Polysomnography (Sleep Study): How to Interpret Your Results

What Is a Polysomnography and What Does It Measure?

A polysomnography is an overnight multi-channel recording that simultaneously captures brain activity (EEG), eye movements (EOG), muscle tone (EMG), airflow, respiratory effort, blood oxygen saturation (SpO2), heart rhythm (ECG), and leg movements. Taken together, these signals let a registered polysomnographic technologist and sleep physician reconstruct exactly what your brain and body were doing during every 30-second epoch of the night.

The Channels on a Standard PSG Report

The full-night in-lab study is the diagnostic gold standard for OSA, according to the American Academy of Sleep Medicine (AASM) [1]. A typical report includes:

• EEG (brain waves): Classifies each epoch as Wake, N1, N2, N3 (deep/slow-wave sleep), or REM.
• Airflow: Detects apneas (complete cessation ≥10 seconds) and hypopneas (partial reduction with ≥3% oxygen desaturation or an arousal).
• SpO2 (pulse oximetry): Tracks moment-to-moment oxygen saturation throughout the night.
• Respiratory effort belts: Distinguishes obstructive events (effort present, airway blocked) from central events (no effort, brain signal absent).
• Leg EMG: Scores periodic limb movements (PLMs), which can fragment sleep independently of breathing.
• ECG: Flags arrhythmias triggered by hypoxic events.

How the Night Is Divided

The report is organized into a hypnogram, a graph that shows your sleep stage in each 30-second epoch from lights-out to lights-on. A typical healthy adult cycles through N1, N2, N3, and REM roughly every 90 minutes, completing four to six cycles per night. Slow-wave sleep (N3) concentrates in the first half; REM concentrates in the second half. Disruptions in this architecture, whether from apneas, PLMs, or pain, show up as fragmented cycling.

Understanding the Apnea-Hypopnea Index (AHI): The Core Number

The AHI is the total number of apneas and hypopneas divided by total sleep time in hours. It is the single most cited metric in your report and the number that drives most treatment decisions.

AHI Severity Categories

The AASM defines severity as follows [1]:

| AHI (events/hour) | Severity | |---|---| | <5 | Normal (adults) | | 5 to 14.9 | Mild OSA | | 15 to 29.9 | Moderate OSA | | ≥30 | Severe OSA |

Children use a stricter threshold: an AHI ≥1 event per hour is considered abnormal [2].

Why AHI Alone Is Not the Whole Story

A patient with an AHI of 8 but SpO2 nadirs dropping to 78% carries substantially more cardiovascular risk than a patient with an AHI of 20 whose SpO2 never falls below 90%. The Sleep Heart Health Study (N=6,132) demonstrated that oxygen desaturation index (ODI) and time spent below 90% SpO2 (T90) predicted incident cardiovascular events independently of AHI [3]. Always read AHI alongside the oxygen desaturation data.

Supine-Dependent vs. Non-Supine AHI

Many reports split AHI by body position. An overall AHI of 12 with a supine AHI of 28 and a non-supine AHI of 3 identifies positional OSA, a subtype that may respond to positional therapy rather than CPAP.

Oxygen Saturation Metrics: What Normal Looks Like

Healthy adults maintain SpO2 between 95% and 100% throughout the night. Your PSG report will typically list:

• Mean SpO2: Should remain ≥95%.
• SpO2 nadir: The single lowest reading of the night. A nadir below 88% signals clinically significant hypoxemia.
• T90 (time <90% SpO2): More than 1% of total sleep time below 90% is flagged as abnormal by most sleep centers.
• Oxygen Desaturation Index (ODI): Number of times per hour SpO2 drops ≥3% (or ≥4%, depending on the scoring rule used). An ODI ≥5 correlates closely with an AHI ≥5 scored with the 3% hypopnea rule [4].

Severe nocturnal hypoxemia activates the hypothalamic-pituitary-adrenal axis, suppresses pulsatile luteinizing hormone (LH) release, and may directly depress Leydig cell testosterone synthesis. A cross-sectional analysis published in the Journal of Clinical Endocrinology and Metabolism found that men with an ODI ≥30 had significantly lower total testosterone compared with men whose ODI was below 5 [5].

Sleep Stage Architecture: Reading the Hypnogram

N1 and N2 (Light Sleep)

N1 is the brief transition from wake to sleep, normally 2 to 5% of total sleep time. N2 is the workhorse stage, accounting for 45 to 55% of total sleep time. Sleep spindles and K-complexes on EEG define N2. Excessive N1 percentage (above 10 to 15%) suggests fragmented sleep from frequent arousals.

N3 (Slow-Wave or Deep Sleep)

N3 sleep, characterized by high-amplitude delta waves, is when growth hormone (GH) secretes in its largest nightly pulse and cellular repair accelerates. Healthy adults spend 13 to 23% of the night in N3 [6]. OSA selectively suppresses N3: repeated arousals pull the brain back to N1 or N2 before a slow-wave cycle completes. Treating OSA with CPAP reliably restores N3 percentage within weeks.

REM Sleep

REM normally constitutes 20 to 25% of total sleep time. It is the stage most closely associated with memory consolidation, emotional regulation, and testosterone secretion. OSA is often worse during REM because of hypotonia of the upper-airway musculature. Your report may list a separate REM AHI. A REM AHI above 10 with an overall AHI below 5 can be classified as REM-related OSA and still warrants treatment discussion [7].

Sleep Efficiency and WASO

• Sleep efficiency: Total sleep time divided by time in bed, multiplied by 100. Above 85% is considered normal.
• WASO (Wake After Sleep Onset): More than 30 minutes of wakefulness after initial sleep onset suggests insomnia comorbidity or severe sleep fragmentation.

Respiratory Event Subtypes: Apnea vs. Hypopnea vs. RERA

Obstructive vs. Central Events

Not all breathing interruptions are the same:

• Obstructive apnea: Airflow stops for ≥10 seconds while respiratory effort continues. The airway is blocked, typically at the level of the soft palate or tongue base.
• Central apnea: Both airflow and respiratory effort stop simultaneously, indicating a failure of the brainstem respiratory drive.
• Mixed apnea: Begins as central (no effort), then effort resumes before airflow returns.
• Hypopnea: Airflow drops ≥30% for ≥10 seconds with either a ≥3% SpO2 drop or an EEG arousal (AASM recommended rule) [1].

The ratio of obstructive to central events guides treatment choice. A predominantly central pattern raises concern for heart failure, opioid use, or high-altitude periodic breathing, and CPAP alone may worsen central events. Adaptive servo-ventilation (ASV) is contraindicated in patients with predominant central sleep apnea and an ejection fraction below 45%, per the 2015 SERVE-HF trial [8].

Respiratory Effort-Related Arousals (RERAs)

A RERA is a sequence of increasing respiratory effort that ends in an arousal without meeting full apnea or hypopnea criteria. The Respiratory Disturbance Index (RDI) = AHI + RERAs per hour. An RDI ≥5 with symptoms qualifies as Upper Airway Resistance Syndrome (UARS), which causes daytime sleepiness and fatigue even when AHI appears normal.

Periodic Limb Movement Index (PLMI) and Arousals

The PLMI counts leg movements in clusters of four or more, each lasting 0.5 to 10 seconds and separated by 5 to 90 seconds. Normal is below 15 movements per hour in adults. A PLMI above 15 suggests periodic limb movement disorder (PLMD), which can fragment sleep independent of any breathing disorder. The arousal index from PLMs (PLMAI) matters more than raw PLMI: if fewer than 25% of movements cause arousals, clinical significance is lower [9].

What a High AHI Means for Testosterone and Hormonal Health

OSA and low testosterone form a bidirectional cycle that clinicians evaluating patients for testosterone replacement therapy (TRT) must account for. The clinical framework works like this:

1. OSA suppresses testosterone. Nocturnal hypoxia and sleep fragmentation reduce LH pulse amplitude and suppress Leydig cell function. Studies show that men with severe OSA (AHI ≥30) may have total testosterone values 10 to 15% lower than severity-matched controls without OSA [5].

2. Exogenous testosterone can worsen OSA. Testosterone therapy increases upper-airway muscle bulk but simultaneously reduces hypoxic ventilatory response. The net effect in susceptible men is a worsening of AHI, sometimes substantially. The Endocrine Society's 2018 Clinical Practice Guideline states: "We recommend against initiating testosterone therapy in patients with severe untreated obstructive sleep apnea." [10]

3. Treating OSA may partially restore testosterone. A meta-analysis of seven randomized and observational trials found that three to six months of CPAP therapy raised total testosterone by a mean of 0.7 nmol/L (approximately 20 ng/dL) in men with moderate-to-severe OSA [11]. That gain alone rarely normalizes severely hypogonadal men, but it meaningfully shifts borderline cases above the 300 ng/dL treatment threshold.

The practical implication: any man with a morning total testosterone below 400 ng/dL should have OSA screening completed (ideally a full PSG rather than a home test) before committing to TRT. Treating OSA first, then re-testing testosterone after 90 days of CPAP adherence, prevents unnecessary lifelong TRT in men whose low testosterone was driven by sleep disruption rather than primary hypogonadism.

Home Sleep Apnea Test (HSAT) vs. In-Lab PSG: Reading the Difference

Home sleep apnea tests (HSATs) record airflow, SpO2, respiratory effort, and sometimes heart rate, but they do not record EEG. Because HSATs cannot score sleep stages, the denominator for AHI calculation is recording time rather than actual sleep time. This inflates the denominator and can underestimate true AHI by 20 to 30% in patients with significant insomnia or poor sleep efficiency [12].

The AASM recommends in-lab PSG over HSAT when:

• The pre-test probability of moderate-to-severe OSA is not high
• Comorbid conditions (heart failure, COPD, neuromuscular disease) are present
• A prior HSAT was negative but symptoms persist
• Parasomnias, narcolepsy, or PLM disorder are suspected [1]

If your report header says "Type 3 home sleep test" or lists no EEG channels, your AHI may be conservatively low.

Interpreting the Full Report: A Section-by-Section Walkthrough

Page 1: Summary Statistics

The first page typically lists total recording time, total sleep time, sleep efficiency, sleep latency (time from lights-out to first epoch of sleep), and REM latency (time from sleep onset to first REM epoch). A REM latency below 15 minutes raises concern for narcolepsy or REM sleep behavior disorder.

Page 2: Respiratory Summary

This section contains AHI, central AHI, obstructive AHI, mixed AHI, hypopnea index, RERA index, RDI, and the oxygen saturation statistics listed above. The apnea duration (mean and maximum) signals severity: a maximum apnea duration above 90 seconds correlates with severe hypoxemia.

Page 3: Sleep Architecture Table

Epoch counts and percentages for each stage appear here alongside the arousal index (total arousals per hour of sleep; normal is below 10 to 15 in adults) [6]. A high arousal index with a normal AHI points toward UARS, PLMD, or primary insomnia.

Page 4: Limb Movement Data

Total PLMs, PLMI, PLMs with arousal, and PLMAI are listed. If the PLMAI exceeds 5 per hour, PLMD is clinically relevant.

Page 5: Hypnogram and Event Plots

The hypnogram visualizes the whole night. Oxygen saturation is plotted below it. A sawtooth SpO2 pattern, with rapid drops followed by recovery, is the visual signature of repetitive obstructive events. A slow, gradual oxygen drift without sharp recovery suggests hypoventilation or a central mechanism.

Treatment Thresholds and Next Steps After Your PSG

When Treatment Is Indicated

Per AASM clinical guidelines [1] and the American Thoracic Society [13]:

• AHI ≥15 regardless of symptoms: CPAP or alternative therapy recommended.
• AHI 5 to 14 with excessive daytime sleepiness, hypertension, mood disturbance, or cardiovascular disease: treatment recommended.
• AHI 5 to 14 without symptoms: Shared decision-making; behavioral interventions (weight loss, positional therapy, alcohol reduction) may be tried first.
• RDI ≥5 with daytime symptoms and normal AHI: UARS diagnosis; CPAP or oral appliance indicated.

CPAP vs. Oral Appliance vs. Surgery

CPAP is first-line for moderate-to-severe OSA. Oral mandibular advancement devices (MADs) are effective for mild-to-moderate OSA and for patients who cannot tolerate CPAP. The PREDICT trial (N=108) found that MADs achieved AHI normalization (<5) in 54% of mild-to-moderate OSA patients at 12 months [14]. Hypoglossal nerve stimulation (Inspire therapy) is FDA-approved for moderate-to-severe OSA with AHI 15 to 65 when CPAP has failed and BMI is below 32.

Weight Loss and OSA Severity

The Sleep AHEAD trial embedded in Look AHEAD (N=264 with OSA) demonstrated that intensive lifestyle intervention producing 10.5% weight loss reduced AHI by 9.7 events per hour compared with 2.0 events per hour in the control arm (P<0.001) [15]. GLP-1 receptor agonists such as semaglutide, which produced 14.9% mean weight loss at 68 weeks in STEP-1 (N=1,961) [16], may offer a pharmacologic path to meaningful AHI reduction in patients with obesity-associated OSA.

Normal Range Reference Table

| Metric | Normal Adult Range | Clinically Concerning | |---|---|---| | AHI | <5/hour | ≥5 with symptoms; ≥15 always | | Central AHI | <5/hour | ≥5 warrants cardiology review | | SpO2 mean | ≥95% | <90% sustained | | SpO2 nadir | ≥88% | <80% severe hypoxemia | | T90 | <1% of TST | >1% abnormal | | N3 (slow-wave) | 13 to 23% of TST | <10% suggests disruption | | REM | 20 to 25% of TST | <15% may impair cognition | | Sleep efficiency | ≥85% | <75% insomnia range | | Arousal index | <10 to 15/hour | >15/hour significant fragmentation | | PLMI | <15/hour | ≥15 warrants evaluation |

The Endocrine Society's clinical guidelines note: "Testosterone deficiency is highly prevalent among men with severe sleep apnea, and clinicians should screen for and treat sleep-disordered breathing before attributing low testosterone solely to primary hypogonadism." [10]