Journal of Asthma, 2013; 50(6): 629–633Copyright © 2013 Informa Healthcare USA, Inc.ISSN: 0277-0903 print/1532-4303 onlineDOI: 10.3109/02770903.2013.794237

DIAGNOSIS

Methacholine-Induced Airway Hyper-Reactivity Phenotypes

ERIC J GARTMAN, M.D.,1,* ERNEST K DININO, M.D.,1 PATRICK KOO, M.D.,1 MARY B ROBERTS, M.S.,2 ANDF. DENNIS MCCOOL, M.D.1

1Division of Pulmonary, Critical Care, and Sleep Medicine, Warren Alpert School of Medicine of Brown University.2The Center for Primary Care and Prevention, Memorial Hospital of RI, Pawtucket, RI USA.

Objective. The incorporation of airways conductance/resistance is a rare feature in clinical methacholine challenge test (MCT) protocols, and themajority of pulmonary laboratories rely solely on the spirometric parameters. The importance and interpretation of an MCT demonstrating asignificant decline in specific airway conductance specific airway conductance (sGaw), but not forced expiratory volume in one second (FEV1),remains undefined. This study sought to elucidate the clinical and physiologic phenotypes of individuals with a �40% sGaw decline but <20%FEV1 change.Methods.All subjects completed the AsthmaQuality of Life Questionnaire (AQLQ), followed by standardMCT, with measurementsof sGaw and an additional independent measurement of resistance and reactance by impulse oscillation system (IOS) before and afterMCT. Results. Of 201 subjects, 47(23.4%) were in Group 1 (FEV1 declined by �20%), 45(22.4%) were in Group 2 (non-significant FEV1

drop, sGaw declined �40%), and 109(54.2%) were in Group 3 (no significant decline in FEV1/sGaw). There was a nearly identical change in alloscillometric parameters and sGaw for Groups 1 and 2 versus Group 3. There were no differences between Groups 1 and 2 in any AQLQ category,and Groups 1 and 2 were statistically different from Group 3. Conclusions. Our prospective study suggests that patients with a significant sGawdecline alone during MCT are a clinically and physiologically important hyper-reactivity phenotype—whose hyper-reactivity independently wasconfirmed to be nearly identical to those with an FEV1 decline. By failing to assess airways conductance/resistance, asthma may be inappropriately“ruled out” in ,20% of the patients referred for MCT. Based on this, standardized incorporation of body plethysmography and/or IOS to MCTprotocols should be considered.

Keywords asthma, bronchial hyper-reactivity, impulse oscillometry, methacholine

INTRODUCTION

The methacholine challenge test (MCT) is the most com-mon method used to test for airway hyper-reactivity (1).There are several standardized ways to perform and inter-pret the test, but the most widely used method is to admin-ister increasing doses of methacholine and performspirometry after each dose increment. The test is completeonce there is a �20% decline in forced expiratory volumein one second (FEV1) (a positive test) or when the maximaldose of methacholine is reached (a negative test). Analternative way to assess for bronchial hyper-reactivity isby performing body plethysmographic determination ofairway resistance/conductance—with a drop in (sGaw) of�40–45% constituting a positive test (2). Given its lowercost, ease-of-use, and large body of literature, assessingchanges in FEV1 with spirometry has been the preferredmethod for methacholine challenge testing—with only anestimated 12% of the tests from clinical pulmonary func-tion laboratories reporting sGaw (3). Consequently, theincorporation of airways conductance/resistance is a rarefeature in clinical MCT protocols.

There is uncertainty as to the interpretation of a test thatshows a significant decline in sGaw, but not FEV1. It hasbeen reported previously that approximately 20% of thepatients tested with both measures will fall into this

indeterminate category (4). The reason for the disparatenature of these two indices is not clear, but it may be due toan increased and unacceptable level of sensitivity of sGawcompared to FEV1, or it possibly could represent twodifferent populations of patients—both of whom exhibitairway hyper-reactivity. Traditionally, a negative MCT byFEV1 criteria is regarded as the gold standard for ruling outasthma;(5, 6) however, if various clinically relevant phe-notypes exist, depending on the testing modality, there isthe potential that the decline in FEV1 criteria, when usedalone, could misclassify a segment of patients with airwayhyper-reactivity.

In this prospective study, we sought (1) to define theclinical phenotype of individuals that solely show a sig-nificant decline in sGaw during MCT, and compare thispopulation to individuals exhibiting a decline in both FEV1

and sGaw and to those who were non-reactive to metha-choline; and (2) to further define the physiologic changesin these groups of individuals following MCT by evaluat-ing respiratory resistance and reactance using impulseoscillometry (IOS).

METHODS

Subjects

We prospectively recruited 201 consecutive patientsreferred for MCT between July 2009 and August 2012.The Memorial Hospital of Rhode Island’s InstitutionalReview Board approved all aspects of the study and the

*Corresponding author: Eric J. Gartman, M.D., Pulmonary, Critical Care,and Sleep Medicine, Memorial Hospital of Rhode Island, 111 BrewsterStreet, Pawtucket, RI, 02860, USA; Tel:þ401 729-2636; Fax:þ401 729-2157; E-mail: eric_gartman@brown.edu

629

J A

sthm

a D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

patients signed an informed consent. Based on theirresponse to methacholine, the patients were divided intothree groups: Group 1 if the FEV1 declined by �20% atany concentration of methacholine; Group 2 if the FEV1

failed to drop by at least 20% but sGaw declined by�40%;and Group 3 if neither significantly declined.

Protocol

Spirometry was performed using standard techniques(Collins Model CPL). Forced expiratory maneuvers wereperformed in triplicate, and the best effort was analyzed.Following spirometry, lung volumes and sGawwere deter-mined by variable-pressure body plethysmography(Collins Model CPLps). The Asthma Quality of LifeQuestionnaire (AQLQ) was completed prior to testing.

As per the American Thoracic Society (ATS) testingguidelines, all patients were screened prior to testing toensure that they were not using specific medications priorto undergoing MCT, and if they had, MCT was not per-formed. (2) Serial dilutions of methacholine chloride weredelivered using an NSPIRE Dosimeter (nSpire Health Inc.,Longmont, CO, USA) and a DeVilbiss Nebulizer (Model646, DeVilbiss, Somerset, PA, USA). Following a controlinhalation of diluent, each patient took five slow inhala-tions from functional residual capacity (FRC) to total lungcapacity (TLC) from the dosimeter starting at a concentra-tion of 0.025 mg/mL. An forced vital capacity (FVC)maneuver was performed within 5 min of methacholineinhalation. If the reduction in FEV1 was <20% from base-line, five inhalations of increasing concentrations ofmethacholine, 0.065, 0.25, 1, 4, and 16 mg/mL, wereadministered. The study was terminated and sGaw wasobtained when FEV1 fell by �20% at any concentrationor when the maximum dose of methacholine had beenadministered.

Respiratory system resistance (Rrs) and reactance (Xrs)were measured using IOS (Masterscreen IOS, VIASYS)before and at the completion of the MCT. Subjects weretested as per the European Respiratory Society recommen-dations (7). The average of three acceptable tests was usedfor the analysis. Calculated parameters included: (1) resis-tance at 5 Hz (R5), total airways resistance; (2) resistanceat 20 Hz (R20), central airways resistance; (3) reactance at

5 Hz (X5), distal capacitance; (4) a frequency-integratedmeasure of reactance at low frequencies (AX), and (5)resonant frequency (Fres).

Statistical Analysis

Descriptive statistics (medians, means, and standard devia-tions for continuous variables) were used to summarize thedata. Chi-squared tests were used to assess differences ingroups for the categorical variables. Analysis of variance(ANOVA) was used to examine differences of the contin-uous variables by group. For continuous variables thatviolated the ANOVA assumptions, a non-parametricequivalent to ANOVA (Wilcoxon rank sum with aKruskal–Wallis test of significance) was used to examinedifferences by group. Pairwise comparisons of grouplevels were tested using specially coded contrasts withinthe ANOVA analysis. A p-value of .05 or less was used todenote statistical significance. Data analyses were per-formed using SAS 9.2.

RESULTS

Of 201 subjects, 47 subjects had a positive MCT based ontraditional FEV1 criteria (Group 1), 45 had a positive testsolely by sGaw decline (Group 2), and 109 were negativeby both criteria (Group 3). Baseline demographics andpulmonary function testing are displayed in Tables 1 and2. At baseline, Groups 1 and 2 had lower FEV1 and a lowerratio of forced expiratory flow between 25% and 75% ofvital capacity and FVC (FEF25-75/FVC) compared toGroup 3. Group 1 individuals had significantly higherbaseline respiratory system resistance (R5), small airwaysresistance (R5–R20), Fres, and reactance area (AX) whencompared to Groups 2 and 3. These findings are consistentwith greater small airways disease at baseline in the Group1 subjects.

Following MCT, there was a nearly identical change inall oscillometric parameters and sGaw for Groups 1 and 2.These changes were significant when compared to Group 3(Table 3). Specifically, there were significant decreases insGaw and increases in overall and percent change respira-tory system resistance (R5), small airways resistance and

TABLE 1.—Patient characteristics.

Group 1 FEV1 (þ) Group 2 FEV1 (�), sGaw (þ) Group 3 FEV1 (�), sGaw (�) p-value

N (% of total) 47 (23.4) 45 (22.4) 109 (54.2)Age (years) 43.0 46.8 45.2 NSSex (% female) 71.4 50.8 60.7 .03BMI 29.2 31.4 30.0 NS% active smoking 29.0 23.0 16.3 NS% former smokers 25.5 55.6 43.1 .01

Asthma Asthma AsthmaReason for testing (% of group) wheezing 61.7 wheezing 55.6 wheezing 39.4

SOB 25.5 SOB 28.9 SOB 27.5Cough 10.6 Cough 8.9 Cough 23.9Other 2.1 Other 6.7 Other 9.2

SOB—shortness of breath.

630 E. J. GARTMAN ET AL.

J A

sthm

a D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

percent change in small airways resistance (R5–R20),resonant frequency (Fres), and reactance area (AX). Themagnitude of change in R5, R5–R20, and AX followingmethacholine was similar for Groups 1 and 2 but differedsignificantly for Group 3. Thus, when measured viaimpulse oscillometry, patients who are positive by sGawcriteria alone (Group 2) behave the same as individualswho have a positive MCT by FEV1 criteria (Group 1), andnot like those who have negative MCT by both sGaw andFEV1 criteria (Group 3).

There were no differences seen between Groups 1and 2 in any category of the AQLQ (Figure 1).However, Groups 1 and 2 were significantly differentfrom Group 3 in all elements, and had a clinicallysignificant difference (>0.5 point difference) in symp-toms, environmental, and overall AQLQ scores (8). Ourfinding that patients in Group 2 are as symptomatic asindividuals in Group 1, in concert with the observationsthat changes in airways resistance following MCT arenearly identical in Group 1 and 2, support the notionthat there are a substantial number of individuals whohave hyper-reactive airways that are not identifiedwhen relying only on FEV1 criteria following MCT.

DISCUSSION

We sought to define the clinical and physiologic pheno-types of patients that have a significant decline in sGawduring MCT without a significant change in FEV1.Apart from the decline in FEV1, both hyper-reactivegroups (Groups 1 and 2) performed nearly identicallyduring MCT when measured via an independent assess-ment of resistance and reactance (body plethysmographyand IOS), and were different from the non-reactivegroup (Group 3). Additionally, when assessed by astandardized questionnaire, Groups 1 and 2 scored simi-larly and statistically differently from Group 3. Sinceboth hyper-reactive groups have similar changes inrespiratory resistance and reactance in response to pro-vocation and are equally affected by their respiratory-related symptoms, the clinical implication is that Group2 individuals should be considered “positive” withMCT. These findings support the notion that bothsGaw and FEV1 should be assessed during MCT, since22.4% of our patients would have been classified asnon-reactive, despite having similar changes in airwaysresistance as those who were positive by the morecommonly used FEV1 criteria.

TABLE 2.—Baseline pulmonary function testing and impulse oscillometry.

Group 1 FEV1 (þ) Group 2 FEV1 (�), sGaw (þ) Group 3 FEV1 (�), sGaw (�) p-value

FEV1 (% predicted) 89.6 (86.4, 92.9) 92.3 (89.0, 95.5) 96.6 (94.7, 99.1) .00032

FVC (% predicted) 98.4 (95.2, 101.6) 97.5 (94.3, 100.8) 100.4 (98.3, 102.6) NSFEV1/FVC (%) 75.9 (74.1, 77.6) 77.9 (76.1, 79.7) 80.3 (79.1, 81.5) .00022

FEF25–75 (% pred) 68.5 (62.5, 74.6) 78.6 (72.5, 84.8) 88.9 (84.8, 93.1) .021

FEF25–75/FVC 0.64 (0.57, 0.70) 0.71 (0.64, 0.77) 0.81 (0.76, 0.85) <.00012

TLC (% pred) 100.0 (96.8, 103.1) 99.2 (96.1, 102.4) 101.0 (98.9, 103.1) NSsGaw 0.15 (0.13, 0.17) 0.18 (0.16, 0.20) 0.18 (0.16, 0.19) NSR5 6.55 (6.03, 7.07) 5.25 (4.71, 5.78) 5.27 (4.93, 5.61) .00061

R20 4.52 (4.18, 4.85) 4.05 (3.71, 4.39) 4.18 (3.97, 4.40) NSR5–R20 2.04 (1.74, 2.34) 1.19 (0.89, 1.50) 1.09 (0.89, 1.28) .00011

X5 �2.12 (�2.41, �1.82) �1.53 (�1.83, �1.23) �1.61 (�1.80, �1.41) .0071

AX 22.77 (18.68, 26.87) 11.67 (7.49, 15.86) 10.4 (7.70, 13.10) .00021

Fres 23.37 (21.54, 25.20) 19.9 (18.03, 21.77) 17.83 (16.62, 19.04) .011

Values expressed as means (95% confidence intervals).1Significant difference between Group 1 vs. Groups 2 and 3.2Significant difference between Group 1 vs. Group 3. No significant difference between Groups 1 and 2.*Units for sGaw, L sec�1 cm H2O; R5, R20, R5-R20, X5; cm H2O L�1 sec; Fres, Hz.

TABLE 3.—Post-methacholine minus baseline values.

Group 1 FEV1 (þ) Group 2 FEV1 (�), sGaw (þ) Group 3 FEV1 (�), sGaw (�) p-value

R5 1.20 (0.78, 1.62) 1.49 (1.07, 1.90) 0.42 (0.15, 0.69) <.00011

% increase R5 20.1 (11.9, 28.3) 28.52 (20.3, 36.7) 10.12 (4.8, 15.4) .00041

R20 0.07 (�0.14, 0.27) 0.3 (0.09, 0.51) 0.12 (�0.02, 0.25) NSR5–R20 1.14 (0.81, 1.46) 1.19 (0.86, 1.51) 0.31 (0.10, 0.52) <.00011

% increase R5–R20 78.89 (�32.3, 190.1) 89.82 (�20.2, 199.8) �9.78 (�81.1, 61.5) .0071

X5 �1.18 (�1.49, �0.87) �1.12 (�1.42, �0.81) �0.36 (�0.56, �0.17) <.00011

AX 16.29 (12.01,20.56) 16.19 (11.91, 20.46) 4.98 (2.21, 7.75) <.00011

Fres 4.86 (3.20, 6.53) 4.47 (2.81, 6.13) 2.96 (1.88, 4.03) .0351

% decrease sGaw 53.4 (50.1, 56.7) 54.4 (51.1, 57.7) 16.2 (14.0, 18.5) <.00011

Values expressed as means (95% confidence intervals).1Significant difference between groups 1 and 2 vs. group 3. No differences between Groups 1 and 2.*Units as in Table 2.

AIRWAY HYPER-REACTIVITY PHENOTYPES 631

J A

sthm

a D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

The simultaneous use of independent methods to assessairways resistance in concert with measures of expiratoryairflow limitation to define reactivity to methacholine isthe major strength of this study—and it calls into questionprevious reports that suggest the lack of utility of sGawduring MCT and the use of FEV1 as the “gold standard” todefine a positive test. Khalid et al. argued that the ATS’srecommended threshold for sGaw during MCT was toolow, and that it should be increased based on the correla-tions between changes in FEV1 and sGaw (9). Our study,however, indicates that using FEV1 as the benchmark toadjust other testing parameters is not correct (i.e., themechanisms underlying the change in FEV1 are not thesame as those underlying the change in sGaw). This issupported by the finding that the 22% of our study popula-tion who did not reach the threshold 20% drop in FEV1,but had significant reductions in sGaw had changes inoscillometric parameters similar to those who are tradition-ally defined as hyper-reactive by FEV1 criteria. Thus,raising the “positive” threshold value for sGaw wouldmis-classify a substantial number of individuals with air-way hyper-reactivity. Additionally, Groups 1 and 2 hadsimilar clinical symptomatology as measured by a standar-dized questionnaire and were significantly more sympto-matic than those in group 3. This finding is consistent withprior observations that changes in oscillometric parametersduring MCT were significantly correlated with asthmasymptoms while spirometric changes were not (10). Ourfindings and those of Mansur et al. highlight the impor-tance of adding measures of resistance (body plethysmo-graphy or IOS) to measures of FEV1 following MCT.

The obvious question that arises is why the two hyper-reactive groups perform differently with respect to spirome-try following MCT. The answer may be related to differ-ences in baseline peripheral airways resistance anddifferences in the maneuvers used to measure FEV1 andairways resistance. The maneuver to measure FEV1 is one

which elicits expiratory flow limitation. If peripheral airwayresistance is increased, expiratory flow limitation would bemore easily provoked and FEV1 would be more readilyreduced in Group 1. Although FEV1 did not differ at base-line between Group 1 and 2, a similar methacholine-induced change in peripheral airway caliber in both groupswould lead to more pronounced expiratory flow limitationin Group 1, thereby reducing FEV1 more in Group 1 thanGroup 2. Our observation that Group 1 had greater periph-eral airways resistance at baseline (R5–R20, Table 2) isconsistent with this explanation and our prior observations(11). In contrast, the maneuvers utilized to measure airwaysresistance do not elicit expiratory flow limitation. sGaw ismeasured during a panting maneuver performed at FRC,and with IOS, all parameters are assessed during tidalbreathing. In this context, a similar methacholine-inducedchange in peripheral airway caliber in both groups wouldlead to similar changes in airways resistance and reactance.

Another explanation for the relative FEV1 change afterMCT between Groups 1 and 2 may reflect different airwayresponses to deep inhalation (DI). It has been welldescribed that DI promotes bronchodilation followingMCT, although the degree of bronchodilation will varydepending on degree of hyper-reactivity or inflammation(i.e., those with more responsiveness or inflammation willreverse less) (12–14). It is possible that those subjects inGroup 1 had more airway inflammation than Group 2, andthus a relative blunting of the DI-induced bronchodilation.To avoid the confounding effects of DI maneuvers, partialmaximal expiratory flows can be used to discriminatebetween asthmatic and healthy subjects (15). Pellegrinoet al. demonstrated that a substantial number of subjectswith chronic airflow obstruction failed to have a significantincrease in FEV1 with bronchodilators but had significantimprovements in partial flows and inspiratory capacity(suggesting a physiologic important reduction in FRC)(16). Although our study did not assess the partial flow

P-values are for Groups 1 and 2 versus Group 3. No differences between Groups 1 and 2.1Differences in absolute scores > 0.5 – indicative of clinically significant differences.

4

4.2

4.4

4.6

4.8

5

5.2

5.4

5.6

Activity Symptoms Emotional Environmental Overall

Group 1

Group 2

Group 3

p = 0.00031p = 0.006 p = 0.01 p = 0.0051 p = 0.0011

FIGURE 1.—Asthma quality of life questionnaire scores.

632 E. J. GARTMAN ET AL.

J A

sthm

a D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

rates, we avoided the confounding effects of DI by mea-suring sGaw during panting maneuvers at FRC and IOSparameters during tidal breathing. Our measures of lungfunction in the tidal breathing range detected significantchanges in airway function that were not detected bymeasures of FEV1. Namely, Groups 1 and 2 respondedidentically to MCT with respect to the parameters tested inthe tidal breathing range and were different from Group 3.It is unlikely that the DI maneuver associated with spiro-metry affected the subsequent measurements of sGaw andIOS parameters, given the effect of the DI during inducedobstruction is transient and reestablishment of the pre-DIphysiology occurs in 30–45 seconds. (17)

Our study represents a “real world” patient populationin whomMCT is commonly performed. We prospectivelyrecruited all-comers referred to our pulmonary functionlaboratory. However, the inclusion of all patients limitedour ability to follow them clinically as the majority ofstudy participants were referred by outside providers.Thus, we have little information regarding the responseto subsequent treatment for patients in Groups 1 or 2.Further studies assessing responses to treatment and incor-poration of serum and breath biomarkers would be extre-mely helpful in further defining these populations.

CONCLUSIONS

This study indicates that subjects experiencing a signifi-cant decline in sGaw alone during an MCT are a separate,yet important physiologic phenotype. Importantly, thisstudy showed that both responsive groups respond identi-cally by two independent methods of assessing airwaysresistance, and they are equally symptomatic. In recogni-tion that most clinical pulmonary function laboratoriesonly routinely assess spirometry during MCT, this studyindicates that the failure to evaluate sGaw or other mea-sures of airways resistance during MCT may lead to inap-propriately “ruling out” bronchial hyper-reactivity in asubstantial proportion of those tested for a clinical questionof asthma.

ACKNOWLEDGMENTS

We would like to thank Gail Dusseault and LaureenSheehan for their assistance in recruitment and performingthe pulmonary function testing.

DECLARATION OF INTEREST

All authors confirm that there exist no conflicts of interest.There was no funding associated with this study.

REFERENCES

1. Cockcroft DW. Direct challenge tests - airway hyperresponsiveness inasthma: Its measurement and clinical significance. Chest 2010; 138(2):18S–24S.

2. Crapo RO, Casaburi R, Coates AL, Enright PL, Hankinson JL, Irvin CG,MacIntyre NR,McKay RT,Wanger JS. Guidelines for methacholine andexercise challenge testing-1999. Am J Respir Crit Care Med 2000; 161(1):309–329.

3. Scott GC, Braun SR. A survey of the current use and methods of analysisof bronchoprovocational challenges. Chest 1991; 100:322–328.

4. Parker AL, McCool FD. Pulmonary function characteristics in patientswith different patterns of methacholine airway hyperresponsiveness.Chest 2002; 121(6):1818–1823.

5. Cockcroft DW, Murdock KY, Berscheid BA, Gore BP. Sensitivity andspecificity of histamine PC-20 determination in a random selection ofyoung college students. J Allergy Clin Immunol 1992; 89:23–30.

6. NieminenMM. Unimodal distribution of bronchial hyperresponsivenessto methacholine in asthmatic patients. Chest 1992; 102(5):1537–1543.

7. Marchal F, Hall GL. Forced oscillation technique. Eur Respir Mon 2010;47:121–136.

8. Juniper EF, Guyatt GH, Willan A, Griffith LE. Determining a minimalimportant change in a disease-specific quality of life questionnaire. J ClinEpidemiol 1994; 47(1):81–87.

9. Khalid I, Morris ZQ, DiGiovine B. Specific conductance criteria for apositive methacholine challenge test: Are the American thoracic societyguidelines rather generous? Respiratory Care 2009; 54(9):1168–1174.

10. Mansur AH, Manney S, Ayres JG. Methacholine-induced asthma symp-toms correlate with impulse oscillometry but not spirometry. RespiratoryMed 2008; 102:42–49.

11. Parker AL, Abu-Hijleh M, McCool FD. Ratio between forced expiratoryflow between 25% and 75% of vital capacity and FVC is a determinant ofairway reactivity and sensitivity to methacholine. Chest 2003; 124(1):63–69.

12. Kariya ST, Thompson LM, Ingenito EP, Ingram RH. Effects of lungvolume, volume history, and methacholine on lung tissue viscance. JAppl Physiol 1989; 66:977–982.

13. Slats AM, Janssen K, van Schadewijk A, van der Plas DT, Schot R, vanden Aardweg JG, de Jongste JC, Hiemstra PS, Mauad T, Rabe KF, SterkPJ. Bronchial inflammation and airway responses to deep inspiration inasthma and chronic obstructive pulmonary disease. Am J Respir CritCare Med 2007; 176(2):121–128.

14. Jensen A, Atileh H, Suki B, Ingenito EP. Airway caliber in healthy andasthmatic subjects: Effects of bronchial challenge and deep inspirations.J Appl Physiol 2001; 91:506–515.

15. Brusasco V, Crimi E, Barisione G, Spanevello A, Rodarte JR, PellegrinoR. Airway responsiveness to methacholine: Effects of deep inhalationsand airway inflammation. J Appl Physiol 1999; 87:567–573.

16. Pellegrino R, Rodarte JR, Brusasco V. Assessing the reversibility ofairway obstruction. Chest 1998; 114:1607–1612.

17. Ingram RH. Deep breaths and airway obstruction in asthma. Trans AmClin Climatol Assoc 1987; 98:80–85.

AIRWAY HYPER-REACTIVITY PHENOTYPES 633

J A

sthm

a D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.