Start · Catalogue · Profile · Table
Sleep BODY HANDBOOK
Sleep · §192
Obstructive Sleep Apnea
If you snore loud enough to be heard through a door, wake unrefreshed after a full night, or have blood pressure that won't budge on medication, your airway is probably collapsing dozens to hundreds of times a night while you sleep. Roughly one adult in six has obstructive sleep apnea; the substantial majority don't know it. Severe untreated cases roughly triple long-term mortality, mostly through quiet cardiovascular damage you cannot feel. The test is two nights at home with a finger sensor and a chest strap. Treatment works — but only if you actually wear it.
Test · Once Evidence Strong Chapter Sleep

The reward is large for the people who have it: the brain fog you've been blaming on age, the 3 PM crash, the morning headache, the high blood pressure that won't quit — most of it clears within weeks. The catch is the device. CPAP is the gold standard, and it's a mask you wear every night for the rest of your life; somewhere between a third and half of users miss the adherence bar in the first months. The dental alternative is gentler but doesn't fully treat severe disease. The first month is the hardest part of any of this.

The pharynx — the tube behind your tongue — is soft. While you're awake, small muscles hold it open. While you sleep, those muscles relax, and in airways that are narrow to start with (a small jaw, a heavy neck, a long soft palate, a tongue that takes up too much room), the negative pressure of trying to draw in air sucks the throat shut. A drink in the evening makes that collapse easier — alcohol relaxes the same airway muscles, which is why cutting the pre-bed drink is one of the simplest first moves. You can't move air. Oxygen drops. Carbon dioxide climbs. After ten to thirty seconds your brain rescues you with a brief micro-arousal — a spike of adrenaline that snaps the airway muscles back open — and you breathe again. You don't remember any of it. It happens dozens to hundreds of times a night.

The damage is three layers deep. Sleep gets shredded into pieces too small for the deep-recovery work that's supposed to happen overnight. The oxygen-drop-and-recover cycle drives oxidative stress and inflammation throughout the body. And the chest-wall pressure swings against a closed throat hammer the heart from the inside, stretching the upper chambers and overloading the left ventricle — which is the mechanical story behind atrial fibrillation and a worn-out heart muscle. None of this feels like anything while it's happening. That is the trick of the disease.

Night-time mouth breathing usually rides along with this. A nose that's chronically blocked — by allergy, a deviated septum, enlarged turbinates — pushes you to breathe through your mouth, which drops the jaw back, which collapses the throat further. Mouth breathing on its own without confirmed apnea still earns a workup; the two conditions overlap heavily.

What we actually know

Sleep apnea is one of the most common chronic conditions on the planet and one of the most undertested. A 2019 literature-based estimate put global prevalence at roughly 936 million adults with mild-or-worse OSA and 425 million with moderate-or-worse Benjafield et al. 2019. The Lausanne HypnoLaus study, which used the more sensitive modern scoring rules, found AHI ≥15 events per hour in half of men and a quarter of women aged 40 and up Heinzer et al. 2015. Best estimates put the undiagnosed share at 80% or more.

The treatment-effect picture for everyday symptoms is unambiguous. A pooled review of dozens of randomized CPAP trials found large improvements in daytime sleepiness, objective wakefulness on standardized tests, and quality-of-life scores — the kind of consistency that anchors a guideline recommendation AASM 2019. The picture for preventing hard cardiovascular events is messier and worth being honest about.

The honest reading isn't CPAP doesn't work. It's that you can't reverse a decade of cardiac damage with three hours of mask time and expect a randomized trial to detect it. Where the strongest signal lives — symptoms in primary-prevention populations, blood pressure, stroke risk at therapeutic adherence, mortality in the long observational cohorts — the case stays clear.

What ignoring it costs you

The version of you with untreated apnea wakes up tired after eight hours in bed and assumes you needed nine. Coffee handles it, mostly. The afternoon crash arrives like clockwork; you schedule meetings around it. Your partner used to mention the snoring; eventually they stopped — they got earplugs, or they moved to the spare room. You don't remember dreams. The morning headache shows up most days and you tell yourself it's stress. Your doctor adds a second blood-pressure pill and then a third without finding a reason. You catch yourself nodding off on a stretch of highway you used to drive without thinking.

That's year one of being symptomatic. Across a decade or two the trajectory bends into the cardiovascular columns: an atrial fibrillation diagnosis that surprises you, hypertension that won't respond, eventually a stroke or a quiet heart attack. The Wisconsin numbers — three to four times the mortality, mostly from the heart — are what that decade looks like at the population level Young et al. 2008. The deaths are not dramatic; you do not feel them coming. You feel tired.

How to test, and what to do if it's positive

Two steps: confirm, then treat. The screening tool is a short eight-question checklist called STOP-BANG — snoring, tiredness, observed pauses in breathing, pressure (high blood pressure), BMI above 35, age over 50, neck circumference over 40 cm, and male sex. Three or more "yes" answers means test; five or more makes severe disease very likely Chung et al. 2008. The actual diagnosis comes from a sleep study. The home version — a finger oximeter, a small airflow sensor under your nose, a strap across the chest — is enough for most adults at high suspicion. The in-lab study is required when you have heart or lung disease, are on chronic opioids, the home test came back negative but symptoms persist, or your doctor suspects something other than straightforward obstructive apnea AASM 2017. If a newer smartwatch has flagged possible apnea overnight, treat that the same way as a high STOP-BANG score — a reason to get the real test, not a diagnosis on its own.

The result is reported as an apnea-hypopnea index (AHI) — the number of breathing events per hour. Mild is 5 to 14, moderate 15 to 29, severe 30 or above. With symptoms, mild already merits treatment; without symptoms, the line tends to move to 15.

CPAP is a small bedside pump that pushes pressurized air through a mask to splint the airway open. The pooled review of randomized trials behind the current treatment guideline shows it cuts daytime sleepiness, lifts objective wakefulness, and improves quality-of-life scores across the board AASM 2019. The dose-response is steep: four hours a night clears most of the daytime sleepiness, but full normalization of cognition and function needs closer to seven and a half Weaver et al. 2007. Modern auto-titrating machines (APAP) adjust pressure breath-by-breath; you don't pick a number, the device picks it for you.

The mandibular advancement device — a custom dental appliance that holds your lower jaw forward a few millimeters — opens the airway at the tongue base. In head-to-head testing it doesn't cut AHI as much as CPAP (residual AHI roughly 7 events higher), but people wear it more reliably, and symptom relief and blood pressure ended up similar Phillips et al. 2013 Bratton et al. 2015. The treatment-effect tie despite the AHI gap is what makes MAD a real first-line option for mild-to-moderate disease.

Weight loss helps in proportion to how much weight comes off — ten percent of body weight tends to drop AHI by twenty to twenty-five percent. Bariatric surgery puts roughly two-thirds of obese OSA patients into remission. The newest entrant is the GLP-1/GIP class: a 52-week trial of tirzepatide in 469 adults with moderate-to-severe OSA and obesity cut AHI by about 25 events per hour, with parallel improvements in weight, blood pressure, and inflammatory markers Malhotra et al. 2024. For people who can't tolerate CPAP and have the right anatomy (tongue-base collapse, BMI under about 32), an implanted nerve stimulator (Inspire) advances the tongue during inspiration; the pivotal trial cut AHI from 29 to 9 Strollo et al. 2014. And for the subset whose apnea is mostly a back-sleeping problem, sleeping position is its own lever — staying off your back through the night eases positional apnea on its own. None of these replace CPAP as the first try — they fill the gap for the people CPAP doesn't fit.

What most guides get wrong

"Only overweight older men get it." Wisconsin found 6% of women aged 30–70 had moderate-or-severe OSA; HypnoLaus, with sensitive scoring, found nearly a quarter of women over 40 at the same severity threshold Peppard et al. 2013 Heinzer et al. 2015. Women often present differently — insomnia, fatigue, depression, morning headache rather than the loud-snoring stereotype — and get told they have anxiety or are perimenopausal. Lean people with small jaws or high-arched palates also get it; the "skinny snorer who wakes tired" is a real and routinely missed phenotype.

"Snoring is just a nuisance." Loud habitual snoring is the most sensitive single symptom of OSA. Primary snoring without apnea exists but is much less common than supposed. If a partner has called your snoring loud enough to leave the room, the threshold for testing is met.

"The SAVE trial proved CPAP doesn't work." SAVE tested CPAP for preventing repeat cardiovascular events in people who already had cardiac disease and weren't sleepy, at an average of 3.3 hours of mask time a night McEvoy et al. 2016. That is a narrow question. It says nothing about symptom relief, mortality in people without established cardiac disease, or stroke risk at full adherence — where the evidence stays positive.

"Catching up on the weekend fixes it." Sleep architecture loss isn't a debt you settle on Sunday. Untreated, the airway collapses again every night you sleep, so does the cardiovascular strain, and the cumulative damage compounds regardless of total hours.

"If I just lose weight, it'll go away." Sometimes — for the BMI-driven phenotype, yes. For people with narrow craniofacial anatomy, the airway stays narrow at any weight. The right test is treat now, retest after weight loss; don't postpone testing because you plan to lose weight.

Where this goes sideways

You quit CPAP in the first month. This is the single biggest failure mode of the whole pipeline; somewhere between a third and half of users don't clear the four-hour-a-night threshold long-term. The first week is the test. Almost every comfort problem is solvable, but the time to escalate is week one, not month six: the mask leaks, refit it; the pressure feels like drowning, ask for the ramp feature or a pressure-relief setting; your mouth dries out, add a heated humidifier or a chin strap; you feel claustrophobic, try a nasal pillow mask instead of a full-face. The single best predictor of whether you'll be on CPAP in a year is whether the first week was tolerable.

The mandibular device causes your bite to drift. Year-over-year, the device's nightly pull can shift teeth and change how your bite comes together. This is real and progressive; the workaround is an annual dental check with photographs and impressions, and morning bite-realignment exercises. Some people get jaw soreness in the first weeks that fades; some develop TMJ pain that doesn't, and that's the signal to stop.

Your anatomy doesn't fit the treatment. If the soft palate collapses in a "circular curtain" rather than back-to-front, MAD won't help much and Inspire won't either — the only fixes are CPAP, weight loss, or anatomic surgery. A sleep specialist with access to drug-induced sleep endoscopy can map the actual collapse pattern.

Central apneas emerge on the mask. A small fraction of OSA patients develop centrally-driven apneas once the obstructive ones are treated — your airway is open, but your brain briefly stops sending the signal to breathe. The fix is a different machine (bilevel-PAP or adaptive servo-ventilation), not abandoning treatment.

What it costs and where you actually get tested

The home sleep apnea test runs about $150 to $500 in the US, almost always insurance-covered with a primary-care referral; the in-lab study is more like $1,000 to $3,000, also typically covered. A CPAP machine is several hundred dollars up front (insured patients usually pay a co-pay only), and supplies — mask cushions, filters, tubing — add up to roughly $200 to $400 a year. A custom mandibular advancement device from a qualified dental sleep specialist is $1,500 to $3,000, with messier insurance coverage that depends on whether your medical plan or your dental plan takes it. The implanted nerve stimulator is in the $30,000 range, but Medicare and most large commercial insurers cover it for patients who meet the criteria. The friction is rarely the money — it's the chain of referrals from primary care to sleep medicine to the equipment supplier, which often takes a couple of months.

What changes when you treat it

The first night the mask actually fits is often the night your partner stops needing earplugs. By week one, the 3 PM crash thins out. By week three or four, the morning headache stops being a daily fixture and you notice you read whole pages of a book without losing the thread. People around you stop asking if you're tired — they ask what you've been doing differently. The afternoon productivity window you used to call your peak hours extends into evening.

By month three, the doctor's office reads your blood pressure and it's lower than it's been in years — the average drop is small in trial data (about 2.5 mmHg) but the people with resistant hypertension are where you see the real swing Bratton et al. 2015. Mood evens out; the irritability and short fuse that the people who live with you had learned to work around quietly retreats AASM 2019.

The cardiovascular and mortality dividend takes years to decades to compound. In the long observational cohort that followed treated and untreated severe OSA over a decade, the people on CPAP tracked the healthy controls; the untreated severe arm tracked a steadily worse curve Marin et al. 2005. You don't feel that part. You just keep waking up rested, and one day the cardiologist looks at a ten-year reading and tells you it's gone the wrong way for someone your age — meaning the right way.

Related, but not this entry

If your home sleep test comes back negative but the symptoms are still there — loud snoring, daytime fatigue, the morning headache — look into upper airway resistance syndrome (UARS). It produces the same felt experience without crossing the AHI threshold, and the in-lab study with an esophageal pressure catheter is what catches it.

If you mouth-breathe at night without confirmed apnea, the night-time nasal-breathing literature — mouth tape, nasal dilator strips, allergic-rhinitis treatment — is the right next step.

Central sleep apnea is a separate condition: the airway is open, but the breathing signal from the brainstem stalls. Different mechanism, different treatment, often associated with heart failure or chronic opioid use.

Pediatric sleep apnea starts with an ENT referral, not a sleep lab — enlarged tonsils and adenoids are the dominant cause and surgery is first-line in a way it isn't for adults.

Related in the handbook
Worsens / aggravates
  • — Sleep apnea is a major engine of AFib; treating it improves the odds of holding a normal rhythm.
  • — Apnea is a leading cause of hidden high blood pressure. A week of home readings catches what a daytime clinic visit misses.
  • — The blood pressure that won't come down on pills is often untreated apnea — the airway collapses spike it all night.
  • — A pre-bed drink worsens apnea events; cutting it is an easy first move.
  • — Chronic nasal congestion stiffens the upper airway and worsens apnea; treating the rhinitis often improves nights and CPAP comfort.
  • — New erection trouble? Sleep apnea is a common, fixable cause - it starves testosterone and the blood vessels at once.
  • — Apnea fragments the REM sleep that drives morning erections, so they often vanish; treating the apnea can bring them back.
  • — Apnea's micro-arousals trigger jaw clenching, so untreated apnea can surface as morning jaw ache and temple headaches.
Helps
  • — For overweight patients, a weekly tirzepatide injection can bring sleep apnea close to normal — the first drug approved for it.
  • — Treating apnea matters more if you carry e4 — it's part of the carrier's prevention playbook.
  • — GLP-1 drugs can roughly halve apnea severity — weight loss is part of why.
  • — For some people apnea is mostly a back-sleeping problem — getting off your back eases it.
Diagnosis
  • — The new apnea-notification on the latest watches is the rare wearable feature worth acting on — it's a screen, not a diagnosis.
Related
  • — Tooth grinding at night is a common flag for an airway that's collapsing — bruxism can point to apnea.
  • — When apnea rides along with COPD, the two stack overnight. Treating the apnea takes real load off struggling lungs.
  • — Apnea and night-time reflux often coexist and feed each other; treating one without the other leaves your sleep broken.
  • — Apnea drives night-time urine production, so treating it can stop the nightly bathroom trips men blame on the prostate.
  • — Chronic mouth breathing and snoring can be the visible edge of sleep apnea; diagnose it before trying mouth-shut fixes.
  • — Sleep apnea is far more common in women with PCOS, and treating it helps the metabolic picture.
  • — A blocked nose makes CPAP unbearable; clearing it is often what makes the mask tolerable.
  • — Sleep apnea quietly drags testosterone down; treating the apnea can raise it without ever needing TRT.
  • — If your sleep study is normal but you still wake wrecked, UARS is the diagnosis that gets missed below the apnea cutoff.
  • — Mouth breathing and snoring overlap with apnea; nasal-breathing habits help a little, but real apnea needs a proper workup.
  • — A nasal dilator won't treat apnea. If breathing stops at night, that needs a proper sleep workup, not a strip.
  • — Apnea risk jumps for women around menopause, so broken perimenopausal sleep deserves a second look beyond hormones.
  • — The overnight oxygen drops that grade your apnea come from a pulse oximeter, which reads high on darker skin — worth knowing when you read the report.
  • — When bedroom fixes like temperature don't end the unrefreshing nights, sleep apnea is the usual culprit underneath.
  • — Plenty of hours in bed but still wrecked? That's not sleep debt — it's an airway collapsing all night.
·
192