• Composite of CV death, MI & ischaemic stroke; components of primary composite endpoint; re-vascularisation procedures; all-cause death; new onset atrial fibrillation; new onset diabetes; OSA symptom scores; mood; health-related quality of life. [ Time Frame: Reviewed 6-monthly; average patient follow up, 4 years. ] [ Designated as safety issue: No ]
• In a sub-sample of 600 subjects pathophysiological mechanisms of CPAP-induced CV event reduction will be explored by assessing various intermediate markers of CV risk [ Time Frame: baseline and at 6-months, 2 and 4 years following randomisation ] [ Designated as safety issue: No ]

OSA is a condition in which a person stops breathing for several seconds at a time due to relaxation of the throat muscles. This can occur many times over during sleep. It is known to cause sleepiness and poor concentration during the day. Research indicates that OSA may be a modifiable risk factor for cardiovascular disease due to its association with hypertension, stroke, heart attack and sudden death. The standard therapy for symptomatic OSA is CPAP. CPAP has been shown to effectively reduce snoring, obstructive episodes and daytime sleepiness and to modestly reduce blood pressure and other risk factors for cardiovascular disease. The overall aim of SAVE is to determine if CPAP can reduce the risk of heart attack, stroke or heart failure for people with OSA.

There is increasing evidence to indicate that OSA is an important modifiable risk factor for CV disease including stroke, MI, and heart failure.

Increased nocturnal arterial blood pressure (BP), hypercoagulability, oxidative stress, inflammation, insulin resistance and cardiac arrhythmias are all associated with OSA. These effects are presumed to accelerate the progression of atheromatous disease, particularly within the coronary or cerebral vasculature. Moreover, OSA also appears to increase the risk of sudden death during sleep, which is different from the circadian pattern of sudden death in those without OSA, suggesting that episodes of apnea may have a direct triggering effect for cardiac arrhythmias or MI.

CPAP is now standard therapy for symptomatic OSA, with adherence to treatment comparable to that of other therapies for common chronic diseases. CPAP can eliminate apneas and improve daytime sleepiness, mood and quality of life. Furthermore, short term (1-3 months) randomised controlled trials of CPAP have shown modest reductions in blood pressure (BP) and other markers of CV disease, including C-reactive protein (CRP) and coagulation. However, the epidemiological data is complicated by potential residual confounding factors and the randomised evidence is limited. Thus, a direct causal link between OSA and CV disease remains inconclusive. The management of OSA, therefore, remains principally directed towards symptom control rather than CV risk modification. The present trial aims to test whether long-term use of CPAP can reduce the incidence of CV events. If the trial shows that CPAP treatment of OSA reduces the incidence of CV events it will influence clinical practice toward the early detection and management of OSA, and add CPAP to the range of strategies available for the prevention of CV disease.

• Sleep Apnea
• Cardiovascular Disease

• Device: Continuous Positive Airway Pressure (CPAP)
• Other: Standard care

• Experimental: CPAP plus Standard care of cardiovascular risk factors
• Active Comparator: Standard care alone

Inclusion Criteria:

1. Males and females, any race, and aged between 45 and 75 years

2. Evidence of established coronary or cerebrovascular disease as evident by:

   • Coronary artery disease

     • Previous MI (equal to or greater than 90 days prior to informed consent)
     • Stable angina or unstable angina (equal to or greater than 30 days prior to informed consent) each with documented multi-vessel coronary artery disease or 50% or more stenosis in at least two major coronary arteries on coronary angiography, or positive stress test (ST depression equal to or greater than 2 mm or a positive nuclear perfusion scintigram)
     • Multi-vessel percutaneous angioplasty (PTCA) and/or stent equal to or greater than 90 days prior to informed consent
     • Multi-vessel coronary artery bypass surgery (CABG) >1 year prior to informed consent

   • Cerebrovascular disease

     • Previous stroke (includes definite or presumed cerebral ischaemia/infarction and intracerebral haemorrhage) equal to or greater than 90 days prior to informed consent
     • Previous transient ischaemic event (TIA) of the brain or retina (symptoms <24 hours) with the diagnosis confirmed by a neurologist 30 days to I year prior to informed consent
3. Patients have moderate-severe OSA (equivalent to apnea plus hypopneas index [AHI] >30 per hour of sleep) as determined by a > 4% oxygen dip rate > 12/ h on overnight testing using the ApneaLinkTM device and confirmed by the SAVE core lab in Adelaide upon receipt of the ApneaLink data
4. Patients are able and willing to give appropriate informed consent

Exclusion Criteria:

Patients will be excluded from entry if ANY of the criteria listed below are met:

1. Any condition that in the opinion of the responsible physician or investigator makes the potential participant unsuitable for the study. For example,

   • co-morbid disease with severe disability or likelihood of death within the next 2 years
   • significant memory, perceptual, or behavioural disorder
   • neurological deficit (eg. limb paresis) preventing self administration of the CPAP mask
   • residence sufficiently remote from the clinic to preclude follow-up clinic visits
2. Any planned coronary or carotid revascularisation procedure in the next 6 months

3. Severe respiratory disease defined as

   • severe chronic obstructive pulmonary disease (FEV1/FVC < 70% and FEV1 < 50% predicted), or
   • resting, awake SaO2 < 90% by ApneaLinkTM device
4. New York Heart Association (NYHA) categories III-IV of heart failure
5. Stroke due to subarachnoid haemorrhage
6. Other household member enrolled in SAVE trial or using CPAP
7. Prior use of CPAP treatment for OSA

8. Increased risk of a sleep-related accident and/or excessive daytime sleepiness, defined by any one of the following:

   • driver occupation (eg truck, taxi)
   • 'fall-asleep' accident or 'near miss' accident in previous 12 months
   • high (> 15) score on the Epworth Sleepiness Scale
9. Severe nocturnal desaturation documented on the ApneaLinkTM device as > 10% overnight recording time with arterial oxygen saturation of < 80%

10. Cheyne-Stokes Respiration (CSResp)

    • CSResp identified on ApneaLinkTM nasal pressure recording by typical crescendo-decrescendo pattern of respiration with associated apneas and/or hypopneas in the absence of inspiratory flow limitation.
    • patients excluded if > 50% of nasal pressure - defined apneas and hypopneas judged to be due to CSResp.

• Respironics
• ResMed
• Fisher and Paykel Healthcare
• The George Institute