Lua Labs Report — Circadian amplitude as a biomarker of perimenopausal transition

Date: 2026-06-19 Researcher: Lua Labs Classification: Chronobiology Line: L4 — Hormonal chronobiology Subtopic: 4.6 — Circadian amplitude as a biomarker of perimenopausal transition

External Sources

1. Ren X, Wang W, Li W, et al. (2025). "Circadian rest-activity rhythms and multimorbidity and mortality risks among menopausal women: a trajectory analysis of a UK Biobank cohort." BMC Public Health. DOI: 10.1186/s12889-025-22536-3. PMID: 40197377.
2. Hou SY, Chiu CJ, Shea JL, et al. (2024). "Role of age, menopausal status, and symptoms in midlife women: Examination of sleep patterns and rest-activity circadian rhythms." Sleep Medicine. DOI: 10.1016/j.sleep.2023.11.015. PMID: 38039943.
3. Hou SY, Chiu CJ, Shea JL, et al. (2024). "Sleep and rest-activity rhythms for women at different menopausal statuses: the role of mental health." Menopause. DOI: 10.1097/GME.0000000000002297. PMID: 38113433.
4. Cohn AY, Grant LK, Nathan MD, et al. (2023). "Effects of Sleep Fragmentation and Estradiol Decline on Cortisol in a Human Experimental Model of Menopause." Journal of Clinical Endocrinology & Metabolism. DOI: 10.1210/clinem/dgad285. PMID: 37207451.
5. Coborn J, de Wit A, Crawford S, et al. (2022). "Disruption of Sleep Continuity During the Perimenopause: Associations with Female Reproductive Hormone Profiles." Journal of Clinical Endocrinology & Metabolism. DOI: 10.1210/clinem/dgac447. PMID: 35878624.
6. Tai Y, Obayashi K, Yamagami Y, Saeki K. (2023). "Association between circadian skin temperature rhythms and actigraphic sleep measures in real-life settings." Journal of Clinical Sleep Medicine. DOI: 10.5664/jcsm.10590. PMID: 37394793.
7. Alzueta E, Gombert-Labedens M, Javitz H, et al. (2024). "Menstrual Cycle Variations in Wearable-Detected Finger Temperature and Heart Rate, But Not in Sleep Metrics, in Young and Midlife Individuals." Journal of Biological Rhythms. DOI: 10.1177/07487304241265018. PMID: 39108015.
8. Lin G, Li JY, Christofferson K, Truong KN, Mariakakis A. (2024). "Understanding wrist skin temperature changes to hormone variations across the menstrual cycle." npj Women's Health. DOI: 10.1038/s44294-024-00037-9.
9. Dittmar M, Stark T, Wedell S. (2024). "Circadian Rhythm of Distal Skin Temperature in Healthy Older and Young Women and Its Relationship with Sleep-Wake Rhythm and Environmental Factors under Natural Living Conditions." Geriatrics. DOI: 10.3390/geriatrics9040102. PMID: 39195132.
10. Perez-Medina-Carballo R, Kosmadopoulos A, Moderie C, et al. (2025). "Dampened circadian amplitude of EEG power in women after menopause." Journal of Sleep Research. DOI: 10.1111/jsr.14219. PMID: 38665057.

Baseline Knowledge (what I know before searching)

Circadian amplitude is not a single variable. It is the robust day-night contrast in activity, temperature, melatonin, cortisol, HRV, heart rate, sleep, hunger, insulin sensitivity, affective state, and clock-gene expression. A strong clock does not only have "timing"; it has contrast. Day is bright and active; night is dark, cool, vagal, and restorative. When that contrast flattens, a person may sleep 7 hours and still live through a biologically weak night.

In women, that amplitude is layered onto a second oscillation: the ovarian cycle or reproductive transition. Estradiol modulates SCN sensitivity, thermoregulation, serotonin, the HPA axis, and sleep architecture. Progesterone and allopregnanolone raise temperature, alter ventilation, affect GABA-A, and modify stress tolerance. Therefore the same light, meal, or shift-work misalignment does not cost the same in the follicular phase, luteal phase, or perimenopause.

Perimenopause does not begin as an "irregular menstruation" switch. It begins as variability: interleaved ovulatory and anovulatory cycles, erratic FSH peaks, high or low estradiol depending on the week, intermittent progesterone, vasomotor load, fragmented sleep, and symptoms that do not obey the calendar. My reading of L4 is that this variability first appears as loss of amplitude and coherence, before the menstrual calendar becomes diagnostic.

Findings From Recent Papers

The strongest longitudinal anchor is Ren et al. 2025 in UK Biobank: 10,138 menopausal women initially free of chronic disease, accelerometry, and mean follow-up of 8.13 years. Low relative amplitude (RA) of rest-activity rhythms was associated with higher risk of transitioning from health to first chronic condition (HR 1.18; 95% CI 1.07-1.31) and from first condition to multimorbidity (HR 1.34; 95% CI 1.12-1.61). This does not prove hormonal causality, but it validates RA as a signal of systemic risk in menopausal women.

Hou et al. 2024 found something counterintuitive in 87 women aged 45-60 years with 7-day actigraphy: perimenopausal and postmenopausal women had higher relative amplitude and stability than premenopausal women in that sample, and psychological symptoms predicted poorer sleep. The companion paper in Menopause showed that mental health modulates RA in a menopausal-status-dependent way. This prevents a simplistic conclusion of "perimenopause = low RA." The useful signal is individual and contextual: amplitude relative to one's own baseline, symptoms, mood, environment, and exposures.

Cohn et al. 2023 separates fragmented sleep from low estradiol: in an experimental menopause model, sleep fragmentation increased bedtime cortisol by 27% and reduced the cortisol awakening response by 57%; estradiol suppression reduced nocturnal cortisol by 22%. Coborn et al. 2022 showed that more awakenings were associated with estradiol in the postmenopausal range and higher FSH, independent of VMS and depressive symptoms. The integrated reading is clear: hormonal transition and sleep fragmentation perturb the HPA axis, and that perturbation can be seen as a signature of broken amplitude.

The temperature papers complete the mechanism. Tai et al. 2023 showed in 2,187 adults that lower intraday variability and higher interdaily stability/relative amplitude of distal temperature were associated with better sleep efficiency, less WASO, and longer total sleep. Alzueta/Gombert-Labedens 2024 showed that wearable-detected finger temperature preserves detectable menstrual oscillation in young and midlife women; in midlife, mesor was higher, but menstrual amplitude and phase of temperature were preserved in those who had an oscillation. Lin et al. 2024 warns that wrist skin temperature is not linearly equivalent to urinary E3G/LH hormones; the wearable needs models that tolerate variance. Dittmar et al. 2024 showed in healthy older women an 18-19% lower distal temperature amplitude and an acrophase 66-73 min earlier versus young women. Perez-Medina-Carballo et al. 2025 adds that postmenopause may have dampened circadian amplitude of sleep EEG power.

Full Molecular/Endocrine Mechanism

Circadian amplitude is formed by convergence of central and peripheral signals:

Morning/night light -> melanopsin/ipRGC -> SCN -> pineal/HPA/autonomic output -> melatonin + cortisol + sympathetic tone
                                            -> peripheral clocks

Food/caffeine/shifts -> liver/pancreas/intestine/adipocyte -> peripheral CLOCK/BMAL1/PER/CRY
                                                          -> insulin, glucose, SHBG, inflammation

Ovary/cycle -> E2/P4/ALLO/FSH/LH -> temperature, GABA-A, PR/ER, vasomotor load, sleep

In a woman with strong amplitude, those layers align: intense light early, low nocturnal melanopic load, last meal far from sleep, stable sleep midpoint, nocturnal temperature with an interpretable pattern, HRV/RHR within her baseline, and symptoms oscillating with the expected phase. In a woman with collapsed amplitude, each signal pushes in a different direction: nighttime screen exposure delays melatonin, late dinner requires insulin when MTNR1B slows beta cells, night shift forces operational cortisol, intermittent P4 raises temperature without stabilizing GABA-A, and the early morning becomes a mixture of heat, awakenings, tachycardia, anxiety, and fatigue.

The perimenopausal endocrine chain looks like this:

E2-P4 decline/variability -> unstable thermoregulation + KNDy/vasomotor load
                           -> awakenings / WASO / fragmentation
                           -> high bedtime cortisol + low CAR
                           -> low HRV / high RHR / low AM energy
                           -> more caffeine, late meals, nighttime light
                           -> lower melatonin and lower circadian amplitude
                           -> amplified perimenopausal symptoms

The important scientific point: low amplitude can be a cause, consequence, or amplifier. It is not an isolated clinical diagnosis. It is a system state that turns a normal hormonal transition into a more symptomatic experience.

Lua Labs Hypotheses

Hypothesis 44: Collapse of circadian amplitude precedes visible perimenopause

Statement: In women aged 42-52 years, sustained loss of individual circadian amplitude predicts increased perimenopausal symptoms before clear menstrual irregularity appears.

Proposed mechanism: Early perimenopausal transition produces intermittent ovulation/P4 and E2 variability. That instability increases awakenings, erratic nocturnal temperature, and vasomotor sensitivity. If high nocturnal light exposure, late meals, or night/rotating work coincide, the darkness signal weakens and autonomic recovery falls. Result: a smaller day-night difference in energy, temperature, HRV/RHR, and sleep. The person feels that "something changed" before having enough cycle irregularity to name it.

Confidence level: Medium. There is strong evidence for sleep/cortisol/temperature/RAR, but a specifically perimenopausal longitudinal cohort with wearables and hormonal diaries is lacking.

How to validate:

• With a formal study: n=150-300 observational cohort, continuous wearable + diaries + monthly FSH/E2/P4/AMH, mixed models per person.

Limitations: Low amplitude can also come from depression, work, motherhood/caregiving, grief, pain, alcohol, apnea, thyroid disease, medications, or infections. It should be modeled as a risk/state biomarker, not as a diagnosis.

Hypothesis 45: Thermo-HPA fracture as a perimenopausal nocturnal signature

Statement: In symptomatic perimenopause, the combination of unstable nocturnal temperature + awakenings + low HRV/high RHR functions as a proxy of thermo-HPA fracture and predicts next-day fatigue/brain fog better than total sleep duration.

Proposed mechanism: Variable E2/P4 narrows the thermoneutral zone and facilitates vasomotor load. Micro-awakenings raise bedtime cortisol and reduce CAR, as Cohn 2023 suggests. The autonomic response is expressed as low HRV and high RHR. Even if 7 hours are recorded, the night loses recovery architecture and parasympathetic contrast.

Confidence level: Medium-high for the sleep-cortisol and sleep-temperature axis; medium for its use as a proxy without salivary cortisol.

How to validate:

• With a formal study: n=60 substudy with wearable, bedtime/CAR salivary cortisol 3 days/month, VMS and phase diaries.

Limitations: Wearables do not directly measure core temperature or cortisol. Intraindividual baseline should be used, and absolute values should not be compared across devices.

Hypothesis 46: Zeitgeber coherence cushions the transition

Statement: Women with higher zeitgeber coherence — morning light, low night light, meals far from sleep, regular sleep, and post-shift recovery — will have lower symptomatic cost from the same hormonal variability.

Proposed mechanism: Perimenopause adds hormonal noise. If the temporal environment is ordered, the SCN and peripheral clocks maintain enough contrast to absorb part of the noise. If the temporal environment is broken, the same E2/P4 variability is amplified in sleep, the HPA axis, thermoregulation, and metabolism.

Confidence level: Medium. Zeitgeber biology is solid, but the specific interaction with perimenopausal symptoms needs validation.

How to validate:

• With a formal study: pragmatic 8-week behavioral intervention, non-therapeutic, focused on temporal regularity and measurement of subjective/physiological outcomes.

Limitations: There is a risk of turning social determinants into individual responsibility. The model must distinguish "cannot regulate because of work/caregiving" from "does not know what to regulate."

Individual Variability

The same low amplitude does not mean the same thing in everyone. In the luteal phase, it may express vulnerability due to P4/ALLO + late dinner. In perimenopause, it may be early transition fragility. In probable PCOS, it may amplify signals of nocturnal insulin, low SHBG, and free androgens. In postmenopause, it may reflect circadian aging, comorbidity, or fragmented sleep.

The main modulators are chronotype, shift work, motherhood/caregiving, sun exposure, latitude/season, alcohol, caffeine, physical activity, depression/anxiety, sleep apnea, thyroid disease, medications, wearable type, and regularity of use. Interpretation must be intraindividual.