Neurostimulation for the Treatment of Central Sleep Apnea
- 1. University of Oklahoma Health Sciences Center and Respicardia, Inc. USA
Abstract
Patients with central sleep apnea (CSA) often have cardiovascular co-morbidities and patients may underestimate the impact of poor sleep on their symptoms and quality of life. Testing for sleep disorders such as CSA has historically been done in the sleep lab, but now may be done at home with newer testing modalities. Few therapeutic options exist for the treatment of CSA. Transvenous phrenic nerve stimulation (TPNS), a fully implantable neurostimulation device, is becoming a common therapeutic option for adult patients with moderate to severe central sleep apnea. By stimulating a single phrenic nerve to stabilize a normal breathing pattern, improvements have been observed in the apnea hypopnea index, central apnea index, arousal index and oxygen desaturation index. These improvements are accompanied by improvements in rapid eye movement sleep, quality of life and daytime sleepiness. Risks are similar to risks of other devices with an implantable lead and pulse generator. Transvenous phrenic nerve stimulation improves central sleep apnea and quality of life and ensures compliance by activating automatically and continuing throughout the night.
Citation
Germany R, Jensen M (2020) Neurostimulation for the Treatment of Central Sleep Apnea. J Sleep Med Disord 6(2): 1104.
BACKGROUND
Central sleep apnea (CSA) is an important medical condition of growing interest among the medical community. Identification of patients with CSA has historically required careful questioning of patients and in-laboratory overnight sleep testing, but new testing devices may allow a clear diagnosis of CSA from home [1,2]. Though CSA is relatively uncommon in the general population, some comorbid conditions are associated with much higher prevalence. For example, rates of CSA have been reported between 20-40% among heart failure patients. CSA has also been associated with a two-fold increase in mortality and heart failure hospitalization in these patients [3,4].
Pathophysiology
Fundamentally, CSA is due to a lack of appropriate signals from the respiratory control center in the medulla to the respiratory muscles. However, the disorder can be further classified into two fundamental types: hypocapnic and hypercapnic [5]. Hypercapnic syndromes include disorders such as obesity hypoventilation syndrome, amyotrophic lateral sclerosis, and some cerebrovascular events, and are associated with a lack of deep breathing. In contrast and more widely studied are the hypocapnic disorders, especially those related to cardiovascular diseases such as heart failure and atrial fibrillation. The hypocapnic forms of the disorder result from an increase in loop gain driven by decreased sensitivity to peripheral chemoreceptors to stimulate breathing. While research continues to understand the complete mechanism, part of this decrease is due to increased sympathetic activation on these chemoreceptors. In patients with reduced ejection fraction, this effect is multiplied, which helps explain why patients with heart failure and reduced ejection fraction are at highest risk for CSA [3,5].
Identification of Patients with Central Sleep Apnea
While CSA is common in patients with reduced ejection fraction HF (HFrEF), CSA may go undetected for years and is significantly underdiagnosed in this high-risk patient population [6]. Screening is complicated by the fact that patients with CSA often do not snore or describe themselves as “sleepy,” which is thought to be due to the high sympathetic activation in these patients [3]. However, further questioning often reveals that patients are fatigued and exhibit signs of poor sleep such as frequent awakenings or nocturia. Asking appropriate questions related to possible symptoms in high-risk patient populations is important for identifying patients suffering from CSA. Known high-risk populations include those diagnosed with heart failure, atrial fibrillation, renal disease, and a history of cerebrovascular accidents [5]. At-risk patients should be asked about signs and symptoms of sleep disordered breathing such as fatigue, daytime sleepiness, falling asleep during normal daytime activities, cognitive decline, and awakenings. Moreover, these patients may believe that the causes of their quality of life and sleep issues are due to their co-morbidities and fail to recognize the impact of sleep disorders [7].
Existing screening tools to identify risk for sleep disordered breathing, such as the Epworth Sleepiness Scale and the STOP Bang questionnaire, have failed to show validity in cardiovascular patient populations using traditional scoring criteria [8]. However, screening patients for reports of daytime fatigue, waking up more than two times per night (including restroom breaks), and partner-witnessed apneas can help identify patients who need sleep testing. Importantly, the bedpartner may witness apneas or identify symptoms and so should be included in the discussion when possible.
Testing Options
The gold standard for the diagnosis of CSA is an in-laboratory, attended polysomnogram [9]. These studies give the greatest amount of information and confirm that the events captured occur while the patient is sleeping. Because many patients are resistant to going into a hospital or sleep center for testing, home testing devices are of growing importance in sleep medicine. A limitation of most home tests is the inability to identify sleep stage and therefore potentially the test may report a falsely low apnea hypopnea index (AHI) if the patient has inadequate sleep time during the night. Further, many home sleep tests cannot adequately distinguish between central and obstructive sleep apnea. However, newer home testing devices (Class III devices) can often identify a Cheyne-Stokes Respiration pattern or CSA and recommend additional testing.
A recent addition to home sleep testing technology is the use of pulse arterial tomography. This technology uses changes in arterial tone and oxygen saturation to identify sleep apnea events. Using an additional sensor on the chest, the device is able to distinguish obstructive from central events and is the only FDA-approved home device at this time for the diagnosis of CSA [2].
Treatment Options for CSA
A common barrier to patient testing for sleep disordered breathing is that patients often believe that there is only a single therapy for all forms of sleep apnea: “a mask”. For patients with heart and lung problems, this perception is so strong that they do not want to be tested and diagnosed with sleep apnea Thus, it is important for clinicians to explain the need for testing and the growing number of therapeutic options for all forms of sleep disordered breathing.
Unfortunately, for patients with central sleep apnea, there are few effective treatment options. Traditionally, positive airway pressure therapies have been utilized for the treatment of CSA. Continuous positive airway pressure (CPAP) has been the most widely utilized therapy for CSA but does not improve the apnea hypopnea index (AHI) in many patients and may induce or worsen CSA in others [9,10]. Bi-level PAP therapy has no randomized data to support its use, but the most advanced PAP therapy, adaptive servo-ventilation (ASV), was studied in a large randomized trial for the treatment of CSA in adult patients with HF and a reduced ejection fraction. The therapy did demonstrate a large improvement in AHI but unexpectedly led to an increased risk of cardiovascular outcomes, which led to a contraindication in patients with HF and an ejection fraction less than 45% [11,12]. Besides PAP therapies, oxygen and a few medications have been evaluated in small studies to treat CSA (e.g., acetazolamide and theophylline) but may not be appropriate in all CSA patients [3,9].
Transvenous Phrenic Nerve Stimulation
Device Implantation and Function: Neurostimulation is a recent addition to the treatment of central sleep apnea. Transvenous stimulation of the phrenic nerve (remed? System, Respicardia, Inc.) is an implantable device designed to restore a natural breathing pattern in patients with CSA. Stimulation of the phrenic nerve moves the diaphragm. With this stimulation, the breathing pattern can be stabilized (Figure 1,2) [13].
Figure 1: Transvenous Phrenic Nerve Stimulation System
Figure 2. Phrenic nerve stimulation on/off in CSA.
Example of a 30-minute segment of sleep from a polysomnogram showing epochs of sleep for a subject with central sleep apnea. The left side (Therapy Off) shows the pattern of sleep with central apnea events prior to initiation of therapy. The right side (Therapy On) shows the immediate cessation of central apneas and resumption of normal sleep when transvenous phrenic nerve stimulation therapy is turned on.
The device is implanted under conscious sedation in the cardiology catheterization suite by electrophysiologists with significant experience in cardiac device implantation. Typically, the pulse generator is placed in a subcutaneous pocket near the clavicle on the right side. The stimulation lead is placed in either the left pericardiophrenic vein or the right brachiocephalic vein depending on patient anatomy. While considered optional, a sensing lead is typically placed in an intercostal branch of the azygos vein to detect respiration by means of transthoracic impedance that improves the diagnostic capability of the device. Patients typically stay in the hospital for several hours to overnight after implant [14]. Following implant, the device is programmed to a monitor-only mode and collects patient data for 4-6 weeks prior to therapy activation. This allows for healing and to acquire adequate data collection to program the device and activate therapy for the particular patient needs. At the therapy activation visit, the device is programmed to initiate therapy automatically when the following programmed criteria are met: appropriate time of day, sleeping position (angle) and lack of movement. Once these criteria are met, a safety check is done by the device (lead impedance) and if the lead impedance is within range, the device begins to deliver therapy. Patients are typically programmed to asynchronous stimulation which is a constant stimulation at a set rate. This is designed to slightly lower the respiratory rate [15]. After stimulation begins, therapy suspends briefly when a patient rolls during sleep and longer when a patient sits up at night to allow the patient to control therapy when staying up late or getting up to use the restroom at night. In the morning, therapy automatically ends at a preset programmed time. Thus, therapy start and stop requires no patient interaction and compliance is assured [15].
Clinical Data: TPNS was studied in several clinical trials. During a feasibility study, Ponikowski, et al., demonstrated that the AHI was improved during one night of therapy compared to a control night [16]. Patients were temporarily implanted with the stimulation lead connected to an external device and underwent two nights of therapy. On therapy, AHI, central apnea index (CAI), oxygenation, and arousals all improved compared to a control night [16]. Following the early feasibility work, subsequent studies implanted the full system.
Zhang et al., conducted a chronic safety study in eight patients for one to six months. They reported improvements in AHI, CAI, ejection fraction and 6-minute hall walk testing. The only serious adverse event was a single patient with a lead dislodgement [17]. This small study was followed by a longer-term pilot study of 47-patients [13].
The remed? System Pilot Study was a prospective, multicenter study with a 3-month safety and efficacy endpoint [13]. The study demonstrated a 55% reduction in AHI from baseline (p<0.0001). Improvements were also noted in the CAI, arousal index, oxygenation, sleep quality and patient global assessment (a patient reported quality of life instrument). The subset of patients with heart failure (36 patients) experienced improvements in quality of life measured by Minnesota Living with Heart Failure Questionnaire (mean 10-point improvement at 6 months, p<0.001). Twenty-six percent of patients had a serious adverse event related to the device or procedure which was primarily due to lead dislodgement requiring a second procedure [13]. The results of this study led to a redesign of the left pericardiophrenic lead and the remed? System Pivotal Trial [18].
Details of the remed? System Pivotal Trial have been described [18]. Briefly, patients were considered for the trial if they demonstrated moderate to severe CSA based on an in-laboratory attended polysomnogram scored by a blinded core laboratory. PSG criteria for inclusion included: at least 4 hours of recording time with 2 hours of sleep, AHI ≥ 20 events/hour, more central apneas than other types of apneas, at least 30 central apneas during sleep, and no more than 20% of all events obstructive apnea. Hypopneas were not classified as central or obstructive during this trial. Important exclusion criteria included CSA due to pain medications, recent cerebrovascular accident, severe COPD, and pacing dependent patients. Patients were required to be medically stable with optimal medical therapy for all comorbidities prior to enrollment in the clinical trial [18].
All (N=151) patients underwent a TPNS device implant procedure and were randomized at the time of implantation attempt to either active TPNS therapy or control (therapy remained inactive for the first 6 months). The primary endpoint was a comparison of the proportion of patients with at least a 50% reduction in AHI at six months. The safety endpoint was device, procedure, and therapy related serious adverse events through 12 months. Additionally, seven hierarchically tested secondary endpoints and multiple exploratory measures were collected and evaluated [19].
There was a 41 percentage point difference in the proportion of subjects achieving a 50% reduction in AHI between the treatment group and control group (p<0.0001), with the outcome favoring the treatment group. At 12 months, freedom from serious adverse events related to the procedure, device or therapy was 91% which is similar to other transvenous devices such as implantable cardiac defibrillators. In addition, the study met all seven pre-specified secondary endpoints including improvements in AHI, CAI, arousal index, oxygen saturation index (ODI) 4%, patient global assessment, Epworth Sleepiness Scale (all p<0.001) and percent of sleep time in REM sleep (p=0.0244) [19].
Therapy was activated in the control group after the sixmonth efficacy endpoint and patients continued to be followed to a maximum of 5 years. Data analyzed through 3 years to date demonstrated sustained improvements in AHI, CAI and ODI 4%.
The 3-year analysis was completed with home sleep testing equipment and therefore arousals and sleep quality were not evaluated, however remained improved through the last in-lab study at 2 years. Safety remained strong through 3 years with 90% freedom from serious related adverse events [20]. Subgroup analysis did not identify predictors of non-response of those enrolled in the trial [21].
The pivotal study provided a single crossover design as therapy was activated in the control group at 6 months. After six months of therapy, the control group demonstrated similar improvements in sleep metrics, oxygenation, and quality of life similar to the therapy group at six months showing reproducibility of results. One sub-group analysis examined patients with heart failure, which was 64% of the study population [22]. The study did include subjects with heart failure that were not ACC/AHA Heart Stage D (exclusion criteria). While the pivotal trial did not specifically study the impact of TPNS in heart failure patients, it did include LVEF as a secondary endpoint. Secondary endpoints were only included to provide additional analyses at 6 and 12 months post therapy initiation visit, for further characterization of outcomes in study subjects. Assessments of changes at 6 months post-therapy initiation visit are comparisons between Treatment and Control groups. Changes at 12 months posttherapy initiation visit are for the Treatment group, and are comparisons to baseline. Patients with heart failure had similar improvements in sleep metrics and quality of life. In addition, improvements were noted in ejection fraction and left ventricular dimension at 12 months as compared to baseline in patients with baseline left ventricular dysfunction (p<0.05). Minnesota Living with Heart Failure scores improved by 6.8 points (p<0.01). While sample sizes were small, the rate of heart failure hospitalization at 6 months showed some evidence of being lower in the treatment group than control (p=0.065) [22].
CONCLUSIONS
Central sleep apnea is associated with high morbidity and mortality. While most often occurring in patients with cardiovascular disease, symptoms of CSA are often mistaken as resulting from their underlying cardiac disease. Clinicians should have a high suspicion for CSA in at-risk patients. Sleep testing for CSA is expanding and now can be completed at home in some patients. Alternate treatment options have limited data, may have compliance concerns, but include positive airway pressure, oxygen, and medications.
Transvenous phrenic nerve stimulation is the most recent therapeutic option for the treatment of CSA. The device activates automatically ensuring patient compliance with therapy. The therapy has demonstrated significant improvements in sleep metrics, quality of life and oxygenation in a multi-center, randomized clinical trial and improvements were stable through 36 months. The effects were reproducible across several studies and groups of patients. Additional data will clarify patient benefits over time in the approved population of adult patients with moderate to severe CSA.
