Mechanisms of Exertional Dyspnea in Fibrotic Interstitial Lung Disease

The purpose of this study is to determine the physiological mechanisms of shortness of breath (dyspnea) in patients with fibrotic Interstitial Lung Disease (ILD) and to determine how breathing supplemental oxygen can manipulate these mechanisms to improve dyspnea and exercise capacity.

The research question is twofold: (Aim 1) To determine the physiological mechanisms of exertional dyspnea in patients with fibrotic ILD; (Aim 2) To determine the effects of hyperoxia on exertional dyspnea and exercise endurance in patients with fibrotic ILD.

Experimental Overview: Participants with fibrotic ILD and control participants will report to the exercise laboratory on four separate occasions separated by a minimum of 48 hours between visits. On visit 1, participants and control participants will complete medical history screening, chronic activity-related dyspnea questionnaires, anthropometric measurements, pulmonary function assessment, and a symptom limited incremental cycle exercise test for familiarization purposes. On visit 2, participants and control participants will perform pulmonary function testing followed by another incremental cycle exercise test. Detailed physiological and sensory measurements will be obtained on both visits but the primary analysis will be based on visit 2 results. Data from visit 2 will address the Aim 1. Visits 3 and 4 will include pulmonary function testing followed by a constant-load cycle exercise test at 75% of peak incremental work rate while breathing, in randomized order, either room air or hyperoxia (60% oxygen). Participants and control participants breathing hyperoxia on visit 3 will breathe room air on visit 4 and vice versa while being blinded to the gas concentration. A multi-pair electrode catheter that combines two balloons will be inserted into the esophagus and near infrared spectroscopy will be used to measure tissue oxygenation on visits 2, 3 and 4. Data from visits 3 and 4 will address Aim 2.

Measurements:

• Pulmonary Function: simple spirometry, plethysmography, diffusing capacity, maximum respiratory pressures, static compliance and recoil pressure will be performed on visit 1. Pulmonary function testing on visits 2-4 will only include spirometry and plethysmography so that total lung capacity and vital capacity can be obtained for the determination of operating lung volumes.
• Dyspnea Evaluation: Dyspnea intensity and perceived leg discomfort will be evaluated at rest, every minute during exercise, and at peak exercise using the modified 10-point Borg scale on all testing visits. Upon exercise cessation, subjects will be asked to verbalize their main reason(s) for stopping exercise (i.e., breathing discomfort, leg discomfort, combination of breathing and legs, or some other reason) and to select qualitative descriptors of breathlessness using an established questionnaire.
• Cardio-respiratory Responses to Exercise: Standard cardio-respiratory measures, including minute ventilation, oxygen consumption (VO2), carbon dioxide production, partial pressure of end-tidal carbon dioxide, tidal volume (VT), and breathing frequency.
• Operating volumes will be derived from dynamic inspiratory capacity (IC) manoeuvres. Arterial oxygen saturation will be measured using pulse oximetry. Electrocardiography and blood pressure will be monitored for safety purposes.
• Respiratory Mechanics: Diaphragmatic electromyography (EMGdi) will be measured using a multi-pair electrode catheter that combines two balloons for measuring esophageal and gastric pressures. The ratio of EMGdi to EMGdimax will be used as an index of neural respiratory drive. The ratio between VT and vital capacity (VC) will be used to represent the mechanical response of the respiratory system. Normalizing for EMGdimax and VC allows the stimulus intensity to be standardized and compared across individuals. Thus neuromechanical uncoupling of the respiratory system will be determined as the ratio (or interaction) between neural drive and the mechanical response of the respiratory system (EMGdi/EMGdimax : VT/VC).
• The mechanical work of breathing (WOB) will be calculated as the area within ensemble averaged esophageal pressure-volume loops.

Statistical Analysis:

Aim 1: Exercise-response slopes (e.g., Borg/VO2) will be determined. Briefly, the investigators will obtain the slope from a plot of Borg vs. VO2 for each participant's and control participant's incremental exercise test performed on visit 2. The investigators will determine the bivariate association of Borg/VO2 slope with VO2 slopes of neuromechanical uncoupling, drive to breathe, and VT response using Spearman correlation coefficients. The investigators will then force all three predictor variables into a multivariate linear regression model with Borg/VO2 slope as the outcome variable in order to identify the independent association of neuromechanical uncoupling with exertional dyspnea. Variables will be transformed to approximate a normal distribution if necessary and predictor variables reaching statistical significance will be assessed for a linear relationship with the outcome variable.

Aim 2: The investigators will first use a paired t-test to identify changes in dyspnea and exercise time comparing room air breathing to hyperoxia during constant-load exercise tests on visits 3 and 4. Multivariate models will then be developed using the between-test difference in Borg dyspnea scale and exercise time for each individual as the outcome variables. Predictor variables will include the between-test difference in neuromechanical uncoupling and its individual components. Outcome and predictor variables for the dyspnea outcome will be based on the between-test difference of these variables at iso-time (i.e., the maximum time for which the patient exercised for both the room air and hyperoxia tests). Predictor variables for exercise time will be measured at the end of the test.

As exploratory analyses, the investigators will repeat the above multivariate linear regression analyses in the IPF subgroup, and the investigators will add a categorical ILD subgroup variable to these analyses in order to identify other inter-group differences. Additional adjustment for potential confounders (e.g., sex, body mass index, high resolution computed tomography fibrosis score) will be used in these exploratory analyses if the subgroup sample size is sufficient. A p value < 0.05 will be considered significant for all analyses. Data analysis will be performed using Stata v11.2 (StataCorp, Texas, USA).