|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Pediatric Pulmonology (Dr. Geller), The Nemours Children's Clinic, Orlando, FL; Medical Affairs (Dr. Kaplowitz), Genentech, Inc, South San Francisco, CA; Division of Pediatric Pulmonology (Dr. Light), The University of California, San Diego, CA; and Division of Pediatric Pulmonology (Dr. Colin), Children's Hospital, Harvard Medical School, Boston, MA.
Correspondence to: David E. Geller, MD, The Nemours Children's Clinic, Division of Pediatric Pulmonology, 83 W Columbia St, Orlando, FL 32806
| Abstract |
|---|
|
|
|---|
Study design: All patients
5 years of age enrolled in
ESCF between December 1993 and May 1996 were eligible. Criteria for the
diagnosis of ABPA were defined by the ESCF guidelines. Prevalence rates
for ABPA were calculated, and potential risk factors for the diagnosis
of ABPA were analyzed, including sex, age, pulmonary function,
diagnosis of asthma, presence of wheeze, and positive respiratory
culture for Pseudomonas.
Results: There were 14,210 eligible patients enrolled in ESCF during this period, and ABPA was diagnosed in 281 patients (2%). Regional prevalence varied from 0.9% in the Southwest to 4.0% in the West. Increased prevalence rates occurred in female patients, the adolescent age group, and subjects with lower lung function, wheeze, asthma, and positive Pseudomonas cultures. Although most ABPA patients had evidence of airway obstruction, 10% had an FEV1 of > 100% of predicted. The rates of wheeze (17%) and asthma (30%) were lower than expected in the ABPA group.
Conclusions: This observational study found a reported prevalence rate of ABPA of 2% of CF patients in a large database. This rate was lower than the 5 to 15% rate reported in smaller studies, suggesting that ABPA is underdiagnosed in the CF population. There was wide regional variation in reported prevalence rates, which is unexplained at this time. The characteristics of the patients with ABPA and the epidemiologic risk factors for diagnosis of ABPA were described. Simplified diagnostic criteria were adapted for ESCF with the intent of increasing awareness of ABPA among the participants in this study.
Key Words: allergic bronchopulmonary aspergillosis cystic fibrosis epidemiology
| Introduction |
|---|
|
|
|---|
ABPA was first described in adult asthmatics in 1952,3 and now has been recognized in infants4 and children5 with asthma. Mearns et al6 reported ABPA in two children with cystic fibrosis (CF) in 1965, and since then it has been diagnosed with increasing frequency in CF. The reported prevalence of ABPA in CF is as high as 15%.7 8 9 10 11 12 13 14 There are discrepancies in reported prevalence, which may have many sources, including lack of accepted uniform diagnostic criteria, variability in the laboratory reagents and expertise of personnel, and underrecognition of the disease by clinicians.15 Similarities in the clinical presentation, laboratory findings, and chest radiograph appearance between CF and ABPA patients contribute to the diagnostic problem.16 17 Therefore, unless there is either a suspicion of ABPA or a formal screening program for ABPA, many cases may go unrecognized.
Diagnostic criteria vary from center to center. Of particular significance is the variability of the IgE level considered to be diagnostic: some studies have required an IgE level of > 1000 IU/mL,18 whereas others have made the diagnosis of ABPA with IgE levels of < 100 IU/mL.11 19 20 The diagnosis hinges on satisfying most of the major criteria defined by Rosenberg et al21 and modified by Laufer et al,9 with supporting evidence from three minor criteria (see column 1 in Table 1 ). Regional differences in ABPA prevalence have not been reported. Because seasonal differences in prevalence have been noted,20 22 23 regional differences in prevalence may occur if there is significantly different exposure to the mold, especially in humid, warm climates, or as a result of indoor exposure in cold climates.
|
Using enrollment data from ESCF, the objectives of this study were to (1) retrospectively estimate the prevalence of ABPA reported for patients with CF; (2) compare prevalence rates across geographic regions; (3) compare prevalence rates across patient subgroups defined by demographic and disease characteristics; and (4) describe the CF population in whom ABPA was diagnosed and compare it with all patients enrolled in ESCF with regards to sex, age, and disease severity.
| Materials and Methods |
|---|
|
|
|---|
5 years of age at enrollment were
eligible for this study. The participating sites in ESCF are divided
into nine geographic areas (Figure 1
) in the United States and Canada. Patient data are collected at
enrollment and at each clinic visit or hospitalization. The information
acquired includes clinical status, routine therapies, physical
examination, measures of pulmonary function, radiologic studies, the
occurrence of acute exacerbations of chronic pulmonary infection,
respiratory microbiology, and other factors predicting poor prognosis.
If ABPA was present in a patient within 6 months of enrollment into
ESCF (based on the recommended diagnostic criteria in Table 1 , column
2), then this was indicated on the enrollment forms.
|
For calculation of ABPA prevalence rates and prevalence rates of risk
factors, the denominators included all patients
5 years of age
enrolled in ESCF with valid data for sex, age, and medical conditions.
Prevalence rates were calculated by dividing the number of patients in
whom ABPA was diagnosed by the total number of CF patients at risk. The
importance of possible risk factors and strength of association was
estimated using multiple logistic regression. Age, FVC, and
FEV1 were entered into the model as categorical
variables rather than continuous variables because exploratory analysis
indicated that ABPA risk did not increase linearly in association with
these variables.
| Results |
|---|
|
|
|---|
5
years enrolled in ESCF. The frequency distributions for patients'
demographic and disease characteristics are shown in Table 2
. Slightly more male than female patients were in this analysis. The
mean (SD) age was 17.1 (9.6) years, with a median of 14.7 years.
Pulmonary function test data are summarized, for the patients who had
reported test values, in Table 2
. Pulmonary function measurements were
not reported for 6% of the patients. In those with data, the mean (SD)
FVC was 84%(25) of predicted value (median, 86%), and the mean (SD)
FEV1 was 71%(28) of predicted value (median,
73%). Wheezing on physical examination at enrollment was noted in 11%
of subjects; asthma was diagnosed in 19% in the 6 months preceding
enrollment; and a positive culture for Pseudomonas spp was reported in
62% in the year preceding enrollment.
|
|
|
Univariate logistic regression models presenting the odds ratio of being diagnosed with ABPA are shown in Table 5 . Age was associated with risk of ABPA, with statistically significant increases in prevalence occurring between the ages of 11 and 25 years. Patients who had an FEV1 < 70% of predicted had twice the risk compared with patients who had better lung function. The presence of wheeze, asthma, or Pseudomonas in sputum increased the relative risk of ABPA by > 1.5 times.
|
|
| Discussion |
|---|
|
|
|---|
The CF Foundation Registry reported an ABPA prevalence of 2.2% in CF patients > 5 years in 1995,27 almost exactly the same as the prevalence reported by ESCF. The CF Foundation Registry is also observational, but does not provide guidelines for the diagnosis of ABPA to the CF centers, instead relying on the clinicians to use their own centers' criteria. Thus, it appears that ABPA may be underdiagnosed in the CF population, and the 6-month window in ESCF was not an important factor in the low reported prevalence of ABPA. The patients observed in ESCF and the CF Foundation Registry data share the characteristic of being a mixed population of those specifically tested for ABPA and those who were not tested, which may lead to an underreporting bias for both studies. While the populations are similar and indeed overlap, the ESCF differs from the CF Foundation Registry in that it captures patients who are not a part of the Registry (those not followed in CF Foundationaccredited centers and patients from Canada). The present study also attempts to address questions related to ABPA that were not previously analyzed on a large scale, such as the association with wheezing, age, and regional effects.
Because of the substantial overlap of clinical and radiographic features of ABPA and CF, the diagnosis of ABPA in patients with CF is challenging. Airway hyperreactivity treated with bronchodilators and corticosteroids is common in CF. Recurrent pulmonary infiltrates and bronchiectasis also frequently occur in CF, and cannot be differentiated from ABPA by standard radiographic techniques.9 Prevalence of Aspergillus spp cultured from sputum in CF patients ranges from 1.3%10 to 57%,8 and is as high as 63% in patients considered for lung transplants.28 Because of the frequent colonization with A fumigatus in CF, immune and allergic reactivity to A fumigatus is frequently observed. A fumigatus precipitins may be found in between 10 and 51% of patients,8 9 10 13 14 26 29 30 31 radioallergosorbent tests for A fumigatus can be positive in 19 to 46%,32 33 and IgG specific for A fumigatus has been reported in up to 70% of CF patients.32 Finally, immediate skin test sensitivity to Aspergillus is found in 30 to 51% of CF patients.8 9 10 14 16 26 34 These immune parameters of A fumigatus sensitization are not static and can spontaneously diminish over time, which further complicates the ability to diagnose ABPA in CF patients.16 35
Most studies report a higher prevalence of one or more of these types of immune responses as the patients get older. This age effect probably represents longer exposure to the mold, progressive lung damage with impaired clearance, and a greater time period to develop sensitivity to A fumigatus. Our data support the increasing prevalence of ABPA with age, with the highest prevalence in patients aged 16 to 20 years, and a slight falloff beyond 20 years. A possible explanation of this observation is that ABPA most frequently affects the heavily colonized, sicker patients who may die earlier. The older survivors have relatively less severe disease, and subsequently less of a tendency to acquire ABPA. It is also possible that genetic factors may play a role in the acquisition of ABPA, and may explain why some of the patients with milder disease acquire ABPA while others with severe lung disease do not. ECSF may be able to test these hypotheses at a later stage, when more genetic information on patients is collected in the database, and when analyses of causes of death and possible associations with ABPA can be undertaken.
Our results agree with those of other studies showing that CF patients who acquire ABPA tend to be older and have more severe lung disease.34 But as Brueton et al36 reported, younger, healthier subjects are not risk-free. In our study, about 10% of ABPA subjects had an FEV1 > 100% of predicted, and 19% had an FVC > 100% of predicted. Therefore, good lung function is not absolutely protective against ABPA. Because the temporal relationship between diagnosis and treatment was not available in this study, it may also reflect that a segment of the ABPA population reported had been treated and recovered lung function. The latter explanation is supported by Light et al,11 who showed a deterioration of FEV1 after the diagnosis of ABPA, but a restoration of baseline function after treatment with prednisone. Only a longitudinal study of ABPA patients will be able to define its prognostic impact in CF.
Analysis of the ESCF data reveals a significant increase in frequency of ABPA in subjects who wheeze, have reported asthma, or are colonized with Pseudomonas. Even so, wheeze was not reported in 83% of ABPA patients and asthma was not reported in 70%. ESCF, unlike many other studies, did not consider asthma a prerequisite for an ABPA diagnosis. With a diagnosis of CF, it seems unreasonable to use asthma as a criterion, or to exclude ABPA in its absence. If asthma or wheezing were a prerequisite for the diagnosis of ABPA, most of these patients would not have satisfied the diagnostic criteria for ABPA and would have been excluded from consideration in the data analyses, resulting in an even smaller prevalence than that reported in this study.
An association of ABPA with Pseudomonas has been reported before,12 although others found no difference in Pseudomonas colonization rates in patients with or without ABPA.9 This study showed a 60% increase in the ABPA prevalence rate in individuals with Pseudomonas in their sputum. Such a relationship does not necessarily reflect interdependence of these conditions, but rather that older, sicker patients are increasingly affected by both. However, there is a theory that Pseudomonas may have a supportive role in the development of ABPA in CF. Pseudomonas aeruginosa inhibits hyphal growth of A fumigatus in vitro, leading to ultrastructural abnormalities that may result in the release of A fumigatus antigens, thus intensifying the immunologic response.37
The regional differences detected in this study have not been previously reported. While regional differences in ABPA prevalence may indeed exist, it is possible that there are other explanations for this observation. For example, the increased reported prevalence in the Western region may be due to the heightened awareness of ABPA in CF by the members of the Western CF Consortium and increased use of screening procedures by that group, resulting in a higher rate of diagnosis than in other regions. ESCF does not collect data to see how many of the participating centers use screening methods, such as periodic IgE levels, to detect ABPA. It is obvious, however, that systematic evaluation increases detection of ABPA. The provision of a uniform, simplified set of diagnostic criteria that is addressed at each patient encounter may be a reasonable method to remedy the problem of underdiagnosis.
Standardized diagnostic criteria and laboratory tests for ABPA are essential for clarifying the epidemiology of the disease.38 The authors recognize that the method offered in Table 1 , column 2using four of the common criteria utilized by many clinicians to diagnose ABPA without an exhaustive list of immune indicatorsis imperfect. One example of questions that can be raised regarding the diagnostic criteria used by ESCF is the nonuniform requirement for A fumigatus skin testing despite our awareness of the importance of this criterion. We recognize, however, that skin testing is not utilized by all clinicians who may prefer to use the other criteria. Another example is our arbitrary decision to use a cutoff of 1,000 IU for IgE for diagnosis, when others use a lower IgE level. This value appears to be reasonable for use in a large study that includes clinicians with varying numbers of patients and experience with the features of ABPA. Overall, the ABPA diagnostic criteria used in ESCF are more liberal than those in other studies, and were conceived to alert the clinician to the possible presence of the diagnosis, rather than establish diagnostic criteria conducive to a decision to treat. In this manner, it is hoped that this study, while observational by design, can improve the diagnostic awareness of ABPA. In the future, increased use of a more specific diagnostic test for ABPA, such as serologic tests toward the allergenic protein A fumigatus antigen I/a, may help CF caretakers reliably differentiate between ABPA and simple sensitization to A fumigatus.39
This study found underdiagnosis of ABPA and outlined epidemiologic features within ESCF. We believe that, when brought to the attention of the study investigators and CF clinicians, this report and the simplified set of diagnostic criteria will have an important impact on improving the diagnosis rate of ABPA. It is the intent of ESCF to repeat these analyses a number of years after study initiation to examine whether this expectation will materialize.
| Acknowledgements |
|---|
| Footnotes |
|---|
All sources of support for The Epidemiologic Study of Cystic Fibrosis in the form of grants, case report forms, and data analysis were provided by Genentech, Inc.
Received for publication November 3, 1998. Accepted for publication March 31, 1999.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. Kraemer, N. Delosea, P. Ballinari, S. Gallati, and R. Crameri Effect of Allergic Bronchopulmonary Aspergillosis on Lung Function in Children with Cystic Fibrosis Am. J. Respir. Crit. Care Med., December 1, 2006; 174(11): 1211 - 1220. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Y. Dolina, M. G. Benninghoff, A. K. Dewan, and R. Bascom PERSISTENT HEMOPTYSIS FROM CAVITATING PSEUDOMONAS PNEUMONIA IN A PATIENT WITH SEVERE BRONCHIECTASIS AS A COMPLICATION OF ALLERGIC BRONCHOPULMONARY ASPERGILLOSIS Chest Meeting Abstracts, October 1, 2006; 130(4): 288S - 288S. [Abstract] [PDF] |
||||
![]() |
D. Shoseyov, K. G. Brownlee, S. P. Conway, and E. Kerem Aspergillus bronchitis in cystic fibrosis. Chest, July 1, 2006; 130(1): 222 - 226. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. A. Belkin, N. R. Henig, L. G. Singer, C. Chaparro, R. C. Rubenstein, S. X. Xie, J. Y. Yee, R. M. Kotloff, D. A. Lipson, and G. R. Bunin Risk Factors for Death of Patients with Cystic Fibrosis Awaiting Lung Transplantation Am. J. Respir. Crit. Care Med., March 15, 2006; 173(6): 659 - 666. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. A. Flume, C. Strange, X. Ye, M. Ebeling, T. Hulsey, and L. L. Clark Pneumothorax in Cystic Fibrosis Chest, August 1, 2005; 128(2): 720 - 728. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. A. Flume, J. R. Yankaskas, M. Ebeling, T. Hulsey, and L. L. Clark Massive Hemoptysis in Cystic Fibrosis Chest, August 1, 2005; 128(2): 729 - 738. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. P. Conway, C. Etherington, D. G. Peckham, K. G. Brownlee, A. Whitehead, and H. Cunliffe Pharmacokinetics and safety of itraconazole in patients with cystic fibrosis J. Antimicrob. Chemother., May 1, 2004; 53(5): 841 - 847. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Saiman and J. Siegel Infection Control in Cystic Fibrosis Clin. Microbiol. Rev., January 1, 2004; 17(1): 57 - 71. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. T. Nguyen The Gloved Finger Sign Radiology, May 1, 2003; 227(2): 453 - 454. [Full Text] [PDF] |
||||
![]() |
H Witt Chronic pancreatitis and cystic fibrosis Gut, May 1, 2003; 52(90002): ii31 - 41. [Abstract] [Full Text] |
||||
![]() |
M. Helmi, R. B. Love, D. Welter, R. D. Cornwell, and K. C. Meyer Aspergillus Infection in Lung Transplant Recipients With Cystic Fibrosis: Risk Factors and Outcomes Comparison to Other Types of Transplant Recipients Chest, March 1, 2003; 123(3): 800 - 808. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. R. Nunley, A. A. Gal, J. D. Vega, C. Perlino, P. Smith, and E. C. Lawrence Saprophytic Fungal Infections and Complications Involving the Bronchial Anastomosis Following Human Lung Transplantation Chest, October 1, 2002; 122(4): 1185 - 1191. [Abstract] [Full Text] [PDF] |
||||
![]() |
J S Parmar, T Howell, J Kelly, and D Bilton Profound adrenal suppression secondary to treatment with low dose inhaled steroids and itraconazole in allergic bronchopulmonary aspergillosis in cystic fibrosis Thorax, August 1, 2002; 57(8): 749 - 750. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Skov, K.M. Main, I.B. Sillesen, J. Muller, C. Koch, and S. Lanng Iatrogenic adrenal insufficiency as a side-effect of combined treatment of itraconazole and budesonide Eur. Respir. J., July 1, 2002; 20(1): 127 - 133. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Maiz, M. Cuevas, S. Quirce, J. F. Canon, A. Pacheco, A. Sousa, and H. Escobar Serologic IgE Immune Responses Against Aspergillus fumigatus and Candida albicans in Patients With Cystic Fibrosis Chest, March 1, 2002; 121(3): 782 - 788. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Mastella, M. Rainisio, H.K. Harms, M.E. Hodson, C. Koch, J. Navarro, B. Strandvik, and S.G. McKenzie Allergic bronchopulmonary aspergillosis in cystic fibrosis Eur. Respir. J., May 1, 2001; 17(5): 1052 - 1053. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |