MAJOR ARTICLE HIV/AIDS

Comparison of the Early Fungicidal Activity of

High-Dose Fluconazole, Voriconazole, and
Flucytosine as Second-Line Drugs Given in
Combination With Amphotericin B for the
Treatment of HIV-Associated Cryptococcal

Downloaded from https://academic.oup.com/cid/article/54/1/121/366746 by guest on 25 October 2020

Meningitis
Angela Loyse,1,2,3,4 Douglas Wilson,3 Graeme Meintjes,1,5,6 Joseph N. Jarvis,4 Tihana Bicanic,4 Leesa Bishop,3
Kevin Rebe,1,5 Anthony Williams,1 Shabbar Jaffar,7 Linda-Gail Bekker,2 Robin Wood,2 and Thomas S Harrison4
1Infectious Diseases Unit, GF Jooste Hospital, Cape Town, 2Desmond Tutu HIV Centre, Institute of Infectious Disease and Molecular Medicine,

University of Cape Town, 3Department of Medicine, Edendale Hospital, Pietermaritzburg, South Africa; 4International Health Group, Research Centre for
Infection and Immunity, Division of Clinical Sciences, St. George's University of London, United Kingdom; 5Division of Infectious Diseases and HIV
Medicine, Department of Medicine, and 6Institute of Infectious Diseases and Molecular Medicine, University of Cape Town, South Africa; and 7Faculty
of Epidemiology and Population Health, London School of Hygiene and Tropical Medicine, United Kingdom

(See the Editorial Commentary by Bennett, on pages 129–30.)

Background. HIV-associated cryptococcal meningitis is associated with an estimated 600 000 deaths worldwide
per year. Current standard initial therapy consists of amphotericin B (AmB) plus flucytosine (5-FC), but 5-FC
remains largely unavailable in Asia and Africa. Alternative, more widely available, and/or more effective antifungal
combination treatment regimens are urgently needed.
Methods. Eighty HIV-seropositive, antiretroviral naive patients presenting with cryptococcal meningitis were
randomized to 4 treatment arms of 2 weeks duration: group 1, AmB (0.7–1 mg/kg) and 5-FC (25 mg/kg 4 times
daily); group 2, AmB (0.7–1 mg/kg) and fluconazole (800 mg daily); group 3, AmB (0.7–1 mg/kg) and fluconazole
(600 mg twice daily); and group 4, AmB (0.7–1 mg/kg) and voriconazole (300 mg twice daily). The primary end
point was the rate of clearance of infection from the cerebrospinal fluid (CSF) or early fungicidal activity (EFA), as
determined by results of serial, quantitative CSF cryptococcal cultures.
Results. There were no statistically significant differences in the rate of clearance of cryptococcal colony-forming
units (CFU) in CSF samples among the 4 treatment groups; the mean (6standard deviation) EFA for treatment groups
1, 2, 3, and 4 were 20.41 6 0.22 log CFU/mL CSF/day, 20.38 6 0.18 log CFU/mL CSF/day, 20.41 6 0.35 log CFU/
mL CSF/day, and 20.44 6 0.20 log CFU/mL CSF/day, respectively. Overall mortality was 12% (9 of 78 patients died)
at 2 weeks and 29% (22 of 75 patients died) at 10 weeks, with no statistically significant differences among groups.
There were few laboratory abnormalities related to the second agents given; in particular, there were no statistically
significant ($grade 3) increases in alanine transaminase level or decreases in neutrophil count.
Conclusions. There was no statistically significant difference in EFA between AmB in combination with
fluconazole and AmB plus 5-FC for the treatment of HIV-associated cryptococcal meningitis. AmB plus fluconazole
(800–1200 mg/day) represents an immediately implementable alternative to AmB plus 5-FC. AmB plus
voriconazole is an effective alternative combination in patients not receiving interacting medications.

Received 5 April 2011; accepted 19 August 2011; electronically published 3 Clinical Infectious Diseases 2012;54(1):121–8
November 2011. Ó The Author 2011. Published by Oxford University Press on behalf of the Infectious
Correspondence: Angela Loyse, MD, Bsc, MRCP, International Health Group, Research Diseases Society of America. All rights reserved. For Permissions, please e-mail:
Centre for Infection and Immunity, Division of Clinical Sciences, St. George's University of journals.permissions@oup.com.
London, Blackshaw Rd, London, SW17 0QT, United Kingdom (angelaloyse@hotmail.com). DOI: 10.1093/cid/cir745

HIV/AIDS d CID 2012:54 (1 January) d 121

There are an estimated 900 000 cases of cryptococcal meningitis Written informed consent was obtained from each patient or
(CM) worldwide per year [1]. The highest burden is in sub- next of kin for patients with altered mental status (Glasgow
Saharan Africa, where it is the most common cause of adult coma scale ,15).
meningitis in many areas [2–5], with an estimated 720,000 cases
and 500,000 deaths per year [1]. This is in keeping with pro- Interventions
spective mortality data that suggest that 10%–20% of all deaths Patients were randomized individually using a computer-
among HIV-infected patients in Africa are attributable to generated program to 1 of 4 treatment arms:
cryptoccocal infection [6–8]. regimen 1: AmB (0.7 or 1 mg/kg/day) plus flucytosine
Current recommended therapy for CM consists of 2 weeks of (25 mg/kg 4 times daily) for 2 weeks,
amphotericin B (AmB) in combination with flucytosine (5-FC) regimen 2: AmB (0.7 or 1 mg/kg/day) plus fluconazole
[9]. However, 5-FC is currently not available in most developing (800 mg daily) for 2 weeks,
countries in Asia and Africa, where CM is most prevalent. This is regimen 3: AmB (0.7 or 1 mg/kg/day) plus fluconazole
in contrast to fluconazole, which is widely available, in generic (600 mg twice daily) for 2 weeks, and
formulation or through a donation program [10]. Voriconazole regimen 4: AmB (0.7 or 1 mg/kg/day) plus voriconazole

Downloaded from https://academic.oup.com/cid/article/54/1/121/366746 by guest on 25 October 2020

is a relatively new broad-spectrum triazole with excellent in vitro (300 mg twice daily; 400 mg twice on day 1) for 2 weeks.
and in vivo activity against Cryptococcus neoformans and good For patients unable to receive oral medication, the second-
cerebrospinal fluid (CSF) penetration [11–14]. Voriconazole has line agents were given by nasogastric tube. Randomization of
yet to be evaluated in comparative studies for the treatment of patients receiving rifampicin was restricted to the first 3 arms of
CM. the study because of the significant interaction between rifam-
Therefore, the primary objective of this study was to compare picin and voriconazole. Randomization was stratified by altered
the rate of clearance of infection from the CSF (early fungicidal mental status at admission. After 2 weeks, patients in all treat-
activity [EFA]) of 3 AmB-containing antifungal regimens ment arms received fluconazole (400 mg daily for 8 weeks and
(AmB 1 fluconazole [800 mg daily], AmB 1 fluconazole 200 mg daily thereafter). These fluconazole consolidation and
[600 mg twice daily], and AmB 1 voriconazole [300 mg twice maintenance doses were increased by 50% in patients receiving
daily]) with the EFA of AmB and 5-FC and to examine the safety concurrent rifampicin.
of these regimens. EFA is associated with 2- and 10-week clinical To reduce significant AmB-related anemia, patients were al-
outcome [15] and provides a means to compare the antifungal located to AmB (0.7 mg/kg/day or 1 mg/kg/day) according to
activity of alternate regimens in small phase II studies [16]. their baseline hemoglobin level. Patients with a baseline hemo-
globin level ,8 g/dL received AmB (0.7 mg/kg), whereas pa-
tients with a hemoglobin level $8 g/dL received AmB (1 mg/kg).
MATERIALS AND METHODS Unless contraindicated, patients received 1 L of 0.9% normal
saline daily to minimize AmB nephrotoxicity. Potassium and
Participants
magnesium were supplemented as required (20 mmol KCl was
From August 2006 through October 2008, 21 study patients
added to the 1 L normal saline, unless the serum potassium level
were recruited at GF Jooste Hospital in Cape Town, Western
was high). If the serum creatinine level increased to .2.5 mg/dL
Cape, South Africa, and 69 patients at Edendale Hospital in
(220 lmol/L), despite adequate hydration and saline loading,
Pietermaritzburg, Kwazulu-Natal, South Africa. The study was
AmB-based induction therapy was discontinued and the pa-
approved by the Research Ethics Committees of the University
tient switched early to fluconazole consolidation therapy. 5-FC
of Cape Town, the University of Kwazulu-Natal, Edendale
was adjusted for creatinine clearance according to standard
Hospital, St George’s Hospital (London), the Kwazulu-Natal
protocols [17].
Department of Health, and the Medicines Control Council of
Follow-up lumbar punctures were performed on days 3, 7,
South Africa. The trial was registered (ISRCTN68133435).
and 14. Patients with CSF opening pressure .35 cm H2O and/or
HIV-infected patients aged $18 years who were hospitalized
headache or other symptoms attributable to increased pressure
with a first episode of CM diagnosed by CSF India ink and had
had additional lumbar punctures performed. After hospital
received #3 doses of AmB were eligible for enrollment. CM was
discharge, participants were counseled, initiated on ART from 2
confirmed by CSF culture for C. neoformans. Patients were ex-
weeks after the start of antifungal therapy [18], and followed up
cluded for alanine aminotransferase levels .5 times upper limit
for 6 months from study enrolment.
of normal (.200 IU/L), absolute neutrophil count,500 3 106
cells/L, platelet count ,50 000 3 106 platelets/L, pregnancy,
Evaluation and Outcomes
lactation, previous serious reaction to study drugs, or prior or
All participants had complete blood count, electrolyte level, urea
current use of antiretroviral therapy (ART).
and creatinine levels, liver enzyme levels, CD4 cell count, and

122 d CID 2012:54 (1 January) d HIV/AIDS

HIV load measured at baseline. Subsequently, renal function was (62%) in treatment group 4 received AmB before enrolment,
evaluated on alternate days, and liver enzyme levels and full compared with the other treatment groups. Treatment group
blood count were measured twice weekly. 3 had higher (median, 5.54 logCFU/mL CSF) and treatment
CSF samples were analyzed for cell count and differential, group 4 lower (median, 4.49 logCFU/mL CSF) baseline
protein, glucose, India ink, cryptococcal antigen titer, and fungal burdens (Table 1, P 5 .03). There were no other sta-
quantitative fungal culture, as described elsewhere [16]. Cryp- tistically significant differences among the 4 treatment arms
tococcal clearance rates were calculated using a summary sta- with regard to established prognostic factors for CM (abnormal
tistic for each patient, defined as the decrease in log CFU per mL mental status, CSF white blood cell count, and CSF opening
CSF per day, with use of the slope of the linear regression of log pressure) or HIV infection (CD4 cell count and HIV load).
CFU against time for each patient. Percentage change in labo-
ratory values was calculated using the following formula: (final EFA
value-baseline value/baseline value) 3 100. There were no statistically significant differences in the rate of
The primary outcome measure was the mean rate of decrease clearance of cryptococcal CFU from the CSF among the 4
in CSF cryptococcal CFU or EFA for each treatment arm. Sec- treatment groups: the mean (6 standard deviation) EFA for

Downloaded from https://academic.oup.com/cid/article/54/1/121/366746 by guest on 25 October 2020

ondary outcome measures were mortality at 2 and 10 weeks and treatment groups 1, 2, 3, and 4 were 20.41 6 0.22 log CFU/mL
drug-related clinical and laboratory adverse events. CSF/day, 20.38 6 0.18 log CFU/mL CSF/day, 20.41 6 0.35 log
CFU/mL CSF/day, and 20.44 6 0.20 log CFU/mL CSF/day,
Statistics respectively, with no statistically significant difference among
We compared baseline characteristics and outcomes in groups groups (P 5 .6, P 5 .9, and P 5.7 for comparison of groups 2, 3,
by the v2 or Fisher’s exact test for categorical variables and the and 4 with group 1, respectively) (Figure 2).
Kruskal-Wallis test for continuous variables. Linear regression In this dataset, rate of clearance was associated with weight
was used to compare mean rates of decrease or EFA by treatment (increase in rate of clearance, or slope, for each increase in
group, adjusting where indicated for other variables, giving weight quartile: 0.07 log CFU/day; 95% CI, 0.02–0.12; P 5 .006).
summary differences with 95% confidence intervals (CIs) and After adjusting for weight and factors previously found to be
significance levels. Laboratory adverse events and median per- associated with rate of clearance (baseline fungal burden, AmB
centage changes in laboratory values in treatment groups and dose, and CD4 cell count), there was still no statistically sig-
among patients receiving the 2 doses of AmB were compared nificant difference among treatment groups (adjusting for
using the v2, Fisher’s exact, Kruskal-Wallis, or Mann-Whitney weight, comparing groups 2, 3, and 4 with group 1, differences
test, as appropriate. Analyses were performed using Stata, in slope were: 0.03 log CFU/day [95% CI, 20.12 to 0.17; P 5 .7],
version 11 (Stata Corp). 20.01 log CFU/day [95% CI 20.16 to 0.14; P 5 .9], and 20.01
log CFU/day [95% CI, 20.19 to 0.17; P 5 .9], respectively). For
the 2 groups of patients receiving fluconazole (treatment groups
RESULTS 2 and 3), mean (6 standard deviation) rate of clearance was
20.44 6 0.32 log CFU/mL CSF/day for patients not receiving
Eighty patients with a first episode of culture-proven CM were concurrent rifampicin treatment (n 5 29) and 20.32 6 0.17 log
enrolled (Figure 1). One patient in treatment group 3 was ex- CFU/mL CSF/day for patients receiving concurrent rifampicin
cluded from the study, because he was found to meet an exclusion (n 5 16; P 5 .2). Nine (43%) of 21, 7 (32%) of 22, 5 (22%) of
criterion on study day 2 (previous episode of CM). One patient 23, and 3 (23%) of 13 had negative culture results at day 14 in
was lost to follow-up at 2–10 weeks in each of treatment arms 2, 3, treatment groups 1–4, respectively.
and 4. One additional patient in study group 1 discontinued all
medical intervention on study day 4 for personal reasons. Mortality
The overall mortality rate was 12% (9 of 78 died) at 2 weeks and
Baseline Characteristics 29% (22 of 75 patients died) at 10 weeks, with no statistically
Baseline clinical and laboratory characteristics and clinical out- significant difference among treatment groups (Table 1). All
comes are shown in Table 1. 57 patients (72%) were known to be surviving patients received ART at a median of 34 days after
HIV seropositive before study enrolment. The median CD4 cell starting antifungal therapy.
count was 24 3 106 cells/L. Twelve patients (15%) had abnormal
mental status (Glasgow coma scale ,15) at baseline. Thirty- Safety
seven patients (47%) received 1–3 doses of AmB before study There were no grade 3 or 4 adverse events thought to be related
enrolment, and baseline fungal burden was significantly asso- to the second drugs given with AmB (Table 2). In particular,
ciated with exposure to AmB before study enrolment (P , .001). there were no grade 3 or 4 decreases in neutrophil count that
Of note, fewer patients in group 3 (30%) and more patients could have been related to 5-FC and no grade 3 or 4 increases

HIV/AIDS d CID 2012:54 (1 January) d 123

 Downloaded from https://academic.oup.com/cid/article/54/1/121/366746 by guest on 25 October 2020
Figure 1. Patient flow chart. Abbreviations: 5-FC, flucytosine; AmB, amphotericin B; bd, twice daily; od, once daily; qds, four times daily.

in alanine aminotransferase level that could have been related Six patients (3 patients in group 2, 2 patients in group 4, and
to fluconazole or voriconazole. The median (interquartile 1 patient in group 1) stopped study drugs early because of
range) percentage change in neutrophil count did not differ grade 4 anemia at study days 9, 10, 11 (2 patients), and 12 (2
comparing group 1 (221%; 238% to 25%) with the other patients). Two patients (1 patient each in group 1 and 4)
3 groups (21%; 243% to 91%; P 5 .4). None of the 13 patients experienced grade 3 and grade 2 renal impairment, re-
receiving voriconazole experienced adverse effects, such as visual spectively, and study drugs were stopped on study day 9 as
disturbance or rash. a result.
AmB-related anemia and renal impairment were common. On the basis of baseline hemoglobin level, 65 patients (82%)
There were no statistically significant differences among received AmB (1 mg/kg) and the remaining 14 patients with
treatment arms in terms of percentage decrease in hemoglobin baseline hemoglobin level #8 g/dL (18%) received AmB
level or percentage increase in creatinine level during the first (0.7 mg/kg). The median decrease in hemoglobin level over the
2 weeks of study enrolment (data not shown). Eight (10%) of first 2 weeks of therapy was 1.9 g/dL in the AmB 1 mg/kg
79 patients discontinued use of study drugs before the end of group and 0.6 g/dL in the AmB 0.7 mg/k group. The median
week 2 because of AmB-related anemia or renal impairment. percentage decrease in hemoglobin level in the amphotericin

124 d CID 2012:54 (1 January) d HIV/AIDS

Table 1. Baseline Clinical and Laboratory Characteristics and Clinical Outcomes for Treatment Groups 1–4

Group 1 Group 2 Group 3 Group 4

Characteristics All patients AmB 1 5-FC AmB 1 FLU 800 AmB 1 FLU 1200 AmB 1 VORI P values
No (%) of men 39 (49) 9 (43) 11 (50) 13 (57) 6 (46) .83
Age (years) 34 (29–39) 34 (32–39) 32 (29–38) 33 (29–38) 35 (33–40) .59
Weight (kg) 56 (49–62) 54 (46–63) 56 (48–62) 54 (49–61) 60 (55–64) .45
No (%) of patients with known HIV 57 (72) 15 (71) 17 (77) 19 (83) 6 (46) .13
No (%) of patients with abnormal 12 (15) 3 (14) 3 (13) 4 (17) 2 (15) 1.0
mental status
No (%) of patients with concurrent 24 (30.4) 8 (38.1) 5 (22.7) 11 (47.8) 0 (0)
tuberculosis at enrollment
CD4 cell count, X 106 cells/L 24 (9–43) 13 (6.5–34) 29 (21–70) 24 (9–69) 27 (9–36) .26
No (%) of patients dosed at AmB 0.7 14(17.7%) 4 (19) 3 (13.6) 5 (20.8) 2 (15.3) .93
CSF data

Downloaded from https://academic.oup.com/cid/article/54/1/121/366746 by guest on 25 October 2020

Opening pressure, cmH20 18 (15–35) 29 (17.5–39) 17 (15–25) 18 (15–35) 25.5(15.5–42) .24
WBC count, cells/mm3 19.5 (0–55.3) 6 (0–50) 10 (0–20) 20 (0–68) 30 (20–60) .21
Baseline fungal burden, logCFU/mL 4.95 (3.79–5.64) 4.83 (4.35–5.89) 4.95 (3.75–5.46) 5.54 (4.86–5.79) 4.49 (3.48–4.82) .03
of CSF
Death, no (%) of patients
By week 2 9/78 (11.5) 1/20 (5) 3/22 (13.6) 4/23 (17.4) 1/13 (7.7) .60
By week 10 22/75 (29) 6/20 (30) 7/21 (33.3) 6/22 (27.3) 3/12 (25) .96
Abbreviations: 5-FC, flucytosine; AmB, amphotericin; CFU, colony-forming units; CSF, cerebrospinal fluid; FLU, fluconazole; HIV, human immunodeficiency virus;
VORI, voriconazole; WBC, white blood cell.
* Values are median (IQR) unless otherwise stated.

(1 mg/kg) group (15%) was greater than that in the amphotericin CFU/mL/day). Pappas et al found that AmB (0.7 mg/kg) plus
B (0.7 mg/kg) group (6%; P 5 .04). fluconazole (800 mg daily) had the highest response rates as
assessed by a composite end point of survival, neurologic sta-
DISCUSSION bility, and negative CSF culture results after 14 days, compared
with AmB alone and AmB plus fluconazole (400 mg/day) [21].
More effective, easily administered, and readily available antifun- AmB plus fluconazole (800 mg/day) was recently adopted as the
gal regimens for the treatment of CM are urgently needed. We standard induction antifungal regimen in a multicenter study
found no statistically significant difference in EFA between the being conducted at African sites on the timing of ART after
combination of AmB and high-dose fluconazole, a safe and readily therapy for CM (Cryptococcal Optimal Antiretroviral Therapy
available second-line agent, and the current standard of AmB and Timing [Coat Trial]; ISCRTN: NCT01075152).
5-FC therapy. The results support the use of AmB plus fluconazole Overall, the second-line drug was well tolerated in this clinical
at a dosage of at least 800 mg daily in Africa and elsewhere where trial, with drug interruptions due to only AmB therapy. In
5-FC is not currently available, pending the results of a phase 3 particular, high-dose fluconazole was well tolerated with no
clinical trial of combination therapy for CM conducted in Viet- adverse liver function abnormalities, in keeping with previous
nam in which AmB alone is being compared with AmB plus 5-FC clinical trial data [22–26]. Of interest, there were no cases of
and AmB plus fluconazole 800 mg; (ISRCTN95123928). grade 3 or 4 neutropenia in the 20 patients assigned to the AmB
Our findings are in keeping with animal model data that have plus 5-FC arm. In prior studies by our group, 5-FC for 2 weeks
demonstrated an additive effect of AmB and fluconazole [19, at 100 mg/kg/day has been associated with grade 4 neutropenia
20]. No prior studies have compared fluconazole at a dosage of in 6 (3.6%) of 163 patients (0 of 32 patients in Thailand [16], 3
$800 mg/day with 5-FC as second-line agents with AmB. of 41 of patients in Malawi [22], and 3 of 90 patients in Cape
Brouwer et al showed a trend toward more rapid CSF sterili- Town [27]). In our view, 5-FC can be used safely in centers in
zation with AmB plus fluconazole (400 mg daily; EFA, 20.39 developing countries, with dose adjustment for any AmB-related
log CFU/mL/day), compared with AmB alone (EFA, 20.31 log renal impairment and complete blood count monitoring and no
CFU/mL/day) [16]. Of note, in contrast to our findings using flucytosine level monitoring [28].
fluconazole at 800 mg daily or 600 mg twice daily, AmB plus A role for voriconazole in the treatment of cryptococcosis
fluconazole (400 mg daily) in that study was significantly less has been supported by in vitro and animal model data [29–31].
rapidly fungicidal than was AmB plus 5FC (EFA, 20.54 log Clinical data regarding the use of voriconazole for the

HIV/AIDS d CID 2012:54 (1 January) d 125

 Downloaded from https://academic.oup.com/cid/article/54/1/121/366746 by guest on 25 October 2020
Figure 2. Decrease in number of cerebrospinal fluid Cryptococcus neoformans colony-forming units over time by treatment group. All patients received
AmB-based therapy. Abbreviations: 5FC, flucytosine; CFU, colony-forming units; Flu, fluconazole; Vori, voriconazole.

treatment of cryptococcosis is otherwise limited to case reports the fact that only patients not receiving rifampicin could be
and small series [14, 32–35]. To our knowledge, this study randomized to receive voriconazole. However, the EFA for
represents the first randomized, comparative clinical trial voriconazole (20.44 log CFU/mL CSF/day) was the same as
evaluating voriconazole for the treatment of HIV-associated the mean EFA for the fluconazole arms in patients not re-
CM. Thus, although the numbers of patients studied was ceiving rifampicin (20.44 log CFU/mL CSF/day). Vor-
small, it appears that the EFA of voriconazole in combination iconazole appeared to be well tolerated at 300 mg twice daily
with AmB did not differ significantly from the EFA of high- for 2 weeks. Voriconazole is useful and effective for rare pa-
dose fluconazole or 5-FC with AmB. The comparison of the tients whose isolates develop secondary fluconazole resistance
voriconazole arm with the 3 other arms is potentially biased by [29, 31, 32, 35]. However, in addition to cost issues, where the

Table 2. Serious Adverse Events Per Treatment Group

AmB 1 5-FC AmB 1 Fluconazole 800 mg AmB 1 Fluconazole 1200 mg AmB 1 Voriconazole
Type/Grade of event Grade III Grade IV Grade III Grade IV Grade III Grade IV Grade III Grade IV
Thrombocytopenia 0 0 0 1 0 0 0 0
Neutropenia 0 0 0 0 0 0 0 0
Anemia 3 3 2 6 0 2 2 3
Transaminase elevation 0 0 0 0 0 0 0 0
Hyponatremia 0 1 1 0 2 1 1 0
Hypokalaemia 0 0 1 0 1 1 0 0
Renal failure 2 0 3 0 3 0 1 0
Total 5 4 7 7 6 4 4 3

Abbreviations: 5-FC, flucytosine; AmB, amphotericin B.

126 d CID 2012:54 (1 January) d HIV/AIDS

HIV and tuberculosis coinfection prevalence is high, use of 3. Hakim JG, Gangaidzo IT, Heyderman RS, et al. Impact of HIV in-
voriconazole for the treatment of CM is limited by interactions fection onmeningitis in Harare: a prospective study of 406 pre-
dominantly adult patients. AIDS 2000; 14:1401–7.
with rifamycins [36] and is complicated by interactions with 4. Gordon SB, Walsh AL, Chaponda M, et al. Bacterial meningitis in
antiretroviral agents [37, 38]. Malawian adults: pneumococcal disease is common, severe, and sea-
As in a prior randomized study [39], the percentage decrease sonal. Clin Infect Dis 2000; 21:53–7.
5. Silber E, Sonnenberg P, Ho KC, et al. Meningitis in a community with
in hemoglobin level was greater in patients assigned to 1 mg/kg/ a high prevalence of tuberculosis and HIV infection. J Neurol Sci 1999;
day of AmB, compared with patients receiving 0.7 mg/kg/day. 162:20–6.
Although reversible, the clinical significance of AmB-associated 6. Harrison TS. The burden of HIV-associated cryptocccal disease. AIDS
2009; 23:531–2.
anemia, especially in patients with low baseline hemoglobin level
7. Okongo M, Morgan D, Mayanja B, et al. Causes of death in a rural,
and in settings where transfusion is restricted, is unclear. Use of population-based human immunodeficiency virus type 1 (HIV-1)
the 0.7 dose of AmB, with fluconazole, in patients with very low natural history cohort in Uganda. Int J Epidemiol 1998; 27:698–702.
baseline hemoglobin level may prevent some grade 4 anaemia, 8. French N, Gray K, Watera C, et al. Cryptococcal meningitis in a cohort
of HIV-1 infected Ugandan adults. AIDS 2002; 16:1031–8.
and it is possible this benefit may outweigh the reduction in rate 9. Perfect JR, Dismukes WE, Dromer F, et al. Clinical practice guidelines
of clearance of infection with the lower dose [39]. However, if for the management of cryptococcal disease: 2010 update by the In-

Downloaded from https://academic.oup.com/cid/article/54/1/121/366746 by guest on 25 October 2020

AmB is given for ,2 weeks, we would still favor the higher AmB fectious Diseases Society of America. Clin Infect Dis 2010; 50:
291–322.
dose, with complete blood count monitoring. 10. Wertheimer AI, Santella TM, Lauver HJ. Successful public/private do-
The overall mortality of 29% at 10 weeks supports the use of nation programs: a review of the Diflucan Partnership Program in South
combination AmB-based therapy in Africa, as opposed to Africa. J Int Assoc Physicians AIDS Care (Chic Ill) 2004; 3:74–9. 84–5.
11. Pfaller MA, Messer SA, Boyken L, et al. Global trends in the antifungal
much less rapidly active fluconazole monotherapy (EFA,
susceptibility of Cryptococcus neoformans (1990 to 2004). J Clin Mi-
20.02 to 20.18 log CFU/mL/day [22, 23, 40]), with an as- crobiol 2005; 43:2163–7.
sociated 10-week mortality of $50% ([22, 23, 41]) or AmB 12. Pfaller MA, Messer SA, Boyken L, et al. In vitro activities of vor-
monotherapy (10-week mortality, 33%–43% [40, 42]), iconazole, posaconazole, and fluconazole against 4,169 clinical isolates
of Candida spp. and Cryptococcus neoformans collected during 2001 and
wherever feasible. In most centers in Africa and Asia, com- 2002 in the ARTEMIS global antifungal surveillance program. Diagn
bination therapy will currently be with AmB and fluconazole. Microbiol Infect Dis 2004; 48:201–5.
If a standard 2-week induction with AmB combination 13. Van Duin D, Cleare W, Zaragoza O, et al. Effects of voriconazole on
Cryptococcus neoformans. Antimicrob Agents Chemother 2004; 48:2014–20.
treatment is not possible, shorter durations of AmB-based 14. Perfect JR, Marr KA, Walsh TJ, et al. Voriconazole treatment for less
induction therapy should be used in preference to fluconazole common, emerging, or refactory fungal infections. Clin Infect Dis
monotherapy. A trial in resource-limited settings that com- 2003; 36:1122–31.
15. Bicanic T, Muzoora C, Brouwer AE, et al. Independent association
pares 1 week with 2 weeks induction with AmB plus fluco-
between rate of clearance of infection and clinical outcome of HIV-
nazole with optimized oral combination therapy with associated cryptococcal meningitis: analysis of a combined cohort of
fluconazole plus 5-FC is planned. 262 patients. Clin Infect Dis 2009; 49:702–9.
16. Brouwer AE, Rajanuwong A, Chierakul W, et al. Combination anti-
fungal therapies for HIV-associated cryptococcal meningitis: a rando-
Notes
mised trial. Lancet 2004; 363:1764–7.
Author contributions. A. L. conducted the trial, collected the data and 17. Kwon-Chung KJ, Bennett JE. Medical Mycology. Philadelphia: Lea &
wrote the manuscript. A. L., J. N. J. and T. S. H. analysed the data. T. B. and Febiger, 1992.
A. L. set up the study. L. B., K. R., G. M. were all involved in study patient 18. South African HIV Clinicians Society. Guidelines for the prevention, di-
care. T. S. H. designed and supervised the study and helped write the agnosis and management of cryptococcal meningitis and disseminated
manuscript. All authors read and approved the final manuscript. cryptococcosis in HIV- infected patients. South Afr J HIV Med 2007: 25–35.
Acknowledgments. We thank all the patients and staff at GF Jooste, 19. Larsen RA, Bauer M, Thomas AM, Graybill JR. Amphotericin B and
Cape Town, and Edendale Hospital, Pietermaritzburg, and, in particular, fluconazole, a potent therapy for cryptococcal meningitis. Antimicrob
Sister Gertrude Ntombomzi Williams and Sister Lungile Maziya. Agents Chemother 2004; 48:985–91.
Financial support. This work was supported by the Medicines Re- 20. Diamond DM, Bauer M, Daniel BE, et al. Amphotericin B colloidal
search Council (UK) and the Wellcome Trust (UK). dispersion combined with flucytosine with or without fluconazole for
Potential conflicts of interest. All authors: No reported conflicts. the treatment of murine cryptococcal meningitis. Antimicrobial Agents
All authors have submitted the ICMJE Form for Disclosure of Potential and Chemotherapy 1998; 528–33.
Conflicts of Interest. Conflicts that the editors consider relevant to the 21. Pappas PG, Chetchotisakd P, Larsen RA, et al. A phase II randomized
content of the manuscript have been disclosed. trial of Amphotericin B alone or combined fluconazole in the treatment
of HIV- associatedcryptococcal meningitis. Clin Infect Dis 2009;
References 48:1775–83.
22. Nussbaum JC, Jackson A, Namarika D, et al. Combination flucytosine
1. Park BJ, Wannemuehler KA, Marston BJ, et al. Estimation of the and high dose fluconazole compared with fluconazole monotherapy
current global burden of cryptococcal meningitis among persons living for the treatment of cryptococcal meningitis: a randomized trial in
with HIV/AIDS. AIDS 2009; 23:525–30. Malawi. Clin Infect Dis 2010; 50:338–44.
2. Jarvis JN, Meintjes G, Williams A, et al. Adult meningitis in a setting of 23. Longley N, Muzoora C, Taseera K, et al. Dose response effect of high-
high HIV and TB prevalence: findings from 4961 suspected cases. BMC dose fluconazole for HIV-associated cryptococcal meningitis in
Infect Dis 2010; 10:67. Southwestern Uganda. Clin Infect Dis 2008; 47:1556–61.

HIV/AIDS d CID 2012:54 (1 January) d 127

24. Milefchik E, Leal MA, Haubrich R, et al. Fluconazole alone or com- 33. Nirenberg NE, Thompson GR, Lewis JS, Hogan BK, Patterson TF.
bined with flucytosine for the treatment of AIDS-associated crypto- Voriconazole use and pharmacokinetics in combination with in-
coccal meningitis. Med Mycol 2008; 46:393–5. terferon-c for refractory cryptococcal meningitis in a patient receiving
25. Haubrich RH, Haghighat D, Bozzette SA, et al. High-dose fluconazole low-dose ritonavir. Med Mycol 2010; 48:532–6.
for treatment of cryptococcal disease in patients with human immu- 34. Carbonara S, Regazzi M, Ciracı̀ E, et al. Long-Term efficacy and safety of
nodeficiency virus infection. J Infect Dis 1994; 170:238–42. TDM-Assisted combination of voriconazole plus efavirenz in an AIDS
26. Menichetti F, Fiorio M, Tosti A, et al. High-dose fluconazole therapy for Patient with cryptococcosis and liver cirrhosis. Ann Pharmacother 2009;
cryptococcal meningitis in patients with AIDS. Clin Infect Dis 1996; 43:978–84.
22:838–40. 35. Chang H, Lee N, Ko W. Voriconazole inhibition of tacrolimus me-
27. Jarvis NJ, Meintjes G, Rebe K, et al. Adjunctive IFN-c immunotherapy tabolism in a kidney transplant recipient with fluconazole-resistant
for the treatment of HIV-associated cryptococcal meningitis: a ran- cryptococcal meningitis. Int J Infect Dis 2010; 14:348–50.
domized controlled trial [abstract 40]. In: Program and abstracts of the 36. Bruggemann RJ, Alffenaar JW, Blijlevens NM, et al. Clinical relevance
18th Conference on Retroviruses and Opportunistic Infections. of the pharmacokinetic interactions of azole antifungal drugs with
Boston, MA. Alexandria, VA: CROI, 2011. other coadministered agents. Clin Infect Dis 2009; 48:1441–58.
28. Brouwer AE, Van Khan HJ, Johnson E, et al. Oral versus intravenous 37. Yakiwchuk EM. Complexity of interactions between voriconazole and
flucytosine in patients with human immunodeficiency virus-associated antiretroviral agents. Ann Pharmacother 2008; 42:698–703.
cryptococcal meningitis. Antimicrob Agents Chemother 2007; 51: 38. Liu P, Foster G, Labadie RR, Gutierrez MJ, Sharma A. Pharmacokinetic
1038–42. interaction between voriconazole and efavirenz at steady state in

Downloaded from https://academic.oup.com/cid/article/54/1/121/366746 by guest on 25 October 2020

29. Nguyen MH, Yu CY. In vitro comparative efficacy of voriconazole and healthy male subjects. J Clin Pharmacol 2008; 48:73–84.
itraconazole against fluconazole-sensitive and -resistant Cryptococcus 39. Bicanic T, Wood R, Meintjes G, et al. High-dose amphotericin B with
neoformans isolates. Antimicrob Agents Chemother 1998; 42:471–2. flucytosine for the treatment of cryptococcal meningitis in HIV-in-
30. Serena C, Pastor FJ, Mariné M, Rodriguez MM, Guarro J. Efficacy of fected patients: a randomized trial. Clin Infect Dis 2008; 47:123.
voriconazole in a murine model of cryptococcal central nervous system 40. Bicanic T, Meintjes G, Wood R, et al. Fungal burden, early fungicidal
infection. J Antimicrob Chemother 2007; 60:162–5. activity, and outcome in cryptococcal meningitis in antiretroviral-naive
31. Silva EG, Paula CR, Dias AL, et al. Combination efficacy of voriconazole or antiretroviral-experienced patients treated with amphotericin B or
and amphotericin B in the experimental disease in immunodeficient mice fluconazole. Clin Infect Dis 2007; 45:76–80.
caused by fluconazole-resistant Cryptococcus neoformans. Mycopathologia 41. Schaars CF, Meintjes GA, Morroni C, et al. Outcome of AIDS-associ-
2011; 171:261–6. ated cryptococcal meningitis initially treated with 200mg/day or
32. Sabbatini S, Manfredi R, Pavoni M, Consales A, Chiodo F. Vor- 400mg/day of fluconazole. BMC Infect Dis 2006; 6:118–23.
iconazole proves effective in long-term treatment of a cerebral cryp- 42. Kambugu A, Meya DB, Rhein J, et al. Outcomes of cryptococcal
tococcoma in a chronic nephropathic HIV-negative patient, after meningitis in Uganda before and after the availability of highly active
fluconazole failure. Mycopathologia 2004; 158:165–71. antiretroviral therapy. Clin Infect Dis 2008; 46:1694–7011694–701.

128 d CID 2012:54 (1 January) d HIV/AIDS