Cryptococcal osteomyelitis of the scapula and rib in an immunocompetent patient

Cryptococcal osteomyelitis is extremely rare and almost always occurs in immunocompromised patients. We describe a case of osteomyelitis due to Cryptococcus neoformans involving both scapula and rib in an immunocompetent and previously healthy patient. The patient received treatment with amphotericin B deoxycholate and flucytosine for 4 weeks, followed by oral fluconazole 400 mg per day for 8 weeks and 200 mg per day for 9 months. The 12-month course of antifungal therapy resulted in complete clinical recovery and undetectable serum cryptococcal antigen. Cryptococcal osteomyelitis should be suspected in any immunocompetent patient with osteolytic lesions on radiological images.     

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.

BACKGROUND. Intravenous amphotericin B, with or without flucytosine, is usually standard therapy for cryptococcal meningitis in patients with the acquired immunodeficiency syndrome (AIDS). Fluconazole, an oral triazole agent, represents a promising new approach to the treatment of cryptococcal disease. METHODS. In a randomized multicenter trial, we compared intravenous amphotericin B with oral fluconazole as primary therapy for AIDS-associated acute cryptococcal meningitis. Eligible patients, in all of whom the diagnosis had been confirmed by culture, were randomly assigned in a 2:1 ratio to receive either fluconazole (200 mg per day) or amphotericin B. Treatment was considered successful if the patient had had two consecutive negative cerebrospinal fluid cultures by the end of the 10-week treatment period. RESULTS. Of the 194 eligible patients, 131 received fluconazole and 63 received amphotericin B (mean daily dose, 0.4 mg per kilogram of body weight in patients with successful treatment and 0.5 mg per kilogram in patients with treatment failure; P = 0.34). Treatment was successful in 25 of the 63 amphotericin B recipients (40 percent; 95 percent confidence interval, 26 percent to 53 percent) and in 44 of the 131 fluconazole recipients (34 percent; 95 percent confidence interval, 25 percent to 42 percent) (P = 0.40). There was no significant difference between the groups in overall mortality due to cryptococcosis (amphotericin vs. fluconazole, 9 of 63 [14 percent] vs. 24 of 131 [18 percent]; P = 0.48); however, mortality during the first two weeks of therapy was higher in the fluconazole group (15 percent vs. 8 percent; P = 0.25). The median length of time to the first negative cerebrospinal fluid culture was 42 days (95 percent confidence interval, 28 to 71) in the amphotericin B group and 64 days (95 percent confidence interval, 53 to 67) in the fluconazole group (P = 0.25). Multivariate analyses identified abnormal mental status (lethargy, somnolence, or obtundation) as the most important predictive factor of a high risk of death during therapy (P less than 0.0001). CONCLUSIONS. Fluconazole is an effective alternative to amphotericin B as primary treatment of cryptococcal meningitis in patients with AIDS. Single-drug therapy with either drug is most effective in patients who are at low risk for treatment failure. The optimal therapy for patients at high risk remains to be determined.     

Comparative evaluation of three antifungal susceptibility test methods for Candida albicans isolates and correlation with response to fluconazole therapy.

In vitro susceptibilities were determined for 56 Candida albicans isolates obtained from the oral cavities of 41 patients with human immunodeficiency virus infection. The agents tested included fluconazole, itraconazole, ketoconazole, flucytosine, and amphotericin B. MICs were determined by the broth microdilution technique following National Committee for Clinical Laboratory Standards document M27-P (M27-P micro), a broth microdilution technique using high-resolution medium (HR micro), and the Etest with solidified yeast-nitrogen base agar. The in vitro findings were correlated with in vivo response to fluconazole therapy for oropharyngeal candidiasis. For all C. albicans isolates from patients with oropharyngeal candidiasis not responding to fluconazole MICs were found to be > or = 6.25 micrograms/ml by the M27-P micro method and > or = 25 micrograms/ml by the HR micro method as well as the Etest. However, for several C. albicans isolates from patients who responded to fluconazole therapy MICs found to be above the suggested breakpoints of resistance. The appropriate rank order of best agreement between the M27-P micro method and HR micro method was amphotericin B > fluconazole > flucytosine > ketoconazole > itraconazole. The appropriate rank order with best agreement between the M27-P micro method and the Etest was flucytosine > amphotericin B > fluconazole > ketoconazole > or = itraconazole. It could be concluded that a good correlation between in vitro resistance and clinical failure was found with all methods. However, the test methods used in this study did not necessarily predict clinical response to therapy with fluconazole.     

Management of cryptococcosis in non-HIV-related patients.

Cryptococcosis is the third most common deep fungal infection in China and has not yet been reported to be associated with HIV infection there. We report the management of non-HIV-related cryptococcosis, including cutaneous cryptococcosis, pulmonary cryptococcosis and cryptococcal infection of central nervous system (CNS). Establishment of the diagnosis of cutaneous cryptococcosis and pulmonary cryptococcosis were mainly based on the histopathologic examination and mycologic culture, with CNS cryptococcal infection based on mycologic examination and latex agglutination test on cerebrospinal fluid. The treatment of cutaneous cryptococcosis included systemic administration of amphotericin B (AMB), 5-flucytosine and triazole agents such as fluconazole and itraconazole combined with topical ketoconazole cream. Treatment of pulmonary cryptococcosis included systemic use of antifungal medication combined with surgical removal of pulmonary lesions. The treatment of CNS cryptococcal infection was challenging. In this study, 53 patients with CNS cryptococcal infection were divided into three groups according to the antifungal regimens applied: eight patients (group I) received intravenous AMB alone or in combination with 5-flucytosine, five patients (group II) received intravenous fluconazole alone or with 5-flucytosine, and 40 patients (group III) received a two-phase therapy, active therapy and consolidation therapy. In active therapy, the patients received intrathecal and intravenous administration of AMB alone or with 5-flucytosine until the mycological culture of cerebrospinal fluid (CSF) became negative. Consolidation therapy followed active therapy by continuous use of oral fluconazole or itraconazole until direct microscopic examination of CSF was negative for three consecutive weeks. In group I, five patients were cured, two improved, one died and one had relapse. In group II, two patients were cured, one improved and two died. In group III, thirty-nine out of forty patients were cured without recurrence. These results indicate that the two-phase protocol was more desirable for the treatment of non-HIV associated cryptococcal infection of CNS.     

A comparison of amphotericin B alone and combined with flucytosine in the treatment of cryptoccal meningitis.

We compared amphotericin B therapy for cryptococcal meningitis with a newer regimen containing both amphotericin B and flucytosine. In 50 patients with 51 courses of therapy adherent to the protocol, 27 courses were with amphotericin B and 24 with the combination. Even though the combination regimen was given for only six weeks and amphotericin B for 10 weeks, the combination cured or improved more patients (16 vs 11), produced fewer failures or relapses (three vs. 11), more rapid sterilization of the cerebrospinal fluid (P less than 0.001) and less nephrotoxicity (P less than 0.05) than did amphotericin B alone. The number of deaths was the same (five) with each regimen. Adverse reactions to flucytosine occurred in 11 of 34 patients but were not life threatening. We conclude that combined flucytosine-amphoericin B therapy is the regimen of choice in cryptococcal meningitis.     

Evaluation of a capacitance method for direct antifungal susceptibility testing of yeasts in positive blood cultures

The feasibility of using a capacitance method (CM) for direct antifungal susceptibility testing of yeasts in positive blood cultures was evaluated. The CM used the same test conditions as those recommended by the National Committee for Clinical Laboratory Standards. After direct inoculation of positive culture broths into module wells (Bactometer; bioMérieux, Inc., Hazelwood, Mo.), the end-point determination was made by monitoring the capacitance change in the culture broths with Bactometer. The MIC of amphotericin B was the lowest concentration at which yeast growth was completely inhibited, while the MICs of ketoconazole, flucytosine, and fluconazole were the concentrations at which a >/=80% reduction in capacitance change was observed. The MICs of the four drugs against each blood isolate obtained on subculture plates were also determined by the macrodilution method. For 51 positive blood cultures tested, the percent agreement (+/-2 log(2) dilutions) between the CM and the macrodilution method were as follows: amphotericin B (98%), ketoconazole (92%), flucytosine (84%), and fluconazole (96%). The CM was further used for breakpoint susceptibility testing of fluconazole (8 and 64 microg/ml) and flucytosine (4 and 32 microg/ml) against yeasts in positive blood cultures. After testing of 74 specimens by the CM, flucytosine and fluconazole produced one (1.4%) major error and two (2.8%) minor errors, respectively. All yeasts that displayed resistance to flucytosine or fluconazole were detected within 24 h after direct inoculation of the positive broths into Bactometer. The CM may be useful for the rapid detection of antifungal resistance in positive blood cultures containing yeasts.     

Treatment of orogastrointestinal candidosis in SCID mice with fluconazole alone or in combination with recombinant granulocyte colony-stimulating factor or interferon-gamma.

Mucosal candidosis is common in acquired immune deficiency syndrome (AIDS) patients, where there is extensive mucosal involvement, but rarely dissemination. To mimic this disease, SCID mice were inoculated orally with Candida albicans, which could be recovered from standardized tissue samples of the esophagus, stomach, small intestine and caecum of all mice. Treatment with fluconazole at 5 or 10 mg kg(-1) per day were equivalent to each other and efficacious in reducing the fungal burden from all four tissues compared with no treatment or lower doses of fluconazole (P < 0.01-0.001). Fluconazole at 5 or 10 mg kg(-1) reduced fungal burden in the stomach by about 200 or 580-fold, respectively, and by approximately 25-fold in the other tissues, with 80 or 100% of mice cleared of esophageal infection, and 40 or 80% cleared of infection in the small intestine, respectively; the same doses cleared < or =20% of stomach infection and none of caecal infection. Treatment with recombinant human granulocyte colony-stimulating factor (G-CSF) up to 500 microg kg(-1) per day or 10(5) U of murine interferon-gamma (IFN-gamma) alone was ineffective, nor were combinations with a suboptimal dose fluconazole synergistic. Overall, fluconazole had dose-responsive efficacy, whereas neither G-CSF nor IFN-gamma alone or in combination with fluconazole improved efficacy. These studies demonstrate the utility of this model for examining antifungal efficacy in a situation that mimics clinical disease in AIDS patients.     

Genetic dissimilarity of two fluconazole-resistant Candida albicans strains causing meningitis and oral candidiasis in the same AIDS patient.

We describe a patient with AIDS who simultaneously developed Candida meningitis with three positive cerebrospinal fluid cultures and oral candidiasis. This patient also had a history or recurrent episodes of oral candidiasis treated with fluconazole. The patient did not respond to this therapy but was cured with amphotericin B and flucytosine. In vitro susceptibility tests revealed that each infection was caused by fluconazole-resistant Candida albicans isolates. Strain delineation by karyotyping, NotI restriction pattern analysis, hybridization with the specific probe 27A, and PCR fingerprinting with the phage M13 core sequence clearly demonstrated that meningitis and oral thrush were caused by two genetically different isolates.     

Genotyping and antifungal susceptibility testing of Cryptococcus neoformans isolates from Cameroonian HIV-positive adult patients

Cryptococcus neoformans is the most common cause of meningitis amongst adult Africans with HIV/AIDS. The widespread use of fluconazole may lead to the emergence of isolates with reduced susceptibility. We studied C. neoformans isolates from HIV-infected patients with cryptococcal meningitis. Genotyping and antifungal testing were performed to assess the genetic diversity, occurrence of mixed infections and in vitro activity of antifungal agents. Isolates were recovered from cerebrospinal fluid prior to systemic antifungal treatment. Six isolates were studied for each sample (a total of 114 isolates from 19 patients). Serotyping was performed via LAC 1 and CAP 64 gene amplification and genotyping was performed using phage M13 core, (GACA)₄ and (GTG)₅ primers and restriction polymorphism analysis of the URA5 gene. Susceptibilities for amphotericin B, flucytosine, fluconazole, voriconazole and posaconazole were tested by the Sensititre YeastOne® method. All strains were identified as C. neoformans var. grubii serotype A. We identified nine major genotypes. Up to two genotypes were identified in the same sample. None of the isolates were resistant to the studied drugs. However, 13 of 114 strains exhibited a reduced susceptibility to fluconazole and 13 of 114 strains exhibited a reduced susceptibility to flucytosine. No correlation was found between the genotype and susceptibility. This study confirms the prevalence of C. neoformans serotype A in Cameroon. Two genotypes may be responsible for a single episode of cryptococcosis. The possibility of mixed infection and diminished susceptibility to fluconazole or flucytosine must be considered for the management of cryptococcosis.     

Global trends in the antifungal susceptibility of Cryptococcus neoformans (1990 to 2004)

The antifungal susceptibilities of 1,811 clinical isolates of Cryptococcus neoformans obtained from 100 laboratories in 5 geographic regions worldwide between 1990 and 2004 were determined. The MICs of amphotericin B, flucytosine, fluconazole, voriconazole, posaconazole, and ravuconazole were determined by the National Committee for Clinical Laboratory Standards broth microdilution method. Isolates were submitted to a central reference laboratory (University of Iowa) from study centers in Africa (5 centers, 395 isolates), Europe (14 centers, 102 isolates), Latin America (14 centers, 82 isolates), the Pacific region (7 centers, 50 isolates), and North America (60 centers, 1,182 isolates). Resistance to amphotericin B, flucytosine, and fluconazole was < or = 1% overall. Susceptibility to flucytosine (MIC, < or = 4 microg/ml) ranged from 35% in North America to 68% in Latin America. Similarly, only 75% of isolates from North America were susceptible to fluconazole (MIC, < or = 8 microg/ml) compared to 94 to 100% in the other regions. Isolates remained highly susceptible to amphotericin B (99% susceptibility at a MIC of < or = 1 microg/ml) over the entire 15-year period. Susceptibility to flucytosine (MIC, < or = 4 microg/ml) increased from 34% in 1990 to 1994 to 66% in 2000 to 2004. Susceptibility to fluconazole (MIC, < or = 8 microg/ml) increased from 72% in 1990 to 1994 to 96% in 2000 to 2004. Voriconazole, posaconazole, and ravuconazole all were very active (99% of isolates susceptible at MIC of < or = 1 microg/ml) against this geographically diverse collection of isolates. We conclude that in vitro resistance to antifungal agents used in the treatment of cryptococcosis remains uncommon among isolates of C. neoformans from five broad geographic regions and has not increased over a 15-year period.     

Granulomatous encephalitis caused by Bipolaris hawaiiensis

We describe a case of granulomatous encephalitis caused by Bipolaris (Drechslera) hawaiiensis in an immunocompetent patient. An 18-year-old man with a seven-month history of seizures and right leg weakness was found by computed tomographic scan to have a left frontoparietal enhancing lesion. Biopsy of the lesion revealed granulomatous inflammation and numerous septate hyphae. Culture of the biopsy specimen yielded a pure culture of B hawaiiensis in four days. Susceptibility studies revealed the organism to be sensitive to amphotericin B (minimal inhibitory concentration [MIC] equals 0.25 mg/L) and miconazole lactate (MIC equals 0.064 mg/L), but resistant to flucytosine (MIC greater than 100 mg/L). No synergy was demonstrated with amphotericin B and flucytosine in vitro. The patient was successfully treated with surgery and systemic and intrathecal amphotericin B therapy, and a negative culture was obtained from a repeated brain biopsy six weeks later.     

Use of fluconazole in the treatment of non-AIDS cryptococcal meningitis.

The treatment of non-acquired immunodeficiency syndrome (AIDS) cryptococcal meningitis has not been well defined as yet. The current recommendation is amphotericin B with or without flucytosine. However, due to the many side effects of these drugs, fluconazole is becoming an attractive alternative in patients who cannot tolerate amphotericin B and flucytosine. This case study demonstrates the successful use of fluconazole in a patient with underlying diabetes mellitus and cryptococcal meningitis.     

Fluconazole versus itraconazole for the prevention of fungal infections in haemato-oncology

AIMS: To compare the efficacy of and tolerance to oral fluconazole and intraconazole in preventing fungal infection in neutropenic patients with haematological malignancies. PATIENTS: 213 consecutive, afebrile adult patients treated with or without autologous stem cell transplantation for haematological malignancies. METHODS: A randomised, double blind, single centre study. Patients were randomly assigned to receive fluconazole 50 mg or itraconazole 100 mg, both twice daily in identical capsules. An intention to treat analysis was performed on 202 patients, 101 in each group. RESULTS: Microbiologically documented systemic fungal infections occurred in four patients in each group. Clinical fungal infection was thought to be present in seven recipients of fluconazole and four of itraconazole. In all 202 patients, 29 proceeded to intravenous amphotericin (amphotericin B), 16 in the fluconazole group and 13 in the itraconazole group. Superficial fungal infection was seen only in three non-compliant patients in the fluconazole group. All these infections were oral. No major differences were noted in the isolates of fungi in mouth washes and fecal samples. Overall mortality was 8.9% (18 deaths; seven in the fluconazole group, 11 in the itraconazole group). Mortality from microbiologically and clinically documented fungal infection was 4.5% (nine deaths; three in the fluconazole group, six in the itraconazole group). Median time to suspected or proven fungal infection was 16 days in both groups. None of these comparisons reached statistical significance (p < 0.05). No major clinical toxicity was noted and compliance was excellent. CONCLUSIONS: In neutropenic patients treated for haematological malignancies with or without autologous stem cell transplantation, fluconazole and itraconazole in low doses result in a similar low frequency of fungal disease. Fluconazole may be the preferable drug because of the smaller number of capsules and lack of need for timing relative to meals.     

In vitro antifungal susceptibility testing of Candida blood isolates and evaluation of the E-test method.

Fungal infections have dramatically increased in recent years, along with the increase of drug-resistant isolates in immunocompromised patients. Ninety eight Candida species obtained from blood cultures at the Tri-Service General Hospital, Taiwan, from 1998 to 2000 were studied. These included 50 Candida albicans, 13 Candida glabrata, 24 Candida tropicalis and 11 Candida parapsilosis isolates. To investigate their susceptibility to commonly used antifungal drugs, minimum inhibitory concentrations (MIC) of amphotericin B, fluconazole, flucytosine, and ketoconazole were determined. Both the National Committee for Clinical Laboratory Standards reference broth macrodilution method and E-test were used in parallel. Ninety five isolates (95/98, 96.94%) were susceptible to amphotericin B at a concentration < or = 1 microg/mL. All isolates (100%, 98/98) were susceptible to flucytosine. Approximately 30% of these Candida isolates were resistant to fluconazole. The MIC for 90% of isolates (MIC90) values for both methods for these isolates were 0.5 microg/mL for amphotericin B, 32 microg/mL for fluconazole, 0.25 microg/mL for flucytosine (0.125 microg/mL by E-test method), and 4 microg/mL for ketoconazole. MIC for 50% of isolates (MIC50) values for these agents were 0.25, 2, 0.06, and 0.06 microg/mL, respectively. The essential agreement of MIC values within 2 dilutions for the 2 methods was 99.0% for amphotericin B, 90.8% for ketoconazole, 92.9% for fluconazole, and 91.8% for flucytosine. This study showed that E-test has equivalent performance to the broth macrodilution method and can be used as an alternative MIC technique for antifungal susceptibility testing.     

Susceptibility testing of Cryptococcus neoformans: a microdilution technique.

We studied a series of test conditions in a microtiter system to define the optimal method for determining the susceptibility of Cryptococcus neoformans to antifungal agents. Twenty-one isolates of C. neoformans were grown for 24 or 48 h in four chemically defined media: yeast nitrogen base (BYNB 7); RPMI 1640; synthetic amino acid medium--fungal (SAAMF), buffered at pH 7.0 to select the medium that best supported growth of this fastidious yeast; and yeast nitrogen base, pH 5.4 (YNB 5.4). Maximum growth of C. neoformans, at 35 degrees C, was obtained in YNB 5.4, with the next highest growth levels in BYNB 7, SAAMF, and RPMI. Growth at 24 h was uniformly poor in all media and lacked reproducibility. In contrast, incubation for 48 h gave adequate growth with low standard deviations, and 48 h was selected as the optimal incubation period for this study. Comparison of the relationship between growth kinetics and initial inoculum size for eight cryptococcal isolates showed that 10(4) cells per ml yielded optimal growth in BYNB 7 and YNB 5.4, whereas 10(5) cells per ml was optimal in RPMI and SAAMF. Furthermore, variation of inocula from 10(3) to 10(5) cells per ml showed small but significant inoculum effects in determining MICs of fluconazole, amphotericin B, and flucytosine for C. neoformans. Therefore, 10(4) cells per ml was chosen as the optimal inoculum for susceptibility testing in this study. Mean MICs of fluconazole, amphotericin B, and flucytosine for 21 crytococcal isolates in RPMI and BYNB 7 were low (for example, fluconazole had mean MICs of 1.2 and 1.3 micrograms/ml in RPMI and BYNB 7, respectively) and differed significantly from medium to medium. In contrast, the MICs obtained in SAAMF were significantly higher (e.g., fluconazole had a mean MIC of 2.2 micrograms/ml). Variance in MICs was large with fluconazole and flucytosine but small with amphotericin B, irrespective of the medium used. A microtiter system employing BYNB 7 as the medium, 48 h as the incubation period, and 10(4) cells per ml as the final inoculum is a simple, accurate, and reproducible method for the testing of C. neoformans susceptibility to fluconazole, amphotericin B, and flucytosine.     

Antifungal combination therapy with granulocyte colony-stimulating factor and fluconazole in experimental disseminated candidiasis

The present studies were performed to determine whether any beneficial interaction results from combination therapy with recombinant human granulocyte colony-stimulating factor (GCSF) and fluconazole in disseminated candidiasis in mice. Fluconazole extended survival and reduced tissue counts. GCSF had no direct benefit, but GCSF combined with fluconazole extended survival beyond that for fluconazole alone and reduced renal tissue counts below those for fluconazole alone. Thus, GCSF and fluconazole may be a useful combined therapy for disseminated candidiasis.     

Infections with Cryptococcus neoformans in the acquired immunodeficiency syndrome

We reviewed the records of 106 patients with cryptococcal infections and the acquired immunodeficiency syndrome (AIDS) treated at San Francisco General Hospital. We examined four issues: the efficacy of treatment with amphotericin plus flucytosine as compared with amphotericin alone, the efficacy of suppressive therapy, the prognostic clinical characteristics, and the course of nonmeningeal cryptococcosis. In 48 of the 106 patients (45 percent), cryptococcosis was the first manifestation of AIDS. Among the 89 patients with cryptococcal meningitis confirmed by culture, survival did not differ significantly between those treated with amphotericin plus flucytosine (n = 49) and those treated with amphotericin alone (n = 40). Flucytosine had to be discontinued in over half the patients because of cytopenia. Long-term suppressive therapy with either ketoconazole or amphotericin was associated with improved survival, as compared with survival in the absence of suppressive therapy (median survival, greater than or equal to 238 vs. 141 days; P less than 0.004). The only clinical features independently associated with a shorter cumulative survival were hyponatremia and a positive culture for cryptococcus from an extrameningeal source. The 14 patients with nonmeningeal cryptococcosis had a median survival (187 days) and rate of relapse (20 percent) similar to those in the patients with meningitis (165 days and 17 percent, respectively). From this retrospective study of cryptococcal infections in patients with AIDS we conclude that the addition of flucytosine to amphotericin neither enhances survival nor prevents relapse, but long-term suppressive therapy appears to benefit these patients.     

Subcutaneous Hyalohyphomycosis Caused by Colletotrichum gloeosporioides

The coelomycete Colletotrichum gloeosporioides was isolated in pure culture from subcutaneous nodules of the left forearm and elbow of a farmer after traumatic injury. To our knowledge, we report the first case involving this fungus as an etiological agent of subcutaneous infection. The in vitro inhibitory activities of amphotericin B, itraconazole, ketoconazole, miconazole, flucytosine, and fluconazole were studied.     

A placebo-controlled trial of maintenance therapy with fluconazole after treatment of cryptococcal meningitis in the acquired immunodeficiency syndrome. California Collaborative Treatment Group

BACKGROUND AND METHODS. In patients with the acquired immunodeficiency syndrome (AIDS), the rate of relapse after primary treatment for cryptococcal meningitis remains high. We conducted a controlled, double-blind trial to evaluate the efficacy of maintenance therapy with fluconazole. At entry into the study, all participants had sterile cultures of cerebrospinal fluid, blood, and urine after following a standardized course of therapy for culture-proved cryptococcal meningitis. The patients were randomly assigned to take either fluconazole or placebo as maintenance therapy. The dose of fluconazole was 100 mg daily in the first phase of study and 200 mg daily in the second phase. RESULTS. Of 84 patients initially enrolled, 16 (19 percent) were found to have silent, persistent infection on the basis of cultures that became positive after entry into the study; 7 other patients were lost to follow-up shortly after entry. Of the remaining 61 patients, 10 of 27 assigned to placebo (37 percent) and 1 of 34 assigned to fluconazole (3 percent) had a recurrence of cryptococcal infection at any site (difference in risk, 34 percent; 95 percent confidence interval, 15 to 53). Of the 11 recurrent infections, 7 were detected in urine obtained after prostatic massage. There were four recurrent meningeal infections in the patients taking placebo, but none in those taking fluconazole (mean duration of follow-up, 164 days) (P = 0.03). In multivariate analyses, the best predictors of recurrence-free survival were fluconazole treatment (P = 0.02; relative hazard, 13.2), a lower serum cryptococcal-antigen titer (P = 0.05; relative hazard, 1.2), and more prolonged primary therapy with flucytosine (P = 0.09; relative hazard, 1.1). Survival and toxicity were similar in the two maintenance-treatment groups. CONCLUSIONS. In patients with AIDS, silent persistent infection is common after clinically successful treatment for cryptococcal meningitis. Maintenance therapy with fluconazole is highly effective in preventing recurrent cryptococcal infection.