Forced midexpiratory flow between 25% and 75% of forced vital capacity is associated with long-term persistence of asthma and poor asthma outcomes

Forced midexpiratory flow between 25% and 75% of forced vital capacity is associated with long-term persistence of asthma and poor asthma outcomes
rie Siroux, PhD,a,b,c Anne Boudier, MSc,a,b,c Ma€ıa Dolgopoloff, MSc,a,b,c Se
bastien Chanoine, PharmD,a,b,c Vale d,e,f g h,i Jean Bousquet, MD, PhD, Frederic Gormand, MD, Jocelyne Just, MD, Nicole Le Moual, PhD,d,e d,e €lle Varraso, PhD,d,e Regis Matran, MD,m and Rachel Nadif, PhD, Christophe Pison, MD, PhD,j,k,l Raphae a,b,c Isabelle Pin, MD Grenoble, Villejuif, Montigny-le-Bretonneux, Montpellier, Lyon, Paris, and Lille, France Background: Whether small-airway obstruction contributes to the long-term evolution of asthma remains unknown. Objectives: Our aim was to assess whether the level of forced midexpiratory flow between 25% and 75% of forced vital capacity (FEF25-75) was associated with the persistence of current asthma over 20 years and the subsequent risk for uncontrolled asthma independently of FEV1. Methods: We studied 337 participants (142 children and 225 adults) with current asthma (asthma attacks or treatment in the past 12 months) recruited to the Epidemiological Study on the Genetics and Environment of Asthma (EGEA1) and followed up at the 12- and 20-year surveys. Persistent current asthma was

From aUniversit
e Grenoble Alpes, IAB, Team of Environmental Epidemiology Applied to Reproduction and Respiratory Health, Grenoble; bInserm, IAB, Team of Environmental Epidemiology Applied to Reproduction and Respiratory Health, Grenoble; cCHU de Grenoble, IAB, Team of Environmental Epidemiology applied to Reproduction and Respiratory Health, Grenoble; dInserm U1168, VIMA (Aging and chronic diseases, Epidemiological and public health approaches), Villejuif; e Universit
e Versailles St-Quentin-en-Yvelines, UMR-S 1168, Montigny-le-Bretonneux; fUniversity Hospital, Montpellier, and MeDALL (Mechanisms of the Development of Allergy, FP7); gCHU de Lyon, Pneumology Department, Lyon; hAssistance Publique-H^ opitaux de Paris, H^opital Armand-Trousseau, Allergology Department, Paris; iUniversit
e Paris 6 Pierre et Marie Curie, Paris; jClinique Universitaire de Pneumologie, P^ ole de Canc
erologie, M
edecine Aigu€e et Communautaire, CHU Grenoble; k Inserm 1055, Grenoble; lUniversite Joseph Fourier, Grenoble; and mUniversit
e Lille Nord de France, Lille. Data collection and analysis funded in part by the Hospital Program of Clinical Research (PHRC)–Paris, PHRC-Grenoble, National PHRC 2012, the scientific committee ‘‘AGIR pour les Maladies Chroniques,’’ Merck Sharp & Dohme (MSD), and the GA2LEN project (Global Allergy and Asthma European Network). Disclosure of potential conflict of interest: V. Siroux receives research support from the Hospital Program for Clinical Research and AGIR ‘‘pour les Maladies Chroniques’’ and receives speakers’ fees from Edimark Sante and Teva. S. Chanoine receives travel support from Actelion France, GlaxoSmithKline, Chiesi, and LFB. J. Bousquet serves as a consultant for Actelion, Almirall, Meda, Merck, Merck Sharp Dohme, Novartis, Sanofi-Aventis, Takeda, Teva, and Uriach. J. Just serves as a board member for Novartis and ALK-Abell
o and receives speakers’ fees from Novartis, ALK-Abell
o, Stallergenes, and Chiesi. R. Nadif provides expert testimony for Anses and receives research support from the Hospital Program for Clinical Research. C. Pison receives consulting fees and travel support from AstraZeneca France, GlaxoSmithKline France, Novartis France, and Boehringer Ingelheim France. I. Pin receives speakers’ fees from GlaxoSmithKline and Novartis and travel support from GlaxoSmithKline, Novartis, and Chiesi. The rest of the authors declare that they have no relevant conflicts of interest. Received for publication July 6, 2015; revised September 21, 2015; accepted for publication October 27, 2015. Corresponding author: Val
erie Siroux, PhD, Institut Albert Boniot, Inserm/Universit
e Joseph Fourrier U823, Equipe d’
epid
emiologie environnementale appliqu
ee a la reproduction et a la sant
e respiratoire, Rond point de la chantourne, 38706 La Tronche cedex, France. E-mail: [email protected]. 0091-6749/$36.00 Ó 2015 American Academy of Allergy, Asthma & Immunology http://dx.doi.org/10.1016/j.jaci.2015.10.029

defined by current asthma reported at each survey. A lung function test and a methacholine challenge test were performed at EGEA1 and EGEA2. Adjusted odds ratios (ORs) were estimated for FEF25-75 decreased by 10% of predicted value. Results: A reduced level of FEF25-75 at EGEA1 increased the risk of long-term asthma persistence (adjusted OR, 1.14; 95% CI, 1.00-1.29). In children the association remained significant after further adjustment for FEV1 and in participants with FEV1 of greater than 80% of predicted value. A reduced FEF25-75 level at EGEA1 was significantly associated with more severe bronchial hyperresponsiveness (P < .0001) and with current asthma a decade later, with an association that tended to be stronger in those with (adjusted OR, 1.44; 95% CI, 1.14-1.81) compared with those without (adjusted OR, 1.21; 95% CI, 1.05-1.41) asthma exacerbation. Conclusion: Our analysis is the first to suggest that small-airway obstruction, as assessed based on FEF25-75, might contribute to the long-term persistence of asthma and the subsequent risk for poor asthma outcomes independently from effects of the large airways. (J Allergy Clin Immunol 2015;nnn:nnn-nnn.) Key words: Asthma, small airways, FEF25-75, epidemiology, longitudinal

Asthma is a chronic inflammatory lung disease characterized by airway obstruction. It has long been considered that the middle and large airways are predominantly involved in asthma. In the last years, there has been renewed interest in the role of small-airways abnormalities in patients with chronic obstructive diseases, including asthma.1,2 The small airways are defined as those less than 2 mm in caliber. These airways, which are difficult to assess and treat in asthmatic patients and have a minimal contribution to overall lung resistance, were labeled the ‘‘quiet zone.’’3 Different noninvasive methods to assess the small airways have recently been reviewed.4,5 By using spirometry, forced midexpiratory flow between 25% and 75% of forced vital capacity (FEF25-75) is considered more reflective of the small airways than FEV1.6 Compared with FEV1, which is a reproducible and appropriate measure of airway obstruction, FEF25-75 has been much less studied in epidemiologic and clinical studies. Nonetheless, previous studies suggested a clinical significance of this measure in managing childhood asthma7 and in deciphering the cause of poor lung function both in children and adults.8,9 The literature supports a role for small-airways dysfunction on the clinical expression of asthma, including worse asthma control10-13 and a higher number of exacerbations.7,14 All these 1

2 SIROUX ET AL

Abbreviations used BHR: Bronchial hyperresponsiveness EGEA: Epidemiological Study on the Genetics and Environment of Asthma, Bronchial Hyperresponsiveness, and Atopy FEF25-75: Forced midexpiratory flow between 25% and 75% of forced vital capacity FVC: Forced vital capacity ICS: Inhaled corticosteroid OR: Odds ratio

studies considered cross-sectional associations or a short-term follow-up of a few months. As recently underlined, the contribution of small-airways abnormalities in the clinical expression of asthma remained to be assessed, both in a cross-sectional and longitudinal manner.15 In particular, how small-airways function drives the long-term evolution of asthma or the long-term subsequent risk of asthma control has not been addressed yet.16 Our aim was to assess the association of FEF25-75 levels with the persistence of current asthma in children and adults followed for 20 years, the subsequent risk for uncontrolled asthma, and the severity of bronchial hyperresponsiveness (BHR) while taking FEV1 into account. We hypothesized that small-airway obstruction contributes, independently of FEV1, to the long-term evolution of asthma and poor asthma outcomes.

METHODS Population The Epidemiological Study on the Genetics and Environment of Asthma (EGEA; https://egeanet.vjf.inserm.fr) is a French cohort including a group of asthmatic patients with their first-degree relatives and a group of control subjects recruited in the early 1990s and followed up for 20 years.17 In total, 2047 adults and children were recruited from 1991 to 1995 (EGEA1). A first follow-up of the EGEA population was conducted from 2003 to 2007 (EGEA2; 1845 subjects),18 and a second follow-up was conducted from 2011 to 2013 (EGEA3; 1558 subjects). All surveys included a detailed respiratory questionnaire (self-completed in EGEA3), and the 2 first surveys included lung function testing, measure of bronchial responsiveness, skin prick tests, and total IgE measurement. No follow-up bias related to asthma status and asthma-related phenotypes was observed.19 A rich biobank, including blood samples, has been constituted (BB-0033-00043). The EGEA study was approved by the appropriate ethics committees. The current analysis was conducted among 367 patients with current asthma at EGEA1 and with available current asthma status at the 12- and 20-year follow-up studies (142 children and 225 adults, Fig 1).

Phenotypes Lung function tests were performed by trained research technicians using a standardized protocol and the European Community Respiratory Health Survey standard operating procedures. Briefly, forced spirometry was performed with regularly calibrated spirometers (Biomedin Srl, Padua, Italy; Spirometer Masterscreen, Jaeger at EGEA1 and SpiroDyn’R, Dyn’R at EGEA2). All measurements were corrected for body temperature, pressure, and saturation. Measurements were performed with the subject sitting straight and wearing a nose clip. The best of 5 forced expirations (FEV1 plus forced vital capacity [FVC]) was selected, according to the American Thoracic Society/European Respiratory Society guidelines.20 Prebronchodilator spirometric data were considered in this analysis. Study of the reproducibility of the spirometric variables showed a coefficient of variation for the best 2 loops (defined by the maximum value for FEV1 plus FVC) of 2.1%, 2.4%, and 5.8% for FEV1, FVC, and FEF25-75, respectively, at EGEA1 (n 5 811) and 1.5%,

J ALLERGY CLIN IMMUNOL nnn 2015

2.0%, and 6.2%, respectively, at EGEA2 (n 5 1190). Percent predicted values were computed by using Global Lung Initiative equations.21 For subjects with FEV1 of 80% of predicted value or greater, a methacholine bronchial challenge test was performed (maximum cumulative dose, 4 mg). The severity of BHR was assessed by using the log slope calculated by regressing the percentage decrease in FEV1 on a log10 dose and further transformed to satisfy the assumption of standard statistical analysis (normality and homogeneity of variance) by using the following transformation: ð100=ðLog slope110ÞÞ:22 A lower slope indicates greater BHR severity. Subjects with a positive answer to the questions ‘‘Have you ever had attacks of breathlessness at rest with wheezing?’’ or ‘‘Have you ever had asthma?’’ or subjects recruited as asthma cases were defined as having ever asthma at EGEA1. Current asthma was defined by the report of having had asthma attacks or asthma treatment in the past 12 months. Persistent current asthma was defined as current asthma reported at each time point (EGEA2 and EGEA 3). The others groups (not reported current asthma at EGEA2, EGEA3, or both) were defined as being in remission, including both transient and persistent remission. Asthma symptom control has been assessed in 3 classes by using responses to EGEA2 survey questions to approximate the Global Initiative for Asthma 2015 definition as closely as possible. Subjects were defined as having controlled, partly controlled, and uncontrolled asthma if they had none, 1 to 2, or 3 to 4 of the _1 following criteria, respectively: frequent daytime symptoms (defined by > _1 episodes of trouble breathing per week in the past 3 months), asthma attack or > any nighttime symptoms (defined as waking because of asthma or an attack of shortness of breath in the last 3 months), frequent use of reliever medication (defined, on average, as more than twice a week in the past 3 months), and any activity limitation (defined by the following answers: ‘‘totally limited,’’ ‘‘extremely limited,’’ ‘‘very limited,’’ ‘‘moderate limitation,’’ and ‘‘some limitation’’ to the question ‘‘Overall, among all the activities that you have done during the last two weeks, how limited have you been by your asthma?’’). Asthma exacerbation was defined at EGEA2 by means of either hospitalization for asthma or the use of oral steroids for breathing difficulties in the past 12 months.

Statistical/strategy of analysis The longitudinal association between FEF25-75 percent predicted at EGEA1 and the long-term persistent current asthma phenotype, taking into account the 20-year follow-up data, was assessed by using logistic regression model. The association between FEF25-75 percent predicted and asthma control phenotypes was assessed in a cross-sectional way at EGEA2. We further estimated the longitudinal association between the level of FEF25-75 percent predicted at EGEA1 and the subsequent risk for partly/uncontrolled asthma and asthma exacerbation assessed at EGEA2 about 12 years later. FEF25-75 percent predicted was first studied as a continuous variable (odds ratios [ORs] were expressed as the risk associated with each decrease of 10% in the level of FEF25-75 percent predicted), and although less statistically powerful, a secondary analysis was conducted by using the 70% threshold. Both cross-sectional and longitudinal analyses were first conducted in the whole studied population and then among participants with preserved FEV1 defined by an FEV1 of 80% of predicted value or greater. To provide a direct comparison between FEF25-75 and more widely used spirometric measures (FEV1 and FEV1/FVC ratio) in terms of their magnitude of association with asthma control outcomes, we estimated the ORs for an increment of 1 SD of each parameter. Any multiple regression model considered age (continuous), sex, body _1 positive skin prick test mass index (continuous), allergic sensitization (> response to any of the 11 allergens at EGEA1 [cat, Dermatophagoides pteronyssinus, Blattela germanica, olive, birch, Parietaria judaica, timothy grass, ragweed pollen, Aspergillus species, Cladosporium herbarum, and Alternaria tenuis] and 12 allergens [cypress added] at EGEA2), smoking status (never, exsmoker, and current smoker), allergic rhinitis (ever when assessed at _4, EGEA1 and active when assessed at EGEA2), and age at asthma onset (<

SIROUX ET AL 3

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

FIG 1. Flowchart of the population. *Asthma attacks or treatment for asthma in the past 12 months.

_16 years) as potential confounders. These factors were measured at 4-16, and > EGEA2 for the EGEA2 cross-sectional analysis and at EGEA1 for the longitudinal analyses, except for current smoking, for which the status at each follow-up time was also considered. Models with further adjustment for FEV1 were conducted to address the independent effect of FEF25-75 on asthma outcomes. Because adjustment for use of inhaled corticosteroids (ICSs) in the past 12 months could reflect asthma severity and therefore lead to an issue of overadjustment, we further adjusted for ICSs in the past 12 months after main analyses. Although the rate of missing data on confounding variables included in the models was low (>92% of the study population had complete data), any missing data on potential confounding variables were imputed by using multiple imputation methods (see the Methods section in this article’s Online Repository at www.jacionline.org).

62.7% had early-onset asthma, and half of the adult population had adult-onset asthma. About half of the adults and one fourth of the children had moderate to severe persistent asthma (as previously defined23). One third of the children and 60% of the adults have used ICSs in the past 12 months. More than one third of the children (36%) and half of the adults (51.4%) experienced an asthma exacerbation in the past 12 months at EGEA2. Mean FEV1 and FEF25-75 percent predicted values at baseline were 96.2% and 92.6% in children and 86.4% and 76.2% in adults, respectively. Over the followup period, 58.5% of the children and 71.6% of the adults had persistent current asthma, and most of those with remission had a persistent remission.

RESULTS Population description For more information, see Table I. The mean ages of the 142 children (39% girls) and 239 adults (56% women) with current asthma at baseline were 10.8 and 37.6 years, respectively. Among the children,

Longitudinal association between FEF25-75 percent predicted and long-term current asthma persistence Unadjusted mean FEF25-75 percent predicted values at baseline (EGEA1) were significantly lower among patients with persistent

4 SIROUX ET AL

J ALLERGY CLIN IMMUNOL nnn 2015

TABLE I. Description of the studied population at EGEA1

Age (y), mean (SD) Female sex, no. (%) Body mass index (kg/m2), mean (SD) Smoking status Never smoker, no. (%) Exsmoker, no. (%) Current smoker, no. (%) Age at asthma onset <4 y, no. (%) 4-16 y, no. (%) >16 y, no. (%) Asthma severity* Intermittent, no. (%) Mild persistent, no. (%) Moderate/severe persistent, no. (%) > _1 Skin prick test to 11 aeroallergens, no. (%) Total IgE >100 IU/mL, no. (%) Allergic rhinitis, no. (%) FEV1 (% predicted), mean (SD) FEV1 <80% of predicted value, no. (%) FEV1/FVC ratio (%), mean (SD) FEF25-75 (% predicted), mean (SD) Inhaled steroid used in past 12 mo, no. (%) Oral steroids used in past 12 mo, no. (%) Hospitalization in past 12 mo, no. (%) Asthma exacerbation  in past 12 mo, no. (%) Change in current asthma status over 20 y Persistent current asthma, no. (%) Current asthma status at each survey, no. (%) CurrentEGEA1 2 currentEGEA2 2 currentEGEA3 CurrentEGEA1 2 currentEGEA2 2 pastEGEA3 CurrentEGEA1 2 pastEGEA2 2 pastEGEA3 CurrentEGEA1 2 pastEGEA2 2 currentEGEA3

Children (n 5 142)

Adults (n 5 225)

10.8 (2.4) 56 (39.4) 17.3 (2.8)

37.6 (13.4) 127 (56.4) 23.1 (3.7)

—

120 (54.0) 65 (29.3) 37 (16.7)

89 (62.7) 53 (37.3) — 53 43 36 124 124 78 96.2 14 98.1 92.6 53 49 5 51

(40.1) (32.6) (27.3) (90.5) (90.5) (54.9) (13.4) (9.9) (7.9) (27.2) (37.3) (35.0) (3.6) (36.7)

43 (19.2) 69 (30.8) 112 (50.0) 44 54 103 157 144 159 86.4 73 90.8 76.2 137 111 18 112

(21.9) (26.9) (51.2) (72.7) (67.3) (71.3) (18.5) (32.9) (12.9) (31.1) (60.9) (49.5) (8.5) (51.4)

83 (58.5)

161 (71.6)

83 (58.5)

161 (71.6)

19 (13.4) 30 (21.1) 10 (7.0)

26 (11.6) 27 (12.0) 11 (4.9)

*Asthma severity was defined by combining a clinical asthma severity score (asthma attacks, asthma symptoms between asthma attacks, and hospitalization) and the type of asthma treatment used (none, treatment not including ICSs, and ICSs) to approximate the Global Initiative for Asthma 2002 Guidelines.  Hospitalization or use of oral steroids.

current asthma compared with values in others (77.0% vs 93.9%, P < .0001; Table II). This association was weaker but remained significant in the restricted population with preserved FEV1 and was mainly driven by observations in children (Table II). The adjusted model showed that a lower FEF25-75 percent predicted level at baseline was significantly associated with persistent current asthma both in children and adults (Fig 2). In children the association remained significant in the model further adjusted for FEV1 (OR for persistent asthma associated with a 10% decrease of FEF25-75 percent predicted, 1.30; 95% CI, 1.05-1.62) and among children with preserved FEV1 (OR, 1.30; 95% CI, 1.03-1.45). In adults the association did not remain significant in the model adjusted for FEV1 or in the population with preserved FEV1. Results remained similar after a further adjustment on ICS use in the past 12 months at EGEA1. No significant association was observed when analyzing FEF25-75 as a categorical variable (<70%).

Cross-sectional association between the FEF25-75 percent predicted with asthma control phenotypes and the severity of bronchial reactivity at EGEA2 Unadjusted means of FEF25-75 percent predicted at EGEA2 decreased, with worse asthma control symptoms and asthma exacerbation at EGEA2 (Table II). A similar trend, although less marked, was observed among subjects with preserved FEV1. Multiple regression models showed that reduced FEF25-75 values were significantly associated with a higher risk for partly controlled/uncontrolled asthma and asthma exacerbation (Table III). For both outcomes, the magnitudes of the association observed for FEF25-75 were similar compared with FEV1 and FEV1/FVC ratio (see Table E1 in this article’s Online Repository at www.jacionline.org). The association remained significant after adjustment for FEV1 and in the population with preserved FEV1 (Table III). Models further adjusted for ICS use in the past 12 months slightly decreased the association, although the association with asthma exacerbation remained borderline significant (OR, 1.29; 95% CI, 0.99-1.69). Low FEF25-75 percent predicted (<70%) values were significantly associated with an increased risk for uncontrolled asthma and asthma exacerbation both in the model adjusted for FEV1 and in the model restricted to subjects with normal FEV1 (see Table E2 in this article’s Online Repository at www.jacionline.org). Lower FEF25-75 values were associated with more severe BHR at EGEA2 (adjusted b value, 20.22; SD, 0.05; P <.0001), and the association remained significant after further adjustment on FEV1 (b value, 20.26; SD, 0.06; P < .0001).

Longitudinal associations between FEF25-75 percent predicted at baseline with asthma control phenotypes and the severity of BHR estimated a decade later An FEF25-75 percent predicted value reduced by 10% at EGEA1 was significantly associated with persistent current asthma a decade later, and the magnitude of the association was similar whether asthma was controlled or partly controlled/ uncontrolled (Fig 3). A further adjustment for FEV1 did not change the association estimates (OR for controlled asthma, 1.23; 95% CI, 1.03-1.45; OR for partly/uncontrolled asthma, 1.24; 95% CI, 1.05-1.45) and associations remained similar in subjects with preserved FEV1. Considering asthma exacerbation, the adjusted risk tended to be stronger in those with asthma exacerbation (adjusted OR, 1.44; 95% CI, 1.14-1.81) compared with those without exacerbation (adjusted OR, 1.21; 95% CI, 1.05-1.45; P comparing these 2 estimated risks 5 .10). Further adjustment on ICS use in the past 12 months did not change the associations. The estimated effect of a lower FEF25-75 value at EGEA1 on asthma control outcomes at EGEA2 was comparable with the effects of lower FEV1 and FEV1/FVC ratio values (see Table E3 in this article’s Online Repository at www.jacionline. org). Low FEF25-75 (<70%) values at EGEA1 were significantly associated with an increased risk for asthma exacerbation at EGEA2 either in the model adjusted for FEV1 (OR, 3.70; 95% CI, 1.20-11.4) or among subjects with normal FEV1 (OR, 4.85; 95% CI, 1.62-14.5; see Table E4 in this article’s Online Repository at www.jacionline.org). Among subjects with normal FEV1 at baseline, the magnitude of the association of FEF25-75 percent predicted at EGEA1 with

SIROUX ET AL 5

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

TABLE II. Unadjusted association of FEF25-75 percent predicted with asthma control symptoms, asthma exacerbation, and longterm asthma persistence Cross-sectional association, EGEA2 FEF25-75 percent predicted (no.), mean 6 SD In all subjects

Long-term current asthma persistence Remittent asthma Persistent current asthma P value In children Remittent asthma Persistent current asthma P value In adults Remittent asthma Persistent asthma P value Asthma symptom control, EGEA2 Remittent asthma Controlled asthma Partly controlled asthma Uncontrolled asthma P value Asthma exacerbation, EGEA2 Remittent asthma Current asthma without exacerbation in last 12 mo Current asthma with exacerbation in last 12 mo P value

(74) (108) (104) (39)

95.7 6 81.1 6 73.5 6 63.2 6 <.0001

26.1 27.9 29.2 30.0

(74) 95.7 6 26.1 (205) 77.4 6 29.4 (53) 65.7 6 27.9 <.0001

In subjects with preserved FEV1*

(68) (87) (73) (22)

98.8 6 89.0 6 85.8 6 82.9 6 .005

24.6 23.6 24.9 22.3

(68) 98.8 6 24.6 (152) 88.8 6 23.8 (33) 80.7 6 21.5 .0008

Longitudinal association, EGEA1 FEF25-75 percent predicted (no.), mean 6 SD In all subjects

In subjects with preserved FEV1*

(120) 93.9 6 29.3 (240) 77.0 6 29.8 <.0001

(110) 98.0 6 26.1 (165) 91.1 6 23.1 .02

(58) 99.1 6 30.5 (82) 88.1 6 23.7 .02

(55) 102.1 6 28.0 (72) 93.1 6 19.3 .03

(62) 89.0 6 27.4 (158) 71.2 6 31.1 <.0001

(55) 93.9 6 23.7 (93) 89.5 6 25.6 .30

(77) (110) (113) (43)

99.2 6 81.0 6 78.5 6 72.4 6 <.0001

29.2 28.4 31.4 28.1

(77) 99.2 6 29.2 (222) 79.6 6 29.3 (55) 71.6 6 30.4 <.0001

(73) 102.4 6 (82) 91.7 6 (84) 90.3 6 (24) 92.5 6 .01

26.1 22.7 25.9 18.0

(73) 102.4 6 26.1 (161) 92.0 6 22.8 (36) 86.0 6 25.5 .001

*Subjects with preserved FEV1 were defined as having an FEV1 of 80% of predicted value or greater.

FIG 2. Adjusted association (ORs and 95% CIs) between FEF25-75 percent predicted at baseline and long-term asthma persistence. ORs are associated for a FEF25-75 level reduced by 10% of predicted value. Values are adjusted on age, sex, body mass index, allergic sensitization, allergic rhinitis, age of asthma onset at baseline, and active smoking ever, as assessed by using baseline follow-up data.

both asthma control symptoms and asthma exacerbation at EGEA2 tended to be higher in children compared with adults, although the difference was not statistically significant (interaction P 5 0.11 and 0.10, respectively; see Table E5 in this article’s Online Repository at www.jacionline.org). FEF25-75 percent predicted reduced by 10% at EGEA1 was significantly associated with more severe BHR at EGEA2 (b estimate in the fully adjusted model, 20.22; SD, 0.06; P < .0001). Among 140 subjects with methacholine tests performed at both EGEA1 and EGEA2, a lower FEF25-75 value at EGEA1 tended

to be associated with an improved BHR over time (unadjusted P 5 .02), but the association did not remain statistically significant in the model adjusted on FEV1 at baseline (P 5 .47) and in the fully adjusted model (P 5 .89).

DISCUSSION Our analysis in a long-term follow-up cohort is the first to suggest that FEF25-75 values might contribute to the long-term evolution of asthma independently from FEV1. This study adds

6 SIROUX ET AL

J ALLERGY CLIN IMMUNOL nnn 2015

TABLE III. Adjusted cross-sectional association of FEF25-75 percent predicted with asthma symptom control and asthma exacerbation at EGEA2 In all subjects, OR (95% CI)*

Asthma symptom control Remittent asthma Controlled asthma Partly/uncontrolled asthma P value, controlled vs partly/uncontrolled Asthma exacerbation in last 12 mo Remittent asthma Current asthma without exacerbation Current asthma with exacerbation P value, current asthma with vs without exacerbation

In subjects with preserved FEV1,y OR (95% CI)*

No.

Adjusted model M1z

Model M1 further adjusted for FEV1

No.

Adjusted model M1z

74 108 143

1 1.17 (1.04-1.31) 1.34 (1.19-1.50) .008

1 1.14 (0.97-1.34) 1.22 (1.03-1.45) .33

68 87 95

1 1.15 (1.00-1.32) 1.25 (1.08-1.44) .23

74 205 53

1 1.23 (1.11-1.37) 1.44 (1.23-1.69) .02

1 1.15 (0.98-1.34) 1.42 (1.12-1.80) .04

68 152 33

1 1.18 (1.04-1.33) 1.35 (1.09-1.67) .16

*ORs are associated with a FEF25-75 level reduced by 10% of predicted value.  Subjects with FEV1 of 80% of predicted value or greater. àModel M1 is adjusted on age, sex, body mass index, smoking status, allergic sensitization, active allergic rhinitis, and age of asthma onset.

FIG 3. Adjusted association (ORs and 95% CIs) between FEF25-75 percent predicted at baseline and subsequent risk for uncontrolled asthma 10 years later. ORs are estimated for an FEF25-75 level reduced by 10% of predicted value. Values are adjusted on age, sex, body mass index, allergic sensitization, allergic rhinitis, age of asthma onset at baseline, and active smoking ever, as assessed by using baseline follow-up data.

to the existing evidence that small-airways obstruction warrants greater attention in epidemiologic studies and in the follow-up of asthmatic patients. One of the strengths of the EGEA relies on the phenotypic characterization of the population, including lung function measurements by trained clinical research technicians and careful standard operating procedures. The high response rate at both follow-up surveys adds the benefit of a 20-year follow-up of a very well-described population. Three assessments of current asthma status over 20 years allow characterizing the long-term

change in current asthma, as supported by a recent study,24 although we could not consider the variability of the expression of asthma between each survey. We were able to take into account a large number of factors that could play a confounding role in the analysis. The use of prebronchodilator versus postbronchodilator spirometry to determine the degree and severity of airway obstruction is a subject of discussion. Although postbronchodilator spirometry is recommended in patients with chronic obstructive pulmonary disease, the question is more open in asthmatic patients, especially in the young population,

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

in whom the reversible part of airway obstruction remains of interest. One main limitation with FEF25-75 relies in its reproducibility, which is usually lower compared with FEV1 or FVC. The observed coefficient of variation based on the best 2 loops in our population showed that the coefficient of variation for FEF25-75 is, as expected, higher compared with FEV1 and FVC, but it remains satisfactory, with values of around 6% at both EGEA1 and EGEA2. From a methodological point of view, a lack of reproducibility in a parameter leads to a random measurement error in the parameter. Because the error measurement in FEF25-75 should not differ according to the subsequent disease evolution, the error is expected to be nondifferential. It has been shown that nondifferential random error in estimating the parameter leads to measures of association being biased toward the null hypothesis, and therefore the measurement error that might result from the lower reproducibility of the indices cannot explain the observed association. It is of clinical relevance to identify patients at increased risk of asthma persistence over decades, but risk factors for the long-term evolution of asthma still remain largely unknown. We observed that reduced FEF25-75 values in asthmatic children increased the risk for long-term persistence of asthma. Our result is in accordance with a 25-year follow-up study of patients with childhood asthma in Tasmania, in which FEF25-75 was identified as a significant predictor of current asthma in adult life, whereas FEV1 was not.25 Our findings add to the existing literature by showing that this association remained significant among children with preserved FEV1, suggesting that it was independent from effects of the large airways. In adults the increased risk of reduced FEF25-75 values for longterm asthma persistence was no longer significant among those with preserved FEV1. The differing results of the associations between children and adults are of interest. Compared with adults, children with asthma have more often preserved FEV1 but might have small-airway dysfunctions that are not reflected by FEV1. In patients with another lung disease, bronchiolitis obliterans syndrome, who underwent lung transplantation, a reduced FEF25-75 values has been shown to occur before a decrease in FEV1.26,27 This might indicate that a reduced FEF25-75 values could be a precursor of a decreased FEV1 and in that way could be an indicator of early disease and poor prognosis. Our cross-sectional results suggest an association between FEF25-75 values and uncontrolled asthma, which is in line with previous studies. In a population of adults with asthma referred from primary care, Manoharan et al12 showed that the presence of small-airway dysfunction assessed by means of spirometry (FEF25-75) and impulse oscillation (difference between resistance at 5 Hz and 20 Rz) was associated with poor asthma control among subjects with preserved FEV1. Comparing children with uncontrolled asthma with those with controlled asthma, Shi et al11 showed a significant difference in both FEF25-75 and impulse oscillometry parameters. In children FEF25-75 in the setting of normal FEV1 has been shown to be associated with poor asthma outcomes.7 In a 3-month follow-up study of patients with initially controlled asthma, the same authors showed that uncontrolled asthma was associated with lower FEF25-75 values at follow-up but not at baseline.28 In this study impulse oscillometry parameters predicted loss of asthma control at follow-up. The present analysis further indicates that FEF25-75 could be

SIROUX ET AL 7

associated with future risk for asthma exacerbation and poorly controlled asthma a decade later. We confirmed results from previous cross-sectional studies indicating that a lower FEF25-75 value was significantly associated with more severe bronchial responsiveness.29-31 We extended these findings by means of a longitudinal analysis indicating that FEF25-75 at baseline independently predicts more severe BHR about 10 years later. However, FEF25-75 values were not independently associated with the change in BHR severity over time. This association with change in BHR severity warrants further investigations in studies with BHR assessed more than twice to avoid the phenomenon of regression toward the mean that might have biased our results. Our study underlines the interest to collect and analyze FEF25-75 more often in clinical and epidemiologic studies and to address factors associated with small-airway obstruction. The limitations related to the measurement of FEF25-75 previously discussed are of greater concern at the individual patient level in clinical practice compared with the population level in clinical and epidemiologic research. Indeed, large and statistically powerful studies might compensate concerns regarding the reproducibility of this measure, as suggested by previous etiological studies. In an international study including 20,000 children, smoking during pregnancy had a higher effect on FEF25-75 (26%) than on FEV1 (21%).8 A reduction of longterm exposure to particulate matter with an aerodynamic diameter of less than 10 mm was associated with a stronger attenuation of annual FEF25-75 decrease (16%) compared with FEV1 (9%).32 Reduced FEF25-75 values were associated with occupational exposure to vapors, gases, dusts, and fumes independently of large-airways obstruction.9 These studies indicate that FEF25-75 might be more sensitive to the deleterious effects of environmental factors than FEV1. From a clinical perspective, although the diagnosis of smallairway dysfunction should be confirmed by addition of pulmonary function tests when feasible,2 the predictive value of FEF25-75 measurement on asthma outcomes might lead to specifically targeting small-airway obstruction in asthma management. It is now accepted that inflammation is present throughout the bronchial tree of asthmatic patients and might be even more pronounced in small than proximal airways.33 Extrafine-particle (<2 mm) formulations of ICSs and long-acting b-agonist inhaler result in increased peripheral deposition compared with the nonextrafine combination.34 They might lead to better asthma control at lower doses and tend to be superior to equipotent doses of the nonextrafine combination in improving closing capacity measured by using the N2 washout test, another measure of the small airways, after 12 weeks of treatment.35 However, it remains to be shown that this treatment, used for longer periods of time, would have long-term benefit, especially in children. In conclusion, FEF25-75 might require more attention because it could be considered a predictor of poor asthma outcomes independently of FEV1. Whether changes in FEF25-75 values over time, either spontaneously or with treatment, are associated with the prognosis of asthma needs to be further addressed. We thank all those who participated in the setting of the study and the various aspects of the examinations. We also thank the 3 CIC-Inserm facilities of Necker, Grenoble, and Marseille, who supported the study and in which participants were examined. We are indebted to all the subjects who participated, without whom the study would not have been possible.

8 SIROUX ET AL

EGEA cooperative group Coordination: V. Siroux (epidemiology, principal investigator since 2013); F. Demenais (genetics); I. Pin (clinical aspects); R. Nadif (biology); F. Kauffmann (principal investigator 1992-2012). Respiratory epidemiology: Inserm U 700, Paris: M. Korobaeff (EGEA1), F. Neukirch (EGEA1); Inserm U 707, Paris: I. Annesi-Maesano (EGEA1-2); Inserm CESP/U 1018, Villejuif: F. Kauffmann, N. Le Moual, R. Nadif, M. P. Oryszczyn (EGEA1-2), R. Varraso; Inserm U 823, Grenoble: V. Siroux. Genetics: Inserm U 393, Paris: J. Feingold; Inserm U 946, Paris: E. Bouzigon, F. Demenais, M. H. Dizier; CNG, Evry: I. Gut (now CNAG, Barcelona, Spain), M. Lathrop (now University of McGill, Montreal, Quebec Canada). Clinical centers: Grenoble: I. Pin, C. Pison; Lyon: D. Ecochard (EGEA1), F. Gormand, Y. Pacheco; Marseille: D. Charpin (EGEA1), D. Vervloet (EGEA12); Montpellier: J. Bousquet; Paris Cochin: A. Lockhart (EGEA1), R. Matran (now in Lille); Paris Necker: E. Paty (EGEA1-2), P. Scheinmann (EGEA1-2); Paris-Trousseau: A. Grimfeld (EGEA1-2), J. Just. Data and quality management: Inserm ex-U155 (EGEA1): J. Hochez; Inserm CESP/U 1018, Villejuif: N. Le Moual; Inserm ex-U780: C. Ravault (EGEA1-2); Inserm ex-U794: N. Chateigner (EGEA1-2); Grenoble: J. Quentin-Ferran (EGEA1-2).

Key message d

This long-term follow-up study indicates that FEF25-75 measurement might provide additional information to FEV1 measurement to predict the long-term evolution of asthma and subsequent poor asthma outcomes.

REFERENCES 1. van der Wiel E, ten Hacken NH, Postma DS, van den Berge M. Small-airways dysfunction associates with respiratory symptoms and clinical features of asthma: a systematic review. J Allergy Clin Immunol 2013;131:646-57. 2. Lipworth B, Manoharan A, Anderson W. Unlocking the quiet zone: the small airway asthma phenotype. Lancet Respir Med 2014;2:497-506. 3. Mead J. The lung’s ‘‘quiet zone’’. N Engl J Med 1970;282:1318-9. 4. Lipworth B. Targeting the small airways asthma phenotype: if we can reach it, should we treat it? Ann Allergy Asthma Immunol 2013;110:233-9. 5. Verbanck S. Physiological measurement of the small airways. Respiration 2012;84: 177-88. 6. McFadden ER Jr, Linden DA. A reduction in maximum mid-expiratory flow rate. A spirographic manifestation of small airway disease. Am J Med 1972;52:725-37. 7. Rao DR, Gaffin JM, Baxi SN, Sheehan WJ, Hoffman EB, Phipatanakul W. The utility of forced expiratory flow between 25% and 75% of vital capacity in predicting childhood asthma morbidity and severity. J Asthma 2012;49:586-92. 8. Moshammer H, Hoek G, Luttmann-Gibson H, Neuberger MA, Antova T, Gehring U, et al. Parental smoking and lung function in children: an international study. Am J Respir Crit Care Med 2006;173:1255-63. 9. De Jong D, Boezen HM, Kromhout H, Vermeulen R, Vonk J, Postma D. Occupational exposure to vapors, gases, dusts, and fumes is Associated with small airways obstruction. Am J Crit Care Med 2014;189:487-90. 10. Farah CS, King GG, Brown NJ, Downie SR, Kermode JA, Hardaker KM, et al. The role of the small airways in the clinical expression of asthma in adults. J Allergy Clin Immunol 2012;129:381-7.e1. 11. Shi Y, Aledia AS, Tatavoosian AV, Vijayalakshmi S, Galant SP, George SC. Relating small airways to asthma control by using impulse oscillometry in children. J Allergy Clin Immunol 2012;129:671-8. 12. Manoharan A, Anderson WJ, Lipworth J, Ibrahim I, Lipworth BJ. Small airway dysfunction is associated with poorer asthma control. Eur Respir J 2014;44:1353-5. 13. Takeda T, Oga T, Niimi A, Matsumoto H, Ito I, Yamaguchi M, et al. Relationship between small airway function and health status, dyspnea and disease control in asthma. Respiration 2010;80:120-6.

J ALLERGY CLIN IMMUNOL nnn 2015

14. Bourdin A, Paganin F, Prefaut C, Kieseler D, Godard P, Chanez P. Nitrogen washout slope in poorly controlled asthma. Allergy 2006;61:85-9. 15. Postma DS, Brightling C, Fabbri L, van der Molen T, Nicolini G, Papi A, et al. Unmet needs for the assessment of small airways dysfunction in asthma: introduction to the ATLANTIS study. Eur Respir J 2015;45:1534-8. 16. McFadden ER Jr. Resurrection men and the FEF25-75. J Allergy Clin Immunol 2010;126:535-6. 17. Kauffmann F, Dizier MH. EGEA (Epidemiological study on the Genetics and Environment of Asthma, bronchial hyperresponsiveness and atopy)—design issues. EGEA Co-operative Group. Clin Exp Allergy 1995;25(suppl 2):19-22. 18. Siroux V, Boudier A, Bousquet J, Bresson JL, Cracowski JL, Ferran J, et al. Phenotypic determinants of uncontrolled asthma. J Allergy Clin Immunol 2009; 124:681-7. 19. Varraso R, Vignoud L, Benmerad M, Bonnin A, Boudier A, Bousquet J, et al. 18-year follow-up of the epidemiological study on the genetics and environment of asthma (EGEA3): study of the follow-up bias. Eur Respir J 2013;42:P983. 20. Quanjer PH, Tammeling GJ, Cotes JE, Pedersen OF, Peslin R, Yernault JC. Lung volumes and forced ventilatory flows. Report Working Party Standardization of Lung Function Tests, European Community for Steel and Coal. Official Statement of the European Respiratory Society. Eur Respir J Suppl 1993;16:5-40. 21. Quanjer PH, Stanojevic S, Cole TJ, Baur X, Hall GL, Culver BH, et al. Multi-ethnic reference values for spirometry for the 3-95-yr age range: the global lung function 2012 equations. Eur Respir J 2012;40:1324-43. 22. Chinn S, Burney P, Jarvis D, Luczynska C. Variation in bronchial responsiveness in the European Community Respiratory Health Survey (ECRHS). Eur Respir J 1997; 10:2495-501. 23. Le Moual N, Siroux V, Pin I, Kauffmann F, Kennedy SM. Asthma severity and exposure to occupational asthmogens. Am J Respir Crit Care Med 2005;172:440-5. 24. Sanchez M, Bousquet J, Le Moual N, Jacquemin B, Clavel-Chapelon F, Humbert M, et al. Temporal asthma patterns using repeated questionnaires over 13 years in a large French cohort of women. PLoS One 2013;8:e65090. 25. Jenkins MA, Hopper JL, Flander LB, Carlin JB, Giles GG. The associations between childhood asthma and atopy, and parental asthma, hay fever and smoking. Paediatr Perinat Epidemiol 1993;7:67-76. 26. Robinson PD, Spencer H, Aurora P. Impact of lung function interpretation approach on pediatric bronchiolitis obliterans syndrome diagnosis after lung transplantation. J Heart Lung Transplant 2015;34:1082-8. 27. Rosen JB, Smith EO, Schecter MG, Mallory GB, Elidemir O. Decline in 25% to 75% forced expiratory flow as an early predictor of chronic airway rejection in pediatric lung transplant recipients. J Heart Lung Transplant 2012;31:1288-92. 28. Shi Y, Aledia AS, Galant SP, George SC. Peripheral airway impairment measured by oscillometry predicts loss of asthma control in children. J Allergy Clin Immunol 2013;131:718-23. 29. van der Wiel E, Postma DS, van der Molen T, Schiphof-Godart L, Ten Hacken NH, van den Berge M. Effects of small airway dysfunction on the clinical expression of asthma: a focus on asthma symptoms and bronchial hyper-responsiveness. Allergy 2014;69:1681-8. 30. Telenga ED, van den Berge M, Ten Hacken NH, Riemersma RA, van der Molen T, Postma DS. Small airways in asthma: their independent contribution to the severity of hyperresponsiveness. Eur Respir J 2013;41:752-4. 31. Simon MR, Chinchilli VM, Phillips BR, Sorkness CA, Lemanske RF Jr, Szefler SJ, et al. Forced expiratory flow between 25% and 75% of vital capacity and FEV1/forced vital capacity ratio in relation to clinical and physiological parameters in asthmatic children with normal FEV1 values. J Allergy Clin Immunol 2010;126:527-34, e1-8. 32. Downs SH, Schindler C, Liu LJ, Keidel D, Bayer-Oglesby L, Brutsche MH, et al. Reduced exposure to PM10 and attenuated age-related decline in lung function. N Engl J Med 2007;357:2338-47. 33. Hamid Q, Song Y, Kotsimbos TC, Minshall E, Bai TR, Hegele RG, et al. Inflammation of small airways in asthma. J Allergy Clin Immunol 1997;100: 44-51. 34. De Backer W, Devolder A, Poli G, Acerbi D, Monno R, Herpich C, et al. Lung deposition of BDP/formoterol HFA pMDI in healthy volunteers, asthmatic, and COPD patients. J Aerosol Med Pulm Drug Deliv 2010;23:137-48. 35. Scichilone N, Battaglia S, Sorino C, Paglino G, Martino L, Paterno A, et al. Effects of extra-fine inhaled beclomethasone/formoterol on both large and small airways in asthma. Allergy 2010;65:897-902.

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

METHODS Although the rate of missing data on confounding variables included in the models was low (>92% of the study population had complete data), any missing data on potential confounding variables were imputed by using multiple imputation methods. Five simulations were performed by using the Markov Chain Monte Carlo method and expectation-maximization algorithm to simultaneously impute missing data for nonmonotonous

SIROUX ET AL 8.e1

quantitative and categorical variables, respectively.E1 A sensitivity analysis using a classical regression model not accounting for missing data was performed. REFERENCE E1. Xiao Y, Song R, Chen M, Hall HI. Direct and unbiased multiple imputation methods for missing values of categorical variables. J Data Sci 2012;10:465-81.

8.e2 SIROUX ET AL

J ALLERGY CLIN IMMUNOL nnn 2015

TABLE E1. Adjusted* cross-sectional association between lung function parameters and asthma control outcomes at EGEA2 OR (95% CI)y No.

Asthma symptom control Remittent asthma Controlled asthma Partly/uncontrolled asthma Asthma exacerbation Remittent asthma Current asthma without exacerbation Current asthma with exacerbation

203 37 68 98 206 37 127 42

FEF25-75

FEV1

FEV1/FVC ratio

1 1.97 (1.23-3.16) 3.04 (1.86-4.96)

1 2.16 (1.23-3.80) 3.13 (1.78-5.51)

1 2.46 (1.34-4.51) 3.61 (1.97-6.62)

1 2.31 (1.48-3.60) 3.76 (2.05-6.89)

1 2.54 (1.50-4.30) 3.16 (1.71-5.84)

1 2.95 (1.64-5.30) 4.33 (2.21-8.46)

*ORs are associated with an increase of 1 SD for each parameter to allow a direct comparison between parameters (19%, 12%, and 31% for FEV1, FEV1/FVC ratio, and FEF25-75, respectively).  Model adjusted on age, sex, body mass index, smoking status, allergic sensitization, active allergic rhinitis, and age of asthma onset.

SIROUX ET AL 8.e3

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

TABLE E2. Adjusted cross-sectional association of low FEF25-75 percent predicted (<70%) values with asthma symptom control and asthma exacerbation at EGEA2 In all subjects, OR (95% CI)* No.

Asthma symptom control Remittent asthma Controlled asthma Partly/uncontrolled asthma Asthma exacerbation Remittent asthma Current asthma without exacerbation Current asthma with exacerbation

325 74 108 143 332 74 205 53

In subjects with preserved FEV1,y OR (95% CI)*

Adjusted model M1z

Model M1 further adjusted for FEV1

Adjusted model M1z

1 2.41 (1.05-5.53) 6.67 (2.99-14.86)

1 1.63 (0.61-4.33) 3.73 (1.45-9.57)

1 1.86 (0.68-5.09) 4.45 (1.69-11.72)

1 3.86 (1.79-8.32) 8.06 (3.11-20.84)

1 2.24 (0.90-5.57) 4.43 (1.45-13.6)

1 2.50 (0.99-6.30) 5.45 (1.69-17.60)

*ORs are associated with an FEF25-75 level of less than 70% of predicted value.  Subjects with FEV1 of 80% of predicted value or greater. àModel M1 is adjusted on age, sex, body mass index, smoking status, allergic sensitization, active allergic rhinitis, and age of asthma onset.

8.e4 SIROUX ET AL

J ALLERGY CLIN IMMUNOL nnn 2015

TABLE E3. Adjusted* longitudinal association between lung function parameters estimated at EGEA1 and asthma control outcomes at EGEA2 OR (95% CI)y No.

Asthma symptom control Remittent asthma Controlled asthma Partly/uncontrolled asthma Asthma exacerbation Remittent asthma Current asthma without exacerbation Current asthma with exacerbation

212 37 71 104 217 37 136 44

FEF25-75

FEV1

FEV1/FVC ratio

1 1.62 (1.02-2.55) 1.98 (1.27-3.08)

1 1.74 (1.01-3.01) 2.17 (1.27-3.68)

1 1.85 (1.12-3.06) 2.19 (1.34-3.58)

1 1.68 (1.10-2.56) 2.41 (1.40-4.13)

1 1.88 (1.13-3.15) 2.42 (1.33-4.40)

1 1.89 (1.19-3.01) 2.46 (1.44-4.21)

*ORs are associated with an increase of 1 SD for each parameter to allow a direct comparison between parameters (19%, 12%, and 31% for FEV1, FEV1/FVC ratio, and FEF25-75, respectively).  Model adjusted on age, sex, body mass index, smoking status, allergic sensitization, active allergic rhinitis, and age of asthma onset.

SIROUX ET AL 8.e5

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

TABLE E4. Adjusted longitudinal association of low FEF25-75 percent predicted (<70%) at EGEA1 with asthma symptom control and asthma exacerbation at EGEA2 In all subjects, OR (95% CI)* No.

Asthma symptom control Remittent asthma Controlled asthma Partly/uncontrolled asthma Asthma exacerbation Remittent asthma Current asthma without exacerbation Current asthma with exacerbation

343 77 110 156 354 77 222 55

In subjects with preserved FEV1,y OR (95% CI)*

Adjusted model M1z

Model M1 further adjusted for FEV1

Adjusted model M1z

1 3.16 (1.45-6.91) 3.40 (1.62-7.17)

1 2.41 (0.91-6.45) 1.94 (0.76-4.94)

1 1.98 (0.74-5.29) 1.88 (0.74-4.84)

1 2.89 (1.41-5.92) 5.20 (2.14-12.60)

1 1.74 (0.71-4.23) 3.70 (1.20-11.40)

1 1.46 (0.60-3.59) 4.85 (1.62-14.5)

*ORs are associated with an FEF25-75 level of less than 70% of predicted value.  Subjects with FEV1 of 80% of predicted value or greater. àModel M1 is adjusted on age, sex, body mass index, smoking status, allergic sensitization, active allergic rhinitis, and age of asthma onset.

8.e6 SIROUX ET AL

J ALLERGY CLIN IMMUNOL nnn 2015

TABLE E5. Adjusted association between FEF25-75 percent predicted at baseline and subsequent risk for uncontrolled asthma 10 years later in children and adults with normal FEV1 separately

Asthma symptom control Remittent asthma Controlled asthma Partly/uncontrolled Asthma exacerbation Remittent asthma Current asthma without exacerbation Current asthma with exacerbation

In children with FEV1 >80% of predicted value, OR (95% CI)

In adults with FEV1 >80% of predicted value, OR (95% CI)

1 1.42 (1.11-1.82) 1.35 (1.08-1.69)

1 1.04 (0.85-1.26) 1.09 (0.90-1.30)

1 1.39 (1.13-1.71)

1 1.02 (0.86-1.22)

1.59 (1.08-2.35)

1.23 (0.95-1.60)

ORs are estimated for an FEF25-75 level reduced by 10% of predicted value. The model was adjusted based on age, sex, body mass index, smoking status, allergic sensitization, active allergic rhinitis, and age of asthma onset.