INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Ozone (O₃) is a major component of urban air pollution which millions of people are periodically exposed to in North America. It is well documented through extensive controlled human exposures that short-term inhalation of O₃ causes dose-dependent decrements in spirometric parameters of lung function, including FEV₁ and FVC, and relatively smaller increases in specific airway resistance (SRaw) (1). Several human studies have demonstrated that short-term O₃ exposures capable of causing acute changes in lung function can also cause cellular and biochemical evidence of injury and inflammation in both proximal airways and in the distal lung (5). Markers of injury/inflammation that are typically elevated after O₃ exposure are neutrophils, total protein, fibronectin, and lactate dehydrogenase (LDH). The results of studies that have examined the relationship of O₃-induced acute changes in FEV₁ and FVC to subsequent airway inflammation (as much as 18 h after exposure) have shown either no correlation or a negative one (7, 8), although we showed in one study that acute increases in SRaw did correlate with subsequent airway inflammation (9) and Weinmann and coworkers (10) showed a correlation between isovolumetrically calculated FEF_25-75% and BAL fibrinogen concentration. With repeated short-term exposures to O₃, animal studies have consistently shown evidence of attenuation of pulmonary function responses but somewhat conflicting findings regarding airway inflammation and injury (11, 12). Repeated daily exposures to O₃ in humans have caused decrements in lung function on the first and second days of exposure, but with succeeding daily exposure, this physiologic response is attenuated (13, 14). What has not been adequately characterized in humans is the nature of the inflammatory response to O₃ with repeated short-term exposures. Our study was designed to test the hypothesis that repeated daily exposures in humans causes a progression of injury and/or inflammation. We compared the effects of a single 4-h exposure to 0.2 ppm O₃, a high ambient level, with those of repeated daily 4-h exposures to this concentration. Because large populations reside in areas with high levels of O₃ for extended periods of time, it is important to determine whether progression of O₃-induced airway inflammation occurs with repeated exposures.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Although 27 subjects were initially enrolled in this study protocol, only 15 completed both single-day and 4-d exposures to O₃. Of the original enrollees, one dropped out because of inability to complete the exercise, two were excused after violating the protocol, four dropped out after the first exposure/bronchoscopy, and five dropped out because of illness and/or scheduling difficulties. This protocol required a substantial time commitment as well as the ability to tolerate repeated periods of exercise. All results are based on the 15 subjects who completed both exposure arms. These healthy, nonsmoking volunteers were informed of the risks of the experimental protocol, and they signed consent forms approved by the Committee on Human Research of the University of California, San Francisco. The subjects, six female and nine male, ranged from 23 to 37 yr of age and specifically denied a history of pulmonary or cardiac disease or respiratory infection within 6 wk of the onset of each exposure sequence. No subject used supplemental vitamins C or E during the study. All of the subjects received financial compensation for their participation. Characteristics of individual study participants are listed in Table 1.

Pulmonary Function Measurements

SRaw was determined as the product of airway resistance and thoracic gas volume, both having been measured in a constant-volume body plethysmograph (Warren E. Collins, Braintree, MA). SRaw was calculated as the average of five measurements taken 30 s apart. Spirometry was performed on a dry rolling-seal spirometer (S400; Spirotech Division, Anderson Instruments, Atlanta, GA). Mean values for FVC and FEV₁ were calculated from three acceptable FVC maneuvers (15) obtained approximately 30 s apart.

Experimental Protocol

After a telephone interview, subjects were scheduled for an initial visit to the laboratory where consent was obtained and a medical history questionnaire was completed. Baseline pulmonary function tests, a methacholine challenge test, and a 15-min exercise test designed to determine a work load that generated the target ventilatory rate of 25 L/min/m² body surface area were also completed on the initial visit. The subjects were randomly exposed to 0.2 ppm O₃ for 1 d or for 4 d, with a minimum break of 4 wk between exposure protocols. Each subject had his or her baseline SRaw, FVC, FEV₁, and peak flow measured 5 to 10 min before undergoing each daily exposure. Exposures were for 4 h on each study day, with subjects exercising for the first 30 min and then resting for the following 30 min of each hour. Each subject was given the option to exercise on a treadmill (Model M9.1; Precor Co., Bothell, WA) and/or a cycle ergometer (Model 818E; Monark, Varberg, Sweden). Tidal volume, respiratory rate, and ventilatory rate were measured 10 and 20 min into each 30-min exercise period and the work load was adjusted as needed to maintain the target ventilatory rate. Ventilation was not measured during rest periods; however, peak flow was measured 10 min into each 30-min rest period to monitor the pulmonary status of subjects during the 4-h exposure period. Subjects remained inside the chamber for the entire 4-h exposure period. Symptom questionnaires consisting of a 5-point rating scale (0 = not noticeable to 4 = severe) were administered immediately before and immediately after each exposure period. Subjects rated the following 11 symptoms: anxiety, chest discomfort or chest tightness, chest pain on deep inspiration, cough, eye irritation, headache, nasal irritation, nausea, phlegm or sputum production, shortness of breath, throat irritation, and wheezing.

Exposure Chamber and Atmospheric Monitoring

All exposures took place in a chamber ventilated with filtered air at 20° C and 50% relative humidity to which O₃ was added. The stainless steel and glass chamber, 2.5 × 2.5 × 2.4 m (Model W00327-3R; Nor-Lake, Hudson, WI), was custom-built and designed to maintain chamber temperature and relative humidity within 1.0° C and 2%, respectively, of the set points (DSC 8500; Johnson Controls, Poteau, OK) (16). Relative humidity and temperature were recorded every 30 s and averaged over each exposure.

O₃ was produced with a corona-discharge O₃ generator (Model T 408; Polymetrics, Inc., San Jose, CA) and analyzed with an ultraviolet light photometer (Model 1004AH; Dasibi, Glendale, CA). O₃ concentration was measured every 30 s, displayed in real-time (Labview 2; National Instruments, Austin, TX), and stored by a microcomputer (Model IIsi; Apple Computer Inc., Cupertino, CA). The O₃ analyzer was calibrated biannually with an O₃ transfer standard (Model 1003PC; Dasibi) by the California Air Resources Board and precision-checked on a monthly basis.

Mean ± SD for O₃ concentration, temperature, and percent relative humidity for the two exposure arms are listed in Table 2. Mean exposure conditions were not significantly different on any day, and there were no differences in mean temperature, relative humidity, or O₃ between the single-day exposure and the 4-d exposures.

Bronchoscopic and Lavage Procedures

Bronchoscopies were performed 19.7 ± 0.8 h (SD) after single-day O₃ exposure and 19.9 ± 0.8 h (SD) after 4-d O₃ exposure, in a dedicated room at San Francisco General Hospital. The procedures of bronchoscopy and BAL have been discussed previously in detail (8). Briefly, intravenous access was established, supplemental O₂ was delivered, and the upper airways were anesthetized with topical lidocaine. The bronchoscope (FB 18x; Pentax Precision Instruments Corp., Orangeburg, NY) was introduced through the mouth and into the airways. The bronchoscope was then directed into the right middle lobe where BAL was performed with three 50-ml aliquots of 0.9% saline warmed to 37° C. The first 15 ml of fluid returned from the first 50-ml aliquot was separated and labeled bronchial fraction, whereas the remaining fluid returned was labeled BAL. All lavage samples were immediately put on ice and taken to the laboratory for processing. After bronchoscopy, each subject was transported to the General Clinical Research Center (GCRC) at San Francisco General Hospital for a two-hour or more recovery period.

Total cells were counted on uncentrifuged aliquots of bronchial fraction and BAL using a hemacytometer. Differential cell counts were obtained from slides prepared using a cytocentrifuge (Cytospin 2; Shandon Southern Products, Ltd, Astmoor, UK), 25 g × 5 min, and stained with Diff-Quik (American Scientific Products, McGaw Park, IL) as previously described (16). Bronchial fraction and BAL fluids were then centrifuged at 180 g for 15 min, and the supernatant was separated and recentrifuged at 1,200 g for 15 min to remove any cellular debris prior to freezing at 70° C.

Measurement of Biochemical Constituents of Lavage Fluid

Lavage fluid and biochemical constituents were measured in bronchial fraction and BAL. Lactate dehydrogenase (LDH) was measured within 30 min of lavage time using a commercially available reagent (No. 228-20; Sigma Chemical Co., St. Louis, MO) and a spectrophotometer (DU 65; Beckman Instruments Inc., Fullerton, CA). The other biochemical assays were performed on lavage supernatants that had been frozen at 70° C. Total protein was assayed by a modification of the Lowry procedure (17). Lavage concentrations of fibronectin were determined with an antibody-capture immunoassay, as described by Miles and Hales (18), with minor modifications (16). Interleukin-6 (IL-6), interleukin-8 (IL-8), and GM-CSF were measured with commercially available immunoassays (R&D Systems, Minneapolis, MN).

Statistical Analysis

All data analyses were performed using Statview 4.5 software (Abacus Concepts, Inc., Berkeley, CA) and data were summarized as means ± SD or SE. If the measured variables had a normal distribution, the Student's paired t test was used to compare paired data within the 4-d exposures and between the exposure arms. If the variable did not have a normal distribution, Wilcoxon's signed rank test was used for comparisons.

The lower respiratory symptom scores that we analyzed included chest discomfort or chest tightness, chest pain on deep inspiration, cough, phlegm or sputum production, shortness of breath, and wheezing. Symptom score differences were determined by subtracting the preexposure symptom score from the postexposure symptom score. The sum of symptom-score differences across the single-day O₃ exposure versus the sum across Day 4 of the 4-d exposure were compared using Wilcoxon's signed rank test. A p value of 0.05 was considered statistically significant in all data analyses.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Pulmonary Function Results

The mean percent change across each study day for FEV₁, FVC, and SRaw are shown in Figure 1.

View larger version (16K): [in this window] [in a new window]   Figure 1.   Mean percent changes in pulmonary function values from baseline of Day 1 (zero values) and percent change across each day (preexposure to postexposure) of single-day and 4-d exposure to O3. Prebronchoscopy data points = percent change calculated from baseline of day prior to bronchoscopy; n = 15.

No differences were found in preexposure baseline measurements for each day of the single-day and 4-d arms of the study protocol. As expected, significant decreases (p < 0.001) were found in the preexposure to postexposure FEV₁ and FVC across the single-day exposure and Day 1 of the 4-d exposure. Significant preexposure to postexposure decreases in FEV₁ and FVC were also found across Days 2 (p < 0.0001) and 3 (p  0.024), but not across Day 4 (p  0.10) of the multiday protocol. Although SRaw increased across exposures on Days 1 to 3, none of these increases achieved statistical significance. Significant differences consistent with attenuation were noted when we compared mean percent change in FEV₁ (p < 0.0001), FVC (p < 0.0001), and SRaw (p < 0.04) across Day 1 of the single-day exposure with the mean percent change across Day 4 of the 4-d exposure.

There were no differences in minute ventilation data when comparing means from all exposure days. During the course of each daily 4-h exposure, ventilation patterns changed in response to O₃, with increased respiratory rate and decreased tidal volume, but the subjects still achieved target ventilatory rates.

Symptom Results

The mean change ± SD in lower respiratory symptom scores across the single-day exposure was 4.7 ± 4.2. In the 4-d exposure arm, the mean changes were 4.3 ± 2.8, 4.7 ± 3.8, 1.9 ± 2.7, and 0.6 ± 1.1, across Days 1 to 4, respectively. The mean lower respiratory score changes on Days 1 and 2 were significantly higher than on Days 3 and 4.

Lavage Fluid Results

The mean bronchial fraction and BAL data ± SE after single-day and 4-d exposures are shown in Table 3. Leukocyte cell counts for bronchial fraction and BAL are shown in Figure 2.

View larger version (22K): [in this window] [in a new window]   Figure 2.   Mean leukocyte counts for bronchial fraction and BAL; n = 15, asterisks indicate p < 0.05.

In the bronchial fraction we found the following significant differences when comparing the results after 4-d exposure to those after single-day exposure: lower number of neutrophils, higher number of macrophages, and a lower fibronectin concentration (p = 0.03, 0.03, and 0.002, respectively). In BAL, significantly lower values after 4-d exposure compared with after single-day exposure were found in total cell count, neutrophil count, fibronectin, and IL-6 concentrations (p = 0.02, 0.006, 0.01, 0.005, respectively). In addition, a higher percentage of macrophages was found after 4-d exposure (p = 0.011). Only one lavage fluid end point showed an increase after 4-d exposure consistent with progressive injury, mean LDH concentration in the bronchial fraction (p = 0.18). We graphed individual LDH data to determine if a subset of subjects was responsible for this mean increase in the bronchial fraction (Figure 3). Of the 13 subjects for whom bronchial fraction LDH data were available, seven showed increases (four with  2-fold increase).

View larger version (21K): [in this window] [in a new window]   Figure 3.   Individual LDH values for subjects from bronchial fraction and BAL after single-day versus 4-d O3 exposure. Group means are indicated by plus signs next to each column of data points; n = 13.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Contrary to our hypothesis, the results of this study did not show progression of the acute inflammatory response to O₃ with repeated short-term exposures. The BAL end points analyzed in our study tended to follow the pattern of attenuation observed previously in the pulmonary function and symptom responses to repeated exposures to O₃. Neutrophils (both in absolute number and as a percentage of total cells), fibronectin, and IL-6 were significantly decreased in BAL after the 4-d exposure. There was lack of attenuation, but no evidence of progressive increases in BAL total protein and LDH. Although there was attenuation of some bronchial fraction end points after the 4-d O₃ exposure (e.g., neutrophils and fibronectin), there was no attenuation of IL-6, IL-8, total protein, and LDH.

When we designed this study, there were no published data available regarding BAL findings in human subjects after multiday O₃ exposure. Recently, however, Devlin and colleagues (19) reported the results of a study comparing lavage end points in 16 subjects after exposure to 0.4 ppm O₃ for 2 h on five consecutive days to those obtained after exposures to filtered air for five consecutive days. The results from this study were compared with the results of previous studies of single-day O₃ exposure by this group. Although the investigators found attenuation of a number of inflammatory end points in BAL, including percentage of neutrophils and levels of IL-6 and fibronectin, there was no attenuation of LDH, IL-8, and total protein responses. Although the different design of the study of Devlin and colleagues precludes direct comparison with ours, the results of the two studies are consistent in showing dampening of O₃-induced neutrophil recruitment to the airways with repeated short-term exposures. The results of the study of Devlin and colleagues also support our finding of persistent O₃-induced airway epithelial injury with multiday exposure.

Attenuation of O₃-induced pulmonary function and BAL neutrophil responses with repeated exposures has been previously reported in animal studies (11, 20). Increased macrophages, which we observed in bronchial fraction, have been demonstrated in rats exposed to O₃ for 7 d (21). Animal data also support our finding of the lack of attenuation of O₃- induced release of LDH with repeated exposures (20). At least two animal studies have shown histologic evidence of progressive tissue injury at the level of the terminal bronchioles after 5 to 7-d of exposure (11, 22).

The major limitation of our study was the relatively small sample size. This is a generic problem in controlled human exposure studies involving bronchoscopy. Despite the small sample size, however, we still had adequate power to detect significant differences in several lavage end points between the single-day and 4-d O₃ exposures. If we had studied more subjects, the trends toward decreases in IL-8 and GM-CSF that were observed in BAL after 4-d exposure as compared with single-day exposure may have reached statistical significance.

In conclusion, this study indicates that although O₃-induced neutrophil recruitment to the respiratory tract is attenuated with repeated short-term exposures, airway epithelial injury may continue to occur. Persistence of such injury may lead to airway remodeling, which has been observed in several animal studies. Our results suggest the need for further investigation of the potential for recurrent or chronic exposures to O₃ to cause structural damage to the respiratory tract in humans.