European Journal of Endocrinology (2003) 148 571–577 ISSN 0804-4643 CLINICAL STUDY Progesterone effects during sequential hormone replacement therapy Lotta Andre´en, Mari Bixo, Sigrid Nyberg, Inger Sundstro¨m-Poromaa and Torbjo¨rn Ba¨ckstro¨m Department of Clinical Sciences, Obstetrics and Gynecology, Umea˚ University, Umea˚, Sweden (Correspondence should be addressed to T Ba¨ckstro¨m, Department of Obstetrics and Gynecology, University Hospital, S-901 87 Umea˚, Sweden; Email: [email protected]) Abstract Objective: The aim was to investigate the effect on mood and the physical symptoms of two dosages of natural progesterone and a placebo in postmenopausal women with and without a history of premenstrual syndrome (PMS). Design: A randomized, placebo-controlled, double-blind, crossover study was performed. Method: Postmenopausal women ðn ¼ 36Þ with climacteric symptoms were recruited. They received 2 mg estradiol continuously during three 28-day cycles. Vaginal progesterone suppositories with 800 mg/day, 400 mg/day, or placebo were added sequentially for 14 days per cycle. Daily symptom ratings using a validated rating scale were kept. Results: Women without a history of PMS showed cyclicity in both negative mood and physical symp- toms while on 400 mg/day progesterone but not on the higher dose or the placebo. Women without a history of PMS had more physical symptoms on progesterone treatment compared with placebo. Women with prior PMS reported no progesterone-induced symptom cyclicity. Conclusion: In women without prior PMS natural progesterone caused negative mood effects similar to those induced by synthetic progestogens. European Journal of Endocrinology 148 571–577 Introduction its conversion to allopregnanolone, active on the GABAA receptor (10). A paradox has emerged as all synthetic progestogens It is well known that sequential hormone replacement tested so far induce negative mood effects in humans, therapy (HRT) causes cyclicity in mood and physical while natural progesterone seems to have the opposite symptoms similar to those symptoms encountered effect on mood, at least in animal experiments. So far, during ovulatory cycles and in premenstrual syndrome in humans, most studies on mood and natural pro- (PMS). It seems plausible to conclude that it is the gesterone have focused on treating PMS. These con- addition of synthetic progestogens in HRT that provokes trolled trials have been unable to confirm any these negative mood symptoms (1–5). superiority of progesterone to placebo in terms of the One major question, however, is whether or not lessening of PMS symptoms (11, 12). Instead, there natural progesterone has the same effect on mood as are indications that natural progesterone is involved synthetic progestogens (6). In laboratory animals, in negative mood changes. The recurrence of physical some of progesterone’s metabolites, i.e. 3a-OH-5a-preg- and psychological symptoms in the late luteal phase nan-20-one (allopregnanolone) and 3a-OH-5b-preg- in women with PMS is related to increased progester- nan-20-one (pregnanolone), have barbiturate- and one serum concentrations during this period (13). benzodiazepine-like effects mediated through their Between menstrual cycles, an intra-individual worsen- binding to the gamma aminobutyric acid (GABAA) ing of symptoms is related to higher progesterone receptor complex in the brain, thereby eliciting hypno- serum concentrations (14). An important question tic, anxiolytic, and anti-epileptic effects (7, 8). In therefore arises: is progesterone qualitatively different humans, progesterone can, in large doses, induce from progestogens in its ability to influence mood or anesthesia, and this effect is mediated via the metab- are the reported differences in symptom-provoking olism of progesterone to allopregnanolone and pregna- effects due to differences in progesterone/progestogen nolone (8, 9). Similarly, the acute anxiolytic effect of dose? To our knowledge, natural progesterone has not progesterone in laboratory animals is mediated through previously been investigated with respect to its possible q 2003 Society of the European Journal of Endocrinology Online version via http://www.eje.org Downloaded from Bioscientifica.com at 10/03/2021 03:38:53AM via free access 572 L Andre´en and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2003) 148 mood-improving or -deteriorating effects in postmeno- deterioration prior to menstruation which decreased pausal women on sequential HRT. and disappeared within 4 days after the onset of men- It has been reported that women with a history of strual bleeding. The influence of PMS symptoms on PMS, during their fertile life, respond with less positive daily life was graded, and women whose family and more negative mood symptoms compared with relations, social activities, or work were negatively controls under treatment with progestogens (3, 5). affected by the symptoms were considered to have a However, women with PMS seem to be less sensitive history of PMS. to the sedative effects of pregnanolone, a progesterone metabolite, and benzodiazepines compared with con- Study design trols (15, 16). In terms of negative mood symptoms, they seem to respond more strongly to oral contracep- The effect of two different progesterone doses on symp- tives than women who do not have PMS (17). Thus, a tom cyclicity was evaluated in a randomized, placebo- history of PMS may be of importance for the severity of controlled, double-blind, crossover design. The women reaction to progesterone in postmenopausal women were divided into three groups and treated for three (3). This motivates recording the presence of PMS 28-day cycles. Estradiolvalerat at a dose of 2 mg (Apote- during fertile life in postmenopausal women. ket AB, Stockholm, Sweden (the national pharmacy The objective of this randomized, double-blind, pla- company)) was given continuously throughout the cebo-controlled study was to investigate if the addition study period. Progesterone vaginal suppositories were of vaginal progesterone to estrogen treatment induces added on days 15–28 of each cycle. One group started effects on mood and physical symptoms and whether the sequential treatment with 800 mg/day progesterone, or not these are dose-dependent. Furthermore, the the second group with 400 mg/day progesterone, and study aimed to evaluate whether a history of PMS influ- the third group with placebo vaginal suppositories. All ences the potential adverse mood effects induced by suppositories were given as 400 mg, 200 mg, or placebo vaginal progesterone. twice a day. A crossover to a new treatment was carried out after each cycle. The two progesterone doses chosen are the two pharmaceutical preparations available in Materials and methods Sweden at present, and it has been shown that there is Subjects a significant difference in progesterone serum concen- trations between the two doses (19). The vaginal formu- Thirty-six women with climacteric symptoms were lation was a waxy suppository containing progesterone recruited and randomly assigned to treatment in a in a base of semi-synthetic glycerides produced from double-blind, crossover study. Subjects were more hydrogenated vegetable oil by interesterification. than 6 months postmenopausal and had not used The vaginal pessaries were prepared by Apoteket AB, any HRT for the past 3 months prior to inclusion in Production and Laboratories. The suppositories were the study. None were receiving any steroid treatment, made to appear identical. Packing and randomization and all had an intact uterus and no contraindications were done by the pharmacy at Umea˚ University to HRT. They had no history of psychiatric illness and Hospital. had not been treated with psychopharmacologic In women with climacteric symptoms, estrogen is drugs within the past 6 months. Subjects with ongoing known to decrease the vasomotor symptoms during psychiatric illness were excluded by the use of PRIME- the first month of treatment. In parallel with MD. PRIME-MD has been developed to help primary the improvement in vasomotor symptoms there is care physicians to screen, evaluate, and diagnose an increase in well-being due to relief of symptoms. mental disorder. This diagnostic tool is constructed to To avoid interference with the mood changes related conform to Diagnostic and Statistical Manual of to improvements in vasomotor symptoms, a run-in Mental disorders, Fourth edition (DSM-IV) criteria and cycle with an ordinary sequential HRT treatment was has been validated for use in a primary care setting given in order to treat the vasomotor symptoms before (18). PRIME-MD is fully described elsewhere (18). The the study proper started (20). Therefore, the study agreement between PRIME-MD and independent began with a 28-day run-in cycle where the patients psychiatric diagnoses guided by a structured interview were treated with 2 mg estradiolvalerat daily and is generally excellent across modules, with an overall 10 mg medroxyprogesterone (MPA) on days 15–28 of accuracy of 88% (18). Before inclusion, patients gave the cycle. The drawback of this procedure is that all their written consent and agreed to keep a daily cycles following a progestogen treatment will have a record of their symptoms. They also underwent a physi- 3- to 4-day period at the beginning of the next cycle cal and gynecological examination, including a routine when the symptoms from the previous cycle decline (3). vaginal ultrasonography. The primary outcome measure was the daily symptom The study included women with and without a his- ratings made by the patients throughout the study. tory of PMS. Premenstrual symptoms during fertile We used a modified form of the cyclicity diagnoser life were defined by a retrospective report of mood ((CD) Diagnoskonsuelt, Umea˚, Sweden), an instrument www.eje.org Downloaded from Bioscientifica.com at 10/03/2021 03:38:53AM via free access EUROPEAN JOURNAL OF ENDOCRINOLOGY (2003) 148 Progesterone effects during HRT 573 for diagnosing cyclic symptoms that has been validated (PMS vs non-PMS), progesterone dose (placebo, for the diagnosis of PMS (21) but also used to evaluate 400 mg, or 800 mg), and progesterone day (days HRT-related symptom changes in postmenopausal 15–28).