Journal of Antimicrobial Chemotherapy (1996) 37, 175-185
Antimicrobial practice
An audit of the use of antifungal agents
Felix Gutierrez*', Patrick G. Wall* and Jonathan Cohen'
"Servicio de Medicina Interna-Infecciosas, Hospital SVS Marina Baixa, 03570
Villajoyosa, Alicante, Spain; bPHLS Communicable Disease Surveillance Centre, 61
Colindale Avenue, London NW9 5EQ, UK; 'Department of Infectious Diseases and
Bacteriology, Royal Postgraduate Medical School, Hammersmith Hospital, Du Cane
Road, London W12 0NN, UK
A three month prospective audit of systemic antifungal therapy was undertaken in
a university hospital in the United Kingdom to determine the patterns of usage of
systemic antifungal drugs. The case notes of all patients receiving systemic antifungal
agents were reviewed daily and the appropriateness of therapy was determined
according to standards in the authoritative literature. One hundred and fifteen
courses of treatment with antifungal agents for 1481 days were administered to 74
patients. When a patient had received more than one course of antifungals, both the
individual courses and the combination of courses, i.e. a regimen, were judged.
Antifungal agents were given as an empirical therapy or as prophylaxis in 68% of
the courses, with fluconazole and amphotericin B desoxycholate being the agents
most frequently used. Therapy was considered unconventional in 26.9% of the
courses and 40.5% of the regimens, mainly because either the indication or the
duration of treatment did not conform to conventional practice. Improvement in
prescribing practices and standards for ongoing audit can only be achieved if a
consensus can be reached on how these agents can be used appropriately.
Introduction
In recent years, several new antifungal agents active against systemic mycoses have been
developed. The availability of these agents has important implications for patient care.
This is particularly true for the azole compounds, a rapidly expanding family of
antifungal agents, as well as for the new formulations of amphotericin B (AmB) using
liposomes or lipid complexes (AmB-LP) as delivery systems. These new drugs may be
less toxic and have a more favourable profile than AmB-desoxycholate and flucytosine
(Lopez-Berestein et al., 1989; Meunier, Prentice & Ringden, 1991; Como & Dismukes,
1994).
The availability of a wider range of less toxic antifungal agents has facilitated a more
aggressive approach to the prophylaxis and treatment of fungal infections and may have
•Corresponding author. Tel: +34-6-685-9247; Fax: +34-6-685-9300.
175
0305-7453/96/010175 + 11 SI2.00/T
£, 1996 The British Society for Antimicrobial Chemotherapy
176
F. Gutierrez et at.
precipitated the uninformed and inappropriate prescription of these drugs leading to
concerns about the emergence of resistant organisms (Warnock et al., 1988; Case et al.,
1991; Persons et al., 1991; Wingard et al., 1991). Several factors may contribute to the
inappropriate prescribing of antifungal agents including the absence of strict guidelines
for their use, the complexity of the patient population for whom antifungal therapy is
most often indicated and the understandable desire of physicians to use the best drugs
available. Novelty and the promotional activities of drug manufacturers may also play
a role.
Understanding the common prescribing patterns of antifungal compounds is
therefore critical in designing future therapeutic strategies.
Medical audit is a systematic process for improving clinical outcome by comparing
performance with agreed best practice, identifying variations and suggesting
improvements and evaluating their effect. However, standards of best practice are not
always easy to establish and, despite an increasing prevalence of fungal infections and
a widening experience in their management, clear guidelines have yet to be established
for their treatment, the absence of which undoubtedly hampers the assessment of the
appropriateness of therapy. To learn more about prescribing practices with a view to
improving them, we undertook a three month prospective audit of systemic antifungal
usage in the Hammersmith Hospital.
Patients and methods
The Hammersmith Hospital is a 500 bed tertiary care teaching hospital which
admits approximately 20,000 patients annually. The hospital has a bone marrow
transplant unit and a cancer treatment unit but is not a major centre for AIDS patients.
During the study period, there was no restriction on the use of antifungal agents and
the systemic antifungal agents routinely used included parenteral amphotericin-desoxycholate (AmB-DC) and liposomal amphotericin (AmB-LP), oral and iv flucytosine,
fluconazole and miconazole and oral ketoconazole and itraconazole. All inpatients who
were given systemic antifungal agents between 3 August and 3 November 1992 were
included in the study. Their clinical notes, drug charts and nursing notes were reviewed
on a daily basis and the underlying disease, the reason for initiating therapy, the dosage,
means of administration and duration of therapy and any adverse reactions were
recorded on data collection sheets. The number of antifungal days was determined by
adding up the total number of antifungal agents given on each day of treatment, e.g.
two agents given for 10 days would represent 20 antifungal days. All the microbiological
investigations requested and the results were obtained directly from the computer
database of the Microbiology Department.
The reasons for using antifungal drugs were classified according to Gross, Pichard
& Perfect (1987) as the treatment of a defined fungal infection, empirical treatment, or
prophylaxis.
Definitions of fungal infection
A defined systemic fungal infection was one in which there was both clinical evidence
of blood or tissue infection and a pathogenic fungus had been cultured from a biopsy
or other appropriate specimen from the site involved (Goodman et al., 1992). A proven
superficial fungal infection was defined as a clinically apparent infection of the
Audit of antifungal use
177
oropharynx, skin, or genitalia that was confirmed by cultures. Any clinically suspected
fungal infection, whether systemic or not, that had not been confirmed by culture or
histology, was considered a possible fungal infection. The diagnosis of a probable fungal
infection was reserved for any haematological patient with neutropenia (granulocyte
counts < 1.0 x 109/L) persisting for more than one week and radiographic evidence
typical of invasive aspergillosis of the sinuses or lungs.
Types of treatment
The treatment of proven or probable infection was considered specific and that of a
possible fungal infection was defined as empirical. Treatment given because a patient
was thought to be at risk of developing a fungal infection but where there was neither
clinical nor mycological evidence was considered as prophylaxis.
Methods of assessment
Courses' lasting less than 24 h were excluded from the analysis. Antifungal agents
given simultaneously comprised a single regimen and were assessed as such except
when assessing a course of therapy in which case the use of each drug was assessed
independently.
Assessment of specific therapy
The appropriateness of drugs used to treat proven fungal infection was determined
jointly by FG and PGW who followed the guidelines published by the Infectious
Diseases Society of America, whereby the information obtained on the indication, the
agents selected, the dose, timing, route of administration and duration of treatment were
evaluated in relation to the current state-of-the-art practices as noted in the
authoritative literature (Marr, Moffet & Kunin, 1988). The use of AmB-LP in
haematological patients was evaluated according to the guidelines introduced in the
hospital in September 1991 which required a patient to have a proven or probable fungal
infection and a deteriorating renal function denned by a serum creatinine > 140 /imol/L
that could not be reversed, e.g. by withdrawing another nephrotoxic drug. AmB-LP
could also be used as a substitute for the standard preparation to treat any patients who
developed renal failure defined by a serum creatinine > 200 ^mol/L. If renal function improved during treatment, an attempt was made to revert to the standard
preparation.
Each course of antifungals was classified as conventional if it was given in accordance
with the practice recommended in the literature (Bennet, 1990; Carr & Dismukes, 1992;
Medical Letter, 1992; Terrel & Hughes, 1992) or, in the case of AmB-LP, in accordance
with the hospital's guidelines. Otherwise it was assessed as unconventional.
The indication was assessed as conventional for any course of drugs used to treat
proven fungal infection as was the use of AmB as the drug of first choice.
Treating infections caused by yeasts with either fluconazole or itraconazole was
considered conventional, but not their use in treating systemic infections caused by
filamentous fungi. For patients with normal renal function 0.3-1.5 mg/kg/day AmB,
50-400 mg/day fluconazole, 50-150 mg/kg/day flucytosine, 100-400 mg/day itraconazole, 600-3600 mg/kg/day parenteral miconazole were all considered conventional
178
F. Gutierrez et at.
dosages. When treating oesophageal candidosis, between five and 14 days' treatment
with AmB or between 2-3 weeks' treatment with azole compounds was considered
conventional. Treatment of oropharyngeal candidosis for 5-14 days with azole
compounds was assessed as conventional as was the treatment of candidaemia without
apparent dissemination with either a total dose of 250-1000 mg of AmB or at least 2
weeks' treatment with azole drugs. Treatment of proven fungal infection with either
formulation of AmB was to be continued until a combined total dose of 1.5 g AmB had
been administered.
Assessment of empirical courses
The indication was assessed as conventional for all immunocompromised patients who
were deteriorating clinically or had persistent fever despite treatment with broad
spectrum antibiotics, or who had developed signs and symptoms of pulmonary disease,
severe oral mucositis or oesophagitis. The use of AmB as the drug of first choice was
accepted as conventional in every case but the use of fluconazole or itraconazole was
only accepted for treating severe oral mucositis or oesophagitis. The standards for the
dose were the same as those used for assessing specific therapy and the duration of
therapy was assessed according to how long the clinical indications still applied. The
duration of treatment was considered unconventional when it was continued after the
granulocyte count had returned to normal.
Assessment of prophylactic courses
Fluconazole given to bone marrow transplant recipients with neutropenia (Goodman
et al., 1992) and AmB given as secondary prophylaxis to any patients with neutropenia
and a past history of invasive aspergillosis was considered conventional, whereas the
use of systemic antifungal agents for any other purpose was deemed unconventional.
The standards for the dose were the same as those used for assessing specific therapy
and the duration of treatment was considered conventional when it was continued only
until the granulocyte count had returned to normal. The course was judged
conventional only when each individual assessment was conventional and a regimen was
deemed conventional if every course was judged to be conventional and the combination
of antifungal agents employed was considered likely to prove effective or to be
reasonable according to the literature. Any trial of therapy was considered an
unconventional indication unless it was initiated because of lack of response to the usual
therapy.
Drug costs
Details of drug expenditure were obtained from the purchasing records of the pharmacy
in which the costs were calculated according to the prices charged during the study
period.
Results
Seventy-four (1.8%) of the 4,111 patients admitted to the Hammersmith hospital
during the three month audit period were given systemic antifungal agents but the
Audit of antifungal use
179
Table I. Course of antifungal drugs used for different categories of
treatment
Drug
Prophylaxis
Empirical
Specific
AmB-DC
AmB-LP
Fluconazole
Flucytosine
Itraconazole
Miconazole
2
—
13
1
-
22
4
36
-
8
6
14
4
4
1
32
10
63
4
5
1
Total
16
62
37
115
Total
proportion of patients treated varied within the hospital with 17 (65%) of the 26 patients
in the Bone Marrow Transplant Unit, 52 (27%) of the 192 in the Haematology
ward, 10 (6%) of the 165 in the Oncology ward and 17 (1.6%) of the 1,052 in the other
Medical wards being given the agents. Patients had been transferred from one ward to
another during their stay in hospital but the antifungal treatment they received was
judged as a whole.
A total of 115 treatment courses were administered and the drugs used for each
indication are illustrated in Table I. Fluconazole was the most frequently employed
agent for each indication and, together with AmB-DC, accounted for 95 (83%) of the
courses of treatment and for 1318 (89%) of the 1481 antifungal days. Ketoconazole was
not given to any patient during the period of audit and itraconazole, flucytosine and
miconazole were seldom used. Prophylaxis was given for a total of 322 days, empirical
therapy for 729 days and specific therapy for 430 days giving an average of 20.1, 11.8
and 11.6 days respectively. Sixteen courses of prophylaxis were given to 15 patients (as
one patient was given two separate courses). The haematology department accounted
for 66 (57%) of the courses, 31 of which were administered in the Bone Marrow
Transplant Unit (Table II). Within each department, fluconazole and AmB-DC were
the agents most frequently used for each indication.
Table II. Use of antifungal agents in different wards
Patients'
Antifungal courses
Haematology
Bone marrow transplant unit
Oncology ward
Other medical wards
Paediatric ward
Intensive care unit
Surgery
Prophylaxis
Empirical
Specific
15
16
8
6
1
1
_
-
43
62
19
23
6
11
2
1
-
20
37
8
2
9
10
4
2
2
•Four neutropenic patients on prophylaxis subsequently received empiric therapy.
'18% (14/74) patients had more than one underlying disease.
180
F. Gutierrez et at.
Table III. Defined fungal infections
Site of infection"
Oesophagus-oropharynx (8)
Blood stream (4)
Peritoneum (4)
Urinary tract (2)
Lung (1)
Soft tissue (1)
Fungi isolated*
Candida albicans (5)
Candida parapsilosis (1)
Toruplosis glabrata (1)
Candida sp. (1)
C. albicans (3)
Fusarium sp. (1)
C. albicans (2)
Candida tropicalis (1)
Paecilomyces sp. (1)
C. albicans (2)
Aspergillus sp. (1)
C. albicans (1)
'Number of patients
'Number of patients in whom the particular fungus was isolated.
Indications and drugs used
Sixteen courses of prophylaxis antifungal were given to eight patients who were
neutropenic, of whom one received two separate courses, to six BMT recipients and to
one HIV-positive patient. Fluconazole was the agent most frequently employed
(Table II). Itraconazole was given as secondary prophylaxis to a patient treated for
acute leukaemia because of neutropenia and a past history of invasive aspergillosis.
Twenty eight courses of empirical therapy were given to patients with fever of
uncertain or undefined aetiology, 27 of whom had been given broad spectrum
antibiotics previously. Twenty seven courses were given to patients complaining of oral
mucositis or oesophagitis. Five courses were given to patients with signs and symptoms
of pulmonary disease and two were given to patients with skin erythema. Thirty six
courses of fluconazole and 22 courses of AmB-DC were given for empirical therapy
accounting for 58 (93%) of the 62 courses and 700 (96%) of the 729 antifungal days,
the drugs being used in combination eleven times. Seventeen neutropenic patients were
given 31 courses of antifungal agents for empirical therapy for 402 (55%) antifungal
days. Twenty-two of these courses were given to neutropenic patients with fever of
uncertain or undefined aetiology and AmB-DC was given in 17 treatment courses. A
total of 31 courses of empirical therapy was given for 327 antifungal days to 25
non-neutropenic patients, of whom sixteen had malignant tumours. Fluconazole was
used in 24 (77%) courses empirically, mainly because of clinical suspicion of mucosal
infection (17 courses).
Specific therapy was given to 17 patients with candidiasis and three with mould
infections (Table III) for 430 antifungal days. Fourteen courses of AmB were given for
200 (47%) antifungal days. Twelve courses were employed for deep seated mycosis and
another two for oral and oesophageal infections. AmB-LP was given for six of the
treatment courses. Fluconazole was given in 14 courses accounting for 155 (36%) of
the antifungal days. Nine courses of fluconazole were given for oral and oesophageal
infections.
Audit of antifungal use
181
Assessment of courses
Overall, antifungal use was considered conventional in 83 (72%) of the courses
(Table IV). Fluconazole was involved in 25 (81%) of the 31 courses considered
unconventional. Six (38%) of the prophylactic courses were assessed as conventional.
Nine prophylactic courses (56%) were considered unconventional because systemic
antifungal agents were given to five patients with chronic leukaemia, to two with aplastic
anaemia, and to one patient suffering chronic graft versus host disease and another with
acute leukaemia. One course given to a patient with neutropenia undergoing BMT was
deemed unconventional because the treatment was continued long after the neutrophil
count had returned to normal.
Forty five (73%) evaluable courses of empirical treatment were considered
conventional. Of the 16 unconventional courses, seven were given for an unconventional
indication, four times for mild, non-specific oral mucositis in non-neutropenic patients,
twice for skin erythema, and once for low grade fever in a nonneutropenic patient. In
another two cases, fluconazole was given to two patients with signs and symptoms of
pulmonary involvement. Six courses were continued after the neutrophil count had
returned to normal. In the final case 400 mg/day fluconazole had been given to a patient
with endstage renal disease when a lower dose was considered more appropriate.
Thirty-two (87%) of the 37 courses of specific treatment were considered
conventional. Of the five courses deemed unconventional, AmB-LP was given to two
patients with normal renal function and fluconazole was given to two patients who had
mould infections. Another course of specific therapy was assessed as unconventional
because fluconazole had only been given for five days to treat oesophageal candidiasis.
Assessment of regimens
Overall, the use of antifungal regimens was considered conventional in 49 (58%) of the
84 regimens (Table V). In all cases of prophylaxis, a regimen consisted of one agent
and six were assessed as unconventional. Twenty eight (58%) of 48 empirical regimens
were considered conventional. Nineteen empirical regimens were assessed as
unconventional because at least one of the courses was unconventional and the other
six because fluconazole and AmB were given together. Among patients with defined
Table IV. Assessment of courses (n = 115) of antifungal therapy
Prophylaxis
Patients'
Courses
Conventional
Unconventional
indication
drug chosen
dose
duration
Not evaluated
No of antifungal days
15
16
6
10
9
Empirical
43
62
45
16
7
2
Specific
20
37
32
5
—
4
i
1
322
1
6
1
1
729
430
'Four neutropenic patients on prophylaxis subsequently received empiric
therapy.
F. Gutierrez et at.
182
Table V. Assessment of regimens of antifungal therapy
Patients
Regimens
Conventional
Unconventional
Not evaluated
Prophylaxis
Empirical
Specific
15
16
6
10
-
43
48
28
19
1
20
20
15
5
-
infection 15 (75%) of the 20 regimens were considered conventional. Three specific
regimens were assessed as unconventional, in one case because at least one of the courses
was unconventional and in two cases because fluconazole and AmB had been given in
combination.
Adverse reactions
Adverse drug reactions were associated with 24 (75%) of the 32 courses of AmB-DC
of which 20 were related to renal toxicity, although renal impairment led to withdrawal
of only six courses of the drug. In contrast, mild to moderate hypokalaemia
(2.5-3.5 mmol/L) was the only side-effect probably related to treatment with AmB-LP
and occurred in two of the 10 courses. Three (5%) courses of treatment with fluconazole
led to side effects probably relating to the drug involving two cases of nausea and
vomiting and one case of hepatotoxicity with an elevated alanine aminotransferase of
87.5 IU/L.
Cost of drugs
The cost of antifungal therapy during the study period accounted for approximately
20% of the hospital's expenditure of £245,000 on antimicrobial agents and 6% of the
total drug budget of £816,666. AmB-LP and fluconazole together accounted for nearly
90% of the total expenditure on antifungal agents (Table VI).
Table VI. Expenditure on antifungals between 3 August and 3 November 1992
Drug
Amphotericin B liposomal
Fluconazole
Amphotericin B-DC
Itraconazolc
Flucytosine
Miconazole
Total
No. of
courses
Antifungal
days
Cost in pounds
sterling (%)
10
63
32
5
4
1
81
843
483
40
34
2
£35,000
£9000
£3000
<£1000
<£1000
-
115
1481
£49,000
(71)
(18)
(6)
(2)
(2)
Audit of antifungal use
183
Discussion
In this study, antifungal agents were often used in an unconventional manner although
this does not necessarily mean that such uses were always inappropriate, as it is difficult
to audit the use of antifungal agents when there are no rigid standards against which
to compare practice. In a retrospective audit of AmB use carried out in the USA, no
attempt was made to determine whether antifungal drugs had been used correctly or
incorrectly mainly because of problems associated with the clinical diagnosis of fungal
infections and the lack of consensus on doses and duration of treatment for certain
infections (Gross et al., 1987). Consequently, we only attempted to assess whether the
antifungal drugs were used according to current accepted clinical practice as reported
in the authoritative literature (Bennet, 1990; Carr & Dismukes, 1992; Medical Letter,
1992; Ten-el & Hughes, 1992).
Antifungal agents, particularly fluconazole, were being used mainly for empirical
therapy or prophylaxis. It was also interesting to note that, unlike at Duke University
Hospital where AmB was used in 70% of the cases in the treatment of documented
infections (Gross et al., 1987), AmB was given more often in the hospital for empirical
therapy than for the specific treatment of defined fungal infections. This was not entirely
unexpected since empirical antifungal therapy has become conventional practice for
managing the febrile, neutropenic patient (Pizzo et al., 1982; EORTC, 1989; Walsh
et al., 1991). However, it was surprising that only half of the empirical courses of
antifungal therapy were given to neutropenic patients.
The use of systemic antifungal agents, mainly fluconazole, for prophylaxis for
recipients of bone marrow transplantation and neutropenia is also considered as a
conventional practice (Goodman et al., 1992). However, during the audit period,
fluconazole was given more often in settings where its efficacy in preventing systemic
fungal infection has not yet been proven.
Nearly half of all regimens were assessed as unconventional largely because of the
unconventional use of prophylactic regimens. However, a substantial proportion of
treatment regimens was also unconventional because fluconazole and AmB were given
together, a practice that is considered both unorthodox and controversial (Schaffer &
Frick, 1985; Gallis, Drew & Pickard, 1990; Albert, Graybill & Rinaldi, 1991; Anaissie,
1992).
Overall, fluconazole was the drug most often employed unconventionally and
the majority of courses were given to nonneutropenic patients for whom fungal
infection was generally not life threatening. Indeed, the most common reason for using
systemic antifungal agents, particularly fluconazole, in this population was to treat
suspected mucosal infection. This suggests that physicians found it easier to give
fluconazole, especially to patients suffering oral mucositis induced by chemotherapy,
rather than confirm the suspected diagnosis of oral candidosis which only requires that
blastospores, hyphae and pseudohyphae are seen on a direct smear from friable,
white curdy lesions characteristic of candidiasis and that a culture yields Candida
albicans. Also, permissive uncontrolled use seems to be the fate of any new drug
which, like fluconazole, is relatively safe, non-toxic and inexpensive and can be given
orally. Only agreed guidelines for use and regular audit will stem this tendency to be
profligate.
The inappropriate use of antifungal agents increases the cost of patient care and may
diminish its quality. Moreover, with the arrival of the competitive market in health care
184
F. Gutierrez et al.
and ever increasing budgetary constraints, the direct cost of prescribing a course of
antifungal treatment inappropriately may mean that any further opportunity for
employing more useful alternatives may be forfeited. This can only be prevented by
improving prescribing practices and employing effective audit, neither of which can take
place until protocols for drug use become available to provide standards against which
to measure performance.
Acknowledgements
This study was supported in part by grant 92/5302 from the Spanish Fondo de
Investigation Sanitaria.
References
Albert, M. M., Graybill, J. R. & Rinaldi, M. G. (1991). Treatment of murine cryptoccocal
meningitis with an SCH 39304-amphotericin B combination. Antimicrobial Agents and
Chemotherapy 35, 1721-25.
Anaissie, E. (1992). Opportunistic mycoses in the immunocompromised host: experience at a
cancer centre and review. Clinical Infectious Diseases 14, Suppl. 1, S43-53.
Bennett, J. E. (1990). Antifungal Agents. In Principles and Practice of Infectious Diseases, 3rd edn
(Mandell, G. L., Douglas, R. G. & Bennet, J. E., Eds), pp. 361-70. Churchill Livingstone,
New York.
Carr, M. & Dismukes, W. E. (1992). Antifungal Drugs. In Infectious Diseases (Gorbach, S. L.,
Bartlett, J. G. & Blacklow, N. R., Eds), pp. 306-13. Saunders, Philadelphia.
Case, C. P., MacGowan, A. P., Brown, N. M., Reeves, D. S., Whitehead, P. & Felmingham, D.
(1991). Prophylactic oral fluconazole and Candida fungaemia. Lancet 337, 790.
Como, J. A. & Dismukes, W. E. (1994). Oral azole drugs as systemic antifungal therapy. New
England Journal of Medicine 330, 263—72.
EORTC International Antimicrobial Therapy Cooperative Group (1989). Empiric antifungal
therapy in febrile granulocytopenic patients. American Journal of Medicine 86, 668-72.
Gallis, H. A., Drew, R. H. & Pickard, W. W. (1990). Amphotericin B: 30 years of experience.
Reviews of Infectious Diseases 12, 308-29.
Goodman, J. L., Winston, D. J., Greenfield, R. A., Chandraseker, P. H., Fox, B., Kaizer, H.
et al. (1992). A controlled trial of fluconazole to prevent fungal infections in
patients undergoing bone marrow transplantation. New England Journal of Medicine 326,
845-51.
Gross, M. H., Pichard, W. W. & Perfect, J. R. (1987). Retrospective review of amphotericin B
use in a tertiary-care medical centre. American Journal of Hospital Pharmacy 44, 1353-7.
Lopez-Berestein, G., Bodey, G. P., Fainstein, V., Keating, M., Frankel, L. S., ZelufF, B. et al.
(1989). Treatment of systemic fungal infections with liposomal amphotericin B. Archives of
Internal Medicine 149, 2533-6.
Marr, J. J., MofTet, H. L. & Kunin, C. M. (1988). Guidelines for improving the use of
antimicrobial agents in hospitals: a statement by the Infectious Diseases Society of America.
Journal of Infectious Diseases 157, 869-76.
Medical Letter (1992). Drugs for treatment of fungal infections. Medical Letter on Drugs and
Therapeutics 34, 14—16.
Meunier, F., Prentice, H. G. & Ringden, O. (1991). Liposomal amphotericin B (AmBisome):
safety data from a phase II/III clinical trial. Journal of Antimicrobial Chemotherapy 28,
Suppl. B, 83-91.
Persons, D. A., Laughlin, M., Tanner, D., Perfect, J., Gockerman, J. P. & Hathorn, J. W.
(1991). Fluconazole and Candida krusei fungaemia. New England Journal of Medicine 325,
1315.
Pizzo, P. A., Robichaud, K. J., Gill, F. A. & Witebsky, F. G. (1982). Empiric antibiotic and
antifungal therapy for cancer patients with prolonged fever and granulocytopenia. American
Journal of Medicine 72, 101-11.
Audit of antifungal use
185
Schaffner, A. & Frick, P. G. (1985). The effect of ketoconazole on amphotericin B in a model
of disseminated aspergillosis. Journal of Infectious Diseases 151, 902-10.
Terrell, C. L. & Hughes, C. E. (1992). Antifungal agents used for deep-seated mycotic infections.
Mavo Clinic Proceedings 67, 69-91.
Walsh, T. J., Lee, J., Lecciones, J., Rubin, M., Butler, K., Francis, P. et al. (1991). Empiric
therapy with amphotericin B in febrile gTanulocytopenic patients. Reviews of Infectious
Diseases 13, 496-503.
Warnock, D. W., Burke, J., Cope, N. J., Johnson, E. M., von Fraunhofer, N. A. & Williams,
E. W. (1988). Fluconazole resistance in Candida glabrata. Lancet ii, 1310.
Wingard, J. R., Merz, W. G., Rinaldi, M. G., Johnson, T. R., Karp, J. E. & Saral, R. (1991).
Increase in Candida krusei infection among patients with bone marrow transplantation and
neutropenia treated prophylactically with fluconazole. New England Journal of Medicine 325,
1274-7.
(Received 2 June 1994; returned 12 July 1994; revised II October 1994; accepted
1 October 1995)