INTRODUCTION — Cryptococcus neoformans meningoencephalitis is one of the leading opportunistic infections seen in patients with untreated AIDS [1]. Management of these severely immunocompromised patients includes antifungal therapy combined with antiretroviral therapy (ART), with careful monitoring for complications related to the invasive fungal infection and the inflammatory syndrome that can develop secondary to immune recovery [2].

This topic will review the treatment of C. neoformans meningoencephalitis in patients with HIV infection. Other topic reviews that discuss cryptococcal infections (C. neoformans and Cryptococcus gattii) include:

●(See "Microbiology and epidemiology of Cryptococcus neoformans infection".)

●(See "Epidemiology, clinical manifestations, and diagnosis of Cryptococcus neoformans meningoencephalitis in patients with HIV".)

●(See "Clinical management and monitoring during antifungal therapy for cryptococcal meningoencephalitis in patients with HIV".)

●(See "Cryptococcus neoformans infection outside the central nervous system".)

●(See "Clinical manifestations and diagnosis of Cryptococcus neoformans meningoencephalitis in HIV-seronegative patients".)

●(See "Cryptococcus neoformans: Treatment of meningoencephalitis and disseminated infection in HIV seronegative patients".)

●(See "Cryptococcus gattii infection: Microbiology, epidemiology, and pathogenesis".)

●(See "Cryptococcus gattii infection: Clinical features and diagnosis".)

●(See "Cryptococcus gattii infection: Treatment".)

TREATMENT

Overview — Patients with HIV and advanced immunosuppression (eg, CD4 cell count <50 cells/microL) are at risk for severe cryptococcal meningoencephalitis, which is uniformly fatal within approximately two weeks if untreated [3]. Common presenting symptoms include fever, headache, photophobia, nausea, and vomiting; patients with fulminant disease may present with coma. Predictors of poor outcome include high cerebrospinal fluid (CSF) cryptococcal antigen levels (titer >1:1024 by latex; >1:4000 by lateral flow assay), low body weight, poor CSF inflammatory response (<20 cells/microL of CSF), and altered mental status on presentation [4]. (See "Epidemiology, clinical manifestations, and diagnosis of Cryptococcus neoformans meningoencephalitis in patients with HIV".)

Management of cryptococcal meningoencephalitis includes all of the following:

●Antifungal therapy. (See 'Antifungal therapy' below.)

●Control of intracranial pressure, which can lead to blindness, herniation, persistent headaches, and/or neuropathies if left untreated. (See 'Management of increased intracranial pressure' below.)

●Immune recovery with potent antiretroviral medications. However, to reduce the risk of developing an immune reconstitution inflammatory syndrome (IRIS), antiretroviral therapy (ART) should be started between 2 and 10 weeks after antifungal therapy has been initiated. In resource-limited areas, the timing of ART is commonly between four and six weeks, but in resource-rich health care systems, ART initiation can be delayed up to 10 weeks. (See 'When to initiate antiretroviral therapy' below.)

There is no role for empiric glucocorticoids during induction therapy. This is in contrast to other conditions (eg, tuberculous meningitis), in which glucocorticoids are routinely administered as part of initial therapy. The use of glucocorticoids for patients with cryptococcal meningitis was evaluated in a randomized trial that included 451 patients from Asia and Africa [5]. Patients received amphotericin B plus fluconazole with or without dexamethasone (starting at 0.3 mg/kg per day and then tapered over six weeks). The trial was stopped early after an interim analysis found no difference in mortality or the rate of IRIS between the two groups at 10 weeks; in addition, patients who received adjunctive glucocorticoids were less likely to have a good neurologic outcome (13 versus 25 percent; odds ratio 0.42, 95% CI 0.25-0.69). Patients who received dexamethasone also had significantly slower rates of fungal CSF clearance after approximately two weeks and were significantly more likely to develop an adverse event (eg, infection, renal or cardiac event) by six months (667 versus 494 adverse events).

Antifungal therapy — The optimal approach to antifungal therapy involves three phases: induction therapy for approximately two weeks, followed by consolidative therapy for eight weeks [2,6], and then maintenance (ie, suppressive) therapy for at least one year to decrease the risk of relapse (table 1). (See 'Induction therapy' below and 'Consolidation therapy' below and 'Maintenance therapy' below and 'Considerations in resource-limited settings' below.)

The primary antifungal agents used for the treatment of cryptococcal meningoencephalitis include intravenous amphotericin B, oral flucytosine, and oral fluconazole. Use of intrathecal or intraventricular amphotericin B is not advised except in extreme circumstances, since systemic administration demonstrates good efficacy and these other direct routes can be associated with arachnoiditis [7]. In addition, echinocandin antifungals should not be used to treat this infection, since echinocandins do not have significant activity against C. neoformans.

The approach to treatment below describes treatment in nonpregnant adults. Treatment of pregnant persons should be managed in conjunction with an infectious disease specialist, since flucytosine and fluconazole may be teratogenic during pregnancy, particularly in the first trimester.

Induction therapy

Regimen — We recommend induction therapy with liposomal amphotericin B (3 to 4 mg/kg intravenously [IV] daily) plus flucytosine (100 mg/kg per day orally in four divided doses). The duration of therapy and special considerations for regimen selection in resource-limited settings are discussed below. (See 'Considerations in resource-limited settings' below.)

For patients with reduced kidney function, the dose of flucytosine must be reduced. Dose modifications are discussed in the Lexicomp drug information topic within UpToDate and within the individual topic review. (See "Pharmacology of flucytosine (5-FC)", section on 'Dose modification'.)

If liposomal amphotericin B is not available, amphotericin B lipid complex (5 mg/kg intravenously [IV] daily) can be used [2]. Amphotericin B deoxycholate (0.7 mg/kg IV daily) can also be used if a lipid formulation is not available (eg, because of cost) and/or the patient is not at risk for developing nephrotoxicity. Although some guideline panels use higher doses of amphotericin B deoxycholate (eg, 1 mg/kg/day) [8], a comparison of 0.7 mg/kg/day with 1 mg/kg/day of amphotericin B demonstrated no difference in mortality at 10 weeks in a small trial of 64 patients with HIV and cryptococcal meningoencephalitis [9].

If flucytosine is not available, fluconazole (800 mg per day orally or IV) should be used in combination with amphotericin B instead of flucytosine. Additional considerations when amphotericin B and flucytosine cannot be used are discussed below. (See 'Patients who have difficulty tolerating amphotericin B and/or flucytosine' below.)

The rationale for this approach is discussed below. (See 'Rationale' below.)

Duration — Induction therapy should be administered for at least two weeks. The duration should be extended if clinical improvement is not observed and/or if CSF sterilization has not yet been achieved [2]. Considerations in resource-limited settings and the timing of repeat lumbar puncture (LP) to assess for CSF sterilization are discussed elsewhere. (See 'Considerations in resource-limited settings' below and "Clinical management and monitoring during antifungal therapy for cryptococcal meningoencephalitis in patients with HIV", section on 'Sterilization of cerebrospinal fluid'.)

Patients who do not achieve rapid sterilization should be managed in conjunction with a provider experienced in the management of cryptococcal disease. (See "Clinical management and monitoring during antifungal therapy for cryptococcal meningoencephalitis in patients with HIV", section on 'Management of persistent or relapsing infection'.)

After completing induction therapy, patients should transition to consolidation therapy. (See 'Consolidation therapy' below.)

Rationale — Data supporting this approach to induction therapy include:

●Efficacy of combination therapy – Rapid sterilization of the CSF is linked to better survival rates and decreased rates of relapse among symptomatic patients with cryptococcal meningoencephalitis [10-13]. However, many patients with HIV and cryptococcal meningoencephalitis have a high burden of infection in the central nervous system (CNS), with yeast concentrations that can reach a million yeasts/mL of CSF. This high burden of infection requires a combination treatment regimen with fungicidal activity, such as IV amphotericin B with flucytosine, to rapidly sterilize the CSF during induction therapy [10,13].

The use of combination therapy with amphotericin B plus flucytosine was supported in a landmark multicenter, double-blind trial in 381 patients with HIV and cryptococcal meningoencephalitis, which evaluated the effectiveness of two weeks of induction therapy with amphotericin B (dosed at 0.7 mg/kg per day), with or without the flucytosine (100 mg/kg per day) [11]. Induction therapy was followed by eight weeks of consolidation therapy with an azole. This trial demonstrated that CSF sterilization at two weeks occurred more frequently among patients receiving combination therapy compared with amphotericin B alone (60 versus 51 percent) without increased toxicity. A multivariate analysis demonstrated that the likelihood of CSF sterilization at the end of 10 weeks was highest among those patients who were assigned combination therapy followed by fluconazole. In this trial, there was no mortality benefit noted among those who received combination therapy. Other trials have confirmed faster rates of CSF sterilization and lower rates of relapse with amphotericin B plus flucytosine than with amphotericin B alone. [14-17]

A subsequent randomized, open-label trial found a survival benefit when amphotericin B plus flucytosine was compared with amphotericin B monotherapy or a different combination regimen [13]. In this trial, 299 HIV-positive patients with cryptococcal meningitis in Vietnam were randomly assigned to monotherapy with amphotericin B deoxycholate 1 mg/kg daily for four weeks, combination therapy with amphotericin B deoxycholate 1 mg/kg daily plus flucytosine 100 mg/kg per day in three or four divided doses for two weeks, or amphotericin B deoxycholate 1 mg/kg daily plus fluconazole 400 mg twice daily for two weeks. In addition to demonstrating increased rates of yeast clearance in the CSF, this trial found that fewer deaths occurred by days 14 and 70 in patients who received amphotericin B plus flucytosine compared with patients who received amphotericin B monotherapy (15 versus 25 deaths by day 14, hazard ratio [HR] 0.57, 95% CI 0.30-1.08; 30 versus 44 deaths by day 70, HR 0.61, 95% CI 0.39-0.97). The mortality reduction was statistically significant at 70 days, but not at 14 days. Similar survival rates were observed in patients who received combination therapy with amphotericin B plus fluconazole and amphotericin B monotherapy. Rates of adverse events were similar in all groups, although neutropenia was more common in patients receiving combination therapy versus amphotericin B monotherapy (approximately 32 percent versus 19 percent).

The mortality benefit of amphotericin B plus flucytosine was also supported in the ACTA trial, which is described below [18]. (See 'Considerations in resource-limited settings' below.)

Clinical trials have examined whether there may be a role for triple combination therapy with amphotericin B/flucytosine/fluconazole compared with dual therapy, and results are mixed. No additional benefit was observed in one study [15], but triple combination therapy appeared to have a more favorable outcome in another study [19].

●Amphotericin B formulation – We suggest that lipid formulations of amphotericin B be administered rather than amphotericin B deoxycholate. The use of lipid formulations is particularly important in patients with or at risk for renal failure (eg, those with diabetes or uncontrolled hypertension, patients taking other nephrotoxic drugs, or patients with suspected HIV nephropathy) [20-23]. Lipid formulations reduce toxicity (eg, nephrotoxicity, infusion reactions, anemia) and, therefore, are able to improve the ability to give an uninterrupted induction period of treatment, which is critical for the successful management of cryptococcal meningoencephalitis [24].

Although most of the clinical trials have evaluated the deoxycholate formulation of amphotericin B in combination with flucytosine, lipid formulations appear effective for the treatment of cryptococcal meningitis [20,23]. The efficacy of conventional amphotericin B (0.7 mg/kg/day) was compared with two dosing regimens of liposomal amphotericin B (3 mg/kg/day in the second arm; 6 mg/kg/day in a third arm) among 267 patients with HIV and cryptococcal meningoencephalitis [20]. The mean serum creatinine amongst the trial participants was approximately 1.1 mg/dL; those with a serum creatinine greater than twice the normal range were excluded. Efficacy was similar in all arms, although there was less nephrotoxicity among patients in the 3 mg/kg/day liposomal amphotericin B arm. Infusion reactions were less frequent in both liposomal treatment groups compared with the standard amphotericin B group.

●Alternative to flucytosine – Fluconazole should be used as part of a combination regimen if flucytosine cannot be used. When measuring rates of CSF sterilization, the combination of amphotericin B plus fluconazole appears to be more effective than amphotericin B monotherapy [15]. However, flucytosine in combination with amphotericin B is superior to amphotericin B plus fluconazole [13,15,18]. In the randomized trial from Vietnam (described above), the combination of amphotericin B plus fluconazole had no significant effect on survival compared with amphotericin B monotherapy; by contrast, there was a survival benefit if amphotericin B plus flucytosine was used [13].

If fluconazole is administered as part of a combination regimen, clinical trial data suggest that high doses of fluconazole should be used [25,26]. In an open-label study, 143 individuals with HIV in the United States or Thailand diagnosed with cryptococcal meningitis were randomly assigned to induction therapy with amphotericin B deoxycholate (0.7 mg/kg/day) alone, or amphotericin B deoxycholate (0.7 mg/kg/day) plus fluconazole (400 mg/day), or amphotericin B deoxycholate (0.7 mg/kg/day) plus fluconazole (800 mg/day) [26]. Following the induction phase of therapy, patients who initially received amphotericin B monotherapy were switched to fluconazole (400 mg/day) for eight weeks, whereas those who initially received amphotericin B plus fluconazole were switched to consolidation therapy with fluconazole at the same dose to which they were randomly assigned (400 or 800 mg per day) for eight weeks. Although this study was not powered to demonstrate statistically significant differences in efficacy among the treatment arms, the arm that received high-dose fluconazole (800 mg per day) showed a trend toward better outcomes, and the regimen was well tolerated.

Consolidation therapy — After completing induction therapy with amphotericin B and flucytosine, patients should receive consolidation therapy with fluconazole for a minimum of eight weeks. We typically extend the duration of consolidation therapy in patients who have had a slow clinical response to therapy or who do not have sterile CSF at two weeks and in patients whose antiretroviral therapy (ART) is delayed for more than 10 weeks after diagnosis. (See 'When to initiate antiretroviral therapy' below.)

For those who received induction therapy with amphotericin B plus flucytosine for two weeks, the preferred dose of fluconazole is 400 mg daily. By contrast, the dose of fluconazole is 800 mg daily for those who received induction therapy with amphotericin B plus fluconazole (instead of flucytosine). Dosing considerations for regimens used in resource-limited settings are described below. (See 'Considerations in resource-limited settings' below.)

The process of transitioning to consolidation therapy can vary and depends upon the patient’s response to induction therapy:

●For patients who rapidly responded to induction therapy, it is reasonable to switch to fluconazole at two weeks pending the results of the repeat CSF culture. For those who received induction therapy with amphotericin B plus flucytosine, some providers switch to the 400 mg dose right away, whereas others prefer to treat with fluconazole 800 mg pending the final culture results.

●For patients who have not had a clear response to therapy, we typically continue induction therapy pending the results of the CSF cultures. If this is not possible (eg, because of toxicity, inability to stay in the hospital), then we switch to the 800 mg dose of fluconazole pending the results.

For consolidation therapy, fluconazole is preferred over itraconazole because it achieves more reliable drug levels and has fewer drug interactions and less gastrointestinal toxicity. The use of fluconazole versus that of itraconazole for consolidation therapy was evaluated in a trial of 381 patients (also discussed above), in which investigators compared eight weeks of fluconazole (400 mg daily) or itraconazole (400 mg daily) as consolidation therapy [11]. Among the patients who underwent a repeat lumbar puncture (LP) at the end of the consolidation phase, the CSF culture was negative in 97 percent of patients on the fluconazole arm (139 of 151) compared with 92 percent in the itraconazole group (93 of 101 patients). However, the data were somewhat limited since a significant percentage of patients assigned to the itraconazole and fluconazole arms did not undergo repeat LP (35 and 26 percent, respectively). Clinical outcomes were similar between the two arms.

Maintenance therapy

●Choice of agent – After completing induction and consolidation therapy, maintenance therapy with a lower dose of fluconazole (200 mg daily) should be continued for long-term suppression.

Early in the AIDS epidemic, primary therapy for cryptococcal meningoencephalitis was followed by frequent relapse after treatment discontinuation [27]. In one study of 84 patients with AIDS and a history of cryptococcal meningoencephalitis, those who were randomly assigned to 200 mg of daily fluconazole (after primary induction and consolidative therapy) had a much lower incidence of relapse than those who were assigned to placebo (none versus 15 percent) [27]. Subsequent trials comparing fluconazole with either weekly intravenous (IV) amphotericin B or oral itraconazole for cryptococcal meningoencephalitis demonstrated that fluconazole was the most efficacious antifungal agent for maintenance therapy [14,28]. As an example, in one clinical trial of 108 patients with HIV, the relapse rate among those treated with itraconazole was 23 percent compared with 4 percent of those taking fluconazole [14].

●Duration and monitoring – The minimum duration of maintenance therapy should be at least one year. After that, maintenance therapy can be discontinued in individuals on ART who have a CD4 cell count greater than 100 cells/microL and have achieved an undetectable viral load on ART for more than three months [2,6]. The duration of maintenance therapy if viral load monitoring is not available is discussed below. (See 'Considerations in resource-limited settings' below.)

After discontinuation of azole maintenance therapy, the clinician must continue to follow patients closely. Reinitiation of maintenance therapy is indicated if the patient's CD4 cell count declines to less than 100 cells/microL and/or there is a significant rise in serum cryptococcal antigen titers. The role of serum cryptococcal antigen monitoring in a patient who discontinues treatment is discussed elsewhere. (See "Clinical management and monitoring during antifungal therapy for cryptococcal meningoencephalitis in patients with HIV", section on 'Monitoring of serum cryptococcal antigen'.)

Small trials and observational studies suggest that maintenance therapy for C. neoformans can be safely discontinued in the majority of patients who have a CD4 cell count >100 cells/microL on effective ART. As an example, in an observational study, 4 of 100 patients who discontinued maintenance therapy had a clinical relapse of cryptococcal infection after approximately two years (incidence, 1.53 events per 100 person-years; 95% CI, 0.42-3.92) [29-31].

Considerations in resource-limited settings — In resource-limited settings, regimens that include shorter courses of amphotericin B deoxycholate, flucytosine, and/or higher doses of fluconazole (1200 mg) are used (table 1) [18]. In these settings, cost and limited access can hinder the provision of the preferred induction regimen for cryptococcal meningitis (amphotericin B plus flucytosine for two weeks), as well as the ability to monitor and manage potential toxicities. (See 'Antifungal therapy' above.)

●Induction therapy – The World Health Organization recommends one of the following for induction therapy [8]:

•The preferred regimen includes one week of amphotericin B deoxycholate (1 mg/kg/day) plus flucytosine (100 mg/kg/day, divided into four doses per day), followed by one week of fluconazole (1200 mg daily).

•When IV therapy is difficult to administer, a two-week oral induction regimen using fluconazole (1200 mg daily) plus flucytosine (100 mg/kg/day, divided into four doses per day) is a reasonable alternative.

•If flucytosine is not available, two weeks of amphotericin B deoxycholate (1 mg/kg/day) plus fluconazole (1200 mg daily) can be used.

Dose reduction of flucytosine is required for those with reduced kidney function. (See "Pharmacology of flucytosine (5-FC)", section on 'Dose modification'.)

In some settings, it is not feasible to administer the combination of amphotericin B deoxycholate plus fluconazole (1200 mg daily) for two weeks. When this occurs, amphotericin B can be given for one week with fluconazole (1200 mg daily), followed by one week of fluconazole (1200 mg daily) alone.

The use of shortened and oral regimens for induction therapy were supported in a randomized trial of 721 patients with HIV and cryptococcal meningitis in Africa (the ACTA trial) [18]. This trial compared a two-week amphotericin-based regimen (amphotericin B in combination with fluconazole or flucytosine) with a shortened amphotericin-based regimen (one week of amphotericin B in combination with flucytosine or fluconazole, followed by fluconazole for one week) and a two-week oral regimen (fluconazole plus flucytosine). The antifungal doses used for induction therapy in this trial were 1 mg/kg/day for amphotericin B, 1200 mg/day for fluconazole, and 100 mg/kg/day (in four divided doses) for flucytosine. These induction regimens were followed by consolidation therapy (800 mg/day of fluconazole until ART was started, and then 400 mg/day of fluconazole until week 10). Maintenance therapy with fluconazole (200 mg/day) was continued thereafter.

The one-week regimen that included amphotericin B plus flucytosine was associated with the lowest 10-week mortality (24.2 percent); this was significantly lower than any other amphotericin B group (unadjusted HR, 0.56 [95% CI, 0.35-0.91]). The lower mortality of the one- versus two-week IV regimen may be due in part to reduced toxicity. The mortality at 10 weeks was comparable between the oral regimen and the amphotericin-based regimens. Although these findings are encouraging for the treatment of cryptococcal meningitis in resource-limited settings, they should not be used to modify the preferred 14-day induction period in resource-rich health care systems. (See 'Induction therapy' above.)

These findings were supported in a subsequent network meta-analysis that included 13 studies from primarily resource-limited settings, in which the use of an induction regimen consisting of one week of amphotericin B deoxycholate plus flucytosine followed by fluconazole for one week was associated with the lowest mortality at 10 weeks [32]. A two-week course of fluconazole and flucytosine appeared to have similar efficacy to amphotericin-containing regimens. In this analysis, the combination of amphotericin B deoxycholate plus fluconazole was not associated with a mortality benefit compared with amphotericin B alone. However, other studies have found that this combination therapy is associated with improved CSF sterilization [15].

Additional investigations are underway to evaluate regimens with a reduced duration of polyene therapy. In one phase II trial, patients who received a single dose of AmBisome (10 mg/kg) in combination with fluconazole (1200 mg/day for 14 days) had the same results as those who received AmBisome (3 mg/kg/day) for 14 days in combination with high-dose fluconazole [33]. A large phase III study to further examine the efficacy of short-course polyene therapy in resource-limited settings is underway.

An additional discussion of the efficacy of alternative regimens for induction therapy is found above. (See 'Rationale' above.)

●Consolidation and maintenance therapy – After completing induction therapy, consolidation therapy with fluconazole (800 mg daily) should be continued for eight weeks. In resource-limited settings, routine LP after two weeks of induction therapy to confirm CSF sterilization is not indicated if the patient has had a clear clinical response to treatment.

After completing consolidation therapy, patients should transition to maintenance therapy with a lower dose of fluconazole (200 mg daily) for at least one year. For patients receiving ART, maintenance therapy can be discontinued after a year if the CD4 count is ≥100 cells/microL and the viral load is undetectable. If viral load testing is not available, maintenance therapy should be continued until the CD4 is ≥200 cells/microL [8].

Adverse events — Adverse reactions to the commonly used antifungal agents are summarized here and discussed in detail in separate topic reviews.

●Amphotericin – Amphotericin B deoxycholate is frequently associated with electrolyte disturbances, anemia, renal insufficiency, and infusion site reaction, such as drug fever and rigors. These adverse events are reduced when liposomal preparations are used [20]. The symptoms of an infusion reaction can be minimized or prevented by premedication with acetaminophen (usual adult dose, 650 to 1000 mg orally) and/or diphenhydramine (usual adult dose, 25 to 50 mg orally or IV). The risk of renal dysfunction associated with amphotericin B may be reduced with infusion of normal saline before and during therapy. Additional information on the management of adverse events related to amphotericin B can be found in separate topic reviews. (See "Clinical management and monitoring during antifungal therapy for cryptococcal meningoencephalitis in patients with HIV", section on 'Monitoring for drug toxicity' and "Pharmacology of amphotericin B", section on 'Infusion-related reactions' and "Amphotericin B nephrotoxicity".)

●Flucytosine – Flucytosine is mainly associated with gastrointestinal intolerance. However, laboratory abnormalities that include elevations in aminotransferases, anemia, leukopenia, and thrombocytopenia can also be seen. (See "Clinical management and monitoring during antifungal therapy for cryptococcal meningoencephalitis in patients with HIV", section on 'Monitoring for drug toxicity'.)

●Azoles – Fluconazole is generally well tolerated [26]; patients occasionally may develop rash or abnormal aminotransferases. By contrast, itraconazole is associated with significant gastrointestinal intolerance and pedal edema and requires drug monitoring due to significant differences in bioavailability from person to person. (See "Pharmacology of azoles".)

Patient and laboratory monitoring during induction/consolidation and maintenance therapy is discussed elsewhere. (See "Clinical management and monitoring during antifungal therapy for cryptococcal meningoencephalitis in patients with HIV".)

Patients who have difficulty tolerating amphotericin B and/or flucytosine — Some patients have difficulty tolerating amphotericin B and/or flucytosine. Alternative regimens are available (table 1), although they generally appear suboptimal for CSF sterilization [2,26,34,35]. If an alternative regimen is used, it is important that patients are monitored closely for evidence of treatment failure. (See "Clinical management and monitoring during antifungal therapy for cryptococcal meningoencephalitis in patients with HIV".)

Before changing a patient's regimen, it is important to see if any of the individual's adverse reactions can be managed without a change in therapy. As examples:

●The use of lipid amphotericin may reduce the risk of adverse events related to amphotericin B deoxycholate. Discussions of how to manage infusion reactions and nephrotoxicity are found elsewhere. (See 'Adverse events' above and "Clinical management and monitoring during antifungal therapy for cryptococcal meningoencephalitis in patients with HIV", section on 'Monitoring for drug toxicity' and "Pharmacology of amphotericin B", section on 'Infusion-related reactions' and "Amphotericin B nephrotoxicity", section on 'Salt loading'.)

●For those with an adverse reaction to flucytosine, serum levels may be helpful to assess if a dose adjustment is required. (See 'Adverse events' above and "Clinical management and monitoring during antifungal therapy for cryptococcal meningoencephalitis in patients with HIV", section on 'Monitoring for drug toxicity' and "Pharmacology of flucytosine (5-FC)", section on 'Management of toxicities'.)

If the patient is still unable to tolerate amphotericin B plus flucytosine, the regimen can be modified.

●For those who are unable to tolerate flucytosine, amphotericin B plus fluconazole (800 to 1200 mg per day) is the preferred polyene-azole combination [26]. The 1200 mg per day dosing of fluconazole in this combination regimen has been best studied in resource-limited settings. (See 'Induction therapy' above and 'Considerations in resource-limited settings' above.)

The adverse events associated with amphotericin B plus fluconazole are mainly related to those expected with amphotericin B administration. (See 'Adverse events' above.)

●For those unable to tolerate amphotericin B, we administer oral fluconazole (1200 mg daily) with flucytosine (100 mg/kg daily orally, in four divided doses) [18]. Dose reduction of flucytosine is required for those with reduced kidney function.

Studies evaluating the use of fluconazole plus flucytosine as an alternative regimen include early open-label studies, in which the combination did better than fluconazole monotherapy [34,35]. These studies used fluconazole doses that were in the 200 to 400 mg range. A subsequent study using high doses of fluconazole (1200 mg/day) and flucytosine (100 mg/kg/day) suggests that a combination oral regimen is comparable to a polyene-containing regimen [18]. In this trial, which is described in detail above, the oral combination regimen had a 10-week mortality of approximately 35 percent, which was similar to the mortality reported when induction therapy with certain amphotericin B-containing regimens were used. (See 'Considerations in resource-limited settings' above.)

●On occasion, if neither amphotericin B nor flucytosine are available or can be tolerated, oral fluconazole alone (1200 mg daily) can be used, although this regimen is not as effective as amphotericin B monotherapy or combination therapy [6,36].

The use of fluconazole alone is only fungistatic (ie, inhibits without killing). The importance of fungicidal (eg, amphotericin B) versus fungistatic therapy (eg, fluconazole) for induction therapy was illustrated in a trial of 194 patients who were randomly assigned in a 2:1 ratio to either fluconazole (200 mg/day) or amphotericin B (0.5 mg/kg per day) [37]. In this trial, the mortality was higher within the first two weeks among patients assigned to the fluconazole arm compared with the amphotericin B arm (15 versus 8 percent). In addition, compared with fluconazole, treatment with amphotericin B resulted in more rapid CSF sterilization (42 versus 64 days).

Management of increased intracranial pressure — The intracranial pressure (ICP) should be measured at the time of the initial LP. Patients suspected of having increased ICP and/or CNS mass lesions should first have neuroimaging (eg, computed tomography [CT] scan or magnetic resonance imaging [MRI]) to rule out the presence of a concomitant space-occupying lesion. The use of neuroimaging and LP for initial diagnosis and assessment of ICP in patients with suspected cryptococcal meningoencephalitis, including those in resource-limited settings, is discussed in greater detail elsewhere. (See "Epidemiology, clinical manifestations, and diagnosis of Cryptococcus neoformans meningoencephalitis in patients with HIV", section on 'Diagnosis' and "Epidemiology, clinical manifestations, and diagnosis of Cryptococcus neoformans meningoencephalitis in patients with HIV", section on 'Special considerations in resource-limited settings'.)

Increased ICP in patients with cryptococcal meningoencephalitis should be managed aggressively to decrease mortality [38,39]:

●LP should be performed to reduce the opening pressure to <20 cm CSF; in symptomatic patients with extremely high CSF pressures (eg, ≥50 cm CSF), the goal is to reduce the ICP by 50 percent of the initial value. We perform daily LPs until the patient is asymptomatic and the CSF pressure has been documented to be normal and/or stable.

●Lumbar or ventricular drains may be preferred in patients who require frequent LPs, and they can be lifesaving. Theoretical concerns about placing drains into infected spinal fluid should not preclude their use as long as effective antifungal therapy has been started.

●The further management of patients who have a continued increase in ICP despite antifungal therapy is discussed elsewhere. (See "Clinical management and monitoring during antifungal therapy for cryptococcal meningoencephalitis in patients with HIV", section on 'Management'.)

There is no role for acetazolamide, mannitol, or corticosteroids to reduce increased ICP in this acute setting [6,40]. In addition, routine use of dexamethasone at the start of induction therapy (similar to tuberculous meningitis) is not effective and should not be used, as it can reduce the fungicidal activity of drugs and increase morbidity [5]. However, corticosteroids should be administered to manage CNS immune reconstitution inflammatory syndromes (IRISs). (See "Clinical management and monitoring during antifungal therapy for cryptococcal meningoencephalitis in patients with HIV", section on 'Management of IRIS'.)

The LP, which leads to transient improvement in ICP, can be lifesaving. As an example, in a study of 248 individuals with HIV and cryptococcal meningitis from Uganda and South Africa, the 75 patients who had at least one therapeutic LP in the first week were more likely to survive compared with the 173 patients who did not have a therapeutic LP (adjusted relative risk of mortality 0.31; 95% CI 0.12-0.82) [39]. Whether reduction of ICP has a beneficial effect on other outcomes, such as visual or auditory acuity, is uncertain [2,41,42].

When to initiate antiretroviral therapy — For patients with cryptococcal meningoencephalitis, we suggest antiretroviral therapy (ART) be started between 2 and 10 weeks after antifungal therapy has been initiated [2,6]. The timing of ART initiation must weigh the potential advantage of early immune recovery against the risk of developing IRIS. Although immune recovery is an important part of the successful treatment of cryptococcal meningoencephalitis, trials comparing early versus delayed initiation of ART in patients undergoing treatment for cryptococcal meningoencephalitis consistently show improved survival with delayed initiation [43-45]. This is in contrast to other opportunistic infections (OIs), such as Pneumocystis pneumonia, for which early ART initiation has been found to be clinically beneficial [46]. Discussions of ART selection are found elsewhere. (See "Selecting antiretroviral regimens for treatment-naïve persons with HIV-1: General approach" and "Use and impact of antiretroviral therapy for HIV infection in resource-limited settings".)

Deciding when to initiate ART within the 2- to 10-week window is less clear. For individuals in resource-limited health care systems, ART should generally be started between four to six weeks after initiating antifungal therapy, since the benefits of a more rapid immune recovery usually outweigh the risks of IRIS [8]. Some experts also initiate ART during this time period in resource-rich settings [47]; however, others may delay ART by up to 10 weeks (ie, after induction and consolidation antifungal therapy) in patients who have close follow-up and access to treatment that can prevent further OIs. This approach minimizes the risk of IRIS and avoids drug interactions associated with high-dose antifungal therapy. (See "Clinical management and monitoring during antifungal therapy for cryptococcal meningoencephalitis in patients with HIV", section on 'Immune reconstitution inflammatory syndrome'.)

The benefit of delayed ART was best illustrated in a trial of 177 patients with HIV from Uganda and South Africa who were randomly assigned to receive either early ART (within one to two weeks after initiating antifungal therapy) or delayed ART (four to six weeks after initiating antifungal therapy) [44]. Antifungal therapy in these resource-limited settings included two weeks of induction therapy with amphotericin and fluconazole followed by 12 weeks of consolidation therapy with fluconazole alone. The mortality at 26 weeks was significantly higher in patients who received early compared with delayed ART (45 versus 30 percent; HR 1.73; 95% CI 1.06-2.82). Most excess deaths associated with early ART initiation occurred two to five weeks after initiating antifungal therapy and among patients who had fewer than five white blood cells/microL in their initial CSF sample (a marker of more severe immunosuppression and a risk factor for paradoxical IRIS).

Patient monitoring — Multiple complications can occur during antifungal therapy related to adverse drug-related events. As an example, patients with severe cryptococcal meningoencephalitis may have persistent increased ICP, which can lead to significant morbidity and mortality. Furthermore, some patients who initiate ART can develop an inflammatory CSF profile with symptomatic increased ICP that is related to immune recovery. A full discussion of the symptoms, signs, and management of these complications is found elsewhere. (See "Clinical management and monitoring during antifungal therapy for cryptococcal meningoencephalitis in patients with HIV".)

PREVENTING SYMPTOMATIC DISEASE — The best way to prevent cryptococcal disease in individuals with HIV is through early initiation of antiretroviral therapy (ART) even at high CD4 counts. ART reduces mortality, as well as serious AIDS- and non-AIDS-related complications, and should be initiated regardless of the CD4 count. (See "When to initiate antiretroviral therapy in persons with HIV".)

However, in patients with a low CD4 count, a strategy involving serum or plasma cryptococcal antigen (CrAg) screening and antifungal therapy in addition to ART may further reduce the risk of developing cryptococcal meningitis [48-52]. Cryptococcal antigen is detectable in serum at least three weeks prior to the onset of neurologic symptoms. (See 'Screening and treatment of early infection' below.)

If CrAg screening is not available, primary prophylaxis with antifungal therapy may be warranted in specific high-risk settings. (See 'If screening is not available' below.)

The approach to antifungal therapy below is for nonpregnant adults. Pregnant women should be managed in conjunction with an infectious diseases specialist since fluconazole may be teratogenic during pregnancy, particularly in the first trimester.

Screening and treatment of early infection

Who should be screened — We suggest serum CrAg screening for asymptomatic patients with a CD4 count ≤100 who are not receiving effective ART. There may also be a benefit to screening at a higher CD4 count threshold (eg, <200 cells/microL) in certain areas of high prevalence for cryptococcal disease, such as resource-limited settings where the accuracy of CD4 counts may be reduced [52,53]; at least 28 countries have adopted screening at this threshold.

This approach to screening is in agreement with recommendations from the World Health Organization (WHO) [8]. In the United States, guidelines also support screening asymptomatic patients with a CD4 count <100 cells/microL [47,54]. (See "Epidemiology, clinical manifestations, and diagnosis of Cryptococcus neoformans meningoencephalitis in patients with HIV".)

Approach to preemptive therapy — Patients who test negative for serum CrAg do not require antifungal therapy and should initiate ART. (See "Selecting antiretroviral regimens for treatment-naïve persons with HIV-1: General approach" and "Use and impact of antiretroviral therapy for HIV infection in resource-limited settings".)

Patients who test positive for serum CrAg should undergo a careful history and exam to elicit signs and symptoms of cryptococcal meningitis, as well as a lumbar puncture (LP) (regardless of signs and symptoms) to evaluate for the presence of cerebrospinal fluid (CSF) CrAg or positive CSF cryptococcal cultures. In one study that evaluated the CSF in 90 CrAg-positive patients who were asymptomatic, approximately one-third had evidence of cryptococcal meningitis [55]. Meningitis is particularly associated with a serum CrAg lateral flow assay (LFA) titer ≥1:160 [36,55].

ART should be held pending return of the CSF studies. Antifungal therapy with fluconazole 400 mg daily can be started in asymptomatic patients with CrAg LFA titers <1:160. By contrast, in patients with symptoms or CrAg LFA titers >1:160, it is reasonable to initiate induction amphotericin B therapy pending the CSF results.

Once the CSF studies return, our approach is as follows:

●If the LP is negative (ie, negative CSF CrAg and culture), antifungal therapy with fluconazole (400 mg daily) is warranted, and ART can be started. We discontinue preemptive fluconazole therapy in those patients receiving ART if they achieve a CD4 cell count >100 cells/microL for at least three months.

●If there is evidence of central nervous system involvement (ie, positive CSF CrAg or culture), patients should be treated for cryptococcal meningoencephalitis, and ART should be initiated 2 to 10 weeks after induction treatment, as discussed above. (See 'Induction therapy' above and 'Considerations in resource-limited settings' above and 'When to initiate antiretroviral therapy' above.)

In certain resource-limited settings, it may not be possible to perform an LP on all asymptomatic patients, and the choice of antifungal therapy is based only on the serum titer and the absence of clinical disease. For such patients with a CrAg <1:160, we would initiate ART two weeks after preemptive therapy with fluconazole has been started. For those with a CrAg >1:160, we would encourage LP and, if no LP, then initiate ART two to four weeks after induction therapy. (See 'When to initiate antiretroviral therapy' above.)

Several studies in resource-limited settings with a high prevalence of disease have suggested a benefit of using a strategy of screening and pre-emptive antifungal therapy to prevent the development of cryptococcal meningitis [48-52,56,57]. As examples:

●In a study of 295 ART-naïve patients in Uganda starting HIV therapy, 26 patients with a CD4 cell count ≤100 cells/microL had a positive serum CrAg. All of the patients with a positive serum CrAg were started on ART [49]. In addition, 21 were treated with fluconazole (200 to 400 mg/day) for two to four weeks, whereas five were treated with ART alone. Clinical cryptococcal meningitis developed in three of the fluconazole-treated persons, and the 30-month survival was 71 percent (95% CI 48 to 89 percent). Of the five CrAg-positive persons who were not treated with fluconazole, all died within two months of ART initiation.

●In another trial that evaluated approximately 2000 patients in Tanzania and Zambia with a CD4 count <200 cells/microL, an intervention that included ART, community adherence support, and CrAg testing with preemptive fluconazole was compared with standard of care (ART plus clinic-based visits) [52]. Patients who received the intervention had a significant reduction in mortality at 12 months compared with those who received standard of care (13 versus 18 percent). It is unclear if the mortality benefit was attributable to CrAg screening or enhanced community support. However, CrAg-positive persons who were preemptively treated had a 70 percent one-year survival.

●In a prospective study of 645 ART-naïve patients with a CD4 count ≤100 cells/microL from South Africa, 28 patients tested positive for serum CrAg using the LFA, and 21 received antifungal therapy [51]. Ten patients agreed to LP and four tested positive for CrAg in the CSF and received amphotericin B. Of the remaining 17, all received fluconazole and none developed cryptococcal meningitis B. These findings were compared with a historical control in which 7 of 25 serum CrAg-positive patients (28 percent) who did not receive preemptive antifungal therapy went on to develop cryptococcal meningoencephalitis, and only half were alive in care after six months [58].

Studies evaluating cost-effectiveness have also suggested that this screen-and-treat strategy can be cost effective in resource-limited areas where the prevalence of cryptococcal antigenemia is >3 percent (based on the CrAg latex agglutination test) [49,50,59,60]. These findings were derived from reductions in mortality found in some of the studies above [49,50]. Although the low quality of the evidence results in substantial uncertainty as to the true effect of this approach on mortality, when assessing the cost of the CrAg LFA and the avoided cost of hospitalizations, CrAg screening is likely cost-saving to the health care system [61].

Despite these potential benefits of screening and pre-emptive treatment, the mortality due to cryptococcal disease remains substantial. In one study of 67 CrAg-positive and 134 CrAg-negative patients, the mortality was 25 and 9 percent, respectively; and cryptococcal disease was an immediate or contributing cause of death in 71 percent of those who were CrAg positive, even with screening and fluconazole therapy [62]. A more potent antifungal pre-emptive strategy may be necessary to further reduce mortality. In addition, more consistent, high-quality data are needed to demonstrate the precise efficacy of this approach in all high-risk regions of the world.

If screening is not available — If screening is not available, we suggest not using routine antifungal prophylaxis for primary prevention of cryptococcal infection in resource-rich countries where the incidence of cryptococcal infection is low (eg, the United States). This approach is supported by major guideline panels because of the lack of overall survival benefit and the increased risk of drug interactions, adverse effects, potential antifungal drug resistance, and cost [2,6]. However, in less-resourced areas, where the incidence of disease is high, the WHO recommends fluconazole prophylaxis in adults with HIV and CD4 counts <100 cells/microL [8].

A 2005 Cochrane systematic review identified five randomized controlled trials using antifungal interventions for primary prevention of cryptococcal disease in a total of 1316 patients with HIV [63]. Although the incidence of cryptococcal disease decreased significantly in patients taking either fluconazole or itraconazole, there was no significant effect on overall mortality. However, a subsequent meta-analysis that was described in the 2018 WHO guidelines appears to favor primary prophylaxis in resource-limited settings if there is no preemptive CrAg screening and the patient has a CD4 count <100 cells/microL [8,64].

PREGNANT PERSONS — Pregnant persons should be managed in conjunction with an infectious diseases specialist, since flucytosine and fluconazole may be teratogenic during pregnancy, particularly in the first trimester. In such patients, the duration of amphotericin may need to be extended.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Cryptococcosis" and "Society guideline links: Opportunistic infections in HIV-infected adults and adolescents".)

SUMMARY AND RECOMMENDATIONS

●Patients with HIV and advanced immunosuppression (eg, CD4 cell count <50 cells/microL) are at risk for severe cryptococcal meningoencephalitis, which is uniformly fatal within approximately two weeks if untreated. (See 'Overview' above.)

●Treatment of cryptococcal meningoencephalitis includes antifungal therapy, control of intracranial pressure (ICP), and immune recovery with HIV therapy. (See 'Overview' above.)

●The optimal approach to antifungal therapy involves three phases: induction therapy for approximately two weeks, followed by consolidation therapy for approximately eight weeks, and then maintenance (ie, suppressive) therapy for at least one year to decrease the risk of relapse (table 1). (See 'Antifungal therapy' above.)

●For most patients with cryptococcal meningoencephalitis, we recommend intravenous (IV) amphotericin B with oral flucytosine during the induction phase of therapy (Grade 1B). (See 'Induction therapy' above.)

•For patients with reduced kidney function and risk factors for renal failure, we recommend liposomal amphotericin B (3 to 4 mg/kg intravenously [IV] daily) rather than amphotericin B deoxycholate (Grade 1B). We also suggest liposomal amphotericin B for patients with normal renal function (Grade 2B). Liposomal amphotericin is associated with fewer toxicities compared with amphotericin B deoxycholate, and therefore, therapy is less likely to be interrupted. However, for patients with normal kidney function, amphotericin B deoxycholate (0.7 mg/kg IV daily) is a suitable alternative if liposomal amphotericin is not available. (See 'Regimen' above.)

•The dose of flucytosine is 100 mg/kg per day in four divided doses, adjusted for renal function. If flucytosine is not available or if it is not tolerated, we administer fluconazole (800 mg daily orally or IV) in addition to amphotericin B during the induction phase of therapy. (See 'Regimen' above.)

•For patients in resource-rich settings, we suggest that induction therapy with amphotericin B and flucytosine be administered for at least two weeks (Grade 2C). It should be extended if clinical improvement is not observed and/or if cerebrospinal fluid (CSF) sterilization has not yet been achieved. (See 'Duration' above.)

•For patients in resource-limited settings, we recommend induction therapy with one week of amphotericin B deoxycholate plus flucytosine followed by one week of high-dose oral fluconazole (1200 mg per day) (Grade 1B). If IV therapy is not possible, then a two-week course of fluconazole plus flucytosine may have similar efficacy to some amphotericin-containing regimens. However, flucytosine is often not available, in which case fluconazole (1200 mg/day) should be administered for two weeks in combination with one to two weeks of amphotericin B. (See 'Considerations in resource-limited settings' above.)

●After completing induction therapy, patients should receive consolidation therapy for a minimum of eight weeks. We suggest fluconazole rather than itraconazole (Grade 2C). Fluconazole is generally preferred over itraconazole due to better bioavailability and a reduced risk of drug interactions and gastrointestinal symptoms. The dose of fluconazole is 400 to 800 mg daily, depending upon the induction regimen. (See 'Consolidation therapy' above and 'Considerations in resource-limited settings' above.)

●For patients who complete the induction and consolidation phases of therapy, we recommend fluconazole (200 mg daily) for maintenance treatment compared with no therapy or itraconazole (Grade 1B). In early studies, patients who received maintenance therapy had a much lower incidence of relapse. The minimum duration of maintenance therapy should be at least one year. After that, maintenance therapy can be discontinued in individuals receiving antiretroviral therapy (ART) who have a CD4 cell count greater than 100 cells/microL and have achieved an undetectable viral load on ART for more than three months. (See 'Maintenance therapy' above.)

●Patients with increased ICP should be managed aggressively to reduce the risk of mortality associated with cryptococcal meningoencephalitis. In general, a lumbar puncture (LP) should be performed to reduce the opening pressure to <20 cm CSF; however, in symptomatic patients with extremely high CSF pressures, the goal is to reduce the ICP by 50 percent of the initial value. We perform daily LPs until the patient is asymptomatic and the CSF pressure has been documented to be normal and/or stable. Lumbar or ventricular drains may be preferred in patients who require frequent LPs. (See 'Management of increased intracranial pressure' above.)

●For patients with cryptococcal meningoencephalitis who are not receiving ART, we suggest initiation of ART be delayed at least two weeks after antifungal induction therapy has been started (Grade 2B). For patients with access to close medical follow-up and preventative therapy, we often start ART 10 weeks after the initiation of antifungal therapy to minimize the risk of drug interactions and development of an immune reconstitution inflammatory syndrome (IRIS). However, for individuals without these resources, we typically start ART four to six weeks after induction therapy has been initiated. (See 'When to initiate antiretroviral therapy' above.)

●The best way to prevent cryptococcal disease in individuals with HIV is through early initiation of ART at high CD4 counts. However, in patients with a low CD4 count, a strategy involving serum cryptococcal antigen (CrAg) screening and preemptive antifungal therapy in addition to ART may further reduce the risk of developing cryptococcal meningitis.

•For asymptomatic patients with a CD4 count <100 cells/microL who are not receiving ART, we suggest serum cryptococcal antigen screening prior to initiation of ART (Grade 2C). There may also be a benefit to screening at a higher CD4 count threshold (eg, <200 cells/microL) in certain areas of high prevalence for cryptococcal disease, such as resource-limited settings where the accuracy of CD4 counts may be reduced. The management of patients who screen positive for cryptococcal antigen is discussed above. (See 'Preventing symptomatic disease' above.)

•If screening is not available, we suggest not routinely administering antifungal prophylaxis for prevention of cryptococcal disease in resource-rich countries where the incidence of cryptococcal infection is low (Grade 2B). However, in certain less-resourced areas with a high prevalence of disease, it is reasonable to initiate primary prophylaxis with fluconazole in asymptomatic adults with CD4 counts <100 cells/microL. (See 'If screening is not available' above.)

●Pregnant persons should be managed in conjunction with an infectious diseases specialist, since flucytosine and fluconazole may be teratogenic during pregnancy, particularly in the first trimester. In such patients, the duration of amphotericin may need to be extended. (See 'Pregnant persons' above.)