Elsevier

Psychoneuroendocrinology

Volume 92, June 2018, Pages 81-86
Psychoneuroendocrinology

Sleep after intranasal progesterone vs. zolpidem and placebo in postmenopausal women – A randomized, double-blind cross over study

https://doi.org/10.1016/j.psyneuen.2018.04.001Get rights and content

Highlights

Abstract

Context

The loss of progesterone during menopause is linked to sleep complaints of the affected women. Previously we demonstrated sleep promoting effects of oral progesterone replacement in postmenopausal women. The oral administration of progesterone, however, is compromised by individual differences in bioavailability and metabolism of the steroid.

Objective

We compared the sleep-endocrine effects after intranasal progesterone (MPP22), zolpidem and placebo in healthy postmenopausal women.

Design

This was a randomized double-blind cross-over study.

Setting

German monocentric study

Participants

Participants were 12 healthy postmenopausal women.

Interventions

Subjects received in randomized order four treatments, 2 doses of intranasal progesterone (4.5 mg and 9 mg of MPP22), 10 mg of zolpidem and placebo.

Outcome measures

Main outcome were conventional and quantitative sleep-EEG variables. Secondary outcomes were the subjective sleep variables and the sleep related concentrations of cortisol, growth hormone (GH), melatonin and progesterone.

Results

Sleep promoting effects were found after the higher dosage of MPP22 and after zolpidem. Zolpidem prompted benzodiazepine-like effects on quantitative sleep EEG as expected, whereas no such changes were found after the two dosages of MP22. Nocturnal progesterone levels increased after 9.0 mg MPP22. No other changes of hormone secretion were found.

Conclusions

Our study shows sleep promoting effects after intranasal progesterone. The spectral signature of intranasal progesterone did not resemble the sleep-EEG alterations induced by GABA active compounds. Progesterone levels were elevated after 9.0 mg MPP22. No other endocrine effects were observed.

Introduction

The hypnotic effect of progesterone is documented by preclinical and human studies. In cats after administration of progesterone into the frontal cortex total sleep time increased (Heuser et al., 1967). Intraperitoneal progesterone administration induced at dose-dependent decrease of wakefulness in rats (Lancel et al., 1996). After oral administration of progesterone to young healthy male subjects non-REM sleep and spectral power of the higher frequency range in quantitative EEG analysis increased, whereas slower activity decreased (Friess et al., 1997). During early pregnancy progesterone plasma levels increase. It is thought that this change prompts tiredness in pregnancy (Lancel et al., 1996; Driver and Baker, 1998). The menopausal transition is discussed to contribute to impaired sleep (Moline et al., 2003; Dzaja et al., 2005; NIH State-of-the-Science Statements, 2005; Polo-Kantola, 2011) and is characterized by loss of both estrogens and progesterone. Estradiol is also known to promote sleep in animals. In the ovariectomized marmoset estradiol replacement prompted higher EEG delta power pointing to improved sleep intensity (Gervais et al., 2016). Sleep homeostasis was investigated during the estrous cycle of the rat. In the light period slow wave sleep activity exhibited high values (Schwierin et al., 1998). In postmenopausal women estrogen replacement therapy induced normalization of sleep-EEG pattern (Antonijevic et al., 2000). Progesterone restored normal sleep in postmenopausal women when sleep was disturbed (Caufriez et al., 2011). In a previous study (Schüssler et al., 2008) we investigated in a randomized double blind cross over design with two treatment intervals of 21 days the effects of a daily oral dose of 300 mg micronized progesterone in healthy post-menopausal women. After progesterone intermittent time spent awake decreased. During the first third of the night rapid eye movement (REM) sleep increased. No significant changes of EEG power spectra were observed. In contrast to oral medication intranasal administration circumvents the intestines and liver (Ducharme et al., 2010) and is an effective method to modulate brain activity (Dhuria et al., 2010; Fréchou et al., 2015). The intranasal route of administration delivered in mice progesterone to blood and brain (Ducharme et al., 2010). Progesterone and GABA A-active substances were reported to modulate nocturnal growth hormone (GH) and melatonin secretion (Caufriez et al., 2011 Hajak et al., 1996; Morera et al., 2009). After diazepam cortisol decreases (Pomara et al., 2005). The effect of intranasal progesterone administration on progesterone plasma levels in humans is unknown.

The objective of our study was to compare the effects of intranasal administration of two dosages of progesterone (MPP22), placebo and of oral administration of the hypnotic zolpidem on sleep EEG, subjective sleep. and on sleep related hormone secretion of cortisol, growth hormone (GH), melatonin and progesterone.

This was a monocentric, randomized, placebo and comparator controlled cross over-study with double dummy conducted at the Max-Planck-Institute of Psychiatry, Munich, Germany. The study is registered with EudraCT number. 2009-013051-30.

The subjects consisted of 12 healthy paid postmenopausal women (mean age ± S.D.: 59.6 ± 1.3 years, range 52–67, range of body mass index between 18.97-28.15 kg/m2). Eleven women were naturally and 1 woman was surgically postmenopausal since at least 4–23 years. Follicular stimulating hormone plasma concentration (FSH) was in a range between 43.2 and 111.7 mU/ml reflecting the postmenopausal status. The subjects were recruited by newspaper advertisement and did not report poor sleep quality as motivation to participate in this trial. All subjects did not take any medicine including hormone replacement therapy for at least 3 months and entered the study after passing rigid psychiatric, physical (including gynecologic and rhinal) and laboratory examinations. Reasons for exclusion from the study were: major chronic diseases (e.g. diabetes, heart failure, hepatitis), previous chronic neurological or psychiatric disorder in the own or family history, stressful life events, a transmeridian flight during the past 3 months, shift work, aberrancies in the blood chemistry or in the waking EEG or electrocardiogram. All subject underwent a polysomnographic examination in the sleep laboratory to exclude sleep disorders including sleep related respiratory disorders (e.g. sleep apnea) and sleep related movement disorders (e.g. restless legs syndrome). Other exclusion factors were abuse of drugs, nicotine (more than five cigarettes per day), alcohol, and caffeine. Caffeine was restricted to 200 ml coffee in the morning.

The experiment was approved by the Ethics Committee of the Faculty of Medicine of the University of Munich. After the purpose of the study had been explained to the subjects, all of gave their written informed consent according to the tents of the declaration of Helsinki.

The study was performed between June 2010 and December 2012 in our sleep laboratory with four sessions separated by at least one week, maximum 14 days. Each session consisted of two nights in the sleep laboratory, one adaptation and one study night. During study nights, the following medications were administered according to randomized schedule:

  • A.)

    4.5 mg MPP 22 intranasally and placebo orally (p.o.)

  • B.)

    9.0 mg MPP 22 intranasally and placebo p.o

  • C.)

    Placebo intranasally and 10 mg zolpidem p.o.

  • D.)

    Placebo intranasally and placebo p.o.

The study medication was obtained by M et P Pharma AG, Emmetten, Switzerland. MPP 22 is a proprietary formulation for intranasal application containing natural progesterone, intranasal placebo is the matrix of MPP 22 without active ingredient. 9.0 mg MPP 22 was the certified maximum dose available at the time of beginning the study. Oral zolpidem (Stilnox®, immediate release) and oral placebo were blinded by encapsulating them into identical capsules.

All subjects were administered either treatment A, B, C or D once in accordance with the randomisation schedule during each of the four study nights, respectively. For intranasal application MPP22 dose was divided into two equal halves in 2 containers, i.e. 2 × 2.25 mg, 2 × 4.5 mg or 2 x placebo, respectively and administered to the left and to the right nostril. The test substances were given intranasally at 22.05 and orally at 22.55. The gap between the each study period was at minimum 7 days to maximum 14 days.

Primary outcome were the conventional sleep-EEG variables total sleep time (TST), sleep period time (SPT), sleep efficiency index (SEI), sleep onset latency (SOL), amount of wake after sleep onset (WASO), the time spent in the various sleep stages and the power spectra derived from quantitative sleep-EEG.For definitions see 2.3.2. Secondary variables were subjective sleep variables and nocturnal hormone secretion. These variables were assessed by the following methods.

Electrodes for polysomnographic recordings (Comlab 32 Digital Sleep Lab, Brain Lab V3.3 Software, Schwarzer GmbH Munich, Germany) were fixed between 21.00 h and 22.00 h. Subjects were not allowed to sleep until the lights were turned off at 23.00 h. Polysomnographic recordings were performed from 23.00 h to 07.00 h according to the international 10–20 electrode system (F3, F4; C3, C4, P3, P4, O1 and O2, all referenced against the contralateral ear lobe), with an electrooculogram and a chin electromyogram. The sampling rate for EEG channels was 250 Hz with a band-pass filter from 0.53–70 Hz.

Related to two EEG channels (C3-A2, C4-A1), EOG and EMG, sleep stages (awake, Stage 1–4 sleep [stage 3 and 4 mean slow wave sleep, SWS] and rapid eyemovement sleep (REM) were scored visually according to the standard guidelines by Rechtschaffen and Kales (1968). The raters were unaware of the treatment for all consecutive thirty- second intervals (epochs) from 23.00h-07.00 h. Sleep continuity parameters referred to TST (total time spent in REM and NonREM sleep during sleep period time), SPT (time between sleep onset and final awakening in the morning), SEI (sleep period time divided by time in bed), SOL (time between lights off and the first occurrence of stage 2 sleep), WASO (time spent awake after sleep onset until end of bedtime) and the time spent in different sleep stages. Sleep-EEG parameters were analyzed for SPT, as well as for the night half and the night thirds (time interval 23.00h-07.00 h)

EEG fragments with artifacts were excluded from analysis using an automatic procedure (focus on high signal amplitude, activity in 0.75–3 Hz and 25–45 Hz) followed by visual inspection. Artifact-free signals were submitted to Fast Fourier Transform (FFT) to obtain the power spectra. FFT was performed on 4-s sliding window with a one-second shift resulting in resolution of 0.25 Hz. Of particular interest for the study were power densities in the EEG frequency bands delta (0.5–4 Hz), theta (4.5–8.0 Hz), alpha (8.5–12.0 Hz), sigma (12.5–16.0 Hz) and beta (16.0–20.0 Hz).

The Leeds Sleep Evaluation Questionnaire consists of 10 standardized self-reporting 100 mm line analogue questions focused on subjective sleep onset, subjective sleep quality, ease of awakening from sleep and alertness and behavior following wakefulness.Subjects filled in the Leeds Sleep Evaluation Questionnaire (Parrott and Hindmarch, 1980, Tarrasch et al., 2003) after awakening in the morning.

During the four study nights blood was collected between 22.00 h and 07.00 h every 20 min by long catheter for later analysis of hormone plasma concentrations of progesterone, cortisol, GH and melatonin. Analsyes of cortisol and melatonin were performed by radioimmunoassays (DRG Instruments GmbH, Marburg, Germany). GH (hGH) was analyzed by Autoanlyses (Immunolite 2000, Siemens, Munich, Germany). Progesterone was determined by Autoanalyser (ELECSYS, Roche Diagnostics, Mannheim, Germany). Intra- and inter assay coefficients of variation for all analyses were less than 9%.

In order to be able to detect large size effects (at least 0.6 large) of treatment on the primary sleep-EEG variables with a type I error of 0.05 and a power of at least 80% a number of minimum 10 subjects was necessary. Thereby a correlation of maximal 0.1 among the repeated measures was presupposed. A final group size of at least 12 subjects was considered, since a dropout rate of 10–15% has been expected.

For randomized allocation of the participants to the sequence of medications a computer-generated list was prepared by the independent statistician (A.Y.).In each of the four study nights participants received a pill and a device for intranasal application. All pills were blinded by encapsulating them into identical capsules.

Accordingly the participants, the investigators and all other staff involved in the trial were blinded for the sequence of medication.

The effect of MPP22 and zolpidem on the considered sleep- and hormone-parameters was tested about significance by one-factorial multivariate analysis of variance (MANOVA) with repeated measures design. The “treatment” was within-subjects factor with four levels (placebo, 4.5 mg MPP22, 9.0 mg MPP22 and zolpidem). Because of the small sample size it was not advisable to use too many parameters as dependent variables in each of the MANOVA designs of the analysis. To avoid collinearities we first partitioned the multitude of sleep parameters into subsets and then performed the MANOVAs on each of these sets separately. Due to the small sample size, care was taken to ensure that only small quantities of 2–5 variables without strong linear dependencies among each other were taken into account in each MANOVA. For the hormones the indicators AUC (area under the curve) and mean concentration were used for statistical comparisons. One sample t-tests (with test value 1) were additionally conducted for testing about significance the ratios of MPP22 to placebo and the relative power densities. A nominal level of significance (α=0.05) was accepted. This was corrected according to Bonferroni procedure for all posteriori tests (univariate F-tests, tests of contrasts, etc.) in order to keep Type I error ≤ 0.05. Most results in the tables and figures are expressed as means ± SEM.

Section snippets

Conventional sleep-EEG analysis

12 subjects participated in the study. Conventional sleep-EEG analysis was performed after removal of two outliers, which showed a low sleep efficiency index (SEI < 0.65) after 4.5 and 9.0 mg MPP22.

SEI revealed significant differences between the four treatments (p = .004, ETA square = 0.383, power = 0.90), it was significantly higher after zolpidem than after the other three treatments. After 9.0 mg MPP22 SEI was elevated also significantly in comparison to placebo (univariate F-tests in

Sleep-EEG analysis

The major effects of 9.0 mg MPP22 on conventional sleep-EEG variables are increases in sleep efficiency, in total sleep time, in the time spent in non-REM sleep and in sleep stage 2 and decreases in intermittent wakefulness and WASO. Also after 4.0 mg MPP22 sleep stage 2 increased, whereas no other sleep-EEG effects were found. These changes suggest dose-dependent sleep promoting properties of MPP22. As expected, hypnotic properties of zolpidem were confirmed. Quantitative sleep-EEG analysis

Conclusions

Our study showed that intranasal progesterone counteracts sleep-EEG changes in menopausal women as after 9 mg MPP 22. NonREM sleep increases and intermittent wakefulness decreases: furthermore nocturnal progesterone plasma level is enhanced. Sleep promoting properties of intranasal progesterone may be used to treat impaired sleep in postmenopausal women.

Contributors

All authors have materially participated in the research and/or article preparation. All authors have approved the final article.

Authorship

All authors have made substantial contributions to all of the following: (1) the conception and design of the study, or acquisition of data, or analysis and interpretation of data, (2) drafting the article or revising it critically for important intellectual content, (3) final approval of the version to be submitted.

Role of the funding source

M et P Pharma AG Emmetten/Switzerland provided the study medication and financial support for the conduct of the study. Claudia Mattern is chief scientific officer of M et P Pharma and is one of the authors. She participated in the study design and in writing the report. The sponsor was not involved in the collection, analysis and interpretation of data; and in the decision to submit the article for publication.

References (27)

Cited by (23)

  • Endocrinology of sleep

    2023, Encyclopedia of Sleep and Circadian Rhythms: Volume 1-6, Second Edition
  • The role of ovarian hormones in the pathophysiology of perimenopausal sleep disturbances: A systematic review

    2022, Sleep Medicine Reviews
    Citation Excerpt :

    A small study found more PSG-awakenings/arousals in the luteal (mixed mid-late) versus the follicular phase in perimenopausal women with ovulatory cycles, however relationships with progesterone were not examined [46]. Progesterone treatment reduced sleep disturbances [54], increased total sleep time and reduced wake after sleep onset as measured with PSG [58], potentially accompanied by increased REM sleep [55]. Only a few studies established opposite results stating that progesterone was related to trouble sleeping [43,59].

  • Sleep and the ovarian axis

    2021, Current Opinion in Endocrine and Metabolic Research
    Citation Excerpt :

    When sleep under placebo was strictly normal, no effect of progesterone was detected [34]. When sleep under placebo was slightly disturbed, a minor decrease of nocturnal awakenings and a possible increase of light non-REM sleep were observed under progesterone [33,35,36]. When sleep was markedly disrupted by an IV catheter with 24-h blood sampling procedure [34], progesterone dramatically reduced wake after sleep onset (WASO) and increased non-REM sleep [34].

  • The pharmacodynamics and safety of progesterone

    2020, Best Practice and Research: Clinical Obstetrics and Gynaecology
    Citation Excerpt :

    In a randomised, double-blind, cross-over study in healthy postmenopausal women, intranasal P4 was shown to have a sleep-endocrine promoting effect, compared to zolpidem and placebo. The spectral signature of intranasal P4 did not mimic the sleep-EEG alterations induced by GABA active compounds [21]. The view that the neuroprotective and regenerative effects of P4 in the brain may be primarily mediated by allopregnanolone has recently been challenged by the observation that P4 receptors play a key role in the viability of neurons after ischemic stroke and in the remyelination of axons after a demyelinating lesion, with potentially significant therapeutic implications in patients with multiple sclerosis and carpal tunnel syndrome [22].

  • Modulation of nose-to-brain delivery of a P-glycoprotein (MDR1) substrate model drug (quinidine) in rats

    2020, Brain Research Bulletin
    Citation Excerpt :

    De Carvalho also tested intranasal dopamine and reported that it attenuates fear responses induced by electric shock to the foot and by electrical stimulation of the dorsal periaqueductal gray matter (de Carvalho et al., 2019). The cerebroprotective effect of intranasal progesterone has also been investigated both in preclinical (Guennoun et al., 2019) and clinical studies (Schüssler et al., 2018). The beneficial central effects of IN insulin were also demonstrated in animal (Simon et al., 2020; Long et al., 2019) and human studies (Heni et al., 2019).

View all citing articles on Scopus
View full text