key: cord-0923185-yclnl87n
authors: Read, R. C.
title: Infection in acute exacerbations of chronic bronchitis: a clinical perspective
date: 1999-12-31
journal: Respiratory Medicine
DOI: 10.1016/s0954-6111(99)90048-3
sha: d1ead1c7266c24611fdb7d4a3f69fb94d083f01b
doc_id: 923185
cord_uid: yclnl87n

Abstract Acute exacerbations of chronic bronchitis (AECB) is an important cause of death and morbidity in developed countries and also has significant economic impact. The disease is characterized by increased dyspnoea, sputum volume and sputum purulence; the most commonly associated pathogens are Haemophilus influenzae, Streptococcus pneumoniae and Moraxella catarrhalis. H. influenzae and S. pneumoniae express virulence determinants that directly and indirectly impair mucociliary clearance and incite other consequences that are permissive to microbial persistence. Regarding the use of antibiotics, there is currently a lack of large-scale clinical trials that are sufficiently powerful to provide good evidence-based information. Nonetheless, antimicrobial chemotherapy has repeatedly been shown to confer benefit in patients with moderately severe features of AECB. Simple clinical criteria can be used to identify patients in whom there is a higher likelihood of treatment failure or mortality during AECB. These include significant cardiopulmonary co-morbidity, frequent exacerbations, advanced decline in lung function, malnutrition or other generalized debility, advanced age (> 70 years) and concurrent treatment with corticosteroids. In such patients, an aggressive antimicrobial approach to AECB may be warranted in order to prevent clinical failure or representation. From a clinical perspective, appropriate drugs include those that are stable to β-lactamases, are bactericidal against causative pathogens, penetrate diseased lung tissue in high concentrations and have a good safety profile.

(1999) 93, [845] [846] [847] [848] [849] [850] Topical Reviews Infection in acute exacerbations of chronic bronchitis: a clinical perspective Introduction Chronic obstructive pulmonary disease (COPD) is characterized by irreversible airflow obstruction due to chronic bronchitis or emphysema; this is defined as a ratio of forced expiratory volume in 1 set: forced vital capacity (FEVI: FVC) of less than 70%. Patients are subject to acute exacerbations in which there is an acute decline in FEV,, often in association with increased dyspnoea, and production of increased quantities of sputum which may be purulent. Recent epidemiological data from the U.S.A. show that COPD affects an estimated 50 million individuals and is the fifth most common cause of death (1) . Development of respiratory failure during an acute exacerbation of COPD occurs in those whose stable airway function is severely impaired (2) . However, respiratory failure is not the only cause of death in this group; as acute exacerbation of COPD can also result in death due to other causes, such as acute myocardial infarction (3) .

The role of infection in AECB is a contentious issue. Bacteria and neutrophils are frequently identified in the sputum of patients with AECB, but it is unclear whether they have a causal relationship with the development of exacerbations, or whether bacterial counts are merely allowed to increase as a result of the acute impairment of mucociliary clearance. Studies with antibiotics have revealed a positive, but small effect compared to placebo, in patients with AECB. In the interests of targeting therapy to those who need it most, there are some data that allow us to identify patients who do not benefit from antimicrobial therapy, and others for whom it is probably a mandatory component of management.

Within the trachea, bronchi and medium to small airways of the lung, the respiratory mucosa is normally sterile. During the evolution of COPD in an individual patient, there is a gradual increase in mucus production, depressed ciliary function and epithelial cell injury. Bacteria that are not normally present in the airway below the larynx are then able to colonize the diseased epithelium (4). The association of purulent sputum in chronic bronchitis with Haemophilus influenzae and Streptococcus pneumoniae was convincingly demonstrated by May (5) . These pathogens, of course, are common colonizers of the upper respiratory tract and they can be recovered from the lower airways even c 1999 HARCOURT PUBLISHERS LTD R. C. READ during stable COPD (6). Monso et al. (7) performed bronchoscopic sampling by protected specimen brush (PSB) of the lower airways of outpatients with stable COPD and compared this with patients with acute exacerbations of COPD. This showed that the prevalence of lower airway bacterial colonization in outpatients with stable COPD is high (25%), mainly due to H. infruenzae in S. pneumoniae, whilst exacerbated COPD is associated with a higher yield (51.7%) of these pathogens and higher numbers of organisms. In patients with more severe AECB, PSB will retrieve a wider range of bacteria including Gramnegative enteric bacilli, Pseudomonas sp. and Stenotrophomonas sp. (8) . Therefore at the very least, even if infection with bacterial pathogens is not the cause of acute exacerbations, they are clearly flourishing in the lower airways in some of these patients, in association with an inflammatory infiltrate (as evidenced by a yield of neutrophils in sputum). Molecular typing of strains of H. influenzae isolated from sputum of patients with COPD has demonstrated that a single strain may persist for many months within the lower respiratory tract mucosa of these patients, and remains present even if there have been acute exacerbations of COPD which have been treated with antibiotics (9). This suggests that Haemophilus infuenzae exists in a state of equilibrium with the host defences, which is lost during exacerbations. During antimicrobial therapy, the equilibrium is restored though the microbe is not eradicated. During chronic colonization there is likely to be a constant balancing act between the host and its bacterial colonizers. On the one hand, host defences include cough, ciliary beating, mucus containing secreted immunoglobulin and lactoferrin, and alveolar macrophages. On the other hand, microbes have adapted to avoid rapid clearance, gain nutrients and survive in the respiratory tract sufficiently to grow and disseminate to other hosts. In the case of Haemophilus injluenzae, the factors which enable the organism to achieve these aims include the production of a polysaccharide capsule, pili, lipopolysaccharide, outer membrane proteins, iron-regulated proteins and factors toxic to ciliary activity (reviewed by Read and Finch) (10).

The pathogens associated with exacerbations of COPD have been summarized by Ball (11) In summary, bacteria are present within the lower airways in most patients with stable COPD. In most cases of AECB, there is evidence of increased inflammation in the lower airways together with an increase in bacterial numbers. Whether the latter is cause or effect has not been elucidated. Table 2 . The studies assessed in this meta-analysis differed in terms of the main outcomes measured (symptom score, days of illness, etc.) so it was able only to report a crude overall effect size relative to placebo, and the effect on peak expiratory flow (PEF), which was reported in the majority of the studies. Overall, the meta-analysis demonstrated a modest but statistically significant benefit from antibiotic therapy (overall effect size 0.22 compared with placebo, 95% CI: 0.10-0.34). The peak expiratory flow improved by a small amount (10.75 1 min-' compared with placebo). None of the studies assessed included any quality of life outcomes. The largest of the studies assessed, by Anthonisen et al. (23) , stratified patients according to the severity of their illness and was able to demonstrate the largest effect size in those patients with the most severe exacerbations. Antibiotics gave the greatest benefit in terms of resolution of symptoms and avoidance of deterioration, in those patients who had two or more of the following features, increased dyspnoea, increased sputum volume and increased sputum purulence.

In Europe, the commonest antimicrobials used in AECB include the aminopencillins (U.K. and France), the macrolides (Spain), tetracyclines (Germany) and third generation cephalosporins (Italy) (25). Both the European Respiratory Society (26) and the American Thoracic Society (27) have produced guidelines for the management of COPD. These endorse the judicious use of simple and inexpensive antimicrobial therapy based on individual clinical assessment by physicians. The British Thoracic Society has recommended antibiotic therapy if any two of worsening dyspnoea, increased sputum volume or increased sputum purulence are present (28) . 

The major pathogen, Haemophilus inJEuenzae, is demonstrating rising resistance to the aminopenicillins which is mainly due to production of p-lactamases. In the U.K., prevalence of resistant H. infuenzae ranges from 424% (29) . Both of the major /?-lactamases are plasmid-mediated and transferable, and so their prevalence is expected to rise with increased use of aminopenicillins in the community. Erythromycin has virtually no activity against Haemophilus injluenzae, but clarithromycin has better activity, though this is based on a study which demonstrated that 96% of H. influenzae isolates had a MIC of 8 pg ml-' or less (29) . Azithromycin has better in vitro activity than the other macrolides (20). Moraxella catarrhalis demonstrates almost universal production of p-lactamases and is also resistant to trimethoprim (reviewed by Amyes and Gemmell) (31) . Resistance to p-lactams amongst Streptococcus pneumoniae isolates is due to a stepwise mutation of the penicillinbinding proteins. There is controversy regarding the impact of penicillin resistance of pneumococci on the management of patients with respiratory tract infection. Patients with pneumonia treated with high dose penicillin and subsequently shown to be infected with penicillin-resistant pneumococci, showed no increase in severity of outcome compared to patients with sensitive strains (32) . However, the effect of penicillin resistance amongst pneumococci on outcomes in community-treated AECB, where lower doses of oral agents are generally used, has not been investigated. Pneumococcal resistance is associated with childcare facilities, hospital facilities, antibiotic use and immunocompromise (33) . Importantly, long-term treatment with low doses of p-lactam antibiotics has recently been shown to contribute to the selective pressure promoting pharyngeal carriage of penicillin-resistant strains of S. pneumoniue (34) . The potential implications of this for patients with AECB treated in the community are obvious.

Identifying those patients most likely to benefit from antibiotic therapy With regard to patients presenting to their general practitioner with cough, it is possible to identify patients for whom no benefit will accrue from the use of antibiotics. These include patients with acute bronchitis and otherwise good respiratory health (15, 35) . A recent meta-analysis of double-blind placebo-controlled trials of antibiotics in patients presenting with acute cough demonstrated that neither resolution of cough nor clinical improvement was significantly hastened by antibiotic therapy (36) .

Regarding patients with COPD, the trial of Anthonisen et al. (23) showed that patients with more than two of (i) increased dyspnoea, (ii) increased sputum volume and (iii) increased sputum purulence can benefit from antibiotic therapy, whilst those patients with only one of these features are unlikely to benefit. With such a heterogeneous population, it is possible that positive benefit from antibiotics in the more severe subgroups has been diluted out in the studies of antibiotic therapy in patients with AECB. The majority of patients presenting to their community physician with AECB do not fulfil these simple clinical criteria and are therefore unlikely to benefit from antimicrobial therapy.

There is a strong case for targeting antibiotic therapy in patients with more severe manifestations of AECB. The major determinants of the survival over the long-term in patients with AECB are age and the degree of airways obstruction (FEVi < 50%) (37) . Co-existing cardiopulmonary disease such as ischaemic heart disease, and also diabetes, have been linked to poor outcome in COPD patients (38) . Use of oral corticosteroids has also been linked to poor outcome (39) . Seneff et al. (40) demonstrated that amongst patients admitted to intensive care units with AECB, variables associated with hospital mortality included age, severity of respiratory and non-respiratory organ system dysfunction, and hospital length of stay before intensive care unit (ICU) admission. These factors imply a worse prognosis of an infective AECB.

Although each of these factors is associated with a poorer outcome of COPD, their effects are not necessarily mediated through infection. Prescott et al. (41) investigated patients who had died from an AECB and determined whether their death was directly linked to infection on clinical grounds. The only historical risk factor that was associated with an infective 'death' was the presence of chronic mucus hypersecretion. Despite this reservation, a strong argument to target therapy at those patients with manifestations of disease associated with a poor prognosis can be made. Future studies should determine whether use of antimicrobial therapy in these patients affects clinical outcome, quality of life measurements and the likelihood of hospital admission. Of further importance, economically, and in terms of quality of life, is the failure of outpatient management resulting in patients needing to return to their physician. One study has determined that these patients can be identified. Those with co-existing cardiopulmonary disease, or who have a history of recurrent exacerbations (> 3 per year) are at greatest risk of failing therapy and needing a return visit (42) .

Appropriate antimicrobial therapy in patients with AECB Table 3 suggests appropriate antimicrobial therapy in patients with AECB. The majority of patients presenting to their general practitioner will not require antibiotic therapy. These are patients with acute bronchitis and simple chronic bronchitis with reasonable lung function and less than two of the three of the 'Anthonisen criteria'. Patients who are ambulant and relatively well with no co-existing illnesses, but who present with two or more 'Anthonisen criteria' can probably be managed by simple therapy with ,&lactams, for example an aminopenicillin, which is cheap and likely to be efficacious in this group. However, local data may point to a high prevalence of filactamase-secreting H. injkenzae, in which case therapy with antibiotics that are stable to p-lactamase is logical. These include co-amoxiclav, the quinolones and modern macrolides (clarithromycin or azithromycin). In patients with advanced age, or with cardiopulmonary comorbidities or moderately to severely impaired lung function (FEVi <50%), it would seem logical to treat infection with more sophisticated antibiotics in order to ensure microbiological clearance. Particularly in those patients with recurrent exacerbations, infection with a p-lactamase-secreting H. inJIuenzae is more likely. In addition, these patients are more likely to be infected with unusual pathogens such as the Gram-negatives such as Enterobacteriaceae and Staphylococcus aureus. In these patients the use of advanced antimicrobials such as co-amoxiclav, quinolones and modern macrolides would seem logical. This has no current evidence basis, and trials to determine whether they have an effect on clinical outcome should be conducted, though some have argued that this will be difficult (43) .

US Bureau of the Census. Statistical Abstract of the United States

Pathophysiology of Chronic Obstructive Pulmonary Disease

Acute Respiratory Tract Infections and Risk of First Time Acute Myocardial Infarction

Interaction of non-typable Haemophilus infruenzae of human respiratory mucosa in vitro

The bacteriology of chronic bronchitis

Bacteriological findings in trans-tracheal aspirates from patients with chronic bronchitis and bronchiectasis

Bacterial infection in chronic obstructive pulmonary disease

Bronchial microbial patterns in severe exacerbations of R.C. READ

The COPD Guidelines Group of the Standards of Care Committee of the BTS. BTS Guidelines for the Management of Chronic Obstructive Pulmonary Disease. Management of acute exacerbations of COPD

Antimicrobial susceptibility of community-acquired lower respiratory tract bacterial pathogens isolated in the U.K. during the 1995-1996 cold season

Spectrum and mode of action of azithromycin, a new 15-membered ring macrolide with improved potency against Gramnegative organisms

Antibiotic resistance

Resistance to penicillin and cephalosporin and mortality from severe pneumococcal pneumonia in

Management of infections caused by antibiotic resistant Streptococcus pneumoniae

Low dosage and long treatment duration of plactam

What will it take to stop physicians from prescribing antibiotics in acute bronchitis?

Quantitative and systematic review of randomised control trials comparing antibiotic with placebo for acute cough in adults

Prognosis in chronic obstructive pulmonary disease

The natural history of chronic bronchitis

Survival of patients with chronic obstructive pulmonary disease receiving longterm domiciliary oxygen therapy

Hospital and one-year's survival of patients admitted to intensive care units with acute exacerbations of chronic obstructive pulmonary disease

Chronic mucus hypersecretion in COPD and death from pulmonary infection

Acute infective exacerbations of chronic bronchitis

Is an objective assessment of antibiotic therapy in exacerbations of chronic bronchitis possible?