Abstract
BACKGROUND: Limited information is available regarding the impact of prior use of inhaled corticosteroids (ICS) in patients subsequently developing community-acquired pneumonia (CAP). We assessed the effects of prior ICS use on severity of illness and microbiology in CAP hospitalized patients.
METHODS: A retrospective cohort study of subjects with CAP (by the International Classification of Diseases, 9th Revision, Clinical Modification) was conducted over a 4-year period at 2 tertiary teaching hospitals. Subjects were considered to be ICS users if they received ICS prior to admission. Primary outcomes were severity of illness and microbiology at admission.
RESULTS: Data were abstracted on 664 subjects: 89 prior ICS users (13.4%) and 575 non-users (86.6%). Prior ICS users had higher severity of illness at admission: mean ± SD Pneumonia Severity Index 100.8 ± 31.4 vs 68.8 ± 33.4, P = .001, and CURB-65 (confusion, urea nitrogen, respiratory rate, blood pressure, ≥ 65 years of age) score 1.56 ± 1.02 vs 1.19 ± 1.02, P = .002. Prior ICS use was independently associated with antimicrobial-resistant pathogens: 11.2% vs 5.9%, odds ratio 2.6, 95% CI 1.1–6.1, P = .04.
CONCLUSIONS: Prior ICS use was associated with higher severity of illness at admission and higher incidence of antimicrobial-resistant pathogens in CAP hospitalized patients.
Introduction
Inhaled corticosteroids (ICS) are anti-inflammatory agents widely used in patients with obstructive airways diseases. Their established efficacy and adequate safety profile have placed these medications in the treatment recommendations for the most prevalent chronic respiratory diseases, such as asthma and COPD.1,2
In COPD, the fourth leading cause of death in the United States,3 ICS reduce the frequency of exacerbations and improve quality of life.4,5 Paradoxically, several large trials have demonstrated that ICS are associated with an increased incidence of community-acquired pneumonia (CAP) in these patients.6–12 There is controversy regarding the association between poor clinical outcomes in CAP patients and prior use of ICS.9,13–15 Although recent and numerous randomized trials and meta-analyses have evaluated the impact of ICS on pneumonia-related or overall mortality,6–15 few data are available regarding the association of prior ICS with the severity of illness at admission or its impact on antimicrobial resistance. Although chronic systemic corticosteroids use has been associated with higher risk of opportunistic infections16–18 and potentially highly drug-resistant pathogens,19,20 these associations are not clear in patients treated with ICS. Therefore, our aim was to investigate the association of prior out-patient ICS use with severity of illness and antimicrobial resistance in hospitalized patients with CAP at the time of admission.
QUICK LOOK
Current knowledge
Inhaled corticosteroids are anti-inflammatories that are widely used by patients with COPD. Limited information is available on the impact of prior use of inhaled corticosteroids in patients with community-acquired pneumonia.
What this paper contributes to our knowledge
Prior use of inhaled corticosteroids was associated with higher severity of illness and incidence of antibiotic-resistant pathogens in patients admitted with community-acquired pneumonia.
Methods
This was a cross-sectional study of patients hospitalized with CAP at 2 academic teaching tertiary care hospitals in San Antonio, Texas. The institutional review board of the University Health Science Center at San Antonio approved the research protocol, with exempt status.
Inclusion and Exclusion Criteria
We identified all patients admitted to the study hospitals between January 1, 1999, and December 31, 2002, who had a primary discharge diagnosis of pneumonia (International Classification of Diseases, 9th Revision, Clinical Modification codes 480.0–483.99 or 485–487.0) or a secondary discharge diagnosis of pneumonia with a primary diagnosis of respiratory failure (518.81) or sepsis (038.xx). Subjects were included if they were > 18 years of age, had an admission diagnosis of CAP, and had a radiographically confirmed infiltrate or other finding consistent with pneumonia on chest x-ray or CT obtained within 24 hours of admission.
Exclusion criteria included having been discharged from an acute care facility within 14 days of admission, transfer after being admitted to another acute care hospital, being on comfort measures only on this admission, and using systemic corticosteroids. If a subject was admitted more than once during the study period, only the first hospitalization was abstracted
Data Abstraction
Chart review data included: demographics, comorbidities, physical examination findings, laboratory data, chest radiographic reports, guideline-concordant empirical antibiotic therapy, and prior antibiotic use. Comorbidities were identified from either the admission or discharge notes or out-patient problem lists. Antimicrobial therapy was considered guideline-concordant if it agreed with the American Thoracic Society guidelines.21 Prior antibiotic use was defined as having taken any kind of antibiotic ≤ 15 days before admission. Information on all ICS out-patient medications that were either reported as currently being taken by the subject at presentation or listed in the electronic medical record was recorded. ICS therapy was defined as receiving inhaled beclomethasone dipropionate, triamcinolone, budesonide, or fluticasone propionate at the time of admission.
Diagnostic Criteria
Microbiologic data results were reviewed, and a microbiologic cause was assigned independently by 2 of the investigators (EMM and MIR). The cause of pneumonia was considered if one of the following conditions were met: blood cultures were positive for bacterial or fungal pathogens (in the absence of an extra-pulmonary source of infection); pleural fluid cultures yielded a bacterial pathogen; endotracheal suctioning samples showed moderate or heavy growth of bacterial pathogens; there was substantial quantitative culture growth from bronchoscopic samples (protected specimen brush culture had ≥ 103 colony-forming units/mL, or bronchoalveolar lavage fluid had ≥ 104 colony-forming units/mL). When 2 or more microbiologic causes were present, the subject was considered to have a polymicrobial infection. A subject was considered to have CAP of unknown cause if microbiologic studies were not performed or inconclusive.
Drug-resistant pathogens were defined as drug-resistant Streptococcus pneumoniae, methicillin-resistant Staphylococcus aureus, Acinetobacter baumannii, extended-spectrum beta lactamases, and multidrug-resistant Pseudomonas aeruginosa (resistance to ≥ 3 antimicrobial categories).22
Clinical Outcomes
The primary outcomes were severity of illness at presentation, defined by the Pneumonia Severity Index23 and CURB-65 (confusion, urea nitrogen, respiratory rate, blood pressure, ≥ 65 years of age),24 and presence of resistant or opportunistic pathogens.
Statistical Analyses
For the statistical analyses, subjects were stratified according to prior ICS use. Categorical variables were analyzed using the chi-square test, and continuous variables were analyzed using the Student t test. Multivariable analysis was performed using a logistic regression model, with drug-resistant pathogens as the dependent variable, and those with P < .10 as independent variables (age, sex, COPD, congestive heart failure, and history of malignancy). Logistic regression results are reported as odds ratio and 95% CI. Subgroup analyses were performed for subjects with and without COPD according to the primary outcomes. COPD represents one the largest groups of ICS users among elderly subjects, and it is the cohort where substantial interest related to complications with ICS has emerged.25 Two-sided P values < .05 were considered statistically significant. All analyses were performed using statistics software (SPSS 19.0, SPSS, Chicago, Illinois).
Results
We identified 664 patients who met the inclusion criteria, 89 (13.4%) of whom had received ICS prior to admission, and 575 (86.6%) had not.
Subject Characteristics
Subject characteristics are shown in Table 1. Prior ICS users were older, more likely to be male, and had more preexisting comorbidities, such as congestive heart failure, history of malignancy, or COPD. No significant differences were found between the groups regarding physical exam, laboratory and radiological data, or processes of care.
Subject Demographic and Clinical Characteristics by Use of Inhaled Corticosteroids
Microbiological Findings
An etiologic diagnosis was found in 161 (24.2%) of hospitalized CAP subjects (Table 2). The prior ICS users and non-users had similar rates of isolated pathogens, including S. pneumoniae, S. aureus and P. aeruginosa. Opportunistic pathogens, such as Aspergillus species, Pneumocystis jiroveci, and Nocardia species and extended-spectrum beta lactamases pathogens, were not found in the ICS users.
Etiologies in Patients With Community-Acquired Pneumonia
Clinical Outcomes
The subjects who had used ICS had higher mean Pneumonia Severity Index (100.8 ± 31.4 vs 68.8 ± 33.4, P = .001) and CURB-65 scores (1.56 ± 1.02 vs 1.19 ± 1.02, P = .002) at presentation. In the multivariate analyses, after adjusting for potential confounders, the use of out-patient ICS was independently associated with drug-resistant pathogens: 11.2% vs 5.9%, odds ratio 2.6, 95% CI 1.1–6.1, P = .04.
Subgroup Analyses
The COPD subjects who had used out-patient ICS (n = 58/176, 33.0%) were more likely to have drug-resistant pathogens (12.1% vs 3.4%, odds ratio 3.9, 95% CI 1.1–13.2, P = .03) than were the COPD subjects without prior ICS use (n = 118/176, 67.0%). However, the COPD subjects with prior ICS use had no significant differences regarding severity of illness at presentation: mean Pneumonia Severity Index 103.1 ± 30.0 vs 100.7 ± 30.8, P = .6; mean CURB-65 score 1.6 ± 0.9 vs 1.6 ± 1.1, P = .90, compared to the COPD subjects with no prior ICS use.
The non-COPD subjects managed with ICS (n = 31/488, 6.4%) were not associated with drug-resistant pathogens (9.7% vs 6.6%, P = .40). However, the non-COPD subjects had higher Pneumonia Severity Index (96.5 ± 33.97 vs 83.2 ± 33.2, P = .04) than the non-COPD subjects without prior ICS use (n = 457/488, 93.6%).
Discussion
In hospitalized CAP subjects, prior out-patient use of ICS was associated with a higher severity of illness at admission and higher incidence of antimicrobial-resistant pathogens. The relationship between ICS and pneumonia is one of the most important unresolved questions since Calverley et al6 reported in 2007 the results of the TORCH (Toward a Revolution in COPD Health) study. Multiple studies have confirmed that COPD subjects treated with ICS have an increased likelihood of developing pneumonia.7–12 However, COPD subjects receiving ICS and treated for pneumonia may or may not have worse outcomes, such as higher mortality, compared with subjects not receiving ICS.6–15 This observation has led to speculation that ICS use may increase CAP risk, but, because of ICS immunomodulatory effects, these patients may be protected against severe pneumonia or pneumonia-related complications.26,27 Our results suggest that prior ICS users who develop pneumonia and require hospitalization have more severe CAP than ICS non-users (higher Pneumonia Severity Index23 and CURB-6524 scores at admission). These results are hypothesis-generating, and require further exploration with larger databases and adjusting for other possible confounding variables. In a study on ICS and pneumonia severity, Singanayaman et al14 found no differences in illness severity at admission in their cohort of COPD patients admitted with pneumonia. These findings are consistent with our subgroup analysis of COPD patients. These data suggest that chronic prior ICS use may play an important role in the lungs' defense mechanisms against infections. We hypothesized that, in addition to the immunomodulatory effects observed in patients treated with ICS, there is the possibility that an immunosuppressive effect might be the cause of the higher disease severity. The immunomodulatory effect may be different between COPD and non-COPD patients. However, further studies are needed to assess the effect of ICS on different airway protective mechanisms against infections, such as local inflammatory response, mucins, and antimicrobial peptides, and between different patient populations.28
Different studies have identified a higher incidence of potentially highly resistant bacteria19,20 and opportunistic lung pathogens such as Aspergillus,17 Pneumocystis jiroveci,18 and Nocardia,29 in patients taking systemic corticosteroids. However, the association of antimicrobial-resistant bacteria with out-patient ICS use was unknown. Our study results suggest that patients with prior ICS use admitted to the hospital with CAP, particularly COPD patients, are more likely to have drug-resistant pathogens. Liapikou et al30 reported that COPD patients treated with chronic ICS had a higher rate of pneumonia due to P. aeruginosa, but fewer Legionella species. However, antimicrobial resistance was not assessed in COPD patients treated with ICS.
We hypothesized that ICS may alter habitual flora and antimicrobial susceptibility, particularly in COPD patients with chronic airway infections.31 Previous antimicrobial courses have been associated with development of multi-drug-resistant pathogens32; this is an important variable to consider in future assessment of the risk of antimicrobial resistance related to the use of ICS.
Our study has several limitations. First, this was a retrospective study involving only 2 centers, with a relatively small simple size. Second, our sample was predominately male, due to the higher population of veterans enrolled in this cohort. It is unknown if prior ICS use in females would have similar results. Third, our cohort did not have information regarding the exact ICS dose, ICS clinical indications, or duration of ICS out-patient therapy. Fourth, our database did not include data on multiple courses of antibiotic prescriptions unrelated to the pneumonia hospitalization event. Finally, diagnosis of COPD was not based on spirometry.
Conclusions
Hospitalized CAP patients receiving prior ICS had a higher severity of disease at presentation and more frequently had antimicrobial-resistant pathogens. Further studies are needed to study mechanisms involved in these important findings, which may affect the current indications of the ICS.
Footnotes
- Correspondence: Marcos I Restrepo MD MSc, South Texas Veterans Health Care System ALMD-7400, Merton Minter Boulevard, San Antonio TX 78229. E-mail: restrepom{at}uthscsa.edu.
Dr Anzueto has disclosed a relationship with Boehringer Ingelheim, GlaxoSmithKline, Forest Laboratories, Astra-Zeneca, Pfizer, Intermune, Amgen, Bayer-Schering Pharma, Lilly, and the National Institutes of Health. Dr Restrepo has disclosed relationships with Theravance, Forest Laboratories, Johnson & Johnson, Trius, Novartis, and Pfizer. This research was supported by Howard Hughes Medical Institute faculty start-up grant 00378–001, and by a Department of Veteran Affairs Veterans Integrated Service Network 17 new faculty grant. Dr Sibila was supported by Instituto de Salud Carlos III (BAE11/00102). Drs Sibila and Laserna were supported by Sociedad Espanola de Neumologia y Cirugia Toracica (SEPAR), Societat Catalana de Pneumologia and Fundacio Catalana de Pneumologia. Dr Restrepo was partly supported grant K23HL096054 from the National Heart, Lung, and Blood Institute. The funding agencies had no role in the preparation, review, or approval of the manuscript. The views expressed in this report are those of the authors and do not represent the views of the Department of Veterans Affairs.
- Copyright © 2013 by Daedalus Enterprises
