The Authors: Professor Kenneth W. Tsang is a consultant physician in respiratory medicine in private practice and honorary consultant to the Faculty of Medicine of the University of Hong Kong. His clinical and research interest focus on respiratory infection, particularly basic and clinical aspects of bronchiectasis. Dr Diana Bilton is a consultant physician at the Royal Brompton National Heart and Lung Institutes in London. She has major research interest in respiratory infection and a very busy clinical practice.
SERIES EDITORS: GRANT WATERER AND KENNETH TSANG
Kenneth W.T. Tsang, LKS Faculty of Medicine, The University of Hong Kong, 1202 Central Building, 1-3 Pedder Street Central, Hong Kong. Email: firstname.lastname@example.org
Bronchiectasis is a common disease in the Asia–Pacific and affected patients suffer from chronic sputum production and recurrent exacerbations. Bronchiectasis is largely idiopathic although there is diverse aetiology. The pathogenesis of bronchiectasis comprises infective, inflammatory and emzymetic elements. These interact to perpetuate continued airway damage in bronchiectasis leading to progressive airway and lung damages. Treatment of bronchiectasis is unsatisfactory and there are only very few trials. Existing data suggest some efficacy of inhaled corticosteroid therapy, which has been shown recently to clinical and anti-inflammatory properties in bronchiectasis. Immunomodulating agent such as low-dose macrolides have also been shown to have some efficacy although more data are needed to advocate their long-term usage. Antibiotic therapy is complex in bronchiectasis and includes short-term empirical treatment for acute exacerbation, and consideration of long-term maintenance of oral, nebulized and i.v. therapy. This long-neglected illness should receive more research attention in order that we can have better understanding of its aetiology, pathogenesis and treatment.
Bronchiectasis, defined pathologically as permanent dilatation of one or more bronchi, is not an uncommon disease in the Asia–Pacific. With the advent of CT, bronchiectasis is also increasing being diagnosed on Western patients among whom this was previously considered rare. Unlike the other inflammatory airway diseases such as asthma and COPD, bronchiectasis has not received much research attention, even in the Asia–Pacific where most of the patients are. Fortunately, similar to the rekindled interests in COPD, more attention has been paid to bronchiectasis over the last decade.1–7
Although the precise prevalence of bronchiectasis is unknown, due to a lack of cross-sectional community surveys, it is likely to be underdiagnosed with many of the sufferers being misdiagnosed as suffering from COPD, particularly if they have been smokers, or asthma.8 The official incidence of bronchiectasis is low or considered rare in the West, but considerably higher in more socioeconomically deprived regions with lower immunization rates. For instance, bronchiectasis was found to be common, with an estimated prevalence of approximately 1 in 6000 in the Auckland paediatric population, among the Maori children.9 In Hong Kong and many parts of Asia, despite high patient load, the exact prevalence of bronchiectasis is unknown. Hong Kong Government statistics show a hospital admission rate of about 16.4 per 100 000 of the population, and a mortality rate of 1 per 100 000 of population in 1990.10 As most patients with bronchiectasis are managed as outpatients and many cases are underdiagnosed, the prevalence is likely to be grossly underestimated.
FINAL COMMON PATH FOR DIVERSE AETIOLOGY
Bronchiectasis could result from numerous systemic and respiratory diseases (Table 1). Intensive evaluation of bronchiectasis patients leads to identification of one or more causative factor in 47% of cases, but only about 15% of patients have aetiological factors amenable to specific treatment or in making a prognosis.11 Nevertheless, it is imperative to evaluate the aetiology of bronchiectasis for any individual patient for diagnostic accuracy and prognostic values.
• Abnormalities of the tracheobronchial tree (such as tracheomegaly and polychrondritis)
○ Primary (pangammaglobulinaemia or selective Ig deficiency including IgG subclass deficiency)
○ Secondary (acquired immunodeficiency syndrome or malignancy)
• Miscellaneous (yellow nail syndrome and alpha-1-antitrypsin deficiency etc.)
A recent British series of 150 adult patients with bronchiectasis showed causes identified as: immune defects (12 patients), cystic fibrosis (CF) (4), Young's syndrome (5), ciliary dysfunction (3), aspiration (6), diffuse panbronchiolitis (1), congenital defect (1), allergic bronchopulmonary aspergillosis (ABPA) (11), rheumatoid arthritis (4), and early childhood pneumonia, pertussis or measles (44). Most Hong Kong patients, like their counterparts in the West, suffer from idiopathic bronchiectasis.12 Among 100 patients with bronchiectasis, 82% of patients suffer from idiopathic disease and 8% post-tuberculous. The latter contradicts previous belief that tuberculosis was a leading cause of bronchiectasis, probably largely due to an effective and free tuberculosis government service in Hong Kong. The remaining patients in this series had bronchiectasis secondary to previous pneumonia, primary ciliary dyskinesia and diffuse panbronchiolitis.12 The advent of effective vaccination against measles and pertussis, and effective antituberculous therapy probably account for the similarity between Hong Kong patients and their Western counterparts. Other necrotizing lung infections, including fulminant and inadequately treated pneumonia caused by Klebsiella and Staphylococcus, could also result in bronchiectasis. In areas with more deprived socioeconomic and health-care provision, bronchiectasis is likely a common complication of such infections.
The high proportion of patients with idiopathic aetiology reflects our poor understanding. Many patients with idiopathic disease have early-onset and slowly progressive disease, and have never smoked. It is likely that these patients exhibit poorly understood immunological mechanisms leading to progressive airway destruction and then pathological permanent dilatation of the bronchi consequently, similar to the situation in other organ-specific autoimmune diseases.12–17 It remains a great challenge to define these mechanisms, not only to prevent further deterioration for patients but also to identify patients who are likely to have rapidly progressive diseases for more intensive therapy.
Cystic fibrosis affects 1/2500 of Caucasians and is caused by a defect of the CF gene with the CF transmembrane conductance regulator located on chromosome 7q31.3.18 It is characterized by an increase in sweat chloride concentration, pancreatic insufficiency, and progressive and then lethal bronchiectasis. Classical CF, with pancreatic insufficiency and progressive lung disease, is extremely rare, if at all present, among the Orientals despite the presence of F508 mutation among indigenous Indians, Lebaneses and Pakistanis.19–21
Microscopic cilia are present on the mucosal surface of the upper and lower respiratory tracts, which beat in a coordinated fashion at 12–18 Hz, thereby creating a flow of mucus which continuously removes inhaled bacteria and airway debris, and maintaining the sterility of the lower respiratory tract.22 When cilia beat abnormally, such as in immotility or dyskinesia, either due to abnormalities or the ultrastructure or for unknown reasons, patients could suffer from sinusitis, bronchiectasis, deafness, infertility and sometimes situs inversus. Kartagener's syndrome (dextrocardia, sinusitis and bronchiectasis) is the most common subset of primary ciliary dyskinesia syndrome.23 Primary ciliary dyskinesia is rare and has an estimated incidence of 1 in 12 500.24 Kartagener's syndrome is most frequently associated with an absence of dynein arms (site of ATPase activity), which can be readily confirmed at transmission electron microscopy examination of cilia, conveniently sampled from the nose23,25 (Fig. 1).
Diffuse panbronchiolitis (DPB) is an idiopathic progressive suppurative and obstructive airway disease, which predominantly affects the Japanese.23,26 Patients classically present with wheezing, rapid development of severe and predominantly lower lobe bronchiectasis, and progression to respiratory failure and death within a few years if untreated. Similar to their counterparts with bronchiectasis and CF, patients suffer from airway infection by Haemophilus influenzae in early and then Pseudmonas aeruginosa later in DPB. Although included in the initial Japanese diagnostic criteria, the presence of the above-mentioned clinical features—sinusitis, obstructive spirometry, and HLA-B54 and HLA-A11—is not specific nor diagnostically useful for DPB.27,28 Patients have diffuse ‘tree-in-bud’ (centrilobular) nodules (<2 mm), which are symmetrical with lower lobe predominance, hyperinflation of the lung fields due to small airway obstruction, and bronchiectasis on high-resolution CT (HRCT) scan (Fig. 2).23,26,29 Characteristic patients could effectively be diagnosed to have diffuse panbronchiolitis based on CT findings and exclusion of small airway sepsis caused by atypical mycobacteria and other bronchiolitides.26 Untreated patients only have 5- and 10-year survival rates of 42% and 25.4%, respectively.30 However, administration of long-term low-dose erythromycin and other macrolides improves survival, symptoms and lung function, and leads to resolution of ‘tree-in-bud’ nodules. For instance, long-term erythromycin therapy (600 mg daily) improves the 10-year survival rate of DPB patients who had Ps. aeruginosa infection from 12.4% to more than 90%.31
Aspiration into the tracheobronchial tree is classically known to cause lower lobe bronchiectasis.32 Gastro-oesophageal reflux (GOR), which entails micro-aspiration of acidic gastric contents and often is clinically silent, is implicated in the aetiology of asthma, COPD, CF and pulmonary fibrosis.33–35 Up to 32% of patients suffer from upper gastrointestinal symptoms, and the presence of GOR or upper abdominal distension is associated with poor FEV1 and FVC in bronchiectasis.36 The use of upper and lower oesophageal pH monitoring, although permitting accurate assessment of reflux frequency, severity and duration,37 is too uncomfortable as a routine investigation in bronchiectasis. Foreign body aspiration, particularly bone fragment lodging in the right lower lobe bronchus, is not uncommon among the Chinese and other Orientals due to eating habits.38–40 Such foreign bodies could evade CT detection and could sometimes only be detectable at bronchoscopy.41
Traction bronchiectasis occurs commonly among patients with established pulmonary fibrosis, when bronchi are forcibly drawn from contracting surrounding parenchymal scar tissue. This is particularly common in post-tuberculous upper lobe bronchiectasis but could also develop in rapidly progressive interstitial pneumonitis and notably among survivors of severe acute respiratory syndrome who could develop severe fibrosis and traction bronchiectasis within weeks of disease onset.42 A recent study shows that traction bronchiectasis could occur in as high as 82% of patients with non-specific interstitial pneumonia.43
Post-transplantation recipients, including patients who have undergone heart–lung, lung and bone marrow transplantation, could develop bronchiectasis.44 The latter occurs as a complication of chronic graft-versus-host disease leading to bronchiolitis obliterans syndrome.45,46
Rheumatological disorders, particularly systemic lupus erythematosus and rheumatoid arthritis, are frequently associated with development of bronchiectasis and other lung diseases, sometimes independent of the associated pulmonary fibrosis and not uncommonly asymptomatic.17,47 Similar to rheumatological disorders, other organ-specific autoimmune disorders such as inflammatory bowel diseases,48 vitiligo49 and thyroiditis are also associated with development of bronchiectasis.50 Immunodeficiency syndromes including HIV infection are associated with an increased incidence of bronchiectasis.51 Respiratory disorders associated with development of bronchiectasis include tracheobronchial wall disorders such as Mounier–Kuhn syndrome (tracheomegaly) and ABPA. The latter is frequently detected among CF patients, but considered rare among Orientals with bronchiectasis.52 Miscellaneous conditions, such as yellow nail syndrome and alpha-1-antitrypsin deficiency, are rare among the Orientals.
The diversity in the aetiology of bronchiectasis not only makes it very difficult to formulate investigation plans, but also could potentially introduce frustrating and costly assessment, often to no avail. Clinicians, particularly those in the less developed parts of the world, are faced with immense challenge. It is imperative that we remain sensible and be guided by clinical features of patients before embark on further evaluation of these patients.
POORLY UNDERSTOOD PATHOGENESIS
Recent data on the basic and clinical research have revealed three important pathogenic and interactively components in bronchiectasis, namely infection, inflammation and enzymatic components of pathogenesis (Fig. 3).1 These components cause chronic and self-perpetuating destruction to the airways, leading to further deterioration in bronchiectasis.
Airways infection is well known in bronchiectasis and often misunderstood as the sole pathogenic element in the pathogenesis of bronchiectasis even among pulmonologists. H. influenzae is a Gram-negative coccobacillus that is the most frequently isolated pathogen in the sputum of patients with early bronchiectasis. In more advanced disease, and in particular patients with copious sputum production (24-h sputum volume >30 mL/h) and significant obstructive lung function (FEV1/FVC ratio <60%), Ps. aeruginosa becomes the predominant organism.53Ps. aeruginosa is a versatile Gram-negative bacterium, which is virtually impossible to eradicate despite intensive antibiotic therapy, and leads to recurrent exacerbations. H. influenzae and Ps. aeruginosa persist in the airways of patients with bronchiectasis in biofilms, producing harmful toxins, which damage mitochrondria and ciliary ultrastructure, and slow down ciliary beating.54–56 The diseased airways, therefore, will have further impairment in clearing these pathogens as the mucociliary clearance mechanisms, which maintain the airway sterility in health, will be progressively impaired in a vicious circle (Fig. 3).1
The role of inflammation in the bronchiectatic airway is now increasingly more understood. There are increased activated CD8 T lymphocytes, neutrophils and macrophages in the epithelial, lamina propria and submucosal layers bronchial mucosa in human bronchiectasis.13,16 Intense neutrophil infiltration, mediated by airway and pro-inflammatory mediators, including leucotriene B4, IL-1β, tumour necrosis factor-α and IL-8, occurs in bronchiectasis. The aforementioned mediators have also been shown to be upregulated both systemically and locally in the airways of patients with bronchiectasis.13,15,57 Leucotriene B4 promotes neutrophil migration and degranulation; IL-1β mediates airway inflammation and fibrosis; tumour necrosis factor mediates elastolytic degradation of lung protoglycans and interacts synergistically with IL-1 in prostaglandin induction; and IL-8 is one of the most potent chemoattractants, which also degranulates neutrophils in bronchiectatic airways.15
Serum levels of adhesion molecules, intercellular adhesion molecule-1, vascular adhesion molecule-1, and E-selectin are upregulated in patients with stable bronchiectasis,58 and these upregulate the migration of intravascular leucocyte traffic into the inflamed airways.57,58 Inflammation in the airways will lead to further damage to the integrity of the mucosal barrier, thereby encouraging bacterial local invasion and persistence (Fig. 3).1
In contrary to asthma and COPD, where airway inflammatory activities are associated with small airway destruction and narrowing, bronchiectasis is unique in displaying progressive airway dilatation in association with airway inflammation. Permanent and pathological dilatation of bronchi could only occur if destruction of airway luminal connective tissues overwhelms the repair process. Activated airway neutrophils release lytic enzymes release, such as elastase and matrix metalloproteinases (MMP), which are capable of digesting airway wall matrix. The sputum in bronchiectasis contains high levels of neutrophil products, such as elastase and superoxide radicals.59 Neutrophil elastase is a serine proteinase which leads to mucous gland hyperplasia, increased airway secretion, damage of the ciliated epithelium and acceleration of airway inflammation, and promotes bacterial colonization by destroying airway IgA and reducing its opsonophagocytic function.60–63 Clinically, elastase levels correlate with 24-h sputum volume, number of bronchiectatic lung lobes and sputum leucocyte density, and negatively with FEV1 and FVC in bronchiectasis.64
An upregulation of MMP-8, MMP-9 and MMP/MMP-2 cascade occurs in bronchiectasis, which is probably of neutrophilic origin.14,65 More recent data also point to the existence of MMP1 G-1607GG and MMP9 C-1562T gene variants, among patients with severe bronchiectasis.66 Although only poorly understood, these recent data strongly emphasize the role of lytic enzymes in the pathogenesis of bronchiectasis. These new findings also have clinical significance as MMP antagonists are already clinically approved for treatment of rheumatoid arthritis and other conditions.
NON-DIAGNOSTIC CLINICAL FEATURES
Classically, bronchiectasis is divided into four types, namely cylindrical, varicose, cystic and follicular pathological types, which are familiar to clinicians.67 Cylindrical bronchiectasis is characterized by uniform dilatation of bronchi, which extends into the lung periphery without tapering. Varicose bronchiectasis has irregular and beaded outline of bronchi, with alternating areas of constriction and dilatation. Cystic or saccular bronchiectasis occurs with air- and mucus-filled cysts originating from grossly dilated bronchi. Follicular bronchiectasis displays extensive lymphoid nodules formation within the bronchial walls, and usually develops after childhood infections. However, this pathological classification does not correlate with clinical features.68
With the advent of modern antibiotics, onset of childhood bronchiectasis has moved from the first decade of life, after severe pneumonia, to early adulthood or middle age in developed countries. There is a wide spectrum of clinical severity. Most patients with bronchiectasis suffer from chronic sputum production and cough, and a slowly progressive clinical course, which is usually punctuated by exacerbations (increase in sputum volume and purulence, and cough) and in some cases haemoptysis. Wheezing is reported in up to 34% of cases and may be due to airflow obstruction subsequent to gradual destruction of the bronchial tree or coexisting asthma in 10% of cases.69 Pulmonary function testing shows airway obstruction in up to 54% of non-smokers and more among smokers.70 Probably secondary to the ‘spilled-over’ systemic inflammation and chronic infection, weight loss and fatigue are frequently reported. Many patients also have chest pain, and pleurisy is not uncommon at the beginning of an exacerbation. Many patients suffer from rhinosinusitis.
Physical examination of the chest is often normal in bronchiectasis. More severe patients develop cachexia, crackles in chest, and possibly signs of respiratory failure in more advanced cases. Finger clubbing is commonly seen among Africans with bronchiectasis.71
As the symptoms of bronchiectasis are non-specific and very similar to those of COPD, many patients with bronchiectasis are misdiagnosed as suffering from COPD instead, particularly if they have had history of smoking. Among 110 patients attending general practice in the UK, who had confirmed ‘COPD’ by lung function and symptom evaluation, 29% were found to have bronchiectasis, on HRCT. Most importantly, 81% of these patients were current or ex-smokers thus provoking the clinicians to consider them as having only COPD, rather than bronchiectasis.72 In a separate British study, 50% of the 54 patients with clinically diagnosed COPD had bronchiectasis on HRCT, predominantly lower lobe in distribution. Those patients with bronchiectasis had higher levels of airway inflammatory cytokines and more severe ‘COPD exacerbations’ than their counterparts.73
There are four stereotypes of patients with bronchiectasis, which are increasingly recognized although still underevaluated and naturally overlapping:1
1Rapidly progressive—these unusual unfortunate patients suffer from early-onset and rapidly progressive disease leading to development of bilateral diffuse bronchiectasis, often cystic type, by their early 20s. They have frequent exacerbations, copious sputum production and general wasting. Some patients with ulcerative colitis have also been reported to show rapidly develop extensive bronchiectasis 1–2 years after colectomy.48,74
2Slowly progressive—most patients with bronchiectasis belong to this category and they appear to show slow but progressive deterioration with insidious increase in exacerbation frequency, sputum production and the extent of bronchiectasis over decades.
3Indolent disease—some fortunate patients with bronchiectasis, possibly those with right middle lobe bronchiectasis, appear to be asymptomatic and do not show progressive deterioration even after decades of follow up. These, therefore, have good prognosis.
4Predominant haemoptysis—some patients present predominantly with recurrent haemoptysis and often with little sputum production during steady state. Among 176 patients with bronchiectasis, patients with recurrent haemoptysis had lower 24-h sputum volume, higher FEV1, less cough, and were less likely to have a smoking history, than their counterparts (Tsang KW, unpubl. data, 2007). These patients often appear to have monosymptomatic haemoptysis, unprovoked by infections.75
NON-UNIFORM INVESTIGATION PLANS
Clinical diagnosis of bronchiectasis is difficult, if not impossible, as the clinical features of sputum production, recurrent exacerbations and occasional haemoptysis are non-specific and could also arise in COPD, tuberculosis, bronchial carcinoma, chronic rhinosinusitis, GOR, asthma and other mimicking diseases.8
In order to establish a diagnosis of bronchiectasis, there needs to be radiological proof of irreversible bronchial dilatation. Previous report on routine chest radiographs being abnormal in 90% of symptomatic patients with bronchiectasis is probably far too high for everyday clinical practice.76 Investigations on a patient with suspected bronchiectasis should initially aim to establish the diagnosis by demonstration of frank bronchial dilatation, before determination of the aetiology, and then activity and severity of the illness. Thereafter, the patient should also have assessment of sputum microbiology, which will help determine the antimicrobial therapy.
HRCT practically prerequisite in diagnosis
Plain CXR finding of ‘tram-lines’ reflecting underlying thickened and dilated bronchi is often unreliable for the diagnosis of bronchiectasis. The use of bronchography is largely phased out by the advent of HRCT,77 which is the current working gold standard for diagnosing bronchiectasis, with sensitivity and specificity of 98% and 99%, respectively.78–80 A more recent study about the role of HRCT in bronchiectasis concluded that small bronchiectases and bronchiolectases may not be visible on CXR and even conventional CT scan and that HRCT is required to confirm the diagnosis, with a high degree of accuracy.81 At HRCT, internal bronchial diameter larger than the adjacent pulmonary artery, lack of normal bronchial tapering and presence of peripheral airways within 1 cm of the costal pleura are diagnostic of bronchiectasis. Other features include bronchial wall thickening, fluid-filled bronchi, centrilobular nodules, and mosaic attenuation due to air trapping on expiration reflecting underlying small airway obstruction. Without the use of a good-quality CT scanner or clinical acumen, bronchiectasis is likely to be underdiagnosed or misdiagnosed, as reflected in the two community studies on COPD.72,73
Difficulties in attributing aetiology
Clinical and HRCT findings may provide clue on the aetiology of bronchiectasis, which is complex (Table 1). Attribution of the aetiology of bronchiectasis for a particular patient requires considerable clinical experience and expertise. For instance, a mere presence of features of previous occurrence of, say, tuberculosis, pneumonia, measles and pertussis etc. does not equate to these being the aetiology for a patient with idiopathic bronchiectasis. Some of the investigations required to delineate the exact aetiology of bronchiectasis, such as ciliary assessment, are not readily available outside highly specialized research units, whereas others are too expensive to use for all patients (Table 2). It is beyond the scope of this paper to exactly define the investigation paths for individual patients and readers need to be familiar with the aetiology of bronchiectasis in order to make an accurate diagnosis using minimal investigations.1 Nonetheless, it is definitely worth considering investigating a patient thoroughly at baseline to determine the aetiology of bronchiectasis.
Table 2. Diagnostic testing for bronchiectasis
High-resolution CT of thorax
Sinus CT scan
Complete and differential blood count
Ig level (IgG, IgM, IgA, IgE) and IgG subclasses
Sweat chloride test
Rheumatoid factor and autoantibodies
Sputum bacterial, mycobacterial, fungal cultures and sensitivity
The use of bronchoscopy has been underemphasized in bronchiectasis although many patients with bronchiectasis show bronchial dilatation, mucosal inflammation, bronchomalacia and presence of obliterative-like lesions at bronchoscopy (Fig. 4).82 Bronchoscopy is indicated for patients with new-onset haemoptysis and particularly in localized bronchiectasis to exclude an obstructive lesion, such as a benign tumour or even foreign body.39 The latter is of particular importance among the Orientals as complete resolution of bronchiectasis has been reported to occur after retrieval of a foreign body at bronchoscopy.83–85 The presence of centrilobular or tree-in-bud shadows warrants a bronchoscopy (BAL and transbronchial biopsy) to exclude atypical mycobacterial infection, diffuse panbronchiolitis and other bronchiolitides.26
Respiratory cilia could be obtained by brushing the respiratory mucosa, on either the nasal or tracheobronchial surfaces. Ciliated epithelium can be examined using light microscopy for beat frequency and motility, and transmission electron microscopy for ultrastructure.22 Functionally, ciliary beat could be normal, slow, dyskinetic or immotile.86 Transmission electron microscopy is used to assess the presence of abnormalities such as absence of outer dynein arms, typically seen in primary ciliary dyskinesia, and presence of other well-described abnormalities such as the cystic cilia that are associated with severe progressive bronchiectasis.25 Measurement of nitric oxide levels in exhaled breath condensate shows very low levels in primary ciliary dyskinesia, but its usefulness is doubtful among other patients with non-CF bronchiectasis.87–89 Despite the identification of a large number of genetic mutations, genetic testing is not very useful diagnostically for primary ciliary dyskinesia.90
Evaluation of the disease activity and severity
Disease activities markers are not commonly assessed, but can be evaluated clinically as the amount of sputum production (often expressed as 24-h sputum volume) and exacerbation frequency (often assessed as that over the previous 12 months).12,14,15,91 Clinical disease activity markers are important to assess as they could reflect better the intensity of the pathogenic components in bronchiectasis. Laboratory parameters, which related to clinical and pathological disease activities, such as sputum inflammatory mediator and MMP levels, leucocyte density, bacterial densities, purulence scale etc., are inconvenient to assess, clinically less applicable, and have no reference ranges available for daily application.92,93 Daily sputum volume has been shown to correlate with chemicals with significant pathogenic roles such as sputum elastase levels, serum ICAM-1 and serum endothelin-1 levels.64,57,58 Patients with 24-h sputum volume of ≥10 mL have significantly different St Georges' Respiratory Questionnaire-HK scores (i.e. quality of life measurement) than their counterparts with bronchiectasis. Exacerbation frequency is also important and correlates with the frequency of hospitalization, physician attendance and St Georges' Respiratory Questionnaire-HK scores.12,14,15,91,94
Severity refers to the extent of bronchiectasis and could be assessed by lung function testing as well as HRCT quantification of the structural damage of the lung. However, severity assessment should also include assessment of quality of life impairment using instruments, such as the St Georges' Respiratory Questionnaire. The latter assesses symptoms, activity and impact of disease on the individual patient, and might be a better measurement of the overall effect of bronchiectasis than simple bedside measurements of lung function.94
Pulmonary function studies may be normal in localized and mild bronchiectasis. With more severe and diffuse disease, pulmonary function tests may show obstructive or combined obstructive and restrictive abnormalities.76 Evidence of hyperinflation and reduced DLCO may also be seen in bronchiectasis. Between 30% and 69% of patients with bronchiectasis show airway hyperresponsiveness.95 Although readily available in most clinical respiratory units, lung function assessment is not an entirely satisfactory assessment of disease severity in bronchiectasis.
High-resolution CT can quantify disease severity according to extent of involvement based on either number of bronchopulmonary segments96 or percentage of lobar involvement.97–99 Bronchial wall thickening and mosaic attenuation appear to be related to lung function indices of obstruction (FEV1, FEF25–75% and FEV1/FVC) in bronchiectasis.97,98,100
Sputum microbiology in bronchiectasis
Unlike healthy non-smokers, the tracheobronchial tree in bronchiectasis is frequently colonized with potentially pathogenic microorganisms.101 The most commonly found microorganism was H. influenza (55%), followed by Pseudomonas species (26%) and Streptococcus pneumoniae (12%).12,102
Other important pathogens include Moxarella catarrhalis, Aspergillus and Mycobacterium avium complex (MAC). Staphylococcus aureus is relatively uncommon in non-CF bronchiectasis, and if repeatedly isolated, should alert the clinician to seriously exclude an undiagnosed CF.103 There are no data that define the role of viruses in acute exacerbations in bronchiectasis.
It is difficult to differentiate whether or not H. influenzae and Ps. aeruginosa are mere colonizers, rather than active infective germs, when a patient is ‘stable’. Not all patients with bronchiectasis who are colonized with Pseudomonas develop infection. However, some patients with repeated sputum isolation of Ps. aeruginosa actually develop anti-Ps. aeruginosa and anti-H. influenzae antibodies thereby suggesting a possible pathogenic role of these bacteria. Unfortunately, these antibodies are not only non-protective against Ps. aeruginosa and H. influenzae infection; they actually form immune complexes with bacterial antigens to trigger further chronic inflammatory changes in the respiratory tract and clinically correlate with a poor prognosis.104 Isolation of Ps. aeruginosa in patients with bronchiectasis was associated with high sputum volume and worse lung function (FEV1/FVC <60%).53
Non-tuberculous mycobacteria (NTM), being positive in culture on 2–10.2% of patients, are important agents causing colonization and disease in bronchiectasis.105,106 In addition, patients with NTM were much more likely to have positive Aspergillus serology and radiographic features of Aspergillus disease.107 NTM infections are increasingly recognized as important factors for deterioration of patients, perhaps particularly among those suffering from right middle lobe or lingula bronchiectasis.108,109 The main HRCT manifestations of Mycobacterium abscessus lung infection are bilateral small nodular opacities, bronchiectasis and cavity formation.110 MAC pulmonary disease with nodules and bronchiectasis is also increasingly recognized in Asia including Japan. A Japanese study using HRCT and bronchoscopic sampling showed that bronchial washing was more sensitive than the routine expectorated sputum for MAC isolation.108
LACK OF TREATMENT GUIDELINES AND PROVEN EFFICACY
There is definitely no cure, nor any proven efficacious therapy that is capable of reversing the permanent airway dilatation in bronchiectasis. There are very limited data addressing the treatment in bronchiectasis, stemming from a lack of research interest for this important debilitating condition. Treatment of bronchiectasis should aim to prevent or reduce the frequency of exacerbations, halt or slow down the continued airways and associated lung destruction, and maintain a good quality of life. However, there has been no treatment proven to be able to achieve any of these goals. In contrary to the wide availability of published guidelines on management of asthma, COPD, pneumonia and lung cancers from professional societies, such are absent for bronchiectasis. There have not been many clinical trials on bronchiectasis treatment and these generally only comprise pilot studies with small sample sizes. Nonetheless, life expectancy in bronchiectasis has improved tremendously from the reported mortality of 49%, probably due to the advances in antibiotic development, physiotherapy, general supportive measures and possibly surgical intervention.111,112
General supportive treatment
General considerations such as the adoption of good nutrition, regular exercise, no smoking strategy and exposure to fresh air are usually considered beneficial to respiratory patients. Pulmonary rehabilitation improves the health status of patients with moderate to severe COPD, and could be useful for patients with bronchiectasis.113 Patients should be monitored regularly for disease activity and severity by a team of physiotherapists, nurses, occupational therapists, psychologists and physicians.
Physiotherapy of the chest including postural drainage, chest percussion, forced exhalation and controlled cough was the main stay of treatment before the antibiotics became available. There is no guideline on the treatment techniques or outcome measures adopted for assessment.114 Patients, except those with virtually no sputum, should undertake regular chest physiotherapy. Postural drainage is particularly important for patients with primary ciliary dyskinesia as they have severely impaired mucociliary clearance for removal of airway secretions. Mechanical devices, designed to provide oscillation or percussion of the chest wall such as ‘oscillation jackets’, have not been shown to have efficacy in bronchiectasis.115 Chest physiotherapy improves pulmonary clearance, as measured by sputum production and radioisotope clearance,116,117 but there is insufficient evidence to support the routine use of chest physiotherapy in bronchiectasis.118 The use of aerosolized recombinant human DNase, approved for CF, actually increases exacerbation frequency and decline in FEV1 among patients with non-CF bronchiectasis.119
Polysaccharide pneumococcal vaccine is recommended for individuals between 2 and 64 years with increased risk of complicated pneumococcal disease, including those who have chronic pulmonary diseases such as bronchiectasis.120 Pneumococcal vaccine is efficacious in preventing invasive pneumococcal disease among all adults, and pneumococcal pneumonia among individuals over the age of 65 years.121,122 Both influenza A and B viruses lead to acute respiratory illness, with high rate of complications, in patients with underlying lung diseases such as bronchiectasis. Vaccination of the elderly can reduce complications or death by 70–85%.122 Influenza vaccine is, therefore, recommended for patients with bronchiectasis.123
Very severely affected patients may benefit from long-term oxygen therapy for respiratory failure and non-invasive positive pressure ventilation using BiPAP.124 The latter could be associated with increased airway mucus stagnation and should be accompanied by intensive chest physiotherapy.
Heart–lung or double-lung transplantation could be successful in severe bronchiectasis with 2-year survival rates of 50–60%. Despite strong advocates from some series on patient survival after surgery, resection surgery in patients with bronchiectasis should only be indicated when medical treatment fails for a highly localized bronchiectasis, in the presence of an obstructing tumour, or in occurrence of life-threatening complications such as uncontrolled haemorrhage.102,125,126 Some patients have symptomatic improvement in cough and sputum production, after resection of localized segment, although often only temporary.127 It must be remembered that resection of a bronchiectatic lung segment could cause further compromise of lung function, and other segments may still develop bronchiectasis, particularly for patient with underlying organ-specific autoimmune disorders or idiopathic disease.
Globulin replacement is efficacious in reducing infections in hypogammaglobulinaemia, and systemic steroid is indicated in ABPA to control the inflammation responsible for the development of asthma and bronchiectasis. Although the principles and practice of gene transfer have been established, gene therapy for CF is still in experimental stages despite initial enthusiasm.128 Although patients with bronchiectasis are frequently treated with long-acting β2 agonists, there are no controlled data to show any efficacy.129 Leucotriene receptor antagonists downregulate neutrophil-mediated inflammation, but have no data to support their efficacy in bronchiectasis.130 Similarly mucolytics have no proven efficacy in bronchiectasis and have potential harmful effects. Antitussive therapy could also be harmful as expectoration of sputum is a protective mechanism.131
Difficulties in the choice of antibiotics
Treatment of infection does not only fail to eradicate established infection with Ps. aeruginosa and H. influenzae, but also is disappointing in curing bronchiectasis as there are other pathogenic elements.132,133 Antibiotics are effective in treatment of acute infective exacerbations by reducing sputum volume and purulence leading to clinical improvement.61,134 Antibiotic therapy reduces sputum microbial load, neutrophil elastase and albumin level and airway responsiveness, although only temporarily.61,135–137
The choice of antibiotics to treat bronchiectasis is difficult as many patients have no pathogens isolated in their sputum even during exacerbations (deterioration in respiratory symptoms, with or without fever or other systemic disturbances). Mere isolation of ‘respiratory pathogens’ in the sputum does not equate to their pathogenic role as some are simply ‘colonizers’ of the bronchiectatic airways. In addition, H. influenzae and Ps. aeruginosa are also frequently isolated in the sputum even when there is no exacerbation. Total eradication of airway bacteria is seldom achieved despite prolonged intensive antibiotic treatment, and the goal of treatment should be to reduce the total microbial load. This reduction in microbial load is beneficial and might also reduce the harmful airway inflammation.64,138 Treatment with antibiotics should be of adequate dosage and duration, as penetration into the damage airways to reach the intraluminal pathogens would be slower and less effective than normal airways.1
As sputum culture will take days to provide results, and thus only provides retrospective information in the initial encounter with a patient with exacerbation of bronchiectasis, empirical choice of antibiotic therapy is usual. Empirical use of antibiotics depends on a knowledge of the likely causative bacterial pathogens, statistically most likely to be H. influenzae or Ps. aeruginosa.53 FEV1/FVC of <60% and sputum volume of >20 mL are independently associated with presence of Ps. aeruginosa in the sputum of patients with stable bronchiectasis.53 It is logical to deploy non-pseudomonal antibiotics for patients who usually produce little sputum and enjoy good lung function. More severely affected patients will require antipseudomonal antibiotics for treatment. Non-pseudomonal antibiotics includes third-generation cephalosporins (usually i.v. administration), and β-lactams with lactamase inhibitor such as Augmentin or macrolides (also in oral preparations). Antipseudomonal antibiotics include ceftazidime or aminoglycosides (i.v.), and quinolones such as levofloxacin and ciprofloxacin (oral).91
Severely affected patients with copious sputum production (e.g. >60 mL/day) even between exacerbations, or those with frequent exacerbations (e.g. more than 6 times/year), will need more intensive therapy. The use of regular elective administration of i.v. antibiotics, long-term nebulized antibiotics and long-term oral antibiotics could be effective.139 Published data on the use of antibiotics for between 4 weeks and 1 year were evaluated recently.139 Six trials were included, in which antibiotics were given for 4–52 weeks, and 302 patients were randomized. Better response rates and lung function were found patients receiving prolonged antibiotic treatment, although not for exacerbation rates. There is some benefit for the use of prolonged antibiotics in the treatment of some patients with bronchiectasis,139 despite concerns on development of antibiotic resistance, adverse reactions, cost consideration and emergence of antibiotic hypersensitivity.
Patients who do not improve on the above-mentioned regimen may benefit from long-term nebulized antibiotic therapy.140 It is attractive to use inhalational antibiotic therapy in bronchiectasis, particularly in light of the efficacy of such in CF, to avoid systemic administration. Inhaled tobramycin and colistin therapy is safe and reduces the decline in spirometry and improves the quality of life of patients.140–144 Inhaled tobramycin, in addition to ciprofloxacin, however, only appears to improve microbiological outcomes, but not clinical parameters in acute exacerbations of bronchiectasis.145 Unlike the CF experience, there does not seem to be an improvement of pulmonary function after treatment with aerosol tobramycin in bronchiectasis patients.146 Inhaled ceftazidime and tobramycin therapy, given over 12 months, reduces the hospitalization although there is no effect on lung function, antibiotic usage or antibiotic resistance for Ps. aeruginosa.144
The use of long-term oral antibiotics should not be encouraged unless the above-mentioned measures have clearly failed in reducing sputum volume and exacerbation frequency of a severely affected patient. Long-term oral antibiotics, such as high-dose amoxicillin, has also been used for very severe cases. Administration of amoxycillin (3 g) over 32 weeks was shown to reduce in severity of exacerbation (but not frequency), sputum purulence, but not sputum bacterial flora.147 Long-term use of ciprofloxacin is associated with rupture of tendons and is no longer recommended.148
The pitfalls and attractiveness of immunomodulation therapy
The presence of an intense airway inflammation, reflected clinically by sputum production and readily measured using surrogate markers such as sputum inflammatory mediators and elastase (and other protease), and exhaled nitric oxide levels makes it logical to attempt to ameliorate the clinical illness with immunomodulation therapy.1 As no data exist on the long-term efficacy or safety of these and anti-inflammatory agents or other immunomodulating agents such as macrolides, these should not be used routinely. Although oral steroids are recommended for bronchiectasis arisen from ABPA,149 there are no controlled trial data to define the role of oral steroid therapy in acute or stable bronchiectasis.150
The safety and efficacy of inhaled corticosteroid (ICS) therapy in the treatment of airway inflammation in asthma is well established.151 Studies have shown some, but not impressive, efficacy on ICS therapy on patients with bronchiectasis,15,129,130,152 despite its proven efficacy in asthma and in some cases of COPD.153 Only a few studies using inhaled15,92,152,154–156 and systemic157–159 corticosteroid have been conducted in bronchiectasis. Alternate day prednisolone (1–2 mg/kg) improved physical growth, spirometry and serum IgG, and IL-2 and IL-1α in CF children,158,160 but was associated with development of glycosuria and pneumothoraces.157 CF patients infected with Ps. aeruginosa and treated with prednisolone (1 mg/kg/alternate day) for 4 years had improved FVC, although a higher dosage (2 mg/kg/alternate day) was associated with increased incidence of diabetes mellitus, growth retardation and cataracts.159 The first ICS therapy on bronchiectasis showed that 16-week inhalation of beclomethasone (0.4 mg/day) had no effects on sputum albumin concentrations, proteolytic activities or immune complex activities after 16 weeks of treatment.154 Another two studies using a higher dosage (1.5 mg/day) of beclomethasone showed improvement of sputum volume152 and spirometry after treatment.156 Similarly, inhaled budesonide (1.6 mg/kg) improved bronchial hyperreactivity, and cough and dyspnoea indices after 16 weeks of treatment.155 Inhaled fluticasone (1 mg daily) appears to be efficacious in reducing sputum chemokine levels in stable bronchiectasis.15 A double-blind placebo-controlled randomized study on 12-month administration of fluticasone on 86 patients with stable bronchiectasis showed that treatment with fluticasone was associated with significantly more patients showing improvement in 24-h sputum volume, but not exacerbation frequency, FEV1, FVC or sputum purulence score. In addition, patients with Ps. aeruginosa infection show significant improvement in 24-h sputum volume and exacerbation frequency.92 It therefore appears that ICS therapy is probably efficacious among some patients with more severe bronchiectasis, although there are no sufficient data for patient selection for such therapy.
There are ‘immunomodulatory’ effects of macrolides on the immune response, defined as suppressing hyperimmunity and inflammation without overt immunosuppression.161 There are small case series and three large randomized controlled trials that have established unequivocal evidence of benefit in CF.162 Low-dose erythromycin is highly efficacious in diffuse panbronchiolitis, which, similar to bronchiectasis, is characterized by copious sputum production, rhinosinusitis, progressive airway destruction and chronic Ps. aeruginosa infection of the airways.28
Several small clinical trials have assessed the effects of macrolides on bronchiectasis, and reported improvement in clinical status, lung function parameters, lung inflammatory markers and sputum volume.2 Administration of low-dose erythromycin (500 mg bd), over 8 weeks, significantly reduces sputum volume and improves lung function in stable severe bronchiectasis.163 A 7-day trial of clarithromycin in patients with bronchiectasis showed reduction in sputum volume.164 A 3-month trial of 17 children with bronchiectasis with clarithromycin was associated with a decrease in IL-8 levels, total cell count, neutrophil ratios in BAL fluid and daily sputum production, but not lung function parameters.165 Existing data on the possible efficacy of macrolides in bronchiectasis are, therefore, only from very small and short-term trials, and not conclusive. Macrolide therapy in bronchiectasis needs to be meticulously considered before introduction to patients, probably those with active severe and progressive disease.
The mechanism of action of macrolides in bronchiectasis is not precisely known although it is unlikely to be bactericidal in view of the low dosage and their poor penetration into the tracheobronchial tree. Respiratory pathogens such as Ps. aeruginosa and H. influenzae produce exotoxins that slow human ciliary beat and cause ultrastructural damage in vitro.54,166,167 In addition, erythromycin inhibits glycoconjugate release168 and macrophage mucus secretagogue production,169 which would result in decreased sputum water contents and volume. Erythromycin affects the bacterial–host interaction in vivo54 by altering bacterial morphology and Ps. aeruginosa exotoxin production,166 and inhibition of neutrophil migration,170 IL-1 production by macrophages,171 and neutrophil production of elastase, myeloperoxidase and collagenase.172 Clarithromycin can also disrupt airway bacterial biofilm, which is one of the mechanism for bacterial persistence in the bronchiectatic airways.55,173