Sleep Apnea and Atrial Fibrillation
Apnea and atrial fibrillation travel together, and treating the apnea is associated with much less AF coming back after ablation. The one randomised trial in that evidence found no difference.
Obstructive sleep apnea and atrial fibrillation occur together far more often than chance explains, and the mechanism is plausible enough that the association is not seriously disputed.
The useful question is whether treating the apnea does anything about the arrhythmia. The evidence there is genuinely encouraging and genuinely weaker than the headline numbers suggest, and both halves belong in the answer.
Why the two go together
Several routes, all pointing the same way.
Mechanical stretch. Attempting to breathe against a closed airway generates strongly negative intrathoracic pressure, which stretches the atria. Atrial stretch is an established driver of electrical remodelling and arrhythmia — the same mechanism behind apnea's effect on overnight urine production. Nocturia and sleep apnea.
Autonomic surges. Each event ends in an arousal with a sympathetic discharge. Repeated hundreds of times nightly, that is a lot of adrenergic stimulation delivered to atrial tissue.
Intermittent hypoxia, driving oxidative stress and inflammation, both implicated in atrial remodelling. The depth and duration of desaturation may matter more than event frequency. What hypoxic burden measures.
Shared risk factors — obesity, hypertension, age — which is the confounder any observational study has to contend with.
What the treatment evidence shows
The best-studied question is narrow and clinically important: after catheter ablation for atrial fibrillation, does treating coexisting sleep apnea reduce the chance the AF comes back?
A 2025 meta-analysis pooled 1 randomised controlled trial and 10 observational studies, covering 1,536 patients with obstructive sleep apnea who underwent catheter ablation for AF, of whom 53% used CPAP, with follow-up from 3 to 42 months (Journal of Interventional Cardiac Electrophysiology, 2025).
AF recurrence was significantly lower with CPAP: odds ratio 0.37 (95% CI, 0.23–0.58; p < 0.01) [J Interv Card Electrophysiol 2025].
The subgroups pointed the same way:
- Studies with follow-up of 18 months or longer: OR 0.28 (95% CI, 0.18–0.46; p < 0.01) [ibid]
- Studies where at least half the patients had persistent AF: OR 0.31 (95% CI, 0.16–0.61; p < 0.01) [ibid]
An odds ratio of 0.37 is a large effect. Taken alone it would be a strong argument for treating apnea before or alongside ablation.
The caveat the authors state themselves
It is in their conclusion, and it changes how much weight the number carries.
Most of the available data is observational, and the only randomised controlled trial showed no difference between groups — so the results should be interpreted with caution [J Interv Card Electrophysiol 2025].
That is a serious qualification and it deserves to be read rather than skipped past.
The reason it matters is selection. In observational data, the people who use CPAP are not a random subset of people with apnea. They are more engaged with their healthcare, more likely to take other medications as prescribed, more likely to attend follow-up, and plausibly healthier in ways that independently predict staying in sinus rhythm. Every one of those differences pushes the odds ratio in the direction observed.
When the single trial that removes selection by randomising found nothing, the honest reading is that some of that 0.37 is the treatment and some is the kind of patient who uses it — and the available evidence cannot say how the split falls.
Tier 2. A consistent, large observational association with a plausible mechanism, undercut by the only randomised evidence available. How we grade evidence.
How this fits the wider CPAP picture
It is the same pattern appearing across cardiovascular endpoints: strong observational signal, weaker trial confirmation.
A 2026 analysis offered an explanation — CPAP's cardiovascular benefit appears concentrated in high-risk apnea rather than distributed evenly, with the authors reporting benefit "observed alongside harm in low-risk OSA." Enrolling by apnoea-hypopnoea index, a metric that predicts cardiovascular outcomes poorly, would dilute a real effect toward nothing. Does CPAP reduce cardiovascular risk.
If that is right, the AF question may have the same shape: a real benefit in the right patients, invisible in trials that did not select for them.
What to do with this
If you have AF, get screened for sleep apnea. This is the part with no downside. The conditions co-occur heavily, apnea is treatable, and the screening is cheap. It is also standard practice in many cardiology services. Risk check · Home sleep testing.
If you have both and ablation is planned, raise the timing. Whether to treat the apnea first is a reasonable question to ask your electrophysiologist, and the evidence above is what it turns on — strong observational support, one null trial.
Do not expect treating apnea to replace AF treatment. Anticoagulation, rate and rhythm control decisions are unaffected by anything here.
And treat the apnea on its own merits. If you have symptomatic moderate-to-severe apnea, the case for treating rests on sleepiness, function and quality of life, which is firmer ground than arrhythmia prevention. What happens after a diagnosis · Getting used to CPAP.
The line
No consumer product treats obstructive sleep apnea or affects atrial fibrillation. If you have both conditions, nothing sold for snoring belongs in the plan — and quietening snoring without assessment removes the symptom most likely to get the apnea found. Snoring versus sleep apnea · What untreated apnea does.