Respiratory conditions

Central Sleep Apnoea

Breathing pauses in sleep because the brain briefly stops sending the signal — not because the throat closes. It is a different problem from the common form of sleep apnoea, and it is usually a sign of something else: heart failure, opioid medication, or altitude.

For patients & health professionals
Cardiomyopathies A–Z of Conditions · 15 of 86 Chronic Bronchitis
Authorship & review
Dr Sean James Ledger, BSc Physio (Hons) MSc PhD FHEA
Director and Principal Physiotherapist
Ahpra registration PHY0002298174
Version
1.0
Last updated
12 August 2026
Next review
12 August 2027
Every guide on this site is reviewed at least once a year, and sooner when the evidence changes.
How these guides are written and reviewed →
Part 1 · In plain language

In the common form of sleep apnoea the throat closes while the brain keeps trying to breathe. Central sleep apnoea is the opposite: the airway is open, but the brain briefly stops sending the signal, so no effort is made at all. It shows up as pauses in breathing during sleep, often in a rising-and-falling pattern, along with broken sleep, waking short of breath, and daytime tiredness. It is much less common than the obstructive form, and it is usually a consequence of something else — most often heart failure, strong pain medicines such as opioids, sleeping at altitude, or a stroke. Because of that, treatment is aimed first at the underlying cause, and the machine used for ordinary sleep apnoea is not always the right answer here.

Definition

Central sleep apnoea is repeated cessation or reduction of breathing during sleep caused by absent or reduced respiratory drive, with no accompanying effort to breathe. This is the defining distinction from obstructive sleep apnoea, in which effort continues against a closed upper airway. The two frequently coexist, and the mix can change during a single night and with treatment.

The main types

TypeSettingKey point
Cheyne–Stokes breathingHeart failure, sometimes strokeCrescendo–decrescendo pattern with a long cycle; marks disease severity
Opioid-relatedLong-term opioid therapyIrregular, ataxic breathing; dose-related
Treatment-emergentAppears when CPAP is started for obstructive apnoeaUsually settles within weeks on continued therapy
High-altitude periodic breathingSleeping above roughly 2,500 mA normal physiological response, not a disease
Disease of the brainstemStroke, tumour, Chiari malformation, neurodegenerative diseaseUncommon; the control centre itself is affected
IdiopathicNo cause foundRare

Pathophysiology

An unstable control loop

Breathing during sleep is governed by a feedback system that senses carbon dioxide and adjusts ventilation. Its stability depends on how strongly it responds to a change (loop gain) and how long the information takes to travel. Where the response is too strong, or the circulation too slow, the system overshoots: the patient over-breathes, carbon dioxide falls below the level that triggers a breath, and breathing stops until it rises again. That oscillation is central apnoea, and it is a control problem rather than a mechanical one.

Why heart failure produces the classic pattern

A failing heart circulates blood slowly, so information about carbon dioxide reaches the brain late; pulmonary congestion simultaneously increases the ventilatory response. The combination of high gain and long delay produces the long, regular crescendo–decrescendo cycle of Cheyne–Stokes breathing. The severity of the pattern tracks the severity of the cardiac disease, which is why it is best understood as a marker of the heart failure rather than as an independent condition.1

Why opioids are different

Opioids suppress the brainstem respiratory centres directly, blunting rather than destabilising the response to carbon dioxide. The result is irregular, ataxic breathing rather than a regular cycle, and the answer is the opioid dose rather than a ventilator setting.

Co-morbidities

Prevalence

Central sleep apnoea accounts for a small minority of sleep-disordered breathing in the general population, but the picture changes entirely in specific groups. It is found in a substantial proportion of patients with symptomatic heart failure with reduced ejection fraction, and in a meaningful share of patients on long-term high-dose opioids. Treatment-emergent central apnoea appears in a small proportion of patients starting CPAP for obstructive apnoea and resolves spontaneously in most of them.2

Causes

Symptoms

What is noticed

Snoring is typically absent or mild, which is a useful distinguishing feature: a patient with witnessed apnoeas who does not snore should raise the question of a central mechanism.

Warning signs

Same-day medical assessmentNew or worsening breathlessness lying flat, waking gasping for air, ankle swelling, or a sudden reduction in exercise tolerance may indicate worsening heart failure rather than a sleep problem alone. In anyone taking opioids, increasing daytime drowsiness or confusion needs urgent review of the dose.

Diagnosis

How it is diagnosed

Diagnosis requires a sleep study that measures respiratory effort as well as airflow — the absence of effort during a pause is what defines a central event. Simple oximetry cannot distinguish central from obstructive apnoea. In-laboratory polysomnography is preferred where central apnoea is suspected, because home studies are less reliable at making the distinction and at characterising the pattern.

What else is assessed

The distinction that matters

Establishing the proportion of central to obstructive events changes treatment fundamentally. Treating predominantly central apnoea as though it were obstructive, or vice versa, is the commonest management error — and in one specific circumstance it is dangerous, as below.

Management

Treat the cause first

This is the principle that distinguishes central from obstructive apnoea. Optimising heart failure therapy — the disease-modifying medications, device therapy where indicated, fluid management — reduces or abolishes Cheyne–Stokes breathing in many patients. Reducing or rotating opioids resolves opioid-related apnoea. Neither is a sleep intervention, and both work better than one.

Positive airway pressure

CPAP is used where obstructive events coexist, and improves some patients with heart failure-related central apnoea. Adaptive servo-ventilation (ASV) is a more sophisticated device that varies support breath by breath to smooth out the oscillation, and it is highly effective at suppressing the events.

An important safety findingAdaptive servo-ventilation is contraindicated in patients with symptomatic heart failure with a reduced ejection fraction and predominantly central sleep apnoea. A large randomised trial found increased cardiovascular mortality in this group despite the device successfully controlling the breathing pattern — a reminder that suppressing an abnormal measurement is not the same as improving an outcome.3

Other options

Supplemental oxygen at night can stabilise breathing in selected patients with heart failure. Transvenous phrenic nerve stimulation, an implanted device that paces the diaphragm during sleep, is available in some centres for moderate to severe central apnoea. Acetazolamide is used occasionally, and is the standard approach for altitude-related periodic breathing. Treatment-emergent central apnoea usually needs nothing beyond continuing CPAP and reassessing.

Living with central sleep apnoea

It is usually part of something larger

Most people with this diagnosis also have heart failure or chronic pain, and the sleep problem is one component of a bigger picture. Managing the underlying condition well is what improves both, and it is worth framing it that way rather than as two separate problems.

Sleep habits

Regular timing, sleeping more upright where breathlessness is worse flat, avoiding alcohol and sedatives, and treating nasal congestion all help. None substitutes for treating the cause.

Living with the machine

Where a device is prescribed, early problems with mask fit, pressure tolerance and dryness are common and almost always solvable — but only if reported. Most people who abandon therapy do so in the first few weeks over problems that had a straightforward fix.

Prognosis

The outlook is largely determined by the underlying condition. In heart failure, the presence and severity of Cheyne–Stokes breathing marks more advanced disease and is associated with worse outcomes — but the evidence indicates it is primarily a marker rather than an independent driver, which is why suppressing it with ASV did not improve survival. Opioid-related apnoea improves as the dose falls. Altitude-related periodic breathing resolves on descent, and treatment-emergent apnoea resolves in most patients within weeks.

Role of the physiotherapist

Recognising it

Physiotherapists in cardiac rehabilitation and chronic pain services see the two highest-risk populations regularly, and often at greater length than the referring clinician. Witnessed apnoeas without snoring, waking short of breath, unrefreshing sleep and disproportionate daytime fatigue are all worth asking about directly — and worth referring on.

Exercise training

Exercise training improves exercise capacity in heart failure and, in several small studies, reduces the severity of sleep-disordered breathing — plausibly by improving cardiac function and circulatory delay. It is one of the few interventions that addresses the mechanism rather than the symptom. Prescribe within heart failure guidelines, monitor for signs of decompensation, and progress on symptoms. See cardiac rehabilitation.

Inspiratory muscle training

Has a supporting role in heart failure where inspiratory muscle weakness is demonstrated, improving capacity and breathlessness. Its effect specifically on central apnoea is unproven. See inspiratory muscle training.

Adherence and education

Where a device has been prescribed, ask about it at every visit. Practical barriers — mask leak, dryness, claustrophobia, pressure intolerance — are the usual reasons for abandonment, and a therapist who asks is often the person who gets them fixed. Reinforce that the cardiac medication and the sleep therapy are working on the same problem.

Part 1 · References

  1. Naughton MT. Cheyne–Stokes respiration: friend or foe? Thorax 2012;67(4):357–360.
  2. Aurora RN, Chowdhuri S, Ramar K, et al. The treatment of central sleep apnea syndromes in adults: practice parameters with an evidence-based literature review and meta-analyses. Sleep 2012;35(1):17–40.
  3. Cowie MR, Woehrle H, Wegscheider K, et al. Adaptive servo-ventilation for central sleep apnea in systolic heart failure (SERVE-HF). N Engl J Med 2015;373(12):1095–1105.

References are numbered in citation order (Vancouver/BMJ style) and were current at the time of writing. Guidelines are living documents — verify against the latest version before clinical use.

How we treat this at the clinic

This is confirmed with a sleep study, and a home study can be arranged without a specialist appointment first.

Sleep Apnoea Check →
Part 2 of 2

Clinical evidence

Part 1 covers the same condition without the technical detail. What follows is the evidence base behind it, written for clinicians — the literature, the reasoning and the gaps.

For clinicians: this summary supports clinical reasoning and is not a protocol. Check current guidelines and local policy before applying it, and read it alongside the key references and guidelines directory.

Framing. Central sleep apnoea in heart failure is the standing example of a surrogate endpoint misleading a whole field. The apnoea–hypopnoea index is reliably reduced by adaptive servo-ventilation; mortality went up.1 The prevailing interpretation is that Cheyne–Stokes breathing in heart failure is largely a marker of circulatory delay and heightened chemosensitivity — and may be partly compensatory — rather than an independent cause of harm. Treatment is therefore directed at the heart, the opioid, or the brainstem lesion, and only then at the breathing pattern.2

Treatment evidence

  • SERVE-HF randomised ASV in patients with HFrEF (EF ≤45%) and predominantly central apnoea. The AHI fell substantially; all-cause and cardiovascular mortality rose. ASV is now contraindicated in this population — the single most important fact in the field, and one that must be checked before any device referral.1
  • CANPAP tested CPAP against control in heart failure with central apnoea. It improved ejection fraction, nocturnal oxygenation and six-minute walk distance but showed no transplant-free survival benefit, and was stopped early. A post-hoc analysis suggested benefit confined to those in whom the AHI was actually suppressed below 15 — hypothesis-generating only, and never confirmed prospectively.3
  • ADVENT-HF, testing peak-flow-triggered ASV in a broader heart failure population including patients with obstructive-predominant disease, did not show harm but also did not demonstrate a survival benefit — leaving ASV without a positive outcome trial in any heart failure phenotype.4
  • Transvenous phrenic nerve stimulation reduced AHI and improved quality of life in the remedē System randomised trial, with sustained effect in follow-up. The trial was powered on AHI, not on mortality — which, after SERVE-HF, is the obvious limitation.5
  • Optimising heart failure therapy reduces central apnoea. Guideline-directed medical therapy, cardiac resynchronisation and management of congestion all lower the AHI, and this is the intervention with the strongest outcome rationale because its benefit is established independently.2
  • Opioid reduction is the treatment for opioid-related central apnoea, with a demonstrated dose–response relationship between morphine-equivalent dose and central event frequency. Device therapy in this group has little evidence and does not address the mechanism.6
  • Exercise training reduces sleep-disordered breathing severity in small heart failure trials, plausibly by improving cardiac output and reducing circulatory delay. Effects are modest and the trials are small, but the mechanism is the right one and the intervention has independent benefit.7

Clinical reasoning

  • Check ejection fraction before any ASV referral. EF ≤45% with predominantly central apnoea is a contraindication. This is the highest-consequence single check in the management of these patients.
  • Effort channels are required. Oximetry and most home studies cannot distinguish central from obstructive events. Where a central mechanism is suspected, laboratory polysomnography with respiratory effort measurement is the appropriate study.
  • Absent snoring with witnessed apnoeas is the most useful bedside discriminator, and sleepiness is often less prominent than in obstructive disease — so an Epworth score in the normal range does not exclude it.
  • Loop gain explains the treatment logic. High controller gain plus long circulatory delay produces instability; interventions that lower gain (oxygen, acetazolamide) or shorten delay (improved cardiac output) stabilise breathing, whereas pressure support alone addresses neither.
  • Distinguish the phenotypes before choosing therapy. Cheyne–Stokes in heart failure, ataxic opioid-related breathing, treatment-emergent apnoea and brainstem disease share a label and share almost nothing else. Treatment-emergent apnoea in particular usually resolves on continued CPAP and should not trigger a device change.
  • A new central pattern warrants a cardiac question. Emergence of Cheyne–Stokes breathing in a known heart failure patient often signals decompensation before symptoms declare it.
  • Physiotherapy contribution is case-finding and mechanism-directed exercise, not device management. Cardiac rehabilitation and persistent-pain services see the two at-risk populations routinely and are well placed to ask the three questions that matter: witnessed pauses, waking short of breath, and unrefreshing sleep.

Evidence gaps

  • No therapy directed at central apnoea has improved survival in any population. Whether any phenotype benefits from suppression remains unknown, and the CANPAP post-hoc suppression hypothesis has never been tested prospectively.
  • The mechanism of harm in SERVE-HF is unexplained — proposed candidates include reduced preload, arrhythmia and loss of a compensatory function — and without it, extrapolation to other devices is guesswork.
  • Central apnoea in HFpEF is largely unstudied despite the prevalence of that phenotype.
  • Opioid-related central apnoea has no randomised treatment evidence; dose reduction is reasoned rather than trialled.
  • Phrenic nerve stimulation has no mortality data, in a field where a device that improved the AHI has already been shown to increase mortality.
  • Exercise training trials are small, short and heterogeneous, and none has been designed with sleep-disordered breathing as its primary endpoint.

References for the clinical evidence summary

  1. Cowie MR, Woehrle H, Wegscheider K, et al. Adaptive servo-ventilation for central sleep apnea in systolic heart failure (SERVE-HF). N Engl J Med 2015;373(12):1095–1105.
  2. Naughton MT. Cheyne–Stokes respiration: friend or foe? Thorax 2012;67(4):357–360.
  3. Bradley TD, Logan AG, Kimoff RJ, et al. Continuous positive airway pressure for central sleep apnea and heart failure (CANPAP). N Engl J Med 2005;353(19):2025–2033.
  4. Bradley TD, Logan AG, Lorenzi Filho G, et al. Adaptive servo-ventilation for sleep apnoea in heart failure (ADVENT-HF): a multicentre, randomised, parallel-group, open-label trial. Lancet Respir Med 2024;12(2):153–166.
  5. Costanzo MR, Ponikowski P, Javaheri S, et al. Transvenous neurostimulation for central sleep apnoea: a randomised controlled trial. Lancet 2016;388(10048):974–982.
  6. Correa D, Farney RJ, Chung F, et al. Chronic opioid use and central sleep apnea: a review of the prevalence, mechanisms, and perioperative considerations. Anesth Analg 2015;120(6):1273–1285.
  7. Yamamoto U, Mohri M, Shimada K, et al. Six-month aerobic exercise training ameliorates central sleep apnea in patients with chronic heart failure. J Card Fail 2007;13(10):825–829.
Important: This page is general information, not medical advice. If your breathing or symptoms change suddenly or severely, seek urgent medical care. For personalised assessment, contact Inspire Clinic.

Corrections: If something on this page is wrong, out of date or unclear, we want to know. Email reception@inspireclinic.au with the page name and what you believe is incorrect. Substantive corrections are made promptly, and the guide’s version and last-updated date are changed to reflect it.