDS patients is 15-25% compared with >50% in persons with AIDS. Prior to the introduction of highly active ART (HAART), patients with AIDS-associated cryptococcal meningitis had to continue chronic suppressive therapy for the remainder of their lives. Repeat lumbar punctures should be performed after 2 weeks of therapy and may be considered after completion of therapy and at any time there is clinical evidence to suggest relapse.(56,94) CSF pleocytosis may persist for up to 6 months after successful treatment, and most patients should have normal CSF findings by 1 year posttreatment. There are no established guidelines for the frequency or timing of surveillance lumbar punctures to evaluate for relapse.
Previous studies conducted by the CCTG and MSG/ACTG have firmly established that fluconazole is the drug of choice for maintenance therapy.(95,96) A phase III, multicenter, randomized, double-blind comparison of fluconazole (200 mg daily) vs itraconazole (200 mg daily) conducted by the MSG was terminated after preliminary results revealed a CSF culture relapse rate of 4% in patients with AIDS who were receiving fluconazole, compared with 24% relapse in the itraconazole group.(78) The dosage of itraconazole may have been too low, and further studies of higher-dose itraconazole may be warranted to determine comparative efficacy.
The guidelines issued in 2004 by the U.S. Public Health Service (USPHS) and IDSA have changed secondary prophylaxis recommendations for cryptococcosis to reflect data revealing that it is safe to discontinue secondary prophylaxis in patients who have had a sustained immunologic response on effective ART.(94) It seems reasonable to assume that it would be safe to discontinue secondary prophylaxis in patients with a history of disseminated fungal disease who have had sustained immunologic responses. There have been 2 small prospective case series of safely discontinuing secondary prophylaxis in patients with cryptococcal meningitis. In one study, 6 patients with disseminated cryptococcal disease, including 1 patient with a history of a brain abscess, discontinued antifungal prophylaxis after 1 year of prophylaxis and a sustained CD4 count >150 cells/µL (range: 178-525 cells/µL) on ART.(97) One patient had both a positive CSF CrAg and serum CrAg at the time of enrollment. In another study, 6 patients including 3 with serum CrAg titers >1:8 discontinued antifungal prophylaxis after a median of 11 months of combination ART with a CD4 count >100 cells/µL.(98) None of the patients in either study has had a recurrence of cryptococcosis. Four small retrospective studies found no recurrences of cryptococcosis in patients who discontinued secondary prophylaxis.(99-102) A larger, retrospective, multicenter study reviewed charts from 100 patients who, at the time maintenance therapy was discontinued, had a median CD4 count of 259 cells/µL and median plasma HIV viral load <2.30 log10 copies/mL.(103) The serum CrAg was undetectable in 56 patients at the time of discontinuation of prophylaxis. During a median follow-up period of 28.4 months (range: 6.7-64.5 months; 262 person-years), 4 events were observed (incidence: 1.53 events per 100 person-years). Three of these patients had a CD4 count >100 cells/µL and a positive serum CrAg result during the recurrent episode. One patient had a CD4 count <100 cells/µL, a reminder of the need to reinitiate prophylaxis at time of immunologic failure in patients who have had a history of cryptococcosis. Thai investigators conducted a prospective, multicenter, randomized study on discontinuing secondary prophylaxis in 60 HIV-infected subjects who were treated successfully for acute cryptococcal meningitis.(104) Subjects were randomized to continue (n = 22) or discontinue (n = 20) secondary prophylaxis when the CD4 count had increased to >100 cells/µL and HIV RNA level had been undetectable for 3 months on zidovudine, lamivudine, and efavirenz. At a median of 48 weeks after randomization, there were no episodes of cryptococcal meningitis in either group.
The USPHS/IDSA guidelines recommend discontinuation of secondary prophylaxis if patients successfully complete a course of initial therapy for cryptococcosis, remain asymptomatic with respect to signs and symptoms of their cryptococcosis, and have a sustained increase (>6 months) in their CD4 counts to >100-200 cells/µL on ART. Prophylaxis should be restarted if the CD4 count declines to <100-200 cells/µL.(94)
| Preventive Therapy|
Fluconazole is highly effective in preventing cryptococcal infection in advanced HIV disease. Fluconazole (100 mg daily) in 329 AIDS patients with CD4 counts <68 cells/µL and no prior history of systemic fungal infections was associated with a 76% reduction in the 1-year incidence of systemic fungal infections compared with 337 historical controls.(105) There were 16 cases of cryptococcosis in the historical controls, compared with only 1 after fluconazole prophylaxis was initiated. Although the results suggest that chemoprophylaxis for cryptococcosis may be effective, the trial was not randomized with concurrent controls and the follow-up period was <6 months.
The ACTG conducted a randomized trial comparing fluconazole (200 mg daily) with clotrimazole troches for prevention of fungal infections in AIDS patients.(106) The researchers estimated a 2-year rate of invasive fungal infection to be 2.8% in the fluconazole group compared with 9.1% in the clotrimazole group. This reduction seen in the fluconazole group was attributed to the reduction of cryptococcosis. The greatest benefit was seen in patients with CD4 counts <50 cells/µL.
The CCTG conducted a double-blind, randomized study comparing weekly fluconazole (400 mg) with daily fluconazole (200 mg) for the prevention of fungal infections.(107) Results of both arms were similar to those of the above-mentioned study by the ACTG (106) as far as prevention of cryptococcosis; however, the weekly dose was less effective in suppressing oral candidiasis.
Additional data support the use of azole antifungals for prophylaxis. In a prospective study in which 231 HIV-infected individuals received fluconazole (200 mg 3 times per week), cryptococcal meningitis occurred in 1 patient (0.4%).(108) A retrospective comparison of 128 patients receiving fluconazole with 121 patients who did not receive fluconazole found an increased incidence of systemic mycoses and esophageal candidiasis in those patients not receiving fluconazole.(109) The MSG conducted a prospective, randomized trial of itraconazole (200 mg daily) compared with placebo in 295 HIV-infected patients with CD4 counts <150 cells/µL residing in areas endemic for histoplasmosis.(110) Prophylaxis failure for oral candidiasis was 19% in the itraconazole arm compared with 29% in the placebo arm. Systemic fungal infections, including 1 case of cryptococcosis, occurred in 6 patients on itraconazole, compared with 17 systemic fungal infections in the placebo group, in which 8 patients developed cryptococcosis. No survival benefit was noted.
The Cochrane Collaboration recently conducted a metaanalysis of 5 randomized controlled studies assessing the use of either fluconazole or itraconazole for primary prophylaxis. Neither treatment was found to provide a significant difference in mortality when compared with placebo.(111) However, a study in Thailand did in fact demonstrate a survival benefit: a randomized, double-blind, placebo-controlled study of 90 HIV-infected patients with CD4 counts <100 cells/µL who received either fluconazole 400 mg weekly (n = 44) or placebo (n = 46) found that, on an intent-to-treat basis, cryptococcal meningitis developed in 3 (6.8%) in the fluconazole group and 7 (15.2%) in the placebo group [hazard ratio = 2.23; 95% confidence interval (CI): 0.58-8.63; p = .245]. The number of deaths per 10,000 person-days was 2.7 for the fluconazole group and 11.7 for the placebo group [rate difference = 9; 95% CI: 0.4-17.5; p = .046]. At this point, the USPHS/IDSA recommendations for prophylaxis in patients with HIV infection do not endorse routine use of primary antifungal prophylaxis. However, the survival benefit reported with the use of fluconazole 400 mg once weekly for primary prophylaxis for cryptococcal meningitis in Thailand may have implications for other resource-poor countries. Studies exploring the routine use of primary prophylaxis are warranted in resource-poor settings where the incidence of cryptococcosis is significantly higher, there is limited access to ART, and generic fluconazole or donated fluconazole is available.
Nevertheless, the efficiency of using azoles as primary prophylaxis for cryptococcosis is unclear. Specific questions remain unanswered. At what dosage or duration is protection optimal? Are there certain patient characteristics that would support the use of primary prophylaxis in certain subgroups? The overall incidence of serious fungal infections is low, and careful monitoring may be more cost effective. Further studies are needed to determine if certain risk factors or predictive factors exist to support the use of fluconazole as primary prophylaxis. There is concern that prolonged usage of fluconazole may result in acquired resistance to fluconazole. In addition, drug-related adverse reactions and drug-drug interactions need to be considered. Most importantly, with effective ART and increased CD4 cell counts, it is likely that specific prophylaxis will not be needed. However, as noted above, prophylaxis may be of significant benefit in countries that are resource poor and have a high prevalence of systemic fungal infections. Studies currently are in development to evaluate the effectiveness of such prophylactic strategies in Africa.
| Summary and Future Directions|
In summary, acute cryptococcal meningitis should be treated with amphotericin B (0.7 mg/kg/day) plus flucytosine (100 mg/kg/day in divided doses) for 2 weeks, followed by fluconazole (400 mg daily) for 8 weeks, followed by lifelong suppression with fluconazole (200 mg daily). Patients who have had immunologic response on ART with sustained CD4 counts >100 cells/µL may be able to discontinue maintenance therapy. This is in agreement with the practice guidelines recommended by the MSG Cryptococcal Subproject and the USPHS.(56,94) Further studies are warranted to explore the use of other antifungals or combination antifungals and immunomodulating agents such as monoclonal antibodies and gamma-interferon in the primary management of cryptococcal meningitis.(112,113) Clearly, the aggressive management of increased intracranial pressure plays a major contributing role in clinical outcome, and all patients with opening CSF pressures >250 mm H2O should have serial lumbar punctures until pressures have normalized. Although multiple studies have shown the clinical efficacy of fluconazole prophylaxis for systemic fungal infections, this strategy may be most effective in countries where ART is unavailable.
|| || Hajjeh RA, Conn LA, Stephens DS, Baughman W, Hamill R, Graviss E, Pappas PG, Thomas C, Reingold A, Rothrock G, Hutwagner LC, Schuchat A, Brandt ME, Pinner RW. Cryptococcosis: population-based multistate active surveillance and risk factors in human immunodeficiency virus-infected persons. Cryptococcal Active Surveillance Group. J Infect Dis 1999; 179:449-54.|
|| ||Aberg JA, Koo JJ. Increase in the Rate of AIDS-Related Opportunistic Infections at San Francisco General Hospital in the Era of Potent Antiretroviral Therapy. XIII International AIDS Conference, July 9-14, 2000; Durban, South Africa. Abstract MoPeB2301.|
|| || Bennett JE, Dismukes WE, Duma RJ, Medoff G, Sande MA, Gallis H, Leonard J, Fields BT, Bradshaw M, Haywood H, McGee ZA, Cate TR, Cobbs CG, Warner JF, Alling DW. A comparison of amphotericin B alone and combined with flucytosine in the treatment of cryptoccal meningitis. N Engl J Med 1979; 301:126-31.|
|| || Dismukes WE, Cloud G, Gallis HA, Kerkering TM, Medoff G, Craven PC, Kaplowitz LG, Fisher JF, Gregg CR, Bowles CA, et al. Treatment of cryptococcal meningitis with combination amphotericin B and flucytosine for four as compared with six weeks. N Engl J Med 1987; 317:334-41.|
|| || Saag MS, Powderly WG, Cloud GA, Robinson P, Grieco MH, Sharkey PK, Thompson SE, Sugar AM, Tuazon CU, Fisher JF, et al. Comparison of amphotericin B with fluconazole in the treatment of acute AIDS-associated cryptococcal meningitis. The NIAID Mycoses Study Group and the AIDS Clinical Trials Group. N Engl J Med 1992; 326:83-9.|
|| || van der Horst CM, Saag MS, Cloud GA, Hamill RJ, Graybill JR, Sobel JD, Johnson PC, Tuazon CU, Kerkering T, Moskovitz BL, Powderly WG, Dismukes WE. Treatment of cryptococcal meningitis associated with the acquired immunodeficiency syndrome. National Institute of Allergy and Infectious Diseases Mycoses Study Group and AIDS Clinical Trials Group. N Engl J Med 1997; 337:15-21.|
|| ||McGuire D, Bromley E, Aberg J, et al. Focal posterior-hemisphere invasive cryptococcal encephalitis- a distinct clinical-neuroimaging entity complicating cryptococcal meningitis in AIDS. In: Program and Abstracts of the 122nd Annual Meeting of the American Neurological Association. September 1997.|
|| ||Powderly WG, Tuazon C, Cloud GA, et al. for the MSG 17/ACTG 159 Team. Serum and CSF cryptococcal antigen in management of cryptococcal meningitis in AIDS. In: Program and Abstracts of 4th National Conference on Human Retroviruses and Related Infections, Washington, DC, 1997.|
|| ||Powderly WG, Keath EJ, Sokol-Anderson M, et al. Amphotericin B-resistant Cryptococcus neoformans in a patient with AIDS. Infect Dis Clin Pract 1992;1:314-316.|
|| ||Smith NH, Graviss EA, Hashmey R, et al. Multi-drug-resistant cryptococcal meningitis in an AIDS patient. In: Program and Abstracts of the 35th Interscience Conference on Antimicrobial Agents and Chemotherapy, September 1997, San Francisco, American Society for Microbiology. Abstract 529.|
|| || Pfaller MA, Rinaldi MG, Galgiani JN, Bartlett MS, Body BA, Espinel-Ingroff A, Fromtling RA, Hall GS, Hughes CE, Odds FC, et al. Collaborative investigation of variables in susceptibility testing of yeasts. Antimicrob Agents Chemother 1990; 34:1648-54.|
|| ||Moskovitz BL, Wiesinger B, and the Cryptococcal Meningitis Research Group. Randomized comparative trial of itraconazole and fluconazole for treatment of AIDS-related cryptococcal meningitis. In: Program and Abstracts of the 1st National Conference on Human Retroviruses and Related infections. December 1993. Washington, DC, American Society for Microbiology. Abstract 34.|
|| || Saag MS, Cloud GA, Graybill JR, Sobel JD, Tuazon CU, Johnson PC, Fessel WJ, Moskovitz BL, Wiesinger B, Cosmatos D, Riser L, Thomas C, Hafner R, Dismukes WE. A comparison of itraconazole versus fluconazole as maintenance therapy for AIDS-associated cryptococcal meningitis. National Institute of Allergy and Infectious Diseases Mycoses Study Group. Clin Infect Dis 1999; 28:291-6.|
|| || Sharkey PK, Graybill JR, Johnson ES, Hausrath SG, Pollard RB, Kolokathis A, Mildvan D, Fan-Havard P, Eng RH, Patterson TF, Pottage JC, Jr., Simberkoff MS, Wolf J, Meyer RD, Gupta R, Lee LW, Gordon DS. Amphotericin B lipid complex compared with amphotericin B in the treatment of cryptococcal meningitis in patients with AIDS. Clin Infect Dis 1996; 22:315-21.|
|| || Leenders AC, Reiss P, Portegies P, Clezy K, Hop WC, Hoy J, Borleffs JC, Allworth T, Kauffmann RH, Jones P, Kroon FP, Verbrugh HA, de Marie S. Liposomal amphotericin B (AmBisome) compared with amphotericin B both followed by oral fluconazole in the treatment of AIDS-associated cryptococcal meningitis. Aids 1997; 11:1463-71.|
|| ||Personal communication with R Haubrich.|
|| || Bozzette SA, Larsen RA, Chiu J, Leal MA, Jacobsen J, Rothman P, Robinson P, Gilbert G, McCutchan JA, Tilles J, et al. A placebo-controlled trial of maintenance therapy with fluconazole after treatment of cryptococcal meningitis in the acquired immunodeficiency syndrome. California Collaborative Treatment Group. N Engl J Med 1991; 324:580-4.|
|| || Powderly WG, Saag MS, Cloud GA, Robinson P, Meyer RD, Jacobson JM, Graybill JR, Sugar AM, McAuliffe VJ, Follansbee SE, et al. A controlled trial of fluconazole or amphotericin B to prevent relapse of cryptococcal meningitis in patients with the acquired immunodeficiency syndrome. The NIAID AIDS Clinical Trials Group and Mycoses Study Group. N Engl J Med 1992; 326:793-8.|
|| || Mussini C, Pezzotti P, Miro JM, Martinez E, de Quiros JC, Cinque P, Borghi V, Bedini A, Domingo P, Cahn P, Bossi P, de Luca A, d'Arminio Monforte A, Nelson M, Nwokolo N, Helou S, Negroni R, Jacchetti G, Antinori S, Lazzarin A, Cossarizza A, Esposito R, Antinori A, Aberg JA. Discontinuation of maintenance therapy for cryptococcal meningitis in patients with AIDS treated with highly active antiretroviral therapy: an international observational study. Clin Infect Dis 2004; 38:565-71.|
|| || Powderly WG, Finkelstein D, Feinberg J, Frame P, He W, van der Horst C, Koletar SL, Eyster ME, Carey J, Waskin H, et al. A randomized trial comparing fluconazole with clotrimazole troches for the prevention of fungal infections in patients with advanced human immunodeficiency virus infection. NIAID AIDS Clinical Trials Group. N Engl J Med. 1995 Mar 16;332(11):700-5.|
|| || Havlir DV, Dube MP, McCutchan JA, Forthal DN, Kemper CA, Dunne MW, Parenti DM, Kumar PN, White AC, Jr., Witt MD, Nightingale SD, Sepkowitz KA, MacGregor RR, Cheeseman SH, Torriani FJ, Zelasky MT, Sattler FR, Bozzette SA. Prophylaxis with weekly versus daily fluconazole for fungal infections in patients with AIDS. Clin Infect Dis 1998; 27:1369-75.|
|| || McKinsey DS, Wheat LJ, Cloud GA, Pierce M, Black JR, Bamberger DM, Goldman M, Thomas CJ, Gutsch HM, Moskovitz B, Dismukes WE, Kauffman CA. Itraconazole prophylaxis for fungal infections in patients with advanced human immunodeficiency virus infection: randomized, placebo-controlled, double-blind study. National Institute of Allergy and Infectious Diseases Mycoses Study Group. Clin Infect Dis 1999; 28:1049-56.|