Continuous Positive Airway Pressure: Evaluation of a Novel Therapy for Patients with Acute Ischemic Stroke

Transcription

1 EVALUATION OF FOR PATIENTS WITH ACUTE ISCHEMIC STROKE doi: /sleep.1254 Continuous Positive Airway Pressure: Evaluation of a Novel Therapy for Patients with Acute Ischemic Stroke Dawn M. Bravata, MD 1,2,4 ; John Concato, MD 5-7 ; Terri Fried, MD 5,7 ; Noshene Ranjbar, MD 9 ; Tanesh Sadarangani, MD 5 ; Vincent McClain, MD 5 ; Frederick Struve, PhD 5 ; Lawrence Zygmunt, BA 5 ; Herbert J. Knight, MD 13 ; Albert Lo, MD, PhD 10,11 ; George B. Richerson, MD, PhD 15 ; Mark Gorman, MD 12 ; Linda S. Williams, MD 1,3,4 ; Lawrence M. Brass, MD 5,8,14 *; Joseph Agostini, MD 5-7 ; Vahid Mohsenin, MD 7 ; Francoise Roux, MD, PhD 7 ; H. Klar Yaggi, MD Center of Excellence on Implementing Evidence-Based Practice, Richard L. Roudebush VA Medical Center, Indianapolis, IN; 2 Department of Internal Medicine, Indiana University School of Medicine, Indianapolis, IN; 3 Department of Neurology, Indiana University School of Medicine, Indianapolis, IN; 4 Regenstrief Institute, Indianapolis, IN; 5 Clinical Epidemiology Research Center, VA Connecticut Healthcare System, West Haven, CT; 6 Medicine Service, VA Connecticut Healthcare System, West Haven, CT; 7 Department of Internal Medicine, Yale University School of Medicine, New Haven, CT; 8 Department of Neurology, Yale University School of Medicine, New Haven, CT; 9 Research Service, VA Connecticut Healthcare System, West Haven, CT; 10 Providence Veterans Administration Hospital, Providence, RI; 11 Department of Neurology, Brown University School of Medicine, Providence, RI; 12 Department of Neurology, University of Vermont College of Medicine, Burlington, VT; 13 Hospital of St. Raphael, New Haven, CT; 14 Department of Neurology, VA Connecticut Healthcare System, West Haven, CT, 15 Department of Neurology, University of Iowa, Iowa City, IA Background: New approaches are needed to treat patients with stroke. Among acute ischemic stroke patients, our primary objectives were to describe the prevalence of sleep apnea and demonstrate the feasibility of providing auto-titrating continuous positive airway pressure (auto-). A secondary objective was to examine the effect of auto- on stroke severity. Methods: Stroke patients randomized to the intervention group received 2 nights of auto-, but only those with evidence of sleep apnea received auto- for the remainder of the 30-day period. Intervention patients received polysomnography 30 days post-stroke. Control patients received polysomnography at baseline and after 30 days. Acceptable auto- adherence was defined as 4 h/night for 75% nights. Change in stroke severity was assessed comparing the NIH Stroke Scale (NIHSS) at baseline versus at 30 days. Results: The 2 groups (intervention N = 31, control N = 24) had similar baseline stroke severity (both median NIHSS, 3.0). Among patients with complete polysomnography data, the majority had sleep apnea: baseline, 13/15 (86.7%) control patients; 30 days, 24/35 (68.6%) control and intervention patients. Intervention patients had greater improvements in NIHSS ( 3.0) than control patients ( 1.0); P = Among patients with sleep apnea, greater improvement was observed with increasing auto- use: 1.0 for control patients not using auto-; 2.5 for intervention patients with some auto- use; and 3.0 for intervention patients with acceptable auto- adherence. Conclusions: The majority of acute stroke patients had sleep apnea. Auto- was well tolerated, appears to improve neurological recovery from stroke, and may represent a new therapeutic approach for selected patients with acute cerebral infarction. Keywords: Acute ischemic stroke, sleep apnea, continuous positive airway pressure Citation: Bravata DM; Concato J; Fried T; Ranjbar N; Sadarangani T; McClain V; Struve F; Zygmunt L; Knight HJ; Lo A; Richerson GB; Gorman M; Williams LS; Brass LM; Agostini J; Mohsenin V; Roux F; Yaggi HK. Continuous positive airway pressure: evaluation of a novel therapy for patients with acute ischemic stroke. SLEEP 2011;34(9): INTRODUCTION Sleep apnea is common following stroke, occurring in 60% to 96% of post-stroke patients Sleep apnea, in the setting of an acute stroke, is associated with early neurological worsening, decreased functional recovery, and increased mortality. 5,6,11,12 Continuous positive airway pressure () safely and effectively treats sleep apnea Among stroke patients during rehabilitation, has been associated with improved outcomes (e.g., reduced depressive symptoms, improved sense of well-being) without serious side effects The safety and effectiveness of in the acute stroke period remains unknown. *Dr. Brass is deceased. Submitted for publication November, 2010 Submitted in final revised form April, 2011 Accepted for publication April, 2011 Address correspondence to: Dawn M. Bravata, MD, Richard L. Roudebush VA Medical Center, Center of Excellence on Implementing Evidence-Based Practice, HSR&D; Mailcode: 11H, 1481 West 10th Street, Indianapolis, IN 46202; Tel: (317) ; Fax: (317) ; dbravata@iupui.edu In general, interventions aimed at improving post-stroke functioning are more effective the earlier they are delivered after symptom onset. Auto-titrating (auto-) senses airflow limitation and delivers pressure in response to apneas and hypopneas. Auto- is equally efficacious compared with fixed-pressure, but it eliminates the need for performing titration during polysomnography making it especially useful in the acute setting. 20,21 The primary objectives of this study were to describe the prevalence of sleep apnea among stroke patients (at baseline and after 30 days) and to evaluate the feasibility of providing auto- to acute ischemic stroke patients. A secondary objective was to examine the effect of auto- on stroke symptom severity. METHODS Sample Patients with acute ischemic stroke were recruited from 3 sites in Connecticut. Patients were included if they had an ischemic stroke, 22 an NIH Stroke Scale (NIHSS) score 2, and age 50 years. Patients were excluded if they had a prior diagnosis of sleep apnea, respiratory distress requiring mechanical SLEEP, Vol. 34, No. 9, Therapy for Patients with Acute Ischemic Stroke Bravata et al

2 Stroke Patients Baseline: 72 hours of stroke Randomize Figure 1 Research design Intervention [Auto-] Control [Sleep Study] Sleep Apnea Present Sleep Apnea Absent Sleep Apnea Present Sleep Apnea Absent After 2 nights Protocol for Intervention Patients All intervention patients received auto-. The auto- machine (AutoSet Spirit, ResMed, Poway, CA) measured and stored information about: respiratory events (apneas and hypopneas), patient use (minutes used, nights used), mask leak, and pressure delivered. Auto- adherence was catventilation, oxygen dependent chronic obstructive pulmonary disease, pregnancy, intracranial hemorrhage, time from symptom onset to emergency department arrival > 72 h, receipt of thrombolytic therapy, life expectancy < 6 months, non-english speaking status, or residence outside of Connecticut. Atrial fibrillation and age < 65 years were both originally included as exclusion criteria in our initial human subjects protocol. After the first several months of recruitment, we examined our screening data and observed that many patients were being excluded on the basis of atrial fibrillation and age. The research team discussed the exclusions and made 2 modifications: a reduction in the minimum age from 65 to 50 years (given that there is no rationale to support the contention that the treatment of sleep apnea should have a differential effect on post-stroke outcomes based on age) and elimination of atrial fibrillation as an exclusion (given that sleep apnea is associated with arrhythmias, and hence treatment of sleep apnea might improve outcomes for patients with atrial fibrillation). Randomization This feasibility study was designed as a randomized controlled trial (clinical trials.gov; NCT ). The randomization was stratified according to NIHSS < 10 or 10, to ensure balance in the groups in terms of baseline stroke severity. Randomization was performed in a 1 (control):1.5 (intervention) ratio, to ensure a sufficient number of patients in the intervention group, given that one of the primary outcomes was related to auto- use. Baseline Measurements Baseline assessments included a review of the medical record, an interview of patients and/or their proxies, and physical measurements. The following data were collected at baseline: demographic factors, comorbidities, medications, blood pressure, heart rate, neck circumference, height, weight, and concomitant stroke care (e.g., antithrombotic agents, assessment of hyperlipidemia or cholesterol-lowering medication). Stroke symptom severity was measured prospectively using the NI- HSS. 23 The NIHSS assesses 11 domains of neurological function, where normal functioning is given a score of zero and higher scores indicate worse stroke severity. In general, points Stop Sleep Study Sleep Study Outcome: 30 d post-stroke are added for additional domains or extent of impairment; a score 2 is generally considered mild, 2-9 moderate, and 10 severe. The NIHSS is the standard measure of stroke severity that is used in stroke clinical trials. 24 The NIHSS was assessed by research staff who received certification in administration of the NIHSS. 25 Protocol for Control Patients Control patients received portable unattended polysomnography at baseline and at 30 days post-stroke (Figure 1). The primary care providers of the control patients were notified of the polysomnography results, but control patients were not given as part of the study. Although it was theoretically possible for control patients with sleep apnea to receive, none of them actually received during the one-month study period. Polysomnography was performed by a trained staff member at the patient s bedside. The patients were not moved off their unit if they were admitted to the hospital. The polysomnography included: electroencephalogram (EEG), respiratory inductive plethysmography (RIP), body position, electrocardiogram (ECG), blood pressure, and oxygen saturation. 26 We used portable unattended polysomnography (LifeShirt, Vivometrics, Ventura, CA) because it has been validated against the gold standard of polysomnography in a sleep laboratory, has been successfully used in clinical trials of sleep apnea, and can be used throughout the hospital or at a patients residence. 27 Diagnosis of Sleep Apnea Apnea was defined as an airflow cessation 10 sec; hypopnea was defined as a reduction in airflow 10 sec or decrease in amplitude of breathing 30% followed by an oxygen desaturation 4% The apnea-hypopnea index (AHI) was calculated from the sum of the number of apneas and hypopneas per hour of sleep. Sleep apnea was diagnosed when the AHI 5 events/ hour. 29 Respiratory events in the presence of abdominal effort were scored as obstructive, whereas respiratory events in the absence of abdominal effort were scored as central events. A respiratory event was scored as a mixed apnea if it met apnea criteria and was associated with absent inspiratory effort in the second portion of the event. For the purposes of distinguishing between obstructive and central sleep apnea, mixed apneas were classified as obstructive respiratory events. Patients were classified has having Cheyne-Stokes respiration if the predominant breathing pattern (> 50%) of respiratory events took the form of a waxing and waning ventilatory effort with arousals typically occurring at the peak of the hyperpnea phase. SLEEP, Vol. 34, No. 9, Therapy for Patients with Acute Ischemic Stroke Bravata et al

3 egorized as: no use; any use (more than no use but < 4 h/night for < 75% of nights); or acceptable adherence ( 4 h/night for 75% of nights). 31,32 The machine was interrogated after 2 nights. If there was no evidence of airflow limitation (AHI < 5 or median pressure < 6 cm H 2 O), the auto- was discontinued. If there was evidence of airflow limitation, the auto- was continued for the remainder of the 30-day period (Figure 1). To support intervention patients efforts to use the auto-, patients received information about sleep apnea, the benefits of treatment, and instructions about machine use daily for 2 days and then weekly for the rest of the one-month study period. Research staff included research associates (not respiratory therapists) who were trained in polysomnography and administration and support specifically for this study. They made visits to the patients (either in the hospital or the patients homes) to refit masks or to troubleshoot equipment as needed. Whenever possible, the staff showed the patients their own auto- data to encourage adherence. Because polysomnography (rather than auto-) is the gold standard method of diagnosing sleep apnea, intervention patients received unattended portable polysomnography at 30 days post-stroke (Figure 1). Outcomes The primary outcomes were: (1) sleep apnea prevalence at baseline (based on the polysomnography data from the control patients), and (2) the proportion of intervention patients with sleep apnea with acceptable adherence. The secondary outcome was change in stroke symptom severity (NIHSS) calculated as the differences between the baseline and 30-day NI- HSS scores (a negative number indicates improvement). Adverse Events The definition of recurrent vascular events included transient ischemic attacks, recurrent stroke, myocardial infarction, hospitalization for congestive heart failure, or any death. 33 All adverse events were reviewed by a panel that included an internist, a neurologist, and a pulmonologist. The panel classified each adverse event as serious or non-serious and as related or unrelated to the study intervention. Analysis Plan Descriptive statistics (e.g., mean with standard deviations, medians, and proportions) were used to describe the baseline characteristics and outcomes. To compare differences in the outcome rates between control and intervention patients and among sleep apnea patients with varying auto- use, χ 2 tests or Fisher exact tests were used for binary or ordinal variables, and Student t-tests or Wilcoxon rank sum tests for dimensional variables. Two-sided p-values were used to evaluate differences between the intervention and control groups. A p-value of 0.05 was used to determine statistical significance. An intention-totreat analysis was performed comparing the change in median NIHSS from baseline to 30 days among intervention patients versus control patients. A secondary prespecified analysis of change in stroke severity was conducted among patients with sleep apnea according to the amount of used. This project was designed to enroll a sufficient number of subjects to ensure face validity for our feasibility assessments; it was not powered to identify differences in outcomes between patients in the intervention versus control groups. No imputations were made for missing data. The analyses were conducted using SAS 9.1 (Cary, NC). This study was approved by all of the participating sites human subjects committees. RESULTS Among 955 screened patients, 136 did not have an ischemic stroke, 620 met at least one exclusion criterion, and 144 patients refused to participate (Figure 2). Therefore, 55 patients were enrolled, with 31 in the intervention group and 24 in the control group. Three control patients (12.5%) and 9 intervention patients (29.0%) withdrew (Figure 2). No statistically significant differences were found in the baseline characteristics of the 2 groups, although the control patients were somewhat heavier (mean weight of versus pounds; Table 1). No clinically or statistically significant differences were observed among patients who withdrew compared with patients who remained in the study (data not shown). The majority of patients received the auto- within 48 h of symptom onset: < 24 h, 5/31 (16.1%); 24 < 48 h, 16/31 (51.6%); and 48 h, 10/31 (32.3%). The control and intervention groups were very similar with respect to the concomitant stroke care they received (data not shown). Sleep Apnea Prevalence Among the 24 control patients, 15 (63%) had polysomnography completed at baseline and 18 (75%) at 30 days poststroke. The main reasons for not being able to complete the polysomnography were: competing diagnostic testing (e.g., patients being sent for brain imaging), poor quality EEG recordings obtained in the inpatient setting, and inability to obtain permission to perform the polysomnography in rehabilitation units. The majority of patients had sleep apnea both at baseline (13/15 [86.7%]) and one month post-stroke, (24/35 [68.6%]; Table 2). The control and intervention groups had similar rates of sleep apnea based on polysomnography at 30 days (72% versus 65%, P = 0.63). Obstructive sleep apnea was the predominant respiratory event, central apneas were rare; no patient had either pure central sleep apnea or Cheyne-Stokes respiration. Among the 14 intervention patients with both auto-titrating and polysomnography data, the diagnostic agreement was: agreement in 12/14 (86%), with a positive predictive value of 75% and a negative predictive value of 90%. Among the 2 patients with disagreement, one had evidence of obstruction by the machine (AHI 0.8, pressure 8.6 cm H 2 O), whereas the polysomnography demonstrated an AHI of 2.2. The second patient had no evidence of obstruction from the machine (AHI 4.4, pressure 4.8 cm H 2 O), whereas polysomnography demonstrated an AHI of 5.5. Auto- Use A total of 16 intervention patients were deemed eligible for, including 15 patients with either an AHI of 5 or a median pressure 6 cm H 2 O (i.e., the patients who were recommended to use auto- for 30 days), plus one patient who insisted upon using auto- machine because it made him feel better. (Note: for this patient who insisted upon SLEEP, Vol. 34, No. 9, Therapy for Patients with Acute Ischemic Stroke Bravata et al

4 955 patients screened for possible eligibility from: Emergency Departments, local physician offices and inpatient wards 136 patients not eligible: Not an ischemic stroke 764 patients potentially eligible: 144 Refused 620 Excluded Unable to provide informed consent (n = 151) Age < 50 years (n = 149) Atrial fibrillation (n = 138) Delayed presentation from symptom onset (n = 122) Life expectancy < 6 months (n = 53) Respiratory distress (n = 48) NIHSS < 2 (n = 35) Thrombolytic therapy (n = 34) Non-English speaker (n = 32) Non-local residence (n = 10) Oxygen dependent (n = 8) 55 patients randomized 24 Control Withdrew (n = 3) More important things in life (n = 2) Did not return telephone calls (n = 1) Available Polysomnography Data Baseline sleep study (n = 15) 90-day sleep study data (n = 18) Figure 2 Patient recruitment and retention diagram 31 Intervention Withdrew (n = 9) Does not want bother of study activity (n = 2) More important things in life (n = 2) Feeling depressed post-stroke (n = 2) Does not believe he had a stroke (n = 1) Did not return telephone calls (n = 1) No IRB approval at the nursing home (n = 1) Available Polysomnography Data 90-day sleep study data (n = 17) USE CATEGORIES Any Use (n = 16) Evidence of sleep apnea, therefore was recommended for 30-days (n = 15) No evidence of sleep apnea, therefore should have stopped after 2 nights, but patient insisted upon using (n = 1) No Use (n = 15) No evidence of sleep apnea, therefore stopped after 2 nights (n = 13) Patient refused any, therefore no recommendation possible (n = 2) continuing auto-, his beginning data suggested no obstruction (AHI 2.5, pressure 4.2 cm H 2 O), but his end- data were consistent with obstruction (AHI 5.4, pressure 4.8 cm H 2 O), as were his polysomnography data (AHI 15.7). All 16 patients had some use, with acceptable adherence observed in 10/16 (62.5%; Table 3). Two intervention patients refused to try auto-, therefore no recommendation for auto- use could be made. Effect of Auto- on Stroke Severity The intervention and control patients had similar stroke severity at baseline (both groups, median NIHSS of 3.0; Table 1). Intervention patients had greater median improve- ment in NIHSS ( 3.0) than control patients ( 1.0; P = 0.03; Table 4). Among stroke patients with sleep apnea, the improvement in the NIHSS was greatest among patients with the most auto- use (median NIHSS change: 3.0 for acceptable adherence, 2.5 for some use, and 1.0 for no use; Table 4). The greatest improvements in NIHSS were observed among patients who started auto-titrating within 48 h of stroke symptom onset: (median NIHSS change: 3.0, < 24 h; 3.0, 24 < 48 h; 0, 48 h). Adverse Events Serious adverse events were observed among 5/24 (20.8%) control patients and 2/31 (6.5%) intervention patients (P = 0.22). SLEEP, Vol. 34, No. 9, Therapy for Patients with Acute Ischemic Stroke Bravata et al

5 No serious adverse events were related to the intervention. Non-serious adverse events were observed among 1/24 (4.2%) control patient and 4/31 (12.9%) intervention patients (P = 0.37). Two patients had non-serious adverse events that were attributed to the study intervention including skin irritation due to the mask (n = 1) and sneezing/nasal irritation due to the (n = 1). DISCUSSION We found that diagnosing and treating sleep apnea with auto- among selected acute ischemic stroke patients is feasible. Our results indicate that sleep apnea is common in the acute stroke period, that acute stroke patients are able to use auto-, and that auto- may improve stroke symptom severity at 30 days post-stroke. Given that the majority of patients had sleep apnea and was eligible for treatment, relatively few are exposed to the intervention who do not have sleep apnea. Therefore, the approach of providing auto- without incurring the delays of a diagnostic sleep study provides the earliest treatment to the greatest number of patients with minimal risk. The prevalence of sleep apnea observed in this study is similar to the rates (60% to 96%) observed in prior studies We found that the prevalence of sleep apnea decreased from the acute period (86.7%) to one month post-stroke (68.6%). This decrement in sleep apnea prevalence with increasing time from the stroke event has been reported previously. 7 This observed pattern in prevalence supports the use of auto- to treat sleep apnea in the acute period. Auto-titrating is not recommended for use among patients with central sleep apnea, which has been demonstrated previously to be rare among patients with stroke and which was not observed in this relatively small study. Certain auto-titrating units use an algorithm to differentiate obstructive from central apneas by positive airway pressure. This detection algorithm works by delivering one or more pressure pulses when spontaneous breathing is not detected. The device evaluates the response of the patient s airway to the pressure pulses and identifies obstructive apneas or central apneas based on the resulting patient flow. The auto-titrating unit does not increase pressure in response to central apneas. These units, therefore, are appropriate for use in patients with obstructive or mixed sleep apnea, which is the cohort that was identified in the present study. Adherence with auto- in this study was similar to prior reports in non-stroke patients, but higher than studies of poststroke patients. For example, a study of stroke patients reported that 15% used in the long term. 9 Our staff visited patients to encourage use and to help them overcome any Table 1 Baseline characteristics (N = 55) Characteristic Intervention (N = 31) Control (N = 24) P-Value* Age (years): range Mean ± standard deviation 70.6 ± ± White race: N (%) 24 (77.4) 17 (70.8) 0.58 Female gender: N (%) 10 (32.3) 8 (33.3) 0.93 Previous stroke: N (%) 4 (12.9) 4 (16.7) 0.72 Past medical history: N (%) Hypertension 24 (77.4) 13 (54.2) 0.07 Congestive heart failure 3 (9.7) 4 (16.7) 0.69 Atrial fibrillation 8 (25.8) 3 (12.5) 0.31 Myocardial infarction 0 (0) 2 (8.3) 0.19 Hyperlipidemia 12 (38.7) 9 (37.5) 0.93 Chronic obstructive pulmonary disorder 1 (3.2) 2 (8.3) 0.57 Diabetes mellitus 7 (22.6) 6 (25.0) 0.83 Current tobacco smoking: N (%) 11 (35.5) 6 (27.3) 0.53 Blood pressure (mm Hg): Mean ± SD Systolic ± ± Diastolic 79.8 ± ± Neck circumference (inches): Mean ± SD 16.1 ± ± Waist circumference (inches): Mean ± 40.9 ± ± SD Weight (pounds): Mean ± SD ± ± Body mass index (kg/m 2 ): SD 26.8 ± ± Stroke severity (NIHSS): median (range) 3.0 (2-12) 3.0 (2-18) - Mean ± SD 4.1 ± ± Time from symptom onset to or 39.0 ( ) 39.5 ( ) - sleep study (h): median (range) Mean ± SD 38.7 ± ± < 24 h 5 (16.1) 1 (4.2) - 24 h, < 48 h 16 (51.6) 16 (66.7) - 48 h 10 (32.3) 7 (29.2) - *P-values refer to 2-sided p-values obtained either from Student t-tests, χ 2 tests, or Fisher exact tests. Table 2 Prevalence of sleep apnea* Characteristic Baseline (N = 15) 30-Day (N = 35) Apnea hypopnea index (events/h): range Mean ± SD 27.8 ± ± < 5 2 (13.3%) 11 (31.4%) 5 < 10 3 (20.0%) 4 (11.4%) 10 < 15 0 (0%) 1 (2.9%) 15 < 20 0 (0%) 4 (11.4%) 20 < 25 3 (20.0%) 1 (2.9%) 25 < 30 1 (6.6%) 3 (8.6%) 30 < 35 1 (6.6%) 3 (8.6%) 35 5 (33.3%) 8 (22.9%) Sleep apnea prevalence: N (%) 13/15 (86.7%) 24/35 (68.6%) *The prevalence of sleep apnea is based on data from the portable unattended polysomnography. The diagnosis of sleep apnea was made if the apnea hypopnea index was 5 events/hour. SLEEP, Vol. 34, No. 9, Therapy for Patients with Acute Ischemic Stroke Bravata et al

6 equipment-related difficulties, and our study had a relatively short duration; both factors may account for the observed higher rate of use. In our analysis of auto- adherence, we excluded patients who did not have sleep apnea and patients whose sleep apnea status was unknown because they refused auto- (a standard analytic practice). 9 Our finding that increasing adherence to auto- was associated with greater improvements in stroke symptom recovery supports the conclusion that auto- may provide a new therapeutic approach for ischemic stroke patients. Possible mechanisms by which auto- may improve neurological recovery include: improved oxygenation, reduced fibrinogen levels, decreased frequency of blood pressure surges associated with apneas, improved left ventricular function, and normalized cerebral vascular reactivity and cerebral blood flow. 13,34-36 We observed that intervention patients had a three-point improvement in their NIH stroke scale from baseline to one month poststroke, compared with a one-point improvement for control patients (P = 0.03). To provide a context in which to evaluate the current results, this absolute change in stroke severity was similar to the results of the NINDS trial that was the basis of the FDA approval for tissue plasminogen activator (tpa) for the treatment of acute ischemic stroke. The NINDS trial, which was adequately powered to observe a difference in outcomes, found that 20% of placebo patients had an NIHSS of 0-1 compared with 31% of patients in the tpa group an 11% abso- Table 3 Auto-titrating continuous positive airway pressure use Auto-titrating continuous positive airway pressure () use category Eligible intervention patients* (N = 16) Proportion of nights used: range Mean ± SD 0.69 ± 0.33 Number of h/night used: range Mean ± SD 5.1 ± 2.3 Auto- Adherence: N (%) - No Use: 0 nights or 0 h/night 0 (0) Some Use: < 75% nights for < 4 h/night 6 (37.5) Acceptable Adherence: 75% nights for 10 (62.5) 4 h/night *Eligible intervention patients (N = 16) include the 15 who had evidence of sleep apnea from the auto-titrating continuous positive airway pressure (auto-) machine and one patient who did not meet these criteria but insisted upon using auto- because it made him feel better. lute increase in the favorable outcome. 24 We found that 43% of control patients had an NIHSS of 0-1 compared with 57% of intervention patients a 14% absolute increase in the favorable outcome, albeit with wide confidence intervals given our sample size. The major strengths of this study were the randomized controlled design and the broad range of patients enrolled from a tertiary-care facility, a community hospital, and a Veterans Administration medical center. Also, we employed a novel intervention treatment strategy and achieved good compliance rates. The major limitations of this study were its relatively small sample size and missing data due to withdrawals and missing polysomnography or auto- data. A 29% withdrawal rate in the intervention arm was higher than expected. Figure 2 provides the reasons for withdrawal, including some related to study activities and others related to the post-stroke clinical course. We designed our protocol to reduce withdrawals as much as possible. For example, we recorded multiple contacts for each patient. We also worked with patients and their caregivers to educate them about the research protocol. We gave several informational sessions about the research to the hospital staff in all of the participating hospitals. As a result, the hospital staff was generally very supportive of this study, yet the attitudes of the patients primary care providers were mixed: some were very interested, whereas others were less supportive of patients participation. For example, some of the stroke patients with sleep apnea did not endorse sleepiness symptoms, and their attending clinicians either doubted the diagnosis of sleep apnea or doubted that asymptomatic patients would benefit from treatment. Future trials of stroke patients that involve complex interventions should engage patients primary care providers as part of the recruitment activities. Another limitation of this study was that the staff members who performed the NIHSS at baseline and at the end of the 30-day study period were not blinded to treatment group assignment. However, the staff members were trained in the administration of the NIHSS and the NIHSS is considered to be a relatively objective measure of stroke severity, therefore mitigating potential bias due to unblinding. In conclusion, these findings suggest that the strategy of diagnosing and treating sleep apnea among selected acute ischemic stroke patients is feasible and may be a novel therapeutic approach to improve outcomes. This strategy, if confirmed in future, larger studies, with a less selected population, could be used to complement existing stroke interventions. Given that Table 4 Clinical outcomes Intention-to-Treat Adherence Analysis No (N = 13) Some (N = 6) Acceptable (N = 10) Intervention Control P-value Outcome (N = 31) (N = 24) Stroke severity (NIHSS) median change from baseline to 30-days Vascular events* 1 (3.2) 3 (12.5) (15.4) 0 (0) 0 (0) *Vascular events were defined as: transient ischemic attacks; recurrent stroke; myocardial infarction; hospitalization for congestive heart failure; or any death. Adherence analysis only among patients with sleep apnea, including 13 control patients in the no group. SLEEP, Vol. 34, No. 9, Therapy for Patients with Acute Ischemic Stroke Bravata et al

7 no specialized systems are needed to implement therapy, this intervention may have the potential to be applicable to most stroke patients, regardless of the size or complexity of the facility at which they receive their care. ACKNOWLEDGMENTS Drs. Bravata, Yaggi, Lo, and Agostini received support from career development awards from the Department of Veterans Affairs (VA) Health Services Research and Development Service, Rehabilitation Research and Development Service, and Clinical Science Service. Dr. Fried was supported by the NIH (K24 AG028443). Dr. Concato was supported by the VA Cooperative Studies Program. This work was funded by the Claude D. Pepper Older Americans Independence Center at Yale (P30AG21342 NIH/NIA), the Robert Wood Johnson Generalist Physician Faculty Scholars Award Program, the ResMed Foundation, a pilot grant from the VA Cooperative Studies Program Clinical Epidemiology Research Center, the Max Patterson Stroke Research Fund at Yale, and a grant from VA HSR&D (IIR ). The authors had full access to all of the data and take responsibility for the integrity and accuracy of the analyses. DISCLOSURE STATEMENT This was not an industry supported study. Dr. Lo has received research grants and is an employee of the United States Veterans Administration. He has also received consulting fees from Acorda and Bayer and a travel stipend from Hocoma. Dr. Mohsenin has received honoraria from Jazz Pharmaceuticals and Cephalon, Inc. Dr. Gorman receives support from Boston Scientific. Dr. Agostini is employed by Aetna. The other authors have indicated no financial conflicts of interest. REFERENCES 1. Sandberg O, Franklin K, Bucht G, Gustafson Y. Sleep apnea, delirium, depressed mood, cognition, and ADL ability after stroke. J Am Geriatr Soc 2001;49: Mohsenin V, Valor R. Sleep apnea in patients with hemispheric stroke. Arch Phys Med Rehabil 1995;76: Turkington P, Bamford J, Wanklyn P, Elliott M. Prevalence and predictors of upper airway obstruction in the first 24 hours after acute Stroke. Stroke 2002;33: Harbison J, Ford G, James O, Gibson G. Sleep-disordered breathing following acute stroke. Q J Med 2002;95: Iranzo A, Santamaria J, Berenguer J, Sanchez M, Chamorro A. Prevalence and clinical importance of sleep apnea in the first night after cerebral infarction. Neurology 2002;58: Dyken M, Somers V, Yamada T, Ren Z, Zimmerman M. Investigating the relationship between stroke and obstructive sleep apnea. Stroke 1996;27: Parra O, Arboix A, Bechich S, et al. Time course of sleep-related breathing disorders in first-ever stroke or transient ischemic attack. Am J Respir Crit Care Med 2000;161: Kapen S, Goldberg J, Wynter J. The incidence and severity of obstructive sleep apnea in ischemic cerebrovascular disease. Neurology 1991;41: Bassetti C, Aldrich M. Sleep apnea in acute cerebrovascular diseases: final report on 128 patients. Sleep 1999;22: Wessendorf T, Teschler H, Wang Y-M, Konietzko N, Thilmann A. Sleepdisordered breathing among patients with first-ever stroke. J Neurol 2000;247: Good D, Henkle J, Gelber D, Welsh J, Verhulst S. Sleep-disordered breathing and poor functional outcome after stroke. Stroke 1996;27: Sahlin C, Sandberg O, Gustafson Y, et al. Obstructive sleep apnea is a risk factor for death in patients with stroke: a 10-year follow-up. Arch Intern Med 2008;168: White J, Cates C, Wright J. Continuous positive airways pressure for obstructive sleep apnoea. Cochrane Database Syst Rev 2002; Wright J, Johns R, Watt I, Melville A, Sheldon T. Health effects of obstructive sleep apnoea and the effectiveness of continuous positive airways pressure: a systematic review of the research evidence. BMJ 1997;314: Takasaki Y, Orr D, Popkin J, Rutherford R, Liu P, Bradley T. Effect of nasal continuous positive airway pressure on sleep apnea in congestive heart failure. Am Rev Respir Dis 1989;140: Naughton M, Liu P, Bernard D, Goldstein R, Bradley T. Treatment of congestive heart failure and Cheyne-Stokes respiration during sleep by continuous positive airway pressure. Am J Respir Crit Care Med 1995;151: Disler P, Hansford A, Skelton J, et al. Diagnosis and treatment of obstructive sleep apnea in a stroke rehabilitation unit: a feasibility study. Am J Phys Med Rehab 2002;81: Sandberg O, Franklin K, Bucht G, Eriksson S, Gustafson Y. Nasal continuous positive airway pressure in stroke patients with sleep apnoea: a randomized treatment study. Eur Respir J 2001;18: Wessendorf T, Wang Y-M, Thilmann A, Sorgenfrei U, Konietzko N, Teschler H. Treatment of obstructive sleep apneoea with nasal continuous positive airway pressure in stroke. Eur Respir J 2001;18: Massie C, McArdle N, Hart R, et al. Comparison between automatic and fixed positive airway pressure therapy in the home. Am J Respir Crit Care Med 2003;167: Fitzpatrick M, Alloway C, Wakeford T, MacLean A, Munt P, Day A. Can patients with obstructive sleep apnea titrate their own continuous positive airway pressure? Am J Respir Crit Care Med 2003;167: Goldstein LB MD. Clinical assessment of stroke. JAMA 1994;271: Goldstein L, Bertels C, Davis J. Interrater reliability of the NIH stroke scale. Arch Neurol 1989;46: The National Institute of Neurological Disorders and Stroke rt-pa Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. New Engl J Med 1995;333: American Heart Association, Professional Education Center. NIH Stroke Scale (NIHSS). [cited; Available from: American Academy of Sleep Medicine Task Force. Sleep-related breathing disorders in adults: recommendations for syndrome definition and measurement techniques in clinical research. Sleep 1999;22: Goodrich S, Orr W. An investigation of the validity of the Lifeshirt in comparison to standard polysomnography in the detection of obstructive sleep apnea. Sleep Med 2009;10: Sleep Disorders Atlas Task Force. EEG arousals: scoring rules and examples: a preliminary report from the Sleep Disorders Atlas Task Force of the American Sleep Disorders Association. Sleep 1992;15: Loube D, Gay P, Strohl K, Pack A, White D, Collop N. Indications for positive airway pressure treatment of adult obstructive sleep apnea patients: a consensus statement. Chest 1999;115: Meoli A, Casey K, Clark R, et al. Hypopnea in sleep-disordered breathing in adults. Sleep 2001;24: Giles T, Lasserson T, Smith B, White J, Wright J, Cates C. Continuous positive airways pressure for obstructive sleep apnoea in adults. Cochrane Database Syst Rev 2006;3:CD Marin J, Carrizo S, Vicente E, Agusti A. Long-term cardiovascular outcomes in men with obstructive sleep apnoea-hypopnoea with or without treatment with continuous positive airway pressure: an observational study. Lancet 2005;365: Johnston S, Gress D, Browner W, Sidney S. Short-term prognosis after emergency department diagnosis of TIA. JAMA 2000;284: Diomedi M, Placidi F, Cupini L, Bernardi G, Silvestrini M. Cerebral hemodynamic changes in sleep apnea syndrome and effect of continuous positive airway pressure treatment. Neurology 1998;51: Chin K, Ohi M, Kita H, et al. Effects of n therapy on fibrinogen levels in obstructive sleep apnea syndrome. Am J Respir Crit Care Med 1996;153: Alchanatis M, Paradellis G, Pini H, Tourkohoriti G, Jordanoglou J. Left ventricular function in patients with obstructive sleep apnoea syndrome before and after treatment with nasal continuous positive airway pressure. Respiration 2000; 67: SLEEP, Vol. 34, No. 9, Therapy for Patients with Acute Ischemic Stroke Bravata et al