Evidence Based Review Article

The Journal of Informed Pharmacotherapy 2003;14:200.

Combining Fluconazole and Amphotericin B to Treat Nonneutropenic Candidemia: Not Contraindicated After All?

Reviewer: Peter J. Jewesson
Reviewer抯 e-mail address: 
Reviewer's profession/specialty: Clinical
pharmacotherapeutics, infectious diseases

Original Citation

Rex JH, Pappas PG, Karchmer AW, Sobel J, Edwards JE, Hadley S, Brass C, Vazquez JA, Chapman SW, Horowitz HW, Zervos M, McKinsey D, Lee J, Babinchak T, Bradsher RW, Cleary JD, Cohen DM, Danziger L, Goldman M, Goodman J, Hilton E, Hyslop NE, Kett DH, Lutz J, Rubin RH, Scheld WM, Schuster M, Simmons B, Stein DK, Washburn RG, Mautner L, Chu TC, Panzer H, Rosenstein RB, Booth J; The National Institute of Allergy and Infectious Diseases Mycoses Study Group.  A Randomized and Blinded Multicenter Trial of High-Dose Fluconazole plus Placebo versus Fluconazole plus Amphotericin B as Therapy for Candidemia and Its Consequences in Nonneutropenic Subjects. Clin Infect Dis 2003;36:1221-8.PubMed Citation

Overall Study Question

The objective of this study was to compare high-dose fluconazole monotherapy with fluconazole plus amphotericin B (AmB) deoxycholate as initial treatment for non-Candida krusei candidemia in non-neutropenic adults. 

The study sample included 236 consenting patients (>= 13 years) with one or more blood cultures positive for Candida species up to 4 days before study entry. Patients were enrolled at 27 U.S. institutions over a 46-month period. Eligible patients were required to demonstrate at least one or more of the following characteristics within 48 hours of study entry: a temperature of at least 37.8 C on two occasions (>4 hours apart); a temperature greater than 38.6 C on one occasion; a systolic blood pressure of <90 mm Hg; a decrease in systolic blood pressure of at least 30 mm Hg from the baseline value; or signs of inflammation (swelling, heat, erythema, or purulent drainage) from a site infected with Candida species. 

Multiple exclusion criteria were applied including the elimination of patients who were or were expected to become neutropenic during therapy; had a history of intolerance or allergy to azole antifungal agents or AmB; had laboratory evidence of liver dysfunction; had previously received full daily doses of AmB, fluconazole; had experienced treatment failure with previous systemic antifungal therapy for the fungal infection being studied;; were receiving or were to receive any other systemic antifungal drug during the course of the study; were infected or found to be infected with Candida krusei; were receiving terfenadine; were thought unlikely to survive for >24 h or had previously participated in this study.

Patients were randomized to receive fluconazole plus placebo (FP) or fluconazole plus AmB deoxycholate (FA). For both study arms, fluconazole (800 mg) was administered intravenously daily for the first 5 days and then orally, unless the patient抯 condition precluded reliable oral therapy. The dose was adjusted for weight extremes and renal impairment. Patients were treated with the fluconazole component of the study regimen for 14 days after resolution of clinical findings related to the episode of candidemia, or 14 days after the last positive deep-site culture result, whichever duration was longest. The placebo or AmB component was provided for the first 5�days of therapy. Patients assigned to receive AmB were administered 0.6�7 mg/kg/day with a total dose goal of 3 mg/kg. Regardless of study assignment, all subjects were to be medicated with diphenhydramine, acetaminophen, and/or ibuprofen before each AmB or placebo infusion.

The primary outcome variable was time to failure in a 搈odified intent-to-treat population� in which all randomized subjects who received 1 dose of study medication and for whom culture of a blood sample obtained within 4 days of study entry positive for a Candida species, except C. krusei were analyzed. 

What was the declared relationship between authors and sponsors for this study?

This study was jointly sponsored by the National Institute of Allergy and Infectious Disease and the National Institutes of Health, in collaboration with Pfizer Inc.  While no specific statements of potential conflicts of interest were provided, five of the authors were employees of the manufacturer of fluconazole (Pfizer Inc.).

Are the Results of the Study Valid?

1. Was assignment of patients randomized?

Yes. After enrollment, each patient was assigned to an organ transplantation or non杘rgan transplantation stratum and were subsequently randomized within these strata in a 1:1 ratio to receive FP or FA on the basis of a sequential list of block-randomized assignments that were maintained by each participating hospital pharmacy. 

2. Were all patients who entered the trial properly accounted for and attributed at its conclusion?

Yes.  Of the 236 enrollees, 219 (93%) were included in the modified intent-to-treat population.  Of the 17 excluded patients, 15 did not have positive blood cultures while 2 received expired drug.  

3. Were patients, their clinicians, and study personnel 'blind' to treatment?

The allocation of patients to the placebo or AmB was apparently blinded to all participants except those hospital personnel responsible for the preparation of the study drugs. Patients assigned to receive placebo were provided an identically colored volume of multivitamin solution in dextrose. Concomitant ordering of hydrocortisone (up to 50 mg per day) or heparin (up to 1000 U per bag) to be added to the AmB/placebo infusion was permitted, but these drugs were only actually added to the infusion if the patient was receiving AmB.

4. Were the groups similar at the start of the trial?

Generally yes.  The two groups were similar with respect to the measured baseline characteristics with the exception of a slightly lower group mean APACHE II score in the FA group.  While statistically insignificant, the only HIV infection patients (N=4) were allocated to the FA group.  Only 5 (2%) study patients received organ transplants, thus analyses were not stratified for this factor as originally planned.  Twenty-eight (13%) of patients had received AmB and 111 (51%) had received fluconazole prior to enrollment in the study.  The distribution of infecting Candida species was similar across treatment arms. 

5. Aside from the experimental intervention, were the groups treated equally?

From the information provided, it appears that the groups were treated equally.  

What were the Results?

1. How large was the treatment effect? 

According to the primary Kaplan-Meier time-to-failure analysis, the treatment groups did not differ with regard to the proportion of patient successfully treatment for the first 30 days of the study (p=0.08).    Secondary analysis revealed that 60 (56%) of 107 FP recipients and 77 (69%) of 112 FA recipients had successful treatment outcomes (ARR 13%; P=0.43). Treatment was considered to be successful if all signs and symptoms thought by the local investigator to be due to the Candida infection resolved, if the results of cultures of blood samples obtained at end of therapy and during follow-up were negative, if the results of cultures of samples of normally sterile sites previously positive for Candida became negative, and if associated signs of local inflammation resolved. Failure to clear the bloodstream of pathogen was more common in the FA group (17% vs. 6%; ARR 11%, P=0.02).  Accordingly to stepwise logistic regression, the odds of treatment failure were increased by APACHE II score, presence of baseline positive blood culture and prior administration of total parenteral nutrition.  In contrast, the odds of failure were decreased by infection with C. parapsilosis.  While the authors report a decreased odds of failure with the administration of FA (OR 0.62), the 95% confidence interval crosses 1 (95% CI 0.33-1.15) and P=0.128.  

There was no evidence of compromised treatment outcomes for patients had previously received fluconazole therapy (70% success for exposed patients vs. 67% for non-exposed patients).  FA appeared to be associated with better treatment outcomes for patients with midrange (i.e. 10-20) APACHE II scores. 

Relatively little information was provided regarding patient intolerance to the study drugs.  Renal dysfunction leading to reduction in study drug dosage was noted in 23% of FA patients versus 3% of those receiving FP (p<0.001), although this was considered a primary cause of failure in a similar proportion of patients in each group (5% vs. 3%).  No detail was provided regarding the magnitude, nor persistence of the renal impairment.  One FA patient was withdrawn from the study following an anaphylactic reaction to amphotericin B.  Hepatic enzyme elevations occurred in a similar proportion of FA and FP recipients (8% vs. 9%).  No description of any amphotericin B infusion-related side effects was provided.

Death within 90 days of study drug receipt was common and noted in 42 (39%) of FP recipients versus 45 (40%) of FA-treated patients.

2. How precise was the estimate of the treatment effect?

The 95% CI around the secondary endpoint of treatment success ARR of 12.9% are (95% CI 0.29 - 25.7, my calculation).  The NNT was  8 (95% CI 4 - 334).  The 95% CI around the secondary endpoint of failure to clear the bloodstream ARR of 11% are (95% CI 2.7% - 19.3%, my calculation).  The NNT was 9 (95% CI 9 - 37).

Will the Results Help Me in Caring for My Patients?

1. Can the results be applied to my patient care?

Partially. Nonneutropenic patients with candidemia are relatively common in the acute care setting and treatment course outcomes are often suboptimal. Since the primary analysis demonstrated only a trend towards treatment course outcome improvements with combination therapy involving high dose fluconazole, one could argue that this more toxic and more costly therapy should be reserved for the initial management of select patients (e.g. those in the APACHE II midrange) when the infecting species has yet to be identified. Considering the lack of evidence that high dose fluconazole monotherapy provides superior clinical outcomes to standard dose regimens (i.e. 400mg/d) employed in previous studies (1), one would have to question whether or not a standard dose fluconazole regimen plus amphotericin B would show similar outcomes, and at lower cost. Unfortunately, this was not addressed in the present study.

2. Were all clinically important outcomes considered?

Most clinically important outcomes were considered. However, more detail regarding treatment course-associated toxicity would have been useful, and treatment costs should have been addressed.

3. Are the likely treatment benefits worth the potential harms and costs?

Possibly. The study results suggest that initial combination therapy with fluconazole and amphotericin B is at least as good, and may possibly improve the likelihood of antifungal treatment course success, including those patients with bloodstream infections. Drug antagonism was not apparent in this study, regardless of previous fluconazole treatment history. Not surprisingly, drug toxicity was more common with combination therapy and this needs to be carefully considered when contemplating the use of this azole-polyene combination.


Invasive fungal infections are common, difficult to diagnose and associated with a high incidence of morbidity and mortality. Despite an expanded therapeutic armamentarium, treatment outcomes with single agents are often suboptimal. Combination anti-infective therapy has shown to be beneficial for other types of infections (e.g. gram-negative bacteremia, tuberculosis) and should theoretically be useful for the management of serious systemic fungal infections. In vitro and in vivo studies involving combinations of azoles and amphotericin B have shown conflicting results, thus the use of these combinations has not been accepted into clinical practice.

This study suggests that a combination of fluconazole and amphotericin B provides treatment course outcomes in nonneutropenic patients that are at least as good, and possibly better, than fluconazole monotherapy. The practical implication of this finding is that initial empiric management of moderately severe fungal infections with broad spectrum combination therapy involving fluconazole and amphotericin B can be used without fear of pharmacological antagonism. Once the causative pathogen(s) is identified, monotherapy with fluconazole can be continued unless a resistant species (e.g. C. krusei) is isolated. The study also suggests that previous exposure to fluconazole does not increase the likelihood of treatment failure with combination therapy. Conversion from fluconazole monotherapy to amphotericin B monotherapy was not addressed in this study, and the results cannot be extrapolated to the neutropenic patient population.

The finding that low APACHE II score patients fair no better on combination therapy suggests that fluconazole monotherapy is likely adequate for less seriously ill patients. Conversely, the most seriously ill patients in this study did not demonstrate improved treatment outcomes, thus one could argue that combination therapy cannot be justified for patients with a pre-existing low likelihood of survival. 

While apparently well tolerated, the high dose fluconazole regimen employed in this study did not appear to improve treatment course outcomes over previously reported by the authors. (1) Thus, there appears to be no reason to use high dose fluconazole regimens for the management of infections involving susceptible fungal species in nonneutropenic patients. 

In summary, this study provides some compelling evidence that fluconazole and amphotericin B do not demonstrate in vivo antagonism when used to manage candidemia in nonneutropenic patients. The use of the combination appears to clear bloodstream infections better and has a higher incidence of overall treatment success, although time-to-failure is similar to fluconazole monotherapy, the incidence of toxicity is higher and (although not examined) the cost is greater. 

Reviewer Competing Interests



  1. Rex JH, Bennett JE, Sugar AM, et al. A randomized trial comparing fluconazole with amphotericin B for the treatment of candidemia in patients without neutropenia. N Engl J Med 1994;331:1325-1330.
  2. Odds FC. Fluconazole plus amphotericin B combinations are not contraindicated and may add benefit for the treatment of candidemia. CID 2003;36:1229-30.

Copyright � 2003 by the Journal of Informed Pharmacotherapy. All rights reserved.