Current issues of ACP Journal Club are published in Annals of Internal Medicine


Fluconazole and amphotericin B were equally effective for candidemia

ACP J Club. 1995 Mar-April;122:42. doi:10.7326/ACPJC-1995-122-2-042

Source Citation

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 Nov 17;331:1325-30.



To compare fluconazole and amphotericin B for treating candidemia in patients without neutropenia or major immunodeficiency.


Randomized controlled trial.


24 university hospitals.


206 patients (mean age 59 y, 51% men) ≥ 13 years old who had ≥ 1 positive blood culture for Candida species and either fever, low or decreased systolic blood pressure, or sign of inflammation at a site infected with Candida. Exclusion criteria were neutrophil counts < 500/mm3, hematologic cancer, congenital immunodeficiency, seropositivity for human immunodeficiency virus, organ transplantation, burns, contraindications to study drugs, liver disease, increased prothrombin times, and conditions likely to cause neutropenia. In 72% of patients, vascular catheters were the probable source of candidemia.


103 patients were allocated to fluconazole, 400 mg/d (intravenously for 7 days, then orally). 103 patients were allocated to amphotericin B, 0.5 to 0.6 mg/kg daily for 7 days and then at least 3 times/wk. Both treatments were continued for at least 14 days after the last positive culture. Changes in therapy were based on predetermined clinical criteria. Patients were followed for 12 weeks after therapy was completed.

Main outcome measures

Successful treatment, responses to initial therapy, relapses, and adverse effects.

Main results

The study was designed to detect a 20% difference in favorable response with a power of 80% and a significance level of 5% with 90 patients in each group. The groups did not differ for number of days of treatment (17 d for amphotericin B vs 18 d for fluconazole), successful outcomes (Table), failure to clear bloodstream infections (Table), overall mortality (Table), time until death, and number of deaths in the first 7 days. Most failures were associated with Candida albicans. Patients taking amphotericin B had higher blood urea nitrogen or serum creatinine levels (37% vs 2%, P < 0.001) and a higher incidence of hypokalemia (10% vs 2%, P = 0.006)


Fluconazole and amphotericin B were equally effective for candidemia in patients without neutropenia or major immunodeficiency. Amphotericin had more adverse effects.

Sources of funding: Roerig-Pfizer and the National Institute of Allergy and Infectious Diseases.

For article reprint: Dr. J.H. Rex, University of Texas Medical School, 6431 Fannin, 1728 JFB, Houston, TX 77030, USA. FAX 713-500-5495.

Table. Fluconazole vs amphotericin B for candidemia in patients without neutropenia or major immunodeficency*

Outcomes at 12 wk Fluconazole Amphotericin RBR (95% CI) NNH
Successful treatment† 70% 79% 11% (-4 to 25) Not significant
Failure to clear bloodstream infection 15% 12% 25% (-37 to 151) Not significant
Mortality 33% 40% 17% (-19 to 42) Not significant

*RBR = relative benefit reduction. Other abbreviations defined in Glossary; RBR, RRI, RRR, NNH, NNT, and CI calculated from data in article.
†Resolution of all signs and symptoms of infection, negative blood cultures at end of therapy and follow-up visits, previously positive cultures from normally sterile sites now negative, and resolution of associated signs of inflammation.


This major study advances our knowledge of how to use fluconazole in patients with candidemia. This is the first randomized study to compare fluconazole with amphotericin B in patients with candidemia and various underlying diseases. Although fluconazole has been used in the treatment of disseminated candidiasis in certain immunocompromised populations (1), none of the patients in the present study was severely immunocompromised.

For these results to be applied properly, several points should be emphasized. First, the dose of fluconazole, 400 mg/d, is higher than that often used to treat less severe candidal infections, and treatment was given intravenously for the first 7 days. Second, most patients in this study were infected with Candidaalbicans, the most common human Candida pathogen. Because few patients had non- albicans candidal infections, these were not analyzed separately. Some non- albicans candidal species may be less susceptible to fluconazole than Candidaalbicans (2).

In applying these results to our patients, it is also important to realize that most patients in this study (72%) had vascular catheters as the probable source of the candidemia. We must use caution in extending these results to candidemia from other sources, such as the peritoneum or mediastinum, which were represented by few patients in this study. The vascular catheters that were in place at the time of candidemia were changed when possible. The study did not control for catheter changes, but the authors note that a similar proportion of patients in each treatment group had the catheter changed.

This study supports the use of fluconazole as an alternative to amphotericin B for candidemia in patients who are not severely immunocompromised, particularly if the infection is caused by Candidaalbicans and originates from a vascular catheter.

David L. Calhoun, MD
Billings ClinicBillings, Montana, USA


1. Anaissie E, Bodey GP, Kantarjian H, et al. Fluconazole therapy for chronic disseminated candidiasis in patients with leukemia and prior amphotericin B therapy. Am J Med. 1991;91:142-50.

2. Wingard JR, Merz WG, Rinaldi MG, et al. Increase in Candidakrusei infection among patients with bone marrow transplantation and neutropenia treated prophylactically with fluconazole. N Engl J Med. 1991;325:1274-7.