Oral candidiasis associated with inhaled

corticosteroid use: comparison of fluticasone

and beclomethasone
Chizu Fukushima, MD; Hiroto Matsuse, MD; Shinya Tomari, MD; Yasushi Obase, MD;
Yoshitsugu Miyazaki, MD; Terufumi Shimoda, MD; and Shigeru Kohno, MD

Background: Inhaled steroids such as fluticasone propionate and beclomethasone dipropionate play a central role in the
treatment of bronchial asthma. Fluticasone exhibits excellent clinical effectiveness; however, oral adverse effects can occur.
Objective: To compare the frequency of oral candidiasis in asthmatic patients treated with fluticasone and beclomethasone,
to evaluate the effect of gargling with amphotericin B, and to measure the inhalation flow rate on candidiasis.
Methods: The study consisted of 143 asthmatic patients who were treated with inhaled steroids, 11 asthmatic patients not
treated with inhaled steroids, and 86 healthy volunteers. Quantitative fungal culture was performed by aseptically obtaining a
retropharyngeal wall swab from these patients. Patients with positive results were treated with gargling using a 1:50 dilution
amphotericin B solution. In asthmatic patients treated with fluticasone, the inhalation flow rate was measured using an inspiratory
flow meter.
Results: The amount of Candida spp. was significantly greater in asthmatic patients taking inhaled steroids compared with
those who were not. It was also significantly greater in patients with oral symptoms than asymptomatic patients and significantly
greater in asthmatic patients treated with fluticasone than in those treated with beclomethasone. Although the presence of
Candida did not correlate with the inhaled dose of beclomethasone, it did increase with the dose of fluticasone. Gargling with
amphotericin B was effective in most asthmatic patients with candidiasis. Candidiasis was not due to inappropriate flow rates
during inhalation of steroids.
Conclusions: Fungal culture of a retropharyngeal wall swab may be useful for predicting the risk of developing oral
candidiasis in asthmatic patients treated with inhaled steroids. The amount of isolated Candida was significantly greater in
asthmatic patients treated with fluticasone than in those treated with beclomethasone. Attention to dosage is required as the
amount of Candida increased with dose of fluticasone. Gargling with a 1:50 dilution of amphotericin B is effective in treating
oral candidiasis of asthmatic patients treated with inhaled steroids.
Ann Allergy Asthma Immunol. 2003;90:646– 651.

INTRODUCTION casone propionate and those treated with beclomethasone

Since airway inflammation was identified as the major patho-
logic condition in bronchial asthma, inhaled steroids have
played a central role in asthma treatment.1 The use of steroids
facilitates disease control in asthmatic patients, resulting in
improved quality of life. The frequency of systemic effects,
such as the suppression of the hypothalamic-pituitary-axis
and the immune system, and other long-term effects on
specific organs, such as the skin, eyes, skeletal system, and
gastrointestinal tract, is markedly lower in asthmatic patients
treated with inhaled corticosteroids than in those treated
systemically.2 However, the localized effects of inhaled cor-
ticosteroids on oral mucosa, such as fungal infections, can be
problematic.
Although several types of inhaled steroids are available,
we were interested in comparing the frequency of topical
adverse effects between asthmatic patients treated with fluti-
dipropionate, because these two major steroids are frequently
used for inhalation therapy in Japan. Since the anti-inflam-
matory effect of fluticasone is about twice that of be-
clomethasone, the clinical dose of fluticasone is half that of
beclomethasone. Despite this, the therapeutic value of fluti-
casone was reported to be similar to or higher than that of
beclomethasone.3 However, oral adverse effects often make
continuous fluticasone administration difficult. In the present
study, we compared the frequency of intraoral adverse effects
between asthmatic patients treated with fluticasone and those
treated with beclomethasone. In addition, we evaluated
whether gargling with amphotericin B is effective as a treat-
ment for these adverse effects. Furthermore, we also exam-
ined the possible role of low inspiratory flow rates in the
development of oral candidiasis by measuring the inhalation
flow rates.

MATERIALS AND METHODS

Second Department of Internal Medicine, Nagasaki University School of Patients
Medicine, Nagasaki, Japan.
Received for publication September 2, 2002. The study consisted of 143 patients with bronchial asthma (76
Accepted for publication in revised form December 9, 2002. men and 67 women; mean ⫾ SD age, 53.7 ⫾ 15.6 years) who

646 ANNALS OF ALLERGY, ASTHMA, & IMMUNOLOGY

were receiving ambulatory treatment with inhaled steroids for
more than 1 year at the Second Department of Internal
Medicine in Nagasaki University Hospital, 11 asthmatic pa-
tients not treated with inhaled steroids (8 men and 3 women;
mean ⫾ SD age, 25.5 ⫾ 4.2 years), and 86 healthy volunteers
(56 men and 30 women; mean ⫾ SD age, 25.2 ⫾ 3.9 years).
Of the 143 asthmatic patients, 96 (41 men and 55 women;
mean ⫾ SD age, 51.2 ⫾ 16.7 years; range, 18 – 82 years;
atopic type, 39.4%) were treated with fluticasone propionate
(Flutide Diskhaler; GlaxoWellcome, London, England; at 50,
100, or 200 ␮g per dose) and the remaining 47 (26 men and
21 women; mean ⫾ SD age, 54.8 ⫾ 15.1 years; range, 31– 81
years; atopic type, 37.5%) were treated with beclomethasone
dipropionate (Aldesin; Schering-Plough, Kenilworth, NY;
100 ␮g per dose). In addition, 69 patients had atopic bron-
chial asthma, whereas the remaining 74 had nonatopic bron-
chial asthma. Furthermore, 32 of the 143 had mild persistent
bronchial asthma, 53 had moderate persistent bronchial
asthma, and the remaining 58 had severe persistent bronchial
asthma. Beclomethasone was inhaled using a pressurized
metered-dose inhaler with a spacer (Inspire Ease; Schering-
Plough), whereas fluticasone was inhaled using a dry powder
inhaler. All asthmatic patients treated with inhaled steroids
gargled with water immediately after inhalation therapy. The
study protocol was approved by the Human Ethics Review
Committee of Nagasaki University School of Medicine, and
a signed consent form was obtained from each patient.
Procedures
The retropharyngeal wall was aseptically wiped with a swab
soaked in 1.8 mL of sterile saline, and the swab was then
stirred in the sterile saline. Subsequently, the throat swab
fluid (50 ␮L) was applied to chromoagar medium, followed
by incubation at 35° C for 48 hours. Furthermore, a 1:10
dilution of the throat swab fluid (50 ␮L) was also applied to
Sabouraud dextrose agar medium and then incubated at 35° C
for 48 hours. Subsequently, the number of fungal colonies
was counted and fungal species were identified. The mini-
2
mum detection limit was 10 CFU/mL. Based on the results of
fungal culture, the following issues were evaluated.
Comparison Based on the Presence or Absence of Asthma,
Inhaled Steroids, and Asthmatic Symptoms
The amount of Candida spp. detected by fungal culture was
compared among healthy subjects; asthmatic patients not
treated with inhaled steroids; those treated with inhaled ste-
roids and complaining of pharyngodynia, unpleasant feeling
in the pharynx, hoarseness, and dysgeusia; and those treated
with inhalation steroids not complaining of these symptoms.
Multiple Regression Analyses Regarding Factors
Associated with the Amount of Candida
Patient age or sex, types of inhalation steroids, inhalation
doses, and doses of oral steroids were evaluated. Since the
inhalation dose was determined based on the dose of be-
clomethasone, the dose of fluticasone was established at 50%
of that of beclomethasone. Furthermore, the following two
VOLUME 90, JUNE, 2003
issues were evaluated because significant correlations were
observed between the amount of Candida and the type and
dose of steroids inhaled. The detection rate and amount of
Candida were compared between asthmatic patients treated
with fluticasone and those treated with beclomethasone. The
correlation between dose of inhaled steroids and amount of
Candida was independently evaluated in asthmatic patients
treated with fluticasone or beclomethasone.
Correlation with Inhalation Flow Rate
In asthmatic patients treated with fluticasone, the inhalation
flow rate was measured by using an inspiratory flow meter
(Incheck; Clement Clark, Essex, England). By using an at-
tached adaptor, it equates with the resistance of the dry
powder inhaler. The inhalation flow rate was measured dur-
ing examination in the outpatient clinic.
Effectiveness of Amphotericin B
In asthmatic patients who tested positive for Candida, the
gargle solution was changed from water to a 1:50 dilution
(2000 ␮g/mL) of amphotericin B solution, and fungal culture
was performed 1 month later. The minimum inhibitory con-
centration (MIC) was independently measured in asthmatic
patients in whom the amount of Candida had decreased to a
level below the minimum detection limit and in those who
showed no such decrease. Patients who did not respond to
amphotericin B were treated with 100 mg daily of flucon-
azole for 7 days.
Statistical Analysis
Multiple regression analysis was performed. All data are
expressed as mean ⫾ SEM. Differences in the detection rate
and amount of Candida between asthmatic patients treated
with fluticasone and those treated with beclomethasone were
evaluated using the ␹2 test, whereas the detection rate of
Candida in the studies described in the other sections was
evaluated using the Mann-Whitney U test. P ⬍ 0.05 denoted
the presence of a statistically significant difference.
RESULTS
Comparison Based on the Presence or Absence of Asthma,
Inhaled Steroids, and Asthma-Related Symptoms
Candida spp. organisms were rarely detected in healthy sub-
jects and asthmatic patients not treated with inhaled steroids.
However, a significantly larger amount of Candida was de-
tected in asthmatic patients treated with inhaled steroids (Fig
1, P ⬍ 0.001). In addition, the amount of Candida was
significantly greater in symptomatic than in asymptomatic
patients (Fig 1, P ⬍ 0.001).
Multiple Regression Analyses of Factors Associated with
the Amount of Candida
Multiple regression analyses of patient age, sex, types of
inhaled steroids, inhalation doses, and doses of oral steroids
revealed that the type and dose of inhaled steroids were
positively correlated with amount of Candida (P ⫽ 0.03 and
0.05, respectively; Table 1).
647
 Figure 2. Comparison of amount of Candida detected by fungal culture
between asthmatic patients treated with beclomethasone dipropionate and
fluticasone propionate (mean ⫾ SEM). Patient background was not signifi-
cantly different between the two groups. However, both the detection rate
and amount of Candida were significantly greater in asthmatic patients
Figure 1. Comparison of the amount of Candida (mean ⫾ SEM) in treated with 50, 100, or 200 ␮g of fluticasone propionate than in those treated
healthy subjects and asthmatic patients not treated or treated with inhalation with 100 ␮g of beclomethasone dipropionate. Horizontal line indicates the
steroid therapy and with or without asthma-related symptoms. A signifi- minimum detection limit.
cantly large amount of Candida was detected in symptomatic patients with
bronchial asthma treated with inhalation steroids. Horizontal line indicates
the minimum detection limit.
patients treated with fluticasone or beclomethasone, the be-
clomethasone dose did not correlate with amount of Candida.
Table 1. Results of Multiple Correlation Regression Analysis
However, the amount of Candida increased with dose of
between the Amount of Candida (Log Scale) and Various Clinical
and Therapy-Related Factors
fluticasone (␹2⫽0.057, P ⫽ 0.02, Fig 3).
Variable t value P value Correlation with the Inhalation Flow Rate
The inhalation rate was measured in asthmatic patients
Age ⫺0.895 0.37
Sex 0.599 0.55
treated with fluticasone using the inspiratory flow meter, and
Type of inhalation steroid 2.138 0.03 the correlation between inhalation flow rate and amount of
Inhalation dose 1.983 0.049 Candida was evaluated. However, most asthmatic patients
Dose of oral steroids ⫺0.643 0.53 were found to inhale fluticasone at an appropriate flow rate,
and no correlation was evident between inhalation flow rate
and amount of Candida (Fig 4).

Comparison between Asthmatic Patients Treated with Effectiveness of Amphotericin B

Fluticasone and Beclomethasone
There were no significant differences in age and sex between
asthmatic patients treated with fluticasone and beclometha-
sone. The inhaled dose of fluticasone was approximately 50%
of that of beclomethasone. Candida was detected in 26% of
asthmatic patients treated with fluticasone, and the amount
detected was 5.6 ⫻ 105 ⫾ 52.3 CFU/mL. Candida was
detected in only 10% of asthmatic patients treated with be-
clomethasone, and the amount was 1.2 ⫻ 103 ⫾ 45.3 CFU/
mL. Although the background did not differ significantly
between the 2 groups, both detection rate and amount of
Candida were significantly greater in asthmatic patients
treated with fluticasone than in those treated with be-
clomethasone (Fig 2, P ⫽ 0.02).
Comparison of Inhalation Doses
When the correlation between dose of inhaled steroid and
amount of Candida was independently evaluated in asthmatic
In asthmatic patients positive for Candida, the gargle solution
was changed to amphotericin B. When fungal culture was re-
peated after 1 month of treatment with amphotericin B in 27
asthmatic patients, the amount of Candida had decreased to a
level below the minimum detection limit in 22 of the 27 patients.
In the remaining 5 patients, the amount of Candida was de-
creased to a level below the minimal detection limit after a
1-week treatment with 100 mg of fluconazole (Fig 5). Of the
Candida species detected in asthmatic patients, Candida albi-
cans was most prevalent (n ⫽ 27), followed by C. parapsilosis
(n ⫽ 3), C. glabrata (n ⫽ 3), C. tropicalis (n ⫽ 2), and C. krusei
(n ⫽ 2), although several different species of Candida were
simultaneously detected in some patients. However, there was
no difference in fungal species between asthmatic patients in
whom gargling with amphotericin B was useful or not. The MIC
of amphotericin B against fungi detected by culture did not differ
significantly among the five asthmatic patients in whom gar-
gling with amphotericin B was ineffective and five other patients

648 ANNALS OF ALLERGY, ASTHMA, & IMMUNOLOGY


Figure 4. Correlation between inhalation flow rate of fluticasone propi-
onate and amount of Candida. The inhalation flow rate of fluticasone did not
correlate with amount of Candida. Vertical line indicates minimum appro-
priate inspiratory flow rate during inhalation; horizontal line, minimum
detection limit.
randomly selected from among 22 asthmatic patients in whom
gargling with amphotericin B was effective (0.14 ⫾ 0.06 ␮L/mL
vs 0.16 ⫾ 0.06 ␮L/mL).
DISCUSSION
In this study, we evaluated the oral adverse effects of inhaled
steroids in association with the amount of Candida spp. in the
retropharyngeal wall as detected by fungal culture. Our re-
sults showed that healthy subjects and asthmatic patients not
treated with inhaled steroids rarely have Candida in the
retropharyngeal wall, indicating that asthmatic patients are
not inherently prone to oral candidiasis. Consistent with pre-
vious reports,4 a significantly large amount of Candida was
detected in asthmatic patients treated with inhalation steroids.
All asthmatic patients treated with inhalation steroids gargled
with water after inhalation therapy. Furthermore, all asth-
matic patients treated with beclomethasone used a spacer.
Several previous studies have reported the usefulness of
gargling with water and the use of a spacer,5 but our data
VOLUME 90, JUNE, 2003
Figure 3. Correlation between dose of inhalation steroids
and amount of Candida. The dose of fluticasone propionate
but not beclomethasone dipropionate correlated with the
amount of Candida. Horizontal line indicates the minimum
detection limit.
Figure 5. Effectiveness of antifungal agents on candidiasis (mean ⫾ SE).
The gargle solution was changed from water to amphotericin B in 27
asthmatic patients in whom Candida was detected, and the amount of
Candida was re-evaluated 1 month after such treatment. The amount of
Candida decreased to a level below the minimum detection limit in 22 of the
27 patients. In the remaining 5 patients for whom gargling with amphotericin
B was ineffective, the amount of Candida decreased to a level below the
minimum detection limit after a 7-day treatment with 100 mg of fluconazole.
Horizontal line indicates the minimum detection limit.
suggest that these measures were ineffective in preventing
such complications in some patients. In our study, the amount
of Candida was significantly greater in asthmatic patients
complaining of an unpleasant feeling in the throat and pha-
ryngodynia than in those without such symptoms. Although
many previous studies have reported the development of oral
candidiasis in asthmatic patients treated with inhaled steroids,
the reported frequency of such fungal infection varied from 0
to 77% due to the different definitions of oral candidiasis
among the respective studies.4,6 –9 When oral candidiasis is
defined as a pathologic condition demonstrating Candida-
induced oral signs such as fur, our results of fungal culture
suggest that this fungal culture is useful for predicting the risk
of developing oral candidiasis in the future.
In asthmatic patients treated with inhaled steroids, the
types of steroids and their doses positively correlated with the
649
amount of Candida. Although a previous study reported that
elderly patients with bronchial asthma are prone to candidi-
asis,10 we could not demonstrate any correlation between
patient age and amount of Candida. Patient background did
not significantly differ between asthmatic patients treated
with fluticasone and beclomethasone, and the inhalation dose
of fluticasone was approximately 50% that of beclometha-
sone. However, a significantly greater amount of Candida
was detected in those treated with fluticasone, probably be-
cause the potency and dosage form of fluticasone (powder)
differed from those of beclomethasone (aerosol) or because
the method of inhalation differed between the two groups. In
their review, O’Byrne and Pedersen11 compared the efficacy
and frequency of systemic adverse effects among be-
clomethasone, fluticasone, and budesonide, although they did
not review the frequency of topical adverse effects. They
found that the frequency of systemic adverse effects appeared
to depend on the dosage form and the type of inhalation
steroids. However, the efficacy of these steroids did not
always correspond with the frequency of systemic adverse
effects. Apart from a few exceptions,12 many other studies
have shown that the frequency of topical adverse effects did
not significantly differ between fluticasone and beclometha-
sone.13–15 We evaluated the dose of inhalation steroids inde-
pendently in asthmatic patients treated with fluticasone or
beclomethasone because the amount of Candida differed,
depending on the type of inhalant. The results showed that the
inhaled dose did not correlate with the amount of Candida in
asthmatic patients treated with beclomethasone. Although a
previous study16 has reported a similar finding, another
study17 reported that the amount of Candida increased with
the dose of the inhaled steroid. Furthermore, other studies4,5,16
have reported that the daily frequency of inhalation therapy is
important.
In our study, the inhalation dose positively correlated with
the amount of Candida in patients treated by inhalation of
fluticasone. These findings suggest that inhaled fluticasone
may dose dependently decrease local immunity, since most
steroids administered orally or by injection dose dependently
decrease systemic immunity and resistance to inflammation.
Therefore, higher doses of fluticasone should be administered
with care. On the other hand, the inhaled dose of beclometha-
sone did not correlate with amount of Candida. The pharma-
cologic characteristics of fluticasone differ from those of
beclomethasone. The lipid solubility of fluticasone was re-
ported to be approximately three times higher than that of
beclomethasone,18 and the affinity of fluticasone to glucocor-
ticoid was shown to be many times stronger than that of
beclomethasone.19 Fluticasone was also found to strongly
inhibit T cell migration and proliferation and cytokine ac-
tions.18 Furthermore, it has been reported that fluticasone and
beclomethasone differentially influence the viability of inter-
leukin 5–induced eosinophils20 and inhibit transcription fac-
tors.21 Therefore, the influence of steroids on intraoral immu-
nity may differ, depending on the magnitude of their efficacy.
On the other hand, fluticasone Diskhaler contains 25 mg of
650
lactose vehicle, which could serve as a substrate for candidal
growth. Studies that use fluticasone that lacks lactose (eg,
metered-dose inhaler-fluticasone) will shed light on the po-
tential effect of lactose on candidal growth. Furthermore,
various proteins in the saliva, such as IgA antibody and
histatin, have also been reported to be involved in develop-
ment of oral candidiasis, and salivary levels of IgA are
reportedly lower in patients with acquired immunodeficiency
syndrome.22,23 Thus, whether fluticasone influences salivary
levels of these proteins requires further investigation. We
initially considered that the frequency of intraoral adverse
effects might be decreased by improving the inhalation flow
rate, and thus we measured the inhalation flow rate in asth-
matic patients treated with fluticasone, using an inspiratory
flow meter. An inhalation rate of 60 L/minute or faster is
recommended. However, most asthmatic patients inhaled ste-
roids at an appropriate flow rate, and Candida was detected in
more than 25% of asthmatic patients in whom the inhalation
rate was within the above limits. These results indicate that
the inhalation flow rate does not seem to correlate with the
frequency of intraoral adverse effects.
In asthmatic patients in whom Candida was detected after
gargling with water, the gargle solution was changed from
water to a 1:50 diluted solution of amphotericin B. We
previously evaluated the efficacy of amphotericin B at vari-
ous concentrations and reported that a 1:50 diluted solution
has adequate antifungal activity.24 In addition, an in vitro
study demonstrated that Candida could be completely elim-
inated by amphotericin B in a dose-dependent manner and
that the viability of 90% Candida was inhibited by 10 to 50
times dilutions of amphotericin B in approximately 15 to 30
seconds (Bristol-Myers Squibb Co, New York, NY, unpub-
lished data, 1992). Treatment of 27 asthmatic patients taking
inhalation steroids by gargling with a 1:50 diluted solution
(2000 ␮g/mL) of amphotericin B resulted in a reduction of
the amount of Candida to a level below the minimum detec-
tion limit in 22 of these patients, resulting in clinical improve-
ment of oral symptoms. In the remaining five patients, the
amount of Candida was decreased to a level below the
minimum detection limit after a 7-day treatment course of
100 mg daily of fluconazole. There were no gargling-related
adverse effects associated with amphotericin B solution, and
the taste of the solution was tolerable. In Western countries,
oral candidiasis is considered to be a minor adverse effect of
inhalation steroids, because most cases of oral candidiasis are
successfully treated with a low-cost nystatin solution. Cur-
rently, however, nystatin is not available in Japan. When
asthmatic patients develop oral candidiasis, in a few cases,
the dose of inhalation steroids must be decreased or inhala-
tion therapy must be discontinued. Our data indicate that
gargling with amphotericin B solution may be useful for
treatment of oral candidiasis. Our results also show that the
MIC of amphotericin B was not significantly different be-
tween those who did not respond to amphotericin B, impli-
cating host-related rather than fungus-related factors, and
those who did. In this regard, uniform instructions were
ANNALS OF ALLERGY, ASTHMA, & IMMUNOLOGY
provided to all asthmatic patients with respect to the most
appropriate method of gargling. Although the effect of slight
individual differences of gargling method cannot be ex-
cluded, changes in oral immunity associated with inhalation
steroid therapy seem to be involved in the development of
oral candidiasis.
Long-term or lifelong inhalation steroid therapy is required
to control bronchial asthma. Fluticasone is useful for treating
bronchial asthma due to its higher efficacy; however, our
results demonstrate that the amount of Candida detected in
the retropharyngeal wall and the risk of developing oral
candidiasis increase with the dose of fluticasone. When asth-
matic patients developed oral candidiasis, the alternation of
delivery method should be considered. Gargling with ampho-
tericin B should also be considered in a subset of patients.
ACKNOWLEDGMENT
We thank Masumi Ishibashi for the excellent technical assis-
tance.
REFERENCES
1. National Asthma Education Program. Expert Panel Report 2:
Guidelines for the Diagnosis and Management of Asthma. Be-
thesda, MD: National Institute of Health, National Heart, Lung,
and Blood Institute; 1997. NIH publication 97– 4051.
2. Sorkness CA. Establishing a therapeutic index for the inhaled
corticosteroids, part 2: comparison of systemic activity and
safety among different inhaled corticosteroids. J Allergy Clin
Immunol. 1998;102(4 pt 2):S52–S64.
3. Bootsma GP, Koenderman L, Dekhuijzen PN, Festen J, Lam-
mers JW, van Herwaarden CL. Effects of fluticasone propionate
and beclomethasone dipropionate on parameters of inflamma-
tion in peripheral blood of patients with asthma. Allergy. 1998;
53:653– 661.
4. Lipworth BJ. Clinical pharmacology of corticosteroids in bron-
chial asthma. Pharm Ther. 1993;58:173–209.
5. Toogood JH. Complications of topical steroid therapy for
asthma. Am Rev Respir Dis. 1990;141(2 pt 2):S89 –S96.
6. Vogt FC. The incidence of oral candidiasis with use of inhaled
corticosteroids. Ann Allergy. 1979;43:205–210.
7. Williamson IJ, Matusiewicz SP, Brown PH, Greening AP,
Crompton GK. Frequency of voice problems and cough in
patients using pressurized aerosol inhaled steroid preparations.
Eur Respir J. 1995;8:590 –592.
8. Double-blind trial comparing two dosage schedules of be-
clomethasone dipropionate aerosol in the treatment of chronic
bronchial asthma. Preliminary report of the Brompton Hospital-
Medical Research Council Collaborative Trial. Lancet. 1974;2:
303–307.
9. Willey RF, Milne LJ, Crompton GK, Grant IW. Beclometha-
sone dipropionate aerosol and oropharyngeal candidiasis. Br J
Dis Chest. 1976;70:32–38.
10. Tanida T, Ueta E, Tobiume A, Hamada T, Rao F, Osaki T.
Influence of aging on candidal growth and adhesion regulatory
agents in saliva. J Oral Pathol Med. 2001;30:328 –335.
VOLUME 90, JUNE, 2003
11. O’Byrne PM, Pedersen S. Measuring efficacy and safety of
different inhaled corticosteroid preparations. J Allergy Clin Im-
munol. 1998;102:879 – 886.
12. Gustafsson P, Tsanakas J, Gold M, Primhak R, Redford M,
Gillies E. Comparison of the efficacy and safety of inhaled
fluticasone propionate 200 ␮g/day with inhaled beclomethasone
dipropionate 400 ␮g/day in mild and moderate asthma. Arch
Dis Child. 1993;69:206 –211.
13. Raphael GD, Lanier RQ, Baker J, Edwards L, Rickard K,
Lincourt WR. A comparison of multiple doses of fluticasone
propionate and beclomethasone dipropionate in subjects with
persistent asthma. J Allergy Clin Immunol. 1999;103:796 – 803.
14. Barnes NC, Marone G, Di Maria GU, Visser S, Utama I, Payne
SL. A comparison of fluticasone propionate, 1 mg daily, with
beclomethasone dipropionate, 2 mg daily, in the treatment of
severe asthma. Eur Respir J. 1993;6:877– 884.
15. Adams N, Bestall JM, Jones PW. Fluticasone versus be-
clomethasone or budesonide for chronic asthma. Cochrane Da-
tabase Syst Rev. 2002;1:CD002310.
16. Smith MJ, Hodson ME. High-dose beclomethasone inhaler in
the treatment of asthma. Lancet. 1983;319:265–268.
17. Toogood JH, Jennings B, Greenway RW, Chuang L. Candidi-
asis and dysphonia complicating beclomethasone treatment of
asthma. J Allergy Clin Immunol. 1980;65:145–153.
18. Johnson M. Development of fluticasone propionate and com-
parison with other inhaled corticosteroids. J Allergy Clin Im-
munol. 1998;101:S434 –S439.
19. Kelly HW. Establishing a therapeutic index for the inhaled
corticosteroids, part 1: pharmacokinetic/pharmacodynamic
comparison of the inhaled corticosteroids. J Allergy Clin Immu-
nol. 1998;102:S36 –S51.
20. Hagan JB, Kita H, Gleich GJ. Inhibition of interleukin-5 medi-
ated eosinophil viability by fluticasone 17-propionate: compar-
ison with other glucocorticoids. Clin Exp Allergy. 1998;28:
999 –1006.
21. Jaffuel D, Demoly P, Gougat C, Balaguer P, Mautino G, et al.
Transcriptional potencies of inhaled glucocorticosteroids. Am J
Respir Crit Care Med. 2000;162:57– 63.
22. Muller F, Froland SS, Hvatum M, Radl J, Brandtzaeg P. Both
IgA subclasses are reduced in parotid saliva from patients with
AIDS. Clin Exp Immunol. 1991;83:203–209.
23. Edgerton M, Koshlukova SE, Lo TE, Chrzan BG, Straubinger
RM, Raj PA. Candidacidal activity of salivary histatins. J Biol
Chem. 1998;273:20438 –20447.
24. Fukushima C, Shimoda T, Kawano T, Tomari S, Mitsuta K, et
al. Effects of amphotericin B gargles on oral colonization of
Candida albicans in asthmatic patients on steroid inhalation
therapy. Respiration. 2001;68:465– 470.
Requests for reprints should be addressed to:
Chizu Fukushima, MD
Second Department of Internal Medicine
Nagasaki University School of Medicine
1–7-1 Sakamoto
Nagasaki 852-8501, Japan
E-mail: chizu-f@ngs2.cncm.ne.jp
651