Amphotericin B versus amphotericin B plus 5-flucytosine: Poor ...

Originalia
P. E. Verweij, J. P. Donnelly, B. J. Kullberg, J. F. G. M. Meis, B. E. De Pauw
Amphotericin B versus Amphotericin B plus 5-Flucytosine: Poor Results in
the Treatment of Proven Systemic Mycoses in Neutropenic Patients
Introduction
Summary: Twenty-eight neutropenic (< 500 granulocytes/gl) adults with microbiologically
or histologically proven systemic mycosis were randomly assigned to receive either ampho-
tericin B alone (0.5 mg/kg/day; n = 14) or amphotericin B (0.5 mg/kg/day) plus 5-flucytosine
(150 mg/kg/day; n = 14) intravenously. Therapy was given for an average duration of 10 days
in both groups, amounting to a total dose of amphotericin B of 338 mg and 308 mg, respec-
tively. The mean duration of granulocytopenia was 18 days in the amphotericin B group and
20 days in the combination group. Only two patients treated with amphotericin B alone and
three given the combination survived. Adverse events were similiar in both groups with an
elevation of the serum creatinine in six cases during the administration of amphotericin B
alone and in seven cases treated with the combination. No other serious adverse events were
encountered. Treatment with both regimens was disappointing partly because mycosis was
too far advanced by the time therapy was begun and neutrophils were recovered in only half
the patients.
Since the introduction of modern broad-spectrum antibac-
terial compounds, which has substantially reduced the mor-
tality due to bacterial infections, fungal infections have
emerged as a major cause of morbidity and mortality in
neutropenic patients [1-4]. The incidence of these infec-
tions has also increased in association with the use of more
aggressive chemotherapy and intravascular devices [5].
Candida and Aspergillus species are the predominant fun-
gal pathogens in acute leukemia [6], but these infections are
difficult to diagnose [7] due to the absence of the classic
inflammatory response and low rate of antibody formation
in these patients. Unfortunately, adequate culturing is ham-
pered by the low specificity of specimens such as sputum
[8-10] and by the need for invasive diagnostic procedures
that may be dangerous because of concomitant thrombo-
cytopenia. Atypical pulmonary infiltrates, as well as unex-
plained disturbances of liver and kidney function and echo-
graphic lesions in these organs may be suggestive of the
diagnosis [11, 12], but these symptoms are certainly not
specific. Consequently, treatment is often commenced on
an empiric basis, most frequently because of fever not re-
sponding to adequate antibacterial therapy [13-15]. More-
over, therapeutic options are qualitatively and quantita-
tively limited. Only three classes of compounds with recog-
nized systemic activity are available: amphotericin B, 5-flu-
cytosine, and, more recently, the azoles. Intravenous am-
photericin B, at a dose of 0.5 mg/kg/day, is considered the
standard therapy for granulocytopenic patients suspected to
have a pulmonary or systemic mycosis. The drug is poten-
tially toxic to the liver and kidney [16], causes hypokale-
mia, and many patients show an immediate reaction to its
administration, which may require co-medication with
agents such as pethidine and/or corticosteroids to amelio-
rate the symptoms. These reactions appear to be dose de-
pendent. 5-flucytosine, an analogue of cytosine, at a dose of
150 mg/kg/day is effective against Candida and Crypto-
coccus infections, but the fungi may become rapidly resist-
ant if the drug is given alone. Potentially dangerous side
effects of 5-flucytosine, particularly bone marrow sup-
pression and liver function impairment, may occur [17].
The concentration in the serum should not fall below
20-25 rag/1 and not exceed 100-120 rag/1. There are data to
suggest that amphotericin B and 5-flucytosine show
synergistic activity, notably in the treatment of cryptococ-
cal meningitis in AIDS patients [18, 19]. However, definite
evidence on the potential of this therapeutic approach in the
granulocytopenic patient with candidiasis is lacking, al-
though a small series does suggest that combination therapy
may be of benefit, particularly in infections caused by Can-
dida tropicalis [20, 21].
The diagnostic shortcomings do not only hamper the estab-
lishment of an early diagnosis, they also restrict the possi-
bilities of evaluation of therapeutic interventions, since
trials commonly include patients in whom the diagnosis of
Received: 11 November 1993/Revision accepted: 6 February 1994
P. E. Verweij, M. D., J. F. G. M. Meis, M. D., Ph. D., Dept. of Medical
Microbiology, J. P. Donnelly, Ph. D., Prof. B. E. De Pauw, M. D., Ph.D.,
Dept. of Hematology~ B.J. Kullberg, M.D., Ph.D., Dept. of General
Internal Medicine, University Hospital St. Radboud, P.O. Box 9101,
NL-6500 HB Nijmegen, The Netherlands.
Correspondence to: Prof. B. E. De Pauw.
Infection 22 (1994) No. 2 © MMV Medizin Verlag GmbH Mtinchen, Mtinchen 1994 81/21

P, E. Verweij et al.: Amphotericin B with or without 5-Flucytosine in Neutropenic Patients
fungal infection is only presumptive. Therefore a trial was
designed to assess the efficacy and safety of amphoteri-
cin B alone in comparison to amphotericin B in combina-
tion with 5-flucytosine in febrile neutropenic patiens with a
histologically or microbiologicaUy documented systemic
fungal infection. It must be emphasized, however, that in
general it is not our policy to await confirmation of the
diagnosis before starting antifungal therapy; in the majority
of cases systemically active antifungal therapy is started on
an eml~iric basis, but in a subset of patients the diagnosis of
systemic mycosis may become established before suspicion
occurs. It is this category of patients who were the subjects
of the present study.
Patients and Methods
Patients: The protocol was approved by the local medical ethical
committee (number CEOM 1988-001647). Patients over 15 years
of age with a granulocyte count of < 500 cells/gl, that was ex-
pected to last for 10 days or more, due to chemotherapy for acute
leukemia, who were suffering from a proven systemic fungal
infection, were-eligible for this study if they had an estimated
creatinine clearance of > 80 ml/min. All patients were nursed in
protective isolation and had been treated for febrile episodes with
broad spectrum antibacterials, consisting of either ceftazidime or
meropenem as single agents or a Combination of piperacillin and
tobramycin. In the majority of cases the patients had also received
a course of teicoplanin for presumed or proven infections by
resistant gram-positive organisms. Antibacterial treatment was
supplemented by the prophylactic use of oral amphotericin B
(500 mg q 6 h).
Fungal infection was considered to be proven if fungi were cul-
tured from deep tissues or normally sterile body fluids or if they
were identified by histological examination of biopsy specimens.
In case of candidemia at least two blood cultures taken from
different veins had to be positive with the same species. Typical
infiltrations on a chest X-ray, in combination with filamentous
fungi (e.g. Aspergillus) in the sputum or assisted sputum produc-
tion (NaC1 0.9% aerosol) specimens with >_ 50 CFU/ml of an
undiluted sample, were regarded as definite evidence for the ex-
istence of pulmonary fungal infection. The results of the cultures
from sputum and bronchoalveolar lavage specimens were com-
pared with the outcome of cultures from a simultaneously taken
oral gargle in order to exclude contamination. Patients with unex-
plained fever not responding to antibacterial therapy were ex-
cluded, as were patients with a presumed fungal infection, i.e.
those with a clinical site of infection without known bacterial or
viral origin.
Patients were examined daily, full hematological profiles were
determined daily and tests of'liver and kidney function, and serum
levels of 5-flucytosine were conducted at least twice weekly.
Chest X-rays, echography of the liver, fundoscopy, and invasive
diagnostic procedures, such as bronchoalveolar lavage, were per-
formed on a case-by-case basis.
Procedure: After giving informed consent the patients were ran-
domized to receive either amphotericin B alone or amphoteri-
cin B in combination with 5-flucytosine. After a test dose of
1 mg, the dose of amphotericin B was rapidly increased to 0.5-
1 mg/kg/day intravenously as a 4 h infusion, depending on the
serum creatinine; 5-flucytosine was given intravenously as a 30
min infusion in a dose of 150 mg&g/day in four divided doses.
During systemic antifungal therapy, the oral antifungal prophy-
Table 1: Demography and results of therapy.
Number 14 14
Age. (years)
Mean 41.2 42.1
Range 20~6 18-67
Male:female 8:6 9:5
Acute myeloid leukemia 10 8
Acute lymphoblastic
leukemia 4 6
Minimum duration ot~
granulocytopenia before
starting therapy (days) 10 12
Survival/number infected
Pulmonary aspergillosis 1/9 2/9
Hepatic candidosis 0/2 0/2
Candidemia* 1/3 1/3
Overall survival 2/14 3/14
Granulocytes < 100 cells/gl
at the start of treatment 11 12
Mean total dose of
amphotericin B (mg) 338 308
* the survivor in each treatment group had catheter-related sepsis due to
Candida parapsilosis and had had the Hickman catheter removed.
laxis was discontinued, while antibacterial coverage was provided
by administering either ceftazidime intravenously or a combina-
tion of colistin and co-trimoxazole orally.
A patient was considered to have responded to therapy when
there was a marked decrease in the size of the initial radiological
abnormalities or when all signs and symptoms of infection had
disappeared. If the fungal infection persisted or progressed, or if
the patient died as a result of this infection, therapy was deemed a
failure. Renal toxicity was defined as an increase.in serum crea-
tinine of more than 30%, and hepatic toxicity as a threefold in-
crease of baseline liver function tests.
Results
Between May 1988 and December 1991 28 patients were
enrolled in this study. The characteristics of the patients are
given in Table 1, showing that the study groups were well
balanced with respect to age, sex, underlying disease, de-
gree of granulocytopenia, and causative fungal organisms.
All candidemias were encountered in patients with central
venous lines in situ. In four patients in the amphotericin B
group and six patients in the combination group the diag-
nosis of systemic mycosis was made after fever had persist-
ed for more than 8 days in spite of broad spectrum antibac-
terial therapy, whereas all other patients had initially re-
sponded to these antibiotids but had then developed a sec-
ond febrile episode.
The results, which are also shown in Table 1, were very
disappointing in both groups with a mortality rate of 86%
for amphotericin B alone and 79% for the combination, the
22/82 Infection 22 (1994) No. 2 © MMV Medizin Verlag GmbH Mtinchen, Mtinchen 1994

difference being one patient surviving pneumonia caused
by Aspergillusfumigatus in the latter group. Only the two
patients with a Candida parapsilosis infection recovered
even though the central venous line was removed in all
cases with candidemia. Two patients with Candida tropi-
calis sepsis died within 3 days in spite of treatment with
both amphotericin B and 5-flucytosine. Chronic dissemi-
nated candidosis proved fatal in all four cases, as did 15 out
of 18 (83%) episodes of pulmonary aspergillosis with pa-
tients dying between day 5 and 14 of therapy. The diagnosis
of aspergillosis was confirmed in all eight patients who
underwent autopsy after treatment with amphotericin B.
Autopsy was performed in seven patients who had been
treated with the combination; A. fumigatus was found in the
lungs of four and Candida in the liver and all major organs
in one case, while in one other patient there was evidence of
generalized aspergillosis together with candidosis. Granu-
locytopenia had been present on average for 18 days in the
amphotericin B group and 20 days in the combination
g_roup, and 16 patients (48%) were still profoundly granulo-
cytopenic with fewer than 100 cells/~tl when they died.
Granulocytes recovered in the three patients who survived
an A. fumigatus pneumonia, whilst the two patients suffer-
ing from C. parapsilosis candidemia were still granulocy-
topenic at the moment of restionse.
The average duration of therapy was 10 days in both
groups, accounting for a total dose of amphotericin B of
338 and 308 mg, respectively. All trough serum levels of
5-flucytosine were between 20 and 30 rag/l, whereas the
peak values did not exceed 120 mg/1. Only 53% of patients
tolerated a dose of more than 0.5 mg/kg/day amphoteri-
cin B. Elevation of the serum creatinine was seen in six
patients treated with amphotericin B alone and in seven of
those treated with the combination, whereas all patients
required suppletion with potassium. No abnormalities in
liver function or bone marrow toxicity, attributable to the
study drugs, were observed.
Discussion
The results of this study were singularly disappointing and
this observation was one of the reasons that led to the early
termination of the study. Neither treatment afforded an ac-
ceptable outcome since only five patients survived. This
may have been due to inadequate dosing with amphoteri-
cin B since 1 mg/kg is now regarded as the optimal dose.
However, the dose is largely determined by patient toler-
ance and any side effects that oCcur require adjusting the
dose downwards and concomitant symptomatic treatment.
It was expected that the addition of 5-flucytosine might
have compensated for any deficiency in the dose of ampho-
tericin B by providing a degree of synergism especially
where Candida was concerned [18, 20]. Clearly, this was
not the case and even had it been so, infection had almost
certainly progressed to such an advanced state that any
treatment would have failed. In fact, the single most im-
portant reason for the poor survival rate was probably the
delay caused by requiting the infection to be proven before
P. E. Verweij et al.: Amphotericin B with or without 5-Flucytosine in Neutropenic Patients
allowing the patient to enter, as was noted in earlier re-
ports [22-27]. In view of the low number of patients en-
tered and the strict criterium of proven systemic infection
no definite conclusion can be drawn about the possible
superiority of the combination of amphotericin B and 5-flu-
cytosine over amphotericin B alone. Although the data are
discouraging, the combination might prove beneficial in an
empiric setting or when the diagnosis can be made at an
early stage of the infection, which makes a prolonged
course of antifungal therapy possible [20]. The failure to
recover granulocytes was also a poor prognostic factor [28]
and perhaps indicates a potential role for hematopoietic
growth factors. It was also noteworthy that all patients had
either a new fever or one that had persisted for more than 8
days despite broad spectrum antibiotics. The latter group
should have been candidates for the current practice of
early empiric therapy with amphotericin B and might ac-
tually have survived [14, 15]. A second reason to stop the
trial was the fact that promising agents like fluconazole
[29] and liposomal amphotericin B [30] became available
for clinical use and therefore the continuation of the study
could no longer be ethically sustained.
The majority of patients had complied with the relatively
high dose of 2 g oral amphotericin B as prophylaxis, yet
still succumbed to infection. This was not surprising for
those who developed aspergillosis since the polyene is
not absorbed and would therefore not have had any effect
in the lung. However, oral polyenes were shown by the
EORTC Antimicrobial Cooperative Group to be an im-
portant negative prognostic factor per se, rather than af-
fording any protection against systemic mycosis [14].
Moreover, prophylaxis with polyenes given orally has
never been shown to be effective in a randomized,
placebo controlled setting, not even for Candida. At the
time of this study, we gave prophylaxis with oral ampho-
tericin B the benefit of the doubt since there was no
alternative.
In fact, the results obtained in the present study are simi-
lar to those reported in previous studies [22-28, 3!] if
only histologically or microbiologically documented
cases are taken into account, indicating that the study
group was representative of a poor risk population com-
promised further by prolonged and profound granulocy-
topenia. Hence, the problem of diagnosing infection
early [7] poses as big a dilemma as when to begin ther-
apy. Too long a delay while awaiting laboratory confir-
mation will result in too high a failure rate, while starting
too early on the grounds of persistent fever will lead to
over-treatment. In trying to resolve this problem, one
must consider the epidemiology of fungal infections, the
diagnostic tools that are readily available such as chest
roentgenogram, the value of surveillance cultures as well
as the newer techniques of antigen and nucleic acid de-
tection. Candida infection is endogenous and appears to
occur only in those patients who are colonised. The gas-
trointestinal tract is the commonest reservoir and there-
fore the most likely portal of entry into the bloodstream.
Infection 22 (1994) No. 2 © MMV Medizin Verlag GmbH Mtinchen, Mtinchen 1994 83/23

P. E. Verweij et al.: Amphotericin B with or without 5-Flucytosine in Neutropenic Patients
Dissemination is facilitated by the ulceration and damage to
the epithelium that results from chemotherapy and possibly
coexistent infection with herpes simplex. Yeasts may also
gain direct access to the bloodstream via indwelling vascu-
lar catheters.
Aspergillosis presents an entirely different problem because
it is mainly confined to the lungs, at least in the early stages
when the organism is difficult to detect. Infection is usually
associated with clinical and roentgenologic findings sug-
gestive of the diagnosis such as focal infiltrates on a chest
roentgenogram, although disseminated infection occurs in
about 20% of the cases. Since these organisms have a pro-
pensity for invading the blood vessels and causing throm-
bosis, organ infarction is frequently a prominent manifesta-
tion. Reduction of spores in the ambient air is the most
Zusammenfassung: Amphotericin B mit und ohne 5-Flucy-
tosin: Schlechte Ergebnisse bei neutropenischen Patienten
mit gesicherten systemischen Mykosen. Achtundzwanzig
neutropenische (< 500 Granulozyten/gl) Erwachsene mit einer
mikrobiologisch oder klinisch nachgewiesenen, systemischen
Mykose wurden randomisiert und behandelt mit Amphoteri-
cin B (0,5 mg/kg/Tag, n = 14) oder mit einer Kombination
von Amphotericin B (0,5 mg/kg/Tag) mit 5-Flucytosin
(100 mg/kg/Tag, n = 14) intraven6s. Im Durchschnitt wurden
beide Gruppen tiber 10 Tage behandelt, so dab sich eine totale
Amphotericin B Dosis yon 338 rag, bzw. 308 mg ergibt. Die
Dauer der Granulozytopenie betrug durchschnittlich 18 Tage in
References
1. Gold, J. W. M.: Opportunistic fungal infections in patients with neo-
plastic disease. Am. J. Med. 72 (1984) 458-463.
2. Horn, R., Wong, B., Kiehn, T. E., Armstrong, D.: Fungemia in a
cancer hospital: changing frequency, earlier onset, and results of ther-
apy. Rev. Infect. Dis. 7 (1985) 646-655.
3. Whimbey, E., Kiehn, T. E., Brannon, P., Bievins, A., Armstrong,
D.: Bacteremia and fungemia in patients with neoplastic disease. Am.
J. Med. 82 (1987) 723-730.
4. Saral, R.: Candida and Aspergillus infections in immunocompromised
patients. Rev. Infect. Dis. 13 (1991) 487-492.
5. Gerson, S., Talbot, G., Huwitz, S., Strom, B., Lusk, E., Cassileth,
P.: Prolonged granulocytopenia: the major risk factor for invasive
pulmonary aspergillosis in patients with acute leukemia. Ann. Int. Med.
100 (1984) 345-351.
6. Bodey, G. P., Bueltmann, B., Duguid, W., Gibbs, D., Hanak, H.,
Hotchi, M., Mall, G., Martino, P., Meunier, F., Milliken, S., Naoe,
S., Okudaira, M., Seevola, D., Van't Wont, J.: Fungal infections in
cancer patients: an international autopsy survey. Eur. J. Clin. Micro-
biol. Infect. Dis. 11 (1992) 99-109.
7. Tang, C. M., Cohen, J.: Diagnosing fungal infections in immunocom-
promised patients. J. Clin. Pathol. 45 (1992) 1-5.
8. Stover, D. E., Zaman, M.B., Hajdu, S. I., Lange, M., Gold, J.,
Armstrong, D.: Bronchoalveolm" lavage in the diagnosis of diffuse
pulmonary infiltrates in the immunosuppressed host. Ann. Int. Med.
101 (1984) 1-7.
9. Yu, V. L., Muder, R. R., Poorsatter, A.: Significance of isolation of
Aspergillus from the respiratory tract in diagnosis of invasive pulmo-
nary aspergillosis. Results from a three-year prospective study. Am. J.
Med. 81 (1986) 249-254.
effective way to reduce the risk of infection in susceptible
patients.
Conclusions
The prognosis of histologically and/or microbiologically
proven systemic mycosis in the febrile neutropenic patient
remains very poor. Under such circumstances it was not
possible to demonstrate any superiority of a combination of
amphotericin B and 5-flucytosine over amphotericin B
None, since the mortality was excessive in both study
groups. This emphasizes the need for an empiric antifungal
strategy, as well as the determination of methods to prevent
these infections or to improve techniques that enable early
diagnosis.
der Amphotericin B Gruppe, und 20 Tage in der Gruppe mit
Kombinationstherapie. Nur zwei Patienten in der Amphotericin
B Gruppe und drei in der Kombinationsgruppe tiberlebten die
Pilzinfektion. Die Nebenwirkungen waren vergleichbar in bei-
den Gruppen, mit einer Steigerung des Serum-Kreatinins bei
sechs Patienten w~ihrend der Verabreichung von Amphotericin
B alleine, und bei sieben Patienten w~rend der Kombinations-
therapie. Andere ernsthafte Nebenwirkungen traten nicht auf. In
beiden Gruppen war das Resultat der Behandlung entt~iuschend,
teilweise dadurch, dab die Mykose sich bereits am Anfang der
Therapie zu weit entwickelt hatte und dag nur bei der H~ilfte der
Patienten ein Anstieg der neutrophilen Granulozyten stattfand.
10. Walsh, T. J., Pizzo, P. A.: Treatment of systemic fungal infections:
recent progress and current problems. Eur. J. Clin. Microbiol. Infect.
Dis. 7 (1988) 460-475.
11. Ho, B., Cooperberg, P. L., Li., D. K. B., Maeh, L., Raiman, S. C.,
Grossman, L.: Ultrasonography and compufed tomography of hepatic
candidiasis in immunosuppressed patients. J. Ultrasound Med. 1
(1982) 157-159.
12. Thaler, M.,Pastakia, B.,Shawker, T. H.,O'Leary, T.,Pizzo, P. A.:
Hepatic candidiasis in cancer patients: The evolving picture of the
syndrome. Ann. Int. Med. 108 (1988) 88-100.
13. Pizzo, P. A., Robichaud, K. J., Gill, F. A., Witehsky, F. G.: Empiric
antibiotic and antifungal therapy for cancer patients with prolonged
fever and granulocytopenia. Am. J. Med. 72 (1984) 101 - 107.
14. EORTC International Antimicrobial Therapy Cooperative
Group.: Empiric antifungal therapy in febrile neutropenic patients.
Am. J. Med. 86 (1989) 668-762.
15. Walsh, T. J., Lee, J., Leeciones, J., Rubin, M., Butler, K., Francis,
P., Weinberger, M., Roilides, E., Marshall, D., Gress, J., Pizzo,
P. A.: Empiric therapy with amphotericin B in febrile greamlocy-
topenic patients. Rev. Infect. Dis. 13 (1991) 496-503.
16. Sacks, P.,Fellner, S. K.: Recurrent reversible acute renal failure from
amphotericin. Arch. Int. Med. 147 (1987) 593-595.
17. Scholer, H.J.: Flucytosine. In: Speller, D. C. E. (ed.): Antifungal
chemotherapy. John Wiley & Sons Ltd., London 1980, p. 86.
18. Polak, A., Seholer, H.J., Wail, M.: Combination therapy of ex-
perimental candidiasis, cryptococcosis and aspergillosis in mice. Che-
motherapy 28 (1982) 461~479.
19. Bennet, J. E., Dismukes, W. E., Duma, R. J., Medoff, G., Sande,
H. A., Gailis, H., Leonard, J., Fields, B. T., Bradshaw, M., Hay-
wood, H., McGee, Y. E., Cate, Th. R., Cobbs, C. G., Warner, J. F.,
24/84 Infection 22 (1994) No. 2 © MMV Medizin Verlag GmbH Miinchen, Miinchen 1994

Ailing, D. W.: A comparison of amphotericin B alone and combined
with flucytosine in the treatment of cryptococcal meningitis. N. Engl.
J. Med. 301 (1979) 127-131.
20. Polak, A.: Combination therapy for systemic mycosis. Infection 17
(1989) 203-209.
21. Polak, A.: Combination therapy for systemic mycosis. J. Chemother.
2 (1990) 211-217.
22. Aisner, J., Sehimpff, S. C., Wiernik, P. H.: Treatment of invasive
aspergillosis: relation to early diagnosis and treatment to response.
Ann. Int. Med. 86 (1977) 539-543.
23. Meunier-Carpentier, F., Kiehn, T. E., Armstrong, D.: Fungemia in
the immuno-compromised host: changing patterns, antigenemia, high
mortality. Am. J. Med. 69 (1981) 363-370.
24. DeGregorio, M. W., Lee, W. M. F., Linker, C. A.: Fungal infections
in patients with acute leukemia. Am. J. Med. 73 (1982) 543-548.
25. Holleran, W. M., Wilbur, J. R., DeGregorio, M. W.: Empiric am-
photericin B therapy in patients with acute leukemia. Rev. Infect. Dis.
7 (1985) 619-624.
26. Haron, E., Feld, R., Tuffnell, P., Paterson, B., Hasseibaeh, R.,
Matlow, A.: Hepatic candidiasis: an increasing problem in immuno-
compromised patients. Am. J. Med. 83 (1987) 17-26.
P. E. Verweij et al.: Amphotericin B with or without 5-Flucytosine in Neutropenic Patients
27. Denning, D. W., Stevens, D. A.: Antifungal and surgical treatment of
invasive aspergillosis: review of 2121 published cases. Rev. Infect.
Dis. 12 (1990) 1147-1201.
28. Maksymiuk, A.W., Thongprasert, S., Hopfer, R., Luna, M., Fain-
stein, V., Bodey, G. P.: Systemic candidiasis in cancer patients. Am.
J. Med. 77, Suppl. D (1984) 20-27.
29. Anaissie, E., Bodey, G. P., Kantarjian, H., David, C., Barnett, K.,
Bow, E., Defelice, R., Downs, N., File, T., Karam, G., Potts, D.,
Shelton, M., Sugar, A.: Fluconazole therapy for disseminated candi-
diasis in patients with leukemia and prior amphotericin B therapy. Am.
J. Med. 91 (1991) 143-150.
30. Lopez-Berestein, G.,Bodey, G. P.,Frankel, L~ S.,Mehta, K.: Treat-
ment of hepatosplenic candidiasis with liposomal amphotericin B.
Oncology 5 (1990) 310-317.
31. Van't Wout, J.W., Novakova, I., Verhagen, C.A.H., Fibbe,
W. E., De Pauw, B. E., Van der Meer, J. W. M.: The efficacy of
itraconazole against systemic fungal infections in neutropenic patients:
a randomised comparative study with amphotericin B. J. Infect. 22
(1991) 45-52.
Infection 22 (1994) No. 2 © MMV Medizin Verlag GmbH Miinchen, Mfinchen 1994 85/25