INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Sputum induction has recently been proposed as the only direct noninvasive or relatively noninvasive method for measuring airway inflammatory indices (1). In subjects with asthma and/or chronic bronchitis, the numbers and types of cells present in induced sputum differ from those in the sputum of healthy subjects, suggesting that evaluation of induced sputum may be a reproducible and valid technique in the management of asthma and/or bronchitis (4).

Studies previously reported in the literature have included small cohorts of subjects as referred to a control population. The reference values and distribution of cells in induced sputum in the general population are therefore not well defined. Moreover, important aspects of induced-sputum cell counts include not only the reference values, but also whether or not the cell counts conform to a normal distribution. We evaluated data from a large number of healthy volunteers in order to determine these parameters in induced sputum.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Study Population

One hundred-fourteen informed healthy volunteers underwent sputum induction at the same time of the day. All subjects were nonsmokers (lifetime never-smokers), were nonatopic (negative skin prick tests), and had no history of asthma or other respiratory symptoms. Their FEV₁ was greater than 80% predicted, and their airway responsiveness to methacholine was normal (provocative dose of methacholine producing a 20% decrease in FEV₁ [PD₂₀ methacholine] > 3,200 µg). No subjects had had reported symptoms of airway infections in the 3 mo previous to the study.

Study Design

The subjects were studied for 3 d that fell within a 2-wk period. On the first day the subjects were examined by means of a questionnaire and skin prick tests with a range of antigens (dog hair, cat fur, Ambrosia, Betulacea, Composite, Graminacea, Parietaria, Salicacea Alternaria alternata, Oleacea, Aspergillus fumigatus, Cladosporium, Dermatophagoides pteronyssinus, D. farinae). On the second day, spirometry and methacholine inhalation tests were performed. On the third day, at least 1 wk after the second visit, sputum was induced by inhalation of hypertonic saline.

The Ethics Committee of the Fondazione Salvatore Maugeri approved the study, and all volunteers gave written informed consent.

Methacholine Challenge

Methacholine (Sigma Chemical Co., St. Louis, MO) was dissolved in distilled water and delivered by an ampule-dosimeter device (Mefar, Brescia, Italy) driven by compressed air at a pressure of 1.5 kg/m² with 1-s activation and 5-s intervals between breaths. Aerosols were inhaled during quiet tidal breathing. After inhalation of isotonic saline as a control, subjects inhaled doubling doses of methacholine from 20 to 3,200 µg. A 3-min interval was allowed before each dose increment. FEV₁ was measured 1 min after each dose, and the best of three acceptable measurements was retained to create dose-response curves. The noncumulative doses causing a 20% decrease in FEV₁ from control FEV₁ (PD₂₀ methacholine) were calculated by interpolation between two adjacent points on the log dose-response curves.

Sputum Induction

Inhalation procedure. After baseline FEV₁ and FVC measurements, salbutamol was given by inhalation (200 µg by metered-dose inhaler), and subjects inhaled hypertonic (4.5%) saline nebulized for periods of progressively increasing length (1, 2, 4, 8, and 16 min). FEV₁ was remeasured 1 min after each inhalation period. An ultrasonic nebulizer (DeVilbiss 65; DeVilbiss Corporation, Somerset, PA) was used to nebulize the saline solutions.

Sputum processing. The collected sputum samples were examined within 2 h. Selected portions of the sputum sample originating from the lower respiratory tract were chosen through examination with an inverted microscope, and were weighed. Dithiothreitol (DTT; Sputo-lysin; Calbiochem Corp., San Diego, CA), freshly prepared in a 1:10 dilution with distilled water, was added in a volume (in microliters) equal to four times the weight of the selected portion (in milligrams) of each sputum specimen. The DTT-sputum mixture was placed in a shaking water bath at 37° C for 20 min and homogenized. It was further diluted with phosphate buffered saline (PBS) in a volume equal to the sputum plus DTT. The suspension was filtered through gauze to remove mucus, and was centrifuged at 2,000 rpm for 5 min. The supernatant was aspirated and frozen at 70°C for later analysis. The cell pellet was resuspended in a volume of PBS equal to that of the sputum plus DTT and PBS preparation described previously. Total cell count (TCC) and viability (Trypan blue exclusion method) were determined with a Burkers chamber hemocytometer. The cell suspension was placed in a Shandon 3 cytocentrifuge (Shandon Southern Instruments, Sewickley, PA) and cytospin preparations were made at 450 rpm for 6 min. Cytospin slides were fixed with methanol and were stained with May-Grunwald-Giemsa for an overall differential cell count of 500 nucleated nonsquamous cells. Only samples with a cell viability > 50% and < 20% squamous cell contamination were considered adequate.

Statistical Analysis

Results are expressed as mean ± SD, and as the median with the interquartile range. The cell count was displayed according to frequency distributions expressed as percentages of total cells. We evaluated whether the results were normally distributed. The goodness of fit to the normal distribution was statistically assessed by using the Kolomogorov-Smirnoff test.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Sputum induction was performed in all 114 subjects without observing any significant adverse effect or a decrease of FEV₁ > 20%. Induction was stopped in five subjects because of general discomfort. Thirteen subjects did not produce adequate sputum samples, and their expectorates were discharged. The analysis of the distribution of different cells in sputum was performed on 96 subjects (50 females and 46 males) who had a normal age distribution. Their induced sputum was rich in macrophages (69.2 ± 13% [mean ± SD]; 1.79 ± 1.50 × 10⁶ cells/ml) and neutrophils (27.3 ± 13%; 0.83 ± 1.11 × 10⁶ cells/ml) and poor in eosinophils (0.6 ± 0.8%; 0.01 ± 0.02 × 10⁶ cells/ml), lymphocytes (1.0 ± 1.2%; 0.03 ± 0.07 × 10⁶ cells/ml), and epithelial cells (1.5 ± 1.8%; 0.04 ± 0.07 × 10⁶ cells/ml) (Table 1). Only macrophages and neutrophils showed a normal distribution; total and differential counts of other cells did not (Figures 1 and 2).

View larger version (30K): [in this window] [in a new window]   Figure 1.   Distribution of actual percentages of macrophages (top), neutrophils (center), and eosinophils (bottom) in induced sputum in comparison with the expected normal curve values for these cells.

View larger version (28K): [in this window] [in a new window]   Figure 2.   Distribution of actual percentages of lymphocytes (top), epithelial cells (center), and total cell count (bottom) in induced sputum in comparison with the expected normal curve values.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In order to utilize sputum induction analysis as a clinical and research tool, it is necessary to determine the reference values and distributions of cells on the basis of data from a large sample of healthy volunteers representative for age and sex of the general population.

The mean values in our sample are in agreement with the findings with other, smaller samples of healthy subjects in which cellular analysis was based on selected sputum plugs (4, 9, 10). The induced sputum in our subjects was rich in macrophages and neutrophils and poor in eosinophils, lymphocytes, and epithelial cells. Our data confirm a balance between neutrophils and macrophages that may depend on the origins of sputum specimens. It has, for example, been previously shown that samples deriving from central airways tend to present more neutrophils and fewer macrophages than do samples from the more peripheral airways (12).

In our study, the percentage of eosinophils was never higher than 2.4%. This finding may be useful both in clinical practice and in research in order to define the lowest eosinophil percentage suggestive of a status of airway inflammation. Our research confirms that lymphocytes are poorly represented in induced sputum; in fact, they were absent from most of our samples. Moreover, our study confirms the importance of considering epithelial cells in the differential cell count; although they were absent from some samples, more than 5% were present in others.

Our data indicate that in induced sputum, only macrophages and neutrophils have a normal distribution. These findings suggest that for eosinophils, lymphocytes, and epithelial cells, nonparametric tests should be applied, since parametric ones could result in a Type I error.

A limit of our study could be the applicability of our data to other studies using a different sputum induction protocol in terms of hypertonic saline concentration and duration of inhalation. However, a previous study found that sputum cell counts were not affected by the tonicity of the saline used for induction (13). Moreover, in our study only nine sputum samples were produced during the last step (16 min) of inhalation; the other samples (93%) were produced during the first 15 min of inhalation. Considering that in other studies inhalation lasted no more than 20 min, we believe that our data could also be applicable to other studies with different durations of inhalation. However, standardization of the duration of inhalation is recommended.

A second limitation of our study could be in the applicability of the study data to other studies that use a different method of sputum processing (e.g., entire expectorate). However, we have previously shown that there was no difference in the differential cell count in a selected portion of sputum and that in the entire expectorate produced by the same group of healthy volunteers (8).

In conclusion, the present study provides reference values and the distribution of cell counts in induced sputum from a large sample of healthy volunteers. We propose that these data be used in comparisons of induced sputum cell counts of normal subjects with those of patients with airway inflammation.