Eur Respir J 2005; 26:904-908
Copyright ©ERS Journals Ltd 2005
Effect of airway clearance techniques on the efficacy of the sputum induction procedure
M. R. Elkins1,2,
T. Lane3,
H. Goldberg4,
J. Pagliuso1,
L. A. Garske3,
E. Hector2,
L. Marchetto2,
J. A. Alison1,2 and
P. T. P. Bye1,2
1 Royal Prince Alfred Hospital, 2 University of Sydney, and 4 Prince of Wales Hospital, Sydney, and 3 Princess Alexandra Hospital, Brisbane, Australia.
CORRESPONDENCE: M. R. Elkins, Dept of Respiratory Medicine, Level 11, Edinburgh Building, Royal Prince Alfred Hospital, Missenden Road, Camperdown NSW 2050, Australia. Fax: 61 295158196. E-mail: elkinsm{at}med.usyd.edu.au
Keywords: Chest physiotherapy, induced sputum, pneumocystis, tuberculosis
Received: March 21, 2005
Accepted July 3, 2005
 |
ABSTRACT
|
|---|
Sputum induction is used in the early identification of tuberculosis (TB) and pneumocystis infections of the lung. Although manual physiotherapy techniques to clear the airways are often incorporated in the sputum induction procedure, their efficacy in this setting is unknown.
This randomised, crossover trial enrolled adults referred for sputum induction for suspected TB and pneumocystis infections of the lung. All participants underwent two sputum induction procedures, inhaling 3% saline via ultrasonic nebuliser. During one randomly allocated procedure, airway clearance techniques (chest wall percussion, vibration, huffing) were incorporated.
In total, 59 participants completed the trial. The airway clearance techniques had no significant effect on how the test was tolerated, the volume expectorated or the quality of the sample obtained (assessed by the presence of alveolar macrophages). The techniques did not significantly affect how often the test identified a suspected organism, nor the sensitivity or specificity of sputum induction.
In conclusion, the study was unable to demonstrate any effect of airway clearance techniques on the sputum induction procedure. The results provide some justification for not including airway clearance techniques as part of the sputum induction procedure.
Early identification of tuberculosis (TB) and pneumocystis infections of the lung is important so that therapy and infection control measures can be instituted appropriately 1, 2. Sputum induction is one of the ways in which a sample of respiratory tract organisms can be obtained from suspected cases. The reported sensitivity of sputum induction varies widely 3–9, emphasising the importance of technical factors in the collection and processing of the sputum sample.
In a survey of major teaching hospitals in Australia and New Zealand, 75% of centres reported using physiotherapy techniques designed to clear mucus from the airways as part of their sputum induction procedure 10. However, an electronic search of the Physiotherapy Evidence Database (PEDro), Medline, PubMed, Embase, Cumulative Index of Nursing and Allied Health Literature (CINAHL) and the Cochrane Library, using the search terms "sputum induction" and "induced sputum", identified no studies on the effect of physiotherapy techniques on the procedure.
A clinical trial was carried out, which aimed to determine whether airway clearance techniques alter the efficacy of the sputum induction procedure. The objectives of the present study were to examine whether adding airway clearance techniques to the sputum induction procedure would: 1) increase sample volume; 2) affect how well and for how long patients tolerate the procedure; 3) improve sample quality assessed by the presence of alveolar macrophages; 4) increase identification of the suspected organism; and 5) affect the sensitivity or specificity of the test.
|
MATERIALS AND METHODS
|
|---|
Participants
The study was conducted at Royal Prince Alfred Hospital, Sydney, and Princess Alexandra Hospital, Brisbane (both Australia), and was approved by the relevant ethics committees. Participants were recruited from inpatients and outpatients referred for sputum induction for suspected pulmonary infection with Mycobacterium tuberculosis or Pneumocystis jiroveci.
Pneumocystis organisms from different host species have been shown to be multiple species with very different DNA sequences. In recognition of this, the organism that causes human Pneumocystis pneumonia is now named Pneumocystis jiroveci. The acronym PCP is retained, referring to Pneumocystis pneumonia. P. carinii now refers exclusively to one of the two Pneumocystis spp. found only in rats 11.
In total, 59 participants completed the trial (table 1
). Participants were excluded if they were aged <18 yrs or were unable to provide informed consent. Of the participants: 36 (61%) were suspected of M. tuberculosis infection only; 10 (17%) were suspected of P. jiroveci infection only; and 13 (22%) were suspected of infection with either organism. In total, 33 (56%) had a nonproductive cough and 26 (44%) had produced a spontaneous sputum sample that had been negative for the suspected organism(s).
Study design
A randomised, crossover design was used. Participants attended on 2 days within a 5-day period, undergoing a standard sputum induction procedure on each day. This involved the inhalation of 3% saline via an ultrasonic nebuliser for as long as tolerated, for up to 15 min. On one of the days (physio), airway clearance techniques were used during saline delivery. These included manual percussion and vibration of the chest wall during deep breathing and huffing and coughing, interspersed with periods of relaxed diaphragmatic breathing. On the other day (control), participants were advised to breathe slightly more deeply than usual, to occasionally take two to three deep breaths, and to cough at least every 5 min and whenever they felt secretions in their airways. The computer-generated randomisation list was sealed in opaque numbered envelopes by a person not otherwise involved in the study to ensure concealment.
Outcome measures
The severity of cough, dyspnoea, wheeze and chest tightness was recorded by the participant on a visual analogue scale (VAS) before and immediately after the test. The volume of expectorate and the duration of the test were recorded by the investigator immediately after the procedure. The sample was examined for the presence of alveolar macrophages by the cytologist and then microbiological testing was carried out by microbiological scientists for the suspected organism(s). The cytologist and scientists were blinded to the randomisation order. To test for the presence of alveolar macrophages, four slides were smeared with sputum and fixed in 95% ethanol. They were then stained using the Papanicolaou method and screened. If five or more alveolar macrophages were found, the specimen was deemed to contain alveolar macrophages. PCR assays were used for the identification of P. jiroveci, and standard smear and culture for M. tuberculosis were performed.
Statistical analyses
Data were analysed by intention to treat. Differences in volume of expectorate, symptom severity and test duration were assessed using the paired t-test or, where not normally distributed, the Wilcoxon signed-rank test. The presence of alveolar macrophages and identification of the suspected organism were compared using McNemar's test.
The sensitivity of the sputum induction procedure was calculated as the probability of a positive test among participants with the infection according to a reference standard. The specificity of the procedure was calculated as the probability of a negative test among participants without the infection according to the same reference standard. The following reference standards were used.
For TB two of the following three criteria were required: 1) clinical symptoms suggesting TB (e.g. cough, fever, night sweats); 2) acid-fast bacilli visible on smear by Ziehl-Nielsen staining or M. tuberculosis cultured from sputum (not including those samples used in the study), bronchoalveolar lavage (BAL) fluid or biopsy samples; and 3) a chest radiograph independently interpreted as highly suggestive of TB. For patients without microbiological confirmation, a clinical and/or radiological response to anti-TB medications was also required.
This definition was used in the trial by Fitzgerald et al. 12, and is based on the definition of the American Thoracic Society's diagnostic standards for TB 13.
For PCP a positive BAL or other lung sample (e.g. lung biopsy/autopsy) was required. In the absence of any such samples having been tested, two of the following three criteria were required: 1) clinical symptoms suggesting PCP (e.g. cough, fever, dyspnoea); 2) bilateral interstitial infiltrates and/or progression to alveolar pattern on chest radiography; and 3) response to anti-PCP therapy.
A respiratory physician determined whether participants met these definitions based on a review of the participant's medical record 6 months after sample collection. The physician was blinded to the test results of the samples obtained at the two study visits. The extended McNemar's test 14 was used to compare the difference in the sensitivities or specificities of the two test conditions.
Sample size calculation
Sample size calculation was based on the outcome of sample quality, assessed by the presence of alveolar macrophages. To determine the required sample size, data from previous sputum induction procedures at Royal Prince Alfred Hospital, where airway clearance techniques were incorporated, was used. The rate at which adequate quality samples were obtained with sputum induction was 92%. If omitting airway clearance techniques reduced this proportion to 60%, this would be adequate justification for the continued use of physiotherapy as part of the procedure. With a power of 80% and 
= 0.05, the current authors calculated that 50 patients (25 per group) would be required to complete this aspect of the study 15.
|
RESULTS
|
|---|
Tolerance of the test
Table 2 shows the comparability of participants' ratings of symptoms on the VAS before commencing the sputum induction procedure on each of the 2 study days.
The inclusion of airway clearance techniques did not significantly affect the change in symptoms that occurred during the sputum induction procedure: cough p = 0.31; dyspnoea p = 0.96; wheeze p = 0.46; chest tightness p = 0.57 (fig. 1).
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 1— Box plots to show the change in cough, dyspnoea, wheeze and chest tightness during the sputum induction procedure with (physio.) and without (control) airway clearance techniques incorporated between the 10th and 90th percentiles. •: outlier. Physio.: physiotherapy; NS: nonsignificant. n = 48.
|
|
There was no significant difference in how long the test was tolerated on the 2 days, p = 0.98 (fig. 2). The volume of expectorate produced on the 2 days was not significantly different, p = 0.42 (fig. 3). Two participants were nonproductive on both days.
View larger version (6K):
[in this window]
[in a new window]
|
Fig. 2— Mean duration for which the test was tolerated. –––: median containing interquartile range; •: outlier. Physio.: physiotherapy; NS: nonsignificant. n = 59.
|
|
View larger version (7K):
[in this window]
[in a new window]
|
Fig. 3— Box plots showing volume of expectorate produced during the test between the 10th and 90th percentile. •: outlier. Physio.: physiotherapy; NS: nonsignificant. n = 59.
|
|
Sample quality
Samples from the first 49 consecutive participants at the Royal Prince Alfred Hospital were assessed for the presence of alveolar macrophages. McNemar's test showed that the airway clearance techniques did not significantly affect how often the test achieved a sample with alveolar macrophages, p = 0.11 (difference 0.12 in favour of control; 95% confidence interval (CI) -0.01–0.25; table 3).
Presence of suspected organisms
The 57 productive participants had their samples assessed for the presence of the suspected organism(s). McNemar's test showed that the airway clearance techniques did not significantly affect how often the test identified a suspected organism on microbiological testing, p = 0.45 (difference 0.05 in favour of physio.; 95% CI –0.04–0.14; table 4).
Sensitivity and specificity
After application of the reference standards, the sensitivity of the procedure with airway clearance techniques was 57% and without was 50%. This 7% difference in the sensitivities of the two test conditions was nonsignificant (95% CI –24–38%). The specificity of the procedure with airway clearance techniques was 94% and without was 100%, which was also nonsignificant (95% CI –3–15%).
|
DISCUSSION
|
|---|
The present study found that the identification of a suspected organism with sputum induction was not significantly altered by the inclusion of airway clearance techniques. In addition, airway clearance techniques did not significantly alter factors which might be expected to improve the efficacy of the diagnostic test, such as the likelihood of obtaining a sample.
As an indication of baseline comparability, the participants' symptoms before the commencement of the sputum induction procedure were reasonably consistent across the 2 study days, with differences in median VAS scores of 
7 mm on a 100-mm scale. These differences were accounted for by examining change in VAS scores.
There was little support for the mechanisms by which airway clearance techniques might improve the efficacy of the sputum induction procedure. The techniques did not significantly affect the test duration or the volume of expectorate. The CI for sample quality excluded the possibility that airway clearance techniques make a clinically worthwhile impact on the likelihood of obtaining a sample with alveolar macrophages.
Airway clearance techniques did not significantly affect the change in symptoms that occurred during the sputum induction procedure. Apart from a few outliers, most participants reported little change in symptoms with either test.
The identification of the suspected organism was not significantly altered by the inclusion of airway clearance techniques. A clinically important effect size was not pre-specified, but the authors acknowledge that the CI did not completely exclude the possibility of a clinically important benefit from airway clearance techniques.
The difference in sensitivities of the tests was 7% better with airway clearance techniques. The CI here was wide enough to contain clinically important effects in either direction. Further research in populations with a higher prevalence of M. tuberculosis and P. jiroveci infection may help determine any effect on sensitivity. The difference in specificities of the test was 6% better without airway clearance techniques. The narrow CI indicates that airway clearance techniques are unlikely to improve the specificity of the procedure to a clinically important extent.
Airway clearance techniques have been shown to accelerate mucus clearance in diseases such as cystic fibrosis 16 and bronchiectasis 17, where there is substantial retention of mucus with abnormal rheological properties. Accordingly, airway clearance techniques are designed to dislodge sticky mucus and use airflow to move it more centrally, from where it may be eliminated with cough or forced expiratory techniques. However, sputum samples induced with hypertonic saline are of small volume and have more favourable rheology for clearance 18, which may render the airway clearance techniques unnecessary. Some centres report using 7, 10 and even 20% saline for the procedure 10. Higher saline concentrations cause a greater improvement in mucus rheology 19, so airway clearance techniques are unlikely to be more effective in such protocols.
In conclusion, airway clearance techniques did not significantly improve the efficacy of the sputum induction procedure in the present study. Some confidence intervals contained differences that could be considered clinically important. Therefore, the results of this study do not completely exclude the possibility that airway clearance techniques may be a clinically useful adjunct to the procedure. However, the lack of evidence for a mechanism by which airway clearance techniques might improve the efficacy of the procedure does not support this scenario. Further studies are required to definitively determine the effect of airway clearance techniques on some aspects of the test, particularly sensitivity. In the interim, the current results provide some justification for not including, or for discontinuing the use of, airway clearance techniques as part of the sputum induction procedure.
|
REFERENCES
|
|---|
- American Lung Association. The American Lung Association Conference on re-establishing control of tuberculosis in the United States. Am J Respir Crit Care Med 1996;154:251–262.[ISI][Medline]
[Order article via Infotrieve]
- Bennett CL, Mathews C, Kosecoff J. Results of a consensus panel on process of care for patients with Pneumocystis carinii pneumonia. In: The Fifth International Conference on AIDS 1989. Montreal, Canada, 1989; pp. 284A
- Al Zahrani K, Al Jahdali H, Poirier L, Rene P, Menzies D. Yield of smear, culture and amplification tests from repeated sputum induction for the diagnosis of pulmonary tuberculosis. Int J Tuberc Lung Dis 2001;5:855–860.[ISI][Medline]
[Order article via Infotrieve]
- Conde MB, Soares SL, Mello FC, et al. Comparison of sputum induction with fiberoptic bronchoscopy in the diagnosis of tuberculosis: experience at an acquired immune deficiency syndrome reference center in Rio de Janeiro, Brazil. Am J Respir Crit Care Med 2000;162:2238–2240.[Abstract/Free Full Text]
- Conde MB, Loivos AC, Rezende VM, et al. Yield of sputum induction in the diagnosis of pleural tuberculosis. Am J Respir Crit Care Med 2003;167:723–725.[Abstract/Free Full Text]
- Huang L, Hecht FM, Stansell JD, Montanti R, Hadley WK, Hopewell PC. Suspected Pneumocystis carinii pneumonia with a negative induced sputum examination. Is early bronchoscopy useful? Am J Respir Crit Care Med 1995;151:1866–1871.[Abstract]
- Miller RF, Semple SJ, Kocjan G. Difficulties with sputum induction for diagnosis of Pneumocystis carinii pneumonia. Lancet 1990;335:112
- Masur H, Gill VJ, Ognibene FP, Shelhamer J, Godwin C, Kovacs JA. Diagnosis of Pneumocystis pneumonia by induced sputum technique in patients without the acquired immunodeficiency syndrome. Ann Intern Med 1998;109:755–756.
- Larson JL, Ridzon R, Hannan MM. Sputum induction versus fiberoptic bronchoscopy in the diagnosis of tuberculosis. Am J Respir Crit Care Med 2001;163:1279–1280.[Free Full Text]
- Elkins M. Sputum induction - current practice in Australia & New Zealand. Respirology 2003;7: Suppl. 1 A63
- Stringer JR, Beard CB, Miller RF, Wakefield AE. A new name (Pneumocystis jiroveci) for pneumocystis from humans. Emerg Infect Dis 2002;8:891–896.[ISI][Medline]
[Order article via Infotrieve]
- Fitzgerald DW, Severe P, Joseph P, et al. No effect of isoniazid prophylaxis for purified protein derivative-negative HIV-infected adults living in a country with endemic tuberculosis: results of a randomized trial. J Acquir Immune Defic Syndr 2001;28:305–307.
- American Thoracic Society Diagnostic standards and classification of tuberculosis in adults and children. Am J Respir Crit Care Med 2000;161:1376–1395.[Free Full Text]
- Hawass NED. Comparing the sensitivities and specificities of two diagnostic procedures performed on the same group of patients. Br J Radiol 1997;70:360–366.[Abstract]
- Sokel RR, Rolf FJ. Biometry. New York, W H Freeman and Co, 1981
- van der Schans C, Prasad A, Main E. Chest physiotherapy compared to no chest physiotherapy for cystic fibrosis. Cochrane Database Syst Rev 2000;2:CD001401
- Jones AP, Rowe BH. Bronchopulmonary hygiene physical therapy for chronic obstructive pulmonary disease and bronchiectasis. Cochrane Database Syst Rev 1998;2:CD000045
- King M, Dasgupta B, Tomkiewicz RP, Brown NE. Rheology of cystic fibrosis sputum after in vitro treatment with hypertonic saline alone and in combination with recombinant human deoxyribonuclease I. Am J Respir Crit Care Med 1997;156:173–177.[Abstract/Free Full Text]
- Scheffner AL, Medler EM, Jacobs LW, Sarett HP. The in vitro reduction in viscosity of human tracheobronchial secretions by acetylcysteine. Am Rev Respir Dis 1964;90:721–729.[Medline]
[Order article via Infotrieve]
TOP
ABSTRACT
MATERIALS AND METHODS
