Skip to main content
NCBI home page
ERJ Open Research logoLink to ERJ Open Research
. 2025 Aug 4;11(4):00736-2024. doi: 10.1183/23120541.00736-2024

Automated computed tomographic analysis of bronchial thickness and mucus plugs in bronchiectasis with asthma

Tjeerd van der Veer 1,2,✉, Eleni-Rosalina Andrinopoulou 3,4, Punitkumar Makani 5, Gert-Jan Braunstahl 1,6, Harm AWM Tiddens 5,7
PMCID: PMC12320113  PMID: 40761656

Abstract

Background

Bronchiectasis disease is characterised by cough, sputum and exacerbations, with chest computed tomography (CT) typically showing bronchial wall thickening and mucus plugging in addition to bronchial dilation. Asthma is a common comorbidity and associated with increased, eosinophilic, airway inflammation. Automated measurements of bronchial wall thickening and mucus plugs may serve as biomarkers for inflammation and are associated with clinical characteristics such as spirometry, blood eosinophil counts and disease severity in patients with bronchiectasis and asthma co-diagnosis.

Methods

In a cross-sectional retrospective cohort of 64 patients with bronchiectasis disease and asthma, we applied automated image analysis to assess bronchial dimensions and mucus plug metrics on chest CT scans. These metrics were correlated with spirometry, blood eosinophil counts as well as FACED and Bronchiectasis Severity Index (BSI) scores using correlations and multiple regression analyses.

Results

In 63 patients, bronchial wall thickness and mucus plugs were quantified. Negative correlations were observed between bronchial wall thickness markers and spirometry (bronchial wall thickness/accompanying artery diameter and forced expiratory volume in 1 s (FEV1), r= −0.37; FEV1/forced vital capacity, r= −0.30). Mucus plugs correlated negatively with spirometry and positively with FACED and BSI scores (number of mucus plugs and BSI, r=0.45). Correlations with blood eosinophil counts were very weak. In multiple regression analyses, independent associations were observed for FEV1, Pseudomonas aeruginosa and frequent exacerbations.

Conclusion

This study identified key relationships between automated measurements of bronchial wall thickness and mucus plugs and clinical characteristics, highlighting their potential as imaging biomarkers to enhance phenotyping, improve risk assessment and facilitate tailored treatment strategies in bronchiectasis.

Shareable abstract

Automated CT analysis of bronchiectasis disease patients with asthma showed that bronchial wall thickening and mucus plugs were linked to decreased lung function, frequent exacerbations and higher bronchiectasis severity, but not blood eosinophil counts https://bit.ly/4hfcPld

Introduction

Bronchiectasis disease is characterised by cough, sputum production and recurrent exacerbations, together with the radiological appearance of abnormal dilatation of bronchi [1]. Chronic airway inflammation is an important driving factor for clinical symptoms and outcomes in these patients, with airway remodelling and altered airway structure as consequences of the inflammatory process [2, 3]. On chest computed tomography (CT), changes such as wall thickening and mucus plugging are indicative of ongoing inflammation and are highly prevalent in bronchiectasis disease patients [4]. In the clinic, these markers have been linked to exacerbations and can potentially be used to differentiate between inflammatory “active” or “inactive” phenotypes [5].

A promising development is the emergence of artificial intelligence (AI)-based automated tools for the precise quantification of airway dimensions on chest CT scans, which may be used for in vivo monitoring of disease activity and progression. Bronchial wall thickening can reduce the airway inner diameter (lumen) and decrease airway compliance, which are key determinants of airflow obstruction. Therefore, airway dimensions are an important link between radiology and lung function. Traditionally, visual radiological evaluation of CT scans in bronchiectasis disease prioritises those airways labelled as bronchiectasis, which have already undergone permanent significant widening and thickening. A more sensitive automatic assessment of bronchial wall thickening and mucus plugging could detect the consequences of airway inflammation earlier, potentially allowing for earlier anti-inflammatory interventions before irreversible widening occurs [3].

Automated assessment of bronchus–artery (BA) pair dimensions has been shown to be sensitive for detecting bronchial wall thickening and bronchial widening in cystic fibrosis (CF) [6, 7], non-CF bronchiectasis and COPD [8, 9], and paediatric asthma [10]. Automated assessment of mucus plugs has recently also been applied on large real-life cohorts of COPD patients as well as non-CF bronchiectasis and CF [11–13].

Additional developments in bronchiectasis disease have focused on identifying inflammatory endotypes. While neutrophilic inflammation is most prevalent, eosinophils have also demonstrated elastolytic capacity and the secretion of substances that can damage the bronchial wall such as eosinophil cationic protein and metalloproteases [14]. In addition, eosinophils have been implicated in the formation of mucus plugs [15]. An eosinophilic endotype was identified in 22.6% of a large cohort of bronchiectasis patients, based on blood eosinophil counts ≥0.3×109 L−1 [16].

As known, particularly in severe asthma, ∼80% of patients have an eosinophilic endotype [17]. Bronchiectasis disease has a high co-occurrence of asthma, as well as COPD, sometimes also called overlap syndromes, which have been associated with higher inflammation and worse outcomes [18–21]. Interestingly, it was shown in the European Multicentre Bronchiectasis Audit and Research Collaboration (EMBARC) cohort that 31% of bronchiectasis disease patients have a co-diagnosis of asthma [22]. Inversely, radiological bronchiectasis is reported as a common comorbidity in asthma, with a significantly higher prevalence in severe eosinophilic asthma compared with mild asthma [23]. Asthma is characterised by bronchial hyperreactivity and chronic inflammation of both the larger and small airways [24].

To investigate the relationship between radiological markers of bronchial wall thickening and mucus plug formation and clinical characteristics such as lung function and blood eosinophil counts, we applied automated image analysis to chest CT scans of a cross-sectional cohort of bronchiectasis disease patients with co-diagnosed asthma. By utilising CT features as biomarkers, the goal is to better understand and phenotype these patients, and tailor anti-eosinophilic and mucolytic treatments to the patients most likely to benefit.

Out hypothesis was that bronchial wall thickening and mucus plugging, as assessed by CT and automated image analysis, are related to clinical characteristics, lung function and blood eosinophil counts in patients with bronchiectasis disease and an asthma co-diagnosis.

We addressed the following research questions:

  • What is the relationship between bronchial wall thickness and mucus plugs (as measured by automated tools on chest CT scans) and spirometry?

  • What is the relationship between bronchial wall thickness and mucus plugs and blood eosinophil counts?

  • How are specific clinical factors (e.g. age, dyspnoea, Pseudomonas colonisation, FACED and Bronchiectasis Severity Index (BSI) scores, and exacerbation frequency) associated with bronchial wall thickness and mucus plugs?

Methods

BASIIS (Bronchiectasis & Asthma Identification and Inflammatory phenotyping Study) is a retrospective cross-sectional cohort using clinical data from 2018 to 2021 from the Erasmus MC University Medical Centre (Rotterdam, The Netherlands) bronchiectasis clinic. Inclusion criteria were ≥18 years, bronchiectasis (based on International Statistical Classification of Diseases and Related Health Problems, 10th Revision code J47 in July 2020) and a diagnosis of asthma in the clinical notes. After screening 255 bronchiectasis patients, 78 with asthma co-diagnosis were selected. Exclusion criteria were no mention of bronchiectasis in the chest CT report, unsuitable chest CT scan (>5 years old, slice thickness >1.5 mm or slice gaps), negative bronchial provocation test and incomplete records. After review, 64 patients were included in this study (figure 1). Patient data were collected and coded for pseudo-anonymity after institutional review board approval (MEC-2020-0061).

FIGURE 1.

FIGURE 1

Open in a new tab

Inclusions flowchart showing the number of participants and reasons for exclusion. ICD: International Statistical Classification of Diseases and Related Health Problems, 10th Revision; CT: computed tomography; BA: bronchus–artery; MP: mucus plug.

Data collection

Data were collected by the first author (T.v.d.V.) using the OpenClinica electronic data capture platform (OpenClinica, Waltham, MA, USA). Spirometry followed American Thoracic Society/European Respiratory Society standards [25]. Height, weight and body mass index (BMI), smoking history, number of exacerbations and hospitalisations, modified Medical Research Council (mMRC) dyspnoea score, and medication use were derived from clinical records and prescription data. Laboratory results (blood eosinophil counts) were collected from digital patient records both closest to the CT scan date and if the patient had historical eosinophil measurements ≥0.3×109 L−1 (yes/no). Bronchiectasis aetiology was determined by the treating pulmonologist after a workup conducted according to Dutch bronchiectasis guidelines [26]. Bacterial colonisation status was retrieved from sputum culture results. All extracted data were coded to ensure blinding. Missing data in the records were marked as missing in the database.

FACED and BSI

We assessed bronchiectasis severity using FACED and BSI scores. The FACED score combines forced expiratory volume in 1 s (FEV1) % predicted, age, colonisation by Pseudomonas aeruginosa, radiological extent of bronchiectasis and mMRC dyspnoea scale, classifying patients into three categories of mortality risk: mild (0–2 points), moderate (3–4 points) and severe (5–7 points) [27]. The BSI includes the additional variables of BMI, number of hospitalisations and exacerbations in the past year, and bacterial colonisation status, to predict hospitalisation and mortality rates. BSI categories: mild (0–4 points), moderate (5–8 points) and severe (≥9 points) [28].

Automated image analysis

Chest CT scans of the participants were retrieved from the hospital radiology system and pseudo-anonymised. The scans underwent pre-processing to identify the optimal inspiratory CT scan reconstruction for each participant based on following criteria: a minimum of 150 scan slices, slice thickness ≤1.5 mm and no slice gap. The selected CT scans were then further analysed to determine BA ratios and the presence of mucus plugs using the fully automated AI-based software LungQ-BA v.2.0.1 and LungQ-MP v.3.0.1 (Thirona, Nijmegen, The Netherlands).

LungQ-BA algorithm steps (see also figure 2):

  • 1) Bronchial tree segmentation.

  • 2) Identification of detected bronchial branches.

  • 3) Matching the adjacent artery for each detected bronchial branch (BA pairs).

  • 4) Identification of the generation (G) number for each BA pair starting from the segmental bronchi (G0).

  • 5) Computation of BA dimensions: bronchial inner diameter (Bin), bronchial wall thickness (Bwt), bronchial wall area (Bwa), bronchial outer area (Boa) and accompanying artery diameter (A).

  • 6) Computation of BA ratios: Bin/A, Bwt/A and Bwa/Boa.

FIGURE 2.

FIGURE 2

Open in a new tab

Bronchus–artery measures: bronchial and arterial dimensions and ratios as automatically measured by LungQ-BA.

Airway measurements were performed for airways proximal to an occluding mucus plug, so that dimensions reflect unobstructed bronchi.

LungQ also determines Pi10, a computed measure that represents the square root of the wall area of a hypothetical airway with an internal perimeter of 10 mm [29].

As a measure of bronchial lumen, we used Bin/A. As measures of wall thickness, we used Bwt/A and Bwa/Boa. Bwa/Boa is independent of arterial diameter, which may be altered by pathological changes in the pulmonary vasculature.

BA dimension results are presented from subsegmental level onward (G1–14), where G1 represents the subsegmental bronchi, G2 represents the subsubsegmental bronchi, etc. For statistical analysis, we used the median BA measurements of G1–6 because these bronchial generations included the highest number of BA measurements in most participants and are less affected by body size, inspiration level and the relatively higher visibility of small airways affected by airway disease [7, 9]. On the request of the reviewer, we conducted an additional analysis focusing on BA measurements in G1–3 to address a concern regarding the reliability of measurements in higher generations. This analysis confirmed that the main findings of the primary analysis were consistent with proximal-generation airways (G1–3) (supplementary material).

LungQ-MP also uses AI-based algorithms to detect mucus plugs throughout the lung and has now been used in multiple real-life cohorts of patients with COPD, CF and bronchiectasis [11–13]. The detection algorithm, trained on expert annotations, identifies full mucus obstructions with clear proximal and distal airways, providing both location and volumetric assessments. The segmentation combines seed-based and voxel-based methods, providing detection and quantification of mucus plugs along the entire bronchi, including the peripheries. The algorithm provides output as the number of detected mucus plugs and their segmental locations and volumes (mL).

Variables of interest

  • Bronchial wall thickness: Bwt/A, Bwa/Boa and Pi10.

  • Bronchial inner diameter: Bin/A.

  • Mucus plugs: total number of detected mucus plugs in the bronchial tree and total mucus plug volume (mL).

  • Spirometry: FEV1 % predicted and FEV1/FVC ratio.

  • Blood eosinophil counts (×109 L−1) and historical measurements ≥0.3×109 L−1.

  • Patient factors: age, BMI, mMRC dyspnoea score, bacterial colonisation (P. aeruginosa and other pathogens), number of exacerbations and hospitalisations, and FACED and BSI scores.

Primary research questions

  • Relationship between:
    •  • BA measurements of bronchial wall thickness (Bwt/A, Bwa/Boa and Bin/A) and spirometry (FEV1 % predicted and FEV1/FVC).
    •  • Mucus plug measurements (total number and volume) and spirometry.
  • Relationship between:
    •  • BA measurements of bronchial wall thickness (Bwt/A and Bwa/Boa) and blood eosinophils (number of cells ×109 L−1 and binary for historical value ≥0.3×109 L−1).
    •  • Measurements of mucus plugs (total number and volume) and blood eosinophils.

Secondary research questions

  • Relationship between automated measurements of Pi10 and Bin/A and spirometry and eosinophil counts.

  • Relationship between mucus plug number and volume and bronchiectasis severity models (BSI and FACED).

  • Relationship between clinical characteristics (age, dyspnoea, exacerbation frequency, eosinophil counts, Pseudomonas colonisation and FEV1 % predicted) and BA measurements of bronchial wall thickness and mucus plugs.

Statistical analysis

Data are presented as median and interquartile range (IQR) or as mean and standard deviation, depending on the data distribution.

Bwt/A, Bwa/Boa, Pi10 and Bin/A were selected as the most relevant measures for the connection with spirometry outcomes, as bronchial internal diameter of the airways and bronchial wall area are likely to determine maximal flows for a spirometry manoeuvre. For the spirometry outcomes, FEV1 % predicted and FEV1/FVC were selected because these outcomes are considered dependent on airway resistance and airway dimensions. The relationship between the total number of mucus plugs, total mucus volume and spirometry was investigated. Correlations were assessed using Spearman's, Pearson's or point biserial coefficients and confidence intervals, depending on data skewness. A correlation coefficient <0.2 was rated as very weak, 0.2–0.4 as weak, 0.4–0.6 as moderate, 0.6–0.8 as strong and 0.8–1 as excellent [30].

Multiple regression analyses were performed using the clinical characteristics of age, mMRC dyspnoea score, number of exacerbations in the preceding year, blood eosinophil counts (×109 L−1), P. aeruginosa colonisation and FEV1 % predicted as independent variables, and measures of bronchial wall thickening (Bwt/A and Bwa/Boa) and mucus plugs (total number and volume) as dependent variables.

Age, blood eosinophil counts and FEV1 were taken as continuous variables, P. aeruginosa as a binary variable, and mMRC dyspnoea score and exacerbations per year as categorical variables using 0, 1 and ≥2, with reference 0. The PP probability plots were checked for normality of the residuals. Initial analyses included all participants; however, residuals suggested deviation from normality. Outlier analysis was conducted based on standard deviation (≥3sd) and Cook's distance (≥0.5). Statistical analysis was performed using SPSS Statistics version 28.0.1.0 (IBM, Armonk, NY, USA), with significance defined as p<0.05, without corrections for multiple testing.

Results

LungQ successfully measured BA pairs and mucus plugs in 63 and 59 participants, respectively. Of the 63 participants, 35 were women (55.6%), mean±sd age 60.8±17.2 years. Most participants were Caucasian (76.2%) and 34.9% had a history of smoking (no current smokers). All participants were prescribed inhaled corticosteroids; no participants were prescribed maintenance oral corticosteroids. In 63 participants, LungQ measured 12 528 BA pairs, of which G1–6 accounted for 93.7% (11 709/12 528). For complete participant characteristics, see table 1. For a depiction of median Bin/A, Bwt/A and Bwa/Boa ratios for bronchial G1–6, see supplementary figure S1a–c, as well as a bar chart of participants sorted by number of mucus plugs in supplementary figure S2. See supplementary table S1 for the CT scanner manufacturers, kernels and slice thicknesses of the included scans.

TABLE 1.

Participant characteristics (n=63)

Clinical characteristics
 Gender (female) 35 (55.6)
 Age (years) 60.8±17.2
 BMI (kg·m−2) 23.77±4.56
 Ethnicity (Caucasian, binary) 48 (76.2)
 Former smoking (binary) 22 (34.9)
 Bronchiectasis aetiology
  ABPA 1 (1.6)
  Asthma 10 (15.9)
  Connective tissue disease 2 (3.2)
  GORD/aspiration 1 (1.6)
  Idiopathic 10 (15.9)
  Immunodeficiency 8 (12.7)
  Non-tuberculous mycobacteria 2 (3.2)
  PCD 2 (3.2)
  Post-infectious 8 (12.7)
  More than one aetiological factor/other 19 (30.2)
 BSI score 6.0 (3.0–8.0)
  0–4 points: mild bronchiectasis 24 (38.1)
  5–8 points: moderate bronchiectasis 24 (38.1)
  ≥9 points: severe bronchiectasis 15 (23.8)
 FACED score 2.0 (1.0–3.0)
  0–2 points: mild bronchiectasis 42 (66.7)
  3–4 points: moderate bronchiectasis 18 (28.6)
  5–7 points: severe bronchiectasis 3 (4.8)
 Number of exacerbations per year 1.0 (0.0–1.0)
  0 25 (39.7)
  1 23 (36.5)
  ≥2 (“frequent exacerbator”) 15 (23.8)
 mMRC dyspnoea score 0.0 (0.0–1.0)
  0: only with strenuous exercise 32 (50.8)
  1: when hurrying or walking up a slight hill 16 (25.4)
  ≥2: walks slower than same age, must stop for breath or worse 15 (23.8)
 Blood eosinophil count (×109 L−1) (continuous) 0.26±0.32
 Eosinophils ever ≥0.3×109 L−1 (binary) 29 (46.0)
 Pseudomonas colonisation (binary) 19 (30.2)
 Other pathogen colonisation (binary) 27 (42.9)
 Use of azithromycin (binary) 35 (55.6)
 FEV1 % predicted 71.30±21.75
 FEV1/FVC 0.64±0.13
Radiological characteristics
 Bwt/A (G1–6) 0.16±0.1
 Bwa/Boa (G1–6) 0.43±0.15
 Bin/A (G1–6) 0.96±0.39
 Pi10 (mm2) 2.47±0.47
 Total number of mucus plugs 9.0 (2.0–20.5)
 Total mucus volume (mL) 0.44 (0.11–1.67)
 Total number of BA pairs (all generations)/mean per participant 12 528/199
  G1 1380/22
  G2 2643/42
  G3 3054/48
  G4 2328/37
  G5 1493/24
  G6 811/15
Open in a new tab

Data are presented as n (%), mean±sd, median (interquartile range) or n/n. BMI: body mass index; ABPA: allergic bronchopulmonary aspergillosis; GORD: gastro-oesophageal reflux disease; PCD: primary ciliary dyskinesia; BSI: Bronchiectasis Severity Index; FACED: forced expiratory volume in 1 s (FEV1) % predicted, age, colonisation by Pseudomonas aeruginosa, radiological extent of bronchiectasis and modified Medical Research Council (mMRC) dyspnoea scale; FVC: forced vital capacity; Bwt: bronchial wall thickness; A: accompanying artery diameter; G1–6: generation 1–6; Bwa: bronchial wall area; Boa: bronchial outer area; Bin: bronchial inner diameter; Pi10: square root of the wall area of a hypothetical airway with an internal perimeter of 10 mm; BA: bronchus–artery.

Relationship between bronchial wall thickness and mucus plugs and lung function

Correlations between bronchial parameters and lung function tests were observed (table 2). Bwt/A and Bwa/Boa demonstrated weak and moderate negative correlations with FEV1 % predicted and FEV1/FVC, indicating that increases in these bronchial wall parameters are associated with lower lung function. Similarly, Pi10 showed moderate negative correlations with both FEV1 % predicted and FEV1/FVC. In contrast, Bin/A showed moderate positive correlation. The total number of mucus plugs and total mucus plug volume also showed moderate to strong negative correlations with both FEV1 % predicted and FEV1/FVC. Correlation plots for all investigated relationships are shown in supplementary figure S3a–c.

TABLE 2.

Correlations between radiological measures and lung function

FEV1 % predicted FEV1/FVC
Bwt/A G1–6 −0.366 (−0.59– −0.09) −0.297 (−0.54– −0.02)
Bwa/Boa G1–6 −0.467 (−0.67– −0.23) −0.452 (−0.66– −0.22)
Pi10 −0.501 (−0.70– −0.25) −0.502 (−0.70– −0.27)
Bin/A G1–6 0.331 (0.10–0.53) 0.379 (0.15–0.58)
Total number of mucus plugs −0.394 (−0.63– −0.11) −0.602 (−0.73– −0.40)
Total mucus plug volume −0.333 (−0.56– −0.07) −0.547 (−0.70– −0.33)
Open in a new tab

Data are presented as Spearman's correlation (ρ); 95% confidence intervals are indicated. FEV1: forced expiatory volume in 1 s; FVC: forced vital capacity; Bwt: bronchial wall thickness; A: accompanying artery diameter; G1–6: generation 1–6; Bwa: bronchial wall area; Boa: bronchial outer area; Pi10: square root of the wall area of a hypothetical airway with an internal perimeter of 10 mm; Bin: bronchial inner diameter.

Relationship between bronchial wall thickness and mucus plugs and blood eosinophil counts

Correlation coefficients revealed very weak associations between bronchial dimension parameters Bwt/A, Bwa/Boa, Pi10 and Bin/A and both continuous and binary eosinophil measures. The total number of mucus plugs and total mucus plug volume were very weakly correlated with eosinophil measures (table 3).

TABLE 3.

Correlations between radiological measures and eosinophil counts

Continuous eosinophils Eosinophils ever ≥0.3×109 L−1
Bwt/A G1–6 −0.039 (−0.33–0.23) 0.079 (−0.23–0.29)
Bwa/Boa G1–6 0.072 (−0.22–0.34) 0.142 (−0.07–0.41)
Pi10 0.085 (−0.22–0.36) 0.127 (−0.13–0.39)
Bin/A G1–6 −0.068 (−0.34–0.21) −0.177 (−0.39–0.12)
Total number of mucus plugs 0.094 (−0.15–0.35) −0.065 (−0.29–0.22)
Total mucus plug volume 0.093 (−0.19–0.37) −0.071 (−0.36–0.16)
Open in a new tab

Data are presented as Spearman's correlation (ρ) for continuous eosinophils or point biserial correlation (r) for eosinophils ever ≥0.3×109 L−1; 95% confidence intervals are indicated. Bwt: bronchial wall thickness; A: accompanying artery diameter; G1–6: generation 1–6; Bwa: bronchial wall area; Boa: bronchial outer area; Pi10: square root of the wall area of a hypothetical airway with an internal perimeter of 10 mm; Bin: bronchial inner diameter.

Relationship between patient factors and bronchial wall thickness and mucus plug parameters

Moderate positive correlations were observed between both BSI and FACED scores and total mucus volume and number of mucus plugs (table 4 and figures 3 and 4; also see supplementary figure S4a and b).

TABLE 4.

Correlations between radiological measures and bronchiectasis severity scores

BSI FACED
Bwt/A G1–6 0.334 (0.10–0.54) 0.329 (0.09–0.52)
Bwa/Boa G1–6 0.151 (−0.09–0.39) 0.205 (−0.02–0.42)
Pi10 0.175 (−0.10–0.42) 0.217 (−0.03–0.45)
Bin/A G1–6 0.022 (−0.23–0.26) −0.022 (−0.28–0.21)
Total number of mucus plugs 0.446 (0.219–0.672) 0.441 (0.217–0.632)
Total mucus plug volume 0.494 (0.274–0.714) 0.473 (0.253–0.647)
Open in a new tab

Data are presented as Spearman's correlation (ρ); 95% confidence intervals are indicated. BSI: bronchiectasis severity index; FACED: forced expiratory volume in 1 s % predicted, age, colonisation by Pseudomonas aeruginosa, radiological extent of bronchiectasis and modified Medical Research Council dyspnoea scale; Bwt: bronchial wall thickness; A: accompanying artery diameter; G1–6: generation 1–6; Bwa: bronchial wall area; Boa: bronchial outer area; Pi10: square root of the wall area of a hypothetical airway with an internal perimeter of 10 mm; Bin: bronchial inner diameter.

FIGURE 3.

FIGURE 3

Open in a new tab

a) Mucus plugs and b) mucus volume by Bronchiectasis Severity Index score category: mild, moderate and severe. Box plots show the median number of plugs or median total mucus volume, interquartile range (IQR) (box), smallest and largest values within 1.5×IQR (whiskers), and outliers.

FIGURE 4.

FIGURE 4

Open in a new tab

Three-dimensional (3D) renderings of mucus plugs in participants with varying degrees of bronchiectasis severity: representative examples of mucus plugs in participants with a) mild, b) moderate and c) severe bronchiectasis, as categorised by the Bronchiectasis Severity Index (BSI). The 3D renderings show the bronchial tree, with mucus plugs highlighted in red. These examples represent participants with mucus plug counts near the median for their respective BSI category.

In multiple regression models including the clinical characteristics of age, mMRC dyspnoea score, blood eosinophil counts, number of exacerbations per year, P. aeruginosa colonisation and FEV1 % predicted, a consistent negative association was found between measures of bronchial wall thickness (Bwt/A and Bwa/Boa) and FEV1 % predicted, a positive association with Pseudomonas colonisation, and a positive association with ≥2 exacerbations per year (“frequent exacerbators”). The same associations were observed for both the total number of mucus plugs and total mucus volume.

Analysis with the full cohort revealed non-normal residuals in PP plots for the mucus models, to the influence of one outlier on the number and volume of mucus plugs at +5sd and Cook's distance 0.53. Upon removal of this outlier, the residuals’ normality improved, and the strength of the associations for FEV1, Pseudomonas and exacerbations increased. In addition, an association for mMRC ≥2 was seen. The comparative results of all models are shown in table 5 and supplementary table S2. The comparative analysis of BA measures in bronchial G1–3 is shown in supplementary table S3a–e.

TABLE 5.

Multiple regression results for the full cohort showing associations between independent participant characteristics as predictor variables for dependent variables of bronchus–artery measures of bronchial wall thickness and mucus plug number and volume

Model:
Bwt/A G1–6 		Model:
Bwa/BoaG1–6 		Model:
Number of plugs		 Model:
Mucus volume
Coefficient p-value Coefficient p-value Coefficient p-value Coefficient p-value
Constant 0.384 <0.001 0.564 <0.001 38.315 0.054 2.147 0.143
Age 0.000 0.799 −0.001 0.279 0.178 0.433 0.014 0.403
Eosinophils −0.012 0.664 0.023 0.499 −2.412 0.835 −0.155 0.857
FEV1 % predicted −0.002 <0.001 −0.002 0.001 −0.526 0.006 −0.034 0.017
Pseudomonas 0.056 0.006 0.066 0.008 22.519 0.007 1.794 0.004
Exacerbations per year 1 versus 0 −0.004 0.850 < −0.0001 0.998 0.559 0.947 0.025 0.968
Exacerbations per year ≥2 versus 0 0.056 0.022 0.045 0.129 21.417 0.026 1.796 0.012
mMRC 1 versus 0 −0.004 0.856 −0.011 0.699 −8.016 0.375 −0.431 0.519
mMRC ≥2 versus 0 −0.034 0.184 −0.019 0.542 −8.155 0.426 −1.146 0.847
Open in a new tab

Bwt: bronchial wall thickness; A: accompanying artery diameter; G1–6: generation 1–6; Bwa: bronchial wall area; Boa: bronchial outer area; FEV1: forced expiratory volume in 1 s; mMRC: modified Medical Research Council. Bold indicates statistically significant at p<0.05.

Discussion

This study aimed to investigate the relationships between automatic image analysis metrics for bronchial wall thickness, mucus plugs and clinical parameters in a cohort of bronchiectasis disease patients with an asthma co-diagnosis. The findings indicate that thicker bronchial walls and greater mucus plug numbers and volumes are associated with reduced FEV1 % predicted and FEV1/FVC ratios. This negative association underscores the impact of structural changes of the bronchial wall and mucus obstruction, as assessed by automated CT analysis, on spirometry metrics in vivo, supporting previous findings on the significance of bronchial wall thickening in asthma, bronchiectasis and COPD [31–33].

The airway lumen dimension, measured as Bin/A, showed a weak positive correlation with lung function, whereas bronchial wall thickness measured as Bwa/Boa and Pi10 showed moderate associations. The airway lumen size varies with thoracic volume during imaging, making it less consistent as a functional indicator. In contrast, increased airway wall thickness is linked to greater stiffness, reduced compliance and reactivity [34]. These findings suggest that these bronchial wall measurements are more reliable markers for linking structure to lung function.

Absent correlations between blood eosinophil counts and bronchial wall thickness or mucus plugs suggest a limited role for eosinophil-driven inflammation in this cohort. Previous research links increased bronchial wall area to higher sputum eosinophil counts and reduced lung function in eosinophilic asthma patients [29]. Mucus plugging is also a common phenomenon in asthma, associated with eosinophils and other type 2 inflammatory features, as well as changes in airflow obstruction [15, 35]. However, in this cohort, the prescription of inhaled corticosteroids in all participants may have attenuated any eosinophilic inflammation. Second, the recruitment of participants from a specialised bronchiectasis clinic suggests a predominance of neutrophilic inflammation or mixed inflammatory pathways. Third, the instability of single blood eosinophil measurements in bronchiectasis may influence these results [36]. Categorising patients based on historical blood eosinophil levels ≥0.3×109 L−1 to indicate an eosinophilic endotype did not reveal an important relationship either [16]. Future studies may use other ways of assessing the presence of eosinophilic inflammation such as longitudinal blood measurements, eosinophil/neutrophil ratios or sputum lateral flow assays [37].

Positive correlations between BSI and FACED scores with mucus plug numbers and volumes support the role of mucus metrics as biomarkers of underlying inflammation and disease severity. According to the “vicious vortex” hypothesis, mucus production and impaired mucociliary clearance act both as a result of and as a contributor to disease progression [2]. Mucus plugs have been associated with all-cause mortality in COPD, underscoring their high clinical relevance [38]. Future studies may also position CT mucus analyses for risk assessment in bronchiectasis disease, potentially offering an efficient alternative to composite risk scores.

BA and Pi10 analyses showed similar correlations with lung function, indicating their comparable value as markers for bronchial wall thickening. Bwa/Boa and Bwt/A ratios may directly reflect airway remodelling across various airway sizes and can also be used for sectional bronchial tree analysis, while Pi10 provides a standardised measure for broader comparisons across patient populations. However, Bwt/A may be influenced by arterial size variations, potentially not reflecting airway changes alone.

In multiple regression analysis, bronchial wall thickness was negatively associated with FEV1 % predicted and positively with Pseudomonas colonisation, supporting the pathogen's detrimental role in exacerbations and disease progression [39, 40]. These results align with the models for mucus plug numbers and volumes, showing similar associations with Pseudomonas and FEV1, and an independent relationship with frequent exacerbations. The linkage between radiological extent of airway changes, Pseudomonas colonisation and severe exacerbations was recently also demonstrated in a large radiological analysis of 524 EMBARC patients [4].

Quantitative CT imaging provides detailed, reproducible data on airway dimensions and mucus plugs, enhancing our understanding of disease mechanisms. These metrics link structural changes to clinical outcomes such as exacerbation frequency and lower lung function [8, 33]. Clinically, quantitative CT analysis may aid in risk stratification and differentiate inflammatory phenotypes to identify patients who may benefit from various treatment options. Moreover, the incorporation of automated quantitative CT analysis meets the widely recognised need for non-invasive and time-efficient biomarkers in bronchiectasis [5]. In this analysis, multiple metrics of bronchial wall thickness and mucus plugs were investigated and each showed important associations with clinical characteristics. The relative value of each metric will be further evaluated in future studies. Also, the total number of mucus plugs has been investigated in previous studies, while the total mucus volume remains largely unexplored. Mucus volume may be particularly valuable in longitudinal studies where mucus plug presence and/or persistence can be weighed against an increase or decrease in volume. Although the algorithm performs highly accurately, occasional underestimation of mucus plugs may occur, similar to visual assessments. This potential underestimation is primarily due to limitations in detecting smaller or peripheral plugs.

A key limitation of our study is its retrospective design and relatively small sample size, limiting statistical power and the number of independent patient characteristics in our regression models. Additionally, the cross-sectional nature of the study prevents investigation of temporal relationships. Prospective longitudinal studies with larger cohorts and control groups of bronchiectasis without asthma, as well as asthma without bronchiectasis, are needed to validate our findings on the impact of bronchial wall thickness and mucus plugs and to further explore the role of eosinophils in bronchiectasis–asthma overlap.

Conclusion

This study observed important relationships between, on the one hand, automated measurements of bronchial wall thickness and mucus plugs and, on the other hand, spirometry metrics and bronchiectasis severity scores, as well as independent associations for FEV1, P. aeruginosa and frequent exacerbations. No clear relationship with blood eosinophil counts was observed in this cohort of patients with bronchiectasis disease and asthma co-diagnosis. These findings illustrate the high potential of quantitative imaging biomarkers to enhance clinical studies, improve risk assessment, enable phenotyping and facilitate the development of tailored treatment strategies for bronchiectasis disease patients.

Acknowledgements

We would like to thank the LungAnalysis – Image Analysis Core Laboratory at the Department of Paediatric Pulmonology and Allergology, Erasmus MC (Rotterdam, The Netherlands) for their invaluable support and resources throughout this research project. We also extend our gratitude to M.M. van der Eerden (Erasmus MC) for his initial contributions and support in the first stage of this project.

Footnotes

Provenance: Submitted article, peer reviewed.

Ethics statements: Local institutional review board approval was obtained for anonymised use of clinical data (MEC-2020-0061).

Conflict of interest: T. van der Veer reports no conflicts of interest involving the work under consideration for publication, no relevant financial activities outside the submitted work and no other relationships or activities that readers could perceive to have influenced, or that give the appearance of potentially influencing, the current manuscript. E-R. Andrinopoulou reports no conflicts of interest and no relevant financial activities or other relevant relationships or activities regarding the work under consideration. P. Makani reports no conflicts of interest and no relevant financial activities or other relevant relationships or activities regarding the work under consideration. G-J. Braunstahl reports honoraria for lectures and consultancy from GSK, AstraZeneca, Novartis and Sanofi Genzyme, as well as research grants from Sanofi Genzyme, GSK and AstraZeneca, not related to the work under consideration. H.A.W.M. Tiddens received, in the last 3 years, multiple grants from the following public and institutional grant institutions for lung structure and function research: NHMRC, NIH, CFF, ECFS, IMI and Erasmus MC Sophia Foundation; received unconditional grants for investigator-initiated research from Novartis and Insmed; acted as a consultant for Insmed, Thirona, Neupharma and Boehringer Ingelheim; and has been chief medical officer for Thirona since April 2022, and vice-chair and faculty for the ADVANCE course sponsored by Vertex.

Supplementary material

Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.

Supplementary material

00736-2024.SUPPLEMENT.pdf (1.5MB, pdf)
DOI: 10.1183/23120541.00736-2024.Supp1
Open in a new tab

00736-2024.SUPPLEMENT

References

  • 1.Aliberti S, Goeminne PC, O'Donnell AE, et al. Criteria and definitions for the radiological and clinical diagnosis of bronchiectasis in adults for use in clinical trials: international consensus recommendations. Lancet Respir Med 2022; 10: 298–306. doi: 10.1016/S2213-2600(21)00277-0 [DOI] [PubMed] [Google Scholar]
  • 2.Flume PA, Chalmers JD, Olivier KN. Advances in bronchiectasis: endotyping, genetics, microbiome, and disease heterogeneity. Lancet 2018; 392: 880–890. doi: 10.1016/S0140-6736(18)31767-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Long MB, Chotirmall SH, Shteinberg M, et al. Rethinking bronchiectasis as an inflammatory disease. Lancet Respir Med 2024; 12: 901–914. doi: 10.1016/S2213-2600(24)00176-0 [DOI] [PubMed] [Google Scholar]
  • 4.Pieters A, van der Veer T, Meerburg JJ, et al. Structural lung disease and clinical phenotype in bronchiectasis patients: the EMBARC CT study. Am J Respir Crit Care Med 2024; 210: 87–96. doi: 10.1164/rccm.202311-2109OC [DOI] [PubMed] [Google Scholar]
  • 5.Johnson E, Long MB, Chalmers JD. Biomarkers in bronchiectasis. Eur Respir Rev 2024; 33: 230234. doi: 10.1183/16000617.0234-2023 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Chen Y, Lv Q, Andrinopoulou ER, et al. Automatic bronchus and artery analysis on chest computed tomography to evaluate the effect of inhaled hypertonic saline in children aged 3–6 years with cystic fibrosis in a randomized clinical trial. J Cyst Fibros 2023; 22: 916–925. doi: 10.1016/j.jcf.2023.05.013 [DOI] [PubMed] [Google Scholar]
  • 7.Lv Q, Gallardo-Estrella L, Andrinopoulou ER, et al. Automatic analysis of bronchus–artery dimensions to diagnose and monitor airways disease in cystic fibrosis. Thorax 2023; 79: 13–22. doi: 10.1136/thorax-2023-220021 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Díaz AA, Nardelli P, Wang W, et al. Artificial intelligence-based CT assessment of bronchiectasis: the COPDGene study. Radiology 2023; 307: e221109. doi: 10.1148/radiol.221109 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Pieters ALP, Lv Q, Meerburg JJ, et al. Automated method of bronchus and artery dimensions of an adult bronchiectasis population. ERJ Open Res 2024; 10: 00231-2024. doi: 10.1183/23120541.00231-2024 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.van den Bosch WB, Lv Q, Andrinopoulou ER, et al. Children with severe asthma have substantial structural airway changes on computed tomography. ERJ Open Res 2024; 10: 00121-2024. doi: 10.1183/23120541.00121-2023 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Chen Y, Pieters A, Van Der Veer T, et al. Automatic analysis of bronchus-artery ratios and mucus plugs of 640 chest CTs of EMBARC bronchiectasis patients. Eur Respir J 2024; 64: Suppl. 68, PA652. doi: 10.1183/13993003.congress-2024.PA652 [DOI] [Google Scholar]
  • 12.McNally P, Lester K, Elnazir B, et al. WS15.05 Improvements in structural lung disease in people with CF aged 12 and above on elexacaftor/tezacaftor/ivacaftor are sustained for up to two years. J Cyst Fibros 2024; 23: Suppl. 1, S29. doi: 10.1016/S1569-1993(24)00197-8 [DOI] [Google Scholar]
  • 13.Tiddens HAWM, Van Der Veer T, Charbonnier J-P, et al. Fully automated mucus plug quantification on chest CTs and its correlation with all-cause mortality. Am J Respir Clin Care Med 2024; 209: A2778. doi: 10.1164/ajrccm-conference.2024.209.1_MeetingAbstracts.A2778 [DOI] [Google Scholar]
  • 14.Guan WJ, Oscullo G, He MZ, et al. Significance and potential role of eosinophils in non-cystic fibrosis bronchiectasis. J Allergy Clin Immunol Pract 2023; 11: 1089–1099. doi: 10.1016/j.jaip.2022.10.027 [DOI] [PubMed] [Google Scholar]
  • 15.Dunican EM, Elicker BM, Gierada DS, et al. Mucus plugs in patients with asthma linked to eosinophilia and airflow obstruction. J Clin Invest 2018; 128: 997–1009. doi: 10.1172/JCI95693 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Shoemark A, Shteinberg M, De Soyza A, et al. Characterization of eosinophilic bronchiectasis: a European multicohort study. Am J Respir Crit Care Med 2022; 205: 894–902. doi: 10.1164/rccm.202108-1889OC [DOI] [PubMed] [Google Scholar]
  • 17.Heaney LG, Perez de Llano L, Al-Ahmad M, et al. Eosinophilic and noneosinophilic asthma: an expert consensus framework to characterize phenotypes in a global real-life severe asthma cohort. Chest 2021; 160: 814–830. doi: 10.1016/j.chest.2021.04.013 [DOI] [PubMed] [Google Scholar]
  • 18.Polverino E, Dimakou K, Hurst J, et al. The overlap between bronchiectasis and chronic airway diseases: state of the art and future directions. Eur Respir J 2018; 52: 1800328. doi: 10.1183/13993003.00328-2018 [DOI] [PubMed] [Google Scholar]
  • 19.Tiotiu A, Martinez-Garcia MA, Mendez-Brea P, et al. Does asthma–bronchiectasis overlap syndrome (ABOS) really exist? J Asthma 2023; 60: 1935–1941. doi: 10.1080/02770903.2023.2203743 [DOI] [PubMed] [Google Scholar]
  • 20.Polverino E, Dimakou K, Traversi L, et al. Bronchiectasis and asthma: data from the European Bronchiectasis Registry (EMBARC). J Allergy Clin Immunol 2024; 153: 1553–1562. doi: 10.1016/j.jaci.2024.01.027 [DOI] [PubMed] [Google Scholar]
  • 21.Chung WS, Lin CL. Acute respiratory events in patients with bronchiectasis–COPD overlap syndrome: a population-based cohort study. Respir Med 2018; 140: 6–10. doi: 10.1016/j.rmed.2018.05.008 [DOI] [PubMed] [Google Scholar]
  • 22.Chalmers JD, Polverino E, Crichton ML, et al. Bronchiectasis in Europe: data on disease characteristics from the European Bronchiectasis registry (EMBARC). Lancet Respir Med 2023; 11: 637–649. doi: 10.1016/S2213-2600(23)00093-0 [DOI] [PubMed] [Google Scholar]
  • 23.Bendien SA, van Loon-Kooij S, Kramer G, et al. Bronchiectasis in severe asthma: does it make a difference? Respiration 2020; 15: 1–9. doi: 10.1159/000511459 [DOI] [PubMed] [Google Scholar]
  • 24.Donovan GM, Noble PB. Small airways vs large airways in asthma: time for a new perspective. J Appl Physiol 2021; 131: 1839–1841. doi: 10.1152/japplphysiol.00403.2021 [DOI] [PubMed] [Google Scholar]
  • 25.Graham BL, Steenbruggen I, Miller MR, et al. Standardization of Spirometry 2019 Update. An Official American Thoracic Society and European Respiratory Society Technical Statement. Am J Respir Crit Care Med 2019; 200: e70–e88. doi: 10.1164/rccm.201908-1590ST [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Federatie Medisch Specialisten . NVALT richtlijnen: bronchiëctasieën. [NVALT guidelines: bronchiectasis.] 2017. Date last accessed: 27 May 2024. https://richtlijnendatabase.nl/richtlijn/bronchiectasieen/startpagina_bronchiectasieen.html
  • 27.Martinez-Garcia MA, de Gracia J, Vendrell Relat M, et al. Multidimensional approach to non-cystic fibrosis bronchiectasis: the FACED score. Eur Respir J 2014; 43: 1357–1367. doi: 10.1183/09031936.00026313 [DOI] [PubMed] [Google Scholar]
  • 28.Minov J, Karadzinska-Bislimovska J, Vasilevska K, et al. Assessment of the non-cystic fibrosis bronchiectasis severity: the FACED score vs the Bronchiectasis Severity Index. Open Respir Med J 2015; 9: 46–51. doi: 10.2174/1874306401509010046 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Grydeland TB, Dirksen A, Coxson HO, et al. Quantitative computed tomography: emphysema and airway wall thickness by sex, age and smoking. Eur Respir J 2009; 34: 858–865. doi: 10.1183/09031936.00167908 [DOI] [PubMed] [Google Scholar]
  • 30.Koo TK, Li MY. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med 2016; 15: 155–163. doi: 10.1016/j.jcm.2016.02.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Kim JH, Dirksen A, Coxson HO, et al. Relationships between high-resolution computed tomographic features and lung function trajectory in patients with asthma. Allergy Asthma Immunol Res 2023; 15: 174–185. doi: 10.4168/aair.2023.15.2.174 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Yamamoto Y, Kuge T, Miki K, et al. Impact of bronchial wall thickness on airflow obstruction in bronchiectasis. Respir Physiol Neurobiol 2022; 295: 103788. doi: 10.1016/j.resp.2021.103788 [DOI] [PubMed] [Google Scholar]
  • 33.Mohamed Hoesein FA, de Jong PA, Lammers JW, et al. Airway wall thickness associated with forced expiratory volume in 1 second decline and development of airflow limitation. Eur Respir J 2015; 45: 644–651. doi: 10.1183/09031936.00020714 [DOI] [PubMed] [Google Scholar]
  • 34.Akio N, Matsumoto H, Takemura M, et al. Relationship of airway wall thickness to airway sensitivity and airway reactivity in asthma. Am J Respir Crit Care Med 2003; 168: 983–988. doi: 10.1164/rccm.200211-1268OC [DOI] [PubMed] [Google Scholar]
  • 35.Al-Shaikhly T, Murphy RC, Parker A, et al. Location of eosinophils in the airway wall is critical for specific features of airway hyperresponsiveness and T2 inflammation in asthma. Eur Respir J 2022; 60: 2101865. doi: 10.1183/13993003.01865-2021 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Martinez-Garcia MA, Murphy RC, Parker A, et al. Reliability of blood eosinophil count in steady-state bronchiectasis. Pulmonology 2024; 31: 2416836. doi: 10.1016/j.pulmoe.2023.11.006 [DOI] [PubMed] [Google Scholar]
  • 37.Ali MM, Mukherjee M, Radford K, et al. A rapid sputum-based lateral flow assay for airway eosinophilia using an RNA-cleaving DNAzyme selected for eosinophil peroxidase. Angew Chem Int Ed Engl 2023; 62: e202307451. doi: 10.1002/anie.202307451 [DOI] [PubMed] [Google Scholar]
  • 38.Diaz AA, Orejas JL, Grumley S, et al. Airway-occluding mucus plugs and mortality in patients with chronic obstructive pulmonary disease. JAMA 2023; 329: 1832–1839. doi: 10.1001/jama.2023.2065 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Araújo D, Shteinberg M, Aliberti S, et al. The independent contribution of Pseudomonas aeruginosa infection to long-term clinical outcomes in bronchiectasis. Eur Respir J 2018; 51: 1701953. doi: 10.1183/13993003.01953-2017 [DOI] [PubMed] [Google Scholar]
  • 40.Vidaillac C, Chotirmall SH. Pseudomonas aeruginosa in bronchiectasis: infection, inflammation, and therapies. Expert Rev Respir Med 2021; 15: 649–662. doi: 10.1080/17476348.2021.1906225 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.

Supplementary material

00736-2024.SUPPLEMENT.pdf (1.5MB, pdf)
DOI: 10.1183/23120541.00736-2024.Supp1
Open in a new tab

00736-2024.SUPPLEMENT

Articles from ERJ Open Research are provided here courtesy of European Respiratory Society

RESOURCES