Eur J Clin Microbiol Infect Dis (2003) 22:457–462
DOI 10.1007/s10096-003-0970-y
ARTICLE
M. L. Uffredi · G. Mangiapan · J. Cadranel · G. Kac
Significance of Aspergillus fumigatus Isolation
from Respiratory Specimens of Nongranulocytopenic Patients
Published online: 25 July 2003

Springer-Verlag 2003
Abstract The aim of this study was to determine the
significance of isolation of Aspergillus fumigatus from
cultures of respiratory specimens in nongranulocytopenic
patients. The medical records of patients with respiratory
specimens positive for Aspergillus fumigatus who were
admitted to an adult pneumology ward were reviewed
during a 2-year period. A total of 80 respiratory speci-
mens from 76 patients yielded Aspergillus fumigatus.
Forty-eight patients were colonized with Aspergillus
fumigatus, whereas the 28 (37%) remaining patients had
pulmonary aspergillosis, manifest as aspergilloma (n=19
patients), chronic necrotizing pulmonary aspergillosis
(n=7 patients), and bronchial aspergillosis (n=2 patients).
The presence of typical hyphae in direct examination of
bronchoscopic specimens was more likely to be found in
infected than in colonized patients (P=0.04). No immu-
nological test was positive in colonized patients, whereas
anti-Aspergillus antibodies were detected in 55% of
infected patients (P<0.001). Pulmonary tuberculosis was
the most common underlying lung disease in patients with
aspergilloma, but it was not found in any patient with
chronic necrotizing pulmonary aspergillosis (P=0.006).
Anti-Aspergillus antibodies were more likely to be
detected in patients with aspergilloma (78%) than in
patients with chronic necrotizing pulmonary aspergillosis
(14%) (P=0.007). The analysis of predisposing factors, in
conjunction with immunological tests and examination of
bronchoscopic specimens, is helpful in distinguishing
between colonization and infection with Aspergillus
This work was presented in part at the 5th Congress of the European
Confederation of Medical Mycology, Dresden, Germany, 1999
M. L. Uffredi · G. Mangiapan · J. Cadranel
Service de Pneumologie, Hpital Tenon,
4 rue de la Chine, 75020 Paris, France
G. Kac ())
Hygine Hospitalire, Hpital Europen Georges Pompidou,
20 rue Leblanc, 75908 Paris Cedex 15, France
e-mail: guillaume.kac@hop.egp.ap-hop-paris.fr
Tel.: +33-1-56092973
Fax: +33-1-56093919
fumigatus, as well as for differentiating between asper-
gilloma and chronic necrotizing pulmonary aspergillosis.
Introduction
Aspergillus is a saprophytic filamentous fungus wide-
spread in the environment. Inhalation of Aspergillus
spores or conidia can give rise to various clinical
conditions, mainly due to the species Aspergillus fumi-
gatus, depending essentially on the host’s immunological
status [1]. In immunocompetent patients, Aspergillus
fumigatus can colonize a pre-existing cavity, causing
pulmonary aspergilloma (PAO), can initiate an exagger-
ated immune response, as seen in allergic bronchopul-
monary aspergillosis, or can form plugs in the bronchi,
causing obstructive bronchial aspergillosis. In immuno-
compromised patients, Aspergillus fumigatus can invade
the pulmonary parenchyma, producing invasive pulmo-
nary aspergillosis, an invasive, often lethal pulmonary
disease. More recently, locally invasive forms in patients
with mild immunosuppression called chronic necrotizing
pulmonary aspergillosis (CNPA) have been described [2,
3, 4, 5].
Laboratory diagnosis of aspergillosis is based mainly
on culture of Aspergillus fumigatus from respiratory
specimens, along with detection of anti-Aspergillus
antibodies and/or Aspergillus circulating antigens in
serum. However, isolation of Aspergillus fumigatus from
respiratory secretions does not discriminate between
colonization and infection. The significance of isolation
of Aspergillus from respiratory cultures has been studied
extensively in immunocompromised hosts who develop
invasive pulmonary aspergillosis [6, 7, 8, 9], but little is
known about the significance of isolation of Aspergillus
from respiratory specimens of immunocompetent or
mildly immunocompromised patients with other forms
of aspergillosis. We therefore conducted a 2-year study in
order to determine the significance of isolation of
Aspergillus fumigatus from respiratory specimens of
nongranulocytopenic patients.
458
Materials and Methods
Setting
The pneumology ward of Tenon Hospital in Paris is a 66-bed ward
that includes an 18-bed intensive care unit. Approximately 1,300
adult patients per year are admitted for a large variety of respiratory
illnesses.
Study Design
Respiratory specimens obtained from nongranulocytopenic patients
admitted to the pneumology ward between 1 January 1996 and 31
December 1997 were cultured for Aspergillus fumigatus. At least
one positive culture was required for inclusion of a patient in the
study. The medical records of these patients, which included
clinical, biological, and treatment data, were reviewed systemati-
cally. Each patient was then classified as colonized or infected
according to the definitions cited below.
Laboratory Methods
Respiratory tract specimens consisted of sputa and bronchoscopic
samples, i.e. bronchial aspirates, protected brush specimens, and
bronchoalveolar lavage fluids. The specimens were cultured on
Sabouraud agar in tubes with chloramphenicol (0.5 g/l) with and
without cycloheximide (0.5 g/l) at 27C and 37C and were
examined every day for 7 days. The pellet of the centrifuged
specimens was examined by means of Giemsa staining and
Toluidine blue staining preparations. Direct microscopic examina-
tion of the specimens was considered positive if typical septate
hyphae were observed. Aspergillus fumigatus was identified on the
basis of colony morphological features and microscopic structures
obtained with lactophenol cotton blue stain. The detection of anti-
Aspergillus antibodies was performed by immunoelectrophoresis
(Beckman Paragon; Beckman instruments, France); results were
considered positive if at least three arcs were identified according
to the manufacturer’s instructions. The detection of circulating
Aspergillus antigens was performed by an immunoenzymatic assay
(Platelia Aspergillus; Sanofi Diagnostics Pasteur, France) that can
detect as little as 1 ng/ml galactomannan. Lung sections were
stained with hematoxylin-eosin and Gomori-Grocott. The histolog-
ical lung examinations were considered positive if septate hyphae
split dichotomously at a 45 angle were observed.
Analysis of Medical Records
The following clinical data were recorded for all patients: age,
gender, alcohol use, clinical signs and symptoms, and chest
radiograph findings. We also analyzed the presence of underlying
lung diseases, i.e. chronic obstructive pulmonary disease, asthma,
bronchial carcinoma, bronchiectasis, pulmonary tuberculosis, non-
tuberculous mycobacteriosis [10], sarcoidosis, idiopathic pulmona-
ry fibrosis, and cystic fibrosis, as well as predisposing
immunosuppressive conditions, i.e. corticosteroid therapy, immu-
nosuppressive therapy, chemotherapy, radiation therapy, non-
bronchial cancer, leukemia, diabetes mellitus, cirrhosis, HIV
infection, and splenectomy. Finally, any treatment initiated follow-
ing a positive Aspergillus fumigatus culture was recorded, along
with the outcome.
Definitions
Infected patients were classified independently as having one
clinical form of pulmonary aspergillosis. PAO was diagnosed when
chest radiographs showed a spherical mass, usually surrounded by a
radiolucent crescent, inside a pre-existing pulmonary cavity
associated with at least one biological test positive for Aspergillus
fumigatus. When histopathological examination was performed,
findings showed the presence of a fungus ball associated with an
inflammatory reaction but without invasion of tissue [11]. CNPA
was diagnosed when chest radiographs showed cavitary chronic
pneumonia in mildly immunosuppressed patients associated with at
least one biological test positive for Aspergillus fumigatus and
improvement following appropriate treatment [4, 12, 13]. Invasive
pulmonary aspergillosis was diagnosed when chest radiographs
showed acute and extensive pneumonia or multinodular pneumonia
resistant to broad-spectrum antibiotics in immunocompromised
patients, and laboratory specimens yielded Aspergillus fumigatus
without pathogenic bacteria [2]. Bronchial aspergillosis was
defined as the presence of endobronchial plugs yielding Aspergillus
fumigatus [14]. Allergic bronchopulmonary aspergillosis was
defined according to the Rosenberg criteria [11].
Colonized patients had at least one culture positive for
Aspergillus fumigatus, but Aspergillus was not considered to cause
clinical symptoms.
Statistical Methods
Proportions were compared using the chi-square test with Yate’s
correction or Fisher’s exact test when appropriate. Mean values of
continuous data were compared with analysis of variance (Kruskal
Wallis test). Statistical tests were two-tailed, and a P value less than
0.05 was considered statistically significant. All calculations were
performed using the Epi-Info software package (Epi-Info 5.01 b;
Centers for Disease Control and Prevention, USA) and Statview 5.0
(SAS Institute, USA).
Results
During the 2-year study period, a total of 80 respiratory
specimens yielding Aspergillus fumigatus (43 sputa and
37 bronchoscopic specimens, including 24 bronchial
aspirates, 5 protected brush specimens, and 8 bronchoal-
veolar lavage fluids) were obtained from 76 patients.
After systematic review of medical records, 48 patients
were found colonized with Aspergillus fumigatus, where-
as the 28 (37%) remaining patients had pulmonary
aspergillosis: PAO in 19, CNPA in 7, and bronchial
aspergillosis in 2. No patient had invasive pulmonary
aspergillosis or allergic bronchopulmonary aspergillosis.
Comparison Between Colonized and Infected Patients
Of the 48 colonized patients (mean age, 54€16 years;
male, 58%) and the 28 infected patients (mean age, 55€12
years; male, 60%), 36 (75%) and 22 (79%) had at least
one chronic underlying lung disease, respectively (Ta-
ble 1). The predominant underlying disease was chronic
obstructive pulmonary disease in colonized patients
(27%) and tuberculosis in infected patients (43%). A
total of 42 (55%) patients, including 25 colonized patients
(52%) and 17 infected (61%) patients, had at least one
immunosuppressive factor and were considered mildly
immunocompromised patients. The remaining 34 (44%)
patients were immunocompetent. The predominant im-
munosuppressive condition in both colonized and infected
patients was systemic corticosteroid therapy, present in
23% and 25%, respectively. Of the 10 alcohol consumers,
 459
Table 1 Comparison of predisposing factors in patients colonized respectively). No significant differences between colo-
(n=48) and patients infected (n=28) with Aspergillus fumigatus nized and infected patients were found with regard to
Results No. (%) of patientsa P immunosuppressive conditions.
value Comparative mycological, serological, and histopath-
Colonized Infected
(n=48) (n=28) ological findings in colonized and infected patients are
shown in Table 2. A comparison of the nature of
Underlying lung diseases respiratory specimens obtained for diagnosis of aspergil-
COPD 13 (27) 9 (32) NS losis showed that the results obtained in colonized and
Asthma 11 (23) 3 (11) NS infected patients were similar except for the rate of
Bronchial carcinoma 8 (17) 3 (11) NS
Bronchiectasis 8 (17) 2 (7) NS bronchoscopic specimens that were positive by direct
Pulmonary tuberculosis 7 (15) 12 (43) 0.01 examination, which was fourfold higher in infected than
Nontuberculous mycobacteriosis 0 (0) 4 (14) 0.02 in colonized patients (P=0.04). The positive predictive
Sarcoidosis 1 (2) 2 (7) NS value of this test was 75%. Infected patients could be also
Idiopathic pulmonary fibrosis 1 (2) 0 (0) NS
Cystic fibrosis 1 (2) 0 (0) NS distinguished from colonized patients by detection of anti-
None 12 (25) 6 (21) NS Aspergillus antibodies; this test was positive in 55% of the
Immunosuppressive conditions infected patients and in none of the colonized patients
Corticosteroid therapy 11 (23) 7 (25) NS (P<0.001). The positive predictive value of this test was
Immunosuppressive therapy 0 (0) 0 (0) NP thus 100%, whereas its negative predictive value was
Chemotherapy 8 (17) 3 (11) NS 76%.
Radiation therapy 2 (4) 5 (18) NS
Nonbronchial cancer 4 (8) 4 (14) NS
Leukemia 1 (2) 2 (7) NS
Diabetes mellitus 1 (2) 1 (3.5) NS Comparison Between Patients with Pulmonary
Cirrhosis 1 (2) 0 (0) NS Aspergilloma and Patients with Chronic Necrotizing
HIV infection 5 (10) 1 (3.5) NS Pulmonary Aspergilloma
Neutropenia 0 (0) 2 (7) NS
Splenectomy 1 (2) 0 (0) NS
None 23 (48) 11 (39) NS Of the 28 infected patients, 19 (68%) patients met the
criteria for diagnosis of PAO (mean age, 57€13 years)
COPD, chronic obstructive pulmonary disease; NS, not significant; and 7 (25%) the diagnosis for CNPA (mean age, 59€15
NP, not performed
a
Patients may be included in more than one category (thus allowing years). A comparison of predisposing factors between
for a total percentage >100%) patients with PAO and those with CNPA showed that
pulmonary tuberculosis was common in patients with
PAO (63%) but was not found in any patient with CNPA
only two consumed alcohol chronically, and both were (P=0.006). Of interest, two (28%) patients with CNPA
infected patients. A comparison of underlying lung had no reported predisposing pulmonary condition, while,
diseases between colonized and infected patients indicat- in contrast, all PAO patients had predisposing pulmonary
ed that tuberculosis and nontuberculous mycobacteria conditions. At least one immunosuppressive factor was
were more predominant in infected than in colonized identified in all but one CNPA patient.
patients (43% vs. 15%, P=0.01; and 14% vs. 0%, P=0.02,

Table 2 Comparison of No. with positive test or specimena/total no. with tests P
biological findings in patients or specimens (%) value
colonized (n=48) and patients
infected (n=28) with Colonized patients (n=48) Infected patients (n=28)
Aspergillus fumigatus
Respiratory specimens
Sputum
Microscopic examination 8/34 (24) 4/21 (19) NS
Culture 28/34 (82) 16/21 (76) NS
Bronchoscopic specimens
Microscopic examination 2/28 (7) 6/18 (33) 0.04
Culture 21/28 (75) 11/18 (61) NS
Immunological tests
Anti-Aspergillus antibodies 0/38 15/27 (55) <0.001
Aspergillus antigenemia 0/38 1/26 (4) NS
Histopathological examination
Presence of typical septate hyphae NP 8/9 (89) NP
NS, not significant; NP, not performed
a
At least one specimen or test positive
460
Table 3 Comparison of biolog-
ical findings in 19 patients with
pulmonary aspergilloma and 7
patients with chronic necrotiz-
ing pulmonary aspergillosis
(CNPA) Respiratory specimens
Sputum
Microscopic examination
Culture
Bronchoscopic specimens
Microscopic examination
Culture
Immunological tests
Anti-Aspergillus antibodies
Aspergillus antigenemia
Histopathological examination
Presence of typical septate hyphae
NP, not performed
a
At least one specimen or test positive
In patients with PAO, hemoptysis was the most
common clinical symptom (63%), followed by dyspnea
(21%), fever (16%), cough (16%), and sputum production
(16%). In patients with CNPA, the primary clinical
symptoms were dyspnea and fever (57% each), followed
by hemoptysis (28%) and cough, sputum production,
chest pain, and weight loss (14% each). Six of the seven
(85%) CNPA patients had more than one symptom upon
admission, and three (43%) presented with three or more
symptoms. Outcome was evaluable in 16 of the 19 (84%)
PAO patients: two died (1 from massive hemoptysis, 1
from lymphoma) and 14 had a favorable outcome
following surgical treatment (n=8) or itraconazole treat-
ment alone (n=6). Outcome was evaluable in only three
CNPA patients because the remaining four patients died
of other causes unrelated to aspergillosis (bacterial sepsis,
n=2; neurological disease, n=1; bronchial carcinoma,
n=1). Outcome was favorable in all three patients
following adequate antifungal therapy with itraconazole
and amphotericin B.
The comparative mycological, serological, and histo-
pathological findings in PAO versus CNPA patients are
shown in Table 3. Of the seven CNPA patients, only one
(14%) was positive for anti-Aspergillus antibodies in
serum (3 arcs), and another tested positive for antigen-
emia. Three CNPA patients had two arcs, which was
below the defined cutoff value. This is in marked contrast
with the results obtained in PAO patients, 78% of whom
tested positive for anti-Aspergillus antibodies (3 arcs),
always in conjunction with a negative test result for
antigenemia (P=0.007). One additional patient had two
arcs. The rate of positivity for culture of bronchoscopic
specimens was significantly higher in CNPA patients
(86%) than in PAO patients (22%) (P=0.04).
No. with positive test or specimena/total no. with tests P
or specimens (%) value
Pulmonary aspergilloma (n=19) CNPA (n=7)
0/13 2/6 (33) NS
9/13 (69) 4/6 (66) NS
2/9 (22) 4/7 (57) NS
2/9 (22) 6/7 (86) 0.04
14/18 (78) 1/7 (14) 0.007
0/16 1/6 (16) NS
7/8 (88) 1/1 (100) NS
Discussion
The analysis of underlying lung diseases associated with
the results of direct examination of bronchoscopic
specimens and immunological tests is helpful in differ-
entiating colonization from infection due to Aspergillus
fumigatus in immunocompetent and mildly immunosup-
pressed patients. Of approximately 2,600 patients admit-
ted to the pneumology ward during the 2-year study
period, 76 (3%) nongranulocytopenic patients had respi-
ratory specimens yielding Aspergillus fumigatus. This
result is in agreement with several studies that reported
isolation rates of Aspergillus in 2–7% of patients with
various lung diseases [8, 15, 16].
The 76 patients in our study were classified as
colonized (n=48) or infected (n=28) (37%) patients. In a
recent prospective multicenter study, Perfect et al. [17]
had found the rate of infected patients to be 20% among
1,209 cases of suspected aspergillosis. In both studies, the
different clinical entities of aspergillosis were defined
according to clinical and radiological criteria, even
though definitions based on histopathological findings
used by some authors may be specific [1, 12, 18, 19].
Clinical diagnostic criteria are needed from a practical
viewpoint because histological diagnosis requires biopsy
specimens, which cannot always be obtained from
immunosuppressed patients or those with respiratory
insufficiency. In studies based on histopathological crite-
ria, the rate of infected patients is much lower; for
example, Treger et al. [15] reported a 10% rate of
demonstrable infection among 89 patients with at least
one positive culture of Aspergillus in respiratory speci-
mens.
In the present work, chronic obstructive pulmonary
disease was the most common underlying lung disease in
colonized patients, which confirms earlier findings by Yu
et al. [8]. Because patients with chronic obstructive
pulmonary disease are often treated with corticosteroids
(15% in our study), they have an increased risk for

developing a more aggressive pattern of infection,
secondary to the immunosuppressive activity of this drug.
Similarly, approximately one-half of the colonized pa-
tients studied had at least one cause of immunosuppres-
sion, which predisposed them to infection. No significant
differences were found between colonized and infected
patients except for those with tuberculosis and nontuber-
culous mycobacteriosis; these high-risk patients should
therefore be monitored regularly for several months to
detect the occurrence of aspergillosis [20].
Two types of respiratory tract specimens (noninvasive,
i.e. sputa, and invasive, i.e. bronchoscopic samples) could
be obtained in patients suspected to have aspergillosis. In
our study, neither direct examination nor culture of
sputum samples was useful in distinguishing colonized
from infected patients, whereas detection of septate
hyphae in bronchoscopic specimens was significantly
higher in infected compared to colonized patients. The
positive predictive value of direct examination performed
in bronchoscopic specimens was thus 75%. In view of the
greater diagnostic value of bronchoscopic aspirates,
bronchoscopy should be strongly considered in mildly
immunosuppressed patients in whom aspergillosis is
suspected.
In colonized patients, the rate of positive microscopic
examination of respiratory specimens was more than
threefold higher in sputa (24%) than in bronchoscopic
specimens (7%), but the difference was not significant
(P=0.10). When microscopic examination of the speci-
mens was negative, contamination of samples during
collection or processing could not be excluded. Addition-
ally, most colonized patients (81%) had only one
specimen positive for Aspergillus, probably as a conse-
quence of the natural elimination of inhaled spores (data
not shown). The fact that antibodies and/or antigens were
not detected in all colonized patients suggests that
examination of respiratory specimens should always be
performed in conjunction with immunological tests in
patients with chronic underlying pulmonary diseases to
distinguish Aspergillus fumigatus infection from coloni-
zation.
In the population studied, PAO was the most common
form of aspergillosis identified (68%). This proportion is
far higher than that usually reported [17], a result that can
be related to the type of patients admitted to the ward.
Indeed, many patients are admitted to our pneumology
ward for hemoptysis because bronchial artery emboliza-
tion can be performed. Since hemoptysis is the main,
often sole symptom of PAO, we naturally have an over-
representation of this pathology among the different
forms of aspergillosis. Our study confirms the high rate of
positivity of anti-Aspergillus antibodies (78%) in PAO
patients and thus confirms the usefulness of antibody
detection when PAO is suspected on the basis of chest
radiograph findings.
A complete distinction between PAO and CNPA may
not be possible clinically or even histologically, mostly
because the findings of histological examination, when
performed in CNPA patients (1 case in our study), could
461
be similar to those in PAO patients [5, 20]. We therefore
determined how these two clinical forms of aspergillosis
could be differentiated.
The present study confirmed that pulmonary symp-
toms such as fever and dyspnea are frequent in CNPA
(57% each), whereas the main symptom in PAO remains
hemoptysis (63%). However, in our series, hemoptysis in
CNPA patients was higher than that usually reported [4,
5]. In addition, a history of tuberculosis in a patient
suspected to have aspergillosis is highly suggestive of
PAO. As reported by Saraceno et al. [4], roughly one-
quarter of the CNPA patients had no predisposing
pulmonary underlying disease. The rate of positive
culture of bronchoscopic specimens was fourfold higher
in CNPA than in PAO, a statistically significant differ-
ence. This could possibly be explained by the fact that
CNPA is considered a more invasive disease than PAO,
associated with local luminal invasion by fungi leading to
easier detection in culture. While clinical and radiological
characteristics of PAO and CNPA have been compared
extensively, very few studies have analyzed serological
data in CNPA patients. Of the CNPA patients reported in
our study, only one (14%) was positive for anti-Asper-
gillus antibodies, but three of the seven had two arcs (just
below the cutoff value for positivity), and another patient
was positive for antigenemia. This is in marked contrast
with findings in PAO patients, in whom positive detection
of antibodies is usually found in conjunction with
negative antigenemia. Finally, the presence of several
pulmonary symptoms and suggestive radiological signs in
conjunction with bronchoscopic samples culture-positive
for Aspergillus fumigatus and negative immunological
tests in a mildly immunosuppressed patient is highly
suggestive of CNPA. In contrast, a history of tuberculosis,
the presence of hemoptysis, and positive detection of anti-
Aspergillus antibodies is suggestive of PAO.
Although the patients in this study have been carefully
characterized, the retrospective design imposes signifi-
cant limitations. The principal limitation is that patients
were not studied systematically. A prospective study in
which uniform diagnostic procedures (examination of
sputum samples, bronchial aspirates, bronchoalveolar
lavage fluids, and detection of serum antibodies and
antigens) and analyses (microscopic examination, culture)
are employed in this population seems therefore warrant-
ed.
The results of this study suggest that the analysis
of predisposing factors, in conjunction with immuno-
logical tests and examination of bronchoscopic speci-
mens, is helpful in distinguishing between colonization
and infection due to Aspergillus fumigatus as well as for
differentiating between PAO and CNPA in nongranulo-
cytopenic patients.
Acknowledgments The authors would like to thank Prof G. Meyer
for his helpful comments.
462
References
1. Sharma OP, Chwogule R (1998) Many faces of pulmonary
aspergillosis. Eur Respir J 12:705–715
2. Denning DW (1998) Invasive aspergillosis. Clin Infect Dis
26:781–805
3. Latg JP (1999) Aspergillus fumigatus and aspergillosis. Clin
Microbiol Rev 12:310–350
4. Saraceno JL, Phelps DT, Ferro TJ, Futerfas R, Schwartz DB
(1997) Chronic necrotizing pulmonary aspergillosis: approach
to management. Chest 112:541–548
5. Caras WE, Pluss JL (1996) Chronic necrotizing pulmonary
aspergillosis: pathologic outcome after itraconazole therapy.
Mayo Clin Proc 71:25–30
6. Horvarth JA, Dummer S (1996) The use of respiratory-tract
cultures in the diagnosis of invasive pulmonary aspergillosis.
Am J Med 100:171–178
7. Levy H, Horak DA, Tegtmeier BR, Yokota SB, Forman SJ
(1992) The value of bronchoalveolar lavage and bronchial
washings in the diagnosis of invasive pulmonary aspergillosis.
Respir Med 86:243–248
8. Yu VL, Muder RR, Poorsattar A (1986) Significance of
isolation of Aspergillus from the respiratory tract in diagnosis
of invasive pulmonary aspergillosis. Am J Med 81:249–254
9. Nalesnik MA, Myerowitz RL, Jenkins R, Lenkey J, Herbert D
(1980) Significance of Aspergillus species isolated from
respiratory secretions in the diagnosis of invasive pulmonary
aspergillosis. J Clin Microbiol 11:370–376
10. American Thoracic Society (1997) Diagnosis and treatment of
disease caused by nontuberculous mycobacteria. Am J Resp
Crit Care Med 156 [Suppl]:1–25
11. Broderick LS, Conces DJ Jr, Tarver RD, Bergmann CA, Bisesi
MA (1996) Pulmonary aspergillosis: a spectrum of disease. Crit
Rev Diagn Imaging 37:491–531
12. Gefter WB, Weingrad TR, Epstein DM, Ochs RH, Miller WT
(1981) “Semi-invasive” pulmonary aspergillosis: a new look at
the spectrum of Aspergillus infections of the lung. Radiology
140:313–321
13. Binder RE, Faling LJ, Pugatch RD, Mahasaen C, Snider GL
(1982) Chronic necrotizing pulmonary aspergillosis: a discrete
clinical entity. Medicine 61:109–124
14. Denning DW (1995) Commentary: unusual manifestations of
aspergillosis. Thorax 50:812–813
15. Treger TR, Visscher DW, Bartlett MS, Smith JW (1985)
Diagnosis of pulmonary infection caused by Aspergillus:
usefulness of respiratory cultures. J Infect Dis 152:572–576
16. Bodey PG, Vartivarian S (1989) Aspergillosis. Eur J Clin
Microbiol Infect Dis 8:413–437
17. Perfect JR, Cox YL, Lee JY, Kauffman CA, Repentigny L de,
Chapman SW, Morrison VA, Pappas P, Hiemenz JW, Stevens
DA (2001) The impact of culture isolation of Aspergillus
species: a hospital-based survey of aspergillosis. Clin Infect Dis
33:1824–1833
18. Thompson BH, Stanford W, Galvin JR, Kurihara Y (1995)
Varied radiologic appearances of pulmonary aspergillosis.
Radiographics 15:1273–1284
19. Miller WT (1996) Aspergillosis: a disease with many faces.
Semin Roentgenol 31:52–66
20. Yousem SA (1997) The histological spectrum of chronic
necrotizing forms of pulmonary aspergillosis. Hum Pathol
28:650–656