Does Treating Sleep Apnea Prevent Heart Attacks?

Short version: the best randomized evidence we have does not show that treating obstructive sleep apnea with a CPAP machine prevents heart attacks. Sleep apnea is worth treating for how it makes people feel, but the case for it as heart protection did not survive the one trial built to test it directly. Understanding why is a small master class in how medical evidence can point two different directions at once.

Key points#

• The SAVE trial randomly assigned more than 2,700 adults with moderate-to-severe sleep apnea and known heart or blood-vessel disease to CPAP or usual care, and found no reduction in cardiovascular death, heart attack, or stroke.
• Decades of observational studies had linked untreated apnea to hypertension, coronary disease, atrial fibrillation, and stroke, with risk rising by severity.
• A likely reason the trial came up flat is that participants used CPAP only about 3.3 hours per night, far less than a full night of sleep.
• CPAP still clearly improves snoring, daytime sleepiness, mood, and quality of life. Those benefits are real and reason enough to treat.
• Whether CPAP protects the heart in people who wear it faithfully remains an open question, not a settled one.

Start with the trial, because it is the surprise#

Most explanations of this topic build up to the trial. It is more honest to lead with it, because the trial is the part that upended expectations.

SAVE, published in the New England Journal of Medicine in 2016, enrolled 2,717 adults aged 45 to 75 who had moderate-to-severe obstructive sleep apnea plus established coronary or cerebrovascular disease. Half were assigned to CPAP added to usual care, half to usual care alone. The primary endpoint bundled the outcomes that matter most: death from cardiovascular causes, heart attack, stroke, and hospitalization for unstable angina, heart failure, or a transient ischemic attack.

After a mean of 3.7 years, that 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.

It is worth sitting with those numbers. The point estimate sits a hair above 1.0, and the confidence interval straddles it on both sides. This is not a promising trend that narrowly missed statistical significance. It is a flat result, with no hint that CPAP lowered hard cardiovascular events in this group. What CPAP did deliver was genuine and measurable: less snoring, less daytime sleepiness, better mood, better health-related quality of life. Those are good reasons to use it. Preventing a heart attack was not on the list of demonstrated benefits.

Why almost everyone expected the opposite#

The flat result landed as a shock precisely because the prior evidence had been so consistent in the other direction.

For years, cohort studies told the same story. People with untreated obstructive sleep apnea went on to develop more high blood pressure, coronary disease, atrial fibrillation, heart failure, and stroke than people without it, and the more severe the apnea, the steeper the risk. A 2025 meta-analysis pooling more than 25,000 people followed for a median of roughly nine years found sleep apnea tied to substantially higher cardiovascular risk, with a graded climb from mild to severe disease.

The mechanism made the link easy to accept. Picture the night of someone with severe apnea. Each time the airway collapses, oxygen drops and the brain jolts the body awake enough to resume breathing, firing off a surge of stress-hormone activity. Repeat that a few hundred times before dawn and you get repeated spikes in blood pressure, plus inflammation, oxidative stress, and damage to the lining of blood vessels, all known drivers of artery disease. If a condition harms the heart through those channels, then switching it off with CPAP ought to help. That was exactly the logic SAVE set out to test.

The asterisk that keeps the question open#

Here is the wrinkle. CPAP only works while the mask is on, and in SAVE the average participant wore it just 3.3 hours a night. A person who sleeps seven or eight hours is therefore leaving the airway to collapse untreated for most of the night. Testing a therapy that is switched off for more than half of sleep is a weak test of whether the therapy can work at all.

Investigators tried to see past this. In an analysis matching participants who used CPAP at least four hours a night against comparable usual-care patients, there was a hint of fewer strokes. A 2023 meta-analysis combining SAVE with the ISAACC and RICCADSA trials found that four or more hours of nightly use lined up with lower recurrent cardiovascular risk, with a hazard ratio near 0.69.

That sounds reassuring until you notice what kind of comparison it is. Once you sort people by how faithfully they used a treatment, the randomization is gone. People who stick with a nightly mask tend to differ from those who abandon it in diet, activity, medication habits, and overall involvement in their own care. The meta-analysis authors flagged this directly: on-treatment comparisons are not shielded from selection bias, a pattern known as the healthy-adherer effect. The tell is that even patients who reliably take a placebo pill tend to fare better than those who do not, which suggests that faithful use is partly a marker of the sort of patient who does well anyway, rather than the cause of doing well.

Reconciling the two stories#

Two explanations compete, and both are probably part of the truth.

The first is that the observational signal was inflated. Sleep apnea keeps company with obesity, older age, and metabolic disease, and sicker patients are also the ones more likely to be sent for a sleep study in the first place. Some of the cardiovascular risk pinned on apnea may really belong to the conditions traveling alongside it.

The second is that CPAP does help a subset of patients, but the benefit was watered down in SAVE by short nightly use, by a population already well treated with cardiovascular medications, and by the deliberate exclusion of people with severe daytime sleepiness, who could not ethically be assigned to no treatment and who may have the most to gain. A later machine-learning analysis of the ISAACC trial went looking for hidden subgroups and found opposing effects concealed inside the flat average: CPAP looked protective in patients whose breathing pauses were shorter, yet in a subgroup with longer pauses and high cholesterol it tracked with more cardiovascular events, not fewer. That is an exploratory finding, not a conclusion, but it fits the idea that an average near zero can hide benefit for some and harm for others.

None of this revives the original claim. It resets it. Whether CPAP prevents cardiovascular events is unproven rather than disproven, and the plain reading is that a well-run trial did not find the effect the associations had predicted.

What this means if you have sleep apnea#

Treat sleep apnea for the reasons that clearly hold up: better sleep, less daytime fatigue, steadier mood, and better control of the blood-pressure and metabolic problems that ride along with it. Just do not accept CPAP being marketed as a proven heart-attack preventive, because the trial evidence does not support that pitch. If CPAP does protect the heart at all, that protection almost certainly depends on wearing the device through the whole night rather than for a few hours.

Decisions about diagnosis and treatment belong with your own clinician, who can weigh your symptoms and your overall cardiovascular risk.

The wider lesson reaches well past sleep medicine. A strong, biologically sensible association is a reason to run the experiment, not a replacement for running it.