Does Treating Sleep Apnea Prevent Heart Attacks?

The strongest randomized evidence we have does not show that treating obstructive sleep apnea with CPAP prevents heart attacks. In the SAVE trial, published in the New England Journal of Medicine in 2016, adults with moderate-to-severe sleep apnea and established heart or blood-vessel disease who were assigned to CPAP had no lower rate of cardiovascular death, heart attack, or stroke than those who received usual care alone. That finding surprised many clinicians, because years of observational research had tied sleep apnea tightly to cardiovascular risk. The distance between those two bodies of evidence is where the real lesson sits.

The observational case looked compelling

For decades, cohort studies painted a consistent picture. People with untreated obstructive sleep apnea developed more hypertension, coronary disease, atrial fibrillation, heart failure, and stroke than people without it, and the risk appeared to rise with severity. A 2025 meta-analysis of prospective cohorts, pooling more than 25,000 participants followed for a median of about nine years, reported that sleep apnea was associated with substantially higher cardiovascular risk, with a graded relationship running from mild to severe disease.

The biology made the association easy to believe. Each apnea produces a fall in blood oxygen followed by a surge of sympathetic nervous activity as the brain forces the body to breathe. Repeated hundreds of times a night, that cycle drives blood-pressure spikes, oxidative stress, inflammation, and endothelial dysfunction, all recognized contributors to atherosclerosis. If a condition raises cardiovascular risk through those pathways, relieving it with CPAP ought to lower the risk. That was the hypothesis the SAVE investigators set out to test directly.

What SAVE actually measured

SAVE randomly assigned 2,717 adults between 45 and 75 years old, all with moderate-to-severe obstructive sleep apnea plus known coronary or cerebrovascular disease, to CPAP plus usual care or to usual care alone. The primary endpoint was a composite of death from cardiovascular causes, heart attack, stroke, or hospitalization for unstable angina, heart failure, or transient ischemic attack. Over a mean follow-up of 3.7 years, the primary endpoint occurred in 17.0 percent of the CPAP group and 15.4 percent of the usual-care group. The hazard ratio was 1.10, with a 95 percent confidence interval of 0.91 to 1.32 and a p-value of 0.34.

Read that carefully. The point estimate sits slightly above 1.0, and the confidence interval crosses it in both directions. The trial did not simply fail to reach significance on a favorable trend; the data were genuinely flat, with no signal that CPAP reduced hard cardiovascular events in this population. What CPAP did do was meaningful in its own right: it reduced snoring and daytime sleepiness and improved mood and health-related quality of life. Those are legitimate reasons to treat sleep apnea. Preventing a heart attack was not among the demonstrated benefits.

The adherence problem

Here is the complication that keeps this question open. CPAP only helps while it is worn, and in SAVE the average participant used the device just 3.3 hours per night. That is well below the seven or eight hours a person actually spends asleep, so for much of the night the airway was collapsing exactly as it would without treatment. A therapy applied for less than half the sleep period is a weak test of the underlying idea.

Investigators tried to work around this. In a propensity-matched analysis comparing participants who used CPAP at least four hours a night against similar usual-care patients, there was a suggestion of fewer cerebrovascular events. A later 2023 meta-analysis pooling SAVE with the ISAACC and RICCADSA trials found that adherence of four or more hours nightly was associated with lower recurrent cardiovascular risk, with a hazard ratio near 0.69. That sounds encouraging until you notice what kind of comparison it is. Once you select people by how faithfully they used a treatment, you are no longer comparing randomized groups. People who stick with CPAP tend to differ from those who abandon it in diet, activity, medication use, and general health engagement. The meta-analysis authors said this plainly, cautioning that on-treatment analyses are not protected against selection bias, a pattern often called the healthy-adherer effect. Even patients who faithfully take a placebo tend to do better than those who do not, which suggests adherence is partly a marker of the kind of patient who does well, as much as a cause of it.

Why observation and randomization diverged

Two explanations compete, and both are probably partly true. The first is that the observational association was inflated by confounding and reverse causation. Sleep apnea travels with obesity, older age, and metabolic disease, and sicker patients are also more likely to be referred for a sleep study and diagnosed. Some of the apparent cardiovascular risk of apnea may belong to the company it keeps.

The second explanation is that CPAP genuinely helps a subset of patients, but the benefit was diluted by short nightly use, by a population already on strong cardiovascular medications, and by the exclusion of people with severe daytime sleepiness, who could not ethically be randomized to no treatment and who may have the most to gain. A later machine-learning analysis of the ISAACC trial searched for subgroups and found opposing effects hidden inside the flat average. CPAP appeared protective in patients whose apneas were shorter, yet in a subgroup with longer breathing events and high cholesterol it was associated with more cardiovascular events rather than fewer. An exploratory result like that fits the idea that an average near zero can conceal benefit for some patients and harm for others. None of this rescues the original claim. It reframes it: whether CPAP prevents cardiovascular events remains unproven rather than disproven, and the honest answer is that a well-run trial did not find the effect that the associations predicted.

What this means if you have sleep apnea

Treating obstructive sleep apnea is worthwhile for how it makes you feel and function, and controlling the blood-pressure and metabolic consequences of apnea remains sensible. But CPAP should not be sold as a proven heart-attack preventive on the strength of the trial evidence, and the cardiovascular protection it offers, if any, likely depends on actually wearing the device through the night rather than for a few hours. This article is educational and is not medical advice; decisions about diagnosis and treatment belong with your own clinician, who can weigh your symptoms and overall cardiovascular risk. The larger takeaway reaches past sleep medicine: a strong, biologically plausible association is a reason to run the experiment, not a substitute for it.