key: cord-0999858-xyw6nqmp
authors: Wedzicha, Jadwiga A; Seemungal, Terence AR
title: COPD exacerbations: defining their cause and prevention
date: 2007-08-30
journal: Lancet
DOI: 10.1016/s0140-6736(07)61382-8
sha: c904e69b9710894015642d967e2581ec3495b089
doc_id: 999858
cord_uid: xyw6nqmp

Exacerbations of chronic obstructive pulmonary disease (COPD) are episodes of worsening of symptoms, leading to substantial morbidity and mortality. COPD exacerbations are associated with increased airway and systemic inflammation and physiological changes, especially the development of hyperinflation. They are triggered mainly by respiratory viruses and bacteria, which infect the lower airway and increase airway inflammation. Some patients are particularly susceptible to exacerbations, and show worse health status and faster disease progression than those who have infrequent exacerbations. Several pharmacological interventions are effective for the reduction of exacerbation frequency and severity in COPD such as inhaled steroids, long-acting bronchodilators, and their combinations. Non-pharmacological therapies such as pulmonary rehabilitation, self-management, and home ventilatory support are becoming increasingly important, but still need to be studied in controlled trials. The future of exacerbation prevention is in assessment of optimum combinations of pharmacological and non-pharmacological therapies that will result in improvement of health status, and reduction of hospital admission and mortality associated with COPD.

Exacerbations of chronic obstructive pulmonary disease (COPD) impose a substantial burden on health-care systems worldwide; they are a major cause of morbidity, mortality, and reduced health status. 1 COPD exacerbations are now the most common cause of medical hospital admission in the UK (accounting for 15·9% of hospital admissions), at a cost to the National Health System of over £253 million a year. 2 Exacerbations are also important outcome measures in COPD, and thus a reduction in their frequency is a key target for intervention.

Although in half of community-treated exacerbations, patients recover to baseline symptoms by 7 days, a study of the time course showed that in 14% of these events patients had still not returned to baseline symptoms within 35 days of onset, and in a small proportion of exacerbations, symptoms never returned to the baseline level. 3 Thus COPD exacerbations can be quite protracted, which accounts for some of the considerable morbidity associated with such an event. An audit of hospital admissions showed that around 30% of patients presenting with an index exacerbation will be seen again and possibly readmitted with another (or recurrent) event within 8 weeks. 4 In a cohort of patients with moderate to severe COPD followed-up after exacerbation, 22% had a recurrent event within 50 days of the fi rst (index) exacerbation. Thus, such events are complex, and an initial exacerbation seems to increase susceptibility to a subsequent one. 5

An exacerbation of COPD is described as an acute worsening of respiratory symptoms associated with a variable degree of physiological deterioration. 3 The guidelines of the WHO and US National Heart Lung and Blood Institute Global Initiative for Chronic Obstructive Lung Disease (GOLD) 6 defi ne an exacerbation as "an event in the natural course of the disease characterized by a change in the patient's baseline dyspnoea, cough, and/or sputum that is beyond normal day-to-day variations, is acute in onset, and may warrant a change in regular medication in a patient with underlying COPD". 6 Defi nitions based on use of health care have also been proposed, eg, unscheduled physician visits, changes or increases in medication, use of antibiotics or oral steroids at exacerbation, and hospital admission. 7 However, health-care use in COPD can vary depending on access, and thus there could be substantial diffi culty in the standardisation of such a defi nition. Additionally, many COPD exacerbations are not reported to health-care professionals and are either self-treated or left untreated. 1 The latest GOLD guidelines also indicate that exacerbations can be self-limiting, especially if of mild severity, and the phrase "may warrant a change in regular medication" has been incorporated. 7 However, health-care use can be used to defi ne exacerbation severity: often defi ned as mild if increases in regular inhaled medication are needed, moderate if courses of steroids or antibiotics are needed, and severe if the patient requires hospital admission.

Exacerbations are usually infl ammatory events, with several airway and systemic infl ammatory markers increasing. 5 There has been substantial interest in developing a systemic biomarker as a diagnostic test for Search strategy and selection criteria We searched Medline ( -May, 2007 , Embase ( -May, 2007 , and CINAHL (1956 -April, 2006 with the search terms "acute exacerbation of COPD", "COPD exacerbation prevention", "COPD exacerbation impact". Thus we largely selected publications in the past 10 years but did not exclude commonly referenced and highly regarded older publications. We also searched the reference lists of articles identifi ed by this search strategy and selected those we judged relevant. Decisions to include specifi c references were based on authors' knowledge of published work, participation in expert meetings, and many years of research on the subject. Thus, this work is not an exhaustive systematic review of the burden and prevention of COPD exacerbations. However, it is an overview of current thinking on the importance of COPD exacerbations and how to prevent them based on research evidence and clinical practice. an exacerbation. Hurst and colleagues 8 did a large study of plasma biomarkers for COPD exacerbations, with 36 candidate molecules assessed in paired baseline and exacerbation plasma samples, and tested against a standard defi nition, meeting both health-care use and symptom-based criteria. To confi rm the diagnosis of exacerbation, the most selective biomarker was C-reactive protein (CRP) but this was neither suffi ciently sensitive nor specifi c alone. However, the combination of CRP with any increased major exacerbation symptom on that day (dyspnoea, sputum volume, or sputum purulence) increased the sensitivity and specifi city. Further research will aid in the understanding of mechanisms of COPD exacerbations and will lead to the development of more specifi c biomarkers.

COPD exacerbations are associated with increased upper and lower airway and systemic infl ammation (fi gure 1). 9-13 There is little information available on the nature of the airway infl ammatory changes especially when studied close to an exacerbation, because taking bronchial biopsies at exacerbation in patients with moderate to severe COPD is diffi cult. In stable COPD there is an increase in the CD8+ lymphocytes and macrophages in the bronchial mucosa and an increase in neutrophils with more severe disease. 14 In one study, where biopsies were done at exacerbation in patients with chronic bronchitis, increased airway eosinophilia was reported, although the patients studied had only mild COPD. 10 Modest increases were seen in neutrophils, T lymphocytes (CD3), and TNF alpha positive cells. However, in patients with more severe COPD, increases have been seen in airway neutrophils when stable that increase further at exacerbation. [10] [11] [12] Qiu and colleagues 11 have studied biopsies from patients with severe COPD, who were treated at exacerbation with tracheal intubation, and showed that there was pronounced airway neutrophilia, neutrophil elastase expression, and upregulation of neutrophil chemokine expression. However, studies in intubated patients with COPD are diffi cult, since the results can be complicated by secondary infection. Oxidative stress is a key factor in the development of airway infl ammation in COPD. A study has shown that patients with severe exacerbations who needed hospital admission or assisted ventilation have evidence of increased large airway interleukin-8 (IL-8) levels and increased oxidative stress. 12 Various markers of oxidative stress have been shown to rise in the airways with exacerbation such as hydrogen peroxide and 8-isoprostane and these markers can take some time to recover to baseline. 15 Upper airway infl ammation is increased in COPD patients, increases further at exacerbation, and is associated with lower airway infl ammatory changes. 13, 16 Systemic infl ammation increases at exacerbation and although the causes of this response in COPD are not clear, there is probably a spill-over of infl ammatory markers from the lungs. By contrast with stable disease, 16 exacerbations seem to be associated with a direct correlation between the degree of airway infl ammation and the size of the systemic acute-phase response. 13 Systemic infl am mation increases when the exacerbation is associated with bacterial and viral infection. 17 Several infl ammatory markers increase at exacerbation, such as plasma fi brinogen and CRP, that have been linked to increased cardiovascular risk. Respiratory infections have been associated with increased cardiac events 18 and thus a COPD exacerbation, especially if triggered by an infection, might also be associated with increased cardiac morbidity.

The airway infl ammatory responses during COPD exacerbations cause airway oedema, bronchospasm, and increased sputum production, leading to worsening airfl ow limitation and development of dynamic hyperinfl ation. 19 Such hyperinfl ation is the main cause of dyspnoea, the most common symptom of an exacerbation, and has other eff ects including modulating gas exchange, mechanical, and cardiovascular eff ects. 19 Generally the more severe the underlying disease, the greater the degree of physiological change at exacerbation leading to worsening of airfl ow limitation, and thus the more likely the patient is to develop respiratory failure.

Changes in airfl ow limitation result in changes in peak expiratory fl ow rate, but available data suggest that changes in peak fl ow are too small to be useful in 20 Novel physiological methods to monitor exacerbations are being developed, and some data now suggest that inspiratory capacity measurements 21 or within-breath forced oscillation measurements 22 might be more useful to assess COPD exacerbations and indicate the degree of physiological impairment.

COPD exacerbations are heterogeneous events that are now thought to be caused by complex interactions between the host, respiratory viruses, airway bacteria, and environmental pollution, leading to an increase in the infl ammatory burden (panel). 23

COPD exacerbations are frequently triggered by upper respiratory tract infections, which are more common in the winter months, when respiratory viral infections are prevalent in the community. Lung function also shows small but signifi cant falls with reduction in outdoor temperature. 24 Exacerbations triggered by respiratory viral infections are more severe and are associated with longer recovery times than those triggered by other factors. 25, 26 Molecular diagnostic techniques have now enabled detection of respiratory viruses at exacerbation, which have been isolated in around half of exacerbations, 27, 28 although this fi nding might be an underestimate due to diffi culties in sampling at onset of symptoms. There have been few such studies in the developing world, but a study from Hong Kong detected viruses in only 22% of exacerbations. 29 However, this study diff ered from others in the choice of assay and in that samples were taken from the upper airway rather than directly from the lower airway. A smaller study of 14 COPD patients admitted to hospital in Singapore reported that 64% of COPD exacerbations were associated with viruses although, like the Hong Kong study, most were associated with infl uenza virus, 30 which was possibly due to less frequent use of the infl uenza vaccine.

The most common viruses isolated are human rhinoviruses (the most frequent viruses associated with exacerbations), and other viruses including coronavirus, respiratory syncytial virus, infl uenza, parainfl uenza, and adenovirus. Since the introduction of infl uenza immunisation for patients with chronic lung disease, the virus has become a less prominent cause of exacerbation, though it is still likely to be an important factor at times of epidemics. Although respiratory syncytial virus infection has been seen at exacerbation, 31 whether it was the sole cause is not entirely clear, since this virus has been detected in the airways of COPD patients when they are stable and is associated with increased airway infl ammation in stable COPD. 32 Latent expression of adenoviral E1A protein in alveolar epithelial cells can amplify the eff ects of lung infl ammation induced by cigarette smoke. 33 Thus, chronic viral infection might be linked to disease severity in COPD and further work is required on the relation between viruses detected in the stable state and at exacerbation.

With PCR techniques, rhinoviruses can be recovered from induced sputum more frequently than from nasal aspirates at exacerbation, 26 suggesting that wild-type rhinoviruses can infect the lower airway and contribute to infl ammatory changes at exacerbation. Low-dose experimental rhinovirus infection in patients with mild COPD has been shown produce symptoms that are typical of an exacerbation, confi rming that respiratory viruses can infect the lower airway. 34 Exacerbations triggered by respiratory viruses are also more severe, associated with longer recovery times, 27 and have more chance of hospital admission than exacerbations where respiratory viruses were not detected. 35

The precise role of bacteria at COPD exacerbations has been diffi cult to assess, since airway bacterial colonisation in the stable state is associated with the same organisms as those isolated at exacerba tions, including Haemophilus Infl uenzae, Streptococcus Pneumoniae, Moraxella catarrhalis, Staphylococcus aureus, and Pseudomonas aeruginosa. 36 In one study in patients with moderate to severe COPD, bacteria were seen in 48·2% of patients in the stable state, rising to 69·6% at exacerbation, with an associated rise in airway bacterial load. 37 Purulent sputum prod uction has been regarded as a surrogate marker of bacterial infection, because COPD exacerba tions associated with purulent sputum are more likely to produce positive bacterial cultures than those where the sputum production was mucoid. 38, 39 Evidence for the involvement of bacteria has come from studies of antibiotic therapy, since exacerbations are often associated with increased sputum purulence and volume, and antibiotics have traditionally been used as fi rst line therapy. A study investigating the benefi t of antibiotics in more than 300 acute exacerba tions showed a greater treatment success rate in patients treated with antibiotics, especially if their initial presentation was with the symptoms of increased dyspnoea, sputum volume, and purulence, than in patients who did not receive such treatment. 40 Patients with mild COPD obtained less benefi t from antibiotic therapy than those with more severe COPD. The results of a meta-analysis have shown that antibiotic treatment off ered a small but signifi cant benefi t in treatment failure and mortality. 41 Substantial progress has been made on investigat ing the role of bacterial infection at exacerbation, with the development of molecular typing methods allowing the detection of changes in bacterial strains, rather than species. Sethi and colleagues 42 have suggested that isolation of a new bacterial strain in COPD patients who were regularly sampled was associated with an in creased risk of an exacerbation. However, this fi nding does not conclusively prove that bacteria are direct causes of exacerbations, because not all exacerbations were associated with strain change, and not all strain changes resulted in exacerbation. The strain-specifi c immune responses to colonising bacterial species 43 provide some further evidence that bacteria are not just innocent bystanders in the lower airways during exacerbations.

However, the situation with airway infection is further complicated, since in many COPD exacerbations both respiratory viruses and bacteria could be isolated. A greater systemic infl ammatory response has been reported in those exacerbations associated with both H infl uenzae and rhinovirus isolation, and if the isolation of H infl uenzae was associated with new or worsening coryzal symptoms (a surrogate of viral infection), such infections were more severe as assessed by changes in symptoms and lung function at exacerbation onset. 37 This change has been confi rmed in a further study which reported greater lung function impairment and longer hospitalisations in patients with exacerba tions associated with viral and bacterial coinfection than in those without coinfection. 44 Thus bacterial coinfection with viruses might be of greater importance than bacterial infection alone at COPD exacerbation but consensus within the fi eld has not yet emerged.

Atypical bacteria such as chlamydia, legionella, and mycoplasma have also been implicated at COPD exacerbation, although evidence on their role is confl icting, and these microorganisms might also interact with airway bacteria and viruses. [45] [46] [47] [48] A recent study using real-time PCR detection methods found no role for these three atypical bacteria at COPD exacerbation. 49 Pollution COPD patients can have increased exacerbations and hospital admissions with increasing environmental pollution. 50 In the APHEA study in six European cities, Anderson and colleagues 51 reported a signifi cant eff ect of air pollution levels on hospital admission for COPD, and similar results have also been seen in Taiwan 52 and Brazil. 53 However, common pollutants, especially nitrogen oxides and particulates, can interact with viral infection in asthma 54 to precipitate exacerbation rather than acting alone, and a similar mechanism might occur in COPD. A study from Hong Kong has shown adverse eff ects of ambient concentrations of air pollutants (sulphur dioxide, nitrogen dioxides, ozone, and particulate matter with an diameter of less than 10 µg/m³ [PM 10 ] and 2·5 µg/m³ [PM 2·5 ]) on hospitalisation rates for COPD, especially during the winter season. 55 Thus measures to improve air quality can have an eff ect on exacerbation frequency.

In general, exacerbations become both more frequent and more severe as the severity of the underlying COPD increases. 56 However, there remain large diff erences in yearly exacerbation incidence rates between patients of similar COPD severity. 1 There is no agreed defi nition of a patient with frequent exacerbations, but in several studies they were defi ned as those with yearly exacerbation rates of greater than the median for the study, usually around three symptom-defi ned exacerbations per year or two per year if the exacerbation is defi ned by the requirement for therapy with courses of antibiotics, corticosteroids, or both.

Patients with a history of frequent exacerbations have worse quality of life than patients with a history of less frequent exacerbations, and have consistent exacerbation frequencies when studied from year to year. 1 These patients also have an increased risk of hospital admission 57 and greater mortality (fi gure 2). 58 A UK study suggested that COPD exacerbations do not aff ect lung function decline, though this study was done in men of working age with fairly good lung function. 59 However, results of subsequent studies suggest that exacerbations do play a part in disease progression in patients who are active 61 although this eff ect is fairly small, at around 25% of the total decline in lung function. This observation might be due to the fact that not all exacerbations recover to baseline levels of symptoms and lung function. 3 The results of one audit showed that around 30% of patients seen at hospital with an index exacerbation will be seen again and possibly readmitted with a recurrent exacerbation within 8 weeks. 62 Patients with a history of frequent exacerbations also have increased airway infl ammation, 9 which could also contribute to the disease progression. 63 Why some patients have frequent exacerbations is not clear, although they possibly have increased susceptibility to respiratory viral infection. 25 Thus this group of COPD patients is important for targeting interventions that have the potential of reducing exacerbation frequency. COPD exacerbations have functional consequences. Spruit and colleagues 64 have shown that peripheral muscle weakness worsens during exacerbation, potentially contributing to reduced functionality and therefore to deconditioning (loss of fi tness). Patients who do not improve their walking distance within a month after exacerbation are more prone to be readmitted to hospital. 65 Donaldson and colleagues 66 have also shown that exacerbations are associated with a decline in outdoor activity for up to 5 weeks after the onset of symptoms. Patients who have frequent exacerbations had a faster decline in functional status, as measured by time spent outdoors, than patients with infrequent exacerbations (fi gure 3). Thus, patients with frequent exacerbations are more likely to become housebound and are a subpopulation that needs targeting for pulmonary rehabilitation programmes.

COPD is associated with other comorbid conditions, and patients who are admitted to hospital are more likely to have associated comorbid conditions such as ischaemic heart disease, pneumonia, and diabetes than patients without a diagnosis of COPD. 67 Increasing blood glucose concentrations have been shown to be associated with adverse clinical outcomes in patients with COPD exacerbations. 68 Thus the disease burden of COPD exacerbations is likely to be much higher when these comorbidities are taken into account. 67 There have been several reports of associations between pulmonary embolism, deep venous thrombosis, and COPD exacerbation. [69] [70] [71] Exacerbations could trigger pulmonary embolic events, since acute infections are known to predispose to deep venous thrombosis and pulmonary embolism. 72 There might also be diagnostic diffi culties because both COPD exacerbations and pulmonary embolism might present solely with dyspnoea. However, a recent study has shown that pulmonary embolism is not a common feature in uncomplicated exacerbations, 73 but some patients with exacerbations can have prolonged recovery periods, complicated by respiratory failure and comorbidity, when the risk of pulmonary embolism could become greater.

A few studies have shown that a decreased exacerbation rate is associated with improved quality of life. [74] [75] [76] [77] Thus lowering the exacerbation rate would be expected to decrease hospitalisations and have important health economic benefi ts. Several classes of drugs with potential for reducing exacerbations have been investigated, with variable evidence for their use: vaccines, bacterial extracts, inhaled steroids and long acting bronchodilators, phosphodiesterase inhibitors, and mucolytic agents.

There are several studies of infl uenza and pneumococcal vaccinations, which are now routinely recommended for all patients with COPD of signifi cant severity. [78] [79] [80] [81] One study that reviewed the outcome of infl uenza vaccination in a cohort of elderly patients with chronic lung disease found that infl uenza vaccination is associated with signifi cant health benefi ts with fewer outpatient visits, fewer hospitalisations, and reduced mortality. 78 A Cochrane database review of four studies in COPD saw no evidence of effi cacy for injectable antipneumococcal vaccines; 82 however, in a study of the 23 serotype pneumococcal polysaccharide vaccine in COPD patients, Alfageme and colleagues 83 showed that the vaccine was eff ective in the prevention of community acquired pneumonia, compared with placebo in patients younger than 65 years or those with severe airfl ow obstruction. However, no diff erence in mortality between the groups was seen in the study. Larger well designed studies are needed to examine the eff ects of pneumococcal vaccine in patients older than 65 years with COPD. Immunostimulatory agents have also been reported to reduce COPD exacerbation frequency. 10 years ago, a study of the immunostimulant OM-85, a detoxifi ed oral immunoactive bacterial extract, reported a reduction in the severe complications of exacerbations and hospital admissions in COPD patients, with a follow-up study confi rming the economic benefi ts of using this agent. 84 A recent randomised study has also found benefi ts but the patients had heterogeneous pathology. 85 A systematic review of 13 trials involving 2066 patients saw no consistent evidence of a benefi t though the agent is currently in use in Europe. 86 Further study is needed to understand the mechanisms of action of this immunostimulant before its role in COPD can be defi ned.

The success of oral corticosteroids in the treatment of COPD exacerbations with reduction of hospital length of stay 87, 88 has prompted much interest in the use of inhaled steroids to reduce exacerbation frequency in COPD. One of the early studies, the ISOLDE (Inhaled Steroid in Obstructive Lung Disease in Europe) study, was a 3-year study, powered to detect a signifi cant change in FEV 1 decline, was negative for the primary outcome but showed that exacerbation frequency can be reduced with inhaled steroids by about 25%. 74 Generally the eff ect of inhaled steroids was greater in patients with more impaired lung function, suggesting that this is the group to target with long-term inhaled steroid therapy. 89 In another study from the USA 90 (The Lung Health Study), the inhaled steroid (triamcinolone) group had signifi cantly fewer visits to a physician due to respiratory illness, suggesting that triamcinolone also reduced the frequency of COPD exacerbations.

Inhaled long-acting beta 2 agonist (LABA) therapy has been shown to cause small reductions in exacerbation frequency, although most studies involved short periods of therapy at 12 weeks. [91] [92] Mahler and colleagues 91 found that the time to the fi rst exacerbation was longer with therapy with the long-acting beta agonist salmeterol, although the overall number of exacerbations during the study was small. The larger TORCH study 93 of 6112 COPD patients reported that salmeterol reduced exacerbation frequency compared with placebo over a 3-year period. In another study comparing salmeterol with the short-acting ipratropium and placebo, there was no diff erence in the eff ect in either treatment arm on exacerbation frequency. 94 Rossi and colleagues 95 compared two diff erent dosages of the inhaled bronchodilator formoterol with placebo or theophylline and concluded that formoterol was more eff ective and better tolerated than therapeutically appropriate doses of oral slow-release theophylline in symptomatic COPD patients. Furthermore, two studies reported that formoterol had no eff ect on total exacerbation frequency relative to placebo, though these studies did fi nd that the rates of steroid-treated exacerbations were decreased in the formoterol group. 96, 97 The balance of evidence favours a positive role for LABAs in decreasing exacerbation frequency, which might be the result of the inhibitory eff ect of β 2 -adrenoceptor agonists on plasma exudation and neutrophil migration, or possibly indicating an additional reduction in the expression of adhesion molecules. However, the likely pathway of action could be through a synergistic action on airway infl ammatory cells in those patients already receiving inhaled corticosteroid therapy. 98, 99 All of the major recent studies of combination therapy 75, 93, 96, 97 have reported that the combination of inhaled steroid and LABA is more eff ective than either individual drug alone in reducing exacerbation frequency. Thus, inhaled steroids are unlikely to be used as sole therapy for COPD patients in the future. The data from the TORCH study have confi rmed the eff ectiveness of the inhaled steroid-LABA combination in reducing exacerbation frequency, and the study has reported that the combination reduces hospital admission rates in COPD patients. 93 However, most of these studies have been done in patients with more severe COPD with an FEV 1 at less than 60% predicted (fi gure 4). The eff ect of the inhaled steroid-LABA combination on reducing exacerbations in patients with milder COPD is not clear, especially in those patients who have an increased exacerbation frequency.

Long-acting anticholinergic agents such as tiotropium also reduce exacerbation frequency, and tiotropium has been shown to reduce exacerbations by 24%, compared with ipratropium when studied over a 1-year period. 76, 77 Niewoehner and colleagues 100 in a well designed study have confi rmed the potential of tiotropium to reduce exacerbation frequency and also have shown a reduction in hospitalisation. Tiotropium does not have any known anti-infl ammatory eff ect, 101 and its eff ect on reducing exacerbations is most likely due to the reduction of dynamic hyperinfl ation that is the major cause of dyspnoea in COPD (fi gure 5). 19 The combination of tiotropium with inhaled LABA and inhaled steroids has been explored in the Optimal Study. 102 This randomised trial compared tiotropium-fl uticasone-salmeterol versus tiotropium-salmeterol versus tiotropium-placebo, and the triple combination reduced hospitalisation as a result of exacerbation, but not the total number of exacerbations. However, a trend was seen in the reduction of the number of exacerbations with the triple combination, which failed to reach signifi cance possibly due to the small size of the study and the high dropout rate. Triple therapy could be more eff ective than dual therapy, but further studies of these combinations are required with adequately powered studies.

The phosphodiesterase inhibitors now form a class of non-steroidal anti-infl ammatory drugs that might be useful in the prevention of COPD exacerbations. Studies of the phosphodiesterase inhibitor theophylline have suggested that small reductions in exacerbation rates can be achieved with therapy, 95, 103 though further studies are needed on the eff ect of low dose theophylline on exacerbation frequency in COPD. Two new receptorspecifi c phosphodiesterase inhibitors have been studied in COPD: cilomilast and rofl umilast, which are both phosphodiesterase-4 inhibitors. A trial of cilomilast in COPD patients showed reduction in exacerbations in the cilomilast group. 104 Two studies of rofl umilast in COPD have been recently reported: Rabe and colleagues 105 have reported a reduction in exacerbations after 24 weeks therapy with rofl umilast, while Calverley and colleagues 106 studied rofl umilast in patients with GOLD stage III and IV in a 1-year trial and showed no overall eff ect on exacerbation rate, although patients with severe disease (GOLD stage IV) had fewer during the study period. Prevention of COPD exacerbations using this class of drugs is evolving and newer agents and more eff ective phosphodiesterase inhibitors will be developed with fewer adverse events. Tumour necrosis factor (TNF)-α has an important role as a key mediator in COPD and has been a target for study. However, a trial of the anti-TNFα antibody, infl iximab, showed no benefi t on any of the main trial outcomes and no eff ect of the therapy on exacerbation rate. 107 

The BRONCUS trial 108 using N-acetylcysteine has shown no overall benefi t of mucolytics on reduction of COPD exacerbations, except a small eff ect was noted in those patients who were not taking inhaled steroids. A follow-up meta-analysis of 26 randomised trials involving mucolytic therapy in COPD revealed a 20% reduction in exacerbations, with a large number of patients treated with mucolytics having no exacerbations. 109 A study from North America has shown that small airways occluded with infl ammatory exudate in COPD patients were associated with early death. This fi nding might stimulate further research into the role of mucolytic agents in COPD therapy. 110 However, the consensus view is that the evidence for mucolytics preventing COPD exacerbations is not convincing; this view is supported by the GOLD guidelines, although mucolytics are still used quite widely in some parts of the world.

Long-term antibiotic treatment has been used by physicians in clinical practice previously in patients with very frequent exacerbations, either continuously or in rotation, though there is little evidence for their eff ectiveness. There are some problems with using long-term antibiotics as resistant bacteria could emerge and cause increased airway infl ammation and exacerbations. However, in view of the presence of lower airway bacterial colonisation in these patients 111 and data from Patel and colleagues 112 that such colonisation is related to exacerbation frequency, there has been renewed interest in this type of intervention and some continuing studies will soon be reporting their fi ndings.

Pulmonary rehabilitation is now an accepted intervention in COPD and although it has important benefi ts for patients, its eff ect on preventing exacerbations is less clear. 113 In a randomised trial from South Wales, UK, Griffi ths and colleagues 114 treated with a pulmonary rehabilitation programme including exercise training and education had shorter hospital stays than the control group and fewer primary-care home visits. This fi nding suggests that a course of pulmonary rehabilitation might reduce exacerbation severity rather than frequency by increasing the patient's knowledge of COPD and how to access health care or self-manage during an exacerbation. Pulmonary rehabilitation could thus reduce hospital stay but also encourage early presentation for exacerbation therapy, which reduces exacerbation length and thus the severity of the event. 115 Garcia-Aymerich and colleagues 116 have also shown in a 1-year study that patients with high levels of usual physical activity were at reduced risk of readmission to hospital. An intensive disease-specifi c self-management programme done in Canada has been shown to reduce hospital admission rate, 117 though a systematic review of nurse-led interventions failed to show a consistent eff ect on hospitalisation). 118 Casas and colleagues 119 have extended this approach by using a similar integrated care plan in two diff erent environments (Barcelona, Spain, and Leuven, Belgium), with similar eff ects on decreased readmission rates for COPD exacerbation. Patients with COPD are elderly, often with a degree of cognitive impairment and might have diffi culty with selfmanagement at exacerbations. How optimum community support should be provided for patients who are at particular risk of hospital admission is not clear.

Long-term oxygen therapy has several benefi ts in COPD patients who are chronically hypoxaemic, including reducing mortality, anxiety, and depression. [120] [121] [122] In an epidemiological study, Garcia-Aymerich and colleagues 52 noted that patients with hypoxaemia but not treated with long-term oxygen therapy had a higher risk of hospital admission. Another observational study from the Danish Oxygen Register 123 has also suggested that long-term oxygen therapy reduces hospital admission rate. Controlled studies on home oxygen therapy and COPD exacerbations are diffi cult, since withholding therapy in a control hypoxaemic group would not be ethically justifi able.

COPD patients with chronic respiratory failure are particularly susceptible to exacerbations. After the early experience of domiciliary long-term non-invasive ventilation in patients with chest wall and neuromuscular disease, non-invasive ventilation has also been assessed in patients with hypercapnic COPD. Early observations on the eff ect of non-invasive ventilation in COPD showed a signifi cant benefi cial eff ect on health status, although data for exacerbations were not recorded. 124 Because health status is such an important determinant of exacerbation frequency, 1 improvement in health status could be due to a reduction of exacerbation frequency. In a controlled study lasting a year, no eff ect was seen of non-invasive ventilation on exacerbations but there was a reduction in hospital admission at the 3-month follow-up point. 125 Thus, larger controlled studies are now required to assess the eff ect of non-invasive ventilation on exacerbation in hypercapnic COPD patients, particularly at risk of hospital admission.

COPD exacerbations are often triggered by airway infection and are an important cause of morbidity, impairment of health status, and mortality. Although many pharmacological and non-pharmacological interventions prevent exacerbations, the degree of reduction in exacerbation frequency is still restricted and we now need new interventions to be urgently developed and studied in well designed and adequately powered randomised trials. Combinations of these interventions will probably be most eff ective and this approach will need future development and assessment. One of the main objectives of therapy for COPD is to reduce the morbidity associated with exacerbations and thus improve the quality of life of patients with this disabling condition.

JAW has received research grant funding from AstraZeneca, GlaxoSmithKline, and Boehringer Ingelheim, who all manufacture pharmacological therapies for prevention of COPD exacerbations. She has also received honoraria for lectures or attended advisory boards from AstraZeneca, GlaxoSmithKline, Boehringer Ingelheim, Bayer Healthcare, and Novartis Pharma. TARS has received honoraria for attendance of research meetings from AstraZeneca, GlaxoSmithKline, and Boehringer Ingelheim.

Airway eosinophilia in chronic bronchitis during exacerbations

Biopsy neutrophilia, neutrophil chemokin and receptor gene expression in severe exacerbations of chronic obstructive pulmonary disease

Oxidative stress and airway infl ammation in severe exacerbations of COPD

Systemic and upper and lower airway infl ammation at exacerbation of chronic obstructive pulmonary disease

The nature of small-airway obstruction in chronic obstructive pulmonary disease

Increased leukotriene B4 and 8-isprostane in exhaled breath condensate of patients with exacerbations of COPD

Relationships among bacteria, upper airway, lower Airway, and systemic infl ammation in COPD

Acute exacerbations of chronic obstructive pulmonary disease are accompanied by elevations of plasma fi brinogen and serum IL-6 levels

Risk of myocardial infarction and stroke after acute infection or vaccination

Eff ects of exacerbations and seasonality on exhaled nitric oxide In COPD

Lung mechanics and dyspnea during exacerbations of chronic obstructive pulmonary disease

Measurement of physiological recovery from exacerbation of chronic obstructive pulmonary disease using within-breath forced oscillometry

Eff ect of environmental temperature on symptoms, lung function and mortality in COPD patients

Epidemiological relationships between the common cold and exacerbation frequency in COPD

Detection of rhinovirus in induced sputum at exacerbation of chronic obstructive pulmonary disease

Respiratory viruses, symptoms and infl ammatory markers in acute exacerbations and stable chronic obstructive pulmonary disease

Respiratory viruses in exacerbations of chronic obstructive pulmonary disease requiring hospitalisation: a case-control study

Epidemiology of respiratory viruses in patients hospitalized with near-fatal asthma, acute exacerbations of asthma, or chronic obstructive pulmonary disease

Viral etiology of acute exacerbations of chronic obstructive pulmonary disease in Hong Kong

Detection of respiratory syncytial virus in adults with chronic obstructive pulmonary disease

Respiratory syncytial virus, airway infl ammation and FEV 1 decline in patients with COPD

Amplifi cation of infl ammation in emphysema and its association with latent adenoviral infection

An experimental model of rhinovirus induced chronic obstructive pulmonary disease exacerbations: a pilot study

Atmar RL Respiratory viral infections in adults with and without chronic obstructive pulmonary disease

Interactions between lower airway bacterial and rhinoviral infection at exacerbations of chronic obstructive pulmonary disease

Relationship to sputum colour to nature and out-patient management of acute exacerbations of COPD

Bronchoscopic validation of the signifi cance of sputum purulence in severe exacerbations of chronic obstructive pulmonary disease

Antibiotic therapy in exacerbations of chronic obstructive pulmonary disease

Antibiotics for exacerbations of chronic obstructive pulmonary disease

New strains of bacteria and exacerbations of chronic obstructive pulmonary disease

Human immune response to nontypeable Hameophilus infl uenzae in chronic bronchitis

Infections and airway infl ammation in chronic obstructive pulmonary disease severe exacerbations

Acute purulent exacerbation of chronic obstructive pulmonary disease and Chlamydia pneumoniae infection

Pneumoniae and chronic bronchitis: association with severity and bacterial clearance following treatment

Serological evidence of Legionella species infection in acute exacerbations of COPD

The role of atypical respiratory pathogens in exacerbations of chronic obstructive pulmonary disease

Health eff ects of an air pollution episode in London

Air pollution and daily admissions for chronic obstructive pulmonary disease in 6 European cities: results from the APHEA project

Air pollution and hospital admissions for chronic obstructive pulmonary disease in a subtropical city: Taipei, Taiwan

Respiratory and cardiovascular hospitalizations associated with air pollution in the city of Sao Paulo, Brazil

Personal exposure to nitrogen dioxide and risk of airfl ow obstruction in asthmatic children with upper respiratory infection

The temporal relationship between air pollutants and hospital admissions for chronic obstructive pulmonary disease in Hong Kong

COPD exacerbations: defi nitions and classifi cations

Risk factors of readmission to hospital for a COPD exacerbation: a prospective study

Severe acute exacerbations and mortality in patients with chronic obstructive pulmonary disease

The natural history of chronic bronchitis and emphysema

Lower respiratory illnesses promote FEV 1 decline in current smokers but not ex-smokers with mild chronic obstructive lung disease: results from the Lung Health Study

The relationship between exacerbation frequency and lung function decline in chronic obstructive pulmonary disease

Clinical audit indicators of outcome following admission to hospital with acute exacerbation of chronic obstructive pulmonary disease

Airway and Systemic infl ammation and decline in lung function, in chronic obstructive pulmonary disease

Muscle force during an acute exacerbation in hospitalised patients with COPD and its relationship with CXCL8 and IGF-I

Physical activity and hospitalization for exacerbation of COPD

Exacerbations, and time spent outdoors in chronic obstructive pulmonary disease

Comorbidity and mortality in COPD-related hospitalizations in the United States

Hyperglycaemia is associated with poor outcomes in patients admitted to hospital with acute exacerbations of chronic obstructive pulmonary disease

Frequency of venous thrombosis in patients with exacerbation of chronic obstructive lung disease

The frequency of deep venous thormobosis and pulmonary embolus in acute exacerbation of chronic obstructive pulmonary disease

Pulmonary embolism in patients with unexplained exacerbation of chronic obstructive pulmonary disease: prevalence and risk factors

Risk of deep vein thrombosis and pulmonary embolism after acute infection in a community setting

Should pulmonary embolism be suspected in exacerbation of chronic obstructive pulmonary disease?

Randomised, double blind, placebo controlled study of fl uticasone propionate in patients with moderate to sever chronic obstructive pulmonary disease: the ISOLDE trial

Combined salmeterol and fl uticasone in the treatment of chronic obstructive pulmonary disease: a randomised controlled trial

Improved health outcomes in patients with COPD during 1 yr's treatment with tiotropium

A long-term evaluation of once-daily inhaled tiotropium in chronic obstructive pulmonary disease

Relation between infl uenza vaccination and out patient visits, hospitalisation and mortality in elderly patients with chronic lung disease

Impact of a winter respiratory virus season on patients with COPD and association with infl uenza vaccination

Acute respiratory illness in patients with COPD and the eff ectiveness of infl uenza vaccination: a randomized controlled study

Impact of infl uenza vaccination on major cause-specifi c mortality

Injectable vaccines for preventing pneumococcal infection in patients with chronic obstructive pulmonary disease

Clinical effi cacy of anti-pneumococcal vaccination in patients with COPD

Eff ect of an immunostimulating agent on acute exacerbations and hospitalization in COPD patients

Swiss-German OM-85 Study Group. Double-blind study of OM-85 in patients with chronic bronchitis or mild chronic obstructive pulmonary disease

Clinical effi cacy of OM-85 BV in COPD and chronic bronchitis: a systematic review

Eff ect of systemic glucocorticoids on exacerbations of chronic obstructive pulmonary disease. Department of Veterans Aff airs Cooperative Study Group

Oral corticosteroids in patients admitted to hospital with exacerbations of chronic obstructive pulmonary disease: a prospective randomised controlled trial

The infl uence of disease severity and the eff ect of fl uticasone propionate on COPD exacerbations in the ISOLDE study

Eff ect of inhaled triamcinolone on the decline in pulmonary function in chronic obstructive pulmonary disease

Effi cacy of salmeterol xinafoate in the treatment of COPD

Use of long-acting inhaled beta2 adrenergic agonist salmeterol xinafoate in patients with COPD

Salmeterol and fl uticasone propionate and survival in chronic obstructive pulmonary disease

Long-term treatment of chronic obstructive pulmonary disease with salmeterol and the additive eff ect of ipratropium

Comparison of the effi cacy, tolerability and safety of formoterol dry powder and oral slow-release theophylline in treatment of COPD

Effi cacy and safety of budesonide/formoterol in the management of chronic obstructive pulmonary disease

Maintenance therapy with budesonide and formoterol in chronic obstructive pulmonary disease

Comparison of the anti-infl ammatory properties of formoterol, salbutamol and salmeterol in guinea-pig skin and lung

Eff ect of beta-agonists on infl ammatory cells

Prevention of exacerbations of chronic obstructive pulmonary disease with tiotropium, a once daily inhaled anticholinergic bronchodilator: a randomised trial

Eff ect of tiotropium on infl ammation and exacerbations in chronic obstructive pulmonary disease

Tiotropium in combination with placebo, salmeterol, or fl uticasone-salmeterol for treatment of chronic obstructive pulmonary disease: a randomized trial

Positive benefi ts of theophylline in a randomized, double-blind, parallel-group, placebo-controlled study of low-dose, slow-release theophylline in the treatment of COPD for 1 year

Cilomilast for COPD: results of a 6-month, placebo-controlled study of a potent, selective inhibitor of phosphodiesterase 4

Rofl umilast-an oral anti-infl ammatory treatment for chronic obstructive pulmonary disease: a randomised controlled trial

Eff ect of one year treatment with rofl umilast in severe chronic obstructive pulmonary disease

The safety and effi cacy of infl iximab in moderate to severe chronic obstructive pulmonary disease

Eff ects of N-acetylcysteine on outcomes in chronic obstructive pulmonary disease (Bronchitis Randomized on NAC Cost-Utility Study, BRONCUS)

Mucolytic agents for chronic bronchitis or chronic obstructive pulmonary disease

Survival following lung volume reduction in COPD: insights from small airway pathology

Association between airway bacterial load and markers of airway infl ammation in patients with chronic bronchitis

Relationship between bacterial colonisation and the frequency, character and severity of COPD exacerbations

COPD exacerbations. 4: Prevention

Results at 1 year of outpatient multidisciplinary pulmonary rehabilitation

Impact of reporting and early therapy on outcome of exacerbations of COPD

for the Estudi del Factors de Risc d'Agudització de la MPOC investigators. Risk factors of readmission to hospital for a COPD exacerbation: a prospective study

Reduction of hospital utilization in patients with chronic obstructive pulmonary disease: a disease-specifi c self-management intervention

Eff ectiveness of nursing and nurse-led chronic disease management innovations for patients with chronic obstructive pulmonary disease: systematic review of evidence

Integrated care prevents hospitalisations for exacerbations in COPD

Medical Research Council Working Party. Long term domiciliary oxygen therapy in chronic hypoxic cor pulmonale complicating chronic bronchitis and emphysema

Continuous or nocturnal oxygen therapy in hypoxaemic chronic obstructive lung disease

Does long term oxygen therapy aff ect quality of life in patients with chronic obstructive pulmonary disease and severe hypoxaemia?

Does long-term oxygen therapy reduce hospitalisation in hypoxaemic chronic obstructive pulmonary disease?

Nasal pressure support ventilation plus oxygen compared with oxygen therapy alone in hypercapnic COPD: a randomised controlled study

Long-term controlled trial of nocturnal nasal positive pressure ventilation in patients with severe COPD

We are very grateful to John Hurst for assistance in the preparation of this Review.