Examining Vascular Outcomes in COPD Inpatients (AIM 1) Examining Early Rehabilitation on Discharged COPD Patients (AIM 2)

AIM 1:

Purpose: To examine pulmonary and cardiovascular outcomes in COPD patients from the time of their acute hospital admission until discharge.

Rationale: AECOPD is associated with systemic inflammation, prolonged bed-rest and increased CV risk. Proper medical management strategies such as appropriate systemic corticosteroids, mobilization, and bedside education may reduce systemic inflammation and improve vascular function, while improving patient self-efficacy and health outcomes and reducing hospital length of stay (LOS).

Hypothesis: Treatment-specific factors such as pharmacological management, patient education and inpatient mobilization will affect LOS, as well as pulmonary and cardiovascular outcomes at discharge.

Study Design: For this cross-sectional study, consecutive patients admitted to the University of Alberta Hospital Pulmonary ward with a diagnosis of an AECOPD (with BNP <500 pg/ml and Troponin <1.0 mcg/L), not requiring ventilatory support in the form of non-invasive ventilation (NIV), will be approached for recruitment. In-hospital patient management will be left to the discretion of the admitting physician, and will follow the new AHS admission order sets for AECOPD, which are based on current guidelines1. Dr Bhutani assisted in the development of these order sets. The order sets standardize pharmacological and non-pharmacological management of AECOPD; however, variance in patients need and time to delivery of these interventions will be present. Treatment will not be otherwise altered during this study, except the timing of the administration of short acting bronchodilators as it relates to the measurement of arterial stiffness and vascular function. 24 hours after recruitment, the research team will begin collecting patient/clinical treatment information on a daily basis (see Appendix B for list). 48 hours after admission, patients will be given an activity monitor to quantify daily physical activity. On the same day, vascular assessment through the measurement of pulse wave velocity, vascular function, lung function as well as serum markers of systemic inflammation (TNF-alpha, MMP-2, IL-6 and CRP) and exhaled nitric oxide (eNO), will be collected (this is in addition to usual blood work required for patient management). This will be repeated on days 5, 10, day of discharge and day 14 post discharge. Self efficacy and a 6-minute walk test will be performed on the day of discharge and on day 14 following discharge, while three-day physical activity will be determined at 14 days post discharge. BODE index will be determined at discharge and at 14 days post discharge (see Appendix C for data collection schedule).

Subject Selection \& Recruitment: All Aims will be registered at (www.clinicaltrials.gov), submitted for approval through the University of Alberta Health Research Ethics Board (HREB), and informed consent will be obtained from each study patient.

Since most of patients with COPD who are hospitalized initially present to, and are assessed in the Emergency Department (ED), they will be recruited from this setting by the ED study team. All patients with a diagnosis of AECOPD who are admitted to the pulmonary ward, meeting the above criteria, will be screened for enrollment. If entered into the study, baseline information (Appendix A) will be collected at that time. Through collaborative and primary ED research activity, Dr Rowe's Emergency Medicine Research Group maintains research nurse coverage for the U of A hospital ED from 07:00 - 23:00 Monday - Friday and 10:00 - 20:00 Saturday and Sunday. These research nurses act as the main recruitment source within the University of Alberta Hospital ED and this site has been a leading Canadian recruitment centre in a number of clinical research studies. Eligible patients with AECOPD will be approached to enter the study as required and informed written consent will be obtained. A Refused, Missed or Otherwise (RMO) excluded database will be maintained to determine the feasibility of a larger study and to assess the generalizability of the enrolled sample to all cases. In the event that a patient presents after research staff hours, the patient will be approached by the research staff to discuss the study and obtain consent and begin collection of all data. AHS data indicates that ~450 COPD patients are admitted at the U of A hospital each year. Dr Rowe's research team has demonstrated success at recruiting through the ED120,121, and therefore recruiting 100 patients over 2 years is feasible

Data Handling: Upon enrollment a data collection form (Appendix A) will be completed for all patients by the ED research team. The research team will collect patient/clinical treatment information daily (see Appendix B), and outcome data per protocol (see Appendix C).

Data Analysis: All data will be entered into a custom-developed secure anonymized database. Data analyses will be performed using StataCorp. 2009. Stata Statistical Software: Release 11 (College Station, TX: StataCorp LP). Continuous data will be reported as means and standard deviations (SD) or median and interquartile ranges (IQR). Chi-square testing will be used for bi-variable analyses of dichotomous variables; continuous variables will be compared using t-tests or Mann-Whitney tests. Two-tailed results p < 0.05 will be considered statistically significant. The influence of patient characteristics and treatment factors on the dependent variables (i.e. inflammation, vascular function, self efficacy and LOS) will analyzed through a multiple linear or logistic regression models as appropriate. For the linear regression model, clinically relevant and statistically significant (at the p < 0.1 level) independent variables will be included. For the logistic regression model, several dependent variables will be explored; clinically relevant and statistically significant (at the p < 0.1 level) independent variables will be included.

Sample size: A convenience sample of COPD patients will be recruited and efforts will be made to provide reasonable confidence intervals on important variables and outcomes; a minimum sample size of 100 COPD cases will be collected. This will be sufficient for multiple regression analyses including up to 10 predictors and would provide narrow 95% confidence intervals (CI) for variable estimates of ~10% for mid-range variables and 6%. for extreme-range variables.

AIM 2:

Purpose: To examine the impact of early PR following hospital discharge on QoL, pulmonary/CV outcomes and AECOPD hospitalizations.

Rationale: In addition to typical improvements in QoL and exercise tolerance, studies have shown that PR increases self-efficacy, physical activity while reducing CV risk in stable COPD patients. Patients recently discharge from hospital following AECOPD represent the sickest patients with greatly reduced QoL, exercise tolerance, self-efficacy and physical activity and greatly elevated CV risk. Exactly how these improve with PR reduce CV risk and hospitalizations following PR requires examination.

Hypothesis: Patients who receive early PR will have improved QoL, pulmonary/CV outcomes and less hospitalizations for COPD in the 6 months following hospital discharge. PR will improve self efficacy, physical activity and QoL while reducing CV risk as compared to usual care.

Study Design \& Subject Recruitment: All patients admitted to the pulmonary ward for an AECOPD, including those who have completed Aim 1, will be offered participation into this arm of the study. Patients found to have an acute cardiac injury during admission, mobility issues or residence outside the greater Edmonton area will be excluded. Consenting patients will be subsequently randomized into one of two groups: early PR versus usual care. Patients randomized to early PR will be enrolled within 1 month of discharge into a PR program, while usual care patients will be followed-up by their most responsible physician as determined by the admitting team. The PR group will be enrolled in the Breathe Easy Program at the Center for Lung Health, and will proceed through the program in a typical fashion. All patients will be followed up 6 months after discharge and will be interviewed to assess disease status, management review and if there has been a history of recurrence or relapse of the AECOPD. Hospital admissions and length of stay data will be obtained through electronic medical records. Patient assessments will include: quality of life, 6min walk, dyspnea, self-efficacy, physical activity, pulse wave velocity, vascular function, systemic inflammation (TNFα, MMP-2, IL-6 and CRP) and FeNO. All data will be collected before, immediately after and 6month after PR. The control group will have the same data collected at the same scheduled time. See above for descriptions of methods.

Data Handling: Data will be entered onto a secure anonymized database.

Data Analysis: The influence of PR on QoL, 6min walk, dyspnea, self-efficacy, physical activity, pulse wave velocity, vascular function, systemic inflammation and eNO will be analyzed using a multivariate mixed-model MANOVA with treatment (PR vs. usual care) being a fixed between-group variable and time (pre, immediate post, 6months post) as a repeated variable.

Sample size: Based on previous work84,95,96,101-105, a sample size of 50 in each group (100 total) will be sufficient to detect a between-group differences in QoL, 6min walk, PWV, dyspnea and hospital readmission rates following PR (α=0.05, β=0.8). Based on our recent work, this sample could detect a 10% difference in physical activity following PR (α=0.05, β=0.8). One hundred patients will also to allow for stratification of physiological and psychological responses with PR.