Sleep Apnea in TIA/Stroke: Reducing Cardiovascular Risk With Positive Airway Pressure (Sleep Tight)
|ClinicalTrials.gov Identifier: NCT01446913|
Recruitment Status : Completed
First Posted : October 5, 2011
Last Update Posted : July 31, 2015
|Condition or disease||Intervention/treatment||Phase|
|Transient Ischemic Attack Stroke||Device: Standard CPAP Intervention Behavioral: Enhanced CPAP Intervention||Phase 2|
|Study Type :||Interventional (Clinical Trial)|
|Actual Enrollment :||255 participants|
|Intervention Model:||Parallel Assignment|
|Official Title:||Sleep Apnea in TIA/Stroke: Reducing Cardiovascular Risk With Positive Airway Pressure|
|Study Start Date :||May 2011|
|Primary Completion Date :||December 2013|
|Study Completion Date :||April 2014|
Active Comparator: Standard Intervention group
This group gets unattended sleep study, auto titrating CPAP, and standard CPAP support.
|Device: Standard CPAP Intervention|
Active Comparator: Enhaced CPAP intervention
This group gets an unattended sleep study, autotitrating CPAP, and enhanced CPAP support.
|Behavioral: Enhanced CPAP Intervention|
No Intervention: Usual Care
This group usual care after TIA/stroke and a sleep study at the end of the study.
- To evaluate whether a diagnosis and treatment intervention strategy for sleep apnea results in a significant reduction in five domains of cardiovascular risk markers (inflammatory, autonomic, insulin resistance, endothelial, and atherosclerotic). [ Time Frame: Change from baseline to final measurements (6-12 months) ]The primary outcomes will include: (a) the impact of CPAP on pathophysiologic markers in the following domains of cardiovascular risk: inflammation (CRP, Il-6), heightened sympathetic activity/parasympathetic withdrawal (plasma catecholamine and heart rate variability (HRV)), insulin resistance (HOMA-IR), endothelial injury (flow mediated vasodilation), and atherosclerosis (carotid intima-media thickness). This analysis will include both an intention to treat analysis for all randomized patients as well as a pre-specified subgroup analysis of only patients with sleep apnea.
- To determine the level of CPAP adherence that corresponds to greatest improvements in markers of cardiovascular risk. [ Time Frame: Change from baseline to final measurements (6-12 months) ]The comparison of change in biomarkers at the time of final measurement between CPAP adherence groups will be restricted to intervention patients with sleep apnea (both the enhanced protocol and the standard protocol patients). We will use mixed models with random effect of patients to test for differences among three CPAP adherence groups (good CPAP use, some CPAP use and no CPAP use, adjusting for follow-up length (in months) (possible interaction term of follow-up length and CPAP adherence groups), site, and TIA/stroke strata. If the overall estimate of the effect for adherence group is statistically significant, we will further assess the differences between pairs of groups (good CPAP use vs. some CPAP use, some CPAP use vs. no CPAP use, and good CPAP use vs. no CPAP use).
- To determine whether an enhanced intervention protocol (targeted education, customized cognitive intervention, increased CPAP support) will result in improved long-term CPAP adherence rates among patients with sleep apnea. [ Time Frame: baseline to final measurements (6-12 months) ]This analysis will be restricted to intervention patients with sleep apnea. We will compare the proportion of patients with good CPAP use in the enhanced versus standard protocols, using logistic regression adjusting for follow-up length (in months) (possible interaction term of follow-up length and treatment), site, and TIA/stroke strata. We will report the observed proportions with 95% confidence intervals.
- To collect the vascular event rate as pilot data for a future effectiveness strategy trial of CPAP among patients with TIA and stroke to reduce cardiovascular events and death. [ Time Frame: Baseline to final measurements (6-12 months) ]We will report the observed vascular event rates (stroke, acute coronary syndrome, and all-cause mortality) in person years (and with 95% confidence intervals) in the intervention patients (both the enhanced protocol and the standard protocol patients) and the control patients. We will not compare these rates with stochastic testing given that this study has not been designed to provide adequate power for this comparison.
- Examine the relationship between CPAP use and various domains of cardiovascular risk markers and the impact of sleep duration and sleep apnea severity. [ Time Frame: baseline to final assessment (6-12 months) ]This analysis will be restricted to (definite TIA/Stroke) intervention patients with sleep apnea. We will begin this exploratory analysis by graphing the association between CPAP use (on the x-axis) and the change in the cardiovascular markers (on the y-axis). For these graphical analyses, we will use the following measures of CPAP use: (a) total cumulative hours of CPAP use over the study period (continuous variable in hours); (b) the mean observed hours of CPAP use per night divided by the estimated average hours of sleep per night (continuous variable that ranges from 0 to 100%) (Figure 11). To establish the relationship between CPAP use and cardiovascular markers we may then engage in regression modeling, controlling for site, and TIA/stroke strata. Analyses will be conducted for all subjects, and by AHI severity groups (5-15, 15-30, ≥ 30).
- 24 hour blood pressure measurements [ Time Frame: basleine to final assessments (6-12 months) ]We will use similar ANCOVA as in primary analysis plan to analyze: (a) mean 24-hour systolic, diastolic BP; (b) the defined daily dose (DDD) of antihypertensive medications (which is defined as the average maintenance dose per day for a drug used for its main indication in adults (http://www.whocc.no/atcdd/)); (c) medication-adjusted 24-hour SBP, which is calculated as [mean 24-hour SBP in mmHg] + [patient's DDD × (8.0 mmHg). We will also use a logistic regression model for the binary outcomes: (a) the patients with a dipping in systolic or diastolic BP ≥ 10% at follow-up or not; and (b) patients with a decrease from baseline to follow-up in 24-hour SBP of 5 mmHg or not, with the independent variables of baseline BP measurements, follow-up time, treatment arms, possible interaction term of time and treatment arms, and baseline characteristics which are clinically different between groups at baseline (p < 0.05).
- Patient-reported measurements [ Time Frame: baseline to final assessments (6-12 months) ]The analysis will be restricted to patients with sleep apnea. For patient-reported measurements (NIH stroke scale for all, stroke patients and TIA patients, modified Rankin score, Epworth sleepiness scale, and PHQ8), results are summarized by means (standard deviations) and medians (minimum, maximum) will be used. Mixed models will be used where the dependent variables are the measurements baseline and follow-up, and the independent variables are follow-up length (in months), treatment arm and possible interaction term of follow-up length (in months) and treatment arm adjusting for baseline measurement, site and TIA/stroke strata.
- Safety Analyses [ Time Frame: baseline to final assessments (6-12 months) ]All adverse events will be reported regardless of severity and relationship to CPAP. The incidence of adverse events will be summarized with frequency tables by three treatment arms and receiving CPAP or not.
Please refer to this study by its ClinicalTrials.gov identifier (NCT number): NCT01446913
|United States, Connecticut|
|New Haven, Connecticut, United States, 06511|
|United States, Indiana|
|University of Indiana|
|Indianapolis, Indiana, United States, 46202|
|Principal Investigator:||Henry Yaggi, MD,MPH||Yale University|
|Principal Investigator:||Dawn M Bravata, M.D.||Indiana University School of Medicine|