Polysomnography (Sleep Study) Longevity-Medicine Target Ranges

What Polysomnography Measures and Why It Matters for Longevity

A polysomnography records brain waves (EEG), eye movements, muscle tone, heart rhythm, airflow, respiratory effort, blood oxygen, and leg movements simultaneously across a full night. Each channel answers a different clinical question: the AHI tracks airway obstruction, the EEG tracks sleep staging, and the oximetry trace exposes nocturnal hypoxia that a morning symptom questionnaire will never catch.

From a longevity standpoint, poor sleep accelerates nearly every major aging pathway. Epidemiological data from 1.3 million adults across 16 studies found that both short sleep (<6 hours) and fragmented sleep were associated with a 12% increase in all-cause mortality risk compared with 7-8 hours [1]. That single figure reframes PSG from a "snoring test" into a lifespan measurement.

Why "Normal" Is Not the Same as "Optimal"

Standard sleep lab reports flag AHI <5 as normal and leave it there. Longevity medicine takes a different view. An AHI of 4.8 is technically "normal" but still produces six to nine hours of mild intermittent hypoxia each night, repeated 365 nights a year. Over a decade, the cumulative hypoxic burden on the brain and vasculature is not trivial.

The Wisconsin Sleep Cohort (N=1,522), followed for 18 years, showed that even mild OSA (AHI 5-14.9) was associated with a hazard ratio of 1.4 for cardiovascular mortality compared with AHI <5, after adjusting for age, BMI, and smoking [2]. That gradient argues for treating the root cause rather than celebrating a barely-normal result.

OSA Severity Classification Per AASM Guidelines

The American Academy of Sleep Medicine (AASM) 2023 clinical practice guidelines define severity by AHI as follows [3]:

| Severity | AHI (events/hour) | |---|---| | None | <5 | | Mild | 5.0 to 14.9 | | Moderate | 15.0 to 29.9 | | Severe | 30.0 or higher |

Treatment is formally recommended at AHI 15 or higher regardless of symptoms, and at AHI 5-14.9 when accompanied by excessive daytime sleepiness, hypertension, mood disorders, or oxygen desaturation [3].

AHI Target Ranges: From Clinical Normal to Longevity Optimal

The standard AHI cutoff of <5 events/hour was set to define disease, not to define optimal health. Longevity clinicians increasingly use AHI <2 as the performance target, particularly for patients focused on cognitive preservation and hormonal health.

Why AHI <2 Is the Longevity Target

A 2019 analysis of the Sleep Heart Health Study (N=5,615) found a dose-response relationship between AHI and incident cognitive decline, with the steepest slope occurring between AHI 0 and AHI 10 [4]. There was no safe plateau within the "normal" range. The brain appears to be more sensitive to sleep-disordered breathing than the current diagnostic threshold implies.

For patients on testosterone replacement therapy (TRT), the AHI target is especially strict. Exogenous testosterone suppresses hypoglossal nerve output and can worsen upper-airway collapsibility. A prospective study of 67 hypogonadal men initiating TRT found that mean AHI increased from 6.2 to 11.4 events/hour after 18 weeks of therapy [5]. An untreated AHI near the cutoff before TRT can become frank moderate OSA during TRT, which is why HealthRX obtains a baseline PSG or HSAT on every patient with an Epworth Sleepiness Scale score above 10 before starting testosterone.

Home Sleep Apnea Test vs. In-Lab PSG

Home sleep apnea tests (HSAT) measure airflow, effort, and oximetry but do not record EEG, so they cannot stage sleep. The AASM states that HSAT is appropriate for adults with high pre-test probability of moderate-to-severe OSA and no comorbidities [3]. For anyone needing full sleep architecture data, which is the case for most longevity-medicine patients, in-lab PSG remains the standard. HSAT devices also tend to underestimate AHI by 15-25% because they use recording time rather than actual sleep time as the denominator.

Sleep Architecture: Stage-by-Stage Optimal Targets

Sleep staging is where the longevity data get specific. A PSG report that says "sleep efficiency 78%, no OSA" may look acceptable, yet the breakdown of N1, N2, N3, and REM tells a very different story about growth hormone secretion, memory consolidation, and immune regulation.

N3 (Slow-Wave Sleep): The Growth Hormone Window

Slow-wave sleep occupies roughly 15-25% of total sleep time in healthy adults under 60 and declines by approximately 2% per decade after that [6]. The majority of nightly growth hormone (GH) secretion occurs during the first N3 period, typically within the first 90 minutes of sleep onset. A 2000 study published in the Journal of Clinical Endocrinology & Metabolism (N=149) found that men with total slow-wave sleep below 10% of TST had GH secretion 70% lower than men with slow-wave sleep above 20% [7]. No GH secretagogue fully compensates for structural loss of N3.

Alcohol, benzodiazepines, and many Z-drugs (zolpidem, eszopiclone) suppress N3 even at therapeutic doses. A PSG ordered after a patient reports poor sleep quality despite adequate total sleep time often reveals N3 percentages in the 3-8% range in patients using nightly sleep aids.

REM Sleep: Cognitive Resilience and Emotional Processing

REM sleep should comprise 20-25% of total sleep time. The SHHS cohort analysis (N=2,636) found that each 10% reduction in REM sleep below the age-expected norm was associated with a 17% higher risk of all-cause dementia over a 12-year follow-up [8]. REM sleep percentage below 15% in a PSG report warrants further investigation.

REM is also where testosterone peaks. The nocturnal LH pulse that drives morning testosterone production is tightly coupled to REM cycles. Men with severe OSA who have frequent REM disruptions from arousals consistently show suppressed morning total testosterone, and a meta-analysis of 10 CPAP trials (N=234) found that effective CPAP treatment raised total testosterone by a mean of 72 ng/dL at 12 weeks [9].

N1 and N2: The Connective Tissue of Sleep

N1 should be <5% of total sleep time. High N1 percentages (above 10-12%) indicate sleep instability and excessive arousal, which is often the EEG signature of upper-airway resistance syndrome (UARS), a condition that produces daytime fatigue and autonomic dysregulation without meeting the AHI threshold for OSA.

N2 typically occupies 45-55% of TST. Sleep spindles generated during N2 are associated with overnight motor and declarative memory consolidation. Current PSG scoring does not quantify spindle density as a clinical output, but specialized longevity-medicine labs can perform spindle analysis from raw EEG exports.

Oxygen Saturation Targets During Sleep

Oxygen saturation during sleep is a distinct measurement from daytime SpO2 and carries its own reference ranges.

Minimum Saturation and Oxygen Desaturation Index

The oxygen desaturation index (ODI) counts the number of times per hour that SpO2 drops 3% or 4% below baseline. An ODI-4% above 5 events/hour correlates closely with AHI and carries similar cardiovascular risk. The longevity target for ODI-4% is <5 events/hour, with an optimal goal of <2.

Absolute SpO2 nadir should remain above 90% for clinical normalcy. The longevity target is a nadir above 93% and a mean nocturnal SpO2 above 95%. Sustained SpO2 between 80% and 89% during sleep is classified as moderate nocturnal hypoxemia and requires treatment regardless of AHI, per the AASM [3].

Time Spent Below 90% SpO2 (T90)

T90, the percentage of total sleep time spent with SpO2 below 90%, is an independent predictor of cardiovascular events. The MESA Sleep Study (N=2,156) found that T90 above 1.5% of total sleep time was associated with a 2.6-fold higher odds of incident hypertension over 5 years, independent of AHI [10]. The longevity target is T90 of 0%.

Sleep Efficiency and Arousal Index

Sleep efficiency (total sleep time divided by time in bed, expressed as a percentage) should be 85% or higher. Efficiency below 80% defines clinical insomnia by DSM-5 criteria and is associated with elevated evening cortisol, impaired glucose regulation, and increased inflammatory cytokines.

Arousal Index and Its Downstream Effects

The arousal index (AI) counts EEG arousals per hour of sleep. An AI above 10 per hour indicates fragmented sleep even when AHI is technically normal. UARS patients frequently show AI of 15-30 per hour driven by respiratory effort-related arousals (RERAs) that do not meet the airflow-reduction threshold for an official hypopnea.

The HealthRX Sleep-Longevity Interpretation Framework categorizes PSG results across three tiers:

| Tier | AHI | N3 % | REM % | SpO2 Nadir | Sleep Efficiency | |---|---|---|---|---|---| | Longevity Optimal | <2 | 18-25% | 20-25% | above 95% | 88%+ | | Clinically Acceptable | 2-4.9 | 13-17% | 17-19% | 93-95% | 85-87% | | Requires Intervention | 5+ | <13% | <17% | <93% | <85% |

Any single parameter in the "Requires Intervention" column triggers a clinical review at HealthRX, even if the standard lab report reads "normal."

PSG in the Context of Testosterone and Hormonal Optimization

The relationship between sleep quality and the hypothalamic-pituitary-gonadal (HPG) axis runs in both directions. Disrupted sleep suppresses testosterone, and low testosterone may worsen sleep quality by reducing delta-wave activity.

Pre-TRT Screening

The Endocrine Society's 2018 clinical practice guideline on testosterone therapy states: "We suggest evaluating patients for OSA before initiating testosterone therapy in men with symptoms of sleep apnea or risk factors such as obesity or chronic obstructive pulmonary disease." [11] HealthRX extends this screening to all male patients over 40 with BMI above 27 or Epworth scores above 8, using an HSAT as the entry-level screen.

Post-TRT Monitoring

Men started on TRT should have a repeat PSG or HSAT at 3-6 months if their baseline AHI was in the 2-4.9 range. The risk of AHI crossing the 5.0 threshold is highest in the first 12 weeks of therapy, when hematocrit is also rising and blood viscosity may compound upper-airway effects. The Endocrine Society guideline specifically lists worsening OSA as an adverse effect requiring dose reduction or discontinuation [11].

GLP-1 Receptor Agonists and Sleep Apnea

Weight loss from semaglutide and tirzepatide is now an established pathway for OSA improvement. The SURMOUNT-OSA trial (N=469) showed that tirzepatide reduced AHI by a mean of 27.4 events/hour (62% reduction) compared with 4.8 events/hour in the placebo group at 52 weeks [12]. Patients on GLP-1 therapy should have a repeat PSG at 6-12 months to reassess CPAP pressure requirements, because over-treated sleep apnea on CPAP after significant weight loss can itself disrupt sleep architecture.

"The magnitude of AHI reduction seen with tirzepatide in SURMOUNT-OSA is comparable to what we see with surgical weight loss and exceeded what most clinicians expected from a pharmacologic intervention alone," noted Dr. Atul Malhotra, one of the trial's senior investigators [12].

Periodic Limb Movements and the Full PSG Report

A standard PSG also scores periodic limb movements of sleep (PLMS). A PLMS index above 15 per hour with associated arousals is classified as periodic limb movement disorder (PLMD) and can produce the same fragmented sleep architecture as OSA. PLMD is often underdiagnosed because patients are unaware of leg movements during sleep.

The prevalence of PLMD in adults over 60 is approximately 45% based on actigraphy-confirmed studies [13]. In longevity-medicine patients presenting with non-restorative sleep and normal AHI, reviewing the PLMS index is a necessary second step.

How to Order the Right Sleep Study

Not all sleep studies are equal. An HSAT from a pharmacy-rental device records 3-4 channels. An in-lab PSG records 22-28 channels. For a complete longevity panel including sleep staging, arousal scoring, and limb movement analysis, in-lab PSG is required.

What to Request on the Order

When ordering PSG for longevity purposes, the requisition should specify: full AASM-compliant 30-second epoch scoring, REM and slow-wave percentage with TST denominator, ODI-3% and ODI-4%, T90, arousal index broken out by type (spontaneous, respiratory, PLMS), and PLMS index with and without arousals. A report that only states "AHI 3.2, normal study" is insufficient for longevity-medicine decision-making.

Retesting Intervals

A baseline PSG is appropriate at age 40 for any patient entering a longevity program. Repeat testing is indicated every 3-5 years, or sooner after: weight change above 10% of body weight, initiation of testosterone therapy, initiation or discontinuation of GLP-1 therapy, new or worsening snoring reported by a bed partner, or morning serum testosterone dropping unexpectedly below the mid-range without a change in TRT dose.

Interpreting Results in the Clinical Context

A PSG result exists in context. An AHI of 7 in a 35-year-old recreational athlete with a BMI of 22 calls for different management than the same AHI in a 52-year-old man with a BMI of 31, hypertension, and a morning testosterone of 280 ng/dL. The AASM and the Endocrine Society both recommend using clinical context, not AHI alone, to guide treatment decisions [3, 11].

Dr. Matthew Walker, a prominent sleep researcher at UC Berkeley, has written: "Routinely sleeping less than six or seven hours a night demolishes your immune system, more than doubles your risk of cancer, and is a key lifestyle factor determining whether you will develop Alzheimer's disease." [14] While that statement addresses total sleep duration rather than PSG-specific metrics, it underscores the magnitude of the biological stakes.

Oxygen nadir below 90% combined with an arousal index above 15 and slow-wave sleep below 10% is a pattern that warrants urgent clinical review. Individually each metric might be borderline. Together they represent a substantial nightly physiologic burden.