© 2003 American Thoracic Society
In Celebration of ExpectorationInduced Sputum Indices as Outcome Measures in Cystic FibrosisDepartment of Respiratory and Sleep Medicine Monash Medical Center Clayton, Victoria, Australia The lungs of a baby with cystic fibrosis (CF) are histologically normal at birth (1). However, infants with CF as young as 2 months of age can develop lower respiratory infection with marked inflammation, characterized by an influx of neutrophils and high concentrations of neutrophil elastase and the proinflammatory chemokine interleukin-8 (2, 3). Although initial episodes of infection can be eradicated with aggressive antibiotic treatment (3), persistent and exaggerated airway inflammation, together with chronic infection, eventually leads to progressive lung injury. Accurate measures of lower airway infection and inflammation are therefore essential, both to monitor disease progression, and to evaluate new therapies for CF lung disease. Unfortunately, neither of the currently used methods for monitoring lung infection and inflammation in CF is ideal. Spontaneously expectorated sputum is often contaminated with oropharyngeal flora, and is sometimes culture-negative in the presence of significant lower airway infection. Moreover, many patients with mild lung disease and most young children are unable to spontaneously expectorate sputum at all. Flexible bronchoscopy and bronchoalveolar lavage can be performed in any age group, although the procedure is invasive, time consuming, costly, and requires general anesthesia in children. These factors also mean that repeated procedures are very difficult to perform. In addition, although sometimes referred to as the "gold standard" for the diagnosis of lower respiratory infection in CF, bronchoalveolar lavage usually samples only one or two lung segments, which may reduce the ability to detect early lower airway infective and inflammatory changes, which may sometimes be focal (4, 5). Sputum induction by inhalation of hypertonic saline is a simple and direct method of obtaining lower airway secretions, and recent studies have shown the utility of induced sputum analysis in a wide range of respiratory disorders, including infection (6, 7), chronic obstructive airways disease (8), and asthma (9). Inhalation of hypertonic saline can be used to obtain sputum from most nonexpectorating patients with CF (10, 11), and provides a larger volume than achieved by spontaneous expectoration. Sputum induction can be performed repeatedly and is less invasive, quicker, and cheaper than bronchoscopy and lavage. A recent study has suggested that induced sputum indices may be useful as outcome measures for CF lung disease by showing a correlation between pulmonary function and measures of airway inflammation in induced sputum (11). In this issue of the Journal (pp. 1471–1475), Ordoñez and coworkers (12) extend these observations. They demonstrate a correlation between improvements in pulmonary function and changes in indices of infection and inflammation in induced sputum after intravenous antibiotics for a pulmonary exacerbation of CF lung disease. They recruited 72 individuals with CF (age range, 8–43 years) who were receiving intravenous antibiotics for an exacerbation of CF lung disease. Sputum induction was performed with 3% hypertonic saline via jet nebulization at the beginning and the end of the treatment course. Induced sputum samples underwent quantitative bacterial culture and measurement of inflammatory indices. Clinical and safety data were prospectively collected. Fifty-five subjects completed the study, including 25 who could not spontaneously expectorate sputum. Mean FEV1 improved by 0.3 L after antibiotic treatment. This improvement was associated with a 2–4 log reduction in lower airway bacterial density, and a 0.5 log reduction in airway inflammatory markers from induced sputum samples. Inhalation of hypertonic saline was associated with transient bronchospasm in a small number of patients. The correlation between changes in pulmonary function and induced sputum markers of lower airway infection and inflammation observed in this study suggests that these markers are valid outcome measures for CF lung disease. The repeatability of sputum induction is good, although the technique is more time consuming than obtaining spontaneously expectorated sputum, and subjects must be trained in the technique to produce an adequate sample (12). A major advantage of sputum induction, however, is that patients who are normally unable to expectorate can almost always produce sputum after inhaling hypertonic saline (10, 11). In the study of Ordonez and coworkers, use of sputum induction approximately doubled the size of the study group (12). The use of induced sputum as an outcome measure for future studies is therefore likely to increase subject recruitment, particularly nonexpectorating individuals with milder lung disease. Sputum induction using inhalation of hypertonic saline appears to be safe for individuals with CF. Although the study of Ordonez and coworkers was performed in the setting of an acute exacerbation of CF lung disease, wheeze occurred in 12 of the 70 subjects, and chest tightness in 6 of the 70. Although the maximum decrease in FEV1 was 46%, inhaled bronchodilator was immediately effective in all cases, and no subject had significant desaturation (12). Other studies using the same or higher concentrations of hypertonic saline for CF sputum induction have shown similar rates and severity of bronchospasm, even in subjects with poor lung function (13). It would appear that increasing age is more predictive of bronchospasm induced by hypertonic saline than a low baseline FEV1 (12, 13). The study of Ordonez and coworkers (12) thus adds to the increasing body of evidence supporting the use of induced sputum markers as valid surrogate outcome measures for future CF clinical trials. The technique is direct, noninvasive, and can be used repeatedly over time. Although bronchospasm was not a major problem in the present study, more information on the use of hypertonic saline in older individuals with CF, and those with low FEV1, is required before we can be confident in the safety of this technique in subjects with severe lung disease. Future studies are needed to determine the relative merits of different concentrations of hypertonic saline, and whether delivery via a jet or ultrasonic nebulizer is preferable. Although induced sputum samples can be obtained over a wide age range, flexible bronchoscopy and lavage will remain useful in infants and very young children who cannot cooperate with instructions for sputum induction. For all individuals over the age of 5 years, however, analysis of induced sputum is likely to become an important component of outcome measurement for future studies of novel therapies for CF lung disease. FOOTNOTES Conflict of Interest Statement: D.S.A. has no declared conflict of interest. REFERENCES
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
in induced sputum from patients with chronic obstructive pulmonary disease or asthma. Am J Respir Crit Care Med 1996;153:530–534.
