Effect of Inhaled Furosemide on the Bronchial Response to Lysine-Aspirin Inhalation in Asthmatic Subjects* Francisco S. Vargas, M.D.; Miguel Croce, M.D.; Lisete R. Teixeira, M.D.; Mario Terra-Filho, M.D.; Alberto Cukier, M.D.; and Richard W Light, M.D., F.C.C.R It has been demonstrated recently that inhaled furosemide inhibits bronchoconstriction induced by cold air, physical exercise, various antigens, and metabisul6te. The goal of the present study was to determine if the inhalation of furosemide would inhibit the bronchoconstriction resulting from the inhalation of lysine-aspirin in aspirin-sensitive asthmatics. Six female subjects with known hypersensitivity to aspirin participated in this crossover study comparing 20 mg of inhaled furosemide and placebo. The volunteers inhaled increasing concentrations of lysine-aspirin after the inhalation of furosemide or placebo. The geometric mean provocative dose causing a 20 percent decrease in the FEV I (PD20) after the inhalation of placebo was 30.4 mglml and
the PD20 was equal or below 90 mg/ml in all patients. In contrast, the FEV. did not decrease by 20 percent in any of the patients pretreated with furosemide when the inhaled concentration was increased to 360 mg/ml. From this study, we conclude that the administration of furosemide blocks the bronchospasm induced by the inhalation of Iysineaspirin in aspirin-sensitive asthmatics. (Chest 1992; 102:408-11)
It has been reported recently that the inhalation of furosemide inhibits the bronchoconstriction provoked by inhaled allergens,I-3 exercise, 4 cold air challenge,S metabisulfite inhalation,6 or adenosine 5'monophosphate inhalation. 7 In contrast, the inhalation of furosemide does not inhibit the bronchoconstriction produced by the inhalation of methacholine l ,s,6 or that produced by the inhalation ofhistamine. H,9 Either the inhalation 10 or the oral ingestionII of aspirin or other nonsteroidal anti-inflammatory drugs (NSAIDs) induces bronchospasm in a significant number of individuals with asthma. The mechanism by which NSAIDs induce bronchoconstriction is not definitely known, but it is probably related to their inhibition of the enzyme cyclooxygenase. This mechanism is probably somewhat different from those involved in other situations where inhaled furosemide has been shown to inhibit induced bronchoconstriction. We therefore conducted a placebo-controlled crossover study to assess the effects of furosemide on the bronchoconstrictor response to inhaled lysineaspirin in a group of asthmatic subjects who had a history of asthma after the ingestion of aspirin.
subjects \\;th a history of asthma induced by aspirin ingestion. All attended the outpatient pulmonary clinic at the Instituto do Cora~o of the University of Sao Paulo. All the volunteers were free from upper respiratory tract infections for the four weeks preceding the study. All the patients were regular users of inhaled sympathomimetic medications, but no other medications, for their asthma. The mean a~e of the participants was 28 years. The forced expiratory volume in 1 s (FEV.) at baseline expressed as percent predicted was 85 ± 10.0 percent.
MATERIALS AND METHODS
lbtients The six study participants (Table 1) were nonsmokin~ female *From the Instituto do Cora~o, University of Sao Paulo, Brazil, and the Department of Medicine, Veterans Administration Medical Center, Long Beach, Calif, and the University of California, Irvine. Manuscript received AlI~st 12; revision accepted Novemher 12. Reprint requests: Dr. Light, VA Medical Center; Umg Beach, California 90822
408
LTE. = leukotriene E.; NSAID = nonsteroidal anti-inflammatory drug; PD20 = mean provocative dose causing a 20 percent decrease in the FEV.
Protocol After the participants gave informed consent approved by the institutional review board, they reported to the pulmonary evaluation laboratory on two different mornings. The interval between the two tests was five days for one patient (case 1) and more than six days for all other patients. They refrained from usin~ their inhaled sympathomimetic medications for the 12 h prior to the study. The patients rested for 15 min after they arrived at the pulmonary evaluation laboratory. Spirometry was obtained three times at intervals of 3 min. The highest FEV t obtained at this time was used for calculation of the mean provocative dose causing a 20 percent decrease in the FEV. (PD20). On the first study day, the patients inhaled a saline solution containing placebo. On the second study day, the patients inhaled a saline solution containing 20 mg of nlrosemide. The solutions (5 ml) were nebulized with a nebulizer
Table 1- Clinical Features of Six Study Patients Subject 1 2 3 4 5 6 Mean±SD
Age, yr
lleight, cm
Weight, kg
FEV., %*
12
156 154 158 150 161 157 156.0±3.7
57.5 63.0 70.5 61.3 110.0 42.0 67.4±22.9
89.3 73.4 100.0 85.8 72.8 88.2 84.9± 10.3
40 36
30 41 12 28.5± 13.3
*FEV. = percent predicted normal prior to the study. Inhaled Furosemide Effect on Bronchial Response in Asthma (Vargas at al)
Table 2-FEV. (Liters) after the Administration of Furosemide and Placebo on the Two Different Study Days Placebo Day
Table 3-PD20 for Lysine-Aspirin after the Inhalation of Saline Solution and 'Furosemide· Lysine-Aspirin PD20, mwn11
Furosemide Day
Subject
Before
After
Before
After
Subject
Placebo
Furosemide
1 2 3 4 5 6 Mean±SD
3.00 1.39 2.71 2.31 1.57 2.32 2.22±0.62
2.66 1.27 2.71 2.25 1.57 2.02 2.08±0.58
2.77 1.73 2.54 2.25 1.76 2.31 2.23±0.41
2.25 1.39 2.54 1.85 1.92 2.20 2.02±0.39
1 2 3 4 5 6
11.25 14.72 28.49 22.50 74.03 90.00
>360 >360 >360 >360 >360 >360
(Wright) over a IO-min period. At the end of the nebulization period, spirometry was obtained three more times. After the patient' received the nebulized furosemide or placebo, she was challenged with inhaled lysine-aspirin following a procedure similar to that described by Phillips et al. III The subjects inhaled the aerosolized solution via a mouthpiece for five breaths from functional residual capacity to full inspiratory capacity with a nosecIip in place. The concentration of the lysine-aspirin solution initially was 11.25 mglml. Three spirometric tracings were obtained 5, 15, and 30 min after each inhalation. The best FEV. was used for analysis. If the FEV. dropped 20 percent below the baseline value at any of these three times, the test was terminated. Otherwise the concentration of the lysine-aspirin was doubled (until the concentration reached 360 mg/ml) and the procedure was repeated. The spirometry was performed at all times with a specific spirometer (Hewlett Packard spirometer model HP047402-A). Data Analysis Data are expressed as the mean ± standard deviation unless otherwise noted. The highest FEV. at each measurement time was used for data analysis. Comparisons between spirometric values at different times were made using the Student's paired t test. The percentage decrease in FEV. from postnebulization baseline was plotted against the cumulative concentration of lysine-aspirin on a logarithmic scale, and the PD20 value was derived by linear interpolation. RESULTS
The baseline FEV I on the placebo day and the furosemide day were nearly identical (Table 2). The mean FEV I decreased to a similar extent after the inhalation of both placebo and furosemide (Table 2). The decrease after placebo or furosemide inhalation did not reach statistical significance (p380 /
\
\
\
I
I
PD20> 380
PD20.22.50 ',1.'251
I
45
22.50
I
80
PD20·74.03
I
180
I
380
i
11.'251
i 45
22.50
occasion, produces catastrophic bronchoconstriction that is difficult to reverse. 10 An alternative approach is to challenge the airways by inhalation, but native aspirin is too irritating and of low solubility in aqueous solution. Inhalation ofa nebulized solution ofthe more soluble and nonirritant lysine-acetylsalicylate (lysineaspirin) is an effective method for investigating aspirininduced asthma and has the advantage of producing reactions that are confined to the respiratory tract and are easier to control. 10,19 Our patients were definitely sensitive to aspirin. In the report by Phillips and coworkers,IO the geometric mean for the PD20 of their 11 patients for lysineaspirin was 64.8 mg/ml. The geometric mean for the PD20 of our six patients was 30.4 mg/ml, which was definitely lower. The bronchoconstricting effects of aspirin were inhibited by inhaled furosemide. As could be seen, the decrease in the FEV1 was less than 20 percent after furosemide in all patients and for three of the six patients, the FEV1 was higher after the highest dose of aspirin than it was at baseline. One possible explanation for our results is that the initial challenge with aspirin left our patients insensitive to a second aspirin challenge. It has been reported that there is a refractory period after ingestion or inhalation of aspirin. However, this refractory period persists for only 72 h. 20,21 Since the two tests in all of our patients were separated by at least five days, this 410
,J
/
\
-.;',----
15 25
I
I
/' ,\ I
10
Petle"t"'8
1 80
1~0
I
380
FIGURE 1. Percentage of change in the FEV. as the subjects were challenged with increasing concentrations of lysine-aspirin. Patients were pretreated either with furosemide (dotted lines) or saline solution (solid lines). The first point for each patient on the graph is the change in the FEV. after the inhalation of furosemide or saline solution. The concentration of the inhaled Iysineaspirin in milligrams per milliliter is shown on the X-axis.
does not appear to be a viable explanation. The fact that furosemide protected against the bronchoconstriction induced by aspirin may provide some clues as to the mechanism of action of furosemide. There have been several different hypotheses proposed concerning the mechanisms of action of furosemide. When Bianco et al4 first reported that inhaled furosemide prevented exercise-induced asthma, they believed that this action was possibly due to the fact that furosemide prevented the hyperosmolarity of the epithelium, which is thought by some to be the mechanism responsible for exerciseinduced asthma. However, when furosemide was subsequently demonstrated to inhibit bronchoconstriction following antigen inhalation,1,3 cold air inhalation, 5 and metabisulfite inhalation, 6 an editorial22 observed that the spectrum of action of furosemide was very similar to that of cromolyn, which might suggest that furosemide acts directly on the luminal mast cells. Subsequent reports demonstrated that furosemide inhibited the bronchoconstriction following adenosine 5' monophosphate inhalation7 and cough induced by low-chloride inhalation 23 and prostaglandin F 2a • The present report demonstrates that furosemide inhibits the bronchoconstriction induced by aspirin inhalation. The multitude of actions suggested that the mechanism of action of furosemide was beyond that which could be attributed to effects on the luminal mast cells. Inhaled Furosemide Effect on Bronchial Response in Asthma (Vargas et al)
Another possibility is that furosemide might act by releasing bronchodilator prostaglandins from airway epithelium. When vascular endothelium is exposed to furosemide, vasodilating prostaglandins are released. 24 This mechanism appears unlikely with aspirin-induced asthma since prostaglandin synthesis is blocked by the ingestion of aspirin. In addition, the effects of furosemide persist with the coadministration of indomethacin, a potent cyclooxygenase inhibitor. 25 Recently, it has been suggested that furosemide works by inhibiting both cholinergic and excitatory nonadrenergic, noncholinergic neurotransmission and that this effect may be related to its inhibitory effects on the sodium-potassium-chloride cotransporter. 25 There is no evidence that furosemide has a direct effect on airway smooth muscle contraction.2.'5 The results of the present study are compatible \\ith this hypothesis. In conclusion, the present study demonstrates that bronchoconstriction that follows the inhalation of lysine-aspirin can be blocked by the inhalation of 20 mg of furosemide. This observation is against the hypothesis that the mechanism of action of furosemide is the release of bronchodilator prostaglandins from the airway epithelium. The observation is consistent with the hypothesis that furosemide works by inhibiting both cholinergic and excitatory nonadrenergic, noncholinergic neurotransmission. REFERENCES 1 Bianco S, Peironi MG, Refini RM, Tottoli L, Sestini ~ Protective effect of inhaled furosemide on a1ler~en-induced early and late asthmatic reactions. N En~l 1 Med 1989; 321:1069-73 2 Verdiani e Di Carlo S, Baronti A, Bian
the PD20 was equal or below 90 mg/ml in all patients. In contrast, the FEV. did not decrease by 20 percent in any of the patients pretreated with furosemide when the inhaled concentration was increased to 360 mg/ml. From this study, we conclude that the administration of furosemide blocks the bronchospasm induced by the inhalation of Iysineaspirin in aspirin-sensitive asthmatics. (Chest 1992; 102:408-11)
It has been reported recently that the inhalation of furosemide inhibits the bronchoconstriction provoked by inhaled allergens,I-3 exercise, 4 cold air challenge,S metabisulfite inhalation,6 or adenosine 5'monophosphate inhalation. 7 In contrast, the inhalation of furosemide does not inhibit the bronchoconstriction produced by the inhalation of methacholine l ,s,6 or that produced by the inhalation ofhistamine. H,9 Either the inhalation 10 or the oral ingestionII of aspirin or other nonsteroidal anti-inflammatory drugs (NSAIDs) induces bronchospasm in a significant number of individuals with asthma. The mechanism by which NSAIDs induce bronchoconstriction is not definitely known, but it is probably related to their inhibition of the enzyme cyclooxygenase. This mechanism is probably somewhat different from those involved in other situations where inhaled furosemide has been shown to inhibit induced bronchoconstriction. We therefore conducted a placebo-controlled crossover study to assess the effects of furosemide on the bronchoconstrictor response to inhaled lysineaspirin in a group of asthmatic subjects who had a history of asthma after the ingestion of aspirin.
subjects \\;th a history of asthma induced by aspirin ingestion. All attended the outpatient pulmonary clinic at the Instituto do Cora~o of the University of Sao Paulo. All the volunteers were free from upper respiratory tract infections for the four weeks preceding the study. All the patients were regular users of inhaled sympathomimetic medications, but no other medications, for their asthma. The mean a~e of the participants was 28 years. The forced expiratory volume in 1 s (FEV.) at baseline expressed as percent predicted was 85 ± 10.0 percent.
MATERIALS AND METHODS
lbtients The six study participants (Table 1) were nonsmokin~ female *From the Instituto do Cora~o, University of Sao Paulo, Brazil, and the Department of Medicine, Veterans Administration Medical Center, Long Beach, Calif, and the University of California, Irvine. Manuscript received AlI~st 12; revision accepted Novemher 12. Reprint requests: Dr. Light, VA Medical Center; Umg Beach, California 90822
408
LTE. = leukotriene E.; NSAID = nonsteroidal anti-inflammatory drug; PD20 = mean provocative dose causing a 20 percent decrease in the FEV.
Protocol After the participants gave informed consent approved by the institutional review board, they reported to the pulmonary evaluation laboratory on two different mornings. The interval between the two tests was five days for one patient (case 1) and more than six days for all other patients. They refrained from usin~ their inhaled sympathomimetic medications for the 12 h prior to the study. The patients rested for 15 min after they arrived at the pulmonary evaluation laboratory. Spirometry was obtained three times at intervals of 3 min. The highest FEV t obtained at this time was used for calculation of the mean provocative dose causing a 20 percent decrease in the FEV. (PD20). On the first study day, the patients inhaled a saline solution containing placebo. On the second study day, the patients inhaled a saline solution containing 20 mg of nlrosemide. The solutions (5 ml) were nebulized with a nebulizer
Table 1- Clinical Features of Six Study Patients Subject 1 2 3 4 5 6 Mean±SD
Age, yr
lleight, cm
Weight, kg
FEV., %*
12
156 154 158 150 161 157 156.0±3.7
57.5 63.0 70.5 61.3 110.0 42.0 67.4±22.9
89.3 73.4 100.0 85.8 72.8 88.2 84.9± 10.3
40 36
30 41 12 28.5± 13.3
*FEV. = percent predicted normal prior to the study. Inhaled Furosemide Effect on Bronchial Response in Asthma (Vargas at al)
Table 2-FEV. (Liters) after the Administration of Furosemide and Placebo on the Two Different Study Days Placebo Day
Table 3-PD20 for Lysine-Aspirin after the Inhalation of Saline Solution and 'Furosemide· Lysine-Aspirin PD20, mwn11
Furosemide Day
Subject
Before
After
Before
After
Subject
Placebo
Furosemide
1 2 3 4 5 6 Mean±SD
3.00 1.39 2.71 2.31 1.57 2.32 2.22±0.62
2.66 1.27 2.71 2.25 1.57 2.02 2.08±0.58
2.77 1.73 2.54 2.25 1.76 2.31 2.23±0.41
2.25 1.39 2.54 1.85 1.92 2.20 2.02±0.39
1 2 3 4 5 6
11.25 14.72 28.49 22.50 74.03 90.00
>360 >360 >360 >360 >360 >360
(Wright) over a IO-min period. At the end of the nebulization period, spirometry was obtained three more times. After the patient' received the nebulized furosemide or placebo, she was challenged with inhaled lysine-aspirin following a procedure similar to that described by Phillips et al. III The subjects inhaled the aerosolized solution via a mouthpiece for five breaths from functional residual capacity to full inspiratory capacity with a nosecIip in place. The concentration of the lysine-aspirin solution initially was 11.25 mglml. Three spirometric tracings were obtained 5, 15, and 30 min after each inhalation. The best FEV. was used for analysis. If the FEV. dropped 20 percent below the baseline value at any of these three times, the test was terminated. Otherwise the concentration of the lysine-aspirin was doubled (until the concentration reached 360 mg/ml) and the procedure was repeated. The spirometry was performed at all times with a specific spirometer (Hewlett Packard spirometer model HP047402-A). Data Analysis Data are expressed as the mean ± standard deviation unless otherwise noted. The highest FEV. at each measurement time was used for data analysis. Comparisons between spirometric values at different times were made using the Student's paired t test. The percentage decrease in FEV. from postnebulization baseline was plotted against the cumulative concentration of lysine-aspirin on a logarithmic scale, and the PD20 value was derived by linear interpolation. RESULTS
The baseline FEV I on the placebo day and the furosemide day were nearly identical (Table 2). The mean FEV I decreased to a similar extent after the inhalation of both placebo and furosemide (Table 2). The decrease after placebo or furosemide inhalation did not reach statistical significance (p380 /
\
\
\
I
I
PD20> 380
PD20.22.50 ',1.'251
I
45
22.50
I
80
PD20·74.03
I
180
I
380
i
11.'251
i 45
22.50
occasion, produces catastrophic bronchoconstriction that is difficult to reverse. 10 An alternative approach is to challenge the airways by inhalation, but native aspirin is too irritating and of low solubility in aqueous solution. Inhalation ofa nebulized solution ofthe more soluble and nonirritant lysine-acetylsalicylate (lysineaspirin) is an effective method for investigating aspirininduced asthma and has the advantage of producing reactions that are confined to the respiratory tract and are easier to control. 10,19 Our patients were definitely sensitive to aspirin. In the report by Phillips and coworkers,IO the geometric mean for the PD20 of their 11 patients for lysineaspirin was 64.8 mg/ml. The geometric mean for the PD20 of our six patients was 30.4 mg/ml, which was definitely lower. The bronchoconstricting effects of aspirin were inhibited by inhaled furosemide. As could be seen, the decrease in the FEV1 was less than 20 percent after furosemide in all patients and for three of the six patients, the FEV1 was higher after the highest dose of aspirin than it was at baseline. One possible explanation for our results is that the initial challenge with aspirin left our patients insensitive to a second aspirin challenge. It has been reported that there is a refractory period after ingestion or inhalation of aspirin. However, this refractory period persists for only 72 h. 20,21 Since the two tests in all of our patients were separated by at least five days, this 410
,J
/
\
-.;',----
15 25
I
I
/' ,\ I
10
Petle"t"'8
1 80
1~0
I
380
FIGURE 1. Percentage of change in the FEV. as the subjects were challenged with increasing concentrations of lysine-aspirin. Patients were pretreated either with furosemide (dotted lines) or saline solution (solid lines). The first point for each patient on the graph is the change in the FEV. after the inhalation of furosemide or saline solution. The concentration of the inhaled Iysineaspirin in milligrams per milliliter is shown on the X-axis.
does not appear to be a viable explanation. The fact that furosemide protected against the bronchoconstriction induced by aspirin may provide some clues as to the mechanism of action of furosemide. There have been several different hypotheses proposed concerning the mechanisms of action of furosemide. When Bianco et al4 first reported that inhaled furosemide prevented exercise-induced asthma, they believed that this action was possibly due to the fact that furosemide prevented the hyperosmolarity of the epithelium, which is thought by some to be the mechanism responsible for exerciseinduced asthma. However, when furosemide was subsequently demonstrated to inhibit bronchoconstriction following antigen inhalation,1,3 cold air inhalation, 5 and metabisulfite inhalation, 6 an editorial22 observed that the spectrum of action of furosemide was very similar to that of cromolyn, which might suggest that furosemide acts directly on the luminal mast cells. Subsequent reports demonstrated that furosemide inhibited the bronchoconstriction following adenosine 5' monophosphate inhalation7 and cough induced by low-chloride inhalation 23 and prostaglandin F 2a • The present report demonstrates that furosemide inhibits the bronchoconstriction induced by aspirin inhalation. The multitude of actions suggested that the mechanism of action of furosemide was beyond that which could be attributed to effects on the luminal mast cells. Inhaled Furosemide Effect on Bronchial Response in Asthma (Vargas et al)
Another possibility is that furosemide might act by releasing bronchodilator prostaglandins from airway epithelium. When vascular endothelium is exposed to furosemide, vasodilating prostaglandins are released. 24 This mechanism appears unlikely with aspirin-induced asthma since prostaglandin synthesis is blocked by the ingestion of aspirin. In addition, the effects of furosemide persist with the coadministration of indomethacin, a potent cyclooxygenase inhibitor. 25 Recently, it has been suggested that furosemide works by inhibiting both cholinergic and excitatory nonadrenergic, noncholinergic neurotransmission and that this effect may be related to its inhibitory effects on the sodium-potassium-chloride cotransporter. 25 There is no evidence that furosemide has a direct effect on airway smooth muscle contraction.2.'5 The results of the present study are compatible \\ith this hypothesis. In conclusion, the present study demonstrates that bronchoconstriction that follows the inhalation of lysine-aspirin can be blocked by the inhalation of 20 mg of furosemide. This observation is against the hypothesis that the mechanism of action of furosemide is the release of bronchodilator prostaglandins from the airway epithelium. The observation is consistent with the hypothesis that furosemide works by inhibiting both cholinergic and excitatory nonadrenergic, noncholinergic neurotransmission. REFERENCES 1 Bianco S, Peironi MG, Refini RM, Tottoli L, Sestini ~ Protective effect of inhaled furosemide on a1ler~en-induced early and late asthmatic reactions. N En~l 1 Med 1989; 321:1069-73 2 Verdiani e Di Carlo S, Baronti A, Bian
