Comparison of impulse oscillometry measurements according to body mass index in patients with asthma

Abstract

Background

Obesity can lead to increased airway resistance, especially in the peripheral airways. Impulse oscillometry (IOS) may detect small airway changes in asthma. In this study, we aimed to investigate the effect of obesity on the small airways of the lungs in asthmatic patients by IOS measurements.

Methods

One hundred newly diagnosed asthmatic patients (28 male) were divided into three groups (BMI (< 25.0 kg/m²: Group-1; 25.0 to 29.9 kg/m²: Group-2; ≥30 kg/m²: Group-3). Demographic data, comorbidities, Beck anxiety and depression questionnaires, PFTs, and IOS measurements of the patients were recorded.

Results

There was no detected significant difference concerning age, gender, smoking history, pack year, comorbidities (except hypertension and anxiety), asthma cardinal symptoms (except dyspnea), and pulmonary function tests among groups (p > 0.05). There were significant differences regarding R5, R5%, R20%, X5, X20, Fres, and R5-R20 among the three groups. Group 1 significantly differed about R5-R20, X5, AX, and Fres compared to groups 2 and 3 in pairwise comparisons. Based on these results, a difference was detected between BMI < 25 and ≥ 25. There was only one significant variable (R20%) between BMI < 30 and ≥ 30. When performing a correlation test between IOS parameters and BMI, it was observed to be significantly correlated (p < 0.05) except X5% and X20% (p > 0.05).

Conclusion

Obesity and being overweight affected peripheral airways and reactance; however, obesity also affects central airways, based on our results. In addition, it was thought that IOS may detect earlier than PFTs on small airway changes.

Background

Asthma is a common disease characterized by airway inflammation presenting clinically variable airway narrowing, airway hyperresponsiveness, and, in some cases, loss of FEV₁. Although asthma pathophysiology has not yet been fully understood, inflammation, structural abnormalities, injury, and repair processes have been shown in small airways despite previously being regarded as in large airways [1, 2]. The diameter of small airways is less than 2 mm, and the contribution of total resistance is less than 20%; for that reason, early diagnosis of small airway diseases (SADs) is challenging to detect utilizing conventional pulmonary function tests [3]. However, multiple-breath nitrogen washout, respiratory oscillometry, and ventilation imagining pave the way for diagnosing SADs. No matter what it is, conventional spirometry will still be used, and it is considered that three peripheral airway measurements will be complementary to draw the whole picture [2].

Impulse oscillometry (IOS), a non-invasive and established technique, provides some data about respiratory mechanical features. R5 (total airway resistance), R20 (resistance of the central airway), and R5-R20 reflect changes in the obstruction of peripheral airways. However, it remains unclear whether R5–R20 reflects small airway caliber [4]. A study showed that R5, R5–R20, area of reactance (AX), and resonant frequency (Fres) significantly increased in patients with SAD (n = 42) compared to those without SAD [5].

Obesity includes mechanical and inflammatory lung and airway effects [6]. It significantly affects both distal and central airway function [7]. Because obesity-related asthma may be a different kind of inflammation (such as TNF-α, IL-1β, and IL-6 and pro-inflammatory cytokines are increased, anti-inflammatory adiponectin is reduced) and contributory factors, asthma control is difficult in these patients [1]. Obesity can lead to increased airway resistance, especially in the peripheral airways, leading to difficulty breathing and reduced FRC and ERV, which indicate peripheral airway dysfunction [8]. Small airway abnormalities present in healthy obese individuals and body weight may negatively affect distal airway function. Therefore, IOS may be an early functional marker of asthma, especially in obese individuals reporting symptoms despite having normal airflow [9].

We hypothesized that obesity affects small airways independently. In this study, we aimed to investigate the effect of obesity on the small airways of the lungs in asthmatic patients by IOS measurements.

Methods

The study was designed as a prospective cross-sectional study, which involved observing a defined population simultaneously. The study was clearly explained to all patients, and an informed consent form was obtained from them. This study complies with the Declaration of Helsinki and was performed according to ethics committee approval. Ethical approval was obtained from the University of Health Sciences Turkey Medical Faculty, Dr Suat Seren Chest Diseases, and Thoracic Surgery Research and Training Hospital.

Patients

The eligible patients aged > 18 years visiting the outpatient clinic with newly diagnosed asthma who had symptoms and confirmation with a 12% and 200 mL change in FEV₁ after a short-acting bronchodilator and/or positive bronchial provocation test (BPT) (PC20 less than 16 mg/mL) were included. Seventeen patients used ICS in a very short time previously (n:6,6,5 patients, separately, for each group); however, they did not use it in the past six months. After the definitive diagnosis of asthma, demographic data (age, gender, BMI, smoking history, treatment), comorbidities, pulmonary function tests, and questionnaire scores were recorded. They were divided into three groups according to BMI, which measures body fat based on height and weight. The groups were Group 1, with a BMI of less than 25.0 kg/m²; Group 2, with a BMI between 25.0 and 29.9 kg/m²; and Group 3, with a BMI of 30 kg/m² or higher. The study did not include a control group consisting of obese patients without a concurrent diagnosis of asthma. Pregnants, current ICS use, immunosuppressive therapy for any reason, patients with other lung disorders (COPD, interstitial lung diseases, emphysema, active/sequel tuberculosis, post-COVID), smoking history > 10 PY, surgery or myocardial infarction in the past three months, insufficient cognitive function were excluded.

Procedures

Pulmonary Function Tests: For the diagnosis, all patients performed pulmonary function testing as a post-bronchodilator test (Zan 100 Bodyplethysmographie, Germany). A 200 ml and 12% change in FEV₁ and clinical agreement were considered asthma. A bronchial provocation test with methacholine was performed in eligible patients, and a 20% decrease in FEV₁ (PC20) was considered a positive bronchial provocation test (Zan 300 Bodyplethysmographie, Germany). A positive BPT was considered asthma if clinical features agreed with asthma [10]. PFTs were performed according to Quanjer 2012 GLI norm.

Impulse Oscillometry: It was measured all variables (R5, R20, R5-R20, X5, X20, Fres, and AX) with Vyntus IOS (CareFusion, Germany 234 GmbH), a non-invasive and established technique that provides data about respiratory mechanical features. IOS measurement was performed at least one day after PFT. Patients were asked to breathe at average tidal volume for 30–60 s with their noses clipped and cheeks supported by their hands. The mean values of three technically acceptable measurements compatible with ERS guideline [4] were used in the analysis.

Beck Anxiety and Depression Inventory: The Beck Depression and Anxiety Inventory consists of 21 questions separately. Each question contains four options. According to total scores for depression, 11–16: Mild mood disturbance, 17–20: Borderline clinical depression, 21–30: Moderate depression, 31–40: Severe depression, over 40: Extreme depression. According to total scores for anxiety, a score of 0–21: low anxiety, a score of 22–35: moderate anxiety, and a score of 36 and above: potentially concerning levels of anxiety.

Statistical Analysis: Analyses were performed using the statistical software SPPS Version 28 (IBM Corp., Armonk, NY, USA). Categorical variables were presented as percentages and numbers. The Chi-Square test was used to compare categorical variables. If numerical variables fit the normal distribution, a t-test was used for two-group comparisons. The one-way ANOVA test was used for three-group comparisons, and these variables were presented as mean ± SD. If the variables did not fit a normal distribution, the Mann-Whitney U test was used for two-group comparisons. The Kruskal-Wallis test was used for three-group comparisons, and these variables were presented as median (IQR25,75). Adjusted p-value was used in post hoc analysis. In addition, the correlation test was performed between BMI and IOS parameters. All variables were analyzed as two-tailed, and p < 0.05 was considered significant.

Results

The patients were categorized into three groups according to BMI (kg/m²) (< 25: 31 patients; ı 25–29.9: 38 patients; >30: 31 patients). There were no significant differences concerning age, gender, smoking history, and pack year among groups (p > 0.05) (Table 1). In addition, there was no detected significance in terms of comorbid conditions such as depression, gastroesophageal reflux, allergic rhinitis, diabetes mellitus, coronary artery disease, or additional comorbidities (the three patients had thalassemia, mitral valve prolapse, and epilepsy, separately for each group and some other comorbidities such as migraine, varicose veins, benign prostatic hypertrophy). We determined a significant difference regarding anxiety (8, 12, 15 of median value; respectively; for groups) (p = 0.035) and hypertension (7%, 21%, 35% of percentage; respectively; for groups) (p = 0.019) (there was a difference between groups 1 and 3 in both variables, respectively; p = 0.030, 0.013) (Table 2). Of all asthma cardinal symptoms, they were similar between groups except for dyspnea (p = 0.027, the adjusted difference between groups 1 and 2 (p = 0.029)) (Table 1). Pulmonary function tests were similar in both pre and post-bronchodilator and changes after bronchodilator (ml and %) between the groups (p > 0.05) (Table 3). Nine patients were treated in step 4; the remaining were treated in steps 3 and 2. When compared to groups in this respect, there was no significant difference between the groups (p = 0.200).

Table 1 Demographic data of the patients

Table 2 Comorbid conditions of the patients

Table 3 Pulmonary function tests of the patients

There were statistical differences between groups regarding IOS measurements except for R20, X5%, and X20%. The distribution of groups were 0.29 ± 0.15, 0.34 ± 0.21,0.44 ± 0.21 kPa/(L/s) for R5 (p = 0.007, difference between group 1 and 3); 103 (79;120), 110 (92;120), 131 (105;155) for R5% (p = 0.005, the difference between group 1 and 3); 94 (79;99), 87.5 (77;105), 100 (90;127) for R20% (p = 0.015, difference between group 2 and 3); −0.04 (−0.11;0), −0.07 (−0.12;−0.05) −0.10 kPa/(L/s) (−0.13;−0.06) for X5 (p = 0.010, difference between group 1,2 and 1,3); 0.16 (0.05;0.49), 0.59 (0.34;1.09), 0.75 (0.34;1.24) kPa/L for AX (p < 0.001, difference between group 1,2 and 1,3); 11.8 ± 6.1, 17.4 ± 5.6, 17.9 ± 7.5 Hz for F_res (p < 0.001, difference between group 1,2 and 1,3) and 19.0 (2.6;432), 44.4 (28.3;63.3), 40.5 (24.8;55.7) kPa/(L/s) for R5-R20 (p < 0.003, difference between group 1,2 and 1,3) (p values presented or all variables were adjusted). There were no differences between groups about R20, X5%, and X20%. Except for those, R20% differed between BMI < 30 and ≥ 30. The major deviations were observed between BMI < 25 and ≥ 25 for the other IOS parameters (Table 4).

Table 4 Impulse oscillometry measurements of the patients

When the correlation test between IOS parameters and BMI was performed, it was determined as p = 0.003 and p < 0.0001 for R5 and Fres (Pearson correlation coefficient; 0.252 and 0.295, respectively). The other IOS parameters were also significant except for two (for R5%, R20, R20%, R5-R20, X5, X5%, X20, X20%, and AX; p-value; <0001, 0.007, 0.007, 0.007, 0.020, 0.403, 0.002, 0.148, < 0.0001; respectively) (Kendall’s tau; 0.247, 0.188, 0.184, 0.183, −0.161, 0.057, −0.222, −0.099, 0.293; respectively). In addition, there was no significant difference between groups for FEF%25–75 in both ml and % (0.156, 0.834) (Fig. 1).

Fig. 1

The correlation test between IOS parameters and BMI

Discussion

In this study, it was determined that there were significant differences concerning R5, R5%, R20%, X5, X20, Fres, and R5-R20 among the three groups. In comparing the two groups, group 1 significantly differed about R5-R20, X5, AX, and Fres compared to groups 2 and 3. Based on these results, a difference was detected between BMI < 25 and ≥ 25 kg/m². Only one significant variable (R20%) between BMI < 30 and ≥ 30. Obesity and being overweight affected peripheral airways and reactance; on the other hand, obesity also affected central airways, based on our results. Remarkable differences were observed in terms of AX and Fres rather than R5-R20.

As for PFT, even though it is not specific to SAD, a study showed FEF25-75% as an indicator of SAD, and it was suggested that FEF25-75% may be more sensitive than FEV₁ for small and middle airway obstruction [11]. R5 and Fres are increased in peripheral airway obstruction for patients with asthma. R5-R20 is higher because R5 increases more than R20. X5 is more negative due to hyperinflation and decreased elasticity. AX increase is observed due to more negative X5 values and increased Fres [12]. In our study, PFTs of our patients did not differ among the three groups, including FEF25%−75%. Increased R5 and R5-R20 and normal R20 are shown in peripheral airway obstruction. Considering our results, the IOS results of the patients showed that obese patients had more peripheral airway obstruction than lean asthmatic patients, even though spirometry results of the patients were distributed homogenous.

There are several studies in this field on asthmatic children. A study on school-age children showed high R5, R5-20, and FEF25-75% at baseline measurement in obese asthmatics compared to lean asthmatics and obese children [13]. In a study by Assumpção et al., patients were divided into three groups (81 children, 30 in control, 21 in the overweight, and 30 in the obesity group). There was a difference between the control group and the obesity group for impedance, R5, Fres, and AX [14]. Another study detected that the median age of 75 participants was seven years old, and R5, Z5, and AX in obesity with asthma were statistically significantly higher than in asthma only. X5’s median was higher in obesity than in asthma and obesity with asthma group. In a comparison of asthma patients and obese patients with asthma, the authors found a significant difference regarding R5-R20 in obese patients with asthma compared to patients with asthma alone [15]. Even though the populations of the studies were different, our results were similar to those of the above studies. However, FEF25-75% was not different among groups in our study; in addition to R5, Fres, and AX, R5-R20 was also significantly different in our groups. Our results were highly similar to those of Sangsawang et al. ‘s study. In addition to the above variables, X5 was also a significant difference in our study. In comparison to all variables, overweight and obese asthmatic patients, some variables that demonstrated peripheral airway obstruction were higher than lean asthmatic patients in our study.

In a study that was performed on 188 patients, they were categorized into four groups according to BMI (kg/m²) (< 25 (control group), 25–30, > 30, and > 40). Demographic data and PFTs of the patients were homogeneous distribution among groups. The authors detected that obesity (BMI 30–39.9) and morbid obesity (BMI ≥ 40) were linked to a significantly worse R5-R20, X5, AX, and Fres compared with normal weight [16]. Similarly, demographic data and PFTs were similar in our study; besides R5-R20, X5, AX, and Fres, we determined that R20% was a significant difference. Conversely, a significant difference was not detected about R20 (p = 0.077). Although R20 increases in central airway obstruction, it also increases in peripheral airway obstruction; however, it does not increase so much as R5. In a study comparing morbid obesity and normal weight in terms of IOS measurements, R5, R20, R5-20, Z5, X5, and Fres were found to be increased [17]. Although our study was not included in morbid obese groups and was included in asthmatic patients, similar to the study, R20% was increased in our patients. Considering that obesity affects both peripheral and large airway functions [8], increasing R20 is not surprising; interestingly enough, the significant difference was between overweight and obese patients; there was no difference between lean and overweight or obese asthmatic patients.

In a cross-sectional study, it was examined whether obesity affects lung mechanics. In this context, 45 patients (22 females) were included in the study to examine BMI and other measurements. There were no differences between BMI < 25 and > 25 regarding IOS parameters and PFTs in males and females [18]. Compared to our results, although there was no significant difference between groups concerning PFTs, similar to the above study, there were notable deviations in IOS parameters between BMI < 25 and > 25 kg/m². The study consisted of a relatively small sample size compared to our study. Asthma might increase airway resistance and reactance in obese patients based on the above study.

The ATLANTIS study found that R5-20, AX, and X5 are all strongly related to small airway disease and have a high prevalence in asthma compared to other physiological measures thought to reflect the small airways [19]. The resonant frequency(Fres), the zero crossing of the reactance spectrum, and the reactance area (AX) are used as indirect markers of the lung periphery, especially for pre-post examinations and provocation tests [4]. In our study, despite indirect markers for peripheral airways, they were strongly significant between BMI > 25 and < 25.

It was detected between groups 1 and 3 regarding hypertension (7,21,35%, respectively) and anxiety (8,12,15 of median value, for groups, respectively). It is well-established that obesity is a significant risk factor for several diseases, such as anxiety [20]. Similarly, the association between obesity and hypertension has been known clearly for a long time [21]. In our study, the more the weight was increased, the more the diagnosis of hypertension was increased. However, there was only one patient diagnosed with CAD in group 3.

Limitations

The study did not include a control group consisting of obese patients without a concurrent diagnosis of asthma. Some comorbid conditions, such as hypercholesterolemia, were based on patients’ self-reporting. On the other hand, there have been few studies in the literature evaluating IOS parameters according to BMI.

Conclusion

The small airways might affect other factors, such as the severity of asthma, genetic features, socioeconomic situation, and obesity. According to our results, obesity and being overweight were detected to negatively affect the peripheral airways in adult asthma patients. Compared with lean asthmatic patients, significant changes related to R5-R20%, R5, Fres, and AX demonstrating small airway disease were worse in overweight and obese asthmatic patients. However, there were no differences concerning PFTs between groups. Therefore, it was thought that IOS may detect small airway changes earlier than PFTs, especially in obese individuals.