Article Text


Analysis of sputum in COPD
  1. Department of Environmental Medicine and Public Health
  2. University of Padova, 35128 Padova, Italy
  3. Department of Medicine, Oncology and Radiology
  4. University of Modena and Reggio Emilia, Italy
  1. Dr P Maestrelli pieromaestrelli{at}
  1. Department of Environmental Medicine and Public Health
  2. University of Padova, 35128 Padova, Italy
  3. Department of Medicine, Oncology and Radiology
  4. University of Modena and Reggio Emilia, Italy
  1. Dr P Maestrelli pieromaestrelli{at}

Statistics from

A number of studies have suggested a pathogenetic role for airway inflammation in the induction of both chronic sputum production and chronic airflow obstruction in smokers.1 It is therefore important to characterise and quantify inflammatory changes in the assessment of subjects with chronic obstructive pulmonary disease (COPD). Assessment of inflammation may be achieved by different means including invasive methods such as bronchial biopsies, bronchoalveolar lavage (BAL) or examination of surgical specimens and non-invasive methods such as spontaneous or induced sputum.

The induction of sputum by inhalation of hypertonic saline is a safe, reliable, and relatively non-invasive method in COPD, provided the technique is performed in a standardised way and measures are used to prevent adverse reactions.2-4 Induced sputum differs from spontaneous sputum by having a higher number of viable cells and less squamous cell contamination.3 There are no differences between spontaneous and induced samples from patients with COPD or asthma in the total and differential cell counts, but there is poor agreement in the fluid phase components.5 6

Bronchial biopsies and BAL can be performed in COPD for investigative use according to the published recommendations.7 8 When the different methods of assessing airway inflammation are compared in the same subjects, a different profile of inflammatory cells is obtained depending on the compartment of the lung examined by each technique—that is, the lumen of the central airways using sputum analysis,9 the airway wall using bronchial biopsies, and the peripheral airways using BAL.10 By combining these techniques, integrated and comprehensive information on cell traffic and inflammatory processes in COPD at different levels of the airway can be obtained. Analysis of induced or spontaneous sputum has contributed to the identification of smokers susceptible to developing COPD, to the characterisation of the inflammatory process during exacerbations, and to the effects of intervention with anti-inflammatory drugs or by smoking cessation.

Characterisation of susceptible smokers

Cigarette smoking is the most important cause of COPD but only a few smokers will develop clinically overt disease.1Cigarette smoke stimulates alveolar macrophages and possibly epithelial cells to release inflammatory mediators such as leukotriene (LT)B4, tumour necrosis factor (TNF)α, and interleukin (IL)-811 12 that may induce influx of neutrophils into the lung.13-15 The mechanisms for the increased susceptibility in some smokers and the identification early in life of high risk smokers remain unknown. A longitudinal study of smokers examined whether the airway inflammatory process is different in smokers susceptible to developing COPD from that in “resistant” smokers. The percentage of sputum neutrophils was greater in smokers with COPD than in asymptomatic smokers and correlated with the annual decline in forced expiratory volume in one second (FEV1).16 In addition, sputum neutrophils in COPD exhibited increased expression of adhesion molecule CD11b/CD18, the ICAM-1 ligand, and this expression was related to the degree of airway obstruction.17 A number of subsequent studies confirmed increased airway neutrophilia in patients with COPD compared with healthy smokers.18-20 It is likely that neutrophils accumulate in the airway lumen by recruitment from the circulation but, because of the rapid migration across the tissue, their numbers in the subepithelial layer of the airways at any time point are low.21 The mechanisms for the airway neutrophilia in COPD are not entirely clear, but imbalance between pro-inflammatory and anti-inflammatory cytokines has been suggested to play a role. Concentrations of IL-10 are reduced, while increased concentrations of IL-8 and TNFα have been reported in the sputum of patients with COPD.11 19 Macrophages and lymphocytes which exhibited activation markers in COPD (CD25, VLA-1, HLA), as well as epithelial cells, may be the source of the cytokines that, in turn, increase expression of adhesion molecules (TNFα) and promote neutrophil chemotaxis (IL-8). Upregulation of E-selectin on vessels in the submucosa and increased expression of epithelial intercellular adhesion molecule (ICAM) 1 on basal epithelial cells in patients with COPD suggest a mechanism for recruitment of these cells from the circulation and for their migration to the airway lumen.22 Indeed, increased numbers of neutrophils were detected in the bronchial epithelium and the mucous glands of patients with COPD.23

Taken together, these data suggest that there is no clear qualitative difference—but rather a quantitative difference—between the inflammatory process in the airway lumen of patients with COPD and that of healthy smokers. This contrasts with asthma where eosinophils and eosinophil products represent the prominent inflammatory feature which is virtually absent in normal control subjects.

Eosinophilic bronchitis

A proportion of patients with COPD exhibited some degree of sputum eosinophilia.18 20 24 25 It is debatable whether the eosinophilia in COPD is related to concomitant features of asthma. Patients with COPD who responded to corticosteroid treatment or exhibited more reversibility of airway obstruction to salbutamol had significantly more eosinophils in their airways than those who did not.26 27 In contrast, patients with exacerbations of chronic bronchitis who exhibited airway eosinophilia were indistinguishable by common diagnostic criteria and pathological findings from those without eosinophilia.25 It has been suggested that the presence of eosinophils in the airways is related to the intensity of the inflammatory process in COPD, leading to non-specific recruitment of these cells18 24 and their activation.28 This is supported by the finding that the more severe impairment in FEV1 occurred in cases in which both sputum neutrophilia11 16 20 and sputum eosinophilia16 20 were more intense, with a direct relationship between neutrophil and eosinophil numbers.20In any case, the ability to detect eosinophilic bronchitis in patients with COPD may have clinical importance since this variant of COPD seems to respond to treatment with corticosteroids.29-31

COPD exacerbations

Patients with COPD are prone to exacerbations which are defined on clinical grounds by increased dyspnoea, cough, and sputum production that causes the subjects to seek medical attention. A proportion of patients with chronic bronchitis and mild airflow obstruction examined during exacerbations were found to have accumulation of neutrophils and eosinophils in sputum and their bronchial biopsy specimens exhibited increased numbers of mucosal eosinophils, similar in degree to that present in asthma.25 However, the influx of eosinophils in COPD was not associated with IL-5 expression, which suggests that a different mechanism induces eosinophilia in asthma and COPD.32 A recent study compared a large cohort of patients with severe COPD during an exacerbation with patients with stable COPD but failed to show any difference in sputum total and differential cell counts between the two groups.33 The only difference in the exacerbated group was a higher concentration of IL-6 in the fluid phase of the sputum. In even more severe patients with an exacerbation of bacterial origin, Crooks et al 34 found striking changes in the spontaneously induced sputum with increased levels of products from neutrophils (myeloperoxidase and elastase) and chemoattractants for neutrophils (IL-8 and LTB4). Moreover, the high proteinase inhibitor sputum to serum ratio suggested increased vascular permeability. The heterogeneity of the disease, the differences in baseline severity of patients, and the variable aetiology of exacerbations account for the discrepancies between the studies. Application to the analysis of induced sputum of molecular biology tools such as the polymerase chain reaction, which is much more sensitive than conventional tools, has been useful in assessing the aetiology of COPD exacerbations. In particular, the detection of rhinovirus nucleic acid in the induced sputum from a significant number of patients with COPD during clinical exacerbations,35 together with the expression of pro-inflammatory cytokines such as IL-6,36 suggests that COPD exacerbations may be at least partly due to an exaggerated inflammatory process primarily caused by a viral infection.

Effect of interventions

The question whether reducing risk factors—for example, smoking cessation and anti-inflammatory drugs—may reverse the inflammatory process in the airways of patients with COPD has been debated and has prompted a series of trials. However, direct assessment of airway inflammation has only been performed in a few studies.

Although it is established that smoking cessation can stop the progression of COPD, there is no evidence that the airway inflammatory process is decreased in ex-smokers compared with current smokers. Sputum neutrophilia and the degree of integrin expression on sputum leucocytes is similar in ex-smokers and current smokers.16 17 A significant increase in the percentage of sputum neutrophils and eosinophils, as well as in the concentrations of IL-8 and eosinophil cationic protein (ECP), has been shown in ex-smokers with COPD compared with ex-smokers without the disease.18 The persistence of granulocyte influx into the luminal compartment of the airways in ex-smokers is consistent with the persistence of the inflammatory process in the walls of the central airways seen in pathological studies.37 38

Treatment with oral and inhaled glucocorticosteroids for 2 weeks in a controlled study failed to modify the inflammatory indices in induced sputum of patients with COPD.39 A more prolonged study of treatment with 1.5 mg inhaled fluticasone propionate for 8 weeks in patients with COPD led to a reduction in chemotactic activity and an increase in elastase inhibitory capacity of spontaneous sputum sol phase, suggesting that fluticasone may reduce neutrophil recruitment to the lung and favour antiproteinases within lung secretions.40 This mechanism could explain the reduction in sputum neutrophils observed in an uncontrolled study of patients with COPD treated with 1.5 mg beclomethasone dipropionate for 8 weeks.41 The effect of steroids on COPD inflammation remains controversial except for their expected activity on eosinophils, when present.42 Further studies are required to determine the long term effects of drugs on lung secretions and to establish whether new treatments addressed to more specific targets will be effective.


Supported by the Italian Ministry of University and Research (MURST 40%) and by the Universities of Padova, Ferrara and Modena.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.