Mode selection for positive airway pressure titration in adult patients with central sleep apnea syndromes

INTRODUCTION — Positive airway pressure (PAP) is used to treat sleep-disordered breathing in patients with central sleep apnea (CSA) and hypoventilation syndromes. Determining the optimal pressure settings is of paramount importance to eliminate apneas and to help assure patient adherence to therapy.

Selecting a titration method to determine the optimal pressure in patients with CSA is discussed in this topic. An overview of the treatment of CSA, selection of a titration method for patients with obstructive sleep apnea, and PAP therapy for hypoventilation syndromes are discussed separately. (See "Central sleep apnea: Treatment" and "Noninvasive ventilation in adults with chronic respiratory failure from neuromuscular and chest wall diseases: Patient selection and alternative modes of ventilatory support" and "Noninvasive ventilatory support and mechanical insufflation-exsufflation for patients with respiratory muscle dysfunction" and "Noninvasive positive airway pressure therapy for the obesity hypoventilation syndrome".)

DIAGNOSIS AND CAUSES OF CSA — CSA syndromes are characterized by the absence of both airflow and ventilatory effort during sleep. CSA is usually best diagnosed during an attended in-laboratory polysomnographic (PSG)-based sleep test but can occasionally be diagnosed on a home sleep apnea test (HSAT). (See "Central sleep apnea: Risk factors, clinical presentation, and diagnosis", section on 'Evaluation'.)

CSA can be primary (ie, idiopathic CSA) or secondary to other conditions (Cheyne-Stokes breathing, medical conditions such as stroke or heart failure, suppressant drug or substance use, high altitude periodic breathing, or treatment-emergent CSA). Alternatively, CSA can be classified as hyperventilation-related (eg, Cheyne-Stokes breathing, primary CSA) or hypoventilation-related (eg, central nervous system diseases, drugs or substances) (table 1). There is also a subset of conditions primarily associated with alveolar hypoventilation (neuromuscular diseases, severe thoracic cage disorders such as kyphoscoliosis) in which CSAs or hypopneas may be present. In many of these conditions, central apnea is not the only or even dominant type of sleep-disordered breathing that is present. It is therefore prudent that the clinician be aware of the proportional spectrum of sleep-disordered breathing type in a given patient such that an appropriate modality targeted at the major type of sleep-disordered breathing is chosen. (See "Polysomnography in the evaluation of sleep-disordered breathing in adults".)

Hypoventilation syndromes without apneic events may be central (eg, spinal cord injury) or peripheral (eg, myasthenia gravis), and when severe enough may be treated with noninvasive PAP, the details of which are discussed separately. (See 'Patients with central hypoventilation syndromes' below and "Respiratory muscle weakness due to neuromuscular disease: Clinical manifestations and evaluation".)

PRINCIPLES AND GOALS OF PAP

General principles — Most cases of CSA are associated with a medical disorder (eg, stroke, heart failure) or medications (eg, opioid). In this population, the primary form of therapy is treatment of the associated medical condition when possible, or removal of the offending medication. The clinician needs to be aware of potential dynamic changes in the severity of CSA related to these factors. This fact also emphasizes the need for medical optimization prior to sleep evaluation. Over time, improvement or resolution of the underlying cause may result in a reduced need for PAP therapy, necessitating re-titration at variable time points. (See 'Goals of PAP titration in CSA' below and 'Caveats of PAP titration in CSA' below.)

PAP titration for the common causes of CSA, including heart failure, opioid use, and treatment-emergent sleep apnea, is discussed in the sections below. (See 'Patients with CSA and heart failure' below and 'Patients with CSA and opioid use' below and 'Patients with treatment-emergent CSA' below.)

It should be noted that most data regarding care of patients with CSA come from the heart failure population.

An overview of specific treatments for CSA, and the data that supports PAP therapy in patients with CSA are reviewed separately. (See "Central sleep apnea: Treatment".)

Goals of PAP titration in CSA — The goal of PAP titration is to determine the minimal pressure required to resolve all apneas, hypopneas (apnea-hypopnea index [AHI]) and other sleep-related respiratory events, in all stages of sleep and in all sleep positions. The use of AHI as an outcome measure comes as an inference from data gathered predominantly in obstructive sleep apnea (OSA) patients [1,2]. However, in many instances, this goal may not be achieved due to the caveats outlined in the section below.

Due to the caveats associated with PAP titration in this population (see 'Caveats of PAP titration in CSA' below), it is difficult to define a "failure" of any therapy that is tried. While some experts use AHI <15 as a goal of therapy (and treat AHI ≥15 as a "failure," or a reason to change and adjust therapy) [3], others use a more strict criterion of AHI <5 as a definition of success.

Caveats of PAP titration in CSA — In contrast with patients who have OSA, PAP therapy may not eliminate all sleep-disordered breathing events in those with CSA, such that titration to this goal is challenging, especially in the absence of alternate well-defined goals. In addition, ventilatory instability associated with high loop gain can lead to overventilation and hypocapnia, which, in turn, can worsen central apnea, particularly when PAP is administered at higher pressures, a phenomenon termed "pressure toxicity" [4,5], or when bilevel PAP is employed. Worsening CSA due to this phenomenon can, in turn, lead to an increase in sleep fragmentation and poor adherence.

Thus, PAP titration with the goal of resolving all central sleep-disordered breathing events may not be feasible in many patients with CSA syndromes. For this reason, attended in-laboratory PAP titrations are preferred with supervised polysomnographic (PSG) monitoring so that fine adjustments can be made to achieve the maximal pressure or optimal PAP modality that allows a minimum number of residual central apneas.

PATIENTS WITH CSA AND HEART FAILURE — The management of sleep-disordered breathing in patients with heart failure should focus on optimizing heart failure medications as well as treating sleep-disordered breathing with PAP. If clinically significant CSA persists despite maximal medical management, continuous PAP (CPAP) is the typical first-line PAP modality used in this population. Most patients should undergo a titration for CPAP in an attended in-laboratory setting. Titration for second-line modes (adaptive servo-ventilation [ASV] and bilevel PAP ventilation) are reserved for those who fail CPAP. (See 'Continuous PAP' below and 'Patients with CSA and heart failure who fail continuous PAP' below.)

Continuous PAP — CPAP provides positive pressure at a relatively constant level throughout inspiration and expiration. Two types of sleep-disordered breathing are common among patients with heart failure, obstructive sleep apnea (OSA) and CSA with Cheyne-Stokes breathing (CSA-CSB). The rationale for using CPAP in patients with heart failure is the correction of airway collapse that is associated with obstructive events but also transiently present at the end of central apneas and may lead to ventilatory overshot, hyperventilation, and oscillatory apneas in susceptible individuals [6-8] (see "Sleep-disordered breathing in heart failure"). Although in-laboratory CPAP titrations and in-hospital "acclimation trials" are sometimes used to determine the optimal CPAP in patients with CSA, there are no comparative data between these titration methods and no official guidelines to direct clinicians. Thus, many centers, including ours, perform attended in-laboratory CPAP titrations for this population. We feel that compared with the acclimation trial, an in-laboratory CPAP titration offers an opportunity to explore and troubleshoot mask interfaces and gives an ability to quickly correct or adjust settings that are clearly not working, thus improving the patient's first experience of PAP, which is an important predictor of long-term compliance.

Attended in-laboratory PAP titration — In a sleep laboratory setting, CPAP is initiated as the first modality at the lowest pressure (usually 4 or 5 cm H₂O) and gradually raised to eliminate obstructive events during sleep as determined by polysomnography (PSG). If central apneas or hypopneas persist at this point, the following are options:

●CPAP may be ramped up in 1 to 2 cm H₂O intervals to no greater than an additional 5 cm H₂O to assure that no residual obstruction (hypopneas, respiratory effort-related arousals [RERAs]) is present, and then gradually brought down to a setting where obstruction is resolved and the number of central apneas is the lowest (ie, the "optimal pressure") [9,10]. We prefer this method, as it assures that obstruction is no longer playing a role, which is important when further adjustments based on the interrogation of the device are needed.

●Alternatively, some experts avoid ramping up pressure and allow a stabilization period of 10 minutes to occur [10]. Concerns about "pressure toxicity" and potential hemodynamic effects of high PAP in patients with CSA and heart failure has led some centers to place a pressure limit on the titration and refrain from using CPAP pressures of >10 to 18 cm H₂O [10-13]. These pressure limits for CPAP are arbitrary and center-specific.

If central apneas persist at this point, low flow supplemental oxygen can be added [13]. This is especially true in patients with heart failure and left ventricular ejection fraction <45 percent for whom use of adaptive servo-ventilation (ASV) may be harmful [14]. The caveat here is that although the addition of oxygen decreases central apnea activity acutely, if the patient does not have significant hypoxemia (SpO₂ <88 percent), they may not qualify for oxygen.

If no control of central apnea activity is accomplished at this stage, and there is sufficient time left during the sleep study (>60 minutes), we typically transition to ASV in a multimodality protocol in patients with normal systolic left ventricular function. (See 'Adaptive servo-ventilation' below.)

Unattended in-home "acclimation" continuous PAP trial — An alternative and less common strategy to a formal in-laboratory CPAP titration, is to initiate CPAP at empiric settings. In the largest randomized study of patients with CSA and heart failure (CANPAP), empiric pressure of 5 cm H₂O was applied in-hospital for the night without PSG monitoring and subsequently raised in increments of 2 or 3 cm H₂O to a level of 10 cm H₂O over two to three nights, with a goal of acclimating the patient to CPAP [15]. This acclimation strategy has been followed by other centers, with acclimation taking place during the day in a one- to two-hour session [16,17]. This trial is discussed separately. (See "Central sleep apnea: Treatment", section on 'Efficacy' and "Central sleep apnea: Treatment", section on 'Titration'.)

Choosing in-laboratory or in-home titration — Most experts choose in-laboratory settings because of the difficulties associated with PAP titration in this population outlined above. In addition, home sleep apnea tests, although validated to assess for obstructive sleep have not been sufficiently validated to accurately identify or monitor CSA, and automatically titrating PAP devices have not been validated to treat CSA. (See 'Caveats of PAP titration in CSA' above.)

Acclimation trials may be offered to those unable to come to the laboratory for a sleep study.

Other titration methods

Multimodality titration — Some centers, including ours, employ a strategy of titrating CPAP and then ASV if necessary, in a sequential fashion, during the same night (multimodality titration study), when feasible. Although this approach is of unproven efficacy, it has the advantages that it may speed up the process, reduce costs, and achieve the main goal of lowering the apnea-hypopnea index (AHI), as would testing with separate modalities on separate nights [18].

Auto-titration — Although there are single reports describing the use of auto-titrating devices to determine the CPAP pressure in treatment of CSA [19], this approach is not recommended [20]. These devices are more commonly used to titrate PAP in patients with OSA, which is discussed separately. (See "Mode selection for titration of positive airway pressure in adults with obstructive sleep apnea" and "Titration of positive airway pressure therapy for adults with obstructive sleep apnea".)

PATIENTS WITH CSA AND HEART FAILURE WHO FAIL CONTINUOUS PAP — Although there is not a universally accepted definition of continuous PAP (CPAP) "failure" in a setting of treatment of CSA, we define it as the residual apnea-hypopnea index (AHI) of 5 or above, with >50 percent of residual events of central origin. Options for PAP therapy in this population are limited to adaptive servo-ventilation (ASV) and bilevel PAP (BPAP) in a spontaneous/timed (S/T) mode. Typically, ASV is preferred because in comparative studies, it achieved a better control of CSA than did BPAP-S/T [18,21]. However, these modes, particularly ASV, are of limited benefit and potentially harmful in those with a low left ventricular ejection fraction (≤45 percent), such that treatment with oxygen may be the only alternative therapy in this population [14,22].

Patients with ejection fraction >45 percent

Adaptive servo-ventilation — ASV uses proprietary algorithms to provide variable amounts of pressure that alternate between expiratory and inspiratory phases of the respiratory cycle with a physician pre-set or automatic back-up rate during periods of central apnea. It is used as second-line therapy for patients with CSA and a preserved ejection fraction (>45 percent) who have failed or do not tolerate CPAP. Although there is no universally accepted definition of such failure, we consider ASV if residual AHI is ≥5. ASV is limited to those with preserved ejection fraction, as a lack of efficacy and an increased cardiovascular mortality has been reported in heart failure patients with CSA and a low ejection fraction (≤45 percent) [14,22]. Patients in whom ASV is indicated should have a formal attended in-laboratory ASV titration with polysomnographic (PSG) monitoring. Empiric, untitrated application of ASV, with the expiratory PAP (EPAP) based on the AHI that has been documented in one study, is not routinely recommended [23]. (See "Central sleep apnea: Treatment", section on 'Patients with ejection fraction ≤45 percent'.)

Devices — ASV is a form of bilevel device that alternates between an expiratory pressure (which may be constant or variable) and inspiratory pressure. The difference between the expiratory and inspiratory pressure ("pressure support") varies from breath to breath. The greater the difference in pressure support, the greater the tidal volume that is delivered. The device is equipped with a feedback circuit that allows it to measure the respiratory output from the patient (either in the form of peak flow or minute ventilation) to individually regulate delivered pressure support. Thus, in periods of increased patient respiratory effort (hyperpnea), the machine's support decreases, while in periods of decreased patient effort (hypopnea/apnea), the machine's support increases [24,25]. The device alternates between expiratory and inspiratory pressure with a physician-preset or automatic back-up rate in periods of central apnea. The algorithms seek to provide a steady level of mild hypoventilation, which may be useful in patients with CSA related to hyperventilation.

Titration modules — Proprietary algorithms employed by individual devices vary such that direct supervision by the technologist in an in-laboratory PSG-monitored setting is preferred; this allows for individual adjustments, which may or may not follow the protocol suggested by the manufacturer of the device. ASV does not perform well when mask leak is high; assuring good mask fit is especially important in the ASV titrations. Most experts use ASV devices that require several pressure limits to be defined in advance but devices that employ an auto-titrating protocol to determine expiratory pressure are also available. The titration method for each type of device is different:

●ASV devices with set pressures – Settings for expiratory pressure and pressure support as well as a back-up rate need to be set manually before titration.

•Expiratory pressure – The expiratory pressure needed to overcome obstruction may be known from the prior CPAP titration, even if it was not successful in controlling central events. In this instance, we and others set expiratory pressure on ASV at the level of CPAP pressure that controlled obstruction [26,27]. Other centers set the EPAP on ASV device 2 cm H₂O below the CPAP level known to control obstructive apneas [13]. If the pressure needed to overcome airway obstruction is not known, or no obstructive apnea is present, then the EPAP is initiated at the lowest setting (4 to 5 cm H₂O) [28,29]. Expiratory pressure is then gradually increased, typically in 1 cm H₂O increments to eliminate obstructive apneas, typically allowing 20 minutes at each pressure level [30,31].

•Pressure support – The level of pressure support is unclear, and there is no agreement among experts.

-We and many experts support all breaths by setting a minimal pressure support in the 2 to 3 cm H₂O range [25]. Maximal pressure support is typically set between 8 and 15 cm H₂O. Some experts limit the pressure support range to 5 cm H₂O [32].

-However, the issue of whether to support each breath during treatment with ASV is controversial, since continued pressure support during periods of hyperpnea may lead to worsening of hypocapnia, which can further enhance central apnea activity and be detrimental to cardiovascular function [33]. Thus, some experts set minimal pressure support at 0 cm H₂O [31].

-Other experts will individualize according to the patients; as an example, in patients with obesity who may have a component of hypoventilation, an increase in minimal pressure support above 5 cm H₂O may be necessary to ensure adequate ventilation. (See 'Caveats of PAP titration in CSA' above.)

•Back-up rate – Whenever an automatic back-up rate is available, we and others use it in our titrations. Some ASV devices allow the use of a set back-up rate rather than relying on the automatic rate. There are no direct comparisons between the set versus automatic rates on the ASV devices during titrations.

●ASV devices with auto-adjusting expiratory pressure – Some ASV devices have auto-titrating technology capable of regulating expiratory pressure [25,34]. This technology is similar to the auto-titrating technology used in CPAP devices. ASV devices with variable expiratory pressure do not require manual alterations of pressures, but auto-adjust expiratory pressures within the preset range in response to the airway obstruction encountered during individual breaths, while delivering pressure support for each breath [27,35]. Typically, the minimum expiratory pressure is set at the level known to control obstruction, and the maximum expiratory pressure is limited to 15 cm H₂O. The use of autotitrating technology may reduce the need for supervised testing. Although in a clinical trial setting, the unsupervised "autotitrations" using the device's algorithm have been successfully performed [36], our practice is still to initiate it during a supervised PSG. This is to assess efficacy for the individual patient and adjust pressure ranges and other settings for maximal efficacy and comfort.

Bilevel PAP — BPAP (usually in the spontaneous/timed [S/T] mode) is typically reserved for those who have failed or not tolerated CPAP and ASV, with a residual AHI of ≥5. In this population, BPAP should only be used with a back-up respiratory rate because BPAP without a back-up rate may exacerbate hyperventilation, hypocapnia, and hypoxemic events associated with worsening central apnea. (See 'Caveats of PAP titration in CSA' above.)

Titration for BPAP in the S/T mode should be performed in an attended in-laboratory setting with PSG monitoring. Settings for expiratory pressure and pressure support as well as a back-up rate need to be set manually before titration:

●Expiratory pressure – Since we typically try BPAP-S/T as a second- or third-line modality, the expiratory pressure needed to overcome obstruction is typically known from the prior CPAP or ASV titrations, even if it was not successful in controlling central events. In this instance, we set expiratory pressure on BPAP-S/T at the level of CPAP pressure that controlled obstructive events. If the pressure needed to overcome airway obstruction is not known, or no obstructive apnea is present, then the expiratory pressure (EPAP) is initiated at the lowest setting (4 to 5 cm H₂O) and gradually increased, typically in 1 cm H₂O increments, to eliminate obstruction.

●Inspiratory pressure – We initially set the inspiratory pressure (IPAP) to 4 to 5 cm H₂O above the EPAP and raise it in tandem with the increases of the EPAP until all obstruction is controlled. If residual central apneas are present, IPAP alone is further increased in 1 to 2 cm H₂O increments. To the extent that the IPAP exceeds the EPAP setting, the machine then provides some degree of assisted ventilation during periods that would otherwise show central apnea.

●Back-up rate – We set the back-up rate at two breaths/minute less than the patient's resting respiratory rate in quiet wakefulness. This minimizes the possibility of hyperventilation during the titration. Thus, in most circumstances, we use the rates of 10 to 14 per minute.

Patients with ejection fraction ≤45 percent — There is an increased mortality in patients with CSA and heart failure with low ejection fraction treated with ASV, but no similar quality data on BPAP-S/T [14,22]. Our personal practice is not to offer BPAP-S/T in this patient population, because its effect and therefore risk may be similar to that of ASV. In patients with persistent CSA who are on maximum medical therapy, we typically treat with CPAP or CPAP with oxygen. Sometimes oxygen is used alone without CPAP. Rarely, some patients have less frequent central apnea events in the non-supine position such that positional therapy including elevation of the head of bed and/or avoiding sleeping in the supine position is advised for selected patients [37]. (See "Central sleep apnea: Treatment", section on 'Supplemental oxygen during sleep'.)

PATIENTS WITH CSA AND OPIOID USE — Treatment of this population involves the administration of PAP therapy while the dose of medication is being lowered or discontinued. The phenotype of sleep-disordered breathing in opioid users differs from CSA due to other conditions. Central apneas, ataxic breathing, prolonged obstructive hypopneas, and hypoventilation can all be seen [38]. For that reason, in most cases, bilevel PAP (BPAP) modalities (eg, adaptive servo-ventilation [ASV] or BPAP in the spontaneous/timed [S/T] mode) is used with better success than continuous PAP (CPAP), which is typically insufficient to fully control sleep-disordered breathing in this population. However, data directly comparing these modalities are sparse, and most data come from small case series and our experience [35,39-42]. In patients whose central apneas are due to opioid use and cannot be controlled to reach an apnea-hypopnea index (AHI) <5 by CPAP, ASV with an auto-titrating expiratory pressure is typically the first-line mode of PAP therapy. For those who fail ASV, we perform titration of BPAP-S/T. In patients with CSA due to opioids, we monitor transcutaneous carbon dioxide (CO₂) levels, a measure of hypoventilation, throughout the polysomnogram, although this technology is not always feasible and this approach is unproven. (See "Sleep-disordered breathing in patients chronically using opioids".)

Attended in-laboratory continuous PAP titration — Similar to CPAP titration in CSA due to heart failure, CPAP is initiated at the lowest pressure likely to be helpful (ie, 4 to 5 cm H₂O) and titrated to eliminate obstruction. If sleep-disordered breathing cannot be controlled with CPAP, we proceed with an ASV titration study, typically during the same titration night. (See 'Attended in-laboratory PAP titration' above and 'Multimodality titration' above.)

Attended in-laboratory adaptive servo-ventilation titration — There are minor differences in ASV titration protocols used in CSA due to opioid use in comparison with the protocol described above for CSA in heart failure (see 'Adaptive servo-ventilation' above). Since hypoventilation may be present, there is frequently a need to increase the pressure support when obstructive and central apneas are controlled [40]. In this situation, after establishing the expiratory pressure that controls obstruction, our practice is to increase both minimal and maximal pressure support in 1 cm H₂O increments until the maximal inspiratory pressure of 25 cm H₂O is reached or mask leak becomes problematic. Other experts use an initial minimal pressure support of 6 cm H₂O, although this approach was not associated with measurable benefit when compared with patients who had minimal pressure support of 0 cm H₂O [12].

Attended in-laboratory bilevel PAP titration — The main difference in the BPAP-S/T titration protocol that we use in CSA due to opioid use, as compared with the protocol described above for CSA in heart failure, is that the difference between expiratory and inspiratory pressures is typically higher. As hypoventilation is present and risk of the BPAP-induced hyperventilation is small, we tend to start with the pressure difference of 5 cm H₂O and increase it further as needed. (See 'Bilevel PAP' above.)

PATIENTS WITH TREATMENT-EMERGENT CSA — A subset of patients (1 to 5 percent) with obstructive sleep apnea (OSA) develops CSAs upon introduction of treatments that restore airway patency, including continuous PAP (CPAP), oral appliance, and surgery. This condition is termed treatment-emergent CSA (TE-CSA, formerly complex sleep apnea). Although the optimal management is unknown, there are two usual sequences of therapy.

●Some experts continue CPAP therapy after an initial "failed" titration (ie, showing residual apnea-hypopnea index [AHI] of ≥5, consisting of mostly central events) for two to three months at the level identified in the initial sleep study to eliminate obstructive events. This is because TE-CSA frequently improves or resolves with this approach. With subsequent evidence of uncontrolled TE-CSA (based on the lack of symptomatic improvement or evidence of residual sleep-disordered breathing on interrogation of the device), an in-laboratory retitration study is then needed. During a repeat study, if TE-CSA has not resolved, CPAP may be continued at a suboptimal level, or an alternate mode of PAP, typically adaptive servo-ventilation (ASV) or, rarely, bilevel PAP (BPAP) in a spontaneous/timed (S/T) mode, may be chosen.

●Other experts, including ourselves, prefer ASV if CPAP is acutely not able to resolve complex sleep-disordered breathing to AHI <5. This ASV trial is performed either as a multimodality study (see 'Multimodality titration' above), if time permits immediately following CPAP titration, or as a separate polysomnogram using the protocol described above (see 'Adaptive servo-ventilation' above). When choosing treatment modalities, based on limited data, ASV seems to be superior to BPAP-S/T in acutely resolving TE-CSA [21].

All titrations should be performed in an in-laboratory setting with polysomnographic (PSG) monitoring:

●ASV titration – Since most patients with TE-CSA have received CPAP prior to ASV titration, in most instances, expiratory pressure that controls airway obstruction is known. However, the expiratory pressures needed to control obstruction in TE-CSA tend to be higher than in CSA from other causes. The suggested protocol for ASV titration (full night or multimodality) is described in the section above. (See 'Adaptive servo-ventilation' above and 'Multimodality titration' above.)

●BPAP-S/T titration – For those in whom BPAP is chosen, a BPAP-S/T titration protocol is followed. Of note, BPAP without a back-up rate should not be used [18,43]. (See 'Bilevel PAP' above.)

As there are some data indicating that addition of oxygen (O₂) to CPAP may help stabilize breathing in TE-CSA [18,39], some centers add O₂ to CPAP after obstructive events are controlled and central events persist. Typically, a flow of 2 to 3 L/minute is used in this setting. A managing physician must bear in mind, however, that most insurances will not recognize TE-CSA as an indication for long-term O₂ supplementation.

The pathophysiology, risk factors, clinical presentation, and diagnosis of TE-CSA as well as treatment strategies for TE-CSA are described in greater detail separately. (See "Treatment-emergent central sleep apnea".)

PATIENTS WITH RARE ETIOLOGIES OF CSA — CSA from other causes, including primary CSA (ie, idiopathic), high altitude periodic breathing, and Cheyne-Stokes breathing, is rare. While continuous PAP (CPAP) is typically used as first-line therapy by most experts, ASV and bilevel PAP have also been used in this population, when CPAP fails. However, no randomized trials have compared these modalities. Thus, titrating PAP is individualized and dependent upon the modality chosen to deliver PAP as well as the underlying cause and spectrum of sleep-disordered breathing. The principles and caveats of PAP therapy and protocols for PAP titration are typically similar to those described for patients with heart failure. (See 'Patients with CSA and heart failure' above.)

PATIENTS WITH CENTRAL HYPOVENTILATION SYNDROMES — In this population, central apneas tend to be a minor component of the sleep-disordered breathing. In patients with CSA whose central apneas are largely due to hypoventilation, bilevel PAP (BPAP) is first-line therapy and an in-laboratory BPAP titration should be performed. (See 'Bilevel PAP' above.)

Examples of these disorders include the following:

●Obesity hypoventilation syndrome (see "Clinical manifestations and diagnosis of obesity hypoventilation syndrome" and "Treatment and prognosis of the obesity hypoventilation syndrome" and "Noninvasive positive airway pressure therapy for the obesity hypoventilation syndrome")

●Idiopathic and congenital central alveolar hypoventilation syndrome (see "Disorders of ventilatory control", section on 'Congenital central hypoventilation syndrome')

●Late-onset central hypoventilation with hypothalamic dysfunction (see "Disorders of ventilatory control", section on 'Late-onset central hypoventilation syndrome')

●Sleep-related hypoventilation due to a medication or substance (see 'Patients with CSA and opioid use' above and "Sleep-disordered breathing in patients chronically using opioids")

●Sleep-related hypoventilation due to a medical disorder (eg, stroke, central nervous system diseases including multiple sclerosis, syringomyelia, spinal cord injury) (see "Sleep-related breathing disorders and stroke")

Hypoventilation due to noncentral etiologies (eg, neuromuscular diseases) is also generally treated with BPAP, the details of which are discussed separately. (See "Noninvasive ventilation in adults with chronic respiratory failure from neuromuscular and chest wall diseases: Practical aspects of initiation" and "Noninvasive ventilation in adults with chronic respiratory failure from neuromuscular and chest wall diseases: Adaptation and follow-up after initiation" and "Noninvasive ventilation in adults with chronic respiratory failure from neuromuscular and chest wall diseases: Patient selection and alternative modes of ventilatory support" and "Respiratory complications in the adult patient with chronic spinal cord injury" and "Respiratory muscle weakness due to neuromuscular disease: Management" and "Treatment of bilateral diaphragmatic paralysis in adults" and "The effect of sleep in patients with neuromuscular and chest wall disorders" and "Noninvasive ventilation in adults with acute respiratory failure: Practical aspects of initiation".)

FOLLOW-UP — Follow-up is particularly important in patients with CSA as PAP requirements may change with treatment of the underlying disorder and close attention needs to be paid to the pressure toxicity effects of PAP. (See 'Caveats of PAP titration in CSA' above.)

In general, following the initiation of PAP therapy, most patients are evaluated at approximately four to six weeks and assessed for symptoms of daytime sleepiness, mask tolerance, and oronasal effects. Device interrogation data can also assess the estimated apnea-hypopnea index (AHI_flow), mask leak, and adherence. Adjustments are sometimes made empirically at that point. Occasionally patients may need retesting in the sleep laboratory, especially when an alternate mode of PAP therapy is indicated, in which case re-titration according to the chosen mode is necessary.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Sleep-related breathing disorders in adults".)

SUMMARY AND RECOMMENDATIONS

●Central sleep apnea (CSA) syndromes are characterized by absent airflow and ventilatory effort during sleep, as identified during sleep laboratory testing (polysomnography [PSG]). CSA can be primary (ie, idiopathic) or secondary to other conditions (eg, heart failure, opioid use). Because central apnea in many cases coexists with other types of sleep-disordered breathing (eg, obstructive sleep apnea [OSA], hypoventilation), an appropriate modality targeted at the major type of sleep-disordered breathing should be chosen. (See 'Diagnosis and causes of CSA' above and "Central sleep apnea: Risk factors, clinical presentation, and diagnosis".)

●Because most patients with CSA have an underlying cause (eg, heart failure or opioid use), the primary form of therapy is often treatment of the associated medical condition or removal of the offending medication. Positive airway pressure (PAP) in patients with CSA, in contrast to those with OSA, often does not eliminate all apneic events. Therefore, titration is more challenging and should occur in an in-laboratory setting with PSG monitoring so that the response can be maximized. (See 'Principles and goals of PAP' above.)

●In patients with CSA and heart failure despite optimized medication therapy, continuous PAP (CPAP) is the typical first-line PAP modality. Patients should undergo a titration for CPAP during an attended in-laboratory PSG. Rarely, an "acclimation trial" may performed in those not suitable for an in-laboratory study. (See 'Patients with CSA and heart failure' above and 'Continuous PAP' above.)

●For patients with CSA and heart failure with a preserved ejection fraction (>45 percent) who fail CPAP, adaptive servo-ventilation (ASV) and bilevel PAP (BPAP) in the spontaneous/timed (S/T) mode are indicated and typically titrated in an attended in-laboratory PSG-monitored setting. We prefer ASV given its better acute efficacy in correcting central apneas. ASV and BPAP are contraindicated in those patients with CSA and a low ejection fraction (≤45 percent) due to a lack of benefit and potential harm; these patients are typically treated with CPAP or CPAP with oxygen supplementation if CSA persists despite maximized medication therapy. (See 'Patients with CSA and heart failure who fail continuous PAP' above.)

●For patients whose central apneas are due to opioid use, CPAP is also the typical first-line of therapy and is titrated in an attended in-laboratory setting. In patients whose sleep-disordered breathing fails to resolve on CPAP, ASV titration using a device with an auto-titrating expiratory pressure is typically performed. Finally, for those who fail ASV therapy, titration with BPAP-S/T mode can be performed. This population generally has titrations performed during in-laboratory PSG and may require higher levels of pressure support to control concurrent hypoventilation. Testing in these populations typically involves monitoring of transcutaneous carbon dioxide levels. (See 'Patients with CSA and opioid use' above.)

●For patients with OSA who have treatment-emergent CSA (TE-CSA), ASV titration during in-laboratory PSG is appropriate, either immediately upon recognition of TE-CSA if time permits during the same night, or during a separate subsequent polysomnogram. Initial treatment with CPAP and reassessment for residual TE-CSA after two to three months is also a reasonable approach. Finally, CPAP with an addition of low-flow oxygen is sometimes done especially in those patients who have another indication for nocturnal oxygen supplementation. (See 'Patients with treatment-emergent CSA' above and "Treatment-emergent central sleep apnea", section on 'Treatment'.)

●For patients with rare etiologies for CSA, choosing a method for PAP titration is individualized and dependent upon the modality chosen to deliver PAP as well as the underlying cause. (See 'Patients with rare etiologies of CSA' above.)

●For patients with central hypoventilation syndromes (eg, obesity hypoventilation, congenital alveolar hypoventilation), central apneas tend to be a minor component of the sleep-disordered breathing. BPAP is first-line therapy and an in-laboratory BPAP titration is generally performed in this population. (See 'Patients with central hypoventilation syndromes' above.)