INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Sputum induction has recently been proposed as a noninvasive alternative to bronchoscopy for the collection of airway secretions from asthmatic subjects (1, 2). Sputum induction yields satisfactory samples of induced sputum in the great majority of asthmatic subjects, and both cellular and biochemical analysis of induced sputum is feasible.

Sputum induction involves the inhalation of hypertonic saline aerosol, a stimulus known to cause bronchoconstriction in asthmatic subjects (3). The mechanism of hypertonic saline-induced bronchoconstriction in asthmatic subjects is unknown but is thought ultimately to involve airway smooth muscle contraction, perhaps provoked by release of tachykinins from sensory neurons (4) or release of mast cell mediators (5). Hypertonic saline-induced bronchoconstriction in asthmatic subjects is inhibited by inhaled steroids (6), nedocromil (4), and inhaled furosemide (7).

When hypertonic saline aerosol is delivered to asthmatic subjects for the purposes of sputum induction, most investigators have administered a pretreatment with beta-agonist from a metered dose inhaler to inhibit hypertonic saline-induced bronchoconstriction (1, 2, 8, 9). Our own practice has been to administer albuterol 180 µg by metered dose inhaler 15 min before the procedure, to limit the duration of sputum induction to 20 min, and to monitor the subject's symptoms closely during sputum induction so that the procedure can be stopped if the subject complains of wheezing or dyspnea. Spirometry is repeated at the end of 20 min (but not during sputum induction), and albuterol is administered to restore the FEV₁ to baseline if the FEV₁ falls by 10% or more. This approach is similar to the approach of Keatings and colleagues (8), who measure FEV₁ before and after a 15-min sputum induction procedure. Other investigators have taken a different approach. For example, Pin and coworkers (1) measure FEV₁ at 5-min intervals during sputum induction, regardless of subject's symptoms, to guard against excessive bronchoconstriction. The necessity for measuring pulmonary function during sputum induction depends on the efficacy of albuterol pretreatment in preventing excessive bronchoconstriction during the procedure. To our knowledge, the frequency of excessive bronchoconstriction during sputum induction has not been reported in any large series, although some studies, including our own (10), have reported falls in FEV₁ as great as 50% in some asthmatic subjects undergoing sputum induction.

To determine the efficacy of pretreatment with albuterol 180 µg in preventing excessive bronchoconstriction during sputum induction, we performed a retrospective review of pulmonary function recorded before and after sputum induction in asthmatic subjects at our institution. We defined excessive bronchoconstriction as a fall in FEV₁ of  20% from the postbronchodilator baseline.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Between October 1992 and March 1995, five research studies involving sputum induction were performed at our center. These protocols enrolled a total of 78 subjects, who underwent a total of 351 sputum inductions (Table 1). Sputum induction was not performed on subjects who had a postbronchodilator baseline FEV₁ of < 60% predicted. Sixty-six subjects underwent more than one sputum induction.

Protocol

Study folders on the 78 subjects were reviewed and data on spirometry performed before and after sputum induction were abstracted and entered into a computerized database. Data on eosinophil percentages and eosinophil cationic protein (ECP) levels in induced sputum from the subjects were also abstracted from the records.

Spirometry

Bronchodilators were withheld for 8 h prior to testing. Spirometry was performed using a rolling seal Ohio 840 spirometer. Flow volume loops were generated by acquiring and digitizing the electrical volume signal versus time produced by a rolling seal spirometer and then scaling these digitized computer counts to liters by application of a calibration factor. Flow was derived from the change in volume over the change in time (in real time) and flow versus volume was displayed on a graphic cathode ray screen as well as printed on hard copy. All spirometric values, including peak expiratory flow (PEF) were calculated from the raw data, i.e., volume versus time and the values reported are the best of three separate maneuvers.

Sputum Induction

The sputum induction protocol consisted of baseline FEV₁, pretreatment with albuterol 180 µg, postbronchodilator spirometry 15 min later, the induction procedure itself (inhalation of 3% saline for 20 min), and postsputum induction spirometry. Nebulized sterile 3% saline was delivered for 20 min from a DeVilbiss 65 ultrasonic nebulizer or a DeVilbiss Ultraneb 99 nebulizer, the reservoir of which was filled with 3% saline (150 ml). These nebulizers generate particles of a mean mass median diameter of 4.5 µm and have an output of approximately 2.4 ml/min. Inhalation of hypertonic saline was interrupted every 2 min so that subjects could expectorate all secretions, including both sputum and saliva, into a clean plastic container. In addition, subjects were encouraged to cough up secretions at any time during the procedure.

Subject safety during sputum induction was ensured by the following protocol stipulations: (1) all sputum inductions were performed in a hospital setting by highly trained and experienced staff; (2) a physician investigator was on call and immediately available for each procedure; (3) subjects with an FEV₁% predicted < 60 were excluded from sputum induction; (4) subjects were monitored closely for the development of symptoms, and the procedure was interrupted for spirometry if symptoms of wheeze or dyspnea developed; (5) a sputum induction worksheet captured all spirometry data, and subjects did not leave the laboratory until their FEV₁ returned to within 10% of baseline; and (6) albuterol was given by metered dose inhaler or by nebulizer in the doses required to reverse hypertonic saline-induced bronchoconstriction.

For the purposes of evaluating the safety of sputum induction, we designated the preinduction postbronchodilator FEV₁ as the baseline FEV₁. This postbronchodilator FEV₁ and the immediate postsputum induction FEV₁ were the values used to calculate the percent fall in FEV₁ during sputum induction.

Sputum Processing and Analysis

Induced sputum was processed as previously described (2). Briefly, 250 µl of homogenized sputum or saliva (diluted in saline to prevent cell crowding on the slide) were spun in a cytocentrifuge (model 7 cytospin; Shandon Scientific, Sewickley, PA) and stained using Diff-Quik^® stain (Baxter Scientific Products, Miami, FL). For sputum, all cells were enumerated and counted, which allowed calculation of both the squamous cell percentage of the total cell sample and the epithelial, macrophage, neutrophil, eosinophil, and lymphocyte percentages of the nonsquamous cells in the sample. At least 200 nonsquamous cells on each sputum slide were read by the same investigator. The portion of the homogenized sputum and saliva samples not used for total and differential cell counts were centrifuged at 1,037 g for 5 min. The supernatant was aspirated and frozen at 70° C for later analysis of ECP levels, which were assayed using a commercially available radioimmunoassay that has a lower limit of detection of 2 ng/ml (Pharmacia Diagnostics Inc., Fairfield, NJ).

Allergen Skin Tests

Skin test reactivity to 11 common aeroallergens was done using methods previously described (11).

Methacholine Challenge

Bronchial reactivity to methacholine was measured using a method from Chai and colleagues (12) as previously described (11). Methacholine reactivity was determined in these protocols 2-7 d before sputum induction.

Statistics

Data are presented as the mean ± standard deviation (SD). For comparative analysis of subgroups, data were first analyzed using the Kruskall Wallis test; where this test indicated a significant difference between the methods, between-group comparisons were made using the Mann-Whitney U test. For correlations between these data, the Spearman rank order test was used. A p value < 0.05, with a two-tailed test, was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Eleven of the 78 subjects (14%) who underwent the 20-min sputum induction had a fall in FEV₁ from the postbronchcodilator baseline of 20% or greater ("excessive bronchoconstriction") (Table 1). A further 6 subjects (8%) had a fall in FEV₁ of between 10% and 20%. The largest fall in FEV₁ was 69%. This subject had a baseline FEV₁ of 2.1 L (51% predicted), which increased to 2.6 L (63%) with albuterol pretreatment. After sputum induction the FEV₁ was 0.8 L (20% predicted). With additional albuterol treatment (360 µg albuterol by metered dose inhaler and 2.5 mg albuterol by nebulizer), the FEV₁ increased to 2.3 L.

All subjects who had a fall in FEV₁ of 20% or more recovered within 1 h to within 10% of their baseline prebronchodilator FEV₁ value following treatment with 2-4 puffs of albuterol or with 2.5 mg of nebulized albuterol (three subjects). No subject developed refractory bronchoconstriction requiring hospitalization or emergency room treatment.

Subjects who had excessive bronchoconstriction were characterized by a lower baseline FEV₁ as a percent of predicted, by a lower PC₂₀ for methacholine, and by higher percentages of eosinophils and higher levels of ECP in their induced sputum (Table 1).

Baseline FEV₁ was a good predictor of decline in FEV₁ during sputum induction. For example, 42% of the subjects whose FEV₁ was less than 80% predicted had excessive bronchoconstriction during sputum induction (Table 2). Only 3% of subjects whose FEV₁ was greater than 90% predicted had excessive bronchoconstriction during sputum induction.

The change in FEV₁ during sputum induction was significantly correlated with the baseline prebronchodilator FEV₁ % predicted, the baseline postbronchodilator FEV₁ % predicted, the PC₂₀ for methacholine, and the eosinophil percentage in induced sputum (Table 3).

The change in FEV₁ during sputum induction after the first sputum induction was generally predictive of the change in FEV₁ during subsequent sputum inductions. For example, 66 of the 78 subjects underwent more than one sputum induction. Fifty-seven of these 66 subjects had a less than 20% fall in FEV₁ during the first sputum induction. Fifty-two of these 57 subjects (91%) had a less than 20% fall in FEV₁ during a subsequent sputum induction. Seven of the 9 subjects (78%) who had a greater than 20% fall in FEV₁ during the first sputum induction also had a greater than 20% fall in FEV₁ during a subsequent sputum induction.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this study we examined the efficacy of pretreatment with albuterol 180 µg in preventing hypertonic saline-induced bronchospasm during sputum induction in asthmatic subjects. We found that, although no subject developed refractory bronchoconstriction requiring hospitalization or emergency room treatment, pretreatment with 180 µg albuterol does not prevent excessive bronchoconstriction (more than a 20% fall in FEV₁ from postbronchodilator baseline) in all asthmatic subjects undergoing sputum induction. We have therefore modified our sputum induction protocol to increase the dose of albuterol given as pretreatment and to monitor peak flow at 4-min intervals during sputum induction to guard against and detect excessive bronchoconstriction.

Overall, 12% of our subjects experienced excessive bronchoconstriction despite pretreatment with 180 µg albuterol. One subject had a very large fall in FEV₁ but recovered within 1 h with nebulized albuterol treatment. Notably, the subjects who developed bronchoconstriction had more severe baseline airway narrowing. For example, all but 1 of the 11 subjects who had excessive bronchoconstriction had a baseline FEV₁ less than 80% predicted. Overall, 10 of the 24 subjects (42%) with baseline FEV₁ less than 80% predicted developed excessive bronchoconstriction during sputum induction.

We did not find it surprising that hypertonic saline-induced bronchospasm was greatest in asthmatic subjects with the most severe asthma, as evidenced not only by lower FEV₁ but also by lower PC₂₀ for methacholine and higher eosinophil percentage in induced sputum. However, these findings were not necessarily predictable. For example, there have been conflicting reports of the strength of the association between bronchial responsiveness to methacholine and hypertonic saline. Although some investigators have found a significant correlation (13), others have not (14, 15). Similarly, Iredale and coworkers (16) did not find a significant correlation between eosinophil percentage in induced sputum and bronchial responsiveness to hypertonic saline in a group of 24 asthmatic subjects.

Recently, there has been a case report of a fatal asthma attack in a 22-yr-old asthmatic subject undergoing an inhalation challenge with ultrasonically nebulized distilled water (17). This report and the significant minority of subjects in our study who developed excessive bronchoconstriction during sputum induction serve as reminders that asthmatic subjects can be extremely sensitive to nonisotonic aerosols. Accordingly, during sputum induction care should be taken to adhere to guidelines that have been presented for bronchial challenges with hypertonic saline aerosols (18, 19). When hypertonic saline is being administered for the purposes of sputum induction, subjects should be pretreated with albuterol and should be supervised constantly. Pulmonary function should be monitored at regular intervals, and the resources required to manage a sudden attack of asthma should be immediately available. Based on the observations made in this study, we have changed our protocol for sputum induction in two ways. First, we have increased the dose of albuterol used for pretreatment from 180 µg to 360 µg. Second, whereas we previously monitored spirometry only at the beginning and end of sputum induction, we now measure pulmonary function at regular intervals during sputum induction. Repetitive spirometry maneuvers for this purpose are time-consuming and can be difficult for the subject, and we have therefore developed a protocol based on peak flow measurement (Figure 1).

View larger version (49K): [in this window] [in a new window]   Figure 1.   Protocol for monitoring pulmonary function during sputum induction.

In summary, we have found that pretreatment with 180 µg of albuterol does not prevent excessive bronchoconstriction in all asthmatic patients undergoing sputum induction. Subjects with an FEV₁ % predicted less than 80% are especially likely to develop excessive bronchoconstriction during sputum induction. Based on our findings, we recommend that albuterol 360 µg be used for pretreating asthmatic subjects prior to sputum induction and that pulmonary function be monitored at least at 4-min intervals during sputum induction.