Chest ACCP Member Benefits
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Guest Access | Sign In via User Name/Password
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (21)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Agustí, C.
Right arrow Articles by Torres, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Agustí, C.
Right arrow Articles by Torres, A.
(Chest. 2003;123:488-498.)
© 2003 American College of Chest Physicians

Pulmonary Infiltrates in Patients Receiving Long-term Glucocorticoid Treatment*

Etiology, Prognostic Factors, and Associated Inflammatory Response

Carlos Agustí, MD, PhD; Ana Rañó, MD, PhD; Xavier Filella, MD, PhD; Juliá González, MD, PhD; Asunción Moreno, MD, PhD; Antoni Xaubet, MD, PhD and Antoni Torres, MD, PhD, FCCP

* From Servei de Pneumologia (Drs. Agustí, Rañó, Xaubet, and Torres), Institut Clínic de Pneumologia i Cirurgía Toràcica; Servei de Microbiologia i Malalties Infeccioses (Drs. Filella, González, and Moreno), Servei de Bioquímica Clínica, Institut Clínic de Infeccions i Inmunitat, Hospital Clínic, Institut d’Investigacions Biomèdiques August Pi i Sunyer, Universitat de Barcelona, Barcelona, Spain.

Correspondence to: Antoni Torres, MD, PhD, FCCP, Servei de Pneumologia, Hospital Clínic, Villarroel 170, 08036 Barcelona, Spain; e-mail: atorres{at}medicina.ub.es


   Abstract
TOP
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Background: Glucocorticoid treatment alters immunoregulatory defense mechanisms and may therefore favor the development of different pulmonary infections.

Methods: The etiology, prognostic factors, and associated inflammatory response of pulmonary infiltrates in 33 patients receiving long-term glucocorticoid treatment (LTGCT) were prospectively evaluated.

Results: Aspergillus spp (n = 9, 31%) and Staphylococcus spp (n = 6, 21%) were the most common causative agents. Using different diagnostic techniques, we obtained a specific diagnosis in 28 of 33 episodes (85%) of pulmonary infiltrates. Bronchoscopic techniques provided the diagnosis in 64% of the cases. Crude mortality was 45%. Variables associated with mortality were as follows: age > 64 years, bilateral radiographic involvement, delay in diagnosis, inappropriate empirical treatment, Simplified Acute Physiology Score (SAPS) II >= 25, and requirement for mechanical ventilation (MV). SAPS II >= 25 (odds ratio [OR], 16; 95% confidence interval, 1 to 260) and MV requirement (OR, 50; 95% confidence interval, 2 to 360) were also significant on multivariate analysis. Pulmonary infections were associated with an increase in the concentration of relevant inflammatory cytokines such as tumor necrosis factor-{alpha} and interleukin-6 both in serum and BAL. This local and systemic inflammatory response was attenuated when compared with the response observed in patients with pulmonary infections but without glucocorticoid treatment or receiving glucocorticoids for a short period of time (< 9 days).

Conclusions: Pulmonary infiltrates in patients receiving LTGCT are often caused by fungi and Gram-positive cocci, and are associated with attenuated local and systemic inflammatory response. Although in most cases, sputum cultures and bronchoscopic techniques are diagnostic, the associated mortality is high, particularly in those requiring MV.

Key Words: BAL • cytokines • diagnosis • glucocorticoids • prognosis


   Introduction
TOP
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The clinical use of glucocorticoids and other immunosuppressive therapies has expanded over the last years. The good results obtained with transplant techniques and the increase in the number of potential candidates requiring immunomodulative therapy to treat different chronic diseases may partially explain the extended use of these drugs.1

Glucocorticoids exert a decisive influence in the immune function of macrophages and granulocytes, the two major immunoregulatory cells in host defenses against opportunistic and bacterial infections.1 Other immunosuppressive drugs such as cyclophosphamide or azathioprine reduce the circulating pool of mononuclear phagocytes.2 As a consequence, one might expect patients receiving long-term immunosuppressive therapy to have depressed resistance to a wide variety of infective agents, as it has been shown in different investigations in animals.3

Numerous reports of opportunistic pulmonary infections (mainly by Pneumocystis carinii and Aspergillus spp) have been described in patients receiving glucocorticoids.4 5 6 7 8 9 However, there are no prospective studies evaluating the etiology, clinical characteristics, and prognostic factors for pulmonary complications in these patients.

In pulmonary infections, the release of cytokines and other inflammatory molecules is an important mechanism in the elimination of invading pathogens.10 However, excessive release can be harmful to the host.11 It has been speculated that adequate modulation of inflammatory response is essential in the evolution of lung infection. Corticosteroids inhibit the transcription of several cytokines that are relevant to inflammatory conditions including interleukin (IL)-1, IL-6, IL-8, and tumor necrosis factor (TNF)-{alpha}, and thus have the potential to modulate the inflammatory response of the lung.12 13 14 Preliminary findings have pointed toward an attenuation of the inflammatory response in nonimmunocompromised patients with pneumonia treated with glucocorticoids for a short period of time.15 To our knowledge, there are no studies evaluating lung inflammatory response in patients with pulmonary complications receiving long-term glucocorticoid treatment (LTGCT). It is not known whether this inflammatory response depends on the dose/duration of glucocorticoids treatment or if it is related to the etiology or outcome of the pulmonary complication.

The aims of the present study were as follows: (1) to determine the etiology, clinical characteristics, and prognostic factors for pulmonary complications in patients receiving LTGCT; and (2) to evaluate local (lung) and systemic (serum) inflammatory response and their potential relationship with the dose/duration of glucocorticoid treatment and with the severity and outcome of the pulmonary complication.


   Materials and Methods
TOP
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Patients
The study included 33 patients (23 patients with different connective tissue disorders and 10 patients with chronic pulmonary disorders) treated chronically with prednisone. Inclusion criteria required current treatment at a dosage > 30 mg/d for > 30 days (LTGCT). Fifteen patients were also receiving another immunosuppressive agent. Clinical, biochemical, and radiographic variables were recorded when symptoms/signs of pulmonary complication were first noticed, with the Simplified Acute Physiology Score (SAPS) II also being calculated.16

To evaluate inflammatory response in LTGCT patients, different cytokines were measured in the BAL fluid and serum and were compared with those obtained in a previously described group of patients with pneumonia requiring mechanical ventilation (MV) who received non-glucocorticoid treatment (NGCT) and in patients with pneumonia requiring MV who received short-term glucocorticoid treatment (STGCT).15 The control group included nine nonsmoking subjects without respiratory disease who underwent bronchoscopy and gave informed consent to complete the examination with BAL.

Diagnostic Procedures
After identification of the pulmonary infiltrates, blood and spontaneous or induced-sputum samples were drawn for culture. A Gram stain was performed to assess the quality of the sputum.17 Bronchoscopy was performed in patients in whom an etiologic diagnosis was not obtained by noninvasive methods and/or there was no clinical or radiographic improvement despite broad-spectrum antimicrobial therapy. The bronchoscopic procedures included protected specimen brush (PSB), bronchial aspirate (BAS), and BAL. The bronchoscope was passed transnasally or transorally into the trachea after topical nasal anesthesia was instilled. BAL was carried out with 150 mL of sterile saline solution in three aliquots (50 mL each) in the involved lobe in patients with localized pulmonary infiltrates, and in the middle lobe or lingula in patients with diffuse pulmonary infiltrates.

Other diagnostic techniques such as transbronchial and bronchial biopsies were carried out in selected patients. Although the primary intention was to perform all the above-referred diagnostic procedures in all patients, the ultimate decision regarding the diagnostic procedures to be performed was conditioned by the patient’s clinical status and the criteria of the physician in charge.

The ethics committee of the hospital approved the study protocol. Informed consent was obtained from the patients referred for a bronchoscopic exploration.

Definitions
The following definitions were taken from Rano et al.18

Pulmonary Comorbidity: Long-term impairment of pulmonary function and/or presence of chronic radiographic involvement.

Etiology of the Pulmonary Infiltrates: Bacterial pneumonia was diagnosed following the accepted clinical criteria and whenever blood or sputum samples grew pathogenic bacteria. Similarly, when a quantitative bacterial culture from BAS, BAL, or PSB specimens grew > 105, 104, or 103 cfu/mL, respectively, the diagnosis of bacterial pneumonia was established. Viruses were considered to be pathogens if they were isolated by cell culture from BAL specimens. P carinii pneumonia was diagnosed by a positive Gomori methenamine silver stain. The identification of Legionella spp and mycobacteria was accepted as a definite diagnosis regardless of colony counts. Fungal pneumonia was diagnosed in the presence of a compatible clinical and radiographic pattern if fungal hyphae were identified by culture or by cytologic evaluation of a Gram stain in at least two different respiratory samples, or when there was histologic evidence for fungal pneumonia or positive blood culture results.19 Pulmonary infection was considered nosocomial (in-hospital) on appearance after 72 h of hospital admission. Noninfectious pulmonary infiltrates were considered when clinical data did not suggest an infectious etiology, no microbiological agents were isolated in any processed sample, and the clinical course and response to treatment were in accordance with an alternative noninfectious etiology.

Inadequate Empirical Treatment: Empirical treatment administered that does not specifically cover the particular etiologies of the pulmonary infiltrates.

Diagnosis Delay: The period of time between the day the pulmonary infiltrates were first demonstrated and the day that results of the diagnostic procedure were available.

Mortality: In-hospital death.

Inflammatory Response
TNF-{alpha}, IL-1, IL-6, IL-8, and IL-10 levels were measured in BAL supernatant and serum using enzyme-linked immunosorbent assay based on the quantitative immunometric sandwich enzyme immunoassay technique (Enzyme Amplified Sensitivity Immunoassay; Medgenix Diagnostics; Fleurus, Belgium) and PerSeptive (Framingham, MA). The following values were regarded as the upper limits for cytokine serum concentrations in normal control subjects in our laboratory: TNF-{alpha}, 20 pg/mL; IL-1, 15 pg/mL; and IL-6, 5 pg/mL.20 Total and differential cell count of the BAL fluid was also performed as described elsewhere.20 Plasma C-reactive protein (CRP) levels were measured on day 1 and day 4 by means of immunonephrometry.

Statistical Analysis
Results are expressed as mean ± SD or as median (range) when appropriate. The influence of several variables on mortality was evaluated by univariate analysis using the {chi}2 test. Thereafter, a multiple logistical regression model was applied to the variables found to be significantly associated with death (p < 0.01) to estimate the odds ratio (OR) and the 95% confidence interval. The Mann-Whitney U test was employed for the comparison of quantitative variables between two groups and the Kruskal-Wallis test for the comparison among the groups. A p value < 0.05 was considered statistically significant.


   Results
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Clinical Characteristics of the Population
The clinical characteristics and demographic data of the 33 patients evaluated are shown in Table 1 . Seventeen patients (51%) had pulmonary comorbidity: COPD (n = 4), chronic asthma (n = 3), idiopathic pulmonary fibrosis (n = 3), and different connective tissue disorders (n = 7). Patients with chronic asthma and COPD were receiving glucocorticoids for pulmonary disease. All patients with idiopathic pulmonary fibrosis and patients with connective tissue disorders were receiving LTGCT for the management of the diffuse interstitial involvement of the lungs. Only three patients were current smokers. Almost all the patients presented with fever (31 of 33 patients, 94%). No patient had neutropenia (neutrophil count < 1 x 109/L). Three patients were already receiving antibiotic treatment (antibacterial) when the pulmonary infiltrate was first detected. No patients were receiving P carinii chemoprophylaxis. Twenty-two patients (67%) were admitted to the ICU because of acute respiratory failure, and 18 patients (55%) required MV.


View this table:
[in this window]
[in a new window]

  		Table 1.. Clinical Characteristics and Demographic Data of the Population Evaluated (n = 33)*

 
All patients showed abnormalities on chest radiography. The infiltrates were focal in 17 cases (51%) and diffuse in 16 cases (49%). In five cases, the infiltrates were cavitated (two cases of Aspergillus spp, one case of Mycobacterium tuberculosis, one Staphylococcus spp, and one Pseudomonas aeruginosa), and in one case adopted nodular configuration (Nocardia asteroides).

Etiology and Diagnosis of the Pulmonary Infiltrates
Overall, a definite etiology could be established during lifetime in 28 cases (85%) and 1 case by necropsy, being of infectious origin in 28 cases (85%) and noninfectious in 1 case (3%). Four patients (12%) remained without a diagnosis (Table 2 ). In 11 cases, the infection was community acquired and in 17 cases was nosocomial. Bacteria (particularly Staphylococcus spp; n = 13, 36%) and fungi (particularly Aspergillus spp; n = 12, 33%) were the most common microorganisms causing pulmonary infiltrates. Interestingly, symptoms of pulmonary infection were present for > 7 days in 58% of the patients with fungal pneumonia and in only 20% of the patients with bacterial pneumonia ({chi}2, p < 0.001). No differences regarding the etiology were observed between patients treated with glucocorticoids and those treated with glucocorticoids and other immunosuppressive drugs (Table 3 ). The diagnostic yield of the different procedures used is summarized in Table 4 . Sputum conferred the diagnosis in 53% of the episodes (mainly bacteria and fungi). Seven of the 10 positive sputum culture findings (70%) were further confirmed by bronchoscopic techniques. Of the bronchoscopic techniques, BAS achieved a specific diagnosis in 75% of the episodes while PSB did so in only 33% of the cases. Moreover, BAS and/or BAL also isolated five of the six causative microorganisms identified by PSB. A specific diagnosis was obtained with BAL in 56% (10 of 18 cases; mean fluid BAL recovery, 64 mL); in 5 cases (15%), BAL was the only technique to provide a definite diagnosis. The diagnosis in the patient with varicella pneumonia was made by previous exposure, typical dermal lesions, and coexisting pulmonary infiltrates. Bronchiolitis obliterans organizing pneumonia was diagnosed by histologic examination of transbronchial biopsy in one case. Bronchial biopsy was obtained in another patient, confirming the diagnosis of pulmonary aspergillosis. Postmortem examination was performed in 6 of the 15 patients who died; pneumonia due to Aspergillus fumigatus was identified in one case not diagnosed during lifetime. Necropsy confirmed one case each of Candida tropicalis, A fumigatus, Aspergillus niger, P aeruginosa, and methicillin-resistant Staphylococcus aureus (MRSA) pneumonia. No statistically significant differences were observed between the different etiologies with regards to mortality.


View this table:
[in this window]
[in a new window]

  		Table 2.. Etiology of Pulmonary Infiltrates and Associated Mortality*

 

View this table:
[in this window]
[in a new window]

  		Table 3.. Pulmonary Complications in Patients Receiving Glucocorticoid Treatment Only and Those Also Treated With Other Immunosuppressive Therapies*

 

View this table:
[in this window]
[in a new window]

  		Table 4.. Diagnostic Yield of the Different Procedures Performed*

 
Outcome and Prognostic Factors
Crude mortality was 45% (15 of 33 patients). Six of the 13 variables included in the univariate analysis were associated with increased mortality (Table 5 ). The variables analyzed were those selected as potential predictors of outcome in this population according to the literature.21 22 23 24 Older patients showed a higher mortality. Variables reflecting the severity of the disease, such as MV requirement, SAPS II >= 25, and the presence of bilateral infiltrates on chest radiography, had decisive influence on mortality. Requirement of MV had the strongest association with mortality since 93% of these patients died (14 of 15 patients), contrary to only 6% of patients not receiving MV (1 of 15 patients; p < 0.0001). Finally, variables amenable to potential medical intervention that were associated with poor outcome were a delay in establishing the diagnosis of > 9 days (median value for the delay in establishing the diagnosis of the overall population) since the beginning of the pulmonary process, and inadequate empirical treatment.


View this table:
[in this window]
[in a new window]

  		Table 5.. Comparison Between Survivors and Nonsurvivors: Univariate and Multivariate Analysis*

 
In order to ascertain whether the existence of previous pulmonary damage may influence the characteristics and outcome of the pulmonary complication, we analyzed separately patients with and without pulmonary comorbidity. Thus, the 17 patients with pulmonary comorbidity (mean spirometric values: FVC, 64%; FEV1, 54%; FEV1/FVC, 84%; and diffusion capacity of the lung for carbon monoxide, 49%) [4 patients with COPD, 3 patients with chronic asthma, 3 patients with idiopathic pulmonary fibrosis, 2 patients with scleroderma, 2 patients with rheumatoid arthritis, and one case each of Sjögren syndrome, systemic lupus erythematosus, and Horton disease] were contrasted with the 16 patients with no apparent pulmonary involvement (no respiratory symptoms, normal chest radiographic findings, and normal pulmonary function test results when available). Patients with pulmonary comorbidity were older (69 ± 11 years vs 58 ± 15 years, p = 0.032) and had lower creatinine levels (1.3 ± 1 mg/dL vs 3.1 ± 3.4 mg/dL, p = 0.04) than patients without apparent lung disease. No differences were observed between these two groups regarding the cumulative dose of glucocorticoids, PaO2/fraction of inspired oxygen (FIO2), radiographic pattern, delay in establishing a specific diagnosis, severity (SAPS II), or outcome of the pulmonary complication. Similarly, the distribution of immunosuppressive drugs other than glucocorticoids between the two groups was alike (Table 1) , and there was no difference in mortality between patients receiving glucocorticoids and those receiving glucocorticoids plus other immunosuppressive agents.

After a multivariate statistical approach, SAPS II >= 25 and MV requirement were the dominant independent variables significantly predicting mortality. Inadequacy of empirical antibiotic treatment was also included in the multivariate model but did not reach statistical significance at the 0.05 level (p < 0.09).

Pulmonary Inflammatory Response
We were able to obtain BAL samples for differential cell counts and cytokine level determinations in 13 of the 33 patients evaluated. No differences were observed in the clinical variables or the etiology and severity of the pulmonary complication among these 13 patients and the 20 subjects in whom BAL was not available (data not shown). The etiology of the pulmonary infiltrates in those 13 patients with BAL cytokine determinations were as follows: bacterial pneumonia in six cases, fungal pneumonia in two cases, P carinii pneumonia in two cases, and one each of viral pneumonia, pulmonary tuberculosis, and nondiagnostic infiltrates. Patients receiving LTGCT showed an increase in BAL neutrophil percentage and significantly higher levels of TNF-{alpha}, IL-8, and IL-6 compared with healthy control subjects (Table 7 ). Five of the 13 patients in whom BAL cytokine levels were available died as a result of the pulmonary infection. No differences were observed in the different cytokines levels between patients with and without pulmonary comorbidity or between survivors and nonsurvivors (data not shown). BAL cytokine levels of these 13 patients with pneumonia were compared with two different groups of patients with pneumonia that have been previously described.15 The first group consisted of patients with pneumonia receiving MV requiring glucocorticoids for bronchial dilatation (STGCT) for a period of 9 ± 7 days. The second group consisted of patients with pneumonia receiving MV and NGCT (Table 7 ).15 As shown in Table 7 , the three groups of patients evaluated were of a similar age but the dose and duration of glucocorticoid treatment differed substantially. For the patients requiring MV (six patients in the LTGCT group), the severity of the pulmonary complication measured by the PaO2/FIO2 ratio, SAPS II, and the number of days requiring MV was similar.


View this table:
[in this window]
[in a new window]

  		Table 7.. Clinical Characteristics, Glucocorticoid Doses, and Mortality of the Three Groups of Patients With Pneumonia*

 
When we compared BAL cytokine levels in the three groups of patients with pneumonia receiving different doses of glucocorticoids (Table 8 ), we observed that the percentage of neutrophils in BAL and the levels of TNF-{alpha} and IL-6 differed significantly. Patients receiving high doses of GC for extended periods of time (LTGCT) showed the lowest pulmonary inflammatory response, while patients receiving NGCT had the highest inflammatory response. Patients receiving STGCT had an intermediate inflammatory response (Fig 1 ).


View this table:
[in this window]
[in a new window]

  		Table 8.. BAL Neutrophil Count, Cytokine Levels (BAL and Serum), and CRP (Serum) Determinations in Three Groups of Patients With Pneumonia Receiving Different Doses of Glucocorticoids*

 


View larger version (28K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1.. Percentage of neutrophils (BAL) and concentrations of TNF-{alpha} (BAL), IL-6 (BAL and serum), and CRP on day 1 and day 4 (serum) for control subjects and patients receiving LTGCT, STGCT, or NGCT. Bars and error bars represent mean and SEM. Cytokines are expressed as picograms per milliliter. See Table 3 for expansion of abbreviation.

 
Systemic Inflammatory Response
Serum values of TNF-{alpha}, IL-1ß, and IL-6 were determined in 13 patients receiving LTGCT the day pulmonary infiltrates were first detected. Similarly, serum CRP was measured the first day and 4 days later. As a group, LTGCT patients showed moderate systemic inflammatory response, with most of the plasma mediators (except IL-1) having median values above the upper limit of normality (Table 6 ). Thus, 9 patients (75%) and 11 patients (88%) had increased serum values of TNF-{alpha} and IL-6, respectively. There was no difference in serum cytokine levels between survivors and nonsurvivors, and poor correlations were observed between serum and BAL cytokine levels.


View this table:
[in this window]
[in a new window]

  		Table 6.. BAL Neutrophil Count and Percentage, Cytokine Levels (BAL and Serum) and CRP (Serum) Determinations of Patients With Pneumonia and LTGCT Compared to Healthy Control Subjects*

 
Patients receiving LTGCT had increased levels of CRP in serum on day 1, with a marked decrease on day 4 (Table 8) . Interestingly, there was no difference in serum CRP values between survivors and nonsurvivors on day 1 (15.7 ± 16 mg/dL vs 15.2 ± 2 mg/dL, respectively); however, in patients who died (n = 5), CRP levels remained high on day 4 (14 ± 11 mg/dL) while in patients who survived (n = 8), this value tended to normalize (3.8 ± 3 mg/dL) [p < 0.03].

When the three groups of patients receiving different doses of glucocorticoids were compared, we observed that IL-6 and CRP levels in serum showed the same trend as BAL cytokines: patients in the LTGCT group had the lowest systemic inflammatory response, while patients in the NGCT group had the highest inflammatory response. Patients receiving low doses of glucocorticoids had intermediate values (Fig 1) . Interestingly, and as can be seen in Table 7 , patients receiving low doses of glucocorticoids (and with an intermediate pulmonary inflammatory response) demonstrated the lowest mortality.


   Discussion
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The present study shows the following in patients receiving high doses of glucocorticoids for prolonged periods of time: (1) the vast majority of the pulmonary complications are due to infections, mainly by fungi (particularly Aspergillus spp) and bacteria (particularly Staphylococcus spp); (2) pulmonary infections carry a poor prognosis with a 50% mortality, and variables associated with mortality are MV requirement, a high SAPS II, and inadequate empirical antibiotic treatment; (3) pulmonary infections are associated with local and systemic inflammatory response reflected by a slight increase in the levels of relevant cytokines in serum and BAL, and by moderate increases in serum CRP; and (4) inflammatory response seems to be attenuated compared with that observed in patients not receiving glucocorticoids. Patients receiving glucocorticoids for a short period of time show intermediate lung inflammatory response.

Etiology and Diagnosis
Although different reports have pointed toward an increase in the incidence of different opportunistic infections of the lung, particularly P carinii and Aspergillus spp,4 5 6 7 to our knowledge, this is the first report prospectively evaluating the etiology and clinical characteristics of pulmonary infiltrates in patients receiving glucocorticoids chronically. The results of the present study demonstrate that fungi (particularly Aspergillus spp) are the most frequent microorganisms causing pulmonary infiltrates in these patients. Interestingly, bacterial infections were almost as prevalent as fungal infections, and particularly S aureus constituted the most common bacteria isolated. These findings are remarkable since, classically, LTGCT has not been considered as a risk factor for Staphylococcus pneumonia.25 Animal studies have shown that the risk of infection due to glucocorticoids is both dose dependent and time dependent.2 In this sense, the relatively high dose received in the population studied may have caused a profound impairment in immunity and may justify the myriad of microorganisms isolated. Chemoprophylaxis for P carinii has been claimed for patients receiving high doses of glucocorticoids.26 Although in the present series only two patients had an episode of pneumonia by this microorganism, both died as a result of this infection, reinforcing the potential benefits of the chemoprophylaxis. Despite the lack of differences in the clinical characteristics among the different etiologies, patients with fungal pneumonia had a longer period of symptoms before seeking medical attention compared to bacterial infections.

The present study confirms our previous findings in a larger group of immunosuppressed patients using different diagnostic tools.18 Bronchoscopic techniques and particularly BAS and BAL have the highest diagnostic yield, while PSB does not seem to provide any additional diagnostic information. Sputum culture is a simple noninvasive technique that provides an specific diagnosis in more than one half of the cases and must be taken into account as a first step in the diagnostic approach of these patients.

Outcome and Prognostic Factors
The present study demonstrates that the mortality of pulmonary infection in patients receiving LTGCT is high (50%) and similar to that observed in patients with other types of immunosuppressive conditions.18 Different variables related to the severity of pulmonary infection had prognostic significance. In particular, MV requirement was a predictive factor of mortality with an OR of 50, confirming this variable as the most determinant in relation to mortality in different groups of immunocompromised patients.18 21 The prognostic value of the SAPS II has already been demonstrated in nonimmunocompromised patients with pneumonia,16 and the present study confirms its utility in patients receiving LTGCT. Although statistical significance was not achieved at the 0.05 level on multivariate analysis (due in part to the small number of patients evaluated), the inadequacy of the empiric antibiotic treatment is a relevant prognostic variable since it is amenable to clinical intervention.27 Based on the high number of cases (58%) in which the empirical treatment did not cover the etiologic agent, and on the associated mortality, antibiotics covering fungi, Staphylococcus, and P aeruginosa should be used as first-line antibiotic treatment. The implications of excessive delay in establishing a specific diagnosis (significant on univariate analysis), and the high diagnostic yield of BAS and BAL suggest that bronchoscopy evaluation should be carried out promptly once pulmonary infiltrates have been identified.18

Pulmonary and Systemic Inflammatory Response
The present study shows that pneumonia in patients receiving LTGCT is associated with a pulmonary inflammatory response reflected by significant increases of relevant proinflammatory cytokines in BAL such as IL-8, TNF-{alpha}, and IL-6. Interestingly, the comparison of levels of proinflammatory cytokines in BAL and serum among patients receiving different doses of glucocorticoids showed that local and systemic inflammatory response are markedly decreased in patients receiving LTGCT compared with patients not receiving glucocorticoids. The acute administration of glucocorticoids seems to have an intermediate effect on the intensity of inflammatory response (Fig 1) . A potential drawback of the comparison between these three groups is that not all the patients receiving LTGCT were receiving MV. Since there is some evidence that MV could initiate or propagate a systemic inflammatory response,28 we performed a separate analysis comparing cytokine levels in patients receiving LTGCT with (n = 6) or without (n = 7) MV requirement. No significant differences were observed for any of the cytokines evaluated (data not shown). Accordingly, we considered all the population (n = 13) for comparison with STGCT and NGCT groups.

Although it is known that glucocorticoids inhibits the expression and action of most inflammatory cytokines, the clinical significance of this effect is controversial. In the present series, serum and BAL levels of the different inflammatory cytokines evaluated had no prognostic significance (probably due to the relatively small number of patients studied). However, the mortality in the group of patients receiving glucocorticoids (showing minimal local and systemic inflammatory response) was similar to that observed in patients not receiving glucocorticoids (with marked inflammatory response). Remarkably, patients receiving STGCT showed an attenuated inflammatory response (but not as evident as the group of LTGCT patients) and a better outcome. These results seem to agree with the clinical experience regarding glucocorticoid treatment in patients with septic shock. Thus, although there is no definitive evidence to support the use of glucocorticoids in patients with severe sepsis,29 the results of different studies suggest that small doses of glucocorticoids over a short period of time attenuate systemic inflammatory response, reduce the duration of shock, and increase survival.30 In patients with acute pneumonia, the adjunctive use of glucocorticoids has an antipyretic effect and can accelerate symptomatic relief, but does not seem to influence survival.31 Whether small doses of glucocorticoids may improve survival in acute pneumonia must be evaluated in future studies.

Serum levels of CRP were also lower in patients receiving LTGCT. Interestingly, although serum levels of CRP on day 1 had no prognostic value, the persistence of elevated levels of this acute phase protein on day 4 reflected a poor outcome. This finding confirms previous reports showing that persistently high or rising CRP levels in patients with pneumonia is a good clinical marker of treatment failure.32

Although the results of the present study must clearly be confirmed in properly designed studies with a larger number of patients, they suggest that profound attenuation of the inflammatory response secondary to LTGCT may be as harmful as an exaggerated production of inflammatory cytokines. Of critical importance in the design of futures studies is to attempt to elucidate not only which patients can potentially benefit from glucocorticoids but also what optimal dose and duration of glucocorticoid treatment are necessary to create an appropriate balance between the beneficial and harmful effects of the inflammatory response.

The following limitations have to be considered for the interpretation of the results. First, the study is based on a third-level hospital population and consequently does not necessarily reflects the whole population receiving LTGCT. Besides, although this was a prospective multidisciplinary study including all patients with pulmonary complications, we cannot definitively rule out that some "benign" cases with a rapid response to a specific treatment were not included, biasing the study to a more severe population. Second, the studied population was rather heterogeneous. Half of them were receiving an additional immunosuppressive regime or had previous pulmonary involvement. Although neither of these two characteristics seemed to have a significant influence in terms of outcome or inflammatory response, we cannot definitively rule out that a different level of immunosuppression may have had some influence in the development of a particular pulmonary complication and outcome or in the BAL cytokine levels. Finally, although the number of patients included is small, this is the first attempt to prospectively evaluate a particular group of immunocompromised patients. We believe that the information provided is remarkable from a clinical point of view and may constitute the rationale for new, properly designed protocols.

In summary, pneumonia in patients receiving LTGCT is mainly due to Aspergillus, Staphylococcus spp, and Gram-negative bacilli., and it carries a high mortality, particularly in patients requiring MV. Empirical treatment should include antibiotics covering these microorganisms. An early bronchoscopic investigation can achieve a specific diagnosis in a high number of cases and may help in deciding on a specific treatment. Pneumonia in patients receiving LTGCT is associated with local and systemic inflammatory response reflected by an increase in BAL and serum levels of relevant cytokines. This inflammatory response seems to be attenuated compared with that observed in patients receiving STGCT or NCGT. The implications of this attenuated inflammatory response in the outcome of pneumonia must be evaluated in properly designed studies.


   Footnotes
 
Abbreviations: BAS = bronchial aspirate; CRP = C-reactive protein; FIO2 = fraction of inspired oxygen; IL = interleukin; LTGCT = long-term glucocorticoid treatment; MRSA = methicillin-resistant Staphylococcus aureus; MV = mechanical ventilation; NGCT = non-glucocorticoid treatment; OR = odds ratio; PSB = protected specimen brush; SAPS = Simplified Acute Physiology Score; STGCT = short-term glucocorticoid treatment; TNF = tumor necrosis factor

Grant support was provided by SOCAP (Societat Catalana de Pneumologia) and FUCAP-Bayer (Fundació Catalana de Pneumologia).

Received for publication February 22, 2002. Accepted for publication July 10, 2002.


   References
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Cato, ACB, Wade, E (1996) Molecular mechanisms of anti-inflammatory action of glucocorticoids. Bioessays 18,371-378[CrossRef][ISI][Medline]
  2. Sen, RP, Walsh, TE, Fisher, W, et al Pulmonary complications of combination therapy with cyclophosphamide and prednisone. Chest 1991;99,143-146[Abstract/Free Full Text]
  3. Berenguer, J, Allende, MC, Lee, JW, et al Pathogenesis of pulmonary aspergillosis granulocytopenia vs cyclosporine and methylprednisone-induced immunosuppression. Am J Respir Crit Care Med 1995;152,1079-1086[Abstract]
  4. Stuck, AE, Minder, CE, Frey, FJ Risk of infectious complications in patients taking glucocorticoids. Rev Infect Dis 1989;69,954-963
  5. Conesa, D, Rello, J, Valles, J, et al Invasive aspergillosis: a life-threatening complication of short-term steroid treatment. Ann Pharmacother 1995;29,1235-1237[Abstract]
  6. Wiest, PM, Flanigan, T, Salata, RA, et al Serious complications of corticosteroid therapy for COPD. Chest 1989;95,1180-1184[Abstract/Free Full Text]
  7. Rello, J, Esandi, ME, Díaz, E, et al The role of Candida sp isolated from bronchoscopic samples in nonneutropenic patients. Chest 1998;114,146-149[Abstract/Free Full Text]
  8. Rodríguez, J, Niderman, MS, Fein, AM, et al Nonresolving pneumonia in steroid-treated patients with obstructive lung disease. Am J Med 1992;93,29-34[CrossRef][ISI][Medline]
  9. Palmer, LB, Greenberg, HE, Schiff, MJ Corticosteroid treatment as a risk factor for invasive aspergillosis in patients with lung disease. Thorax 1991;46,15-20[Abstract]
  10. Stevens, DL Cytokines: an updated compendium. Curr Opin Infect Dis 1995;8,175-180
  11. Bone, RC Toward a theory regarding the pathogenesis of the systemic inflammatory syndrome: what we do and do not know about cytokine regulation? Crit Care Med 1996;24,163-172[CrossRef][ISI][Medline]
  12. Jantz, MA, Sahn, SA Corticosteroids in acute respiratory failure. Am J Respir Crit Care Med 1999;160,1079-1100[Free Full Text]
  13. Barnes, PJ Anti-inflammatory mechanisms of glucocorticoids. Biochem Soc Trans 1995;23,940-945[Medline]
  14. Brattsand, R, Linden, M Cytokine modulation by glucocorticoids: mechanisms and actions in cellular studies. Aliment Pharmacol Ther 1996;10(suppl 2),81-90
  15. Montón, C, Ewig, S, Torres, A, et al Role of glucocorticoids on inflammatory response in nonimmunosuppressed patients with pneumonia: a pilot study. Eur Respir J 1999;14,218-220[Abstract]
  16. Le Gall, JR, Lemeshow, S, for the ICU Scoring Group. The SAPS-II: a new score with new objectives. Vincent, JL eds. Yearbook of intensive care and emergency medicine 1994,795-804 Springer-Verlag New York, NY.
  17. Murray, PR, Washington, JA, II Microscopic and bacteriologic analysis of expectorated sputum. Mayo Clin Proc 1975;50,339-344[ISI][Medline]
  18. Rañó, A, Agustí, C, Jiménez, P, et al Pulmonary infiltrates in non-HIV immunocompromised patients: a diagnostic approach using non-invasive and bronchoscopic procedures. Thorax 2001;56,379-387[Abstract/Free Full Text]
  19. El-Ebiary, M, Torres, A, Fàbregas, N, et al Significance of the isolation of Candida species from respiratory samples in critically ill, non-neutropenic patients: an immediate postmortem histologic study. Am J Respir Crit Care Med 1997;156,583-590[Abstract/Free Full Text]
  20. Montón, C, Torres, A, El-Ebiary, M, et al Cytokine expression in severe pneumonia: a bronchoalveolar study. Crit Care Med 1999;27,1745-1753[CrossRef][ISI][Medline]
  21. Vincent, JL Prevention of nosocomial bacterial pneumonia. Thorax 1999;54,544-549[Free Full Text]
  22. Heath, CH, Grove, DI, Looke, DF Delay in appropriate therapy of Legionella pneumonia is associated with increased mortality. Eur J Clin Microbiol Infect Dis 1996;15,286-290[CrossRef][ISI][Medline]
  23. Poe, RH, Wahl, GW, Qazi, R Predictors of mortality in the immunocompromised patients with pulmonary infiltrates. Arch Intern Med 1986;146,1304-1308[Abstract]
  24. Rañó, A, Agustí, C, Benito, N, et al Prognostic factors of non-HIV immunocompromised patients with pulmonary infiltrates. Chest 2002;122,253-261[Abstract/Free Full Text]
  25. Rello, J, Quintana, E, Ausina, V, et al Risk factors for Staphylococcus aureus nosocomial pneumonia in critically ill patients. Am Rev Respir Dis 1990;142,1320-1324[ISI][Medline]
  26. Godeau, B, Coutant-Perronne, V, Huong, DL, et al Pneumocystis carinii pneumonia in the course of connective tissue disease: report of 34 cases. J Rheumatol 1994;21,246-251[ISI][Medline]
  27. Kollef, MH, Ward, S, Sherman, J, et al Inadequate treatment of nosocomial infections is associated with certain empiric antibiotic choices. Crit Care Med 2000;28,3456-3464[CrossRef][ISI][Medline]
  28. Slutsky, AS, Trembley, LN Multiple system organ failure: is mechanical ventilation a contributing factor? Am J Respir Crit Care Med 1998;157,1721-1725[Free Full Text]
  29. Lefering, R, Neugebauer, EA Steroid controversy in sepsis and septic shock: a meta-analysis. Crit Care Med 1995;23,1294-1303[CrossRef][ISI][Medline]
  30. Weitzman, S, Berger, S Clinical trial design in studies of corticosteroids for bacterial infections. Ann Intern Med 1974;81,36-42[ISI][Medline]
  31. McGowan, JE, Chesney, PJ, Crossley, KB, et al Guidelines for the use of systemic glucocorticoids in the management of selected infections. J Infect Dis 1992;165,1-13[ISI][Medline]
  32. Smith, RP, Lipworth, BJ, Cree, IA, et al C-reactive protein: a clinical marker in community-acquired pneumonia. Chest 1995;108,1288-1291[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 ThoraxHome page
C Agusti, A Rano, M Rovira, X Filella, N Benito, A Moreno, and A Torres
Inflammatory response associated with pulmonary complications in non-HIV immunocompromised patients
Thorax, December 1, 2004; 59(12): 1081 - 1088.
[Abstract] [Full Text] [PDF]

Home page
 Ann. Thorac. Surg.Home page
J.-Y. Wang, Y.-L. Chang, L.-N. Lee, J.-H. Chen, J.-L. Tang, P.-C. Yang, and Y.-C. Lee
Diffuse pulmonary infiltrates after bone marrow transplantation: the role of open lung biopsy
Ann. Thorac. Surg., July 1, 2004; 78(1): 267 - 272.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (21)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Agustí, C.
Right arrow Articles by Torres, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Agustí, C.
Right arrow Articles by Torres, A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS