How Untreated Sleep Apnea May Worsen Seizure Control in Adults With Epilepsy

How sleep apnea and epilepsy are connected

Sleep apnea and epilepsy overlap in roughly one adult in three. A 2017 meta-analysis by Lin and colleagues in Sleep & Breathing found mild-to-severe obstructive sleep apnoea in 33.4% of people with epilepsy (95% CI 20.8-46.1%). People with epilepsy were more susceptible than healthy controls (odds ratio 2.36), men more than women (odds ratio 3.00).

Untreated night-time breathing disrupts sleep where seizure control is fragile. Repeated micro-awakenings and cortical arousals break up slow-wave NREM N3 sleep, while intermittent hypoxaemia, meaning repeated drops in blood oxygen, loads the brain all night. Sleep fragmentation is widely described as lowering the seizure threshold, and interictal epileptiform discharges, the abnormal EEG spikes between seizures, are commonly reported to rise when sleep is unstable.

Nothing cited here shows that apnoea causes epilepsy. The practical point is that untreated apnoea is a modifiable load on an already irritable brain, much as it is after a stroke, where screening survivors is now standard practice.

What the trials really show about treating sleep apnea and epilepsy

Positive airway pressure reliably fixes the breathing, the seizure signal is real, and the one randomised trial was too small to prove it.

The only sham-controlled randomised trial enrolled 68 adults with medication-resistant epilepsy and randomised 35, with 22 on therapeutic CPAP and 13 on sham (Malow et al., Neurology, 2008). The apnoea-hypopnoea index fell from 16.1 to 3.1 events per hour on therapeutic pressure, against 19.4 to 18.0 on sham (p<0.0001). A seizure reduction of at least 50% occurred in 28% of the treated group and 15% of the sham group. That difference was not statistically significant (p=0.40).

Observational data carries the stronger signal. Among 132 adults with epilepsy referred for polysomnography, 57.6% had obstructive sleep apnoea, and 73.9% of those treated with positive airway pressure achieved at least a 50% seizure reduction, against 14.3% of the untreated apnoea group (Pornsriniyom et al., Epilepsy & Behavior, 2014). The 2017 Lin meta-analysis also found better seizure control on CPAP than without it (odds ratio 5.26). Such cohorts compare people who took up therapy with those who did not, and the groups differ in more than breathing.

Why adherence, not the prescription, changes seizure counts

Adherence separates the studies where seizures fell from those where nothing happened. Vendrame and colleagues (Epilepsia, 2011) split patients by actual use rather than prescription. CPAP-compliant patients went from 1.8 to 1.0 seizures per month (p=0.01). Non-compliant patients went from 2.1 to 1.8, not a significant change (p=0.36).

The seizure-freedom figures are starker. In that 2011 series, 16 of 28 compliant patients became seizure-free, against 3 of 13 non-compliant patients (relative risk 1.54, p=0.05).

So the question is not whether to try CPAP but whether you can wear it most of the night, most nights. Raise mask type, humidification, pressure comfort and nasal congestion at the first appointment, because those details decide whether therapy survives month three.

How a bed partner can tell an apnoea from a nocturnal seizure

A bed partner's observation is often what gets the sleep study ordered. Obstructive apnoeas repeat many times an hour, usually with loud snoring between them. Nocturnal seizures behave differently, and four things separate them.

Which screening questionnaire to ask for

Ask for the STOP-BANG questionnaire or the Sleep Apnea Scale of the Sleep Disorders Questionnaire (SA-SDQ) rather than the Epworth Sleepiness Scale alone. The SA-SDQ was the screening tool used in the 2008 randomised trial. STOP-BANG scores snoring, tiredness, observed apnoeas, blood pressure, body-mass index, age, neck size and sex, items medication cannot mask.

The Epworth scale measures daytime sleepiness. In epilepsy that signal is unreliable, because sedating anti-seizure medication can flatten sleepiness and hide the symptom the questionnaire depends on. Sleepiness still counts when present: in 370 consecutive video-EEG admissions, pathological excessive daytime sleepiness independently predicted moderate-to-severe sleep-disordered breathing (odds ratio 10.35, p=0.004), alongside older age and higher body-mass index (Sivathamboo et al., Neurology, 2019).

Drug-resistant epilepsy has the strongest case for testing. An earlier study of medically refractory epilepsy found one third had apnoea at a respiratory disturbance index of 5 or more, and 13% above an index of 20 (Malow et al., Neurology, 2000).

How medication and vagus nerve stimulation affect the airway

Some epilepsy treatments make night-time breathing worse. Sedating anti-seizure medication (ASM), including benzodiazepines, and drugs linked to weight gain such as valproate, gabapentin and perampanel act on the two things that matter most for the upper airway: muscle tone and neck size.

Vagus nerve stimulation is the more specific issue. VNS is linked to both obstructive and central apnoeas, and rapid-cycling settings can make that pattern worse. So new or worsening snoring after implantation is first a device-parameter conversation with your epilepsy team, not automatically a CPAP referral. Stimulation settings can often be adjusted.

None of this justifies changing medication on your own. Withdrawal carries clinical risks and, in the EU, its own driving restrictions.

SUDEP, night-time breathing and what the numbers actually say

SUDEP means sudden unexpected death in epilepsy, and it is usually the real reason people search this at three in the morning. People with epilepsy have a 24-fold increased risk of sudden death versus the general population. Incidence is roughly 0.22 to 1.2 per 1,000 patient-years in general epilepsy, rising to 3 to 9 in drug-resistant epilepsy and up to 18 per 1,000 with frequent generalised tonic-clonic seizures (Translational Research Symposium on SUDEP, Epilepsy Currents, 2026).

Breathing is central to the mechanism. In that same 2026 review, most SUDEP events occurred at night after a generalised tonic-clonic seizure, with early postictal apnoea, then slowing of the heart, then terminal asystole. Postconvulsive central apnoea develops in 22% of seizures.

Now the boundary that matters. Postconvulsive central apnoea is a seizure-driven failure of respiratory drive, not obstructive sleep apnoea, and it is not treated with the same tools. No study cited here shows that treating obstructive apnoea reduces SUDEP risk. Cutting convulsive seizures, especially night-time ones, has the clearest link.

What happens between the neurologist and the sleep study in Europe

The pathway is short in principle and slow in practice. Your neurologist or GP refers you to a sleep physician or pneumologist, who orders either home respiratory polygraphy, a portable overnight recording of airflow, oxygen and breathing effort, or full in-lab polysomnography with EEG channels.

Polysomnography is preferred when seizures are part of the question, because only the in-lab study records brain activity alongside breathing. A home sleep apnoea test is faster and often enough for straightforward apnoea, but it cannot characterise nocturnal seizures. If moderate or severe disease is confirmed, pressure is set by an automatic device or a CPAP titration study.

National gatekeeping is where the waiting happens. In France, treating apnoea with PPC (pression positive continue) or an orthèse d'avancée mandibulaire, a mandibular advancement device, requires an accord préalable, prior authorisation from Assurance Maladie, before anything is reimbursed. Continued cover is tiered by telemonitored use: LPP tariff bands reference at least 112 hours per rolling 28-day period, roughly four hours a night, with a reduced band below 56 hours. French paperwork calls the index the IAH (indice d'apnées-hypopnées); German readers see Apnoe-Hypopnoe-Index and Schlafapnoe; UK readers see OSAS.

What sleep apnea and epilepsy mean for your EU driving licence

In the EU, epilepsy and obstructive sleep apnoea syndrome are two separate conditions in Annex III of the driving-licence rules, each with its own criteria and review clock. A driver with both must satisfy each independently.

Directive 2014/85/EU, amending Annex III of Directive 2006/126/EC, writes the bands into law: moderate OSAS is an apnoea-hypopnoea index of 15 to 29, severe is 30 or more. Licences require adequate control of the condition plus compliance with appropriate treatment, and Member States transposed this by 31 December 2015.

Epilepsy sits in section 12 of Annex III as amended by Directive 2009/113/EC, defined as two or more seizures less than five years apart. Driving is allowed six months after a first unprovoked seizure with medical assessment, and a neurologist should assess syndrome and seizure type.

One clause matters especially here. A driver who has never had seizures other than during sleep can be declared fit to drive, provided that pattern has lasted the seizure-free period required for epilepsy. Anything destabilising your nights therefore carries licence consequences, not only clinical ones. The directive also allows advice not to drive from the start of medication withdrawal and for six months after. National rules vary, so confirm with your licensing authority.

What if your sleep study comes back mild

This is the most likely outcome, and almost nothing prepares you for it. The 33.4% pooled figure covers mild through severe disease combined, and only a minority sit above the AHI 15 line where EU driving rules and reimbursement schemes engage. Many people who get tested are told they have snoring or mild apnoea, then offered nothing.

Be clear about the boundary. If your result is moderate or severe apnoea, CPAP is the treatment with the evidence behind it, and no over-the-counter device, nasal stents included, is an alternative to it. Every study linking apnoea treatment to better seizure control used positive airway pressure, mostly in moderate-to-severe disease with drug-resistant epilepsy.

If the result is simple snoring or confirmed mild apnoea, the options thin out fast. Positional therapy, weight management, alcohol timing and treating nasal obstruction are the usual first moves. Back2Sleep sits in that gap: a CE-certified Class I soft silicone intranasal stent that keeps the nasal airway open during sleep, with no prescription, no electricity, no tubing and four sizes in the starter kit. It is a snoring and mild-to-moderate apnoea device, not a seizure treatment, and nothing here suggests it changes SUDEP risk.

Whatever you choose, keep the epilepsy team and the sleep physician in the same loop, and keep counting seizures. Sleep quality is one of the few levers you can pull without touching medication, and its reach shows up in the early cognitive warning signs linked to untreated apnoea.

Frequently Asked Questions