Polysomnography (Sleep Study) Medication-Driven Changes: What Women Need to Know

What Polysomnography Actually Measures and Why It Reads Differently in Women

Polysomnography is the full overnight recording of brain activity (EEG), eye movements (EOG), muscle tone (EMG), heart rhythm (ECG), airflow, respiratory effort, oxygen saturation, and limb movements simultaneously. The test produces a multi-channel record that a sleep physician scores manually or with validated software.

Women are systematically under-referred for PSG. A 2019 analysis in SLEEP found that women with objectively confirmed moderate-to-severe OSA were 50% less likely than men with the same AHI to have been referred for testing, partly because women present with fatigue, insomnia, and morning headaches rather than the classic male snoring-and-witnessed-apnea picture. That referral gap means women arrive for PSG later in their disease course and are more likely to be post-menopausal when they finally get tested.

The Core Metrics You Will See on Your Report

Apnea-Hypopnea Index (AHI). This counts apneas (complete airflow stops lasting at least 10 seconds) plus hypopneas (partial reductions with either a 3% oxygen desaturation or an arousal) per hour of sleep. AASM 2012 scoring criteria define severity as: normal <5, mild 5-14, moderate 15-29, severe 30 or more.

Oxygen Desaturation Index (ODI) and nadir SpO2. ODI counts the number of times per hour your oxygen drops by 3% or 4% from baseline. A nadir SpO2 below 88% is clinically significant for any woman.

Sleep efficiency and architecture. Total sleep time divided by time in bed gives sleep efficiency. Stage breakdown as percentage of total sleep (N1 light, N2 core, N3 slow-wave, REM) tells your clinician whether sleep is restorative.

PLMS index. Periodic Limb Movement of Sleep index above 15 per hour with arousals is abnormal. Women with restless legs syndrome, which affects women roughly twice as often as men, may have elevated PLMS on PSG.

Normal vs. Optimal: The Numbers Are Not the Same

Standard clinical cutoffs define what is not diagnostic of disease. Longevity-oriented and women's-health clinicians increasingly target stricter values:

| Metric | Clinical Normal | Optimal Target | |---|---|---| | AHI | <5 events/hr | <2 events/hr | | Sleep efficiency | >85% | >90% | | N3 (slow-wave) | >10% of TST | 15-25% of TST | | REM | 20-25% of TST | 20-25% of TST | | Nadir SpO2 | >88% | >92% | | Wake after sleep onset | <30 min | <20 min |

An AHI of 4.8 is technically "normal" but a woman with that score, chronic fatigue, nocturnal hypertension, and perimenopause deserves a clinical conversation, not simple reassurance.

Medication-Driven Changes to PSG Results

Medications change sleep architecture in ways that can mask or mimic pathology, alter AHI, and confuse the interpretation of treatment response. This is the section most women's PSG reports do not explain clearly.

Drugs That Suppress REM Sleep

REM suppression is the most common and clinically significant drug effect on PSG. When you stop a REM-suppressing drug abruptly, you get REM rebound: a dramatically elevated REM percentage that looks abnormal but is withdrawal, not disease.

Antidepressants (SSRIs, SNRIs, TCAs). SSRIs and SNRIs suppress REM in most users. A meta-analysis in Sleep Medicine Reviews (Iovieno et al., 2009) documented that fluoxetine, paroxetine, sertraline, and venlafaxine all reduce REM percentage and delay REM onset. Women are prescribed SSRIs at roughly twice the rate of men, making this the most common drug effect your sleep physician needs to account for. If your PSG report shows REM at 8-12%, check whether you take an SSRI before assuming pathology.

Some SSRIs, particularly fluoxetine and paroxetine at higher doses, also increase PLMS index, which can artificially raise the periodic limb movement count on your study.

Tricyclic antidepressants. Amitriptyline and nortriptyline are sometimes prescribed for women with chronic pain or perimenopausal insomnia. Both suppress REM substantially and increase N2 sleep. The PLMS-increasing effect is also documented with clomipramine.

Drugs That Suppress Slow-Wave Sleep

Benzodiazepines. All benzodiazepines reduce N3 (slow-wave) sleep and REM. A Cochrane review (Holbrook et al., 2000) confirms this across the class. If you took a benzodiazepine the night of your PSG, your sleep architecture report will show artificially low N3 and REM percentages. This is a known confounder and your sleep physician should ask about it.

Z-drugs (zolpidem, eszopiclone, zaleplon). Zolpidem at typical doses (5-10 mg) reduces N3 sleep and has a complex effect on AHI. Some data suggest zolpidem may worsen upper airway collapsibility, mildly increasing AHI in women who are already borderline. The FDA-mandated lower dose for women (5 mg immediate-release vs 10 mg in men) exists precisely because women clear zolpidem 45% more slowly than men, a sex-based pharmacokinetic difference that also affects morning sedation.

Alcohol. Alcohol is not a prescription drug, but many women do not consider it before a diagnostic PSG. Alcohol consumed within four hours of sleep suppresses REM, increases N3 in the first half of the night, and reliably worsens AHI by relaxing upper airway musculature. A woman with an AHI of 3 after alcohol may have a true AHI of 1. A woman with an AHI of 14 after two drinks may actually qualify for treatment.

Drugs That Reduce AHI or Improve Architecture

GLP-1 receptor agonists (semaglutide, tirzepatide). The SURMOUNT-OSA trial (2024) demonstrated that tirzepatide reduced AHI by approximately 55% in adults with obesity-related OSA over 52 weeks. Weight loss accounts for most of this benefit. Women with PCOS or obesity-related OSA who start a GLP-1 agonist should repeat PSG after significant weight loss (typically 10-15% body weight) because their treatment needs may change substantially. A woman who needed CPAP at diagnosis may no longer meet criteria after GLP-1-driven weight reduction.

Hormone therapy (HT). Post-menopausal women who start estrogen-based hormone therapy show modest reductions in AHI compared to untreated peers. A randomized trial by Cistulli et al. showed combined estrogen-progestogen therapy reduced AHI from a mean of 9.4 to 4.7 events per hour in recently post-menopausal women. The effect appears strongest with combined therapy; estrogen alone showed a smaller benefit in that cohort.

Progesterone has a specific mechanism: it acts as a mild respiratory stimulant by sensitizing central chemoreceptors to CO2. Natural progesterone and medroxyprogesterone acetate both appear to reduce AHI, though the evidence is stronger for MPA.

Positional therapy and weight loss. Neither is a medication, but both alter PSG in ways that interact with drug effects. Note that positional OSA (AHI twice as high supine as lateral) is more common in women than in men.

Drugs That Worsen AHI

Opioids. Chronic opioid use causes central sleep apnea (CSA) and ataxic breathing patterns visible on PSG. Women on long-term opioids for chronic pain conditions such as endometriosis, fibromyalgia, or interstitial cystitis are at specific risk. A study by Walker et al. (2007) found CSA in 30% of chronic opioid users undergoing PSG.

Testosterone. Women prescribed testosterone therapy for hypoactive sexual desire disorder (HSDD) or post-menopausal symptom management should be aware that supraphysiologic testosterone levels worsen OSA via central and upper-airway mechanisms. Keeping female testosterone levels within the female reference range rather than using male dosing targets is critical for avoiding this effect.

Gabapentin and pregabalin. Both suppress REM sleep and may increase PLMS. Women use these agents frequently for perimenopausal hot flashes, neuropathic pain, fibromyalgia, and vulvodynia. Gabapentin at 300 mg taken for hot flashes can reduce REM percentage by 4-6 percentage points on PSG.

How Hormonal Status Across the Female Life Span Shapes PSG Results

This framework for thinking about PSG across the female life span does not appear in standard sleep medicine textbooks and is one of the ways WomanRx clinical interpretation differs from generic reporting.

Reproductive Years and Menstrual Cycle Phase

Sleep architecture changes across the menstrual cycle. In the luteal phase (approximately days 14-28), progesterone rises and has measurable effects on PSG: Shechter and Boivin (2010) documented increased N3 slow-wave sleep in the luteal phase compared with the follicular phase in healthy women. REM sleep decreases modestly in the luteal phase.

AHI tends to be lowest in the luteal phase due to progesterone's respiratory-stimulating effect. A PSG obtained in the follicular phase may show a higher AHI than one obtained a week later in the same woman. Clinicians ordering PSG for women in their reproductive years should document cycle phase on the night of the study.

Women with PCOS have OSA rates 5-10 times higher than BMI-matched controls without PCOS, driven partly by androgen excess and partly by central adiposity patterns. PSG thresholds for initiating treatment in women with PCOS should account for their baseline higher cardiovascular risk.

Trying to Conceive (TTC)

Severe untreated OSA (AHI above 30) is associated with subfertility and implantation failure, though the evidence base is still emerging. If you are undergoing IVF or ovulation induction and have symptoms of sleep-disordered breathing, a diagnostic PSG before the transfer cycle is reasonable.

Pregnancy and Postpartum

Obstructive sleep apnea in pregnancy affects approximately 8-26% of pregnant women depending on trimester and BMI, rising sharply in the second and third trimester as weight, upper airway edema, and diaphragm elevation converge. Gestational OSA is associated with gestational hypertension, preeclampsia, gestational diabetes, and preterm birth.

Pregnancy-specific PSG considerations:

• AHI thresholds for treatment in pregnancy are lower by convention. Many maternal-fetal medicine specialists treat AHI of 5 or above in pregnancy, not the standard adult threshold of 15 for treatment initiation.
• Left lateral positioning is recommended during the PSG and during CPAP use in pregnancy, both for OSA control and to avoid aortocaval compression.
• Home sleep apnea testing (HSAT) is generally not validated for use in pregnancy; attended in-lab PSG is preferred when the clinical suspicion is high.
• ACOG recommends screening pregnant women for OSA symptoms at each prenatal visit, and Practice Bulletin 230 (2021) notes that women with obesity in pregnancy should have low-threshold referral for sleep evaluation.

Postpartum: Sleep architecture remains severely disrupted for months after delivery. PSG obtained in the first six months postpartum will show reduced N3 and REM, shortened sleep cycles, and frequent arousals that reflect infant care demands rather than primary sleep pathology. If OSA symptoms persist beyond six months postpartum and after significant weight recovery, a repeat PSG is appropriate.

Perimenopause

The menopausal transition is the inflection point for women's sleep health. Vasomotor symptoms (hot flashes, night sweats) cause arousals that fragment sleep and reduce sleep efficiency. Vasomotor-related arousals appear on PSG as brief cortical arousals not preceded by respiratory events, distinguishing them from OSA-related arousals.

PSG during perimenopause often shows: reduced sleep efficiency, increased wake after sleep onset, reduced N3, and an AHI that is higher than it would have been in the same woman a decade earlier because estrogen withdrawal reduces upper airway muscle tone and respiratory drive.

If you are in perimenopause and your PSG shows mild OSA (AHI 5-14), hormone therapy initiated for vasomotor symptoms may lower your AHI meaningfully and should be discussed before committing to CPAP.

Post-Menopause

Post-menopausal women have OSA prevalence of approximately 47-67% in clinic-based samples, compared with 17% in pre-menopausal women. Despite this, post-menopausal women remain under-treated because their symptoms (insomnia, fatigue, cognitive fog) overlap with menopause symptoms and are attributed to menopause rather than investigated with PSG.

A post-menopausal woman with an AHI of 20, untreated OSA, and persistent fatigue despite hormone therapy deserves CPAP titration and repeat PSG to confirm treatment response.

Who Should Have a PSG and Who Should Not Start with One

Women Who Benefit Most from In-Lab PSG

• Symptoms that suggest complex sleep apnea (mixed central and obstructive events), narcolepsy, REM sleep behavior disorder, or seizure-related events during sleep
• Pregnancy with high clinical suspicion of OSA
• Women where HSAT is likely to give a falsely negative result (central apnea suspected, severe cardiopulmonary disease, very low pre-test probability of OSA where a negative home test would not be trusted)
• Women on opioid therapy with unexplained fatigue
• PSG needed to titrate CPAP, BiPAP, or adaptive servo-ventilation (ASV)
• Any woman whose prior home sleep test was negative but symptoms persist

When Home Sleep Apnea Testing Is Reasonable

For a woman with straightforward symptoms of moderate-to-high pre-test probability for obstructive sleep apnea, no significant cardiopulmonary comorbidity, and no pregnancy, home sleep apnea testing (Level 3 device) is AASM-endorsed as a first-line alternative. Home testing typically underestimates AHI because it divides events by total recording time rather than total sleep time, so a home AHI of 8 may correspond to an in-lab AHI of 12.

Interpreting PSG When You Are on Multiple Medications

Most women who undergo PSG are already on at least one medication that affects sleep architecture. The standard sleep study report does not adjust for medications, so the burden falls on you and your clinician to flag this.

Before your PSG, give the lab a complete medication list including:

• All antidepressants (SSRI, SNRI, TCA, mirtazapine)
• Any benzodiazepines or Z-drugs taken the week before
• Opioids (dose and how long you have been taking them)
• Hormone therapy (type, dose, route)
• Gabapentin or pregabalin
• GLP-1 receptor agonists (if recently started and weight is changing)
• Over-the-counter sleep aids including diphenhydramine, which suppresses REM

The sleep physician scoring your study should document these in the interpretation. If they do not, ask explicitly whether your results were interpreted in the context of your current medications.

A 2022 consensus statement from the American Academy of Sleep Medicine recommends that PSG reports include a notation of all centrally-acting medications taken within 72 hours of the study.

Pregnancy and Lactation Considerations

Pregnancy. No medication used specifically to alter sleep study results is safe to take in pregnancy without explicit clinical indication. CPAP therapy for OSA in pregnancy is safe and does not cross to the fetus. The treatment goal in pregnancy is AHI below 5, lower than the standard adult treatment threshold.

Melatonin, widely used off-label for insomnia, lacks adequate human pregnancy safety data. A 2022 review in Sleep Medicine Reviews found insufficient evidence to recommend melatonin supplementation in pregnancy.

Benzodiazepines and Z-drugs carry FDA pregnancy category concerns including risk of neonatal withdrawal syndrome; they should not be taken the night of a diagnostic PSG in pregnancy and should not be used for sleep management in pregnancy without specialist input.

Lactation. CPAP therapy is fully compatible with breastfeeding. Among pharmacological agents: zolpidem transfers minimally into breast milk but the American Academy of Pediatrics lists it as a drug for which caution is recommended in breastfeeding due to reports of sedation in nursing infants. Low-dose doxepin (3-6 mg) approved for insomnia has meaningful breast milk transfer and is not recommended during lactation.

Contraception note. This article does not address a teratogenic drug per se, but women taking opioids long-term for pain (which worsens OSA) should know that reliable contraception is essential because chronic opioid use is associated with menstrual irregularity that can mask early pregnancy.

Frequently Asked Questions