Real-world efficacy of the impulse oscillometry system for the management of bronchial asthma

Background: Bronchial asthma (BA) has different phenotypes, and it requires a clinically effective subtype classification system. The impulse oscillometry system (IOS) is a novel device used in respiratory functional tests. However, its efficacy has not been validated. Therefore, this study aimed to assess the relationship between BA and the IOS parameters, and the difference in the therapeutic effects of inhaled corticosteroids (ICSs) among the subtype classifications was evaluated using the IOS. Methods: The following ICSs were randomly prescribed in daily medical care: coarse-particle ICS (fluticasone propionate [FP]), fine-particle ICS (mometasone furoate [MF]), and moderate-particle ICS (budesonide [BUD]). The treatment effects were assessed using the Asthma Health Questionnaire (AHQ) and the Asthma Control Test (ACT), and were compared among the three subtypes (central predominant, peripheral predominant, and resistless) using the IOS. Results: In the central predominant type, the AHQ score of the MF group was significantly higher than that of the FP and BUD groups. In the peripheral predominant type, the MF group had a significantly lower AHQ than the FP group. In the resistless type, no significant difference was observed in the AHQ scores among the three groups. Moreover, similar results were obtained in the study using ACT Conclusions: A strong association was observed between IOS subtype classification and ICS particle size in terms of therapeutic efficiency in BA. This result indicates that the ICS particle size must be identified using the IOS subtype classification before treatment.

phenotypes or treatment responses are not correlated to pathological features [5]. Thus, further research should be performed to assess the clinical efficacy of a phenotypic classification system for asthma. The impulse oscillometry system (IOS) can be used to assess the parameters of large and small airway functions. Moreover, it is a novel and noninvasive device used for the assessment of respiratory function using the forced oscillation technique, which was first described by Dubois et al. [6][7][8]. This device uses sound waves to rapidly detect airway changes. These pressure signals can independently quantify the degree of obstruction in the central and peripheral airways. Moreover, ICSs with different particle sizes are commercially available. In asthma, therapeutic effects may differ based on the drug aerosol's particle size and location of inflammation [9,10]. Our recent study revealed that a relationship exists between the therapeutic effect based on the particle size of ICSs and the parameters of IOS in CVA patients [11]. Using the IOS, the CVA subtypes were classified as follows: the central, peripheral, and resistless type. Coarse-and fine-particle ICSs are effective for patients with central and peripheral airway resistance, respectively. IOS values are reported to more correlate with clinical symptoms in asthma compared with spirometry as well as CVA, because it is possible to detect subtle airway changes earlier than spirometry [12,13]. Airway inflammation in BA may extend from the central to the peripheral airways [14]. Thus, several studies have assessed the clinical application of the IOS for the treatment of BA [15,16].
In the current study, the clinical role of the IOS in BA was assessed by dividing the participants according to three subtypes with the R20 and R5−R20 values. Moreover, whether the IOS can be utilized in the diagnosis and therapeutic evaluation of BA was assessed.

Participants and Treatments
This single-center retrospective observational study was conducted at the Sugawara Internal Medicine and Respiratory Clinic. All consecutive patients who presented with ICS-naïve BA between April 2017 and June 2019 were included in this study. BA was defined as the presence of variable expiratory airflow limitation and history of respiratory symptoms, such as wheezing, shortness of breath, chest tightness, and cough, which vary over time and in intensity. Patients who had respiratory symptoms and detailed history and examination results supporting asthma diagnosis were diagnosed with asthma based on variable airflow limitation, as evidenced by reduced forced expiratory volume in 1 second (FEV1) and FEV1/forced vital capacity (FVC) ratio, and a significant increase in lung function 4 weeks after anti-inflammatory treatment [1]. Patients diagnosed with bronchial asthma were treated according to the Japanese asthma prevention and management guidelines [17]. Three ICS were used for treatment and were randomly assigned to the patient groups: fluticasone propionate (FP: average particle size = 4.4 um [coarse particle]), mometasone furoate (MF: average particle size = 2.0 um [fine particle]), and budesonide (BUD; average particle size = 2.4 um [moderate particle]). In addition to the ICS therapy, all patients were treated with salmeterol or formoterol (long-acting beta2agonists) and montelukast (leukotriene receptor antagonists). Therapeutic efficacy was assessed at baseline and 2 and 4 after treatment using the Asthma Health Questionnaire (AHQ)-33-Japan and at baseline and 4 weeks after treatment with the asthma control test (ACT). The experimental protocols and the purpose of the research were explained to all participants. The study was conducted in accordance with the declaration of Helsinki and was approved by the institutional ethical committee of Sapporo Medical Association.

Measurements of IOS and Pulmonary Function
IOS measurements were assessed using a commercially available IO device (Master Screen IOS, Jaeger, Germany) according to the manufacturer's recommendations [8] .The resistance at 5 Hz (R5: indicating total airway resistance), resistance at 20 Hz (R20: representing central airway resistance), difference between R5 and R20 (R5−R20: index of the small airways), reactance at 5 Hz (X5: relating to compliance), resonant frequency (Fres), and integrated area of low-frequent X (AX) values were evaluated [18][19][20]. The use of Fres and AX for detecting expiratory flow limitations was proposed.
After the IOS measurement, spirometry was performed by using MasterScreen IOS-Jaeger (Germany).
To prevent the occurrence of any negative effects caused by forced expiration on the airway, spirometry was not performed before IOS measurement. The percentage of FVC (%FVC), percentage of FEV 1 (%FEV 1 ), FEV 1 /FVC ratio, percentage of the maximal mid-expiratory flow (%MMEF), and percentage of the peak expiratory flow (%PEF) were assessed.
The predicted values (PV) of the parameters of IOS were calculated using the equations reported by Vogel et al. [21]. The patients diagnosed with BA were divided based on three subtypes, as shown in Table 1.

AHQ and ACT
All patients with asthma completed the AHQ at baseline and 2 and 4 weeks after treatment and the ACT at baseline and 4 weeks after treatment.
The Japanese version of the AHQ, which is a disease-specific, health-related quality of life questionnaire, was developed [22,23]. The clinical validity of the AHQ was evaluated, and it was found to be reliable and effective for discriminative purposes. Thus, it can be used with confidence in clinical research. The AHQ has six subscales (asthmatic symptom, emotion, daily activity, factors that worsened symptoms, social activity, and economics) and comprises 32 items (grades 0-4) and one face scale (grades 1-5). A higher AHQ score reflect a worse health status with respect to these 33 items.
The ACT is a validated, patient-completed measure of asthma control, which comprises five questions used to assess activity limitation, shortness of breath, night-time, symptoms, use of rescue medication, and overall rating of asthma control within the last 4 weeks [24]. The questions are scored from 1 (worst) to 5 (best), and the ACT score is obtained by obtaining the sum of the responses, with a maximum best score of 25. An ACT score of 19 indicated the highest area under the ROC curve. Thus, a score ≥20 is considered the optimal cutoff point for a well-controlled asthma within the last 4 weeks.

Measurement of FeNO
Fractional exhaled nitric oxide (FeNO) measurement was performed according to the 2011 recommendations of the American Thoracic Society [25] and prior to spirometry to prevent modifying its values. Niox® VERO (Circassia AB, Sweden), a portable analyzer, was used to assess whether the expiratory flow was maintained at 50 mL/s using acoustic emission signal.

Statistical Analysis
Statistical analysis was performed using an intention-to-treat strategy. Numeric variables were expressed as means ± standard error of mean. Differences among the groups were assessed with non-repeated analysis of variance with or without the Student-Newman-Keuls test. The differences before and after treatment were compared using paired t-test or Wilcoxon signed-rank test.
Categorical variables were tested using the chi-square test. A p value <0.05 was considered statistically significant. Microsoft Excel 2007 (Microsoft Corporation, the USA), Excel Statistical Program File (ystat2008.xls, Igakutosho-shuppan Ltd., Tokyo, Japan), and GraphPad Prism v8 software (GraphPad, Inc., San Diego, CA, the USA) were used for data analysis and graph generation.
Spearman's correlation analysis and JMP13.0 (SAS Institute, Cary, NC, the USA) were used to evaluate the coefficients of determination (ρ), residuals, and significance (p) to identify the associations between the pulmonary function test score and IOS index.

Selection of participants
Of the 245 patients with bronchial asthma who were screened, 137 were excluded due to a history of treatment (n=33), asthma and COPD overlap (n=10), and refractory asthma (n=3), use of ICS other than dry powder inhalation (n=50), and dropout (n=41). In total, 108 patients were enrolled in this study. These patients were divided based on three subtypes according to the IOS result, as described in the Methods section: CP type (n=34), PP type (n=58), and resistless type (n=16). In addition, these patients were randomly prescribed with any ICS (FP, MF, or BUD) to compare the therapeutic effects

Baseline characteristics of the participants
The characteristics of the patients are summarized in Table 2. The number of young participants in the CP group was slightly higher than that in the other groups. No difference was observed among the three groups in terms of sex ratio, duration of disease, smoking rate, BMI, IgE level, number of eosinophils in the peripheral blood, and FeNO. In terms of spirometry parameters, the %FEV1, FEV1/FVC ratio, and %MMEF of the PP type were significantly lower than those of the CP and resistless types. The %PEF of the resistless type was higher than that of the CP and PP types. In addition, the characteristics of the three ICS groups are summarized in Table S1. No statistically significant differences were observed in terms of characteristics among the groups.

Therapeutic evaluation with the AHQ
The severity of asthma symptoms and patient-related outcome were evaluated with the AHQ, which is a simple and objective measurement tool for asthma symptoms (Fig 2A). In the CP type, the FP group had a higher AHQ score than the MF and BUD groups at baseline. The AHQ score did not significantly differ among the three ICS groups in the CP type at 2 weeks after treatment and in the PP type at baseline. By contrast, there were significant differences among the subtypes. In the CP type, the AHQ score of the MF group was significantly higher than that of the FP group ( that the improvement of asthma symptoms was closely correlated to the treatment effect based on the particle size of ICSs.

Therapeutic evaluation with ACT
The GINA guidelines indicate that the classification of asthma control directly reflects the efficacy of therapeutic interventions. Thus, it may be more useful clinically. In relation to this, the ACT was effective in predicting GINA-defined asthma control categories. Although there was no statistically significant difference in ACT scores among the three ICP groups in the CP type at baseline, major differences were observed after treatment. The ACT score of the FP group was significantly higher than that of the MF and BUD groups (24.3 vs 21.7, 22.3, respectively, p<0.05) at 4 weeks after treatment. By contrast, in the PP type, the ACT score of the FP group was lower than that of the MF and BUD groups (22.8 vs 24.6, 24.4, respectively, p<0.01) at 4 weeks after treatment (Fig 2B). In the resistless type, the scores of the three ICS groups did not significantly differ.

Therapeutic effects on pulmonary function and IOS
We examined whether the classification using the IOS value was correlated to the clinical data before and after treatment. FeNO is useful for the diagnosis and identification of treatment effect in asthma.
In this analysis, FeNO improved with treatment. However, no difference was observed between the three groups. The spirometric value was a clinical index relatively similar to IOS, and it had a similar treatment behavior to IOS, but the difference between the three groups after treatment was not significant in FEV1 and FVC. The IOS value itself significantly improved with treatment in all groups, but unlike the respiratory function test values, the difference between the three groups was still observed even after treatment, indicating that IOS may be more effective in reflecting airway constriction (Table 3). When the IOS and respiratory function test results of the participants were compared (Table 4), most items were found to be correlated. However, no significant association was observed between most spirometric value and R20, which is an indicator of the central airway function, and only %PEF was correlated to the IOS value. This result indicates that %PEF might relatively reflect central airway obstruction in asthma.

Discussion
The current study showed that the differences in asthma phenotypes based on the IOS parameters might be correlated to the efficacy of ICSs. Moreover, IOS was significantly correlated to the severity of asthma. Our previous study has revealed that the phenotypic differences in IOS parameters were associated with the efficacy of ICSs in CVA patients [11]. Thus, we examined the role of IOS in BA in this study. From the observation of IOS, it became clear that phenotype of BA predominating in central airway lesions and BA predominating in peripheral airway lesions exist. Based on the result, the participants were divided based on three subtypes, which were as follows: the CP, PP, and R types. In terms of background characteristics, the participants did not significantly differ (Table 2).
However, the age of the participants with the CP type was lower than that of participants with the PP and R type. This result may reflect that age at onset in the CP type was earlier than that in the PP type. The spirometry values differed at baseline in the three groups. The %FEV1, FEV1/FVC, and %MMEF of the PP type were lower than those of the CP and R types ( Table 2). Some studies showed that the small airways were the major site of inflammation and obstruction in asthma [26][27][28][29][30][31]. Thus, we believed that the PP type may be more severe than the CP and R types. Moreover, IOS is a useful device used for the simultaneous assessment of diseases in the peripheral and large airways.
Therefore, the role of IOS in BA therapy was examined in this study, which first reported about the subtype of bronchial asthma using IOS. The efficacy of FP and BUD was highest in the CP type. In the PP type, the AHQ score decreased in the MF group compared with FP group (Fig.2A). Regarding the ACT in the CP type, the score of the FP group was significantly higher than that of the MF group. In the PP type, the ACT score of the FP group was significantly lower than that of the MF group (Fig 2B).
These results indicate that ICSs with coarse-and fine particles are most suitable for the CP and PP type, respectively. Considering these results, the differences in phenotypes based on the IOS parameters might be correlated to the efficacy of ICSs. Thus, IOS examination should be performed on patients with BA prior to ICS administration, and it is better to use ICS with a coarse-particle size for the CP type and ICS with a fine-particle size for the PP type.
When we reconsidered the patient group using the IOS values, we focused on a group called the whole airway type, which was characterized by high R20 and high R5−R20 (Table S2). This group had the lowest spirometry value, the highest AHQ score, and the lowest ACT score. The whole airway type is believed to present with inflammation or remodeling in both the central and peripheral airways. As previously mentioned, coarse-particle ICSs are suitable for CP type and fine-particle ICSs for PP type.
Thus, if a patient with refractory asthma have high R20 and R5−R20, combination therapy with coarse-and fine-particle ICSs should be considered.
The current study had some limitations. Since IOS is a novel method, data about the exact meaning, interpretation, and clinical application of IOS parameters versus spirometric parameters are limited.
Reference values used in this study are based primarily on data obtained from Caucasians, and they might cause bias when adapted in Japanese. Several ICSs were randomly prescribed by multiple physicians in daily asthma treatment. However, some bias might exist in the prescription of ICS.
Furthermore, the follow-up period in the current study was short, this study was retrospective observational and the aerosol type of ICS were not examined. Thus, further research must be conducted to identify the most effective individualized treatment for BA using IOS.

Conclusions
The effect of treatment with ICSs and their correlation to the parameters of IOS in patients with BA were investigated. The phenotypic differences in IOS parameters might be associated with the efficacy of ICSs. That is, coarse-particle ICSs were found to be effective for patients with central airway resistance and fine-particle ICSs for those with peripheral airway resistance. In addition, IOS could detect the diversity in airway dysfunction, thereby indicating its efficacy considering that the lesion sites are the central and peripheral airways. Hence, we recommend the use of IOS in the examination of BA before ICS administration. Moreover, IOS was effective when used in the selection of ICS and the evaluation of the BA phenotype.       budesonide. The bars were expressed as mean ± standard error of mean. The differences among the ICS group were analyzed using the non-repeated analysis of variance with the Student-Newman-Keuls test. *: p<.05, **: p<.01