Research Analysis · Sleep & circadian health
CPAP improved symptoms—not cardiovascular events in major trials
In SAVE and ISAACC, CPAP did not significantly reduce the primary cardiovascular outcome in the intention-to-treat analyses. SAVE separately found better sleepiness, quality-of-life and mood outcomes.
Discovery, counterevidence, claim mapping, identifiers, funding, correction state and prose were reviewed before release.
Evidence at a glance
Two large null cardiovascular trials
- Study type
- Large randomized cardiovascular-outcome trials
- Studied in
- Humans
- Participants / sample
- 2,717 adults in SAVE; 1,264 adults in ISAACC
- Publication status
- Peer reviewed; no linked correction or retraction identified as of verification
- Outcome type
- Major cardiovascular events, with symptoms and quality of life reported separately
- Evidence maturity
- Large human randomized evidence in selected cardiovascular populations
- Conflicts / funding
- SAVE and ISAACC disclosed public and industry support; average nightly CPAP use was limited.
- Our assessment
- CPAP treats OSA and can improve symptoms, but the featured trials did not demonstrate fewer major cardiovascular events.
Read this first
Three takeaways
- SAVE reported primary cardiovascular events in 17.0% with CPAP plus usual care and 15.4% with usual care alone.
- ISAACC reported events in 16% and 17%, respectively; its confidence interval also crossed no difference.
- SAVE found symptom and quality-of-life benefits. Those outcomes should not be rewritten as cardiovascular or longevity benefits.
Keep outcomes separate
Symptoms, adherence and events answer different questions
Primary events
The prespecified intention-to-treat cardiovascular outcomes were null in both featured trials.
Symptoms
SAVE reported better sleepiness, mood and quality of life with CPAP.
Adherence
Average nightly use was 3.3 hours in SAVE and 2.78 hours in ISAACC, limiting but not erasing interpretation.
Longevity
Neither trial was designed as a lifespan or biological-aging study.
The bottom line
CPAP is an established treatment for obstructive sleep apnea, and symptom improvement matters. But in two major randomized trials involving people with cardiovascular disease, assigning CPAP did not significantly reduce the primary cardiovascular outcome.
What SAVE tested
SAVE randomized 2,717 adults age 45 to 75 with moderate-to-severe OSA and cardiovascular or cerebrovascular disease. The primary composite outcome occurred in 17.0% of the CPAP group and 15.4% of usual care; adjusted hazard ratio 1.10, 95% CI 0.91 to 1.32.
What ISAACC tested
ISAACC randomized 1,264 non-sleepy adults with OSA after acute coronary syndrome. Cardiovascular events occurred in 16% with CPAP and 17% with usual care; hazard ratio 0.89, 95% CI 0.68 to 1.17.
Why adherence matters
Average CPAP use was modest in both trials. That constrains the question the trials answer, especially for highly adherent users, but does not justify replacing the randomized intention-to-treat results with a favorable subgroup story.
What did improve
SAVE reported improvements in sleepiness, quality of life, mood and work attendance. These are patient-relevant outcomes, but they are distinct from myocardial infarction, stroke, cardiovascular death and lifespan.
What this does not show
The trials do not show that CPAP is useless, that diagnosed OSA should go untreated, or that every cardiovascular subgroup has the same result. They also do not establish longer life or slower biological aging.
Safety and clinical context
Diagnosis, device selection, pressure settings, mask fit and comorbidities belong in clinical care. Trial results from selected non-sleepy cardiovascular populations cannot be turned into individualized stop-or-start advice.
What happens next
Trials with better adherence support, clearer phenotype selection and prespecified cardiovascular endpoints could narrow who may benefit beyond symptoms while preserving the null intention-to-treat record.
Educational boundary
This article explains population trials. It is not personalized medical advice and does not diagnose sleep apnea or tell a person to start, stop or change CPAP. Untreated OSA and treatment decisions require qualified clinical assessment.
Primary sources
Sources, roles and limits
- SAVE randomized cardiovascular-outcomes trial
- Identifier
- PMID:27571048 · DOI:10.1056/NEJMoa1606599 · NCT00738179
- Role
- Primary randomized cardiovascular and symptom support
- Limitation
- Established CVD population; mean use 3.3 hours/night; no lifespan endpoint.
- ISAACC randomized cardiovascular-outcomes trial
- Identifier
- PMID:31839558 · DOI:10.1016/S2213-2600(19)30271-1 · NCT01335087
- Role
- Independent randomized cardiovascular support
- Limitation
- Non-sleepy post-acute-coronary-syndrome population; mean use 2.78 hours/night.
- AASM positive-airway-pressure guideline
- Identifier
- PMID:30736887 · DOI:10.5664/jcsm.7640
- Role
- Treatment and implementation context
- Limitation
- Guidance is not proof of universal cardiovascular or longevity benefit.
Connected topics
Continue through the knowledge system
Continue with real coverage
Keep reading
These are learning paths, not rankings or recommendations.
Another cardiovascular trial where absolute outcomes matter
SPRINT reduced cardiovascular events—not aging itself
Intensive treatment lowered cardiovascular events and mortality in eligible high-risk adults, with more hypotension, electrolyte problems and kidney injury.
Evidence scopeHuman clinical outcomes · large randomized trial
Another large trial separating disease and longevity claims
High-dose flu vaccine reduced influenza—not immune aging
The randomized result was 1.4% versus 1.9% influenza across two seasons, not proof of immune rejuvenation or longer life.
Evidence scopeHuman infection outcome · large randomized trial
A function evidence base with a different intervention
Exercise has strong evidence for function—not proof of slower biological aging
Human trials support mobility and physical function in defined populations. Lifespan and biological-age claims require separate evidence.
Evidence scopeHuman function · mature randomized evidence
Disclosures and history
- August 2, 2026 — source discovery and independent verification completed before prose drafting.
- August 2, 2026 — atomic sleep, function, disease, safety and longevity boundaries reviewed.
- Corrections: no linked correction or retraction identified for the featured primary papers as of publication.