IIR 06-233
Diagnosis and Treatment of Sleep Apnea in Cerebrovascular Disease
Dawn M. Bravata, MD Richard L. Roudebush VA Medical Center, Indianapolis, IN Indianapolis, IN Funding Period: April 2008 - September 2011 Portfolio Assignment: Care of Complex Chronic Conditions |
BACKGROUND/RATIONALE:
Sleep apnea is a risk factor for stroke, is present in at least half of patients who survive a stroke or transient ischemic attack (TIA), and is associated with recurrent stroke and death. Continuous positive airways pressure (CPAP) is a safe and effective treatment for sleep apnea. Among patients with stroke and TIA, treatment of cerebrovascular risk factors, such as hypertension, is a major clinical priority. CPAP lowers blood pressure modestly among patients with sleep apnea. This background provides a rationale for treating sleep apnea among veterans with cerebrovascular disease. Unfortunately, few patients with cerebrovascular disease are screened for sleep apnea or referred for treatment with CPAP. Traditional laboratory-based testing may represent one potential barrier in a population of post-stroke veterans. The "Diagnosis and Treatment of Sleep Apnea in Cerebrovascular Disease" (Go To Sleep) study was designed to provide timely diagnosis and treatment of sleep apnea by using ambulatory home-based polysomnography and auto-titrating continuous positive airway pressure (CPAP). OBJECTIVE(S): The primary objectives were to determine whether a diagnostic and therapeutic intervention strategy among veterans with cerebrovascular disease and hypertension improved: (1) detection of sleep apnea; (2) appropriate treatment for sleep apnea; and (3) control of hypertension. Twenty-four hour blood pressure assessments were made at baseline and at the end of the one-year study period for both groups. Antihypertensive medications and their doses were recorded at the time of the 24-hour blood pressure measurements and converted into total antihypertensive defined daily dose (DDD). The medication-adjusted systolic blood pressure (in mm Hg) was calculated as follows: (mean systolic blood pressure + (DDD x 8.0)). CPAP adherence was categorized as follows: "none," zero use; "poor," <10% of total nights used; "some," 10% of total nights used but cumulative hours used <4 hours for 70% of nights; "excellent," cumulative hours used 4 hours for 70% of nights. Secondary aims included identifying predictors of the presence of sleep apnea among this cohort of veterans with cerebrovascular disease and to compare the total healthcare costs for veterans in the intervention versus control groups. Supplemental funding was obtained to collect qualitative data from participants in the randomized controlled trial. The primary objective of the project supplement was to improve understanding of how veterans make the decision to adhere to CPAP. METHODS: This prospective, multi-site, randomized, controlled strategy trial randomly assigned veterans with cerebrovascular disease and hypertension to an intervention or control group. Intervention group patients received unattended polysomnography at baseline and those with sleep apnea received auto-titrating continuous positive airway pressure (CPAP) therapy for up to one year. Control group patients received usual care and unattended polysomnography at the end of the study to identify the rate of undiagnosed sleep apnea. Qualitative interviews were conducted at the time of sleep apnea diagnosis and after 30 days of CPAP therapy on a sub-sample of patients. FINDINGS/RESULTS: A total of 225 patients were randomized, 115 control and 110 intervention patients. Approximately three-quarters of the patients had a prior stroke as the index cerebrovascular event (68.2% of intervention and 75.7% of control, p=0.21) and the average time from this event was 6.23 years (+/- standard deviation, 7.44) for intervention patients and 7.25 years (+/- 9.15) for control patients (p=0.45). The two groups were similar with regard to body mass index (intervention, 38.5 +/-12.9; control 39.6 +/-12.9; p=0.45) and excessive daytime sleepiness (based on a score of 10 on the Epworth Sleepiness Scale; intervention, 32.4%; control, 24.14%, p=0.16). The patients had relatively well controlled hypertension at baseline: the 24-hour mean systolic blood pressure was 134 mmHg (+/-14) in the intervention group and 136 mm Hg (+/-13) in the control group (p=0.32). Overall, 56% of patients in this study had obstructive sleep apnea, 6% had central sleep apnea, and 38% had no sleep apnea. The intervention strategy was successful in diagnosing sleep apnea: among patients with sleep apnea, 100% were diagnosed in the intervention group whereas only 9.4% of control patients received a diagnosis as part of usual care; therefore 90.6% (48/53, 95%CI 82.7-98.4%) of sleep apnea was undiagnosed in the control group. In general, patients in this study had moderately severe sleep apnea with a mean Apnea Hypopnea Index (AHI) of 15.0 events/hour (+/-16.6) in the intervention group and 13.6 events/hour (+/-15.7) in the control group. Daytime sleepiness and the Berlin questionnaire were not useful in predicting sleep apnea prevalence in this cohort, however, patients with sleep apnea did have a larger neck circumference: 41.9 cm (+/-3.4) versus 40.8 (+/-3.3; p=0.045). The intervention strategy successfully improved the rate of sleep apnea treatment: 66.7% of intervention patients versus 2.1% of control patients had some or excellent use (64.6% improvement, 95%CI 51.7-77.5%). The intervention strategy did not significantly improve blood pressure control among this cohort of patients with very well controlled hypertension at baseline, the final mean 24-hour systolic blood pressure was nearly identical in the two groups: intervention, 133 mm Hg (+/-14); control, 134 mm Hg (+/-14; p=0.48). Among patients with sleep apnea, excellent CPAP use was associated with an improvement in the medication-adjusted systolic blood pressure, but this finding was not statistically significant: not-excellent CPAP use, increase of 1 mm Hg (+/-17), excellent CPAP use, decrease of 3 mm Hg (+/-16; p=0.45). Twenty three patients were interviewed at baseline and 22 at 30-days post-CPAP delivery. Key findings from these interviews suggest that there are numerous avenues to intervene to improve adherence to CPAP. Some of these include veterans' perceptions of their CPAP diagnosis (e.g., denial, surprise, acceptance), their initial impressions of using the CPAP machine and how it might either integrate or interfere with their existing sleep habits, and veterans' understandings of how CPAP use might benefit them (e.g., reducing risk factors for subsequent vascular events, improved energy, improved mental acuity). Of particular importance seemed to be the face-to-face dialogue that took place in veterans' homes between themselves and field staff in which veterans could candidly discuss their questions and concerns about using CPAP, and field staff could respond to these concerns in an effort to increase motivation and adherence to CPAP. This was valuable with respect to preemptively addressing potential problems and obstacles to CPAP use and creating a positive association in the veteran's mind between CPAP use and outcomes that were personally important to the individual veteran (e.g., avoiding another stroke, spending time with grandchildren, being more independent, not sleeping during the day). IMPACT: These results confirm that patients with cerebrovascular disease have a very high prevalence of sleep apnea (over half of the cohort had sleep apnea). Traditional methods of screening for sleep apnea are not useful in this population (e.g., reliance upon reports of daytime sleepiness). A home based diagnostic and treatment intervention strategy using portable polysomnography and auto-titrating CPAP in patients' homes substantially improves both the diagnosis and treatment rates for sleep apnea compared with usual care. Our data suggest that over 90% of veterans with TIA/stoke are currently undiagnosed with sleep apnea. These data also suggest that the very low rate of observed CPAP adherence in usual care (2%) could be markedly improved (to 67%) with an in-home management strategy. These results suggest that all veterans with a history of cerebrovascular disease should receive polysomnography post-cerebrovascular event to make the diagnosis of sleep apnea including those without prominent symptoms of daytime sleepiness. Moreover, the strategy of using unattended polysomnography and auto-titrating CPAP treatment is an effective strategy for delivering this care and does not require a facility to have an on-site sleep laboratory. In this way, these results are potentially relevant across the full spectrum of VAMCs: those with existing sleep laboratories and those with less developed sleep medicine programs. The high prevalence of sleep apnea, known links with adverse outcomes, and known benefits of CPAP suggests that the potential public health impact of these findings in this population is significant. External Links for this ProjectDimensions for VADimensions for VA is a web-based tool available to VA staff that enables detailed searches of published research and research projects.Learn more about Dimensions for VA. VA staff not currently on the VA network can access Dimensions by registering for an account using their VA email address. Search Dimensions for this project PUBLICATIONS:Journal Articles
DRA:
Aging, Older Veterans' Health and Care, Health Systems Science, Cardiovascular Disease, Other Conditions
DRE: Treatment - Comparative Effectiveness, Diagnosis, Treatment - Observational, Prevention Keywords: Cardiovasc’r disease, Sleep disorders, Stroke MeSH Terms: Sleep Disorders |