Study Recap: Does Sticking With CPAP Actually Protect Your Heart?

If you've been handed a CPAP machine, you've probably also been handed a promise: use this, and you'll lower your risk of a heart attack or stroke down the road. It's a reasonable thing to tell someone. Sleep apnea is tightly linked to cardiovascular disease, and CPAP is the gold-standard treatment. The logic almost writes itself.

But here's the uncomfortable thing about that promise: when researchers ran the big randomized trials to prove it, the results didn't cooperate. Trial after trial assigned people with sleep apnea to CPAP or usual care, followed them for years, and found no clear reduction in cardiovascular events. For a treatment everyone believed in, that was a genuine head-scratcher.

A 2023 analysis published in JAMA went back into those trials and found something the original studies had partly obscured — and the answer comes down to a single word that turns out to matter enormously: adherence.

What the study actually did

The paper, led by Manuel Sánchez-de-la-Torre and colleagues, was published in JAMA in October 2023 ^[1]. It wasn't a brand-new trial. Instead, it was something arguably more powerful: an individual-participant-data meta-analysis, which pools the raw, patient-level data from multiple existing trials rather than just combining their summary results. That lets researchers re-analyze everyone together as if they'd been in one big study.

They pulled together three of the most important randomized trials of CPAP and heart health — known by their acronyms SAVE, ISAACC, and RICCADSA — for a combined total of 4,186 patients ^[1]. These weren't healthy volunteers. The group skewed older (average age 61), mostly male, mostly overweight, and crucially, all of them already had established cardiovascular disease. They had moderate-to-severe sleep apnea (average AHI of about 31 events per hour) and a high burden of other risk factors — 71 percent had hypertension ^[1]. In other words, this was a high-risk population being studied for secondary prevention: people who'd already had cardiovascular trouble and were trying not to have more.

Half were assigned to CPAP, half to usual care, and they were followed for an average of about 3.25 years. The outcome the researchers tracked was the first MACCE — a major adverse cardiac or cerebrovascular event, the umbrella term covering things like heart attack, stroke, and cardiovascular death ^[1].

The headline result has two halves

Here's where it gets interesting, and where a careless summary would go wrong.

Half one — the disappointing part. When the researchers compared everyone assigned to CPAP against everyone assigned to usual care — the standard "intention-to-treat" analysis that respects the original randomization — there was essentially no difference. The hazard ratio was 1.01, which is statistical shorthand for "no detectable effect" ^[1]. On its face, this echoes the frustrating earlier findings: hand people a CPAP machine, and on average their cardiovascular outcomes don't improve.

Half two — the revealing part. But the researchers also ran an "on-treatment" analysis, using statistical methods designed to estimate the effect of actually using CPAP consistently — defined as at least 4 hours per night. And there, a real signal emerged. Good adherence was associated with a 31 percent lower risk of recurrent cardiovascular events (hazard ratio 0.69) ^[1].

The gap between those two numbers is the whole story. The reason the simple comparison showed nothing wasn't that CPAP doesn't work — it's that a lot of people assigned to CPAP barely used it. Across the pooled group, average usage was only about 3.1 hours per night, below the 4-hour threshold generally considered the minimum effective "dose" ^[1]. When you average together the people who used it faithfully and the people who left it in the closet, the benefit of the first group gets diluted into invisibility by the second.

It's a bit like studying a medication where half the participants never filled the prescription, then concluding the drug doesn't work. The problem isn't the pill. It's that you measured prescribing, not taking.

The catch every honest recap has to mention

Before anyone treats that 31 percent figure as gospel, there's an important caveat, and the researchers themselves were upfront about it. An on-treatment analysis is no longer a clean randomized comparison — it's observational, and observational data carry a well-known trap in this field called the "healthy adherer" effect ^[2].

The concern is this: people who use a demanding nightly therapy reliably tend to be different from people who don't, in ways that have nothing to do with the therapy. Faithful CPAP users may also be the kind of people who take their other medications on schedule, keep their appointments, exercise, and eat better. So when their hearts do better, it's genuinely hard to know how much credit goes to the CPAP and how much goes to the broader conscientiousness that made them stick with it.

Good statistical methods (the study used marginal structural models with inverse-probability weighting — a technique built specifically to reduce this kind of bias) can shrink that problem but can't fully eliminate it ^[1]. So the fair conclusion isn't "CPAP definitely cuts your cardiac risk by 31 percent." It's "consistent CPAP use is associated with meaningfully better cardiovascular outcomes, the biological story is plausible, but residual confounding means we can't call it airtight proof." That's a more modest claim — and a more accurate one.

Where the evidence is firmer: blood pressure

If the cardiovascular-event evidence is suggestive-but-hedged, the blood-pressure evidence is sturdier — and it helps explain why adherence might matter for the heart.

Sleep apnea drives blood pressure up through a fairly well-understood chain. Each time the airway collapses and oxygen drops, the sympathetic nervous system fires, releasing a surge of stress hormones that constrict blood vessels and raise pressure. Repeated hundreds of times a night, this keeps the cardiovascular system in a state of chronic overactivation, and the elevated pressure often persists into the daytime. It's one reason sleep apnea is a recognized cause of resistant hypertension — high blood pressure that won't come down despite multiple medications.

CPAP measurably lowers blood pressure, though the size of the effect depends on who you are. In the general population of people with sleep apnea and hypertension, CPAP produces a modest reduction — typically in the range of 2 to 3 mmHg ^[3]. That sounds small, and for any one person it is. But across a population, even a couple of points of systolic blood pressure shifts the rate of strokes and heart attacks.

The effect is considerably larger in the people who need it most. A meta-analysis focused specifically on patients with resistant hypertension and sleep apnea found that CPAP reduced 24-hour ambulatory systolic pressure by about 7.2 mmHg and diastolic by about 5.0 mmHg ^[3]. For someone whose blood pressure won't budge on three or four drugs, a 7-point drop from treating their undiagnosed apnea is a clinically meaningful result. And because 24-hour ambulatory pressure is one of the better predictors of future cardiovascular damage, this is exactly the kind of change that plausibly translates into fewer events over time.

There's also a useful practical finding tucked into this literature: combining CPAP with weight loss lowers blood pressure substantially more than CPAP alone — one meta-analysis found roughly 8 to 9 mmHg of additional systolic reduction from adding effective weight loss ^[4]. Which lines up with everything we know about how intertwined sleep apnea, weight, and cardiovascular risk really are.

What this actually means for you

Pulling the threads together, here's the honest bottom line a sleep specialist would stand behind:

The machine only works if you wear it. This is the single most important takeaway, and it's not a throwaway line. The entire reason the big trials looked disappointing is that average use was too low. The 4-hour-a-night threshold isn't arbitrary bureaucratic box-ticking — it's roughly where the cardiovascular and blood-pressure benefits start showing up in the data. If you're using CPAP two hours a night, you're getting a fraction of what it can offer.

The benefit is most established for blood pressure, especially if you have hypertension that's hard to control. If your pressure is stubbornly high on multiple medications and you haven't been screened for sleep apnea, that's a conversation worth having — treating an unrecognized apnea is sometimes the missing piece.

The cardiovascular-event benefit is probable but not proven. Consistent use is associated with fewer heart attacks and strokes in high-risk patients, the mechanism makes biological sense, but the "healthy adherer" problem means we should hold the precise number loosely. This is an area of active research, and the trials are being redesigned to better isolate the effect of adherence itself.

If you're struggling to tolerate CPAP, that's a problem to solve, not a reason to quit. Mask discomfort, dryness, claustrophobia, and pressure intolerance are common and usually fixable — different mask styles, humidification, pressure adjustments, or alternative therapies for people who truly can't adapt. Given that the entire benefit hinges on consistent use, getting the adherence problem addressed is arguably the highest-value thing you can do.

The takeaway

The 2023 JAMA analysis is, in a sense, a study about the difference between having a treatment and using one. The original trials seemed to say CPAP doesn't protect the heart. A closer look says something more precise and more useful: CPAP, used consistently, is associated with meaningfully better cardiovascular outcomes and clearly lowers blood pressure — while CPAP that sits unused does nothing at all, because of course it doesn't.

That's not as clean as the promise printed on the box. But it's the truth the data actually support, and it points to something you can act on. The hours you put in are the part you control.

This article summarizes published research for general education and isn't medical advice. Decisions about CPAP, blood pressure medication, or sleep apnea treatment should be made with your own clinician, who knows your history.