The Impact of Age and Disease Entity on Small Airway Dysfunction in Obstructive Airway Diseases

Introduction

Small airways are critical sites of inflammation and obstruction in asthma and chronic obstructive pulmonary disease (COPD).^1,2 In asthma, small airway dysfunction (SAD), assessed by the forced expiratory flow between 25% and 75% of vital capacity percent predicted (FEF_25-75%), is linked to asthma symptoms and healthcare utilization.³ SAD is present across all asthma severities but is particularly prevalent in severe cases.⁴ In COPD, SAD, characterized by abnormal small airway resistance and reactance, correlates with disease severity⁵ and is associated with poor spirometric outcomes, severe lung hyperinflation, and reduced quality of life.⁶ Despite growing evidence on the impact of SAD in obstructive lung diseases,^5,7–10 factors influencing its development and severity remain unclear.

Aging-related lung function decline, including loss of small airways, has been reported in healthy non-smokers.¹¹ Aging is also linked to functional small airway abnormalities, as observed through computed tomography in never- and ever-smokers.¹² Asthma, which often begins in childhood, is also prevalent in adults aged 65 years or older, with elderly asthma rates ranging from 1.3% to 15.3% in Asian populations.¹³ As life expectancy rises, it is crucial to explore whether there are differences in the prevalence and severity of SAD between younger and older adult asthmatics, and thereby to explore the impact of aging on the development of SAD in obstructive lung diseases. Conversely, COPD is commonly diagnosed after age 60. While asthma and COPD differ in origin, they share physiological features, such as airflow limitation, bronchoreversibility, and type 2 (T2) inflammation.¹⁴ The factor of disease entity may be related to various degrees of physiological impairment and SAD in obstructive lung diseases, and the impact of disease entity on SAD can be studied by analyzing the patient population of age-matched asthma and COPD.

Spirometry, the standard for lung function assessment, poses challenges in elderly people (more than 65 years old).¹⁵ Interpreting airway obstruction and SAD and differentiating asthma from COPD in elderly patients using spirometry alone is difficult. Impulse oscillometry (IOS), a patient-friendly method, can measure lung mechanics in an effort-independent manner, assessing respiratory resistance and reactance during tidal breathing.^16,17 Patients with respiratory symptoms and preserved pulmonary function may still have SAD, which can be identified with the aid of IOS in addition to spirometry.^18,19 IOS is more sensitive than spirometry-based FEF_25-75% in detecting SAD in both asthma and COPD.^20,21 Moreover, bronchodilator response (BDR) assessed via IOS, particularly changes in respiratory reactance, identifies more cases of poor asthma control compared to spirometry,²² highlighting the advantages of IOS to evaluate SAD in COPD and asthma.

This study aims to evaluate the impact of aging and disease entity on SAD in age-specific asthma and COPD patients, with additional analysis of BDR on IOS and its relationship to T2 inflammatory biomarkers.

Materials and Methods

Study Subjects

This retrospective cross-sectional study analyzed the medical records of individuals aged 20 years or older who were newly diagnosed and treatment-naïve for asthma and COPD. The study population underwent spirometry and IOS testing with bronchodilator administration during a single visit at Taipei Veterans General Hospital between May 2017 and August 2021. The diagnosis of COPD was based on a clinical history, including progressive exertional dyspnea, and pulmonary function characterized by not fully reversible airflow obstruction, which was defined as forced expiratory volume in 1 second (FEV₁)/forced vital capacity (FVC) <70% after bronchodilator (BD) administration, in accordance with the Global Initiative for Chronic Obstructive Lung Disease (GOLD) guidelines.²³ Asthma diagnosis was based on GINA guideline and a clinical history of variable respiratory symptoms, including wheezing, shortness of breath, chest tightness, and cough.²⁴ COPD patients with prior asthma diagnosis or recurrent wheezing before adolescence were excluded. All COPD patients had a smoking history of >10 pack-years, while asthmatic patients were non-smokers.

Study Design

This study included a total of 77 COPD patients and 323 asthmatic patients, and those aged ≥60 years were classified into the elderly groups. After age- and gender-matching, 73 COPD patients and 73 elderly asthmatic patients were selected and categorized as COPD and EA groups, respectively. The remaining asthmatic patients under 60 years old were categorized into the younger asthma (YA) group (n = 132). The first pulmonary function tests obtained when patients were treatment-naïve were analyzed. Additionally, baseline demographic variables were collected through chart review, including sex, age, body mass index, and laboratory data.

Measurement of Spirometry and IOS with Bronchodilator Test

Herein, patients were advised to avoid inhaled bronchodilators 12 hours before spirometry and reversibility testing. The two tests were conducted using combined spirometry and IOS equipment (MS-IOS Jaeger, Würzburg, Germany, and Vmax 22 Sensor Medics, Yorba Linda, Calif) following the protocols of the European Respiratory Society.^25,26 Considering that forced expiration can affect resistance and reactance, IOS measurement was performed before each spirometry assessment. The pulmonary function test and IOS were repeated 15 minutes after inhalation of 400 mg of Salbutamol from a metered-dose inhaler with a valve-bearing spacer device. Then, from the forced expiratory volume-time curve, the values of FEF_25-75%, FVC, and FEV₁ were collected. The percentage of predicted indices was obtained. The Bronchodilator Reversibility (BDR) was positive if FEV₁ both improved by 12% and 200 mL when compared with baseline values.²⁷ The predicted values were based on our previous study.²⁸

For IOS measurement, nasal clips and manual compression of the cheeks were used to reduce the confounding factors of cheek vibration and escape of air through the nostrils. Herein, patients breathed quietly, through a loudspeaker-activated mouthpiece for 30–45 seconds without contraction of the glottis or interposition of the tongue. The loudspeaker emitted 120–150 pulses varying from 5 to 35 hz in frequency and overlapping normal breathing sounds (0.2–2 hz). Thus, the resulting pressure and flow rates allow the calculation of respiratory impedance (Z). Airway impedance includes total resistance of the respiratory system (Rrs) and reactance (X), which is resistance to inertia and elasticity of the tissues of the respiratory tract. Although Rrs at different frequencies reflects total resistance (resistance at 5 hz [R₅]) or central airway resistance (resistance at 20 hz [R₂₀]), the difference between R₅ and R₂₀ (R₅ – R₂₀) reflects the resistance of small or peripheral airways.¹⁶ Resonant frequency (F_res) represents the frequency at which reactance is 0, thereby indicating that the elastic and inert properties are equal and opposite. The area under the reactance curve between 5 hz and the resonant frequency is AX, known as the “Goldman Triangle”, thus indicating an index related to respiratory compliance and small airways patency.^29,30 We evaluated the following IOS parameters: R₅–R₂₀, reactance at 5 hz (X₅), F_res, and AX. Changes with Salbutamol inhalation in FEV₁ and IOS measurements (ΔFEV₁, ΔR₅- R₂₀, ΔX₅, ΔF_res, and ΔAX) were defined as (post-BD value – pre-BD value)/pre-BD value. Notably, the majority of ΔIOS values exhibited a negative trend. To enhance comprehension of the relationship between the relative changes in IOS and other variables, we employed the percentage reduction of IOS from baseline, instead of ΔIOS, when investigating correlation associations. The SAD was defined as FEF _25–75% less than 65% predicted and AX > 0.44.^18,31 We considered AX rather than R₅- R₂₀ for the several reasons. First, while R₅- R₂₀ reflects resistance in the small airways, AX captures the elastic properties of the entire peripheral airway network, offering a more comprehensive assessment.³² Second, AX is highly sensitive to changes in airway function, particularly small airway dysfunction, and correlates with asthma severity and patient-reported outcomes.^33,34 Furthermore, AX was better association with diagnosis of adult asthma compared to R₅- R₂₀, and AX combined with FEV1 enhances diagnostic accuracy for adult asthma.³² Third, a study also shows that AX responds more markedly to bronchodilator treatment compared to R₅- R₂₀ in patients with COPD.³⁵ As for FEF_25–75%, while no universal guideline exists, arbitrary cutoffs typically range between 60–75% predicted,⁷ with studies indicating the 95th percentile for normal is closer to 56%.³⁶ We use an FEF_25–75% <65% predicted to define SAD because it aligns with evidence suggesting this cutoff is sensitive for detecting peripheral airway impairment. A study demonstrated its abnormality and clinical relevance in a real-world study.³¹ Additionally, our previous work confirmed its correlation with IOS parameters, reinforcing its validity.¹⁸ Thus, <65% provides a practical, evidence-based threshold.

Statistical Analysis

The distribution of variables was evaluated by Kolmogorov–Smirnov goodness-of-fit test. The variables are presented as either mean ± standard deviation or median (interquartile range (IQR)) based on their distribution. One-way analysis of variance (ANOVA) with the Scheffe post hoc test and the Kruskal–Wallis test were used for continuous variables, while χ2 tests was employed for categorical variables. Meanwhile, coefficients of correlation and regression analysis were used to identify the correlations between percentage changes after bronchodilator administration in spirometry measurements and IOS parameters. The logistic regression was used to investigate the relationship between age, disease entity and SAD. Statistical significance was defined as a p value <0.05. All scientistic analyses were conducted using SPSS (IBM Corp. Released 2017.IBM SPSS Statistics for Windows, Version 25.0. Armonk, NY: IBM Corp.) and MedCalc Statistical Software version 23.1.1 (MedCalc Software bv, Ostend, Belgium).

Results

Demographic and Clinical Characteristics Related with Small Airway Dysfunction in the Study Subjects

The demographic and baseline characteristics of the study subjects are shown in Table 1. Male was predominant in the COPD and EA groups as well as female was predominant in the YA group. The mean age of YA, EA, and COPD were 44, 73, and 73 years, respectively. There was no significant difference observed in BMI, blood eosinophil counts, and total IgE levels among these three groups. The prevalence of patients with bronchodilator reversibility is 31.1%, 37.0%, and 13.7% in the YA, EA, and COPD groups, respectively (p=0.004). In comparison to YA, EA group had significantly lower values of FEV₁, FEV₁/FVC, FEF_25-75% and worse IOS parameters of X5, F_res, and AX except R₅- R₂₀. The prevalence of SAD defined as FEF _25–75% ≤ 65% predicted and AX ≥0.44 was higher in EA as compared to YA group (65.8% versus, 45.5%; 82.2% versus 62.9%).

Table 1 Demographic and Clinical Characteristics Related with Small Airway Dysfunction in the Study Subjects

In the COPD group, as compared to asthmatics, we found: (1) the values of FEV₁, FEV₁%, and FEV1/FVC were significantly lower, (2) the airway reactance X₅ and FEF_25-75% indicating small airway function were significantly worse, (3) the prevalence of small airway dysfunction defined as FEF _25–75% ≤65% was higher. Notably, the COPD group as compared to age-matched EA group presented with significantly lower values of FEV₁, FEV₁%, FEV₁/FVC, FEF_25-75% and worse X5. This result may indicate that disease entity plays a role determining the lung function impairment and SAD in obstructive lung disease. Smoking is the only different baseline characteristic between COPD and age-matched EA (p < 0.001), aside from differences in lung function parameters. This highlighted that smoking-related pathogenesis in COPD may have a greater impact on SAD compared to age-matched EA. Although, the prevalence of bronchial reversibility defined as the change of FEV₁ between pre and post-bronchodilator administration was higher in the EA as compared to COPD patients (37.0% vs 13.7%, p = 0.001). The small airway reversibility calculated by ΔR₅- R₂₀, ΔX₅, ΔF_res, or ΔAX was not statistically different among three groups.

Correlation Between Spirometric and IOS Parameters at the Baseline and After Bronchodilator Testing

The spirometric parameters of FEV₁, FVC, FEV₁/FVC, and FEF_25-75% were significantly correlated with IOS parameters of R₅- R₂₀, X₅, F_res, and AX at baseline in the asthmatics and COPD groups, respectively (Table 2). Particularly, FEF_25-75% indicating small airway function by spirometry was significantly correlated with IOS parameters of R₅- R₂₀, X₅, F_res and AX for both asthma and COPD. After the bronchodilator test, the reduction of R₅- R₂₀ and AX were significantly associated with the increase of FEV₁ in both COPD and asthma (Table 3).

Table 2 Correlation Between Spirometry and IOS Parameters at the Baseline

Table 3 Correlation Between Spirometry and IOS Parameters After Bronchodilator Test

Burden and Severity of SAD in the Study Subjects Stratified by Forced Expiratory Volume in One second

The study subjects were divided into two categories based on their FEV₁ (≥ or <80% predicted) to further compare SAD and airflow obstruction across the three groups (Table 4). There were no significant differences of IOS parameters at baseline or in their relative changes following bronchodilator inhalation among the three groups with FEV₁ ≥80% predicted. The airflow obstruction (FEV₁/FVC) was more severe in COPD patients compared to the asthmatics regardless of the categories of FEV₁% with statistical significance. In the study subjects with lung function impairment of FEV₁ < 80% predicted, the spirometric values and IOS parameters (X₅, F_res, and AX except R₅- R₂₀) were significantly worse in COPD patients compared to the asthmatics. In addition, there was a significant difference in reactance parameters between the COPD and YA groups, but no significant difference was observed between the EA and YA groups, or between the EA and COPD groups.

Table 4 Burden and Severity of SAD in the Study Subjects Stratified by Forced Expiratory Volume in One second

The Relationship Between Type 2 Biomarkers and Small Airway Reversibility

The blood eosinophil counts (BECs) and total IgE levels were significantly correlated with the reduction of R₅- R₂₀, F_res, and AX after bronchodilator test in all asthmatics. The correlation did not exist in COPD groups (Table 5). Furthermore, asthmatic patients with a ≥ 35% reduction in AX exhibited significantly higher BECs and total IgE level compared to those with a relative change of AX less than 35% (Table 6).

Table 5 Correlation Between Type 2 Inflammatory Markers and IOS Reversibility in COPD and Asthma

Table 6 Relationship Between Type 2 Inflammatory Markers and Ax Reversibility with Cut-off Value of 35% in Asthma Patients

The Factors Associated with SAD in Patients with Obstructive Airway Diseases

The supplementary tables demonstrated the factors that were associated with SAD as defined by AX (Table S1) or FEF_25-75% (Table S2). The results indicated that age and disease entity (YA, EA, or COPD) were independent risk factors associated with SAD in both univariate and multivariate logistic regression analyses.

Discussion

We have demonstrated the functional parameters of small airways assessed by IOS and their association with age and spirometric values in younger and elderly asthmatics versus COPD patients in this study. In comparison to YA, EA group had significantly lower FEV₁, FEV₁/FVC, FEF_25-75% and worse IOS parameters except R₅- R₂₀. The age- and gender-matched EA exhibited higher values of FEV₁, FEV₁/FVC, FEF_25-75% and X₅ in comparison to the COPD patients. The values of FEV₁ and FEV₁/FVC were significantly lower in the COPD group compared to the asthmatics. In addition, the airway reactance X₅ and FEF_25-75% indicating small airway function were significantly worse in COPD compared to the asthmatics. Age and disease entity (YA, EA, or COPD) were independent risk factors associated with SAD. The results may indicate that aging and disease entity simultaneously contribute to the development of SAD in obstructive lung disease. We also demonstrated that COPD significantly possessed a higher burden and severity of SAD than EA and YA while the occurrence of lung function impairment with FEV₁ <80% predicted. Our study elucidated the impact of age and disease entity on SAD in obstructive lung disease and provided the critical information on the disease management.

The age is one of the critical factors to determine the normal values of lung function in healthy subjects.²⁸ Meanwhile, aging has been reported to be associated with computed tomography-defined functional small airway abnormality in never- and ever smokers without airflow obstruction.¹² The development, prevalence and severity of SAD increased over aging had been confirmed by computed tomography, IOS and spirometry.³⁷ In this study, we selected a cohort of age- and gender-matched elderly asthmatics in comparison to COPD patients to eliminate this confounding factor. EA had higher FEV₁, FEV₁/FVC, FEF_25-75%, and less negative X5, suggesting the difference in pathophysiological change with the variable severity of peripheral airway and lung parenchymal damage between asthma and COPD. By contrast, we enrolled the treatment-naïve asthmatics and reported the EA presented with more serious severity of lung function impairment and SAD assessed by spirometry and IOS than YA. These results suggest that the aging process significantly influences lung function.

FEV₁/FVC <0.7 or < lower limit of normal, usually defined as fixed airway obstruction (FAO), is mainly attributed to the airway and parenchymal remodeling.³⁸ FAO appears to be more severe in COPD compared with YA and EA groups or even in asthma patients with FEV₁ ≥ 80% predicted observed in this study. Airway remodeling including thickening of the basement membrane, smooth muscle, and mucus gland has been demonstrated in fatal and non-fatal asthma.³⁹ The parenchyma remodeling includes loss of alveolar attachment, parenchymal destruction, and reduction in elastic fibers which are pathological characteristics of COPD.⁴⁰ Structural parenchymal abnormalities were also observed in post-mortem lungs of non-smoker patients with chronic severe asthma.⁴¹ Aging is an additional contributing factor to FAO, including the changes of the airway wall, remodeling, stiffness, and loss of terminal bronchioles and elastic recoil.¹¹ Airway resistance and reactance measured by IOS as well as FEV₁ were significantly correlated with intraluminal area, particularly in distal bronchi confirmed by three-dimensional computed tomography in patients with COPD.⁴² The inverse relationship between reactance X₅ and expiratory residual volume was found in COPD, indicating the role of reactance on airway narrowing associated with communicated lung volume.⁴³ We have found that the small airway reactance and resistance determined by IOS show a significant correlation with the spirometric values of FEV₁, FVC, FEV₁/FVC and FEF_25-75% in both asthma and COPD. These physiological parameters may be surrogate markers to evaluate the severity of small airway remodeling in asthma and COPD. Notably, the results of our study indicated that the severity of small airway remodeling assessed by spirometry and IOS was higher in COPD than EA and YA. The phenomenon of small airway remodeling may be associated with the impaction of age and disease entity in obstructive lung diseases.

Blood eosinophil counts and IgE level serve as crucial biomarkers of type 2 inflammation in individuals diagnosed with asthma. The presence of positive bronchodilator reversibility indicates a higher blood eosinophil count and IgE level and is a physiologic indicator of lower lung function in asthma.^44,45 On the other hand, type 2 airway inflammation is acknowledged as a specific phenotype in COPD and is associated with a favorable response to inhaled corticosteroid.^46–48 In a longitudinal cohort for COPD, a significant inverse correlation was observed between total serum IgE and FEV₁/FVC, independent of smoking and asthma status.⁴⁹ However, when evaluating airway reversibility by using IOS, a weak but significant correlation was observed between ΔAX and blood eosinophils or total IgE levels in individuals asthma, whereas no such correlation was evident in the COPD group, as depicted in Table 5. This observation aligns with our previous study and provides further support for the potential utility of IOS in accessing small airway reversibility in adult asthma.⁴⁴ Individuals with a ΔAX value of ≥35% exhibited significantly elevated blood eosinophil counts and total IgE levels as compared to those with a ΔAX value of <35%. This finding provides additional evidence supporting the involvement of eosinophilic airway inflammation in obstructive airway diseases and contributing to small airway reversibility.

There are several limitations that should be addressed in this study. First, disease severity cannot be classified in asthma and COPD groups for comparison because of limited patient numbers. The effect of disease severity on SAD in asthma and COPD could not be determined in this study. However, there are significant correlations between spirometric and IOS parameters in both asthma and COPD patients. Second, in the age- and gender-matched EA group versus COPD, the effect of smoking on SAD in asthma and COPD cannot be determined. Third, this study had a lack of type 2 biomarkers from the airways such as sputum eosinophils or exhaled nitric oxide, which could be more reliable to reflect airway inflammation in COPD. Finally, a prospective study with a large cohort of healthy subject, asthmatics, and COPD is needed to examine the difference in small airway function and exclude confounding factors such as age, sex, and smoking.

Conclusion

The airflow obstruction and SAD assessed by spirometry and IOS are more severe in the COPD group compared with asthmatics. The age- and disease entity-related pathophysiological change may be involved in the airway remodeling and contribute to the severity of SAD in obstructive lung disease.