Progesterone Reduces Wakefulness in Sleep EEG and Has no Effect on Cognition in Healthy Postmenopausal Women

Objectives/ Mode of administration: To characterise the effects of progesterone in a dose of 300mg each for 21 days on changes of sleep EEG and cognition in postmenopausal healthy women, in comparison with placebo.

Methodology:The study was performed as double blind randomized placebo controlled crossover design with two treatment intervals of 21 days duration separated by 2 weeks washout. Progesterone was administered as tablets in a dose of 300mg each for 21 days. At the beginning and the end of the two intervals a sleep EEG was recorded after one night of adaptation to laboratory conditions, during which EEG electrodes were attached without recording an EEG.For the registration nights the subjects arrived at the sleep laboratory at 21.00 h. Sleep EEG was recorded from 23.00 h until 07.00 h hours next morning.Cognitive performance was tested after the registration nights at 09.00h.All pharmacodynamic investigations were carried out in the sleep laboratory of the Max Planck Institute of Psychiatry in Munich.

Main criteria for inclusion: Subjects consisted of 10 healthy women (age: range 54-70 years) who entered the study after passing rigid psychiatric, physical and laboratory examinations. They had been drug free at least 4 weeks. Reasons for exclusion from the study were: psychiatric disorder in the subjects own or family history, including dementia or other cognitive impairment; recent stressful life events; substance abuse; a transmeridian flight during the last three months; shift work; medical illness; and aberrations in blood chemistry or in the waking EEG or electrocardiogram.All subjects underwent a polysomnographic examination in the sleep laboratory before entering the study to exclude sleep disorder including sleep related respiratory disorders and sleep related movement disorders.

Statistical methode: Expolatory and inferential statistical analysis of certain aggregated sleep variables after classifying the sleep EEG in dinstinct stages according to Rechtschafen/Kales and of the results of EEG spectral analysis. Inferential statistics were based on multivariate analyses of variance (MANOVAs) with treatment as influential factor and on one sample t-tests for placebo related power changes of progesterone in non-REM sleep as well.