Effects of inhaled bronchodilators on dynamic lung mechanics during exercise in COPD: protocol for a randomised, placebo-controlled crossover trial

Graphical abstract

Effects of inhaled bronchodilators on dynamic lung mechanics during exercise in COPD: protocol for a randomised, placebo-controlled crossover trial. CPET: cardiopulmonary exercise testing; di: diaphragmatic; D_LCO: diffusing capacity of the lung for carbon monoxide; EELV: end-expiratory lung volume; EMG: electromyography; ERV: expiratory reserve volume; F_CO2: carbon dioxide fraction; ga: gastric; IC: inspiratory capacity; IOS: impulse oscillometry; IRV: inspiratory reserve volume; MBW: multiple-breath washout; MEP: maximum expiratory pressure; MIP: maximal inspiratory pressure; oes: oesophageal; P: pressure; PFT: pulmonary function test; P_tcCO2: transcutaneous carbon dioxide tension; RV: residual volume; SABA: short-acting β₂-agonists; SAMA: short-acting muscarinic antagonists; SBW: single-breath washout; S_pO2: peripheral oxygen saturation; TLC: total lung capacity; V_cap: volumetric capnography; V_T: tidal volume.

Abstract

Background

Activity-related dyspnoea in moderate–severe COPD is centrally related to heightened wasted ventilation in the physiological dead space colliding with constraints to tidal expansion due to acute-on-chronic gas trapping. Both phenomena are ultimately related to small airway dysfunction. However, it remains unknown whether the positive effects of pharmacological deflation on exertional dyspnoea in COPD can be ascribed to improved small airway conductance, enhancing ventilation–perfusion matching and neuromechanical coupling.

Main hypotheses

We postulate that inhaled bronchodilators will enhance small airway function at rest and during exercise, improving gas exchange efficiency, lung mechanics and exertional dyspnoea in hyperinflated patients with COPD.

Methods

A double-blind, placebo-controlled study involving 25 symptomatic patients (modified Medical Research Council score ≥2) showing functional residual capacity >120% pred and/or >upper limit of normal will receive nebulised salbutamol sulphate (2.5 mg) plus ipratropium bromide (0.5 mg) or normal saline. On different days, patients will undergo a constant-load test at 75% peak to symptom limitation. Single- and multiple-breath N₂ washout and impulse oscillometry (IOS) will assess resting small airway function. Key exercise measurements will include 1) alveolar and physiological dead space by volumetric capnography, 2) operating lung volumes, 3) oesophageal catheter-based respiratory manometry and inspiratory neural drive via diaphragm electromyography, 4) exercise IOS, and 5) exertional dyspnoea (0–10 Borg scale).

Implications

Confirmation of the study's hypotheses will shed new light on the mechanisms of dyspnoea improvement after inhaled bronchodilators in patients with COPD. Moreover, it will provide a novel integrative platform for assessing the effects of therapeutic approaches targeting their small airways.

Shareable abstract

This study will amalgamate measurements of small airway function during exercise with detailed lung mechanics to advance the knowledge of the mechanisms of improved exertional dyspnoea after inhaled bronchodilators in hyperinflated patients with COPD https://bit.ly/44IPhBZ

Introduction

Background and rationale

Activity-related dyspnoea in patients with COPD arises when the (neural) drive to breathe is not adequately rewarded by the output of the lungs (mechanics), a phenomenon appropriately called neuromechanical dissociation [1]. Neuromechanical dissociation stems mainly from acute-on-chronic air trapping, a phenomenon strongly influenced by impaired patency of the noncartilaginous airways <2 mm diameter, the so-called small airway [2]. During exercise, when ventilatory demands increase, the reduced time available for lung emptying worsens dynamic air trapping. This upwardly shifts the operating lung volumes, accelerating the attainment of critically low inspiratory reserves, which reduces dynamic lung compliance, i.e., “constrained” breathing [3]. Moreover, ventilation distribution inhomogeneities throughout the small airways may increase wasted ventilation in the physiological dead space and/or decrease ventilation in well-perfused units, causing hypoxaemia, both leading to “excessive” breathing [4]. “Constrained” and “excessive breathing” conflate to elicit exertional dyspnoea and impair patients’ tolerance to physical effort (figure 1).

FIGURE 1.

A pragmatic framework to translate the neurobiological underpinnings of exertional dyspnoea to cardiopulmonary exercise testing measurements. Increased inspiratory neural drive to the diaphragm and accessory inspiratory muscles occurs in the presence of a) heightened ventilation or b) inspiratory constraints to tidal expansion. c) When tidal displacement is not mechanically restrained, increased dyspnoea/work rate will coexist with relatively preserved dyspnoea/ventilation, i.e., a pattern descriptively termed “excessive breathing”. d) If ventilation is partially impeded due to inspiratory constraints, both dyspnoea/work rate and dyspnoea/ventilation will increase, i.e., “constrained breathing”. These abnormalities frequently coexist in a given patient since excessive ventilation hastens the mechanical constraints; conversely, the latter worsens the ventilatory requirements due to lower tidal volume increasing the physiological dead space (V_Dphys). Shadowed areas in a) and b) represent values commonly associated with increased dyspnoea. Inhaled bronchodilators may lessen “constrained breathing” and, secondarily, “excessive breathing” with the net effect of ventilation output depending on the relative contribution of each of these mechanisms. Grades of dyspnoea severity are according to [44]. Modified, with permission of the publisher, from [4]. ↓: low; ↑: high; ↔: preserved; IC: inspiratory capacity; P_aCO2: arterial partial pressure; Q̇_c: capillary perfusion; V̇_A: alveolar ventilation.

In this context, a long-term effort has been made to deliver bronchodilator treatment to the distal airway in COPD [5]. It is known that β₂-adrenergic receptor expression increases with each airway generation, with the greatest density in the distal airways and alveoli [6]. In fact, more than 90% of all β-adrenergic receptors in the human lung are located in the alveoli [7]. Although cholinergic innervation is higher in large than peripheral airways, muscarinic receptors are found in large and small human airways [8]. Given the inconsistent activation of dry power inhalers in hyperinflated patients [9], nebulised short-acting β₂-adrenergic receptor agonist (SABA) treatment combined with a short-acting antimuscarinic (SAMA) is particularly suitable to assess the effects of bronchodilators on the smaller airways [10], dynamic air trapping and exertional dyspnoea in a single-dose study [11].

Due to lung geometry and the mechanism of particle deposition, however, treatment of small airway disease remains challenging and a major unmet clinical need [11, 12]. Recent advances in respiratory medication delivery have purported to improve medication delivery to the small airways [12, 13]. Despite intensive research efforts, advances in the field have been hampered by the lack of robust platforms for physiological assessment of small airway dysfunction and its corresponding effects on neuromechanical dissociation, gas exchange inefficiency and dyspnoea during exercise. Although our laboratory has extensively investigated the effects of bronchodilators on the determinants of dyspnoea in COPD (reviewed in [14]), none of our studies measured dynamic airway function and detailed lung mechanics coupled with electromyographic metrics of inspiratory neural drive during exercise [1, 15–17] pre- and post-active intervention.

Study objectives and hypotheses

Investigating the effects of inhaled bronchodilators, the cornerstone of COPD treatment, on small airway dysfunction and its deleterious mechanical and gas exchange consequences is paramount to efficacy studies. Our laboratory is proficient in simultaneously assessing small airway dysfunction with impulse oscillometry (IOS)) [18], impaired respiratory mechanics with oesophageal catheter and crural diaphragm electromyography (EMG_di) [1], and impaired pulmonary gas-exchange with breath-by-breath volumetric capnography [19] under the dynamic conditions of exercise (figure 2 and e-figure 1).

FIGURE 2.

Schematic representation of the study protocol. On visit 1, subjects will be assessed for informed consent and potential participation after medical assessment. Patients will then undergo comprehensive pulmonary function tests (PFTs) and, after salbutamol metered dose inhaler, spirometry and an incremental cardiopulmonary exercise test (CPET) to peak work rate. On visits 2 and 3, they will undergo measurements of 1) small airway function (impulse oscillometry (IOS)) and 2) lung mechanics with an oesophageal catheter and diaphragm electromyography (EMG_di). After nebulised placebo (saline) or short-acting β₂-adrenoceptor (SABA)+short-acting antimuscarinic (SAMA), a constant work rate CPET with continuous lung mechanics/EMG_di assessment and serial IOS measurements will be performed to time to exercise intolerance (T_lim). Note that IOS data will be collected prior to inspiratory capacity (IC) manoeuvres. D_LCO: lung diffusing capacity for carbon monoxide; MEP: maximal expiratory pressure; MIP: maximal inspiratory pressure; N₂MBW: nitrogen multiple breath washout; N₂SBW: nitrogen single breath washout; P_tcCO2: transcutaneous partial pressure for CO₂; S_pO2: peripheral oxygen saturation.

We hypothesise that, compared to placebo, dual (SABA-SAMA) bronchodilator treatment in dyspnoeic patients with COPD showing resting hyperinflation would be associated with improved small airway function at rest and during exercise, leading to lung deflation and lower operating lung volumes (figure 3). Both effects are expected to contribute to improved neuromechanical dissociation and decreased wasted ventilation in the physiological dead space (figure 4). The corollary is lower exertional dyspnoea at a given work rate and ventilation, signalling improved “constrained breathing” (figure 1).

FIGURE 3.

Expected physiological and sensory changes secondary to improved small airway function after bronchodilator treatment compared to placebo. The study's main outcomes are highlighted. See the main text for further details. ↑: higher; ↓: lower; CR: category ratio; EELV: end-expiratory lung volume; EMG_di: diaphragm electromyography; ERV: expiratory reserve volume; P_CO2: carbon dioxide pressure; F_CO2: carbon dioxide fraction; IC: inspiratory capacity; IOS: impedance oscillometry; P_di: trans diaphragmatic pressure; P_ga: gastric (∼ abdominal) pressure; P_oes: oesophageal (∼pleural) pressure; RV: residual volume; TLC: total lung capacity; V_cap: volumetric capnography; V_T: tidal volume.

FIGURE 4.

Overview of the putative mechanical consequences of lung deflation during exercise based on a previous work from our group [1] involving normal subjects (blue) and patients showing progressively worse resting hyperinflation (from green to red). Lower functional residual capacity (FRC), the expected effect of pharmacological deflation, was associated with the following at given ventilation: a) lower inspiratory neural drive; lower b) oesophageal and c) gastric tidal pressures, d) ventilatory muscle recruitment (VMR) towards low values in keeping with less activation of the accessory inspiratory muscles, and improved e) neuromechanical dissociation of the respiratory system during inspiratory capacity and f) tidal volume. In the present study, we hypothesise similar within-subject findings if inhaled bronchodilators effectively decrease resting and exercise lung volumes secondary to improved small airway function. See the main text for further details. EMG_di: diaphragm electromyography; IC: inspiratory capacity; P_ga: gastric (∼abdominal) pressure; P_gaTidal: gastric (∼abdominal) pressure at tidal volume; P_oes: oesophageal (∼pleural) pressure; P_oesTidal: gastric (∼abdominal) pressure at tidal volume; V̇_E: minute ventilation; V_T: tidal volume. Modified, with permission of the publisher, from [1].

Methods

Study design

This is a single-centre (Respiratory Investigation Unit, Kingston Health Sciences Centre, Queen's University, Kingston, ON, Canada), double-blinded, randomised, placebo-controlled crossover study designed to demonstrate the superiority of single-dose nebulised SABA-SAMA over placebo on metrics of small airway function and exercise performance in hyperinflated patients with COPD (table 1). Participants will be randomised in a 1:1 ratio to one of two treatment sequences, with each participant receiving both placebo and bronchodilators on different days (see the online supplement for additional methodological details).

TABLE 1.

The study's administrative information as per the SPIRIT (Standard Protocol Items: Recommendations for Interventional Trials) guidelines

SPIRIT requirement	Information
Study design, population, interventions, and trial acronym	A prospective, randomised controlled trial on the effects of single dose nebulised salbutamol sulphate (2.5 mg) and ipratropium bromide (0.5 mg) on Small airway function, Operating lUng volumes, and Neuromechanical Dissociation (SOUND) in hyperinflated patients with COPD
Trial identifier and registry name	ClinicalTrials.gov (ClinicalTrials.gov Identifier: NCT06825013) Shortened title: Bronchodilators and Lung Mechanics During Exercise in COPD (SOUND)
Items from the World Health Organization trial registration data set
1) Primary registry and trial identifying number	ClinicalTrials.gov (ClinicalTrials.gov Identifier: NCT06825013)
2) Date of registration in primary registry	13 February 2025
3) Secondary identifying numbers	None
4) Source(s) of monetary or material support	William M. Spear Endowment Fund in Pulmonary Research and the Richard K. Start Memorial Fund for Respiratory Diseases in the Faculty of Health Sciences/Kingston General Hospital, Queen's University
5) Primary sponsor	Queen's Health Sciences Research Queen's University Botterell Hall, 6th Floor, Room 650 18 Stuart Street, Kingston, Ontario K7L 3N6. E-mail: qhs_research@queensu.ca
6) Secondary Sponsor(s)	None
7) Contact for public queries	J. Alberto Neder, MD, PhD, DSc, FRCPC, FERS. Respiratory Investigation Unit (RIU)-Laboratory of Clinical Exercise Physiology (LACEP), Kingston General Hospital, Connell 2-200. 76 Stuart St., K7L 2V7. Kingston, ON, Canada. E-mail: alberto.neder@queensu.ca
8) Contact for scientific queries	J. Alberto Neder, MD, PhD, DSc, FRCPC, FERS. Respiratory Investigation Unit (RIU)-Laboratory of Clinical Exercise Physiology (LACEP), Kingston General Hospital, Connell 2-200. 76 Stuart St., K7L 2V7. Kingston, ON, Canada. E-mail: alberto.neder@queensu.ca
9) Public title	Effects of inhalers in persons with COPD: how can they improve shortness of breath on exertion?
10) Scientific title	The effects of pharmacological deflation on small airway function, operating lung volumes and neuromechanical dissociation in COPD: the SOUND study
11) Countries of recruitment	Canada, single centre
12) Health condition(s) or problem(s) studied	Shortness of breath, COPD
13) Intervention(s)	Active comparator: bronchodilator Participants will receive a single dose of a combination of short-acting bronchodilator (salbutamol sulphate (2.5 mg)+ipratropium bromide (0.5 mg)) via nebuliser Placebo comparator: placebo Participants will receive a single dose of normal saline via nebuliser.
14) Key inclusion and exclusion criteria
Inclusion criteria	Male or female ≥40 years of age Current or former smokers with ≥20 pack-years FEV1/FVC<lower limit of normal Δ pre–post FVC ≥10% pred after 400 μg inhaled salbutamol pMDI on visit 1 FRC ≥120% pred and/or the upper limit of normal Clinically stable as defined by no exacerbations in the preceding 6 weeks Ability to provide informed consent and perform all study procedures
Exclusion criteria	Major cardiopulmonary diseases other than COPD (particularly asthma, interstitial lung disease, pulmonary hypertension and congestive heart failure) Neuromuscular or musculoskeletal disease Any other disorder that may contribute to exertional dyspnoea Any contraindication to cardiopulmonary exercise testing
15) Study type	A phase 4, interventional, randomised, placebo-controlled, cross-over trial Randomisation of treatment sequences is 1:1 and generated using a computer-generated, random number sequence
16) Date of first enrolment	1 October 2024
17) Sample size	n=20
18) Recruitment status	Recruiting: participants are currently being recruited and enrolled
19) Primary outcome(s)	Airway resistance measured by impulse oscillometry at isotime (maximum exercise time achieved by all participants) during cardiopulmonary exercise testing (cycle ergometer) The outcome will be obtained every 2 min during exercise Key secondary outcomes: Inspiratory neural drive by diaphragm activation at isotime (maximum exercise time achieved by all participants) during cardiopulmonary exercise testing (cycle ergometer)
	Trans-diaphragmatic pressure at isotime (maximum exercise time achieved by all participants) during cardiopulmonary exercise testing (cycle ergometer) Inspiratory capacity to obtain end-inspiratory and end-expiratory lung volumes at isotime (maximum exercise time achieved by all participants) during cardiopulmonary exercise testing (cycle ergometer) Perceptual (e.g. intensity of dyspnoea) responses at isotime (maximum exercise time achieved by all participants) during a standardised cardiopulmonary exercise testing (cycle ergometer) Exercise tolerance will be assessed as the time from start of loaded pedalling to end of loaded exercise perceptual (e.g. intensity of dyspnoea) responses at isotime (maximum exercise time achieved by all participants) during a standardised cardiopulmonary exercise testing (cycle ergometer)
20) Ethics review	Approved. Health Sciences REB, Queen's University, Annual Renewal Approval on 13 January 2025 up to 17 January 2026. TRAQ: 6040014. Health Sciences and Affiliated Teaching Hospitals Research Ethics Board (HSREB). Vice-Principal Research 355 King St. West, Kingston ON K7L 2X3, Canada
21) Completion date	The study has not been completed
22) Summary results	No data have been analysed
23) IPD sharing statement	Deidentified clinical trial IPD will be made available upon request
Date and version identifier	Protocol version: 1.1 (6 December 2024)
Sources and types of financial, material and other support	William M. Spear Endowment Fund in Pulmonary Research and the Richard K. Start Memorial Fund for Respiratory Diseases in the Faculty of Health Sciences/Kingston General Hospital, Queen's University
Names, affiliations and roles of protocol contributors	J.A. Neder, M.D. James and S.G. Vincent, (Respiratory Investigation Unit, Queen's University, Kingston, ON, Canada) developed the study concept J.A. Neder, K. Alosta, R. Smyth, M.D. James, N.D., Domnik and S.G. Vincent (Respiratory Investigation Unit, Queen's University, Kingston, ON, Canada) developed the study design and protocol All authors (Respiratory Investigation Unit, Queen's University, Kingston, ON, Canada) contributed to the pilot testing and refinement of the submitted trial protocol J.A. Neder, K. Alosta and M.D. James wrote the first version of the manuscript All authors contributed to the revisions and approval of the final manuscript
Name and contact information for the trial sponsor	Queen's Health Sciences Research. Queen's University Botterell Hall, 6th Floor, Room 650 18 Stuart Street, Kingston, Ontario K7L 3N6 Email: qhs_research@queensu.ca
Role of study sponsor	The sponsor had no role in study design, collection, management, analysis and interpretation of data, writing of the report, and the decision to submit the report for publication, including whether they will have ultimate authority over any of these activities
Other responsibilities	Single-centre study coordinating all activities There is no steering committee, end-point adjudication committee, data management team and other individuals or groups overseeing the trial

FEV₁: forced expiratory volume in 1 s; FRC: functional residual capacity; FVC: forced vital capacity; IPD: individual participant-level data; pMDI: pressurised metered-dose inhaler.

Study population

We will recruit 20 dyspnoeic patients (modified Medical Research Council score ≥2) with COPD with a post-bronchodilator forced expiratory volume in 1 s (FEV₁)/forced vital capacity (FVC) [20], showing resting lung hyperinflation (functional residual capacity >upper limit of normal and/or >120% pred, i.e., typically moderate to severe airflow obstruction) and a volume response to salbutamol pressurised metered-dose inhaler (pMDI) on visit 1 [21–23]. Participants will give written informed consent (see online supplement) and be able to perform all study procedures. Participants will refrain from smoking ≥1 h before visits, avoid caffeine and alcohol, and exercise ≥4 h before testing. Table 1 provides a complete list of eligibility criteria.

Study intervention

Participants will receive a nebulised SABA-SAMA combination (salbutamol sulphate (2.5 mg)+ipratropium bromide (0.5 mg)) at one study visit (visit 2 or 3) or nebulised placebo (normal saline) on the alternate study visit. The active study intervention and placebo will be delivered using an ultrasonic nebuliser device. Any bronchodilators will be discontinued prior to the visits according to standard recommendations, i.e., 6, 12 and 24 h before testing for short-acting, long-acting and “ultra” long-acting bronchodilators. They will return to their usual treatment between the visits and after the last visit.

Study outcomes

We a priori consider that the study's main hypotheses to be confirmed if the following changes are observed post-bronchodilation compared to baseline (figures 3 and 4):

• lower difference between IOS-determined airway resistance at 5 Hz (R₅) and 20 Hz (R₂₀) at rest and during exercise, reflecting improved small airway function [24];

• lower functional residual capacity, indicating resting lung deflation [3];

• higher inspiratory capacity, reflecting a greater volume for tidal expansion [1];

• lower isotime EMG_di (% max):tidal volume/inspiratory capacity, indicating improved neuromechanical coupling [1];

• lower isotime EMG_di:EMG_{di, max}, in keeping with lower inspiratory neural drive [25]; and, as a corollary,

• lower isotime exertional dyspnoea and longer exercise tolerance [26].

Significant cross-correlations between within-subject decrements in ΔR₅–R₂₀ on exercise with lung deflation and improved neuromechanical dissociation would provide support for a mechanistic association between improved small airway function and their sensory–mechanical consequences.

Participant timeline

After written informed consent, all participants will be asked to attend our laboratory thrice. Each visit will be conducted at the same time, in the morning, 2–7 days apart (figure 2 and e-figure 1).

Visit 1 (eligibility assessment)

A medical history will be obtained from each participant, along with an assessment of self-reported activity-related dyspnoea and symptom status. Lung function will be assessed with comprehensive pulmonary function tests (PFTs) including spirometry (pre- and post-bronchodilator, 400 μg salbutamol), lung volumes by body plethysmography, diffusing capacity of the lungs for carbon monoxide, maximum voluntary ventilation, single- and multiple-breath nitrogen washout, maximum inspiratory and expiratory mouth occlusion pressures, and IOS. These resting tests will be followed by a symptom-limited incremental cardiopulmonary exercise test (CPET) on a cycle ergometer to determine peak work rate.

Visit 2–3 (efficacy evaluation)

Participants will undergo spirometry, body plethysmography, IOS and measurements of static and dynamic lung compliance prior to and 20 min after the administration of nebulised salbutamol (2.5 mg)+ipratropium (0.5 mg) or placebo (saline). The participants will then complete a constant-load test at 75% peak to time to exercise intolerance (T_lim), with continuous monitoring of lung mechanics and discrete IOS measurements, as well as serial (every 2 min) measurements of inspiratory capacity subsequent to IOS data collection and dyspnoea ratings using the 0–10 category-ratio Borg scale [26] (figure 2).

Sample size

Given the absence of similar studies using the same array of physiological outcomes to address the effects of bronchodilators in COPD, an effect size could not be properly calculated. Based on our previous studies with similar interventions, however, we estimated that 20 patients would be sufficient to show an increase in T_lim (reviewed in [27]) beyond the minimal clinically significant difference of 100 s and/or 33%. [26].

Study recruitment

Eligible participants will be identified and recruited from 1) institutional outpatient respirology/COPD clinics, 2) a database of individuals who have previously participated in research studies with our research group and 3) through community advertisement. Study information will be recorded only after obtaining informed consent from each participant. Participants will be assigned a unique study identifier and all information will remain anonymised and securely stored to ensure participant privacy.

Randomisation and blinding

Eligible participants will be randomly assigned to one of two treatment sequences, namely 1) placebo followed by SABA-SAMA or 2) SABA-SAMA followed by placebo. Randomisation of treatment sequences will be 1:1, generated using a computer-based random number generator. Each participant will receive the two treatments assigned in the sequences, with a washout period of at least 7 days between study visits. Preparation of the study medication will be conducted by an unblinded study team member who will not participate in patient testing. The unblinded team member can perform emergency unblinding in the event of an adverse reaction to the study intervention. The study interventions will be pre-prepared and use the same delivery device to ensure blinding of the participants and study team.

Study procedures

Standard PFTs

All tests will be performed using an automated equipment (Vmax 229d, SensorMedics) according to current recommendations. Measurements will be expressed as absolute values and compared to the standards proposed by the Global Lung Function Initiative [21–23].

Small airway function

IOS (Masterscreen IOS Digital, Jaeger; Leibnizstrasse, Germany) will be performed according to current technical standards [28]. The main variable will be the difference ΔR₅–R₂₀ with high values being biased to reflect the mechanical properties of the distal lung. Secondary variables will be respiratory system reactance at 5 Hz, the integrated area of low-frequency reactance from 5 Hz to resonant frequency and respiratory impedance at 5 Hz [24]. Additional metrics related to small airway function will be obtained at rest by single-breath nitrogen washout (slope of phase III calculated between 25%–75% of vital capacity) and multiple-breath nitrogen washout (S_cond assessing the convection-dependent ventilation inhomogeneity and S_acin interrogating diffusion–convection-dependent inhomogeneity) (Vmax Encore 29C, SensorMedics) [29]. IOS measurements during CPET will always precede the inspiratory capacity manoeuvres (figure 2).

CPETs

Incremental and constant load tests (75% peak to symptom limitation) will be conducted on a stationary cycle ergometer (Ergoselect 200s, Erogline GmbH, Bitz,Germany) using a computerised metabolic system (Vmax 229, SensorMedics). Pilot tests indicated that reproducible IOS measurements during constant work rate exercise can be obtained by applying external pressure to the participant's cheeks after switching him from the CPET mouthpiece to that used on the IOS system. Measurements will be recorded in duplicate every 2 min, lasting ∼40 s. Additionally, participants will be instrumented throughout with an oesophageal catheter for continuous measurement of respiratory mechanics and inspiratory neural drive and a transcutaneous CO₂ monitor for dead space assessment by volumetric capnography (see details below). Exertional dyspnoea will be assessed with the 0–10 category-ratio Borg dyspnoea scale [30] and exercise tolerance will be assessed by exercise intolerance time (T_lim) [26]. Additional protocol details are provided in the online supplement.

Lung mechanics and inspiratory neural drive

At the commencement of visit 2/3, participants will be instrumented with a small-bore, high-compliance catheter with two pressure balloons and five paired electrodes. The catheter will be inserted by experienced personnel via the nose and down into the oesophagus and stomach (after application of a topical anaesthetic spray to the nasal passages and throat) [31]. The oesophageal pressure (P_oes) will be considered a surrogate of pleural pressure, while the gastric pressure (P_ga) will be considered an index of abdominal pressure; their difference (P_ga−P_oes) will provide the trans-diaphragmatic pressure (P_di). Maximal P_oes (P_{oes, max}) will be determined during a sniff manoeuvre (largest pre- or post-exercise value). Maximal P_di (P_{di, max}) will be selected from the largest value measured during serial inspiratory capacity manoeuvres. Based on these measurements, the following will be calculated [1, 15–17]:

• diaphragm muscle effort: the fraction of tidal swings in P_oes relative to P_{oes, max},

• total respiratory muscle effort: the fraction of tidal P_di relative to P_{di, max}, and

• ventilatory muscle recruitment: the difference between end-inspiratory P_ga and end-expiratory P_ga divided by the difference in end-inspiratory P_oes and end-expiratory P_oes: a negative value indicating greater diaphragm contribution and a positive value indicating greater rib cage contribution to inspiration [31].

Moreover, the multi-paired electrodes will allow breath-by-breath measurements of EMG_di (Guangzhou Yinghui Medical Equipment Ltd, Guangzhou, China) [1, 16, 25, 32]. This signal will be integrated with respiratory manometry, yielding the following:

• inspiratory neural drive as the average amplitude of the EMG_di (μV) relative (%) to EMG_{di, max}, i.e., the largest value measured during serial inspiratory capacity manoeuvres (to be consistent with P_{di, max}), and

• the ratio of EMG_di (%max) to tidal volume/inspiratory capacity will be used as an index of neuromechanical coupling of the respiratory system, with a value of >0.75 being defined as neuromechanical dissociation.

Data will be acquired and analysed using LabChart v8.1.2.1 (ADInstruments, Castle Hill, Australia). A detailed description of these measurements is available in the online supplement.

Volumetric capnography

Volumetric capnography will be utilised in conjunction with standard CPET variables and transcutaneous carbon dioxide tension [33]. Using an analogue-to-digital converter integrated into our CPET system (Vmax 229, CareFusion, San Diego, USA), the respiratory gas flow and CO₂ concentration signals will be recorded at a 100 Hz sampling rate using LabChart. We will record breath-by-breath anatomical, alveolar and physiological dead space (e-figure 2) based on software developed initially at the Lundquist Institute for Biomedical Innovation at Harbor-UCLA Medical Center, USA (e-figure 3). To align these signals with CPET data, they will also be averaged as a 20 s arithmetic mean. Further details are available in the online supplement.

Data management

Data will be anonymised and entered into customised data sheets immediately after the experiments. Data will be securely stored with a password known only to study members involved in data analysis. EMGdi and respiratory manometry data will be reviewed on a breath-by-breath basis to ensure maximal quality and freedom from artifacts, e.g., sighing or swallowing.

Statistical analysis

Depending on the variable distribution, paired t or Wilcoxon's rank sum tests will determine the effects of active treatment versus placebo. Between-variable correlations will be assessed using Pearson or Spearman correlation calculations. Intra-test repeatability of IOS measures will be tested using intraclass correlation coefficients. Generalised estimating equations for a repeated measure design will compare interventions (bronchodilators versus placebo) at different exertional time-points; in variables with significant main effects for intervention, a paired t-test will be used to identify significant between-intervention differences at submaximal exercise iso-work rate or iso-ventilation comparisons. A p-value <0.05 will set the limit for statistical significance.

Monitoring

A data monitoring committee was considered unnecessary for this clinical physiology, proof-of-concept study involving medications long used in the treatment of COPD. There will be no interim analysis. Any adverse events will be communicated to the Queen's University Health Sciences and Affiliated Teaching Hospitals Research Ethics Board and to the participant's family physician.

Ethics and dissemination

The study protocol has been approved by the Queen's University Health Sciences and Affiliated Teaching Hospitals Research Ethics Board (TRAQ: 6040014). Any changes in the protocol will be communicated to this Board and updated on ClinicalTrials.gov. Informed consent (online supplement) will be obtained by pre-specified members of the research team. The principal investigator will not be involved in this task. The authors will declare any financial and other competing interests. All data will be anonymised and kept securely stored in the laboratory's computers. Regardless of the results, they will be submitted to peer-reviewed journals. Data will not be shared until completion of the trial and final analysis. Anonymised data will be made available upon reasonable request after appropriate ethical approval and institutional data-sharing agreements.

Discussion

The consequences of small airway dysfunction on the respiratory physiological and sensory responses to exercise remain incompletely understood in patients with COPD. This study will innovate by directly interrogating the relationship between small airway dysfunction during exercise and key mechanical and gas exchange underpinnings of exertional dyspnoea. Our results might provide novel evidence for the clinical relevance of targeting the small airway to improve relevant patient-related outcomes, including activity-related dyspnoea and exercise intolerance.

Some of the methodological aspects of the present study deserve special attention. As a proof-of-concept study, we will focus on the subpopulation of patients with COPD who are more likely to benefit from pharmacological lung deflation, i.e., dyspnoeic and hyperinflated patients [27]. We reasoned that this would be more appropriate than any metric of airflow obstruction severity since there is a large variability on exertional dyspnoea and hyperinflation at a given FEV₁ in COPD [14]. Moreover, patients will be required to present with volume recruitment in response to inhaled salbutamol pMDI on visit 1, i.e., a significant increase in FVC, signalling greater decrements in residual volume than total lung capacity. Given the key relevance of small airway function in determining residual volume [34], these patients are well suited to place our main research question under proper scrutiny [35]. Given the negative effects of air trapping on the inspiratory flows required for homogenous lung deposition with dry powder inhalers [9, 36] and inconsistent deposition after a single inhalation with metered-dose devices [37], we opted for the nebulised administration. A single visit for active treatment is acceptable since the deflating effects are seen after an acute dose of bronchodilators. Constant work rate exercise testing has been widely recommended to test the effects of interventions in patients with COPD [26].

IOS has been used primarily on resting conditions due to the potential confounding effects of high respiratory pressures and lung volumes on airway resistance [24]. However, the method has been successfully used during incremental exercise in COPD with acceptable reproducibility (particularly ΔR₅–R₂₀) and responsiveness [18]. We carefully chose to use IOS under relatively constant (and low) ventilatory demands during a square wave test, before the inspiratory capacity manoeuvre, which is likely to interfere with IOS variables (figure 2). Assuming a mean FEV₁ close to 1 L (giving estimated maximal voluntary ventilation ∼40 L·min⁻¹), it seems unlikely that our patients will ventilate substantially above 30 L·min⁻¹ at 75% peak. Moreover, tidal volume under mechanical constraints shows lower variability compared to normal subjects and any increase in tidal volume post-bronchodilator is expected to be relatively modest (typically within 200 mL) [35]. Other metrics based on single- and multiple-breath N₂ washout [29] might provide further corroborative evidence for small airway dysfunction indicated by IOS; however, they were not included as outcomes due to the complexity of the study. Significant correlations between high exercise ΔR₅–R₂₀ and other IOS variables under placebo, with the immediate consequences of small airway dysfunction (lung hyperinflation) and their reciprocal changes after active treatment, would strengthen our line of interpretation.

Regarding the assessment of lung mechanics and gas exchange, several lines of evidence indicate that crural EMG_di (%max) is responsive to chemical and mechanical loading of the respiratory system and is not impeded by disease-specific constraints on ventilation. These include 1) the effective linear rise in conjunction with increasing ventilation in healthy controls, 2) the rise during exercise even in the presence of dynamic mechanical loading and ventilatory constraints, 3) the rise with CO₂ loading, and 4) inspiratory threshold loading (reviewed in [1]). We extensively validated crural EMG_di (%max):tidal volume/inspiratory capacity as a metric of neuromechanical dissociation in COPD and interstitial lung disease [16]. Elevated crural activation correlates well with increasing dyspnoea intensity during exercise in health and disease [16]. Although a previous study from our group did not find a significant effect of bronchodilators on Bohr–Enghoff physiological dead space [38], recent functional imaging found improved ventilation–perfusion matching after bronchodilation [39]. Continuous measurements of alveolar dead space with volumetric capnography may enhance our ability to detect relatively minor improvements in gas exchange efficiency after active treatment missed in our previous study [38].

There is a renewed interest in treating the distal lung via improved drug delivery, including extra-fine formulations [40] and dual or triple therapy as co-suspension [13]. Despite the key role of small airway disease in the pathogenesis of COPD [41] and physiological considerations that span several decades [5], the evidence that treating the so-called “silent zone” improves patient-centred outcomes remains surprisingly meagre. For instance, recent studies showing encouraging results based on functional imaging [42] have neither assessed whether these pictorial changes translate into improved lung mechanics nor examined their consequences for exertional dyspnoea. Our study has the potential to fill this important gap in knowledge, advancing a field that has neglected the value of clinical physiology in the current era of molecular biology [14, 43]. The present study is also designed to establish a novel physiologic pipeline for studying the putative effects of newer and emerging bronchodilator formulations targeting the distal airways in COPD and other prevalent obstructive diseases.

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

01323-2024.SUPPLEMENT

Data availability

Data will not be shared until completion of the trial and final analysis. Anonymised data would be made available upon reasonable request after appropriate ethical approval and institutional data sharing agreements.