Polysomnography (Sleep Study): Evidence-Based Ways to Improve Your Results

What Polysomnography Actually Measures

A polysomnography (PSG) is not a single number. It is an overnight recording of at least 16 channels of physiological data collected while you sleep in a monitored setting. The test captures electroencephalography (EEG) for brain waves, electrooculography (EOG) for eye movements, electromyography (EMG) for muscle tone, nasal and oral airflow, thoracic and abdominal respiratory effort, pulse oximetry, electrocardiography, and leg movement sensors 1.

The American Academy of Sleep Medicine (AASM) considers PSG the gold-standard diagnostic tool for obstructive sleep apnea (OSA), central sleep apnea, narcolepsy, periodic limb movement disorder, and REM sleep behavior disorder 2. Your sleep technologist scores the raw data into discrete metrics. The apnea-hypopnea index (AHI) gets the most clinical attention because it drives treatment decisions, insurance approvals, and follow-up intervals. But sleep efficiency, oxygen desaturation index (ODI), arousal index, and percentage of time in each sleep stage all carry clinical weight.

For patients in hormone-therapy or GLP-1 programs, PSG results matter doubly. Untreated OSA suppresses nocturnal testosterone pulses and blunts growth hormone secretion, both of which undermine the goals of TRT and peptide protocols 3.

Normal Polysomnography Ranges and What They Mean

A normal AHI is fewer than 5 respiratory events per hour. The AASM classifies mild OSA as 5 to 14, moderate as 15 to 29, and severe as 30 or above 4. Sleep efficiency (total sleep time divided by total time in bed) should reach 85% or higher. Sleep latency, the time it takes to fall asleep, normally falls between 10 and 20 minutes. An oxygen nadir below 90% is flagged as clinically significant desaturation.

Short sleep latency (under 5 minutes) may suggest sleep deprivation or hypersomnolence disorders. A very high sleep latency (over 30 minutes) can indicate insomnia. REM sleep should account for 20% to 25% of total sleep time, and N3 (deep sleep) typically makes up 15% to 20% in adults under 60 5.

The Endocrine Society's 2018 clinical practice guideline on testosterone therapy recommends screening men starting TRT for OSA, noting that "clinicians should inform patients of the potential for worsening of sleep apnea during testosterone therapy" 6. This makes baseline and follow-up PSG results especially relevant for HealthRX patients on hormone optimization.

CPAP Therapy: The First-Line Intervention

Continuous positive airway pressure (CPAP) remains the primary treatment for moderate-to-severe OSA. It works. A 2019 Cochrane systematic review of 36 trials (N=2,926) found that CPAP reduced AHI to below 5 events per hour in the majority of adherent users and improved daytime sleepiness scores by a mean 2.5 points on the Epworth Sleepiness Scale compared with sham CPAP 7.

Adherence is the bottleneck. Medicare defines adequate CPAP use as 4 hours per night on at least 70% of nights. Roughly 50% of patients fail to meet this threshold at 1 year. Auto-titrating PAP (APAP) devices that adjust pressure in real time have shown better adherence than fixed-pressure CPAP in a meta-analysis of 9 RCTs (N=1,215), with patients using APAP an average of 0.4 hours more per night 8.

Practical steps to boost adherence: use a heated humidifier to reduce nasal dryness, try multiple mask styles to find a comfortable fit, and use the ramp feature to start at a lower pressure that gradually increases. A 2020 trial showed that telemedicine-based CPAP coaching improved 90-day adherence by 15 percentage points compared with standard care 9.

Weight Loss: The Most Powerful Modifiable Factor

Body weight is the single strongest modifiable predictor of AHI. The Wisconsin Sleep Cohort Study (N=690) demonstrated that a 10% weight gain predicted a 32% increase in AHI, while a 10% weight loss predicted a 26% decrease 10. The Sleep AHEAD trial, a substudy of the Look AHEAD diabetes trial, followed 264 overweight adults with type 2 diabetes and OSA for 4 years. Participants who lost an average of 10.8 kg through intensive lifestyle intervention saw their AHI drop by 9.7 events per hour compared with 3.3 events per hour in the diabetes-education control arm 11.

GLP-1 receptor agonists have added a pharmacological path to this outcome. The STEP 1 trial (N=1,961) showed semaglutide 2.4 mg produced 14.9% mean body weight loss at 68 weeks versus 2.4% with placebo 12. In the SURMOUNT-OSA trial, tirzepatide specifically targeted adults with moderate-to-severe OSA and obesity. After 52 weeks, tirzepatide reduced AHI by 27.4 events per hour in CPAP non-users and by 23.0 events per hour in CPAP users, compared with reductions of 4.8 and 1.3 events per hour with placebo, respectively 13. Roughly half of tirzepatide-treated participants no longer met the AHI threshold for moderate-to-severe OSA.

Dr. Atul Malhotra, chief of pulmonary, critical care, and sleep medicine at UC San Diego and a SURMOUNT-OSA investigator, stated: "These are the largest AHI reductions we have seen with any pharmacotherapy for sleep apnea" 13.

Positional Therapy and Sleep Hygiene

Supine-dependent OSA (where AHI is at least twice as high on the back compared with lateral positions) affects roughly 50% to 60% of patients with mild-to-moderate OSA 14. For these patients, positional therapy alone can cut AHI substantially.

A 2015 meta-analysis of 8 RCTs (N=323) found that the newer generation of vibrotactile positional devices reduced supine sleep time by a mean 84% and lowered overall AHI by approximately 54% 15. These small devices attach to the chest or neck and vibrate gently when the wearer rolls onto their back. They outperformed the old "tennis ball technique" in both efficacy and adherence.

Sleep hygiene measures do not directly lower AHI, but they can improve sleep efficiency and architecture. Evidence-supported practices include:

• Maintaining a consistent bed and wake time (within a 30-minute window, even on weekends)
• Keeping the bedroom at 65 to 68 degrees Fahrenheit
• Limiting alcohol within 3 hours of bedtime, as alcohol relaxes upper airway muscles and increases apnea frequency by 25% to 60% in dose-dependent fashion 16
• Avoiding caffeine after 2 PM (caffeine's half-life ranges from 3 to 7 hours depending on CYP1A2 genotype)
• Elevating the head of bed by 30 degrees, which reduced AHI by a mean 31.8% in a small crossover trial of 13 OSA patients 17

Oral Appliances and Surgical Options

Mandibular advancement devices (MADs) are the primary alternative for patients with mild-to-moderate OSA who cannot tolerate CPAP. These custom-fitted dental appliances hold the lower jaw forward by 6 to 10 mm, widening the retroglossal airway. A 2021 Cochrane review (27 trials, N=1,982) reported that MADs reduced AHI by an average of 11.3 events per hour versus no treatment, though CPAP still achieved greater AHI reductions overall 18.

Surgical approaches target identifiable anatomical obstruction. Uvulopalatopharyngoplasty (UPPP), the most common OSA surgery, reduces AHI by about 33% on average but carries variable success rates depending on patient selection 19. Hypoglossal nerve stimulation (Inspire device) has stronger data for select patients. The STAR trial (N=126) followed implanted patients for 5 years and found sustained AHI reduction from a median of 29.3 to 6.2 events per hour 20. FDA approval requires a BMI <35, moderate-to-severe OSA (AHI 15 to 65), and documented CPAP failure.

Maxillomandibular advancement (MMA) surgery advances both the upper and lower jaw by 10 to 12 mm. A systematic review of 45 studies reported a pooled success rate of 86% (defined as AHI <20 and at least a 50% reduction) and a cure rate (AHI <5) of 43% 21.

Medications That Affect Sleep Study Results

No FDA-approved drug directly treats OSA as of May 2026, but several medications influence PSG findings.

Acetazolamide (250 mg twice daily) reduces central apneas at altitude and has shown modest AHI reductions in OSA (approximately 38% in a small crossover trial, N=12) by stimulating ventilatory drive 22. Atomoxetine plus oxybutynin, a combination targeting noradrenergic and antimuscarinic pathways, reduced AHI by 63% in a proof-of-concept trial (N=20) 23. This combination remains investigational.

Medications that can worsen PSG results include benzodiazepines (which reduce upper airway muscle tone), opioids (which cause central apneas), and exogenous testosterone at supraphysiologic doses. The Endocrine Society guideline notes that TRT can be safely administered alongside CPAP treatment in men with OSA, but untreated severe OSA (AHI >30) is listed as a relative contraindication to initiating testosterone therapy 6.

The OSA-Testosterone Connection

OSA and low testosterone share a bidirectional relationship. Fragmented sleep reduces the amplitude of nocturnal LH pulses, lowering morning testosterone levels. A cross-sectional analysis of 3,015 men in the European Male Ageing Study found that those with severe OSA had total testosterone levels 2.2 nmol/L lower than men without OSA after adjusting for age, BMI, and comorbidities 24.

CPAP treatment partially reverses this deficit. A meta-analysis of 9 studies (N=232) reported a modest but statistically significant increase in total testosterone of 1.1 nmol/L after 3 months of adequate CPAP use 25. For men on TRT protocols at HealthRX, optimizing CPAP adherence may complement exogenous testosterone by improving sleep quality, reducing cortisol, and increasing growth hormone secretion during N3 sleep.

The AACE 2017 clinical practice guidelines for comprehensive management of obesity state: "All patients with obesity and suspected OSA should undergo polysomnography, as treatment of OSA may improve metabolic and hormonal profiles independent of weight change" 26.

When to Retest and What to Track

The AASM recommends repeat polysomnography when a significant clinical change occurs: a weight change of 10% or more, initiation or discontinuation of CPAP, surgical intervention, or persistent symptoms despite treatment 4. Routine annual retesting in stable, well-treated patients is not supported by current evidence.

For tracking progress between formal sleep studies, home sleep apnea tests (HSATs) offer a lower-cost alternative. HSATs measure airflow, respiratory effort, and oxygen saturation but do not capture EEG data, so they cannot assess sleep architecture or diagnose non-respiratory sleep disorders 27. The AASM position paper notes that HSATs are appropriate for confirming moderate-to-severe OSA in patients with a high pretest probability but should not replace in-lab PSG when the clinical picture is ambiguous.

Modern CPAP machines with built-in data tracking (reporting residual AHI, leak rate, and hours of use) provide useful interim monitoring. A residual AHI consistently above 5 on CPAP data suggests the need for pressure adjustment or mask refit before ordering a repeat in-lab study.