Abstract
Background. Investigation of the safety of sputum induction in patients with moderate-to-severe chronic obstructive pulmonary disease (COPD) has been limited. Objective. to evaluate this issue in 100 patients with a mean FEV1 of 1.2 (0.4) L. After 200 μg inhaled salbutamol, sputum induction was performed with inhaled saline in increasing and tailored concentrations (0.9% to 5%) until an adequate sample of sputum was obtained or the FEV1 fell by > 20%. Main Findings. Sputum induction was successful in 92% of occasions. The mean (SD) fall in FEV1 was 13.5 (8.6)%. Five patients had a fall > 20% but all recovered to 10% of baseline after inhaled salbutamol. The magnitude of fall in FEV1 correlated weakly with salbutamol reversibility (r = 0.37, p < 0.001), baseline FEV1/VC (r = −0.32, p = 0.001) and baseline FEV1% predicted (r = −0.36, p = 0.003) but not with age, smoking history or post-Salbutamol FEV1. Principle conclusion. Sputum induction can be performed safely using a patient-tailored approach in patients with moderate-to-severe COPD, supporting its use in research and clinical practice.
| ABBREVIATIONS | ||
| AUC | = | Area under the curve; |
| COPD | = | Chronic obstructive pulmonary disease; |
| FEV1 | = | Forced expiratory volume in 1 second; |
| VC | = | Slow vital capacity; |
| SD | = | Standard Deviation. |
INTRODUCTION
Sputum induction using hypertonic saline for quantitative cell counts has been used for over 10 years as a non-invasive method to assess lower airway inflammation in the research of asthma (Citation[1]) and, more recently, in short and long term clinical management of asthma (Citation[2]). It has also been shown to have clinical relevance in the management of COPD in order to document its association with eosinophilic or infective bronchitis. Sputum eosinophilia predicts significant clinical improvement in quality of life and FEV1 after corticosteroid therapy (Citation[3], Citation[4]).
The risk of saline bronchoconstriction in patients or research subjects with asthma is acceptable but this assessment in COPD is limited. With the increasing number of publications reporting sputum induction in patients with COPD, we felt it was necessary to address the safety of this procedure. Previous studies evaluating the safety of induction have either excluded subjects with severe disease (Citation[5]), included small numbers of patients (Citation[6]) or used a non-incremental dosing regimen (i.e., nebulisation of 0.9% or 3% saline) (Citation[7]). Indeed in a European Respiratory Society Task Force Working Group report on the safety of sputum induction, it was evident that few studies had reported on the safety of sputum induction in patients with COPD compared to those with asthma, and there was a paucity of data in patients with severe disease (Citation[8]). We have therefore evaluated the risk and possible predictors of bronchoconstriction in a large sample of subjects with moderate-to-severe COPD.
METHODS
Participants
The data were obtained from 100 consecutive patients with COPD with a post-salbutamol FEV1/VC of < 70% who attended screening for entry into two clinical studies (Citation[9], Citation[10]) (). The severity, as indicated by FEV1% predicted, was moderate (69–50%) in 36, severe (49–30%) in 47, and very severe (< 30%) in 17. Twenty-three patients had significant reversibility as defined by improvement in FEV1 after bronchodilator of ≥ 200 ml and > 20% of predicted. All but 1 patient, aged 37 years, were older than 40 years and all but 2 subjects (8 and 10 pack years) were current or ex-smokers of > 15 pack years. The patients were stable on regular treatment with salbutamol or ipatropium. None had had received inhaled long-acting anticholinergic or leukotriene receptor antagonists. Patients were not excluded if they had a history of cardiovascular disease or were taking concomitant medication. None received corticosteroids for at least 2 months. Both studies from which the data were obtained were approved by the Research Ethics Board of St. Joseph's Healthcare and all patients provided written informed consent.
Table 1. Characteristics of all patients and those with and without a 20% fall in FEV1 during sputum induction
Design
This was a cross-sectional study of baseline data from two clinical studies in which safety of sputum induction was a secondary objective (Citation[9], Citation[10]). Patients attended to document clinical characteristics by questionnaire and for sputum induction after withholding bronchodilators for their duration of action.
Measurements
Spirometry was performed using a Koko-Trek Spirometer (Trudell Medical, London, Ontario, Canada) before and 20 minutes after 400 μ g inhaled salbutamol via a pressurized metered-dose inhaler and dedicated spacer. Sputum induction was performed as described by Pizzichini et al. (Citation[11]) with modifications. Patients with a post-salbutamol FEV1 of 1.0 L or less started nebulisation with 0.9% saline using a Medix ultrasonic nebulizer with an output of 0.87 ml/min and particle size of 5.58 μ m aerodynamic mass median diameter (Canadian Medical Products, Ltd., Markham, Ontario). Other subjects commenced nebulisation with 3% hypertonic saline. In subjects with an FEV1 < 1.5 L, spirometry was repeated after 1, 2, 5 and 7 minutes and in those with an FEV1 > 1.5 L it was performed at 2 and 7 minutes. If the FEV1 fell by < 10% then the concentration of saline was increased from 0.9% to 3% in subjects with an FEV1 < 1.5 L, and from 3% to 4% and then to 5% at 7-minute intervals in those subjects with post salbutamol > 1.5 L. If the FEV1 fell by 10–20% the same concentration of saline was used throughout and the induction was ended if the FEV1 fell by ≥ 20%. Further salbutamol 200 μ g was inhaled if there was a fall in FEV1 of ≥ 20% and spirometry was repeated. Pulse oximetry was not monitored during this study. However, the procedure could be stopped at any time if patients experienced any adverse or unpleasant event related or not to a fall in the FEV1.
The sputum was selected from the expectorate and examined as described by Pizzichini et al. (Citation[12]). Total cell count of non-squamous cells was obtained in a modified Neubauer haemocytometer with cell viability being determined by trypan blue exclusion. Sputum induction was defined as successful if a total and differential cell count could be obtained and the cell viability was greater than 40%.
Statistical analysis
The main outcome measure was the maximum fall in FEV1 after sputum induction compared to the value 20 minutes after salbutamol, expressed as the percentage from post-salbutamol levels. The saline load was calculated as the area under the curve (AUC) for the saline concentration–time of inhalation curve. Age, pack year smoking history, FEV1% predicted, FEV1/VC, reversibility after salbutamol, post-salbutamol FEV1 and AUC saline-time curve were examined as predictors for percentage fall using a backward multiple linear regression analysis method. Spearman's rank correlation coefficient was used to determine the effect of continuous variables and percentage fall, and independent t-test was used to determine differences between those with and without a 20% fall in FEV1. The analysis was performed using SPSS for Windows 10.0 (Chicago, IL, USA).
RESULTS
Sputum induction was successful in 92 of 100 samples. In a further three, cell counts were obtained but the cell viability was less than 40%. The success was achieved using 0.9% saline in 30 patients and a maximum of 3% in 41, 4% in 12 and 5% in 4 patients. Data on final saline concentration was not available in 5 patients. Of the 37 patients with an FEV1 < 1.0 L, only 1 received a maximum concentration of 5% saline and 2 a maximum concentration of 4%. Of 64 patients with a post-bronchodilator FEV1 < 50% predicted, 1 received a maximum concentration of 5% saline and 7 received 4%.
The safety of induction was examined by fall in FEV1 (). The mean (SD) percentage fall in FEV1 was 13.5 (8.6)% overall and 14.0 (1.7)% in those with an FEV1 < 1 L. Of the latter patients, 4 had a fall in FEV1 between 20–30%, 2 between 30–40% and one of 44%. In these 7 patients the FEV1 recovered to within 10% of baseline after inhalation of salbutamol. In the patient with the lowest baseline FEV1 of 0.49 L, this increased to 0.78 L after salbutamol, fell to 0.44 L during induction and then recovered to 0.59 L after further salbutamol.
Table 2. FEV1 and percent predicted FEV1 (mean and SD) before and during sputum induction
We examined the influence of various factors on the bronchoconstrictive effect of saline induction. Those patients with an FEV1 fall > 20% had a lower baseline FEV1, lower FEV1% predicted, lower FEV1/VC and greater reversibility with salbutamol. There was a weak correlation with FEV1% predicted (r = −0.36, p = 0.003), FEV1/VC (r = −0.36, p−0.001) and reversibility (r = 0.037, p < 0.001). The only two significant independent predictors of a % fall from linear regression analysis were reversibility and FEV1/VC with a regression equation of % fall = 20.28 + 0.149*bronchodilator reversibility − 0.203*FEV1/VC. There was no significant difference between the mean (SD) fall in FEV1 comparing males with females [208 (132) vs. 139 (102) ml], current vs. ex-smokers [200 (153) vs. 175 (112) ml], patients receiving therapy with LABA or not [178 (123) vs. 182 (117) ml] or beta-blockers or not [247 (30) vs. 182 (120) ml]. The recovery of FEV1 after salbutamol was not significantly different in subjects taking LABA or not [1.05 (0.39) vs. 1.27 (0.47) L] or beta-blockers or not [1.57 (0.40) vs. 1.21 (0.46) L].
DISCUSSION
The results of this study have shown that sputum induction with a relatively low output nebulizer and isotonic or hypertonic saline is successful and safe in patients with moderate to severe COPD when a modified method patient-tailored to ensure safety is used. In patients with more severe airflow obstruction (FEV1 < 1 L), significant bronchoconstriction during induction is not uncommon, indicating that the procedure needs to be tailored for each patient and done carefully. The bronchoconstriction was readily reversed by salbutamol. The best, but weak predictors, of fall in FEV1 were reversibility to salbutamol and baseline FEV1/VC, with high reversibility and low FEV1/VC being associated with the greatest fall in FEV1.
The patients in the present study were taken from the screening data for two studies in patients with a clinical diagnosis of COPD. However many patients had considerable reversibility to salbutamol, which is frequently seen in patients with COPD and, by definition, indicates that they also have asthma. For example, Nisar et al. (Citation[13]) showed that 56% of subjects with COPD had reversibility to salbutamol when defined as a rise in FEV1 of 15%, or more than 200 ml. The change in patients with very low FEV1, when expressed as a percentage, can be a result of the low denominator. We believe that by not excluding patients with reversibility from the study our results are more applicable to the real-life clinical situation. In this respect, we also included patients with a cardiovascular history or those on concomitant medications.
In discussing the safety of sputum induction in patients with COPD, we therefore need to consider the reversibility to inhaled salbutamol and the severity of airflow limitation. Rytila et al. (Citation[6]) induced sputum in a group of 28 subjects with a mean FEV1 of 1.8 L and showed that although the induction was well tolerated, three subjects had greater than a 20% fall in FEV1. In contrast to our findings, they found that reversibility to salbutamol was not associated with a greater fall in FEV1 during sputum induction; indeed they observed a weak inverse relationship (r = −0.4, p = 0.03). The probable explanation for their negative finding is that they excluded patients with reversibility of more than 10%. In another study, Sutherland et al. (Citation[7]) induced sputum in 42 patients with moderate to severe COPD with a mean FEV1 1.5 L, who were not recruited on the basis of their reversibility to salbutamol. They monitored peak expiratory flow (PEF). Twelve subjects had a fall in PEF of greater than 20% requiring measurement of FEV1 and termination of the study procedure in 8 (19%) subjects. In keeping with our results, they found a positive relationship (r = 0.71 p < 0.001) between reversibility and a fall in PEF during the induction. Brightling et al. (Citation[5]) examined the repeatability of sputum induction, in 61 patients with a mean FEV1 of 1.09 L. They reported a mean fall of 11% in the FEV1 after the full induction and 5 subjects had a fall in FEV1 of more than 20%. No details of characteristics predicting bronchoconstriction during induction were given. These data in combination suggests that measurement of baseline spirometry and reversibility to salbutamol are important predictors of excessive bronchonstriction during sputum induction.
The safety of sputum indication also needs consideration of the volume of aerosol delivered, as reflected by nebulizer output, and the concentration of saline, and duration of inhalation. We used an ultrasonic nebulizer with a relatively low output, isotonic followed by hypertonic saline and inhalation periods of up to 7 minutes with each inhalation. There were differences in the method of sputum induction used in the three studies mentioned above. Rytila et al. (Citation[6]) induced the sputum by the same method we have used without modifications, but saline (0.9 to 3.0%) was delivered by a high-output nebulizer (2.5 ml/min). Sutherland et al. (Citation[7]) performed sputum induction with saline concentrations of 0.9% or 3% given for a maximum of 12 minutes delivered by a high-output nebulizer. Brightling et al. (Citation[5]) used the same nebulizer and method as ourselves with the exception of not using 0.9% saline but not performing the procedure if the FEV1 was less than 0.5 L. High output nebulizers in COPD and asthma produce greater bronchoconstriction than those with a lower output. For example, Kelly et al. (Citation[13]) compared this effect by using a high (3.0 ml/min) and a low output (0.4 ml/min) in 10 subjects with COPD and observed a fall in FEV1 of more than 20% in 3 patients induced with the low output and in 7 induced with the high output nebulizer.
Predictors of bronchoconstriction after sputum induction have also been studied in patients with asthma. For example, Wong and Fahy (Citation[15]) observed that the change in FEV1 during sputum induction of asthmatics was related to the post-salbutamol as well as pre-salbutamol FEV1 and reversibility. They gave the same concentration (3%) of saline aerosol and duration of induction to all patients whereas in the present study the initial saline dose was determined by the post-salbutamol FEV1. We did not find the post-bronchodilator FEV1 predicted a fall in FEV1; however, we used an algorithm based on this measurement to choose the starting combination of saline. We cannot therefore comment on the use of post-bronchodilator FEV1 to predict bronchoconstriction during sputum induction.
As the patients with the greatest fall in FEV1 were those with the highest degree of reversibility, they were more likely to respond to salbutamol after the induction. We therefore examined whether patients on treatment with LABA or beta-blockers seemed to limit the return to baseline by their properties of tachyphylaxis to bronchodilation or functional antagonism. However, there was no difference in the fall in FEV1 or recovery comparing patients who were or were not taking these medications or not taking. This is in keeping with a previous report in COPD in which LABA usage was not a predictor for bronchoconstriction. It is in contrast to that observed in patients with asthma where LABA (Citation[16]) or frequent doses of short-acting bronchodilators (Citation[11]) were more likely to have a fall in FEV1 with sputum induction. We did not observe any effect of beta blocking or agonist therapy on fall or recovery of FEV1.
In conclusion, when examining our data along with the previous findings of other authors, it has been shown that sputum induction can be performed safely and with high success in patients with moderate-to-severe COPD. As baseline spirometry and bronchodilator reversibility predicts bronchoconstriction (especially if patients are not selected in terms of reversibility to salbutamol), both of these measures, as well as post-bronchodilator FEV1, should be assessed prior to sputum induction. Apart from nebulizer output, no other variable predicts excessive bronchoconstriction. We therefore recommend that in subjects with COPD the induction procedure should be patient-tailored based on baseline spirometry and reversibility to salbutamol and this should be monitored closely during sputum induction.
Richard Leigh has received funding from CIHR, and Andrew Wilson and Frederick Hargreave from the Father Sean O'Sullivan Research Centre. None of the authors have competing interests.
REFERENCES
- Pin I, Gibson P G, Kolendowicz R, Girgis-Gabardo A, Denburg J, Hargreave F E, Dolovich J. Use of induced sputum cell counts to investigate airway inflammation in asthma. Thorax 1992; 47: 25–29, [INFOTRIEVE], [CSA]
- Green R H, Brightling C E, McKenna S, Hargadon B, Parker D, Bradding P, Wardlaw A J, Pavord I D. Asthma exacerbations and sputum eosinophil counts: a randomised controlled trial. Lancet 2002; 360: 1715–1721, [PUBMED], [INFOTRIEVE], [CSA], [CROSSREF]
- Pizzichini E, Pizzichini M M, Gibson Parameswaran P, Berman L, Efthimiadis A, Gleich G G, Dolovich J, Hargreave F E. Sputum eosinophilia predicts benefit from prednisone in smokers with chronic obstructive bronchitis. Am J Respir Crit Care Med 1998; 158: 1511–1517, [INFOTRIEVE], [CSA]
- Brightling C E, Monteiro W, Ward R, Parker D, Morgan M D, Wardlaw A J, Pavord I D. Sputum eosinophilia and short-term response to prednisolone in chronic obstructive pulmonary disease: a randomised controlled trial. Lancet 2000; 356 1480–1485, [PUBMED], [INFOTRIEVE], [CSA], [CROSSREF]
- Brightling C E, Monterio W, Green R H, Parker D, Morgan M D, Wardlaw A J, Pavord I D. Induced sputum and other outcome measures in chronic obstructive pulmonary disease: safety and repeatability. Respir Med 2001; 95: 999–1002, [INFOTRIEVE], [CSA], [CROSSREF]
- Rytila P H, Lindqvist A E, Laitinen L A. Safety of sputum induction in chronic obstructive pulmonary disease. Eur Respir J 2000; 15: 1116–1119, [INFOTRIEVE], [CSA], [CROSSREF]
- Sutherland E R, Pak J, Langmack E L, Silkoff P E, Martin R J. Safety of sputum induction in moderate-to-severe chronic obstructive pulmonary disease. Respir Med 2002; 96: 482–486, [INFOTRIEVE], [CSA], [CROSSREF]
- Pizzichini E, Pizzichini M M, Leigh R, Djukanovic R, Sterk P J. Safety of sputum induction. Eur Respir J Suppl 2002; 37: 9s–18s, [INFOTRIEVE], [CSA], [CROSSREF]
- Leigh R, Kamada D H, Pugsley S O, Efthimiadis A, Kitching A D, Hargreave F E. The prevalence of cardiac dysfunction in patients with COPD. Am J Respir Crit Care Med 2001; 163: A908, [CSA]
- Pizzichini E, Leigh R, Pizzichini M M, Hargreave F E. Sputum eosinophilia predicts clinical benefit for inhaled budesonide treatment in patients with COPD. 2003; A320, [CSA]
- Pizzichini M M, Pizzichini E, Clelland L, Efthimiadis A, Mahony J, Dolovich J, Hargreave F E. Sputum in severe exacerbations of asthma: kinetics of inflammatory indices after prednisone treatment. Am J Respir Crit Care Med 1997; 155 1501–1508, [INFOTRIEVE], [CSA]
- Pizzichini E, Pizzichini M M, Efthimiadis A, Evans S, Morris M M, Squillace D, Gleich G J, Dolovich J, Hargreave F E. Indices of airway inflammation in induced sputum: reproducibility and validity of cell and fluid-phase measurements. Am J Respir Crit Care Med 1996; 154: 308–317, [INFOTRIEVE], [CSA]
- Nisar M, Walshaw M, Earis J E, Pearson M G, Calverley P M. Assessment of reversibility of airway obstruction in patients with chronic obstructive airways disease. Thorax 1990; 45: 190–194, [PUBMED], [INFOTRIEVE], [CSA]
- Kelly M G, Brown V, Martin S L, Ennis M, Stuart Elborn J. Comparison of sputum induction using high-output and low-output ultrasonic nebulizers in normal subjects and patients with COPD. Chest 2002; 122: 955–959, [INFOTRIEVE], [CSA], [CROSSREF]
- Wong H H, Fahy J V. Safety of one method of sputum induction in asthmatic subjects. Am J Respir Crit Care Med 1997; 156: 299–303, [INFOTRIEVE], [CSA]
- Vlachos-Mayer H, Leigh R, Sharon R F, Hussak P, Hargreave F E. Success and safety of sputum induction in the clinical setting. Eur Respir J 2000; 16: 997–1000, [INFOTRIEVE], [CSA], [CROSSREF]