The annual rate of moderate or severe COPD exacerbations which occurred during treatment was assessed. Moderate exacerbations were defined as exacerbations that required treatment with oral/systemic corticosteroids and/or antibiotics (not involving hospitalization or resulting in death). Severe exacerbations were defined as exacerbations that required hospitalization or resulted in death. Analysis performed using a generalized linear model assuming a negative binomial distribution. ITT population was used which comprised of all randomized participants, excluding those who were randomized in error. Only those participants with non-missing co-variates were included in the analysis.

Pulmonary function was measured by forced expiratory volume in one second (FEV1), defined as the maximal amount of air that can be forcefully exhaled from the lungs in one second. Trough FEV1 measurements were taken electronically by spirometry on Days 2, 14, 28, 56 and 84. BL was defined as the mean of the assessments made 30 minutes pre-dose and immediately pre-dose on Treatment Day 1.Trough FEV1 was defined as the mean of the FEV1 values obtained 23 and 24 hours after previous morning's dosing. Change from BL was calculated as the average at each visit minus the BL value. Analysis was performed using a repeated measures model with covariates of treatment, reversibility status (stratum), BL, region, day, day by BL and day by treatment interactions.

RVEDVI is a measure of the volume of blood in the right ventricle at the end of diastole, normalized over body surface area and was measured using Cardiac Magnetic Resonance (CMR) imaging. RVEDVI is calculated as the right ventricular end diastolic volume (RDEDV) divided by the body surface area (BSA). The change from Baseline in RVEDVI was analyzed using a mixed model analysis with period, treatment group, and Baseline RVEDVI fitted as fixed effects and participants fitted as a random effect. The Baseline is defined as the assessment performed pre-dose at Day 1 of Treatment Period 1. The change from Baseline is calculated as the RVEDVI value at the end of each treatment period minus the Baseline value. The Per Protocol (PP) Population was comprised of all participants in the modified intent-to-treat (mITT) Population not identified as having deviations considered to impact the primary efficacy analysis.

Pulmonary function was measured by forced expiratory volume in one second (FEV1), defined as the maximal amount of air that can be forcefully exhaled from the lungs in one second. The weighted mean was calculated from the pre-dose FEV1 and post-dose FEV1 measurements at 5, 15, 30 minutes and 1, 2, 4, 6, 8, 12, 13, 14, 16, 20, and 24 hours on Treatment Day 84. Baseline trough FEV1 was the mean of the two assessments made 30 and 5 minutes pre-dose on Treatment Day 1. Change from Baseline (BL) was calculated as the average of the Day 84 values minus the Baseline value. Analysis was performed using an analysis of covariance (ANCOVA) model with covariates of BL FEV1, exacerbation history and reversibility stratum, smoking status at screening, country, and treatment group.

Pulmonary function was measured by forced expiratory volume in one second (FEV1), defined as the maximal amount of air that can be forcefully exhaled in one second. Trough FEV1 was defined as the pre-dose and pre-bronchodilator FEV1, which was obtained at each clinic visit. Baseline is defined as the mean of the two assessments made 30 minutes pre-dose and 5 minutes pre-dose on Treatment Day 1.Trough FEV1 is defined as the mean of the FEV1 values obtained 23 and 24 hours after dosing at each clinic visit. Change from Baseline was calculated as the average at each clinic visit minus the Baseline value. Analysis was performed using a repeated measures model with covariates of treatment, smoking status at screening (stratum), baseline - mean of the two assessments made 30 minutes pre-dose and immediately pre-dose on Day 1, day, day by baseline and day by treatment interactions.

PWV is defined as the speed of travel of the pressure pulse along an arterial segment and can be obtained for any arterial segment accessible to palpation. aPWV is measured with tonometers positioned transcutaneously at the base of the common carotid artery and over the femoral artery. PWV increases with arterial stiffness and is defined by the Moens-Korteweg equation: PWV=square root of (Eh/2ρR), where E is Young's modulus of the arterial wall, h is the wall thickness, R is the arterial radius at the end of diastole, and ρ is the blood density. Change from BL was calculated as the Day 168 value minus the BL value. The analysis was performed using a repeated measures model with covariates of treatment, visit, age, gender, smoking history, history of exacerbation strata, geographical region, BL aPWV and interaction terms of BL by visit and treatment by visit.

Pulmonary function was measured by forced expiratory volume in one second (FEV1). The weighted mean was calculated from the pre-dose FEV1 and post-dose FEV1 measurements at 5, 15, 30, and 60 minutes (min) and 2, 4, 6, 8, 12, 13, 14, 16, 20, and 24 hours on Treatment Day 84. Baseline trough FEV1 was the mean of the two assessments made 30 and 5 minutes pre-dose on Treatment Day 1. Change from Baseline was calculated as the average of the Day 84 values minus the Baseline value.

Death from any cause: which occurred from the day of starting IP until the Commone End Date (CED). Common End Date (CED) is the study end date that was pre determined where approximately 1000 deaths would have occurred in the Intent-toTreat Efficacy (ITT-E) Population. Only deaths which occurred on or before the CED were used for the primary analysis. Those who had not died by CED, but who were known to be alive on or after the CED, were censored at the CED. Cox Proportional Hazards (PH) Model was adjusted for age, and gender, including all 4 arms. A hazard ratio of less than 1 indicates a lower death rate versus placebo or other arm. ITT-E Population consisted of all participants in the Safety Population (i.e. randomized to IP and who received at least one dose of IP), with the exception of those recruited at sites that were closed.