Indomethacin Does Not Inhibit the Ozone-induced Increase in Bronchial Responsiveness in Human Subjects1 •2
RODNEY L. YING, KENNETH B. GROSS, THOMAS S. TERZO, and WILLIAM L. ESCHENBACHER
Introduction
Exposure to ozone in humans can, depending on the concentration and duration of exposure,induce transient changes in lung function. These changes include decreases in FVC, FEV 1, flow rates (FEF25-75), and increases in specific airway resistance (SRaw) (1-9). In addition, ozone causes increases in airway responsiveness to nonspecific stimuli such as methacholine (9-14). The magnitude of these physiologic effects has clearly been shown to be determined by the dose of ozone, which is related to the concentration of ozone, the duration of exposure, and the minute ventilation (1, 3, 8, 10). For example, previous results from this laboratory (9) revealed that exposure to 0.4 ppm of ozone for 2 h with intermittent moderate exercise(approximately 52 to 55 L/min of minute ventilation) resulted in an average decrease in FEV 1 of 13.20/0 for normal subjects and 24.00/0 for subjects with mild asthma. Although the etiology of these ozone-induced pulmonary changes is unknown, human and animal studies have suggested that products of the cyclooxygenasepathway of arachidonate metabolism may be involved (1416). In addition, mediators such as thromboxane A 2 , PGF2 a , and PGE2 , when pharmacologically applied to the airways, are known to have an irritant effect in the lung with resultant cough, pain, and changes in airway caliber (17-19). Indomethacin, an inhibitor of the cyclooxygenase pathway, has been shown to block the increase in airway responsiveness induced by high levels of ozone (3.0 ppm) in dogs (20). More recently,Schelegleand coworkers (16) have shown that indomethacin can partially attenuate the ozone-induced lung function changes in human subjects. Specifically, they demonstrated that pretreatment with indomethacin preventedmuch of the ozoneinduced decrease in FEV 1 and FVC. Thus, it was our purpose not only to evaluate the effect that indomethacin might have on ozone-induced lung function changes, as Schelegle had, but more
SUMMARY Exposure of human subjects to sufficiently high levels of ozone can result in reversible changes In lung function (restrictive In nature) and increases In nonspecific airway responsiveness. Several studies have implicated products of cyclooxygenase metabolism In the mediation of these changes. The purpose of this study was to determine If Indomethacin (a cyclooxygenase Inhibitor) would alter the changes In the ozone-induced Increase in responsiveness to methacholine or the ozone-Induced decrease In lung function. Thirteen male subjects underwent three randomly assigned 2-h exposures to 0.4 ppm ozone with alternating 15-min periods of rest and exercise on a cycle ergometer (30 Umln/m 2 , body surface area). For the 4 days before each of the exposures, the subjects received either Indomethacin (150 mg/day) or placebo, or no medication. Of the 13 subjects, only seven had both detectable Indomethacin serum levels on the Indomethacin StUdy Day and a significant increase in bronchial responsiveness to methacholine on the No Medication Day. For this group of seven subjects, we found that Indomethacin did not alter the ozone-Induced increase in bronchial responsiveness to methacholine (decrease in PC1 ooSRaw for the different stUdy days: no medication, -78.4 ± 5.3% [mean ± SEM]; placebo, -48.9 ± 12.2%; indomethacin, -64.5 ± 6.3%; p > 0.2), although indomethacin did attenuate the ozone-induced decrease in lung function. The decrease In the FEV1 for the different study days was as follows: no medication, - 20.7 ± 5.0% (mean ± SEM); placebo, -19.2 ± 6.3%; indomethacin, -4.8 ± 3.7% (p < 0.001). These results suggest that ozone-induced changes In lung function are mediated in part by cyclooxygenase products, but that the ozone-induced increase in bronchial responsiveness occurs by some other mechanism. AM REV RESPIR DIS 1990; 142:817-821
importantly to determine if indomethacin could inhibit the ozone-induced increase in bronchial responsiveness in humans as it had in the canine model. Methods Subject Selection Approval for this study was obtained from the Human Research Committees at both the University of Michigan Medical Center and the General Motors Research Laboratories. Volunteers were recruited through advertisements at the University of Michigan. Informed consent was obtained prior to entry into the study. Thirteen nonsmoking male volunteer subjects between 18 and 31 yr of age were recruited and completed the entire protocol. All subjects underwent an initial characterization, which included a medical history, spirometry, and bronchoprovocation testing with methacholine (see below). Each subject had no history of recent viral infection, had normal pulmonary function tests, and had a provocation concentration-l00 for methacholine (PC 10 oSRaw) of 5.0 mg/ml or greater. The PC 10 0SRaw is calculated from a methacholine provocation test and refers to the concentration of methacholine in milligrams per milliliter that would have resulted in a 100070 increase in the baseline specific airway resistance (SRaw). Characteristics of the 13 subjects are shown in table 1.
Apparatus and Exposure Chamber Ozone exposures were conducted in a stainless steel chamber 4 x 4 x 3 m maintained at 22 0 C and 50070 relative humidity. Ozone was generated from pure oxygen using a corona discharge ozone generator (Model T-148; Welsbach Ozone System, Philadelphia, PAl. The ozone was metered with a mass flow controller and diluted with filtered dry air before entry into the chamber. Ozone concentration was continuously monitored with an ultraviolet ozone photometer (Modell003-PC; Dasibi, Glendale, CAl. During the exposure period, the subjects exercised on a cycleergometer (TYpe KEM-2; Mijnhardt, Odijk, The Netherlands). During exercise, exhaled minute ventilation was measured by integration of the flow signal from a pneumotachograph (System 2001; Medical Graphics Corp., St. (Received in original form November 8, 1989 and in revised form April 2, 1990) 1 From Pulmonary and Critical Care Medicine, University of Michigan Medical Center, Ann Arbor, and the Biomedical Science Department, General Motors Research Laboratories, Warren, Michigan. 2 Correspondenceand requestsfor reprintsshould be addressed to William L. Eschenbacher, M.D., F988, The Methodist Hospital, Pulmonary and Critical Care Medicine, Baylor College of Medicine, 6565 Fannin, Houston, TX 77030.
817
YING, GROSS, TERZO, AND ESCHENBACHER
818 TABLE 1 SUBJECT CHARACTERISTICS Subject No. 1 2 3 4 5 6 7 8 9 10 11 12 13
Indomethacin t (fJg/m/)
Age (yr)
Height (em)
Weight (kg)
PC10oSRaw* mg/ml
22 22 26 31 27 30 28 27 22 24 23 18 26
170 178 163 188 170 173 185 185 188 190 178 180 185
68 73 61 89 72 66 73 76 75 91 57 66 68
> 8.0
0.7
7.7 > 8.0 > 8.0 > 8.0 > 8.0 > 8.0 > 8.0 8.0 6.7 > 8.0 > 8.0 5.4
< DL 0.7
< DL 0.6
< DL < DL 0.4 0.4 0.4 1.1 < DL 1.2
* PC100SRaw is the baseline bronchial responsiveness for each subject defined as the concentration of methacholine in mg/ml that would have resulted in a 100% increase in the baseline specific airway resistance. t Indomethacin serum concentration on the Indomethacin Study Day; DL is the detectable limit of indomethacin for this assay: 0.3 J,1g/ml.
Paul, MN). Airway resistance, thoracic gas volume, and functional residual capacity were measured using a constant-volume, variablepressure body plethysmograph (System 1085; Medical Graphics). Slow vital capacity and forced expiratory volumes and flow rates were measured from a pneumotachograph in a pulmonary function analyzer (System 1070; Medical Graphics).
Pulmonary Function Measurements and Bronchoprovocation Testing Specific airway resistance was calculated as the product of airway resistance and thoracic gas volume. Mean values of specific airway resistance and thoracic gas volume were derived from fivesuccessivemeasurements. Slow vital capacity was chosen as the greater of at least two successive efforts, and inspiratory capacity was then derived from the chosen effort. Total lung capacity was calculated as the sum of the mean thoracic gas volume and the inspiratory capacity. Forced expiratory volumes and flow rates were derived from the better of at least two forced expiratory maneuvers as determined by American Thoracic Society guidelines (21). The best test was defined as the one with the greater sum of FVC and FEV 1 • Bronchoprovocation testing was performed using methacholine delivered as an aerosol from a DeVilbiss no. 646 nebulizer equipped with a dosimeter (DeVilbissCorp., Somerset, PAl. The dosimeter was built locally at the General Motors Research Laboratories and consisted of a manually triggered solenoid valve that was electronically controlled to deliver 1.0 s of compressed air at 20 psi. Each concentration of methacholine was administered during five successive slow inhalations from FRC to TLC. Successive concentrations weregiven at approximately 5-min intervals, and SRaw was measured after each. The initial concentration of methacholine was 1 mg/ml. The methacholine concentration was doubled until a 100070 increase in the base-
line SRaw was achieved or a maximum concentration of 128 mg/ml was given. The concentration required to double the baseline SRaw (PClOoSRaw) was determined by interpolation.
Protocol Each subject underwent three separate exposures to 0.4 ppm ozone. Exposures were separated by at least 2 wk. Each subject was randomized in a double-blind fashion to receive either indomethacin (75mg twice a day for 4 days prior to exposure day) or placebo (twice a day for 4 days prior to exposure day), or no drug prior to each of the exposures. In addition, on the day of exposure if randomized to indomethacin or placebo, each subject was to receive a dose prior to arrival at the study site. At the beginning of each study day, subjects underwent a complete series of pulmonary function tests as a baseline, including measurement of SRaw,thoracic gas volume, slow vital capacity and forced expiratory volumes and flow rates. Measurement of SRaw was repeated after the forced expiratory maneuver to determine if airway resistance had increased. If SRaw was found to be elevated, it was remeasured at 5-min intervals until it had returned to within 10070 of the baseline value. This value was then used as the starting point for the bronchoprovocation testing. Progressively increasing concentrations of methacholine were then administered as described above, and the baseline PC 100SRaw was determined. A period of 90 min was used to allow SRaw to return to baseline, and SRaw and forced expiratory volumes and flow rates were then measured. Subjects then entered the exposure chamber and remained there for 2 h. The exposure period was divided into alternating 15-min intervals of rest and exercise on the cycle ergometer. During the first exercise period, subjects breathed through a mouthpiece, and minute ventilation was monitored. Work load on the cycle ergometer was adjusted so that a target
minute ventilation of 30 Lzmin/m" body surface area based on ideal body weight was achieved. Ideal weights were obtained from standard tables (22). During the remaining three exerciseperiods, the same work load was applied to the ergometer, and exhaled gas was not collected. This procedure was followed on each of the 3 study days for each subject. During all exercise periods, subjects wore noseclips, and continuous ECG monitoring was performed. After the final exercise period, subjects left the chamber, and measurements of SRaw, thoracic gas volume, slow vital capacity and forced expiratory volumes and flow rates were performed immediately. After these measurements, each subject had blood drawn for indomethacin drug levels. Ninety minutes after leaving the chamber, subjects underwent bronchoprovocation testing. The 90-min period allows for resolution of the ozone-induced changes in lung function before beginning the postexposure methacholine challenge. After completion of the protocol, subjects were offered a bronchodilator and were discharged when all symptoms had resolved. Indomethacin measurements were performed on all the blood samples that had been drawn after the exposure period. The methodology for the analysis used high performance liquid chromatography (Smith Kline Bioscience Laboratories, Philadelphia, PAl.
Data Analysis and Statistical Methods It was the purpose of this study to evaluate the effectiveness of indomethacin in preventing the ozone-induced increase in methacholine bronchial responsiveness. Therefore, it was decided to analyze the data from those subjects who had increases in bronchial responsiveness after ozone exposure. We arbitrarily decided that subjects who had ~ 50070 decrease in PC 100 SRaWafter ozone exposure on the No Medication Study Day would qualify as having an ozone-induced increase in bronchial responsiveness. This magnitude of change is equivalent to at least a log decrease in the concentration dose of methacholine. These subjects were therefore included in the analysis for the evaluation of indomethacin on altering the ozone-induced increase in bronchial responsiveness and lung function. Because we were interested in whether indomethacin would alter the ozone-induced pulmonary changes and because we had obtained indomethacin serum levels, we further decided to include in our data analysis only those subjects with a positive indomethacin serum concentration on the Indomethacin Study Day. For each pulmonary function parameter measured (FVC,FEV1 , etc.), mean values and mean percent change from baseline were determined using all combinations of the following two variables: drug (no medication, placebo, or indomethacin) and time (baseline and postexposure). Because doubling concentrations of methacholine were used, PC 100SRaw values were converted to logarithms, and the differences between preex-
INDOMETHACIN DOES NOT INHIBIT OlONE-INOUCED INCREASES IN BRONCHIAL RESPONSIVENESS
posure and postexposure values on each study day were compared using analys is of variance. For each parameter in which the null hypothesis was rejected, Duncan's multiple range test was used to determine where the differences existed (23). A significance level of p < 0.05 was applied in all statistical analyses.
Results
Of the thirteen subjects who completed the entire protocol, five had less than the detectable limit of indomethacin in their blood sample on the Indomethacin Study Day (table 1). The ozone-induced percent changes in PClOoSRaw and FEV 1 for all 13subjects on all 3 study days are shown in figures 1 and 2. The actual values for the PClOoSRaw and the FEV 1 before and after ozone exposure are shown in tables 2 and 3. The eight subjects with positive indomethacin concentrations on the Indomethacin Study Day are identified with asterisks. Of these eight subjects, seven had a > 50070 decrease in their PClOOSRaw for methacholine after ozone exposure on the No Medication Day. The line representing a 50% decrease in the PClOoSRaw or a decrease in one concentration level of methacholine (see above) is shown in figure 1. These seven subjects made up the group used for the data analysis to evaluate the efficacy of indomethacin in altering the ozone-induced increase in bronchial responsiveness. For these seven subjects, the average values for PClOoSRaw before and after the ozone exposure for the three separate study days are shown in figure 3. The ozone-induced percent changes in PClOoSRaw for the three study days (No Medication, Placebo, and Indomethacin) were -78.4 ± 5.3% (mean ± SEM) , -48.9 ± 12.2%, and -64.5 ± 6.3%, respectively. There was no significant difference for the percent decrease in PClOoSRaw on the Indomethacin Study Day when compared with those on either the No Medication Day or the Placebo Study Day (p > 0.2). The average values for the FEV 1 and FVC before and after ozone exposure for these seven subjects on the 3 study days are shown in figure 4. The average percent decreases in FEV 1 for the 3 study days (No Medication, Placebo, and Indomethacin) are -20.7 ± 7 .5% (mean ± SEM), -19.2 ± 6.3070, and -4.8 ± 3.7%, respectively (figure 5). The percent decreases in FVC for the 3 study days (No Medication, Placebo, and Indomethacin) are - 20.5 ± 3.4% (mean ± SEM), -17.5 ± 3.8%, and -4.1 ± 1.8%, respectively.The percent decreases in FEV 1
819 No Mediclltion
Plllcebo
rnuo met necrn
Placebo
I ndom ethaci n
125
o o I
U
Fig. 1. Percent change in PC lO .SRaw for methacholine after ozone exposure for 13 normal volunteers on 3 separate stUdy days. 'Subjects with positive in· domethacin levels on the indomethacin study day.
100 75
c,
c:
'" "
en c:
50 25
.&:
U
~
c:
...u'"
'"
c,
-25 -50 -75 -100
No Medication
->
20
I
10
UJ
....
c:
'"
Fig. 2. Percent change in FEV, after ozone exposure for 13 normal volunteers on 3 separate study days. 'Subjects with positive indomethacin levels on the lndomethacin study day.
en c: CD
- 10
.&:
U
~
- 20
...u'" '" c,
- 30 -4 0
- 50
TABLE 2 PC,ooSRaw' BEFORE AND AFTER EXPOSURE TO OZONE ON DIFFERENT STUDY DAYSt No Medicat ion
Subject No.
Before
After
1 2 3 4 5 6 7 8 9 10 11 12 13
128.0* 2.8 20.4 23.0 6.7 11.6 16.6 128.0 33.7 10.9 19.0 7.1 4.4
10.0 5.3 1.8 1.0 1.5 6.8 19.2 128.0 10.0 1.4 4.3 3.0 2.1
Placebo
Indomethacin
Before
After
Before
After
65.2 9.7 2.1 17.5 4.6 4.9 21.4 128.0 9.6 2.6 5.2 4.0 2.9
4.4 1.6 1.4 3.9 2.9 1.7 9.2 128.0 2.1 1.0 2.8 4.7 3.0
21.7 1.4 17.0 28.2 128.0 5.1 43.5 58.3 9.0 5.7 12.7 4.6 2.4
13.3 1.0 5.4 2.3 8.7 3.4 59.2 128.0 3.2 2.4 3.5 3.1 1.0
• PClOoSRaw is the concentration of methacholine in mg/ml that would have resulted in a 100% lncrease in baseline specific airway resistance. t Before and after the 2·h exposure to 0.4 ppm ozone. When the PC,ooSRaw was greater than 128 mg/ml, the value of 128 was used in the analysis.
*
and FVC for the Indomethacin Study Day were significantly less than for either the No Medication Day or the Placebo Study Day (p < 0.001). Discussion
In this study, we have demonstrated that indomethacin does not inhibit the ozone induced increase in bronchial responsiveness in humans. This finding is in con -
trast with the results of studies in which dogs were exposed to a markedly higher concentration of ozone (20). It had been shown in those studies thatindomethacin (20) and more specifically a thromboxane synthetase inhibitor (15)successfully prevent the ozone-induced increase in bronchial responsiveness in dogs. Despite its lack of effectiveness in preventing the increase in bronchial responsive-
YING, GROSS, TERZO, AND ESCHENBACHER
820 TABLE 3
20
u
> ....
FEV, BEFORE AND AFTER EXPOSURE TO OZONE ON DIFFERENT STUDY DAYS' Subject No.
1 2 3 4 5 6 7 8 9 10 11 12 13
. -> o
Placebo
No Medication
Indomethacin
Before
After
Before
Afte r
Before
After
4.52 5.21 3.10 4.25 3.19 4.24 4.41 4.30 4.76 4.38 4.49 4.66 5.08
3.92 4.14 2.27 3.16 2.81 2.23 4.29 4.20 3.50 2.71 3.19 4.17 5.13
4.47 5.30 3.22 4.35 3.19 3.62 4 .49 4 .20 5.29 4.88 4.59 4.70 4.75
3.54 4.97 2.19 4.36 2.93 2.44 4.73 4.15 4.81 2.93 3.16 4.38 5.08
4.37 5.49 3.20 4.48 3.30 3.78 4.44 4.18 4.50 4.67 4.53 4.66 4.47
4 .22 4 .38 3.03 3.36 3.14 2.87 4.41 4.34 4.07 3.89 4.15 4.44 5.13
'0
FE Y- l
Fye
I
U.I ....
C
'cc"n " B
-1 0
-2 0
~
c
..'" u
-30
'"
0..
-40
Fig. 5. Percent change in FEV, and FVC after ozone exposure on 3 separate study days for the seven subjects who met the inclusion criteria for the study (see text). Values showo are mean ± SEM. Closed bars = no medication; open bars = placebo; hatched bars = indomethacin .
• Before and after the 2·h exposure to 0.4 ppm ozone.
ness, indomethacin did appear to prevent the ozone-induced decrease in lung function (FEV! and FVC). It is possible that the dosage of indomethacin used by the subjects was not sufficient to inhibit cyclooxygenase enzymes in the cells of the airways for these subjects. However,(1)all the subjects that were included in the data analysis had positive serum concentrations of indomethacin on the Indomethacin Study Day; (2) these serum concentrations of indomethacin obtained were within the range believed to provide the antiinflammatory therapeutic effect (24); and (3) the ozone-induced decrease in lung function was reduced significantly on the Indomethacin Study Day. The magnitude of the inhibition of the ozone-induced lung function changes by indomethacin in our results matches quite closely that
.,c
60
'0
so
in the results of Schelegle and coworkers (16). These results suggest that cyclooxygenase products may be involved in the ozone-induced decreases in lung function. As mentioned earlier, animal and human studies (17-19) have shown that 6
5
..J
...
:>
3
W
u..
o
~
u
....," ~
40
E
....E en
30
E
o'
20
2
~ I
U
10
C-
o
After Before Ozone Exposure Fig . 3. PC, ••SRaw for methacholine before and after ozone exposure on 3 separate study days tor the seven subjects who met the inclusion criteria for the study (see text). Values shown are mean ± SEM . Closed bars = no med ication; open bars = placebo; hatched bars = indomethacin.
Before
After Ozone Exposure
Fig. 4. FEV, and FVC before and after ozone exposure on 3 separate study days for the seven subjects who metthe inclusion criteria for the study (see text). Values shown are mean ± SEM. Open squares = no medication ; closed diamonds = placebo; closed squares = indomethacin.
metabolites of arachidonic acid can cause irritation and bronchoconstriction. In addition, Leikauf and coworkers (25) have shown that ozone can enhance the production of arachidonic acid metabolites, including PGF2a from bovine tracheal epithelial cells. Thus, it is possible that products such as PGF2a , PGD 2 , or ThA2 are being released as a result of ozone exposure and that these products are producing effects that result in restriction of lung function. These results also suggest that there are separate mechanisms for the ozoneinduced increases in bronchial responsivenessand the ozone-induced decreases in lung function (FEV! and FVC). As stated above, the mechanism for the latter effect does seem to involve products of cyclooxygenase metabolism. The mechanism for ozone-induced increases in bronchial responsiveness in humans remains unknown. It is known from other studies that ozone can alter airway epithelial function in humans (26). It is possible that alterations in the airway epithelial lining could result in a decrease in a normally present epithelial-derived smooth muscle relaxant factor (27, 28) or that the increase in epithelial permeability could allow increased access of the inhaled methacholine to the airway smooth muscle. Another possibility is that in addition to ozone-induced alterations in the epithelial cells, there could be alterations in the neural elements in the airway lining that could result in an enhancement in the release of neuropeptides that could in turn contribute to the increase in airway smooth muscle responsiveness to methacholine. Finally, it is also conceivable that ozone-induced cellular production of other mediators is occurring.
INDOMETHACIN DOES NOT INHIBIT OlONE-INDUCED INCREASES IN BRONCHIAL RESPONSIVENESS
These other mediators could include agents such as platelet-activating factor (PAF) that, if present in increased concentrations, could be responsible for the increase in bronchial responsiveness. In conclusion, we have shown that although indomethacin can inhibit the ozone-induced decreasesin lung function in human subjects, it does not appear to prevent the ozone-induced increases in methacholine bronchial responsiveness in these same subjects. Ozone-induced increases in bronchial responsiveness in humans does not appear to involve cyclooxygenase metabolites. Acknowledgment The writers thank Messrs. James D'Arcy and Robert Wooley for their operation and maintenance of the human chamber exposure facility. References 1. Adams WC, Savin WM, Christo AE. Detection of ozone toxicity during continuous exercise via the effective dose concept. J Appl Physiol1981; 51:415-22. 2. Folinsbee LJ, Bedi JF, Horvath SM. Pulmonary function changes after 1 hour continuous heavy exercise in 0.21 ppm ozone. J Appl Physiol 1984; 57:984-8. 3. McDonnell WF, Horstman DH, Hazucha MJ, et al. Pulmonary effects of ozone exposure during exercise: dose-responsecharacteristics. J Appl Physiol 1983; 54:1345-52. 4. Gibbons SI, Adams WC. Combined effects of ozone exposure and ambient heat on exercising females. J Appl Physiol 1984; 57:450-6. 5. DeluciaAJ, Adams WC. Effects of 0 3 inhalation during exercise on pulmonary function and blood biochemistry. J Appl Physiol1977; 43:75-81.
6. Beckett WS, McDonnell WF, Horstman DH, House DE. Role of the parasympathetic nervous systemin acute lung responseto ozone. J Appl Physiol 1985; 59:1879-85. 7. Adams WC, SchelegleES. Ozone and high ventilation effects on pulmonary function and endurance performance. J Appl Physiol1983; 55:805-12. 8. Silverman F, FolinsbeeLJ, Barnard J, Shephard RJ. Pulmonary function changes in ozone-interaction of concentration and ventilation. J Appl Physiol 1976; 41:859-64. 9. Kreit JW, Gross KB, Moore TB, Lorenzen TJ, D'Arcy J, Eschenbacher WL. Ozone-induced changes in pulmonary function and bronchial responsiveness in asthmatics. J Appl Physiol 1989; 66:217-22. 10. Hazucha MJ. Relationship between ozone exposure and pulmonary function changes. J Appl Physiol 1987; 62:1671-80. 11. DimeoMJ, GlennMG, HoltzmanMJ, Sheller JR, Nadel JA, Boushey HA. Threshold concentration of ozone causing an increase in bronchial reactivity in humans and adaptation with repeated exposures. Am Rev Respir Dis 1981; 124:245-8. 12. Golden JA, Nadel JA, Boushey HA. Bronchial hyperirritability in healthy subjects after exposure to ozone. Am Rev Respir Dis 1978; 118: 287-94. 13. Holtzman MJ, Cunningham JH, Sheller JR, IrsiglerGB, Nadel JA, BousheyHA. Effect of ozone on bronchial reactivityin atopic and nonatopic subjects. Am Rev Respir Dis 1979; 120:1059-67. 14. Seltzer J, Bigby BG, Stulbarg M, et al. Ozone induced change in bronchial reactivityto methacholine and airway inflammation in humans. J Appl Physiol 1986; 60:1321-6. 15. Aizawa H, Chung KF, Leikauf GD, et al. Significance of thromboxane generation in ozone induced airway hyperresponsiveness in dogs. J Appl Physiol 1985; 59:1918-23. 16. Schelegle ES, Adams WC, Seifkin AD. Indomethacin pretreatment reduces ozone induced pulmonary function decrements in human subjects. Am Rev Respir Dis 1987; 136:1350-4. 17. Gardiner PJ, Copas JL, Elliot RD, Collier HOJ. Tracheobronchial irritancy of inhaled pros-
821
taglandins in the conscious cat. Prostaglandins 1978; 15:303-15. 18. Smith AP, Cuthbert MF, Dunlop LS. Effects of inhaled prostaglandins E 1,E2 , F 2 U on the airway resistance of healthy and asthmatic man. Clin Sci Mol Med 1975; 48:421-30. 19. Svensson J, Strandberg K, Tuvemo T, Hamberg M. Thromboxane A 2 : effects on airway and vascular smooth muscle. Prostaglandins 1977; 14: 425-36. 20. O'Byrne PM, Walters EH, Aizawa H, Fabbri LM, Holtzman MJ, Nadel JA. Indomethacin inhibits the airway hyperresponsiveness but not the neutrophil influx induced by ozone in dogs. Am Rev Respir Dis 1984; 130:220-4. 21. American Thoracic Society. Standardization of spirometry: 1987 update. Am Rev Respir Dis 1987; 136:1285-98. 22. Foster DW. Eating disorders: obesity and anorexia nervosa. In: Wilson JD, Foster DW,eds. Williams textbook of endocrinology.Philadelphia: WB Saunders, 1985; 1083. 23. Duncan DB. Multiple range and multiple F tests. Biometrics 1955; 11:1-42. 24. Flowers RJ, Moncada S, Vane JR. Analgesicantipyretics and anti-inflammatory agents; drugs employed in the treatment of gout. In: Gilman AG, Goodman LS, RaIl TW, Murad F, eds. The pharmacologic basis of therapeutics. New York: Macmillan, 1985; 695. 25. Leikauf GD, Driscoll KE, Wey HE. Ozoneinduced augmentation of eicosanoid metabolism in epithelial cells from bovine trachea. Am Rev Respir Dis 1988; 137:435-42. 26. Kehrl HR, Vincent LM, Kowalsky RJ, et al. Ozone exposure increases respiratory epithelial permeability in humans. Am RevRespir Dis 1987;135: 1124-8. 27. Stuart-Smith K, Vanhoutte PM. Airway epithelium modulates the responsiveness of porcine bronchial smooth muscle. J Appl Physiol1988; 65: 721-7. 28. Jones GL, Lane CG, Daniel EE, O'Byrne PM. Release of epithelium derived relaxing factor after ozone inhalation in dogs. J Appl Physiol1988; 65: 1238-43.
RODNEY L. YING, KENNETH B. GROSS, THOMAS S. TERZO, and WILLIAM L. ESCHENBACHER
Introduction
Exposure to ozone in humans can, depending on the concentration and duration of exposure,induce transient changes in lung function. These changes include decreases in FVC, FEV 1, flow rates (FEF25-75), and increases in specific airway resistance (SRaw) (1-9). In addition, ozone causes increases in airway responsiveness to nonspecific stimuli such as methacholine (9-14). The magnitude of these physiologic effects has clearly been shown to be determined by the dose of ozone, which is related to the concentration of ozone, the duration of exposure, and the minute ventilation (1, 3, 8, 10). For example, previous results from this laboratory (9) revealed that exposure to 0.4 ppm of ozone for 2 h with intermittent moderate exercise(approximately 52 to 55 L/min of minute ventilation) resulted in an average decrease in FEV 1 of 13.20/0 for normal subjects and 24.00/0 for subjects with mild asthma. Although the etiology of these ozone-induced pulmonary changes is unknown, human and animal studies have suggested that products of the cyclooxygenasepathway of arachidonate metabolism may be involved (1416). In addition, mediators such as thromboxane A 2 , PGF2 a , and PGE2 , when pharmacologically applied to the airways, are known to have an irritant effect in the lung with resultant cough, pain, and changes in airway caliber (17-19). Indomethacin, an inhibitor of the cyclooxygenase pathway, has been shown to block the increase in airway responsiveness induced by high levels of ozone (3.0 ppm) in dogs (20). More recently,Schelegleand coworkers (16) have shown that indomethacin can partially attenuate the ozone-induced lung function changes in human subjects. Specifically, they demonstrated that pretreatment with indomethacin preventedmuch of the ozoneinduced decrease in FEV 1 and FVC. Thus, it was our purpose not only to evaluate the effect that indomethacin might have on ozone-induced lung function changes, as Schelegle had, but more
SUMMARY Exposure of human subjects to sufficiently high levels of ozone can result in reversible changes In lung function (restrictive In nature) and increases In nonspecific airway responsiveness. Several studies have implicated products of cyclooxygenase metabolism In the mediation of these changes. The purpose of this study was to determine If Indomethacin (a cyclooxygenase Inhibitor) would alter the changes In the ozone-induced Increase in responsiveness to methacholine or the ozone-Induced decrease In lung function. Thirteen male subjects underwent three randomly assigned 2-h exposures to 0.4 ppm ozone with alternating 15-min periods of rest and exercise on a cycle ergometer (30 Umln/m 2 , body surface area). For the 4 days before each of the exposures, the subjects received either Indomethacin (150 mg/day) or placebo, or no medication. Of the 13 subjects, only seven had both detectable Indomethacin serum levels on the Indomethacin StUdy Day and a significant increase in bronchial responsiveness to methacholine on the No Medication Day. For this group of seven subjects, we found that Indomethacin did not alter the ozone-Induced increase in bronchial responsiveness to methacholine (decrease in PC1 ooSRaw for the different stUdy days: no medication, -78.4 ± 5.3% [mean ± SEM]; placebo, -48.9 ± 12.2%; indomethacin, -64.5 ± 6.3%; p > 0.2), although indomethacin did attenuate the ozone-induced decrease in lung function. The decrease In the FEV1 for the different study days was as follows: no medication, - 20.7 ± 5.0% (mean ± SEM); placebo, -19.2 ± 6.3%; indomethacin, -4.8 ± 3.7% (p < 0.001). These results suggest that ozone-induced changes In lung function are mediated in part by cyclooxygenase products, but that the ozone-induced increase in bronchial responsiveness occurs by some other mechanism. AM REV RESPIR DIS 1990; 142:817-821
importantly to determine if indomethacin could inhibit the ozone-induced increase in bronchial responsiveness in humans as it had in the canine model. Methods Subject Selection Approval for this study was obtained from the Human Research Committees at both the University of Michigan Medical Center and the General Motors Research Laboratories. Volunteers were recruited through advertisements at the University of Michigan. Informed consent was obtained prior to entry into the study. Thirteen nonsmoking male volunteer subjects between 18 and 31 yr of age were recruited and completed the entire protocol. All subjects underwent an initial characterization, which included a medical history, spirometry, and bronchoprovocation testing with methacholine (see below). Each subject had no history of recent viral infection, had normal pulmonary function tests, and had a provocation concentration-l00 for methacholine (PC 10 oSRaw) of 5.0 mg/ml or greater. The PC 10 0SRaw is calculated from a methacholine provocation test and refers to the concentration of methacholine in milligrams per milliliter that would have resulted in a 100070 increase in the baseline specific airway resistance (SRaw). Characteristics of the 13 subjects are shown in table 1.
Apparatus and Exposure Chamber Ozone exposures were conducted in a stainless steel chamber 4 x 4 x 3 m maintained at 22 0 C and 50070 relative humidity. Ozone was generated from pure oxygen using a corona discharge ozone generator (Model T-148; Welsbach Ozone System, Philadelphia, PAl. The ozone was metered with a mass flow controller and diluted with filtered dry air before entry into the chamber. Ozone concentration was continuously monitored with an ultraviolet ozone photometer (Modell003-PC; Dasibi, Glendale, CAl. During the exposure period, the subjects exercised on a cycleergometer (TYpe KEM-2; Mijnhardt, Odijk, The Netherlands). During exercise, exhaled minute ventilation was measured by integration of the flow signal from a pneumotachograph (System 2001; Medical Graphics Corp., St. (Received in original form November 8, 1989 and in revised form April 2, 1990) 1 From Pulmonary and Critical Care Medicine, University of Michigan Medical Center, Ann Arbor, and the Biomedical Science Department, General Motors Research Laboratories, Warren, Michigan. 2 Correspondenceand requestsfor reprintsshould be addressed to William L. Eschenbacher, M.D., F988, The Methodist Hospital, Pulmonary and Critical Care Medicine, Baylor College of Medicine, 6565 Fannin, Houston, TX 77030.
817
YING, GROSS, TERZO, AND ESCHENBACHER
818 TABLE 1 SUBJECT CHARACTERISTICS Subject No. 1 2 3 4 5 6 7 8 9 10 11 12 13
Indomethacin t (fJg/m/)
Age (yr)
Height (em)
Weight (kg)
PC10oSRaw* mg/ml
22 22 26 31 27 30 28 27 22 24 23 18 26
170 178 163 188 170 173 185 185 188 190 178 180 185
68 73 61 89 72 66 73 76 75 91 57 66 68
> 8.0
0.7
7.7 > 8.0 > 8.0 > 8.0 > 8.0 > 8.0 > 8.0 8.0 6.7 > 8.0 > 8.0 5.4
< DL 0.7
< DL 0.6
< DL < DL 0.4 0.4 0.4 1.1 < DL 1.2
* PC100SRaw is the baseline bronchial responsiveness for each subject defined as the concentration of methacholine in mg/ml that would have resulted in a 100% increase in the baseline specific airway resistance. t Indomethacin serum concentration on the Indomethacin Study Day; DL is the detectable limit of indomethacin for this assay: 0.3 J,1g/ml.
Paul, MN). Airway resistance, thoracic gas volume, and functional residual capacity were measured using a constant-volume, variablepressure body plethysmograph (System 1085; Medical Graphics). Slow vital capacity and forced expiratory volumes and flow rates were measured from a pneumotachograph in a pulmonary function analyzer (System 1070; Medical Graphics).
Pulmonary Function Measurements and Bronchoprovocation Testing Specific airway resistance was calculated as the product of airway resistance and thoracic gas volume. Mean values of specific airway resistance and thoracic gas volume were derived from fivesuccessivemeasurements. Slow vital capacity was chosen as the greater of at least two successive efforts, and inspiratory capacity was then derived from the chosen effort. Total lung capacity was calculated as the sum of the mean thoracic gas volume and the inspiratory capacity. Forced expiratory volumes and flow rates were derived from the better of at least two forced expiratory maneuvers as determined by American Thoracic Society guidelines (21). The best test was defined as the one with the greater sum of FVC and FEV 1 • Bronchoprovocation testing was performed using methacholine delivered as an aerosol from a DeVilbiss no. 646 nebulizer equipped with a dosimeter (DeVilbissCorp., Somerset, PAl. The dosimeter was built locally at the General Motors Research Laboratories and consisted of a manually triggered solenoid valve that was electronically controlled to deliver 1.0 s of compressed air at 20 psi. Each concentration of methacholine was administered during five successive slow inhalations from FRC to TLC. Successive concentrations weregiven at approximately 5-min intervals, and SRaw was measured after each. The initial concentration of methacholine was 1 mg/ml. The methacholine concentration was doubled until a 100070 increase in the base-
line SRaw was achieved or a maximum concentration of 128 mg/ml was given. The concentration required to double the baseline SRaw (PClOoSRaw) was determined by interpolation.
Protocol Each subject underwent three separate exposures to 0.4 ppm ozone. Exposures were separated by at least 2 wk. Each subject was randomized in a double-blind fashion to receive either indomethacin (75mg twice a day for 4 days prior to exposure day) or placebo (twice a day for 4 days prior to exposure day), or no drug prior to each of the exposures. In addition, on the day of exposure if randomized to indomethacin or placebo, each subject was to receive a dose prior to arrival at the study site. At the beginning of each study day, subjects underwent a complete series of pulmonary function tests as a baseline, including measurement of SRaw,thoracic gas volume, slow vital capacity and forced expiratory volumes and flow rates. Measurement of SRaw was repeated after the forced expiratory maneuver to determine if airway resistance had increased. If SRaw was found to be elevated, it was remeasured at 5-min intervals until it had returned to within 10070 of the baseline value. This value was then used as the starting point for the bronchoprovocation testing. Progressively increasing concentrations of methacholine were then administered as described above, and the baseline PC 100SRaw was determined. A period of 90 min was used to allow SRaw to return to baseline, and SRaw and forced expiratory volumes and flow rates were then measured. Subjects then entered the exposure chamber and remained there for 2 h. The exposure period was divided into alternating 15-min intervals of rest and exercise on the cycle ergometer. During the first exercise period, subjects breathed through a mouthpiece, and minute ventilation was monitored. Work load on the cycle ergometer was adjusted so that a target
minute ventilation of 30 Lzmin/m" body surface area based on ideal body weight was achieved. Ideal weights were obtained from standard tables (22). During the remaining three exerciseperiods, the same work load was applied to the ergometer, and exhaled gas was not collected. This procedure was followed on each of the 3 study days for each subject. During all exercise periods, subjects wore noseclips, and continuous ECG monitoring was performed. After the final exercise period, subjects left the chamber, and measurements of SRaw, thoracic gas volume, slow vital capacity and forced expiratory volumes and flow rates were performed immediately. After these measurements, each subject had blood drawn for indomethacin drug levels. Ninety minutes after leaving the chamber, subjects underwent bronchoprovocation testing. The 90-min period allows for resolution of the ozone-induced changes in lung function before beginning the postexposure methacholine challenge. After completion of the protocol, subjects were offered a bronchodilator and were discharged when all symptoms had resolved. Indomethacin measurements were performed on all the blood samples that had been drawn after the exposure period. The methodology for the analysis used high performance liquid chromatography (Smith Kline Bioscience Laboratories, Philadelphia, PAl.
Data Analysis and Statistical Methods It was the purpose of this study to evaluate the effectiveness of indomethacin in preventing the ozone-induced increase in methacholine bronchial responsiveness. Therefore, it was decided to analyze the data from those subjects who had increases in bronchial responsiveness after ozone exposure. We arbitrarily decided that subjects who had ~ 50070 decrease in PC 100 SRaWafter ozone exposure on the No Medication Study Day would qualify as having an ozone-induced increase in bronchial responsiveness. This magnitude of change is equivalent to at least a log decrease in the concentration dose of methacholine. These subjects were therefore included in the analysis for the evaluation of indomethacin on altering the ozone-induced increase in bronchial responsiveness and lung function. Because we were interested in whether indomethacin would alter the ozone-induced pulmonary changes and because we had obtained indomethacin serum levels, we further decided to include in our data analysis only those subjects with a positive indomethacin serum concentration on the Indomethacin Study Day. For each pulmonary function parameter measured (FVC,FEV1 , etc.), mean values and mean percent change from baseline were determined using all combinations of the following two variables: drug (no medication, placebo, or indomethacin) and time (baseline and postexposure). Because doubling concentrations of methacholine were used, PC 100SRaw values were converted to logarithms, and the differences between preex-
INDOMETHACIN DOES NOT INHIBIT OlONE-INOUCED INCREASES IN BRONCHIAL RESPONSIVENESS
posure and postexposure values on each study day were compared using analys is of variance. For each parameter in which the null hypothesis was rejected, Duncan's multiple range test was used to determine where the differences existed (23). A significance level of p < 0.05 was applied in all statistical analyses.
Results
Of the thirteen subjects who completed the entire protocol, five had less than the detectable limit of indomethacin in their blood sample on the Indomethacin Study Day (table 1). The ozone-induced percent changes in PClOoSRaw and FEV 1 for all 13subjects on all 3 study days are shown in figures 1 and 2. The actual values for the PClOoSRaw and the FEV 1 before and after ozone exposure are shown in tables 2 and 3. The eight subjects with positive indomethacin concentrations on the Indomethacin Study Day are identified with asterisks. Of these eight subjects, seven had a > 50070 decrease in their PClOOSRaw for methacholine after ozone exposure on the No Medication Day. The line representing a 50% decrease in the PClOoSRaw or a decrease in one concentration level of methacholine (see above) is shown in figure 1. These seven subjects made up the group used for the data analysis to evaluate the efficacy of indomethacin in altering the ozone-induced increase in bronchial responsiveness. For these seven subjects, the average values for PClOoSRaw before and after the ozone exposure for the three separate study days are shown in figure 3. The ozone-induced percent changes in PClOoSRaw for the three study days (No Medication, Placebo, and Indomethacin) were -78.4 ± 5.3% (mean ± SEM) , -48.9 ± 12.2%, and -64.5 ± 6.3%, respectively. There was no significant difference for the percent decrease in PClOoSRaw on the Indomethacin Study Day when compared with those on either the No Medication Day or the Placebo Study Day (p > 0.2). The average values for the FEV 1 and FVC before and after ozone exposure for these seven subjects on the 3 study days are shown in figure 4. The average percent decreases in FEV 1 for the 3 study days (No Medication, Placebo, and Indomethacin) are -20.7 ± 7 .5% (mean ± SEM), -19.2 ± 6.3070, and -4.8 ± 3.7%, respectively (figure 5). The percent decreases in FVC for the 3 study days (No Medication, Placebo, and Indomethacin) are - 20.5 ± 3.4% (mean ± SEM), -17.5 ± 3.8%, and -4.1 ± 1.8%, respectively.The percent decreases in FEV 1
819 No Mediclltion
Plllcebo
rnuo met necrn
Placebo
I ndom ethaci n
125
o o I
U
Fig. 1. Percent change in PC lO .SRaw for methacholine after ozone exposure for 13 normal volunteers on 3 separate stUdy days. 'Subjects with positive in· domethacin levels on the indomethacin study day.
100 75
c,
c:
'" "
en c:
50 25
.&:
U
~
c:
...u'"
'"
c,
-25 -50 -75 -100
No Medication
->
20
I
10
UJ
....
c:
'"
Fig. 2. Percent change in FEV, after ozone exposure for 13 normal volunteers on 3 separate study days. 'Subjects with positive indomethacin levels on the lndomethacin study day.
en c: CD
- 10
.&:
U
~
- 20
...u'" '" c,
- 30 -4 0
- 50
TABLE 2 PC,ooSRaw' BEFORE AND AFTER EXPOSURE TO OZONE ON DIFFERENT STUDY DAYSt No Medicat ion
Subject No.
Before
After
1 2 3 4 5 6 7 8 9 10 11 12 13
128.0* 2.8 20.4 23.0 6.7 11.6 16.6 128.0 33.7 10.9 19.0 7.1 4.4
10.0 5.3 1.8 1.0 1.5 6.8 19.2 128.0 10.0 1.4 4.3 3.0 2.1
Placebo
Indomethacin
Before
After
Before
After
65.2 9.7 2.1 17.5 4.6 4.9 21.4 128.0 9.6 2.6 5.2 4.0 2.9
4.4 1.6 1.4 3.9 2.9 1.7 9.2 128.0 2.1 1.0 2.8 4.7 3.0
21.7 1.4 17.0 28.2 128.0 5.1 43.5 58.3 9.0 5.7 12.7 4.6 2.4
13.3 1.0 5.4 2.3 8.7 3.4 59.2 128.0 3.2 2.4 3.5 3.1 1.0
• PClOoSRaw is the concentration of methacholine in mg/ml that would have resulted in a 100% lncrease in baseline specific airway resistance. t Before and after the 2·h exposure to 0.4 ppm ozone. When the PC,ooSRaw was greater than 128 mg/ml, the value of 128 was used in the analysis.
*
and FVC for the Indomethacin Study Day were significantly less than for either the No Medication Day or the Placebo Study Day (p < 0.001). Discussion
In this study, we have demonstrated that indomethacin does not inhibit the ozone induced increase in bronchial responsiveness in humans. This finding is in con -
trast with the results of studies in which dogs were exposed to a markedly higher concentration of ozone (20). It had been shown in those studies thatindomethacin (20) and more specifically a thromboxane synthetase inhibitor (15)successfully prevent the ozone-induced increase in bronchial responsiveness in dogs. Despite its lack of effectiveness in preventing the increase in bronchial responsive-
YING, GROSS, TERZO, AND ESCHENBACHER
820 TABLE 3
20
u
> ....
FEV, BEFORE AND AFTER EXPOSURE TO OZONE ON DIFFERENT STUDY DAYS' Subject No.
1 2 3 4 5 6 7 8 9 10 11 12 13
. -> o
Placebo
No Medication
Indomethacin
Before
After
Before
Afte r
Before
After
4.52 5.21 3.10 4.25 3.19 4.24 4.41 4.30 4.76 4.38 4.49 4.66 5.08
3.92 4.14 2.27 3.16 2.81 2.23 4.29 4.20 3.50 2.71 3.19 4.17 5.13
4.47 5.30 3.22 4.35 3.19 3.62 4 .49 4 .20 5.29 4.88 4.59 4.70 4.75
3.54 4.97 2.19 4.36 2.93 2.44 4.73 4.15 4.81 2.93 3.16 4.38 5.08
4.37 5.49 3.20 4.48 3.30 3.78 4.44 4.18 4.50 4.67 4.53 4.66 4.47
4 .22 4 .38 3.03 3.36 3.14 2.87 4.41 4.34 4.07 3.89 4.15 4.44 5.13
'0
FE Y- l
Fye
I
U.I ....
C
'cc"n " B
-1 0
-2 0
~
c
..'" u
-30
'"
0..
-40
Fig. 5. Percent change in FEV, and FVC after ozone exposure on 3 separate study days for the seven subjects who met the inclusion criteria for the study (see text). Values showo are mean ± SEM. Closed bars = no medication; open bars = placebo; hatched bars = indomethacin .
• Before and after the 2·h exposure to 0.4 ppm ozone.
ness, indomethacin did appear to prevent the ozone-induced decrease in lung function (FEV! and FVC). It is possible that the dosage of indomethacin used by the subjects was not sufficient to inhibit cyclooxygenase enzymes in the cells of the airways for these subjects. However,(1)all the subjects that were included in the data analysis had positive serum concentrations of indomethacin on the Indomethacin Study Day; (2) these serum concentrations of indomethacin obtained were within the range believed to provide the antiinflammatory therapeutic effect (24); and (3) the ozone-induced decrease in lung function was reduced significantly on the Indomethacin Study Day. The magnitude of the inhibition of the ozone-induced lung function changes by indomethacin in our results matches quite closely that
.,c
60
'0
so
in the results of Schelegle and coworkers (16). These results suggest that cyclooxygenase products may be involved in the ozone-induced decreases in lung function. As mentioned earlier, animal and human studies (17-19) have shown that 6
5
..J
...
:>
3
W
u..
o
~
u
....," ~
40
E
....E en
30
E
o'
20
2
~ I
U
10
C-
o
After Before Ozone Exposure Fig . 3. PC, ••SRaw for methacholine before and after ozone exposure on 3 separate study days tor the seven subjects who met the inclusion criteria for the study (see text). Values shown are mean ± SEM . Closed bars = no med ication; open bars = placebo; hatched bars = indomethacin.
Before
After Ozone Exposure
Fig. 4. FEV, and FVC before and after ozone exposure on 3 separate study days for the seven subjects who metthe inclusion criteria for the study (see text). Values shown are mean ± SEM. Open squares = no medication ; closed diamonds = placebo; closed squares = indomethacin.
metabolites of arachidonic acid can cause irritation and bronchoconstriction. In addition, Leikauf and coworkers (25) have shown that ozone can enhance the production of arachidonic acid metabolites, including PGF2a from bovine tracheal epithelial cells. Thus, it is possible that products such as PGF2a , PGD 2 , or ThA2 are being released as a result of ozone exposure and that these products are producing effects that result in restriction of lung function. These results also suggest that there are separate mechanisms for the ozoneinduced increases in bronchial responsivenessand the ozone-induced decreases in lung function (FEV! and FVC). As stated above, the mechanism for the latter effect does seem to involve products of cyclooxygenase metabolism. The mechanism for ozone-induced increases in bronchial responsiveness in humans remains unknown. It is known from other studies that ozone can alter airway epithelial function in humans (26). It is possible that alterations in the airway epithelial lining could result in a decrease in a normally present epithelial-derived smooth muscle relaxant factor (27, 28) or that the increase in epithelial permeability could allow increased access of the inhaled methacholine to the airway smooth muscle. Another possibility is that in addition to ozone-induced alterations in the epithelial cells, there could be alterations in the neural elements in the airway lining that could result in an enhancement in the release of neuropeptides that could in turn contribute to the increase in airway smooth muscle responsiveness to methacholine. Finally, it is also conceivable that ozone-induced cellular production of other mediators is occurring.
INDOMETHACIN DOES NOT INHIBIT OlONE-INDUCED INCREASES IN BRONCHIAL RESPONSIVENESS
These other mediators could include agents such as platelet-activating factor (PAF) that, if present in increased concentrations, could be responsible for the increase in bronchial responsiveness. In conclusion, we have shown that although indomethacin can inhibit the ozone-induced decreasesin lung function in human subjects, it does not appear to prevent the ozone-induced increases in methacholine bronchial responsiveness in these same subjects. Ozone-induced increases in bronchial responsiveness in humans does not appear to involve cyclooxygenase metabolites. Acknowledgment The writers thank Messrs. James D'Arcy and Robert Wooley for their operation and maintenance of the human chamber exposure facility. References 1. Adams WC, Savin WM, Christo AE. Detection of ozone toxicity during continuous exercise via the effective dose concept. J Appl Physiol1981; 51:415-22. 2. Folinsbee LJ, Bedi JF, Horvath SM. Pulmonary function changes after 1 hour continuous heavy exercise in 0.21 ppm ozone. J Appl Physiol 1984; 57:984-8. 3. McDonnell WF, Horstman DH, Hazucha MJ, et al. Pulmonary effects of ozone exposure during exercise: dose-responsecharacteristics. J Appl Physiol 1983; 54:1345-52. 4. Gibbons SI, Adams WC. Combined effects of ozone exposure and ambient heat on exercising females. J Appl Physiol 1984; 57:450-6. 5. DeluciaAJ, Adams WC. Effects of 0 3 inhalation during exercise on pulmonary function and blood biochemistry. J Appl Physiol1977; 43:75-81.
6. Beckett WS, McDonnell WF, Horstman DH, House DE. Role of the parasympathetic nervous systemin acute lung responseto ozone. J Appl Physiol 1985; 59:1879-85. 7. Adams WC, SchelegleES. Ozone and high ventilation effects on pulmonary function and endurance performance. J Appl Physiol1983; 55:805-12. 8. Silverman F, FolinsbeeLJ, Barnard J, Shephard RJ. Pulmonary function changes in ozone-interaction of concentration and ventilation. J Appl Physiol 1976; 41:859-64. 9. Kreit JW, Gross KB, Moore TB, Lorenzen TJ, D'Arcy J, Eschenbacher WL. Ozone-induced changes in pulmonary function and bronchial responsiveness in asthmatics. J Appl Physiol 1989; 66:217-22. 10. Hazucha MJ. Relationship between ozone exposure and pulmonary function changes. J Appl Physiol 1987; 62:1671-80. 11. DimeoMJ, GlennMG, HoltzmanMJ, Sheller JR, Nadel JA, Boushey HA. Threshold concentration of ozone causing an increase in bronchial reactivity in humans and adaptation with repeated exposures. Am Rev Respir Dis 1981; 124:245-8. 12. Golden JA, Nadel JA, Boushey HA. Bronchial hyperirritability in healthy subjects after exposure to ozone. Am Rev Respir Dis 1978; 118: 287-94. 13. Holtzman MJ, Cunningham JH, Sheller JR, IrsiglerGB, Nadel JA, BousheyHA. Effect of ozone on bronchial reactivityin atopic and nonatopic subjects. Am Rev Respir Dis 1979; 120:1059-67. 14. Seltzer J, Bigby BG, Stulbarg M, et al. Ozone induced change in bronchial reactivityto methacholine and airway inflammation in humans. J Appl Physiol 1986; 60:1321-6. 15. Aizawa H, Chung KF, Leikauf GD, et al. Significance of thromboxane generation in ozone induced airway hyperresponsiveness in dogs. J Appl Physiol 1985; 59:1918-23. 16. Schelegle ES, Adams WC, Seifkin AD. Indomethacin pretreatment reduces ozone induced pulmonary function decrements in human subjects. Am Rev Respir Dis 1987; 136:1350-4. 17. Gardiner PJ, Copas JL, Elliot RD, Collier HOJ. Tracheobronchial irritancy of inhaled pros-
821
taglandins in the conscious cat. Prostaglandins 1978; 15:303-15. 18. Smith AP, Cuthbert MF, Dunlop LS. Effects of inhaled prostaglandins E 1,E2 , F 2 U on the airway resistance of healthy and asthmatic man. Clin Sci Mol Med 1975; 48:421-30. 19. Svensson J, Strandberg K, Tuvemo T, Hamberg M. Thromboxane A 2 : effects on airway and vascular smooth muscle. Prostaglandins 1977; 14: 425-36. 20. O'Byrne PM, Walters EH, Aizawa H, Fabbri LM, Holtzman MJ, Nadel JA. Indomethacin inhibits the airway hyperresponsiveness but not the neutrophil influx induced by ozone in dogs. Am Rev Respir Dis 1984; 130:220-4. 21. American Thoracic Society. Standardization of spirometry: 1987 update. Am Rev Respir Dis 1987; 136:1285-98. 22. Foster DW. Eating disorders: obesity and anorexia nervosa. In: Wilson JD, Foster DW,eds. Williams textbook of endocrinology.Philadelphia: WB Saunders, 1985; 1083. 23. Duncan DB. Multiple range and multiple F tests. Biometrics 1955; 11:1-42. 24. Flowers RJ, Moncada S, Vane JR. Analgesicantipyretics and anti-inflammatory agents; drugs employed in the treatment of gout. In: Gilman AG, Goodman LS, RaIl TW, Murad F, eds. The pharmacologic basis of therapeutics. New York: Macmillan, 1985; 695. 25. Leikauf GD, Driscoll KE, Wey HE. Ozoneinduced augmentation of eicosanoid metabolism in epithelial cells from bovine trachea. Am Rev Respir Dis 1988; 137:435-42. 26. Kehrl HR, Vincent LM, Kowalsky RJ, et al. Ozone exposure increases respiratory epithelial permeability in humans. Am RevRespir Dis 1987;135: 1124-8. 27. Stuart-Smith K, Vanhoutte PM. Airway epithelium modulates the responsiveness of porcine bronchial smooth muscle. J Appl Physiol1988; 65: 721-7. 28. Jones GL, Lane CG, Daniel EE, O'Byrne PM. Release of epithelium derived relaxing factor after ozone inhalation in dogs. J Appl Physiol1988; 65: 1238-43.
