Abstract
BACKGROUND: The spirometric response to fast-acting bronchodilator is used clinically to diagnose asthma and in clinical research to verify its presence. However, bronchodilator responsiveness does not correlate with airway hyper-responsiveness measured with the direct-acting stimulus of methacholine, demonstrating that bronchodilator responsiveness is a problematic method for diagnosing asthma. The relationship between bronchodilator responsiveness and airway hyper-responsiveness assessed with indirect-acting stimuli is not known.
METHODS: Retrospectively, the spirometric responses to inhaled bronchodilator and a eucapnic voluntary hyperpnea challenge (EVH) were compared in 39 non-smoking adult subjects with asthma (26 male, 13 female; mean ± SD age 26.9 ± 7.8 y; mean ± SD body mass index 26.3 ± 4.7 kg/m2). All subjects met one or both of 2 criteria: ≥ 12% and 200 mL increase in FEV1 after inhaled bronchodilator, and ≥ 10% decrease in FEV1 after an EVH challenge.
RESULTS: Overall, FEV1 increased by 9.9 ± 7.9% after bronchodilator (3.93 ± 0.97 to 4.28 ± 0.91 L, P < .001) and decreased by 23.9 ± 15.0% after the EVH challenge (3.89 ± 0.89 to 2.96 ± 0.88 L, P < .001). However, the change in FEV1 after bronchodilator did not correlate with the change after EVH challenge (r = 0.062, P = .71). Significant bronchodilator responsiveness predicted a positive response to EVH challenge in 9 of 33 subjects (sensitivity 27%). Following EVH, the change in FEV1 strongly correlated with the change in FVC (FEV1 percent change vs FVC percent change, r = 0.831, P < .001; FEV1 ΔL vs FVC ΔL, r = 0.799, P < .001).
CONCLUSIONS: These results extend previous findings that demonstrate a lack of association between bronchodilator responsiveness and methacholine responsiveness. Given the poor concordance between the spirometric response to fast-acting bronchodilator and the EVH challenge, these findings suggest that the airway response to inhaled β2-agonist must be interpreted with caution and in the context of its determinants and limitations.
- airway hyperresponsiveness
- bronchoconstriction
- bronchoprovocation challenge
- bronchodilator responsiveness
- spirometry
Introduction
Asthma is characterized by variable episodes of airway narrowing that result in dyspnea, chest tightness, wheeze, and cough. The airway narrowing is consequent to an underlying airway inflammation and airway hyper-responsiveness (AHR), which is increased sensitivity and responsiveness of the airway smooth muscle to a variety of stimuli. The spirometric response to inhaled fast-acting β2-agonist is used clinically to diagnose asthma and in clinical research to verify its presence. However, a recent review highlights the poor diagnostic utility of bronchodilator responsiveness and the subsequent costs of misdiagnoses on the individual and health care system.1 Indeed, bronchodilator responsiveness does not correlate with AHR measured with the direct-acting stimulus, methacholine,2,3 which indicates that bronchodilator responsiveness is a problematic method for diagnosing asthma. However, the association between the spirometric response to inhaled fast-acting β2-agonist and indirect assessment of AHR is not known. Compared with direct-acting bronchoprovocation tests, the indirect tests of AHR more accurately reflect the biological mechanisms of airway narrowing in asthma.4 Eucapnic voluntary hyperpnea (EVH) is a well-established indirect bronchoprovocation challenge that causes airway narrowing through the actions of biological mediators released from resident and nonresident airway cells.5,6
We compared the spirometric responses to inhaled fast-acting β2-agonist and EVH in 39 adult males and females who demonstrated either significant bronchodilator responsiveness or significant AHR in response to an EVH challenge, or both. Given the wide inter-individual variation in asthma pathophysiology and phenotype, and the temporal variability in airway function within individual patients with asthma,7-9 we hypothesized that the spirometric responses to the inhaled β2-agonist and EVH challenge would not correlate with each other.
QUICK LOOK
Current Knowledge
The airway response to an inhaled fast-acting bronchodilator is routinely used to diagnose asthma and to verify its presence in clinical research. However, studies indicate that bronchodilator responsiveness does not correlate with airways hyper-responsiveness (AHR) assessed via methacholine challenge.
What This Paper Contributes to Our Knowledge
In a group of 39 adults with asthma, the spirometric response to inhaled fast-acting bronchodilator did not associate with the response to a eucapnic voluntary hyperpnea challenge (EVH). In the subjects in whom EVH caused significant airway narrowing (n = 33), a minority (9 of 33) demonstrated significant bronchodilator responsiveness. In response to EVH, the spirometric response was characterized by reductions in both FEV1 and FVC, demonstrating involvement of the peripheral airways in response to the indirect bronchoprovocation challenge.
Methods
All subjects in this retrospective analysis were participants in research protocols that took place between August 2008 and July 2019 in the Exercise Physiology Lab at Northern Vermont University-Johnson. Subjects were recruited by a combination of advertisement in local print or online publications (eg, neighborhood forums, recreational clubs), flyers posted on campus and in the surrounding community, and by word of mouth. All subjects provided written consent to participate in the studies, and all research was approved by the Northern Vermont University institutional review board for research involving human subjects. All subjects were identified with asthma by meeting at least 1 of 2 criteria: ≥ 12% and 200 mL increase in FEV1 after inhalation of a fast-acting β2-agonist, or ≥ 10% decrease in FEV1 after EVH challenge.
In all subjects, bronchodilator responsiveness was conducted as a preliminary screening study prior to participation in one of several research projects. The EVH challenge was conducted either as a preliminary screening study or as an experimental intervention in previous research.10,11 In all cases, bronchodilator responsiveness was assessed during each subject’s first visit to the lab. The EVH challenge was completed between a second and sixth visit to the lab, depending on the research study protocol being completed by the subject. In all subjects, at least 24 h separated the bronchodilator responsiveness and EVH challenge visit.
All subjects were non-smokers between the ages of 18 and 45 y, had a negative history for cardiovascular disease and other chronic illness (excepting asthma), and had an absence of respiratory infection during the 4 weeks prior to participation. Many of the subjects used an inhaled fast-acting β2-agonist on an as-needed basis; these subjects were instructed to refrain from such use for 8 h prior to each study. Subjects were instructed to avoid caffeinated food and beverages for 4 h prior to each study, and to avoid strenuous exercise for 8 h prior to each study. None of the subjects were using inhaled corticosteroids at the time of their participation.
An automated pulmonary function system was used to collect all spirometry data reported here (MGC Diagnostics, St Paul, Minnesota). Maximum forced exhalations were completed in the seated, upright position according to recommendations by the American Thoracic Society and European Respiratory Society.12 During each measurement, subjects completed maximum volitional forced exhalations to determine FVC, FEV1, forced expiratory flow during the middle half of the FVC maneuver (FEF25-75%), and peak expiratory flow (PEF). Predicted values for FVC and FEV1 are from Quanjer et al.13 Predicted values for FEF25-75% and PEF are from Hankinson et al.14
Following baseline spirometry, subjects inhaled 4 actuations of a fast-acting β2-agonist (4 × 90 μg albuterol sulfate). During each inhalation, subjects exhaled to residual volume, depressed the actuator, inhaled slowly through a holding chamber (Aerochamber, AbbVie, North Chicago, Illinois) to total lung capacity, and held their breath for 5 s prior to exhaling. Spirometry was assessed 5 min after β2-agonist and up to, but never after, 30 min after drug inhalation. In all cases, the post-bronchodilator maneuver with the highest FEV1 was selected for analysis. The largest FVC and PEF were selected from the same measurement time as the highest FEV1; however, the highest values did not necessarily occur during the single maneuver with the highest FEV1. On average, the highest postbronchodilator FEV1 occurred 15.4 ± 6.6 min after inhalation (range 5–30 min). The highest FEV1 and highest FVC were used to calculate FEV1/FVC. An increase in FEV1 of > 12% and 200 mL from baseline was considered a significant response.15
Following baseline spirometry, subjects completed an EVH challenge according to previously published methods.16 Briefly, subjects ventilated dry gas from a tank of compressed gas (21% O2, 5% CO2, balance nitrogen) for 6 min at a target ventilation equal to FEV1 × 30. Spirometry was assessed serially (at 5, 10, 15, 20, and 30 min) following the EVH challenge. The lowest post-hyperpnea FEV1 measured after the challenge was selected for analysis. The FVC and PEF were selected from the same post-EVH time point with the lowest FEV1; however, they were not necessarily from the same maneuver. A ≥ 10% decrease in FEV1 was considered a positive EVH response.16
All tabular data are reported as mean ± SD. Subjects were partitioned into 3 groups (Groups 1–3) on the basis of their responses to inhaled fast-acting β2-agonist and the EVH challenge. Due to the resulting unequal group sizes, a nonparametric test (ie, Kruskal-Wallis) was used to compare demographic and spirometry results among the 3 groups. Pairwise comparisons using the Mann-Whitney U test, including Bonferroni-corrected P values, were used to determine significant group differences. Within-group observations were analyzed using dependent t tests. Pearson correlation coefficients were used to assess associations between variables. Statistical significance was set at α < 0.05. The statistical software SYSTAT 12 (IBM, Armonk, New York) was used for all analyses.
Results
Subject characteristics and results for baseline spirometry are shown in Table 1. The table contains results for all subjects and for the 3 subgroups based on subject responses to the 2 interventions. On average, subjects were 26.9 ± 7.8 y old and had a mean body mass index of 26.3 ± 4.7 kg/m2. When all subjects were analyzed collectively, FVC was larger than predicted values (P < .05), whereas FEV1 was not different from the predicted values. On average, PEF was 93% of predicted values (8.48 ± 1.86 vs 9.04 ± 1.62 L/s for measured vs predicted, P = .02), FEF25-75% was 76.8% of predicted values (3.35 ± 1.34 vs 4.30 ± 0.61 L for measured vs predicted, P = .002), and FEV1/FVC was 0.75. Collectively, these spirometry results suggest an overall mild airway obstruction.
Figure 1 shows the correlation between the FEV1 response to inhaled fast-acting β2-agonist and the EVH challenge. There was no association between the response to β2-agonist and EVH (r = 0.06, P = .71). A positive bronchodilator responsiveness predicted a positive response to EVH challenge in 9 of 33 subjects (sensitivity 27%). Thus, in the 33 subjects with a significant decrease in FEV1 after EVH, 73% (24 of 33) did not have significant bronchodilator responsiveness.
As shown in Figure 1, the subjects were divided into 3 groups on the basis of their FEV1 responses to the 2 tests: Group 1 had a positive response to fast-acting β2-agonist and negative response to EVH; Group 2 had a positive response to both challenges; and Group 3 had a negative response to β2-agonist and positive response to EVH. Subject characteristics and baseline spirometry in the 3 groups are shown in Table 1. In Group 1 and Group 2, FEV1 was lower than predicted values (P = .045 and .02, respectively), whereas it was not different from predicted values in Group 3. FEF25-75% was significantly lower than predicted values in all 3 groups. FEF25-75% was significantly lower in Group 2 than in Group 3 (P = .008), and there was a nonsignificant trend for it to be lower in Group 1 than in Group 3 (Bonferroni-corrected P = .08). In Group 2, PEF was lower than predicted values (P = .03), whereas it was not different from predicted values in Group 1 or Group 3. FEV1/FVC was lower in Group 2 compared with Group 3 (0.65 vs 0.79, P = .008). In Group 1, FEV1/FVC was 0.71; however, this was not statistically different from the other 2 groups. In summary, the spirometry results demonstrate mild-to-moderate airway obstruction in Group 1 and Group 2, whereas airway caliber was largely normal in Group 3.
Figure 2 presents correlations between FEV1 and FVC following fast-acting β2-agonist (A, C) and the EVH challenge (B, D) in all subjects. The responses are shown both as percentage changes from baseline and as the absolute differences in liters following the 2 interventions. With inhaled β2-agonist, the change in FEV1 was significantly correlated with the change in FVC (FEV1 ΔL vs FVC ΔL, r = 0.63, P < .001). However, 95% of the subjects had a larger absolute increase in FEV1 than FVC (37 of 39 subjects). With β2-agonist, FEV1 increased by an average of 9.86%, whereas FVC increased by 1.18% (+346 ± 236 vs +61 ± 175 mL for FEV1 vs FVC). Following the EVH challenge, the change in FEV1 was strongly and significantly correlated with the change in FVC (FVC percentage change vs FEV1 percentage change, r = 0.83, P < .001) (Fig. 2B). FEV1 decreased by a greater amount than FVC in 82% (32 of 39) of subjects (–932 ± 614 vs –603 ± 653 mL for FEV1 vs FVC). In the 7 subjects in whom FVC decreased by a greater amount than FEV1, the decrease in FVC was 214 ± 181 mL greater than the decrease in FEV1 (–1.26 ± 0.87 vs –1.05 ± 0.74 L for FVC vs FEV1).
Results from correlations between baseline spirometry (FEV1 percent of predicted; FVC percent of predicted; FEV1/FVC) and the FEV1 response to inhaled β2-agonist and the EVH challenge are shown in Figure 3. Baseline FEV1 and FEV1/FVC were negatively correlated with the response to β2-agonist (P < .001), but they did not correlate with the response to the EVH challenge. Baseline FVC was not significantly associated with the responses to either β2-agonist or EVH.
Figure 4 depicts the group mean maximum changes in FEV1 and FVC after inhaled fast-acting β2-agonist and EVH in Groups 1–3. Following β2-agonist, the increase in FEV1 in Group 3 was less than that for Group 2 (P < .05). The increase in FEV1 after β2-agonist was the same in Group 1 and Group 2. After the EVH challenge, the decreases in FEV1 and FVC were greater in both Group 2 and Group 3 than Group 1 (P < .05); however, the decreases in FEV1 and FVC were not different between Group 2 and Group 3. Figure 5 shows individual subject and group mean values for FEV1 at baseline and after inhaled β2-agonist in Groups 1–3. There was a significant main-effect for the differences in baseline FEV1 among the groups (P = .007); however, after β2-agonist, FEV1 was not different among the 3 groups. Following β2-agonist, FEV1 in all 3 groups was at or above 100% of predicted.
Discussion
We compared the spirometric response to inhaled fast-acting β2-agonist with the response to EVH challenge in 39 adults with asthma. The most important new finding in this study is that the spirometric response to inhaled β2-agonist did not correlate with the response to the EVH challenge (Fig. 1). Second, in many of the subjects with significant airway narrowing after the EVH challenge, FVC also decreased substantially (Fig. 2). This finding indicates dysfunction of the peripheral airways in response to the EVH challenge. Finally, baseline FEV1 and FEV1/FVC correlated with the response to inhaled β2-agonist but not with the response to the EVH challenge (Fig. 3). Given the poor concordance between the spirometric response to inhaled β2-agonist and the EVH challenge, these findings support the conclusion that the airway response to inhaled β2-agonist must be interpreted with caution and in the context of its determinants and limitations.
With regard to the principal new finding in this study (ie, that the airway response to inhaled fast-acting β2-agonist and EVH challenge did not correlate in a group of adults with asthma), previous work has shown weak concordance between the airway responses to inhaled β2-agonist and methacholine responsiveness.2,17,18 However, whereas methacholine is a direct-acting smooth muscle agonist that acts via binding to muscarinic receptors on airway smooth muscle, EVH causes airway narrowing indirectly through the actions of pro-inflammatory mediators released by resident and nonresident airway cells.19 Similarly, the amount of bronchoconstriction following exercise (another indirect bronchoprovocation test) did not correlate with bronchodilator responsiveness in a group of 21 adults with asthma in whom both methacholine responsiveness and degree of airway inflammation demonstrated a range of severity.20
The poor concordance between bronchodilator responsiveness and EVH is not altogether surprising upon consideration of the factors that mediate the airway responses to the 2 tests. First, bronchodilator responsiveness is very much dependent on baseline airway caliber; in general, responsiveness increases as a function of reduced baseline caliber.2 Our findings also demonstrate a strong dependence of bronchodilator responsiveness on initial airway caliber, as baseline FEV1 and FEV1/FVC were significantly associated with the change in FEV1 after fast-acting β2-agonist inhalation (Fig. 3). Because daily airway caliber varies in patients with asthma, bronchodilator responsiveness should also be variable. In support of this, Silkoff et al9 reported that within-subject bronchodilator responsiveness was highly variable across 5 visits over 12 months in patients with asthma. Moreover, temporal variability in bronchodilator responsiveness was seen in subjects with both moderate and severe disease, all of whom were prescribed low-moderate and high-dose inhaled corticosteroids, respectively.9 Clinicians should therefore be mindful of the fact that the bronchodilator responsiveness will vary over time in adults with asthma, which appears to span both disease severity and phenotype.9 Conversely, previous work showed that baseline airway caliber did not correlate with the spirometric response to EVH.21,22 Similarly, in our subjects, baseline FEV1 did not correlate with the response to the EVH challenge (Fig. 3B, D, F).
Second, bronchodilator responsiveness is affected by chronic maladaptations to airway structure, such that maximum bronchodilation will be limited by increases in airway wall thickness or altered interdependence between the airways and surrounding lung parenchyma. Whereas treating airway inflammation will improve airway function by resolving the mutable inflammatory sequelae such as excessive bronchiolar smooth muscle contraction and airway wall edema,23,24 its influence on features of airway remodeling are not clear.25 In contrast, analyses suggest that structural maladaptations to the airway walls will, if anything, increase the airway response to bronchoprovocative challenges by causing exaggerated narrowing in response to a given amount of airway smooth muscle shortening.26-28 We note, however, that remodeled airways may be stiffer and thus not only more resistant to collapse but also would provide an increased load on airway smooth muscle.29
Finally, one can argue that bronchodilator responsiveness and EVH assess 2 fundamentally different phenomena. On the one hand, the bronchodilator responsiveness measures the acute increase in airway caliber due principally to airway smooth muscle relaxation. On the other hand, EVH assesses airway narrowing in response to an indirect stimulus. Airway narrowing requires some combination of the presence of inflammatory cells in the airway wall, hyperresponsive airway smooth muscle, and altered airway wall structure.4 Thus, the stimulus (ie, exogenous drug vs endogenous mediators), response (bronchodilation vs bronchoconstriction), and physiology of the response are different.
In this study, subjects were post priori placed into 1 of 3 groups depending on their spirometric responses to the 2 interventions. Several implications arise from comparisons among the 3 groups. While baseline spirometry did associate with bronchodilator responsiveness in our subjects, it did not correlate with their response to EVH; AHR was seen in our subjects both with (Group 2) and without (Group 3) significant bronchodilator responsiveness. Furthermore, while baseline spirometry was compromised in subjects in Group 2, it was largely normal in subjects in Group 3. This indicates that the lack of bronchodilator responsiveness in Group 3 was not due to remodeling of the airways and an inability to dilate. The baseline airway obstruction in subjects in both Group 1 and Group 2 was largely reversed with β2-agonist inhalation (Fig. 5). Collectively, the differences in baseline airway function, bronchodilator responsiveness, and AHR among the three groups reflects the important, ongoing challenge to unravel the phenotypic expressions of asthma. The group comparisons also highlight a critical point that baseline spirometry should not be used as a proxy for AHR.
In our subjects with significant AHR, the decrease in FEV1 was significantly associated with a decrease in FVC following the challenge (Fig. 2). The decrease in FVC suggests that peripheral airway dysfunction with premature small airway closure is an important component of the airway narrowing induced by voluntary hyperpnea and, ostensibly, other indirect stimuli (eg, exercise, inhaled mannitol). Given findings that peripheral airway dysfunction is associated with asthma severity and control,30,31 our results provide additional evidence that characterization of a small airway phenotype might benefit clinical care of patients with asthma.
Limitations
In all subjects, bronchodilator responsiveness was completed first and the EVH challenge second; the order of the 2 interventions was not randomized. However, the fact that baseline spirometry was the same on the 2 experimental days (Table 1) suggests that the results were not affected by visit number. The number of days separating the 2 visits varied among the subjects. In individuals with asthma, temporal variability in airway function is inconsistent and unpredictable, and thus in most cases will not display any consistent time-dependent behavior.9 Additionally, although we made every effort to study each subject at the same time of the day, some subjects were not able to complete both visits at the same time of day. Finally, given the unequal group sizes, these data must be interpreted with caution. We intend to continue to add subjects to this data set as they are studied.
Conclusions
In this study, bronchodilator responsiveness did not correlate with the spirometric response to EVH challenge in a group of adults with asthma. Thus, our data demonstrate that a significant proportion of patients with asthma (ie, demonstrated AHR to one or more stimulus) will not routinely exhibit a significant bronchodilator responsiveness. This finding complements previous work that indicated no correlation between bronchodilator responsiveness and methacholine re-sponsiveness in adults with asthma. While bronchodilator responsiveness should be included in the assessment and ongoing care of patients with asthma, our results should be interpreted with caution and with an understanding of the complexity of functional and structural features determining the degree of responsiveness.
Footnotes
- Correspondence: Hans Christian Haverkamp PhD, Washington State University-Health Sciences Spokane, Spokane, Washington. E-mail: hans.haverkamp{at}wsu.edu
This work was supported in part by National Institutes of Health grant P20GM103449. The authors have disclosed no conflicts of interest.
- Copyright © 2021 by Daedalus Enterprises