Rationale: Inhaled corticosteroids (ICS) are the mainstay of asthma treatment. Studies in chronic obstructive pulmonary disease reported increased rates of pneumonia with ICS. Concerns exist about an increased pneumonia risk in patients with asthma taking ICS.
Objectives: To evaluate the risks of pneumonia in patients with asthma taking ICS.
Methods: A retrospective analysis evaluated studies of the ICS budesonide in asthma. The primary data set were all double-blind, placebo-controlled trials lasting at least 3 months, involving budesonide (26 trials, n = 9,067 for budesonide; n = 5,926 for the comparator) sponsored by AstraZeneca. A secondary data set evaluated all double-blind trials lasting at least 3 months but without placebo control (60 trials, n = 33,496 for budesonide, n = 2,773 for fluticasone propionate). Cox proportional hazards regression modeling was used to estimate the relative effect of ICS on pneumonia adverse events (AEs) or serious adverse events (SAEs).
Measurements and Main Results: In the primary data set, the occurrence of pneumonia AEs was 0.5% (rate 10.0 events/1,000 patient-years [TPY]) for budesonide and 1.2% (19.3 per TPY) for placebo (hazard ratio, 0.52; 95% confidence interval, 0.36–0.76; P < 0.001); the occurrence of pneumonia SAEs was 0.15% (2.9 per TPY) for budesonide and 0.13% (2.1 per TPY) for placebo (hazard ratio, 1.29; 95% confidence interval, 0.53–3.12; P = 0.58). In the secondary data set, the percentage of patients reporting pneumonia AEs was 0.70% (12.7 per TPY), whereas the percentage of patients reporting pneumonia SAEs was 0.17% (3.1 per TPY). There was no increased risk with higher budesonide doses or any difference between budesonide and fluticasone.
Conclusions: There is no increased risk of pneumonia in patients with asthma, identified as an AE or SAE, in clinical trials using budesonide.
The use of inhaled corticosteroids has been associated with an increased risk of pneumonia in patients with chronic obstructive pulmonary disease. This risk has not been previously evaluated in patients with asthma using inhaled corticosteroids.
The results of this study do not support any increased risk of pneumonia in patients with asthma using inhaled corticosteroids.
ICS have also been extensively studied for the treatment of chronic obstructive pulmonary disease (COPD) (6–8). Several recent studies in patients with COPD reported increased rates of pneumonia among patients treated with the ICS fluticasone propionate (9, 10). In addition, a Canadian nested case-control study in a cohort of patients with COPD (11) suggested that treatment with any ICS may be associated with an increased risk of pneumonia. These conclusions are supported by a number of metaanalyses of pneumonia risk with the use of ICS in COPD (12–14). However, a similar risk was not identified in an evaluation of the use of the ICS budesonide in COPD (15). These findings have resulted in the inclusion of pneumonia and bronchitis as common adverse reactions in the product monograph for the combination of fluticasone/salmeterol.
Because ICS are so widely prescribed in asthma, there is a possibility of an increased risk of pneumonia in patients with asthma who are taking ICS. Therefore, the present retrospective analysis was undertaken to address this question in patients with asthma treated with the ICS budesonide.
To use the full safety database for budesonide, two data sets were defined: a primary data set, including only double-blind, placebo-controlled trials, used to calculate the hazard ratio (HR) for budesonide versus placebo; and a much larger secondary data set, including the ICS-randomized patients from all double-blind trials (regardless of placebo control or not), used for risk factor analyses, dose–response analysis, and a comparison between budesonide and fluticasone.
The studies used for the primary analysis were all double-blind, placebo-controlled clinical trials in patients with asthma with a duration of at least 3 months in patients aged 4 years and older, involving the use of budesonide (Pulmicort) or budesonide/formoterol (Symbicort) conducted by AstraZeneca, completed by December 2007 and with individual patient data present in the clinical study database at AstraZeneca, Lund, Sweden. All trials had at least one treatment arm involving exposure to budesonide (as Pulmicort and/or Symbicort) and one placebo arm, and all patients were allowed to use short-acting β2-agonists as rescue medication. A total of 26 trials with different formulations of budesonide fulfilled the selection criteria. Among these were 17 trials with budesonide in a dry powder inhaler (Turbuhaler) (3 with budesonide and formoterol in separate inhalers), 2 trials with budesonide in a metered dose inhaler (pMDI), 3 trials with budesonide nebulized inhalation suspension, and 3 trials with budesonide/formoterol in pMDI.
These trials had a total of 14,993 patients in the primary data set. Patients randomized to budesonide or budesonide plus formoterol were assigned as patients on active treatment (n = 9,067), whereas the patients randomized to placebo or formoterol alone were assigned as the comparator group (n = 5,926). The results are heavily influenced by the results from one large trial (the Steroid Treatment As Regular Therapy [START] trial) (16). This trial includes more than 7,000 patients treated with budesonide or placebo, on top of regular asthma treatment for 3 years. Because this was also the only trial that had duration longer than 1 year, the current primary analysis on pneumonia only considered events that occurred in the first 366 days of the START trial. As a sensitivity analysis, we also looked at the results for the full 3 years from the START study.
A secondary analysis was performed on the ICS-randomized patients from all double-blind, placebo and/or active-controlled trials with a duration of at least 3 months in patients with asthma aged 4 years and older, involving the use of budesonide or budesonide/formoterol.
A total of 60 trials with different formulations of budesonide fulfilled these criteria, which included 42,686 patients, of which 36,269 were ICS-treated patients, 33,496 exposed to budesonide and 2,773 to fluticasone.
An analysis of a potential dose–response effect of a higher dose (640 μg/d or more) and a lower dose (500 μg/d or less) of budesonide and/or budesonide plus formoterol on the incidence of pneumonia was performed on a subset of 14 trials in the second data set, in which these doses had been compared. In addition, a comparison between budesonide and fluticasone regarding the incidence of pneumonia was performed on a subset of five trials from the secondary data set.
Pneumonia events that occurred during randomized treatment have been included. Information has been taken from the clinical safety database at AstraZeneca Lund (ADVER 2000), the global AstraZeneca safety database (CLINTRACE/SAPPHIRE), and the respective clinical study reports (CSRs). The pneumonia events were adverse events (AE, both serious or nonserious) coded to the MedDRA (version 9) with preferred terms “Pneumonia,” “Bronchopneumonia,” “Lobar pneumonia,” “Lung infection,” “Pneumonia mycoplasmal,” “Mycoplasma infection,” “Pneumonia aspiration,” “Pneumonia bacterial,” “Pneumonia primary atypical,” “Pneumonia staphylococcal,” “Pneumonia viral,” or “Pneumonia pneumococcal.”
For each patient in both analyses, the person-time of participation in the trial was measured and cumulated to obtain person-years of exposure. The rates of events were expressed per 1,000 patient-years (TPY) computed for each treatment group, considering only the first pneumonia event, thus ignoring the negligible number of multiple events of the same type that occurred in a few patients. For the primary analysis, the risk of pneumonia as AE or serious AE (SAE) was compared between patients who were assigned to budesonide and those assigned to non-ICS treatments. The participants in each trial were followed from the date of enrollment to the date of withdrawal, the first pneumonia event, or study completion, whichever came first. Kaplan-Meier curves were generated to compare the time pattern of the pneumonia events as adverse events or serious adverse events between the budesonide and non-ICS treatment arms. Cox proportional hazards regression modeling, both adjusted and not adjusted by study, was used to estimate the relative effect of ICS on pneumonia adverse events or serious adverse events. HR and the nominal 95% confidence intervals (CIs) are presented. The overall relative risk (RR) using a Mantel-Haenszel approach stratified by study and adjusted for treatment exposure was calculated as a complementary analysis and expressed as the pooled Mantel-Haenszel RR and 95% CI. Additional details are given in the online supplement. As a sensitivity analysis, the Cox HR was calculated for the START trial alone and for all trials excluding START. The Cox HR was also calculated by sex and by age 4 to 11 years and more than 12 years.
The comparison of high-dose versus low-dose budesonide, and of budesonide and fluticasone, was also performed using the Cox regression adjusting for study. The risk factor analysis was performed by using linear tailed restricted cubic splines (17) in a Cox regression to model potential nonlinearities for age, body mass index (BMI), and baseline FEV1 expressed as percent of predicted normal. This is presented as HR compared with, respectively, an age of 35 years, BMI of 25 kg/m2, and an FEV1 of 100% of predicted normal. Sex as a risk factor was analyzed in a separate model.
A total of 14,993 patients were included in the primary analysis, involving 8,624 person-years of exposure (for details of the individual trials, see Table E1 in the online supplement). The overall occurrence of pneumonia AEs was 0.5% (rate 10.0 events/TPY) among the budesonide-treated patients compared with 1.2% (19.3 per TPY) among the placebo-treated patients (Table 1). The overall occurrence of pneumonia SAEs was lower, being 0.15% (2.9 per TPY) among the budesonide-treated patients compared with 0.13% (2.1 per TPY) among the placebo-treated patients (Table 1). The time to first pneumonia AE was statistically significantly in favor of budesonide-treated patients versus placebo-treated patients (HR, 0.51; 95% CI, 0.35–0.74; P < 0.001) (Figure 1A). The time to first pneumonia SAE did not show significant differences between budesonide-treated patients and placebo-treated patients (HR, 1.30; 95% CI, 0.53–3.14; P = 0.57) (Figure 1B).
No. (%; Rate per TTY) of Patients Reporting Pneumonia AEs
No. (%; Rate per TTY) of Patients Reporting Pneumonia SAEs
|Subgroup||Number on ICS||Number on Non-ICS||Exposure on ICS||Exposure on Non-ICS||ICS||Non-ICS||ICS||Non-ICS|
|All trials excluding START||5,437||2,335||1,528||630||15||(0.28, 9.8)||12||(0.51; 19.0)||6||(0.11; 3.9)||2||(0.09; 3.2)|
|START alone||3,630||3,591||3,275||3,190||33||(0.91; 10.1)||62||(1.7; 19.4)||8||(0.22; 2.4)||6||(0.17; 1.9)|
|Patients aged 4–11 yr||2,185||1,474||1,283||1,081||25||(1.1; 19.5)||38||(2.6; 35.2)||7||(0.32; 5.5)||6||(0.41; 5.6)|
|Patients aged 12–78 yr||6,882||4,452||3,520||2,739||23||(0.33; 6.5)||36||(0.81; 13.1)||7||(0.10; 2.0)||2||(0.04; 0.7)|
|All trials||9,067||5,926||4,804||3,820||48||(0.53; 10.0)||74||(1.2; 19.3)||14||(0.15; 2.9)||8||(0.13; 2.1)|
The incidence of pneumonia was higher among the children than among the adults in both treatment groups; among adults, there was also evidence of increasing rates of pneumonia with age in the group without exposure to ICS (Figure 2). The relative risk for pneumonia was reduced in the high and low age groups, as judged by the upper 95% confidence limit (values < 1) risk reductions on budesonide compared with placebo (Figure 2, lower panel).
The overall RRs using a Mantel-Haenszel approach were almost identical to the HRs obtained by the Cox regression model; for pneumonia AE the RR was 0.52 (95% CI, 0.36–0.76; P < 0.001) (Figure 3), and for pneumonia SAE the RR was 1.29 (95% CI, 0.53–3.12; P = 0.58) (Figure 4). As a sensitivity analysis, the Cox HR was calculated for the START trial alone and for all trials excluding START, and for the split by sex and by age 4 to 11 years and age 12 years and older (Table 2).
|Model||HR||95% CI||P Value||Interaction||HR||95% CI||P Value||Interaction|
|Not adjusted for study||0.51||0.35, 0.73||<0.001||1.32||0.55, 3.15||0.54|
|Adjusted for study||0.51||0.35, 0.74||<0.001||1.30||0.53, 3.14||0.57|
|Split by START/other||0.99||0.92|
|All other trials||0.51||0.24, 1.10||0.085||1.17||0.24, 5.8||0.84|
|START alone||0.51||0.34, 0.79||0.002||1.30||0.45, 3.75||0.63|
|Split by age||0.78||0.30|
|4–11 yr||0.54||0.33, 0.90||0.017||0.93||0.31, 2.78||0.90|
|≥12 yr||0.49||0.29, 0.82||0.007||2.55||0.53, 12.31||0.24|
|Split by sex||0.86||0.24|
|Female||0.49||0.30, 0.81||0.005||2.24||0.61, 8.31||0.23|
| Male||0.53||0.31, 0.90||0.02||0.76||0.22, 2.62||0.66|
The secondary data set included 33,496 patients (exposure time 18.9 TPY) exposed to budesonide, plus 2,273 patients (exposure time 1.2 TPY) exposed to fluticasone (Table E2). Data from the latter patients were only used in the budesonide and fluticasone comparison. In the secondary data set, the percentage of patients reporting pneumonia AE was higher in the budesonide-treated patients (0.70%; 12.4 per TPY) than for the budesonide-treated patients in the primary data set (0.5%; 10.0 per TPY) but lower than for the non-ICS–treated patients (1.2%; 19.3 per TPY) (Table E3A). The percentage of patients reporting pneumonia SAE was similar between the primary and secondary data sets (0.17%; 2.6 per TPY) (Table E3B). The risk difference between women and men was relatively small (HR, 1.07; 95% CI, 0.83–1.39), and there was no indication of a heterogeneous risk between different race groups (Table E6.) The risk for pneumonia AE was nonlinear with regard to age and baseline FEV1, but relatively flat for BMI (Figure 5).
There were 14 trials that had a direct comparison between budesonide assigned as low dose and budesonide assigned as high dose. They included 2,705 patients (exposure 1.4 TPY) on low-dose and 3,111 patients on high-dose therapy (exposure 1.5 TPY). The overall occurrence of pneumonia AE was 0.59% (9.9 per TPY) in the low-dose budesonide patients and 0.48% (11.4 per TPY) in the high-dose patients. The HR for high-dose versus low-dose budesonide was 0.88 (95% CI, 0.43–1.78) (Table E4), with almost identical results when adjusted for age and sex. In addition, there were five trials that had a budesonide arm as well as a fluticasone arm in which 4,051 patients (exposure 1.8 TPY) were exposed to budesonide and 2,773 (exposure 1.2 TPY) were exposed to fluticasone. The overall occurrence of pneumonia AEs with budesonide treatment was 0.81% (18.1 per TPY) compared with 0.76% (17.1 per TPY) with fluticasone treatment (HR, 1.01; 95% CI, 0.58–1.75) (Table E5).
As a sensitivity analysis, the full 3-year data from the START trial were calculated. The results from the primary data set limited to Year 1 regarding pneumonia AEs was 0.5% (10.0 events per TPY) for budesonide and 1.2% (19.3 per TPY) for placebo. When Years 2 and 3 from START were added, the results for pneumonia AE became 1.1% (9.6 per TPY) for budesonide and 2.1% (13.5 per TPY) for placebo, and for pneumonia SAE 0.23% (2.0 per TPY) for budesonide and 0.32% (2.0 per TPY) for placebo.
In three studies, the Asthma Control Questionnaire (ACQ5) was available at baseline. An analysis was performed to evaluate whether baseline asthma control in the 6,242 patients in these studies was associated with asthma risk. Uncontrolled asthma was identified by an ACQ5 score of 1.5 or greater and well-controlled asthma by an ACQ5 score of less than 1.5. The event rate for ACQ5 1.5 or greater versus less than 1.5 was 33/4,096 versus 13/2,146 (ratio, 1.346; 95% CI, 0.708–2.558; P = 0.3650).
This retrospective analysis pooled individual data analysis from the randomized trials, which have evaluated the effects of the ICS budesonide in patients with asthma, did not demonstrate any increase in the risks of pneumonia either identified as an AE or SAE, when compared with non-ICS treatment. In addition, there was no evidence of any increased risk of pneumonia AE with higher doses of budesonide, or any difference when budesonide was compared with the ICS fluticasone. All of the double-blind studies, sponsored by AstraZeneca, that have evaluated budesonide for the treatment of asthma have been included in this analysis. These included studies with a placebo control (the primary analysis) and those with a placebo and/or active control (the secondary analysis). It was important to include both analyses, as the secondary analysis was larger and also allowed an evaluation of the dose response, of another ICS, and of the possible risk factors for pneumonia as an AE in asthma clinical trials.
A significant reduction was seen, in the primary analysis, in the budesonide-treated patients for pneumonia AEs. A possible explanation for this difference may be incorrect allocation of respiratory events, which are known to be improved by ICS, as pneumonia AEs. These would include segmental atelectasis due to mucous impaction, more often seen in children with poorly controlled asthma; increased cough and mucous production; or mild asthma exacerbations. This interpretation is supported by the fact that the most frequent reporting of pneumonia AEs, as well as the most striking differences between budesonide and non-ICS treatment for pneumonia AEs, were in children aged 4 to 11 years (Table 1), in whom atelectasis is more frequently seen as a consequence of an asthma exacerbation, than in adult patients. Also, pneumonia SAEs are more likely to be fully investigated to establish whether pneumonia is present and whether an infective causation can be identified, and the risks of pneumonia as an SAE was not different between the budesonide- and non-ICS–treated patients. It is also possible, however, that improving asthma control with ICS, by reducing airway inflammation, reduces the risk factors for pneumonia, particularly segmental atelectasis due to mucous impaction (16). Only a prospective study with strictly defined criteria for pneumonia can resolve this question.
These findings in patients with asthma differ from studies that have evaluated pneumonia risk in patients with COPD treated with ICS (particularly fluticasone) (9, 10, 14), in which the risks were higher when higher inhaled doses were used (11). This increased risk was as high as 70% (11) above and beyond the already know risk of pneumonia in patients with COPD, particularly those with severe disease (17, 18, 20). For budesonide, alone or in combination with formoterol, a metaanalysis showed no difference between budesonide-treated and non-ICS–treated patients (15). The mechanisms for the increase in risk is not understood, but a factor of potential importance may be that approximately 25% of patients with COPD may be colonized with bacteria in the lower airways (19). A local impairment of host defense by some ICS may therefore be harmful in the presence of pathogenic microorganisms. Of interest, in the secondary data set, there was a higher pneumonia risk in those patients with the lowest FEV1 values (Figure 5). In general, studies of patients with COPD have a lower baseline FEV1 value when compared with the studies in patients with asthma. Thus, a low baseline FEV1 value may be an independent risk for pneumonia with treatment with ICS; however, this would need to be properly evaluated in a prospective trial.
The lack of an association between the use of budesonide in asthma and increased risks of pneumonia are supported by the fact that there was no dose response found when higher inhaled doses were compared with low inhaled doses, nor was an association seen in the smaller number of studies that used fluticasone as a comparator in asthma. Also, there was no increased pneumonia with ICS treatment in patients with uncontrolled asthma at the baseline visit.
A limitation of this study is that the analyses were performed retrospectively; however, a study to prospectively evaluate the risks of pneumonia in patients with asthma taking ICS would need to have a very large study population and be conducted over several years. These studies are unlikely to be done. Another limitation is that the term “pneumonia” likely has different connotations in different age groups and in different countries, so its use to define an AE is probably quite imprecise. However, the double-blind design of the trials and the within-country balance in randomization protect against any bias from differential miscoding of these events. When SAEs are evaluated in the same way, where the diagnosis is more likely to be accurate, there is still no significant effect of ICS treatment. Finally, the only other ICS for which data were available in the secondary data set was fluticasone, so these results cannot be generalized to other ICS used to treat asthma.
An increased risk of pneumonia AE was identified in children less than 12 years and patients more than 60 years of age. These are the age ranges in which the prevalence of pneumonia is highest in the general population. This may in part explain why there is an increased risk of pneumonia in patients with COPD treated with ICS, as these patient populations are in general older than patients with asthma entered into clinical trials. However, there was no evidence of an increased pneumonia risk in those older patients with asthma treated with inhaled budesonide (Figure 2). A previous study has identified an increased risk of serious pneumococcal disease in both patients with asthma and atopic patients (20, 21). However, pneumonia as an SAE was a rare event in the studies evaluated, occurring in a total of 22 patients of the 14,993 patients in the primary data set.
In conclusion, the results of this retrospective analysis do not support the concern that the use of ICS increases the risk of pneumonia in patients with asthma, identified as either an AE or SAE in clinical trials using budesonide.
AstraZeneca provided the data set and initial analysis.
|1.||Expert Panel Report 3 (EPR-3): guidelines for the diagnosis and management of asthma-summary report 2007. J Allergy Clin Immunol 2007;120:S94–S138.|
|2.||Bateman ED, Hurd SS, Barnes PJ, Bousquet, J, Drazen JM, Fitzgerald M, Gibson P, Ohta K, O'Byrne PM, Pedersen S, et al. Global strategy for asthma management and prevention: GINA executive summary. Eur Respir J 2008;31:143–178.|
|3.||Barnes PJ, Pedersen S, Busse WW. Efficacy and safety of inhaled corticosteroids. New developments. Am J Respir Crit Care Med 1998;157:S1–S53.|
|4.||Busse WW, Chervinsky P, Condemi J, Lumry WR, Petty Tl, Rennard S, Townley RG. Budesonide delivered by Turbuhaler is effective in a dose-dependent fashion when used in the treatment of adult patients with chronic asthma. J Allergy Clin Immunol 1998;101:457–463.|
|5.||Juniper EF, Kline PA, Vanzieleghem MA, Ramsdale EH, O'Byrne PM, Hargreave FE. Long-term effects of budesonide on airway responsiveness and clinical asthma severity in inhaled steroid-dependent asthmatics. Eur Respir J 1990;3:1122–1127.|
|6.||Pauwels RA, Lofdahl CG, Pride NB, Postma DS, Laitinen LA, Ohlsson SV. European Respiratory Society study on chronic obstructive pulmonary disease (EUROSCOP): hypothesis and design. Eur Respir J 1992;5:1254–1261.|
|7.||Burge PS, Calverley PM, Jones PW, Spencer S, Anderson JA, Maslen TK. Randomised, double blind, placebo controlled study of fluticasone propionate in patients with moderate to severe chronic obstructive pulmonary disease: the ISOLDE trial. BMJ 2000;320:1297–1303.|
|8.||Vestbo J, Sorensen T, Lange P, Brix A, Torre P, Viskum K. Long-term effect of inhaled budesonide in mild and moderate chronic obstructive pulmonary disease: a randomised controlled trial. Lancet 1999;353:1819–1823.|
|9.||Calverley PM, Anderson JA, Celli B, Ferguson GT, Jenkins C, Jones PW, Yates JC, Vestbo J. et al. Salmeterol and fluticasone propionate and survival in chronic obstructive pulmonary disease. N Engl J Med 2007;356:775–789.|
|10.||Wedzicha JA, Calverley PM, Seemungal TA, Hagan G, Ansari Z, Stockley RA. The prevention of chronic obstructive pulmonary disease exacerbations by salmeterol/fluticasone propionate or tiotropium bromide. Am J Respir Crit Care Med 2008;177:19–26.|
|11.||Ernst P, Gonzalez AV, Brassard P, Suissa S. Inhaled corticosteroid use in chronic obstructive pulmonary disease and the risk of hospitalization for pneumonia. Am J Respir Crit Care Med 2007;176:162–166.|
|12.||Nannini LJ, Cates CJ, Lasserson TJ, Poole P. Combined corticosteroid and long-acting beta-agonist in one inhaler versus inhaled steroids for chronic obstructive pulmonary disease. Cochrane Database Syst Rev 2007;CD006826.|
|13.||Drummond MB, Dasenbrook EC, Pitz MW, Murphy DJ, Fan E. Inhaled corticosteroids in patients with stable chronic obstructive pulmonary disease: a systematic review and meta-analysis. JAMA 2008;300:2407–2416.|
|14.||Singh S, Amin AV, Loke YK. Long-term use of inhaled corticosteroids and the risk of pneumonia in chronic obstructive pulmonary disease: a meta-analysis. Arch Intern Med 2009;169:219–229.|
|15.||Sin DD, Tashkin D, Zhang X, Radner F, Sjobring U, Thoren A, Calverley PM, Rennard S. Budesonide and the risk of pneumonia: a meta-analysis of individual patient data. Lancet 2009;374:712–719.|
|16.||Taylor DA, McGrath JL, O'Connor BJ, Barnes PJ. Allergen-induced early and late asthmatic responses are not affected by inhibition of endogenous nitric oxide. Am J Respir Crit Care Med 1998;158:99–106.|
|17.||Farr BM, Bartlett CL, Wadsworth J, Miller DL. Risk factors for community-acquired pneumonia diagnosed upon hospital admission. British Thoracic Society Pneumonia Study Group. Respir Med 2000;94:954–963.|
|18.||Almirall J, Bolibar I, Balanzo X, Gonzalez CA. Risk factors for community-acquired pneumonia in adults: a population-based case-control study. Eur Respir J 1999;13:349–355.|
|19.||Monso E, Ruiz J, Rosell A, Manterola J, Fiz J, Moreta J, Ausina V. Bacterial infection in chronic obstructive pulmonary disease. A study of stable and exacerbated outpatients using the protected specimen brush. Am J Respir Crit Care Med 1995;152:1316–1320.|
|20.||Juhn YJ, Kita H, Yawn BP, Boyce TG, Yoo KH, McGree ME, Weaver AL, Wollan P, Jacobson RM. Increased risk of serious pneumococcal disease in patients with asthma. J Allergy Clin Immunol 2008;122:719–723.|
|21.||Jung JA, Kita H, Yawn BP, Boyce TG, Yoo KH, McGree ME, Weaver AL, Wollan P, Jacobson RM, Juhn YJ. Increased risk of serious pneumococcal disease in patients with atopic conditions other than asthma. J Allergy Clin Immunol 2010;125:217–221.|