INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Asthma is an inflammatory airway disease, and histologic examination of the airway wall has revealed increased vascularity and vascular engorgement in severe asthma (1, 2) and less markedly increased vascularity and vascular engorgement in mild to moderate asthma (3). These changes suggest the presence of altered airway vascular hemodynamics. We have previously confirmed this by showing that airway mucosal blood flow (aw) is increased in patients with stable mild to moderate asthma, that glucocorticosteroid (GS)-naive patients tend to have higher aw values than patients who regularly use inhaled GS, and that -adrenergic-agonist-induced increases in aw as an index of vasodilation are blunted in patients with asthma (4). These observations were consistent with the hypothesis that asthma is associated with inflammation-related vascular hyperperfusion and blunted -adrenergic vasodilation. However, the evidence that the altered bronchial hemodynamics that we observed were a reflection of GS-sensitive airway inflammation was circumstantial. We therefore conducted the present open, prospective study (without placebo control) to assess the effect of an inhaled GS administered for 14 d on aw and its responsiveness to inhaled albuterol in patients with stable asthma. The results were compared with aw and aw responsiveness in normal control subjects. Since only GS-naive asthmatic individuals were eligible, the asthmatic study population consisted predominantly of patients with mild disease.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Test Population

Thirty one nonsmokers participated in the study (Table 1). They denied having cardiovascular disease, and none of them was taking vasoactive or antiinflammatory medications. Subjects who had taken antibiotics or inhaled or systemic GS, and subjects who had had an acute respiratory infection during the 6-wk period preceding the study were excluded. Twelve subjects were healthy and 19 had mild intermittent asthma and used a short-acting inhaled -adrenergic agonist on demand as their only asthma treatment. Informed consent was obtained from all subjects, and all received financial remuneration for their participation.

Spirometry was done with a Model 6200 Autobox DL (Essential Medic, Yorba Linda, CA). The highest FEV₁ of three FVC maneuvers was determined and expressed as both an absolute value and as a percent of the predicted value (5).

Airway Mucosal Blood Flow

A soluble inert gas uptake method was used to measure aw (4, 6, 7). The subjects were seated in front of a valve system that allowed them to inhale, through a mouthpiece (with the nasal passages occluded by a nose clip), room air or a gas mixture from a Teflon bag containing 10% dimethylether (DME), 5% helium, and a balance of oxygen, and to then exhale into a rolling-seal spirometer (Model 842; Ohio Instruments, Houston, TX). The subjects first inhaled room air to TLC and then exhaled 500 ml from TLC, and subsequently rapidly inhaled the same volume of gas mixture from the Teflon bag. They held their breath for a predetermined period and then exhaled into the spirometer through a critical-flow orifice to standardize expiratory flow. The maneuver was performed with two breathhold times each of 5, 10, 15, and 20 s, in random order. During exhalation, the instantaneous concentrations of DME, nitrogen, and helium were measured at the airway opening with a mass spectrometer (Perkin-Elmer, Pomona, CA), along with the expired gas volume. The mass spectrometer inlet was not heated, and no corrections were made for water pressure. This resulted in a negligible overestimation of DME concentration (approximately 0.3%). The mass spectrometer was also used to verify the gas concentrations in the Teflon bag before inhalation of the gas mixture. Anatomic dead space was determined from the expired nitrogen concentration curve as described by Fowler and coworkers (8). The helium-corrected decrease in the DME concentration over time was obtained through the least-squares fit procedure, using the two measurements for each of the four breathhold times. This was done in the expired volume fraction corresponding to the anatomic dead space minus the most proximal 50 ml, which was designated DS. From the helium-corrected DME slope multiplied by DS (DME), the mean DME concentration in DS (F), and the solubility coefficient for DME in blood and tissue (), aw was calculated by using Fick's principle (aw = DME/ · F). aw was normalized for DS and expressed as µl/min/ml.

Protocol

The subjects were asked to come to the research laboratory in the morning of the study day without having had any coffee or caffeinated beverages. The subjects were asked to abstain from ingesting alcoholic beverages on the night before the study day. The asthmatic subjects were asked not to use their inhaled -adrenergic agonist for at least 12 h before the study. On arrival, the subjects underwent spirometry and proceeded with the measurement of aw. The subjects then inhaled 2 puffs of albuterol (180 µg) from a metered dose inhaler, using a spacer, and 15 min later spirometry and the measurement of aw were repeated in the same order. All subjects participated in this part of the protocol (Day 1).

The asthmatic subjects were then begun on inhaled fluticasone propionate (FP; 220 µg/puff), at 1 puff twice daily (440 µg/d) for a period of 2 wk. Between 12 and 18 h after the last dose of FP, the asthmatic subjects returned to the laboratory for a second day, and underwent the same measurements that were made on Day 1.

Eleven of the 19 asthmatic subjects were willing to discontinue FP at this time and/or were available for further study. They returned to the laboratory 2 wk later (Day 3) for a final set of measurements identical to those made on Days 1 and 2. Two patients were excluded because they continued to use inhaled GS, five were excluded because they failed to return, and one was excluded because of an acute respiratory infection.

Data Analysis

For the measurement of aw, the mass spectrometry and spirometry signals were fed through analog-to-digital converters to a computer for processing and storage of raw data. The aw values were calculated after completion of the study.

Statistical comparisons between and within groups were made with the appropriate unpaired and paired variates of Student's t test or by analysis of variance. A value of p < 0.05 was considered significant. The data are presented as mean ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Baseline

Baseline FEV₁ was lower in asthmatic than in healthy subjects (2.78 ± 0.2 L versus 3.40 ± 0.2 L [mean ± SE]; p < 0.05), whereas baseline aw was higher in asthmatic (55.5 ± 0.7 µl/ min/ml) than in healthy subjects (44.2 ± 0.7 µl/min/ml; p = 0.01). Albuterol increased FEV₁ both in asthmatic (by 7 ± 1%; p < 0.01) and in healthy subjects (by 6 ± 1%; p < 0.05) (Figure 1). In contrast, albuterol increased aw only in healthy subjects (by 27 ± 3%; p < 0.01), having no effect on aw in asthmatic subjects (2 ± 1%; p = NS) (Figure 2).

View larger version (14K): [in this window] [in a new window]   Figure 1.   FEV1 before (B) and after (A) albuterol in healthy and asthmatic subjects before (BSL) and after 2 wk of treatment with inhaled FP (440 µg/d). *p < 0.01 versus prealbuterol value, +p < 0.01 versus corresponding value in healthy subjects. Values are given as mean ± SE.

View larger version (16K): [in this window] [in a new window]   Figure 2.   aw before (B) and after (A) albuterol in healthy and asthmatic subjects before (BSL) and after 2 wk of treatment with inhaled FP (440 µg/d). *p < 0.01 versus prealbuterol value, +p < 0.05 versus corresponding value in healthy and asthmatic subjects after FP. Values are given as mean ± SE.

Short Study

After 2 wk of treatment with FP, the asthmatic subjects' aw decreased by 11.3% from baseline, to 49.2 ± 0.8 µl/min/ml (p < 0.01) (Figure 2). In addition, albuterol increased the asthmatic subjects' aw by 21 ± 2% (p < 0.01); this response was comparable to the response observed in the healthy subjects (+27 ± 3%; p < 0.01). After FP, FEV₁ (2.79 ± 0.22 L) and the FEV₁ response to albuterol (9 ± 1%) were unchanged from baseline (p = NS) in the asthmatic subjects (Figure 1).

Long Study

In the subset of asthmatic subjects that was available for restudy after discontinuation of FP, aw was 55.7 ± 0.9 before treatment with FP, 49.5 ± 1.0 after completion of treatment, and 55.2 ± 0.9 µl/min/ml 2 wk after the discontinuation of FP treatment (p < 0.01 for baseline versus completed FP treatment) (Figure 3). The aw response to albuterol was 1 ± 1%, 22 ± 25%, and 2.7 ± 1% on the three evaluation days, respectively (p < 0.01 for Day 1 versus Day 2) (Figure 4). The pre- and postalbuterol FEV₁ was similar on the three days (Figures 3 and 4).

View larger version (11K): [in this window] [in a new window]   Figure 3.   aw and FEV1 in 11 asthmatic subjects before (pre), after 2 wk of treatment with inhaled FP (440 µg/d; fluticasone) and 2 wk after cessation of FP treatment (post). *p < 0.05 versus values before and after FP. Values are given as mean ± SE.

View larger version (12K): [in this window] [in a new window]   Figure 4.   Effect of inhaled albuterol (180 µg) on aw and FEV1 before (pre), after 2 wk of treatment with FP (440 µg/d; fluticasone) and 2 wk after cessation of FP treatment (post) in 11 asthmatic subjects. *p < 0.05 versus values before and after FP. Values are given as mean ± SE.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Airway Mucosal Blood Flow

On the basis of theoretical considerations, responsiveness to vasoactive drugs, and microsphere validation in sheep, we considered the soluble gas uptake technique used in the present study to be an acceptable measurement of aw (6, 7).

We believe that the GS-induced decrease in aw in the asthmatic subjects in our study was an expression of the antiinflammatory action of GS and therefore suggest that their increased aw was due to airway inflammation. Unfortunately, we were unable to assess the effect of the GS in the control subjects for ethical reasons. Therefore, the possibility that GS decreased aw by an action unrelated to its antiinflammatory properties was not excluded. The reduction in aw after a 2-wk course of 440 µg FP daily was modest (mean: 11.3%) but consistent (present in all 19 patients). From a clinical perspective, this dose of FP can be considered intermediate. The post-GS mean aw in the asthmatic subjects did not decline to and was still significantly above the mean aw of normal controls. It is possible that a longer duration of GS treatment would have resulted in a further decrease of aw in the asthmatic subjects. In the patients who agreed to discontinue GS after the 2-wk treatment period and who were available for restudy (11 of 19 patients), mean aw returned to its pretreatment level 2 wk later. Again, this change was consistent (11 of 11 patients).

The airway can undergo remodeling in patients with severe asthma, and this includes development of an increased number of bronchial vessels (1, 2). New vessel formation has also been reported in patients with mild to moderate asthma (3). It would be difficult to design an in vivo study to determine whether the increased aw in asthmatic individuals is caused by inflammatory vasodilation, angiogenesis, or both. We limited the GS treatment in our study to 2 wk, with the expectation that the resulting decrease in aw would favor initial inflammatory vasodilation over new vessel formation. To our knowledge, it is not known whether inhaled GS reverses airway remodeling, or what the time course of the reversal process might be. However, more likely than not, the reversal process would require more than 2 wk. Therefore, the rapid decrease in aw during GS treatment, and its rapid increase following such treatment, is more consistent with inflammatory vasodilation than with inflammatory angiogenesis.

Topical GS has an acute vasoconstrictor effect, as demonstrated by the McKenzie and Stoughton skin blanching test (9). We confirmed this phenomenon by showing that inhaled FP caused a transient decrease in aw in both normal and asthmatic subjects, with a nadir at 30 min and recovery at 90 min after administration (10). The mechanism underlying the acute GS-induced vasoconstriction remains to be elucidated. In order to avoid this nongenomic action of the inhaled GS used in our study on aw, we measured aw at least 12 h after the last dose of the drug.

-adrenergic Responsiveness of Airway Mucosal Blood Flow

As in our previous study (4), we observed a blunted vasodilator response to inhaled albuterol (180 µg) in our asthmatic subjects prior to GS treatment. In normal controls, 180 µg albuterol increased mean aw by 11.7 µl/min/ml, or 27%. GS treatment brought -adrenergic responsiveness in the asthmatic subjects back toward normal levels (mean aw = 10.1 µl/min/ml, or 21%), and this effect of GS was lost 2 wk after cessation of treatment. The asthma-associated blunting of -adrenergic vasodilation in the airway mucosa may be related to the downregulation of -adrenoceptor function, possibly as a consequence of airway inflammation. The demonstrated restoration of -adrenergic responsiveness by GS treatment is in keeping with this notion. Inflammatory downregulation of -adrenergic-agonist-mediated responses and its reversal by GS have been documented in the past for nonvascular physiologic endpoints (11). Our observation suggests that this process can involve the vasculature as well.

There are other, less likely explanations for the blunted -adrenergic vasodilator responsiveness in asthma. For example, agonist-induced downregulation of -adrenergic responses has been shown to occur in the airway with respect to airway smooth muscle (11). This tolerance to -adrenergic agonists could include the airway vascular smooth muscle. However, our subjects did not use -adrenergic agonists regularly before or during our study, and the maximum frequency of use of albuterol did not exceed 4 puffs/wk. This would argue against agonist-induced tolerance. Maximal inflammatory vasodilation before inhalation of albuterol could be another possibility for the blunted -adrenergic vasodilator responsiveness. We found that the mean postalbuterol aw value was similar in our healthy and asthmatic subjects, an observation that is consistent with maximal baseline vasodilation in asthmatic subjects. The blunted vasodilator response to albuterol is in variance with new vessel formation as the cause of the increased baseline aw because if this were the case, one would expect that the vascular bed could still undergo drug-inhaled dilatation.

Airway Function

Mean baseline FEV₁ was slightly but significantly lower in the asthmatic subjects than in the normal controls, and GS treatment had no effect on baseline FEV₁ in the former group. This is in keeping with the study by Djukanovic and coworkers (12), who found a marginal improvement in FEV₁ after 2-wk of treatment with inhaled GS in subjects with mild asthma. The full effect of inhaled GS on airway caliber is usually seen after 4 wk of treatment (13). We were surprised to find that treatment with GS did not increase the bronchodilator response to 180 µg albuterol while increasing albuterol-induced vasodilation in our study. The albuterol-induced increase in mean FEV₁ was 5.7% before, 7.5% immediately after, and 6.5% at 2 wk after discontinuing GS treatment. The restoration of -adrenergic responses by GS may require more time than this, or a higher dose for airway smooth muscle than for airway vascular smooth muscle.

Airway Mucosal Blood Flow as an Index of Airway Inflammation

The assessment of airway inflammation in patients with asthma has prognostic and therapeutic relevance. Several invasive and noninvasive methods have been used to determine the presence and severity of airway inflammation. These include the measurement of cellular and acellular components of induced sputum or bronchoalveolar lavage fluid, histologic or cellular examination of bronchial forceps and brush-biopsy specimens, assay of serum markers of inflammation, determination of exhaled NO concentration, recording of peak-flow variability, and bronchial provocation testing. Although these tests may be sensitive to different aspects of airway inflammation, correlations have been reported for some of them before and after GS therapy (13). Some of the tests address inflammation directly (e.g., bronchial biopsy and induced sputum), whereas others are indirectly related to inflammation (e.g., airway hyperresponsiveness and peak flow variability). On the basis of the results of the present investigation, and the role of vascular hyperperfusion as an integral part of tissue inflammation, we suggest that aw is another quantitative index of airway inflammation.