Correspondence
The Treatment of Candidemia
N Engl J Med 1995; 332:1100-1102April 20, 1995
- Article
To the Editor:
The prospective study by Rex and colleagues (Nov. 17 issue)1 comparing amphotericin B with fluconazole for the treatment of candidemia concludes that the two treatments are equally effective. This opinion is echoed and endorsed in the accompanying editorial.2 Unfortunately, this randomized, controlled study may in some respects provide the right answer to the wrong question. It may be misconstrued as a mandate for the treatment of all patients with candidemia.
Overall, patients with candidemia treated with either amphotericin B or fluconazole do not appear to have fared much better than untreated historical controls.3 I contend that no therapy is still a viable option for appropriately selected patients with transient candidemia, which is usually catheter-related.3-5 There is little evidence to suggest that patients whose candidemia clears quickly after the removal of the catheter and who have no evidence of metastatic infection require any drug therapy. Even if a small number of these patients subsequently relapse with focal infections,5 they can still probably be treated successfully and at an overall cost lower than the cost of universal treatment.
This carefully designed trial1 perhaps proves only that fluconazole is as ineffective as amphotericin B in lowering the early mortality rate in patients with candidemia but without neutropenia. The trial is not pertinent to the issue of the complications of self-limited, catheter-related fungemia.
5 ReferencesMark J. DiNubile, M.D.
Cooper Hospital–Robert Wood Johnson Medical School, Camden, NJ 081031
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
Full Text | Web of Science | Medline2
Meunier F . Management of candidemia. N Engl J Med 1994;331:1371-1372
Full Text | Web of Science | Medline3
Fraser VJ ,Jones M ,Dunkel J ,Storfer S ,Medoff G ,Dunagan WC . Candidemia in a tertiary care hospital: epidemiology, risk factors, and predictors of mortality. Clin Infect Dis 1992;15:414-421
CrossRef | Web of Science | Medline4
Elis CA ,Spivack ML . The significance of candidemia. Ann Intern Med 1967;67:511-522
Web of Science | Medline5
Klein JJ ,Watanakunakorn C . Hospital-acquired fungemia: its natural course and clinical significance. Am J Med 1979;67:51-58
CrossRef | Web of Science | Medline
To the Editor:
There is a wide consensus that catheter-related fungemia clears up with catheter removal in most cases. In this context antifungal therapy is a prophylactic measure mostly needed to prevent relapse and the development of secondary, deep-seated infection.1 However, in patients with candidemia due to an invasive infection, antifungal therapy has a crucial role.
In the study by Rex et al.,2 a central venous catheter was considered to be the apparent source of candidemia in 77 of 103 patients treated with amphotericin B (75 percent) and 72 of 103 patients treated with fluconazole (70 percent). Moreover, 62 patients in each group underwent catheter removal and replacement in the same day. In only a minority of cases was a deep-seated, invasive candidiasis documented.
These data certainly prove that fluconazole is a well-tolerated alternative to amphotericin B in catheter-related candidemia but do not prove this to be true of deep-seated infection unrelated to catheter use. Thus, the question of whether fluconazole is as efficacious as amphotericin B in patients with deep-seated, invasive infections but without neutropenia remains unanswered.
2 ReferencesCorrado Girmenia, M.D.
Pietro Martino, M.D.
University La Sapienza, 00161 Rome, Italy1
Lecciones JAA ,Lee JW ,Navarro EE , et al. Vascular catheter-associated fungemia in patients with cancer: analysis of 155 episodes. Clin Infect Dis 1992;14:875-883
CrossRef | Web of Science | Medline2
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
Full Text | Web of Science | Medline
To the Editor:
Some patients in the study by Rex et al. were treated with either amphotericin B or fluconazole before randomization. Of the 103 fluconazole recipients, 24 had received an average of 20 mg of amphotericin B before randomization. At times a single dose of amphotericin B can have major beneficial effects in patients with candidemia. It would be interesting to know whether the results would be different if one eliminated the patients in the fluconazole group who had previously received amphotericin B. Those who did not receive any amphotericin B may not have done as well as the 24 who did.
Richard D. Cofsky, M.D.
Brookdale Hospital Medical Center, Brooklyn, NY 11212To the Editor:
The well-designed clinical trial reported by Rex et al. addresses the difficult issue of the treatment of patients with candidemia. The authors assert that “fluconazole and amphotericin B do not differ significantly as therapy for candidemia,” and in the accompanying editorial Dr. Meunier states that “fluconazole can now be considered a safe alternative to amphotericin B.” We believe that caution is in order.
Several species of candida (particularly Candida krusei and C. glabrata) are known to be less susceptible to fluconazole in vitro than C. albicans, and therapy with this agent has been associated with treatment failure.1,2 Because the study by Rex et al. included only 27 patients infected with C. krusei or C. glabrata, the trial may not be sufficiently large to demonstrate important differences in outcome between patients treated with amphotericin B and those treated with fluconazole. Furthermore, there was a higher rate of treatment failure and relapse among the patients with C. krusei or C. glabrata infection who received fluconazole (5 of 12) than among those who received amphotericin B (2 of 15).
Over the past 12 months at our institution, nine patients without human immunodeficiency virus infection or neutropenia have had candidemia, among whom five died. C. albicans accounted for only 45 percent of all fungal isolates. In addition, there were two infections with C. tropicalis and one each with C. krusei, C. kefyr, and C. parapsilosis. given the high proportion of cases of candidemia due to isolates other than C. albicans in our population, we will continue to recommend amphotericin B as the first-line treatment, pending the results of a test to confirm that the isolate is C. albicans. If that test is positive, the data presented by Rex et al. support the use of fluconazole as an alternative to amphotericin B. However, if the test is negative, we suggest that fluconazole therapy should not be instituted until the causative species has been formally identified and the pattern of susceptibility established.
2 ReferencesC.M. Tang, M.D.
P. Howe, B.Sc.
D.W. Crook, M.D.
Oxford Radcliffe Hospital, Oxford OX3 7LJ, United Kingdom1
Van't Wout JW ,Mattie H ,van Furth R . A prospective study of the efficacy of fluconazole (UK-49,858) against deep-seated fungal infections. J Antimicrob Chemother 1988;21:665-672
CrossRef | Web of Science | Medline2
McIlroy MA . Failure of fluconazole to suppress fungemia in a patient with fever, neutropenia, and typhlitis. J Infect Dis 1991;163:420-421
CrossRef | Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: The letters by Dr. DiNubile and Drs. Girmenia and Martino raise a series of related questions. Do all patients with candidemia require therapy? Does therapy prevent late complications? And what is the best therapy for established primary or secondary deep candida infections? Although it is doubtless true that some patients with candidemia may not require therapy, it is unclear how to identify those who can safely be left untreated. Given the data demonstrating substantial attributable mortality associated with candidemia,1 we think it prudent to treat all patients with candidemia, regardless of the apparent source of the infection. Unlike Dr. DiNubile, we do not believe that waiting for complications to develop is appropriate and, as with other serious infections, we would be unwilling to leave a group of patients with candidemia untreated simply to show that late complications would be prevented.2 We agree that our study provides only limited data on the relative efficacy of the two drugs as therapy for established deep candida infections.
Dr. Cofsky asks whether the small dose of amphotericin B received by some patients before therapy with fluconazole may have influenced outcomes. When the patients who received any prior amphotericin B therapy are removed from the primary analysis group, successful outcomes are seen in 66 of 85 amphotericin B–treated patients (78 percent) and 55 of 79 fluconazole-treated patients (70 percent, P = 0.24). These results are identical to those obtained for the group as a whole.
Finally, Tang et al. are concerned about the treatment of infection due to isolates other than C. albicans. We agree that our trial was too small to permit strong conclusions to be drawn with regard to species other than C. albicans. Although it is apparent that C. krusei generally responds poorly to fluconazole, the data for C. parapsilosis, C. tropicalis, and C. glabrata are less clear. We have recently reviewed this area and analyzed the effect of minimal-inhibitory-concentration (MIC) values on the outcomes in our trial.3,4 We found that the cases of therapeutic failure we reported were often associated with the use of less than 400 mg of fluconazole per day and that the higher MICs of species other than C. albicans did not predict failure in our trial. These data, along with data demonstrating the ability of fluconazole (at a dose of 400 mg per day) to prevent infection with species other than C. krusei, 5 suggest that fluconazole at doses of at least 400 mg per day is acceptable initial therapy in institutions where C. krusei infections are rare.
5 ReferencesJohn H. Rex, M.D.
University of Texas, Houston, Medical School, Houston, TX 77030John E. Bennett, M.D.
National Institute of Allergy and Infectious Diseases, Bethesda, MD 20892Alan M. Sugar, M.D.
Boston University Medical Center Hospital, Boston, MA 021181
Wey SB ,Mori M ,Pfaller MA ,Woolson RF ,Wenzel RP . Hospital-acquired candidemia: the attributable mortality and excess length of stay. Arch Intern Med 1988;148:2642-2645
CrossRef | Web of Science | Medline2
Edwards JE Jr . Should all patients with candidemia be treated with antifungal agents? Clin Infect Dis 1992;15:422-423
CrossRef | Web of Science | Medline3
Rex JH ,Pfaller MA ,Barry AL ,Nelson PW ,Webb CD . Antifungal susceptibility testing of isolates from a randomized, multicenter trial of fluconazole versus amphotericin B as treatment of nonneutropenic patients with candidemia. Antimicrob Agents Chemother 1995;39:40-44
Web of Science | Medline4
Rex JH ,Rinaldi MG ,Pfaller MA . Resistance of Candida species to fluconazole. Antimicrob Agents Chemother 1995;39:1-8
Web of Science | Medline5
Goodman JL ,Winston DJ ,Greenfield RA , et al. A controlled trial of fluconazole to prevent fungal infections in patients undergoing bone marrow transplantation. N Engl J Med 1992;326:845-851
Full Text | Web of Science | Medline
Author/Editor Response
The question raised by Tang et al. is indeed very appropriate for current practice in patients with candidemia, as I discussed in my editorial. It is always dangerous to extract parts of sentences from publications. The editorial mentioned that “several issues remain controversial, including . . . species identification before the initiation of antifungal treatment.” The lack of efficacy of fluconazole against species such as C. krusei was also discussed, including the need for the physician to “be familiar with the local epidemiologic features, resistance patterns, and the patient's colonization status.”
So far, it has not been proved that fluconazole is a safe treatment alternative for candidemia caused by C. krusei or C. glabrata, and I stated that “the study by Rex et al. involved too few of these unusual candida species to make it possible to draw conclusions about outcomes.” State-of-the-art treatment for candidemia will result only from a comprehensive approach that includes close collaboration between the microbiologist, the specialist in infectious diseases, the surgeon, and other physicians involved in patient care.
Routine practice should not be based on anecdotal reports. Given the increasing number of cases of candidemia caused by species other than C. albicans, it would be of great benefit to undertake clinical studies to evaluate the optimal management of candidemia caused by those species. Only multicenter studies that include adequate numbers of patients and use sound methods will be able to answer this important question. In addition to increasing the awareness of clinicians about recent advances in the field of antifungal treatments, one should promote effective clinical research, including research on candidemia. Such initiatives have currently been taken in the United States and Europe under the auspices of the Invasive Fungal Infections Cooperative Group of the European Organization for Research and Treatment of Cancer. These activities will make a substantial contribution to the care of patients with candidemia.
Françoise Meunier, M.D., Ph.D.
European Organization for Research and Treatment of Cancer, 1200 Brussels, Belgium
