Psychotropic medications and other non-hormonal treatments for premenstrual disorders

Antidepressant medications

Antidepressant medications that enhance serotonergic activity are established treatments for premenstrual dysphoric disorder (PMDD), approximating 60% efficacy. Most of the studies have examined fluoxetine, sertraline and paroxetine. Each of these selective serotonin re-uptake inhibitors (SSRIs) has Food and Drug Administration approval in the USA for the treatment of PMDD but they are not licensed in the UK or continental Europe. SSRIs may be taken continuously (every day of the cycle), intermittently (from ovulation to the first or second day of menses), or with the onset of symptoms whenever that occurs during the luteal phase with discontinuation at menses. It is important to confirm that a woman has a core premenstrual disorder (PMD), i.e. moderate–severe premenstrual symptoms, absence of symptoms during the follicular phase and functional impairment. ¹ The choice of continuous versus intermittent dosing is generally tailored to the individual woman, taking into account symptom profile, adverse effects, cost and patient preference. Generally, intermittent dosing can start at the full dose and be discontinued abruptly the first or second day of menses. The lack of discontinuation symptoms may be due to the brief (2 week) exposure and rapid but unknown mechanism of action of the SSRI for premenstrual symptoms. ² Doses that are reported to be effective are in the low-to-mid range of recommended doses for major depressive disorder.

Selective serotonin re-uptake inhibitors

A meta-analysis of 29 placebo-controlled randomized controlled trials (RCTs) (2964 women) reported that SSRIs were effective for treating both PMDD and severe premenstrual syndrome (PMS) with an odds ratio (OR) of 0.40. ³ The meta-analysis included studies of fluoxetine with continuous dosing ^4–7 and intermittent dosing; ^8,9 sertraline with continuous dosing ^10–12 and intermittent dosing; ^11,13–15 paroxetine with continuous dosing ^16–18 and intermittent dosing; ^17,19 citalopram; ²⁰ and fluvoxamine. ²¹ This meta-analysis concluded that continuous dosing regimens were more effective (OR 0.28) than intermittent regimens (OR 0.55). The pooled effect size of SSRI treatments for PMDD and PMS were essentially equivalent and the effect size for both has decreased in more recent studies, potentially due to the inclusion of more treatment-resistant women in recent clinical trials. ³

A Cochrane review examined the efficacy of SSRIs from 40 placebo-controlled RCTs (2294 women) and combined data from 22 of the 40 studies into a meta-analysis. ²² Studies included in the meta-analysis that were not included in the previous Shah et al. meta-analysis included a continuous dosing study with paroxetine ²³ and an intermittent study with clomipramine. ²⁴ This review reported that SSRIs were effective for behavioral and mood symptoms, physical symptoms and functioning symptoms. This study reported that both continuous and intermittent dosing were effective. Three studies demonstrated overall weak superiority of SSRIs compared with non-serotonergic antidepressants. ^5,10,16 The most common side-effects reported were nausea, insomnia, headache and asthenia. ²²

Since these reviews were published, a study comparing luteal phase escitalopram 20 mg, 10 mg and placebo in 151 women has been published. ²⁵ Escitalopram 20 mg was superior to both 10 mg and placebo in reducing a composite of mood and anxiety premenstrual symptoms. Another study published since the recent meta-analyses compared luteal phase paroxetine 20 mg, 10 mg and placebo in 99 women. ²⁶ Similarly, the higher dose of paroxetine (20 mg), but not 10 mg, was superior to placebo on several premenstrual mood and anxiety symptoms. In this study, neither dose of paroxetine used in intermittent dosing was effective in reducing premenstrual physical symptoms. Both of these studies, as well as previous studies using fluoxetine, ⁸ suggest that when intermittent dosing of a SSRI is used, lower doses (10 mg) compared with 20 mg, may have no significant effect on decreasing premenstrual physical symptoms.

The issue of whether SSRIs differentially treat subtypes of symptoms was recently examined in 476 women who had participated in three studies with sertraline. ²⁷ Response to sertraline was examined for specific emotional, behavioral and physical premenstrual symptoms. Sertraline was clearly superior to placebo for women with prominent psychological symptoms (e.g. irritability) or mixed psychological and physical symptoms (e.g. breast tenderness and abdominal bloating). However, sertraline was not effective for women with predominantly physical premenstrual symptoms and not very severe emotional symptoms. This is the first large study to suggest a differential efficacy of an SSRI for different premenstrual symptom profiles. ²⁷ Several other studies have confirmed the lack of efficacy of SSRIs for premenstrual headaches.

Serotonin norepinephrine re-uptake inhibitors

Studies with the serotonin norepinephrine re-uptake inhibitors (SNRIs) were not included in the meta-analyses mentioned above. One RCT suggested that continuous dosing of venlafaxine was superior to placebo, ²⁸ and an open study of intermittent use of venlafaxine reported a lack of problematic discontinuation effects with discontinuation at menses. ²⁹ It is possible that the efficacy of venlafaxine may be due to its acting as a SSRI at lower doses. Two open studies reported improved premenstrual symptoms with continuous dosing of duloxetine 60 mg, ^30,31 suggesting that this medication warrants systematic study.

Length of treatment

One RCT has been published examining the relapse of PMD with long-term use of a SSRI and following its discontinuation. ³² Women who responded to sertraline 50 or 100 mg daily were discontinued after four months or one year in a blinded trial. Relapse occurred in 60% of women after four months of treatment (median time to relapse was 4 months) compared with a 41% relapse rate in women after one year of treatment (median time to relapse was 8 months). Relapse was more likely if a woman had more severe symptoms prior to treatment and if she had not achieved full symptom remission with sertraline prior to discontinuation. ³²

Special populations

Adolescents are thought to have similar prevalence rates of PMD as women older than age 18, ³³ and treatment studies in adolescents are virtually non-existent. Women trying to get pregnant or who are at risk of becoming pregnant should choose an antidepressant with optimal safety data (i.e. not paroxetine or newer drugs with minimal data). Some women who are trying to conceive choose intermittent or symptom-onset dosing to minimize fetal exposure, or may use alprazolam as needed. Some women presenting for treatment for a PMD, have a variant PMD with follicular symptoms due to an underlying disorder. ¹ Treatment of women with premenstrual exacerbation of underlying mood and anxiety symptoms with semi-intermittent dosing, or ‘'bumping up’ the dose during the luteal phase only, has had a few positive reports, ^20,34 but systematic study is needed.

Antidepressant treatment during perimenopause

Many women at midlife develop the onset of bothersome emotional and physical symptoms. Although the menopause is a time of risk for the expression of a first episode of major depression or a recurrence of major depression, many women have symptomatic but subsyndromal mood changes, anxiety, decreased concentration and sleep difficulties. Vasomotor symptoms are common during the menopausal transition and hot flushes can be bothersome. Although estrogen is clearly effective in the relief of vasomotor symptoms, multiple studies report that SSRIs and SNRIs provide moderate relief for hot flushes during the natural menopausal transition as well as in women who have received cancer treatments. ³⁵ The increase in serotonin and norepinephrine transmission achieved with SSRIs and SNRIs may help modulate thermoregulation, although the precise mechanism of action is unknown. ³⁵ Although not studied, it is possible that midlife women taking SSRIs for their PMD may achieve the benefit of reduction in perimenopausal mood and anxiety symptoms and insomnia, in addition to the decrease in hot flush frequency and bother.

A challenging issue that arises during the menopausal transition is the likely necessity of dosage adjustments as menses become irregular. Women who achieved relief of their PMD with intermittent dosing during regular menstrual cycles may no longer know when to start the SSRI if their menses occurs every two weeks or every 3–4 months. Some women report that their premenstrual symptoms become less severe as they age ³⁶ and ovulate less regularly, other women report that their premenstrual symptoms are markedly more severe ‘than usual’ prior to the infrequent menses that occur during the perimenopause. Women who feel that their symptoms are decreasing in intensity may choose to change to symptom-onset dosing with an SSRI, ^37,38 using the medication for a few symptomatic days every few months. Other women, particularly those with severe symptoms, may choose to change to continuous dosing since they cannot predict when an ovulatory cycle will occur. Women with PMDs should taper and discontinue the SSRI following the final menstrual period. If the SSRI had been treating a core PMD and not an underlying disorder with premenstrual exacerbation, women should not need a psychotropic medication with the absence of menstrual cycles.

Anxiolytics

Some studies reported benefit with luteal phase administration of alprazolam ^39–42 while others did not. ⁴³ Alprazolam is not considered a firstline treatment due to the side-effect of sedation, potential for abuse and need for tapering at menses. Buspirone has also been reported to be helpful; ^44,45 however, the efficacy of both of these anxiolytics is less than that for SSRIs. These medications currently have a role as an adjunctive medication for specific premenstrual symptoms not successfully treated by a SSRI or hormone regimen.

Dietary, lifestyle changes and alternative treatments

Multiple dietary recommendations have been proposed for treating PMDs, but few have been systematically evaluated. Recommendations include frequent feedings containing complex carbohydrates or protein, decreased refined sugar, decreased salt, decreased red meat and elimination of caffeine. It is theorized that serotonin levels in the central nervous system may be increased following ingestion of complex carbohydrates. Two small RCTs reported that a beverage containing simple and complex carbohydrates was superior to isocaloric placebo beverages in reducing premenstrual symptoms when taken on symptomatic days during the luteal phase. ^46,47

A study conducted in 466 women with severe PMS reported that calcium 600 mg twice a day was superior to placebo in improving premenstrual emotional and physical symptoms. ⁴⁸ Increasing calcium levels may regulate periovulatory calcium homeostasis but the exact mechanism is unclear. Meta-analyses reported that vitamin B₆ 100 mg daily demonstrated weak superiority compared with placebo in improving premenstrual symptoms. ^49,50 A recent RCT reported superior efficacy for an essential fatty acid preparation (containing linoleic acid, gamma-linolenic acid, oleic acid and vitamin E) in improving PMS compared with placebo. ⁵¹ However, previous RCTs with evening primrose oil (containing linoleic acid and gamma-linolenic acid) have failed to demonstrate superiority compared with placebo. ⁴⁹ Mixed results have been reported for magnesium and vitamin E. ⁴⁹

Chasteberry (Vitex agnus castus) has been reported to be efficacious for decreasing premenstrual emotional and physical symptoms compared with placebo. ^49,52 Chasteberry has been reported to be equivalent to fluoxetine and superior to vitamin B₆. Chasteberry has dopaminergic effects but its potential mechanism of action in PMDs is unknown. Herbs that may deserve further study include Hypericum perforatum, Ginkgo biloba and Crocus sativus/saffron. ^49,52

Several studies have been conducted with cognitive-behaviour therapy (CBT) for women with PMDs. The efficacy of cognitive therapy may include modification of negative thoughts and increasing coping strategies. Reviews of RCTs of CBT studies suggest weak superiority compared with placebo or waitlist controls. ^53,54 One study that compared CBT with fluoxetine reported that while both treatments were effective, fluoxetine was superior for anxiety and CBT had superior long-term efficacy at six months. ⁵⁵ Other psychological treatments for PMDs have undergone minimal study.

Exercise has been reported to improve premenstrual dysphoria, fatigue and bloating, but no RCTs of exercise have been conducted in women with prospectively confirmed PMDs. ⁵⁶ Exercise may improve premenstrual symptoms through increasing beta-endorphin levels and wellbeing. Small RCTs have suggested efficacy of sleep deprivation and light therapy. ⁵⁷ These treatments are theorized to regulate circadian rhythms; however, the duration of therapeutic effects are unclear. Acupuncture was reported to be superior to sham acupuncture and other treatments in a recent meta-analysis, but many of the studies had methodological issues. ⁵⁸

Conclusion

SSRIs have clear efficacy in PMDD, both with continuous and intermittent dosing. Although some studies suggest that intermittent dosing may be somewhat less effective than continuous dosing, it may be preferable for women with side-effects or cost concerns. Symptom onset dosing may have a role for milder symptoms or symptoms that are problematic on few luteal days. Symptom-onset dosing may become a desirable option at perimenopause, when menses become irregular and infrequent. Other women may choose to switch to continuous dosing due to difficulty predicting ovulation or an increase in symptom severity. Relapse of premenstrual symptoms following discontinuation of SSRIs is common; however, it is less likely with longer treatment, treatment to remission of symptoms, and less severe symptoms prior to treatment. Anxiolytics during the luteal phase can have an adjunctive role, and non-pharmacological treatments such as calcium, chasteberry and CBT may be of benefit to women with PMDs.

Competing interests

Received research support from Pfizer.