Obstructive Sleep Apnea (OSA) in Special Populations: Diagnosis and Treatment

What Is OSA and How Is It Defined?

OSA occurs when the upper airway collapses repeatedly during sleep, causing oxygen desaturation and sleep fragmentation. The American Academy of Sleep Medicine (AASM) defines OSA as an AHI of 5 or more obstructive events per hour accompanied by daytime symptoms (sleepiness, fatigue, witnessed apneas), or an AHI of 15 or more events per hour regardless of symptoms. Severity is graded as mild (AHI 5-14), moderate (AHI 15-29), or severe (AHI 30 or more).

Why the AHI Threshold Matters Clinically

The AHI cutoff is not arbitrary. A 2009 analysis of the Sleep Heart Health Study (N=6,441) found that participants with an AHI of 15 or more had a statistically significant increase in cardiovascular mortality compared to those with an AHI below 5, even after adjustment for BMI and hypertension [1]. Choosing the right threshold determines who qualifies for insurance-covered CPAP, who enters clinical trials, and, as of 2025, who qualifies for tirzepatide under the FDA label.

Diagnosis: Home Sleep Apnea Testing vs. Polysomnography

In-lab polysomnography (PSG) is the diagnostic gold standard, measuring airflow, respiratory effort, oxygen saturation, EEG, EMG, and leg movements simultaneously. Home sleep apnea testing (HSAT) measures a narrower channel set and typically underestimates AHI by 10-20%, which is clinically relevant in populations where the real AHI sits close to the diagnostic threshold. The AASM recommends HSAT only for adults with a high pretest probability of moderate-to-severe OSA and no significant comorbid sleep disorders [2].

OSA in Adults With Obesity

Obesity is the single most modifiable risk factor for OSA. Excess adipose tissue in the parapharyngeal region reduces airway lumen, and visceral fat restricts thoracic compliance, worsening hypoxic events. Every 10-unit increase in BMI raises OSA risk approximately 6-fold in population-based cohort data [3].

Weight Loss as Disease-Modifying Therapy

A meta-analysis of 7 randomized controlled trials (N=362) published in Sleep Medicine Reviews found that a 10% reduction in body weight reduces AHI by a mean of 26% [4]. Diet alone rarely produces durable weight loss sufficient to normalize AHI, so pharmacologic and surgical adjuncts have become standard co-management tools.

Tirzepatide (Zepbound) for OSA: The SURMOUNT-OSA Evidence

Tirzepatide, a dual GIP/GLP-1 receptor agonist, received FDA approval in January 2025 specifically for moderate-to-severe OSA in adults with obesity (BMI 30 or above), making it the first drug approved for this indication. The approval rested on two phase 3 trials, SURMOUNT-OSA-1 and SURMOUNT-OSA-2 (combined N=469).

In SURMOUNT-OSA-1, participants not using CPAP who received tirzepatide 10 mg or 15 mg weekly for 52 weeks achieved a mean AHI reduction of 27.4 events per hour versus 4.8 events per hour with placebo (P<0.0001) [5]. Body weight fell by a mean of 20.1% in the tirzepatide arm. In SURMOUNT-OSA-2, which enrolled CPAP users who wished to discontinue CPAP, the tirzepatide arm showed an AHI reduction of 29.3 events per hour versus 5.5 with placebo.

The FDA label requires concurrent treatment for OSA (CPAP or behavioral interventions) and does not position tirzepatide as a CPAP replacement in isolation.

Bariatric Surgery as an Option

For adults with BMI 35 or above and moderate-to-severe OSA who fail behavioral and pharmacologic weight loss, Roux-en-Y gastric bypass produces AHI reductions exceeding 60% at one year in registry data [6]. Sleeve gastrectomy shows similar, though slightly smaller, AHI reductions. OSA should be assessed before and approximately 6-12 months after bariatric surgery to determine whether CPAP can be discontinued.

OSA in Pregnant Women

Prevalence and Risk Amplification During Pregnancy

Pregnancy creates a unique OSA phenotype. Rhinitis of pregnancy, diaphragm elevation, progesterone-driven upper airway muscle changes, and gestational weight gain all compress airway reserve. Estimated OSA prevalence rises from roughly 4% in the first trimester to 8-26% by the third trimester, based on PSG-confirmed data from a 2020 systematic review (N=3,792 pregnancies) [7].

Gestational diabetes, preeclampsia, and preterm birth all occur at higher rates in pregnant women with untreated OSA. A 2014 study in Sleep (N=175) found preeclampsia in 40.5% of pregnant women with OSA compared with 20.5% in matched controls without OSA [8].

Diagnosis Considerations in Pregnancy

HSAT is generally not recommended in pregnancy because positional hypoxia, fetal movement artifacts, and the higher pretest probability of non-respiratory sleep disorders reduce its accuracy. Full PSG is preferred when OSA is suspected. Screening tools such as the modified STOP-BANG questionnaire (which replaces the "neck circumference" item with a "snoring plus observed apnea" item) have been adapted for obstetric populations, though sensitivity remains around 66% [9].

Treatment During Pregnancy

CPAP is safe throughout pregnancy and is the first-line treatment. Tirzepatide, semaglutide, and other GLP-1 class agents are contraindicated in pregnancy, as are most weight-loss drugs. Auto-titrating CPAP (APAP) may require resetting pressure ranges across trimesters as anatomy changes. Positional therapy (avoiding the supine position) reduces AHI modestly but does not reach therapeutic targets as a standalone measure in moderate-to-severe disease. Postpartum, OSA often improves but does not always resolve, and a follow-up sleep study or HSAT is recommended 6-12 weeks after delivery.

OSA in Children and Adolescents

How Pediatric OSA Differs From Adult OSA

Pediatric OSA is not a scaled-down version of adult OSA. The dominant anatomic cause in children aged 2-8 is adenotonsillar hypertrophy rather than obesity, although the obesity-related pediatric OSA phenotype is rising. The AASM defines abnormal AHI in children as 1 or more obstructive events per hour, a much lower threshold than the adult cutoff of 5 events per hour [10].

Symptoms differ too. Children with OSA often present with hyperactivity, attention deficits, and behavioral problems rather than daytime sleepiness. Misattribution of these symptoms to ADHD is common and delays diagnosis by a mean of 2.5 years in retrospective pediatric cohort data.

Adenotonsillectomy as First-Line Therapy

Adenotonsillectomy (AT) is the recommended first-line treatment for most children with OSA and adenotonsillar enlargement. The CHAT trial (Childhood Adenotonsillectomy Trial, N=464) showed that early AT improved behavior, quality of life, and PSG measures of OSA more than watchful waiting at 7-month follow-up, though PSG normalization occurred in only 79% of children without obesity [11]. Children with obesity, Down syndrome, or craniofacial anomalies have lower post-AT cure rates and typically require CPAP.

CPAP in Pediatric Populations

CPAP in children requires careful mask fitting and caregiver training. Adherence averages 4.5 hours per night in pediatric studies, which is below the recommended 6 hours. Pressure titration in children is performed in a PSG laboratory rather than via APAP, as algorithms are not validated for pediatric airway mechanics. Oral appliances are not routinely recommended for children whose craniofacial development is still active.

Pediatric OSA in Adolescents With Obesity

The overlap of obesity and adolescent OSA represents a growing subgroup. A 2022 analysis from the TEEN-LABS consortium found that adolescents who underwent metabolic-bariatric surgery had a mean AHI reduction from 9.4 to 2.1 events per hour at one year [12]. Pharmacologic options for adolescent OSA-obesity overlap are limited. The FDA has approved GLP-1 receptor agonist semaglutide (Wegovy) for adolescents aged 12 and above for weight management, but no pediatric OSA-specific drug indication exists for any agent as of mid-2025.

OSA in Older Adults

Why OSA Looks Different After Age 65

OSA prevalence exceeds 50% in adults aged 65 and older, based on data from the MrOS Sleep Study (N=2,911 men, mean age 76) [13]. Despite this high prevalence, older adults are systematically underdiagnosed because daytime sleepiness is attributed to aging, medications, or comorbidities rather than sleep-disordered breathing.

The phenotype shifts with age. Upper-airway muscle hypotonia, increased thoracic rigidity, and higher rates of central apnea events make the respiratory disturbance index more complex in this population. Central and mixed apneas account for a greater share of total events, which has implications for both diagnosis and treatment.

CPAP Adherence and Tolerability in Older Adults

CPAP remains effective in older adults. A randomized trial of CPAP versus sham CPAP in older adults with OSA (mean age 71, N=278) showed a significant reduction in daytime sleepiness and blood pressure at 6 months [14]. Adherence in older patients is similar to or slightly better than in middle-aged cohorts, possibly because older adults have more structured sleep schedules and stronger motivation related to fall prevention and cognitive concerns.

Mask interface selection matters more in older adults. Edentulous patients may struggle with nasal pillow masks, and denture removal at night can change oral anatomy enough to affect mask seal. Full-face masks often work better but increase claustrophobia risk.

Cognitive Impairment and OSA

OSA is an independent risk factor for cognitive decline. The landmark HYPNOS trial and a 2020 JAMA Internal Medicine analysis (N=4,288 older adults) found that AHI above 15 was associated with a 35% higher risk of incident mild cognitive impairment over 5 years compared to those with AHI below 5 [15]. CPAP treatment in cognitively impaired older adults requires caregiver involvement for mask management and pressure adjustments.

Pharmacologic weight management with tirzepatide in older adults with OSA and obesity is feasible, though SURMOUNT-OSA trials enrolled relatively few participants over age 70, and dose escalation should be slower to reduce gastrointestinal adverse events.

OSA in Patients With Cardiovascular Disease

Cardiovascular disease and OSA are bidirectionally linked. OSA causes intermittent hypoxia, sympathetic surges, and intrathoracic pressure swings that promote hypertension, atrial fibrillation, and coronary artery disease. Resistant hypertension, defined as blood pressure above 130/80 mmHg on three or more antihypertensive agents, has OSA as a contributing factor in an estimated 30-83% of cases [16].

CPAP and Cardiovascular Outcomes: What the Evidence Shows

The SAVE trial (N=2,717 adults with moderate-to-severe OSA and established cardiovascular disease) found that CPAP did not reduce the primary composite cardiovascular outcome versus usual care over a mean follow-up of 3.7 years [17]. This was a significant finding, though mean nightly CPAP use was only 3.3 hours, below the therapeutic threshold. Post-hoc analyses suggest that participants using CPAP for 4 or more hours per night had lower stroke rates, but these subgroup findings require confirmation.

The American Heart Association and American College of Cardiology include OSA screening in their 2023 hypertension guideline, recommending evaluation in all patients with resistant hypertension [18].

Atrial Fibrillation and OSA

AF recurrence after cardioversion or pulmonary vein isolation is substantially higher in patients with untreated OSA. A 2013 meta-analysis (N=3,995 patients) found that CPAP use was associated with a 42% lower AF recurrence rate at 12 months compared to untreated OSA [19]. Electrophysiologists increasingly request OSA evaluation before planning AF ablation procedures.

OSA in Patients With Type 2 Diabetes

OSA and type 2 diabetes share overlapping pathophysiology through insulin resistance, visceral adiposity, and sympathetic activation. An estimated 58-86% of adults with type 2 diabetes have OSA based on PSG-confirmed data [20].

GLP-1 receptor agonists prescribed for glycemic control (liraglutide, semaglutide, dulaglutide) produce weight loss that may secondarily reduce AHI, though only tirzepatide carries a specific FDA OSA indication. Glycemic improvement from CPAP alone, without weight loss, is modest, averaging about a 0.4% reduction in HbA1c in randomized trial data, which is clinically meaningful but insufficient as diabetes monotherapy.

The American Diabetes Association 2024 Standards of Care state: "Clinicians should consider screening for OSA in adults with type 2 diabetes, particularly those with obesity, as treatment of OSA may improve glycemic control and quality of life" [21].

First-Line and Adjunct Treatments Across All Populations

Continuous Positive Airway Pressure (CPAP)

CPAP delivers a pneumatic splint that prevents airway collapse. Therapeutic pressure ranges from 4 to 20 cm H2O and is determined by auto-titration or in-lab PSG titration. Adherence, defined as 4 or more hours per night on 70% or more of nights, is achieved by roughly 50-60% of patients in real-world studies. Heated humidification and mask interface options (nasal pillow, nasal mask, full-face mask) are the primary levers for improving adherence.

Mandibular Advancement Devices

Oral appliances (OAs) that advance the mandible 50-75% of maximum protrusion reduce AHI by a mean of 47% in meta-analytic data across mild-to-moderate OSA [22]. OAs are less effective than CPAP at AHI normalization but may achieve equivalent cardiovascular outcomes due to higher nightly use hours. They are not recommended for severe OSA (AHI 30 or above) as a standalone therapy.

Hypoglossal Nerve Stimulation

The Inspire system (upper airway stimulation, UAS) uses a surgically implanted electrode to stimulate the hypoglossal nerve during inspiration, protracting the tongue and opening the airway. The STAR trial (N=126) demonstrated a 68% reduction in AHI at 12 months in adults with moderate-to-severe OSA who had failed CPAP [23]. The FDA has approved UAS for adults with AHI between 15 and 65, BMI below 32, and without a complete concentric palatal collapse pattern on drug-induced sleep endoscopy.

Positional Therapy

Positional OSA, defined as AHI more than twice as high in the supine position as in the lateral position, affects approximately 56% of OSA patients. Positional devices (vibrating chest bands, specialized pillows) reduce supine sleep time and lower AHI by 40-60% in positional-OSA subgroups. They are most appropriate for mild-to-moderate positional OSA in patients who decline or cannot tolerate CPAP.