| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the VA Western New York Healthcare System (Drs. Murphy and Sethi), Buffalo, NY; and Winthrop-University Hospital (Dr. Niederman), Mineola, NY.
Correspondence to: Michael S. Niederman, MD, FCCP, Pulmonary and Critical Care Medicine, Winthrop-University Hospital, 222 Station Plaza N, Suite 400, Mineola, NY 11501; e-mail: mniederman{at}winthrop.org
Abstract
The role of infection in exacerbations of COPD remains controversial and incompletely understood. Although some investigators believe that bacteria are not important for patients with exacerbation, we disagree and believe that patients with at least two of the three cardinal symptoms of exacerbation should receive antibiotic therapy. With an open-minded view of the area, we review the data, showing that bacteriologic studies, pathologic investigations, and clinical trials all support roles for bacteria and antibiotic therapy in this disease. Still, many questions remain, and future studies will be needed to better define the mechanisms of bacterial invasion in the bronchitic patient and to develop effective vaccines to prevent exacerbations. In the meantime, we must rely on antibiotic therapy, and we will need prospective studies to corroborate preliminary findings showing that different patients may require different therapies; thus, patient subsetting may be vital in the selection of antibiotic therapy for exacerbations of COPD.
Key Words: antibiotic therapy • bacterial bronchitis • chronic bronchitis • exacerbation • immune response • nontypeable H influenzae • vaccines
The role of bacterial infection in COPD is not completely understood despite decades of investigation by many authors. Obstacles to defining the role of bacteria in COPD in general, and their role in causing exacerbations in particular, include the following: (1) bacteria colonize the lower airways in COPD patients, even in the absence of symptoms of an exacerbation; (2) patients with COPD are highly heterogeneous, and, as a result, bacteria may vary from patient to patient and are likely to play different roles in different individuals; (3) the information obtained from cultures of expectorated sputum samples is limited since the samples do not reliably reflect conditions in the lower airways; (4) even if bacteria play a role in exacerbations, they probably account for only half of all infectious episodes in COPD patients; (5) the three bacteria most strongly implicated in causing exacerbations (nontypeable Haemophilus influenzae, Moraxella catarrhalis, and Streptococcus pneumoniae) are exclusively human pathogens, limiting the use of animal models; and (6) studies of antibiotic therapy for exacerbations of COPD have not generally accounted for differences among patients, and they have not been adapted to account for the rising rate of antimicrobial resistance among the commonly identified bacteria.
In the preceding article (page 193), Hirschmann reviews and interprets selected studies that have addressed the potential roles of bacteria as causes of exacerbations of COPD. Hirschmann concludes that the literature "fails to provide compelling evidence that bacteria cause COPD exacerbations," and that "antibiotics provide no additional benefit" in the management of exacerbations. We disagree with the interpretations of the literature and the conclusions reached in this article. Hirschmann makes a number of conclusions that are incomplete and misleading, including the following suggestions: (1) bacterial colonization is not more common during exacerbation than in a stable state of COPD; and (2) antibiotic trials for COPD exacerbation show no benefit, and, thus, antibiotic therapy for exacerbation is unhelpful. Most importantly, Hirshmann’s perspective is a naysayer’s view about the role of antibiotics in the management of exacerbations of COPD, a view that is far too pessimistic and not very helpful. Most clinicians have not had an experience to support the conclusion of the article, in which Hirschmann recommends against the use of antibiotic therapy because "at most only 40% of exacerbations might benefit and even in those the results are not dramatic."
A more objective and comprehensive review of the literature with attention to new information about mechanisms of microbial pathogenesis in the respiratory tract, assaying the human immune response using appropriate assays and antigens, novel approaches to assess airway microbiology and inflammation, and a more critical assessment of prospective randomized antibiotic trials leads us to the conclusion that bacteria do play an important role as causes of exacerbations of COPD. We briefly outline the current state of knowledge in several areas, review the available literature in what we believe is a more "open-minded" fashion, and propose constructive avenues for future research to advance the field with regard to the role and management of bacterial infection in patients with COPD.
Microbiological Studies
Investigators have been performing quantitative cultures on sputum samples for 50 years. Such studies have not provided definitive answers regarding the role of bacteria in exacerbations of COPD. These studies show that it is an oversimplification to assume that pathogens are more likely to be present, or to be present in higher numbers, in sputum cultures during exacerbations than in stable periods. Sputum cultures will not distinguish between colonization and infection. Furthermore, the absence of a bacterial pathogen does not exclude an organism as a cause of lower respiratory tract infection. Recall that in the classic study by Austrian and Gold,1 S pneumoniae was absent from the sputum in one quarter of patients with pneumococcal pneumonia whose etiology was proven unequivocally by the presence of bacteremia.
Another 50 years of quantitative sputum cultures will not shed additional light on the role of bacteria in exacerbations of COPD. More innovative approaches are required. One approach has been to culture samples obtained by the protected-specimen brush technique. Such studies have established that bacteria are present in the lower airways during exacerbations at thresholds normally used to define pneumonia (> 1,000 cfu/mL), an observation that has been reproduced in four separate studies.2 3 4 5 This approach provides compelling evidence that bacteria are present in the lower airways during exacerbations in concentrations sufficient to cause invasive infection. In fact, in one outpatient study, 40 patients were studied at a time when their illness was stable, while another 29 patients were studied during an acute exacerbation.2 In that study, only 25% of the outpatients in stable condition had a significant concentration (ie, > 1,000 cfu/mL) of bacteria recovered by the protected-specimen brush technique from the lower airway, while 52% of patients with exacerbation had this amount of bacteria present. In fact, if a concentration of > 10,000 cfu/mL was used, then only 5% of the patients in stable condition had that many bacteria, compared to 24% of the patients with exacerbation. Although Hirschmann might be unhappy with these data, since the same patients were not studied serially, we believe that the data in this study are of high quality and relevant to the issues being considered.
Serologic Studies
Over the past decade, our understanding of the antigenic structure of surface antigens of nontypeable H influenzae and M catarrhalis has grown considerably. In evaluating the literature on immune responses to bacteria in COPD, it is critical that the following two features be considered in detecting a relevant immune response: (1) the homologous infecting strain must be used in immunoassays; and (2) immunoassays that detect antibodies to antigens on the bacterial surface must be utilized. The surface of nontypeable H influenzae, for example, has an enormous amount of antigenic heterogeneity among strains. Since humans generate immune responses preferentially to strain-specific epitopes, antigens from homologous strains must be used in order to detect an immune response.6 7
Hirschmann discusses five studies published between 1959 and 1973.8 9 10 11 12 All used a laboratory strain in immunoassays, so, as one would predict, specific immune responses were not observed. The literature is replete with such studies, emphasizing the importance of an awareness of more modern work to accurately interpret the literature on the human immune response to bacterial pathogens in COPD. More recent studies in which appropriate immunoassays with homologous strains are employed show unequivocally that a specific immune response to the infecting bacterial strain is observed following exacerbations of COPD.6 13 14 These observations provide another line of compelling evidence that bacteria cause exacerbations of COPD. An analysis of both systemic and mucosal immune responses to bacteria is an important area of further research. Such studies will further characterize the role of bacteria in COPD, elucidate the elements of a protective immune response, and guide vaccine development.
Vaccine Studies
There is considerable interest in developing effective vaccines to prevent bacterial infections in adults with COPD. The currently licensed 23-valent pneumococcal vaccine has produced equivocal results in patients with COPD.15 An important factor that may contribute to the variable efficacy in COPD is that the immune response to the vaccine is lower among elderly populations, which comprise a large proportion of those who suffer from COPD.16 17 18
A series of clinical trials with a killed whole-cell vaccine using a strain of nontypeable H influenzae produced a short-lived reduction in the incidence of exacerbations of COPD.19 20 21 22 These provocative studies are promising with regard to a potential role for mucosal vaccines for patients with COPD. However, more work is needed to understand the mechanism of the reduction in exacerbations to guide the formulation of future vaccines.
In view of the variable immunogenicity of the pneumococcal polysaccharide vaccine and the need for more work on the mechanism of action of the killed whole-cell vaccine of H influenzae, one cannot draw meaningful conclusions regarding the role of bacteria in exacerbations of COPD based on vaccine studies at this time. However, this is a fruitful area for future research. Clearly, the development of vaccines to prevent bacterial infections in COPD should receive a high priority in our national research effort.23
Pathologic Investigations
A hallmark of COPD is airway inflammation. Indeed, the level of
airway neutrophilia correlates with sputum production and airway
obstruction.24
Studies of sputum, BAL fluid, and bronchial
biopsy specimens reveal the recruitment of neutrophils and eosinophils
to the airways during exacerbations of COPD. In all studies,
neutrophils outnumber eosinophils during
exacerbations.25
26
27
More recent studies of inflammatory
markers in the airways during exacerbations of COPD reveal that acute
bacterial exacerbations are associated with greater airway inflammation
than nonbacterial exacerbations with regard to interleukin-8, tumor
necrosis factor-
, neutrophil elastase, and myeloperoxidase,
which are all markers of neutrophilic
inflammation.28
More work is needed to elucidate the
mechanisms of airway inflammation in COPD, especially the role
of bacteria in the process. In view of the association
of eosinophilia with a variety of viral and bacterial
infections,29
30
31
32
33
it will be interesting to test the
hypothesis that specific bacterial antigens are involved in the
recruitment of neutrophils and eosinophils during exacerbations.
An analysis of lung tissue from patients with COPD with immunoperoxidase staining and polymerase chain reaction has revealed that nontypeable H influenzae is present in the epithelium, the submucosa of the bronchi, the bronchioles, the interstitium, and the alveolar epithelium.34 Elegant studies with normal human bronchial epithelial cells reveal that nontypeable H influenzae adheres to and enters exclusively nonciliated cells by the process of macropinocytosis.35 Such studies require us to change our thinking regarding colonization and infection of the human respiratory system by nontypeable H influenzae. In addition to being present in the airway lumen, bound to mucus and adhering to respiratory epithelium, it is now clear that nontypeable H influenzae invades beyond the mucosal surface during colonization and infection. Investigation into the mechanisms of invasion and survival in respiratory tissue will be important and may lead to novel ways to treat and prevent bacterial infections in COPD.
Trials of Antibiotic Therapy
It is our belief that the available data about antibiotic therapy for exacerbations of COPD support a role for their use, probably because bacteria are often responsible for exacerbations. The data show a benefit from antibiotic therapy, but there is no question that more data are needed, particularly from studies that are designed to test whether different patients require different therapies.36 If this latter concept, which is supported by several observational studies, is proven, then it may explain why the benefits of antibiotics, tested in trials without accounting for patient differences, have been present, but not dramatic.
Hirschmann has interpreted the study by Anthonisen et al37 and the meta-analysis by Saint et al38 as generally showing no benefit for antibiotics, when compared to placebo in prospective randomized trials. Our reading of these studies is not quite so negative. The study by Anthonisen et al37 is clearly one of the largest and best-designed studies on this topic and showed a definite benefit for antibiotics in the > 80% of patients with chronic bronchitis with exacerbations who had at least two of the following three cardinal symptoms: increased dyspnea; increased sputum volume; and increased sputum purulence. The benefits of antibiotics were more dramatic for patients with all three symptoms than for those with two symptoms, but the benefits included more antibiotic-treated patients who responded to therapy than those who did not, compared to placebo-treated patients. When the duration of the exacerbation was examined, those who received antibiotics had a significantly shorter course of illness (2.2 days) than those receiving placebo. In addition, peak flow rates improved more rapidly in those patients receiving antibiotics. Although Hirschmann considers this latter benefit minor, "attributed to the expected variation in testing" rather than to a therapeutic benefit, the small changes seen certainly could be clinically relevant and could, in fact, provide a physiologic explanation for the finding of a shorter duration of illness in the antibiotic-treated patients. For reasons that are unclear to us, Hirschmann is so overly pessimistic about these data that he incompletely describes the results when he states that the use of corticosteroids confused the trial and that antibiotics were not better than placebo in those patients receiving corticosteroids. That is not exactly what was shown. In the trial, steroids were used as adjunctive therapy in both antibiotic- and placebo-treated patients about 40% of the time, and, thus, the number of patients who actually received steroids was relatively significant. Although not statistically significant (but limited by the small numbers), the authors found that for those patients treated with steroids, "the success rate with antibiotic was higher than that with placebo." In addition, the authors found that the benefits of antibiotics compared to placebo were the same whether steroids were used or not, calling into question Hirschmann’s belief that steroids alone, without antibiotics, are all that is needed for exacerbations of COPD.
The meta-analysis by Saint et al38 also shows benefits from antibiotic therapy for exacerbations, compared to placebo. In nine prospective randomized trials conducted from 1957 to 1992, the overall effect size (defined as the SD of benefit with therapy vs placebo for the effect measured) was 0.22, but seven of the nine trials showed a benefit for antibiotics. Although this benefit may have been small, it is likely important, and, given the consistent findings of the studies, it seems difficult to withhold antibiotic therapy from these patients. In addition, it is clear that there could be other benefits of therapy not reported in the meta-analysis. As discussed above, the trial by Anthonisen et al37 showed a shorter duration of illness with therapy; other similar benefits could be present and need to be examined in future trials, including the prevention of hospitalization, a more rapid return to usual activities, and fewer days lost from work.
We believe that although antibiotics have benefit compared to placebo, we should continue to collect more information about this form of therapy. However, future studies should not be asking whether antibiotics are helpful, but rather whether choosing one agent over another, with specific patient profiles in mind, is truly beneficial for patient outcome.36 If, in fact, trials can be designed with this question in mind, and if differences among patients can be defined, the results can be used to explain why the benefits of antibiotics have not been more dramatic in some of the older trials that assumed that all patients were equivalent and all antibiotics were of similar benefit. Several studies have suggested that patients with different severities of chronic lung disease have exacerbations with different organisms.39 40 Eller et al39 found that if patients with better lung function were compared to those with worse lung function (measured by FEV1), the bacteria responsible for exacerbation shifted from pneumococcus and H influenzae to more complex organisms such as the Enterobacteriaceae and Pseudomonas species. Similarly, Miravitlles et al40 found that H influenzae and Pseudomonas aeruginosa were more common in patients with FEV1 values < 50% of predicted than in patients with better lung function. These findings suggest that patients with more advanced lung disease harbor different organisms and, thus, may need a different therapy than patients with milder disease. These concepts should be tested prospectively, and were not incorporated into the older antibiotic therapy trials.
Several studies have suggested that patient subsetting, for the purpose of choosing therapy during exacerbations, may be beneficial. Grossman et al41 studied the use of ciprofloxacin in 115 patients compared to usual antibiotic therapy in 107 patients with exacerbation. Although there were no differences in outcomes between the two groups when examined in aggregate, there was a subset of patients who benefited clinically and economically from the quinolone therapy. The subset that benefitted was the group with moderate or severe bronchitis plus at least four exacerbations in the preceding year. The benefits were even greater if additional risk factors were present, which included the following: > 10 years’ duration of illness; age > 56 years; and multiple medical comorbidities. In another study of a more complex population of exacerbation patients, Destache et al42 found that the use of "third-line" antibiotics (amoxicillin/clavulanate, ciprofloxacin, or azithromycin) was associated with fewer hospitalizations and longer time between exacerbations than the use of "first-line" therapy (amoxicillin, trimethoprim/sulfamethoxazole, erythromycin, or tetracycline). It is interesting that the first-line antibiotics, which had limited efficacy in this study, were the same agents that were used in the study by Anthonisen et al.37 Thus, the benefits seen in the study by Anthonisen et al,37 although undeniable, may have been even more dramatic if antibiotics had been chosen with patient characteristics in mind. Future studies still are needed to answer this question. In addition, new trials will need to account for the rising rate of antimicrobial resistance found among the common organisms that lead to exacerbation. Currently, many pneumococci are resistant to penicillins and other agents, while both H influenzae and M catarrhalis are commonly associated with ß-lactamase production.
In conclusion, we agree with Hirschmann that more information is needed about the role of bacteria and antibiotics in exacerbations of COPD. However, based on all available data, we think that bacteria are important and that antibiotics should be used for any patient with at least two of the cardinal symptoms of exacerbation. Future studies will be needed to better define disease pathogenesis and the roles of immunization and the immune response itself in the patient with COPD. Therapy studies that examine differences among patients and among antimicrobial agents also will be important. However, we should not be pessimistic about the data that exist, and we think that a constructive approach to this field is necessary rather than the nihilistic and negative view presented by Hirschmann.
Received for publication March 22, 2000. Accepted for publication March 23, 2000.
References
This article has been cited by other articles:
|
|
 |
|
  D. C. Look, C. L. Chin, L. J. Manzel, E. E. Lehman, A. L. Humlicek, L. Shi, T. D. Starner, G. M. Denning, T. F. Murphy, and S. Sethi Modulation of Airway Inflammation by Haemophilus influenzae Isolates Associated with Chronic Obstructive Pulmonary Disease Exacerbation Proceedings of the ATS, August 1, 2006; 3(6): 482 - 483. [Full Text] [PDF]
|
|
|
|
 |
|
 R Wilson, P Jones, T Schaberg, P Arvis, I Duprat-Lomon, P P Sagnier, and for the MOSAIC Study Group Antibiotic treatment and factors influencing short and long term outcomes of acute exacerbations of chronic bronchitis Thorax, April 1, 2006; 61(4): 337 - 342. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Wilson Treatment of COPD exacerbations: antibiotics Eur. Respir. Rev., September 1, 2005; 14(94): 32 - 38. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L. Johnston Overview of Virus-induced Airway Disease Proceedings of the ATS, August 1, 2005; 2(2): 150 - 156. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. L. Chin, L. J. Manzel, E. E. Lehman, A. L. Humlicek, L. Shi, T. D. Starner, G. M. Denning, T. F. Murphy, S. Sethi, and D. C. Look Haemophilus influenzae from Patients with Chronic Obstructive Pulmonary Disease Exacerbation Induce More Inflammation than Colonizers Am. J. Respir. Crit. Care Med., July 1, 2005; 172(1): 85 - 91. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Buhl and S. G. Farmer Current and Future Pharmacologic Therapy of Exacerbations in Chronic Obstructive Pulmonary Disease and Asthma Proceedings of the ATS, April 1, 2004; 1(2): 136 - 142. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. S. Niederman Does the Presence of Antibodies Justify the Use of Antibiotics in Exacerbations of Chronic Bronchitis? Am. J. Respir. Crit. Care Med., February 15, 2004; 169(4): 434 - 435. [Full Text] [PDF]
|
|
|
|
 |
|
 S. Sethi, N. Evans, B. J.B. Grant, and T. F. Murphy New Strains of Bacteria and Exacerbations of Chronic Obstructive Pulmonary Disease N. Engl. J. Med., August 15, 2002; 347(7): 465 - 471. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Niederman, S. Sethi, and F. Martinez Acute Exacerbations of Chronic Obstructive Pulmonary Disease Ann Intern Med, April 2, 2002; 136(7): 557 - 557. [Full Text] [PDF]
|
|
|
|
|
|
V. Snow Acute Exacerbations of Chronic Obstructive Pulmonary Disease Ann Intern Med, April 2, 2002; 136(7): 557 - 558. [Full Text] [PDF]
|
|
|
|
 |
|
 J. V. Hirschmann, T. F. Murphy, S. Sethi, and M. S. Niederman Bacteria and COPD Exacerbations Redux Chest, February 1, 2001; 119(2): 663 - 667. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
