Proteins present in bronchoalveolar lavage (BAL) fluid have been the object of study in recent years, both in normal subjects and in a variety of pulmonary diseases. Albumin, immunoglobulin G (IgG), and immunoglobulin A (IgA) are found in BAL fluid in normal subjects, but immunoglobulin M is rarely detected. Increases in all three classes of immunoglobulins have been described in various interstitial pulmonary diseases. These may originate from the intravascular space, reaching the lower respiratory tract by a process of transudation across the endothelial and pulmonary epithelial membranes or from local synthesis.
There is little information about changes in immunoglobulin concentration in BAL fluid in asbestosis, but it has been suggested that immunoglobulin profiles in asbestosis may differ markedly from other interstitial pulmonary diseases.
The rate of clearance of inhaled solutes from lung to blood is thought to be an index of pulmonary epithelial permeability. Increased solute clearance occurs in asbestosis and in other interstitial pulmonary diseases. Little is known of the relationship between solute clearance and protein concentrations in BAL fluid, which may also reflect alveolar permeability.
The purposes of the present study were to compare the rate of clearance of pentetic acid labelled with 99mTc-DTPA from the lungs and BAL protein concentrations as methods of estimating the permeability of the alveolocapillary barrier and to examine immunoglobulin profiles of BAL fluid in asbestosis.
Materials and Methods
We studied 28 men (mean age, 57 years; range, 32 to 74 years) with asbestosis. The mean time since first exposure to asbestos was 34 years (range, 18 to 53 years), and the mean duration of exposure was 17 years (range, 1 to 43 years). Exposure had ceased a mean of 16 years previously (0 to 31 years). TVventy men had worked in the thermal insulation industry, and eight had carried out mixing, milling, or sawing of asbestos. Fourteen were current cigarette smokers, and 14 were nonsmokers, of whom 11 had ceased smoking at least two years previously and three had never smoked. Sixteen patients showed clubbing of the fingers, and all had crackles over the pulmonary fields and radiologic shadowing compatible with asbestosis. Ttoenty-six had experienced breathlessness for a mean period of 4.5 years (range, 1 to 17 years). This population is different from but overlaps with that reported in previous studies in which physiologic profiles are described.
A control range of immunoglobulin levels in BAL fluid was established by studying 11 patients (ten men and one woman; mean age, 40 years; range, 24 to 62 years) undergoing fiberoptic bronchoscopy. The reasons for bronchoscopy were suspected bronchial neoplasm in five, of which two were confirmed, unexplained hemoptysis in four, and unexplained dyspnea in two. Where hemoptysis was the indication, this had occurred at least five days previously, and no active bleeding site was observed at the time of bronchoscopy in any of the cases. Six patients were current smokers, and five were nonsmokers. Bronchoalveolar lavage was performed in each case in a radiologically and bronchoscopically normal lung.
Normal values for the half-time clearance rate in minutes of an inhaled aerosol of 99mTc-DTPA from lungs to blood (tJiLB) were obtained by studying a control group of 31 nonsmokers, of whom 23 had never smoked and eight had ceased smoking at least five years previously. Each patient gave informed written consent, and approval of the ethics committee was obtained.
Bronchoalveolar lavage was carried out in all 28 patients with asbestosis and in the 11 control subjects. The procedure was performed through a fiberoptic bronchoscope using three sequential aliquots of 50 ml of sterile physiologic saline solution at 37°C, and the sample was prepared for total and differential cell counts as previously described. Manual counts of regular-shaped and regular-segmented ferroprotein-coated fibers were carried out on samples treated with 0.5 ml of 0.1M sodium hydroxide as previously described. Asbestos body counts were standardized to 1 ml of fluid.
Serum and BAL Estimations
The same method was used for assays of IgG, IgA, IgM, and albumin in serum and BAL fluid. The assay system used was a rate nephelometric immunoassay. Assays were performed using an analysis system (Beckman ICS) as previously described. The system was used according to the manufacturers instructions, except that we did not use Beckman antisera but substituted our own. Calibration was achieved using the Beckman calibrant. The system is fully automatic, sampling unadulterated specimens and giving results in grams per liter directly. All methods were monitored using conventional within-batch and between-batch quality assurance systems, based on human serum pools, a United Kingdom national reference preparation for specific proteins, and a national external quality assessment scheme. Immunoglobulins in BAL fluid were not detectable using this system below a concentration in unadulterated BAL fluid of 0.01 g/L.
Validation of BAL IgA Measurement
Preliminary work had been carried out to assess the validity of using the Beckman ICS system for measurement of BAL (predominantly dimeric ULS) IgA. A study was performed using samples of saliva as a source of secretory IgA, since there are no well-characterized 11S IgA preparations currently available in the United Kingdom. A comparison was made between the results obtained using the Beckman ICS system and a standardized radial immunodiffusion assay; the relative reactivities of serum (monomeric 7S) and secretory (dimeric 11S) IgA have been established and a conversion factor (xl.4) determined for the calculation of US IgA concentrations from results obtained using a 7S standard with the radial immunodiffusion assay.“ The radial immunodiffusion assay was performed using the same antisera (anti-a-chain Fc) and calibrants as the Beckman ICS assay to ablate any reagent or calibration problems.
Saliva (mixed) was obtained from 30 healthy members of the laboratory staff*and was analyzed in duplicate by both methods. The results of radial immunodiffusion assay were corrected for the known difference in reactivity between secretory and serum IgA in this system, which is mainly due to the slower diffusion of the larger secretory IgA molecule, and the results obtained by both methods were compared. Concentrations of IgA in BAL fluid quoted in this study are direct, not corrected results.
Clearance of 99mTc-DTPA
The half-time clearance in minutes of 99mTc-DTPA from lungs to blood (tftLB) was measured in 14 nonsmoking and 12 smoking patients with asbestosis and in 31 normal subjects. An aerosol was generated from a jet nebulizer (Acorn) containing 20 mCi of 99mTc-DTPA in 4 ml of saline solution. The output was modified by passage through an array of stainless steel ball bearings to remove large particles, as described by Jones and colleagues. A cascade impactor demonstrated that 77 percent of the particles were less than 0.9|jl, and only 4 percent were greater than 2|i in diameter. Subjects inhaled for 5 minutes from the nebulizer and retained about 5 percent of the radioactivity. Data were collected over 15 minutes by a gamma scintillation camera (Siemens 37 ZLC) linked to a computer (MDS A2), and correction for background tissue activity was achieved by scanning the area between the kidneys and by administering an intravenous injection of 0.5 mCi of 99mTc-DTPA. A proportion of the interrenal counts, derived by the method described by Jones and colleagues for thigh counts, was subtracted from the pulmonary counts.
The Mann-Whitney U test (two-tailed) was used for group comparisons of quantitative data, and correlations were assessed using the Spearman rank correlation coefficient. Results of salivary IgA measurements were analyzed using the Demings regression analysis, which takes into account the variance of both methods of measurement