New Treatment Approaches for Premenstrual Disorders

Published Online: December 01, 2005
Andrea J. Rapkin, MD

Several approaches to alleviating the symptoms of premenstrual disorders are available to women and can be tailored according to individual needs and preferences. This article discusses methods that entail changes to lifestyle and diet and managing life stresses without relying on drug therapy, as well as a variety of medications that may be necessary in addition to or in place of recommended lifestyle modifications. New pharmacologic research is promising and is discussed along with the need to provide empathetic counseling for patients to determine the approach that will work best for each individual.

(Am J Manag Care. 2005;11:S480-S491)

The term "premenstrual disorders" covers a spectrum of premenstrual symptom combinations, from mild premenstrual syndrome (PMS) to premenstrual dysphoric disorder (PMDD) that is severe enough to interfere with work and social functioning. Effective evaluation and treatment of PMS were hampered until the mid-1980s by the lack of established criteria for diagnosing this common condition. The American College of Obstetricians and Gynecologists (ACOG) published a practice bulletin in 2000 that included criteria for PMS, based on an earlier article by Mortola and colleagues, as well as a discussion of different approaches for treating PMS,1,2 including lifestyle modifications such as regular aerobic exercise and dietary changes. Pharmacologic options studied for treating severe PMS include selective serotonin reuptake inhibitors (SSRIs), anxiolytic agents, gonadotropin-releasing hormone (GnRH) agonists, the diuretic spironolactone, and combination oral contraceptives (OCs).

PMDD is defined as a psychiatric disorder in Appendix B of the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR).3 Currently, selected SSRIs are the only pharmacologic agents with a US Food and Drug Administration (FDA) indication for PMDD.

Prior to the early 1990s when PMDD was defined as "late luteal dysphoric disorder" in DSM-III, various severities of PMS and PMDD were often investigated and discussed without differentiation. Therefore, many of the early studies in which the term "PMS" was used probably included patients with PMDD as well. It is necessary to keep this fact in mind when reviewing the literature on premenstrual disorders.

In this article, a number of available nonpharmacologic and pharmacologic treatments are reviewed, as well as recent advances in pharmacotherapy for premenstrual disorders.

Lifestyle Modifications

Lifestyle modification rather than drug therapy may be the most appropriate treatment approach for women with mild PMS symptoms. Physicians should always inform their female patients about lifestyle changes that may ameliorate their premenstrual symptoms and advise them to evaluate the effect of various approaches during the 2 months for which they keep a daily symptom diary. (At least 2 months of prospective daily symptom recording are required for a diagnosis of PMS or PMDD.) Regular aerobic exercise, for example, eases premenstrual symptoms for many women.4 The decline in endorphin levels that normally occurs in the late luteal phase of the menstrual cycle has been suggested to be an underlying mechanism for premenstrual symptoms in some women. Because regular aerobic exercise leads to the release of endorphins in the central nervous system, physicians should recommend that women perform at least 20 to 30 minutes of aerobic exercise per day for at least 3 days each week.5

Dietary and nutritional modifications have also been used over the years to treat premenstrual symptoms. One such approach, calcium supplementation, was studied by Thys-Jacobs and colleagues in 466 evaluable women with moderate-to-severe premenstrual symptoms that had been documented over 2 cycles.6 Participants were randomized to receive 1200 mg/day of elemental calcium or a placebo for 3 cycles. Premenstrual symptoms were significantly lower in the calcium-treated group than in controls in the second (P = .007) and third (P <.001) treatment cycles. Therefore, calcium supplementation appeared to reduce premenstrual symptoms in some women.

Other studies have suggested that excess alcohol, salt, and caffeine intakes may actually worsen premenstrual symptoms by decreasing magnesium levels.7 For example, Walker et al conducted a double-blind, placebo-controlled, crossover study in which 41 evaluable women were randomized to 200 mg/day of magnesium or placebo for 2 cycles before being crossed over to the alternate treatment for 2 additional cycles.8 Walker and colleagues observed that daily magnesium supplementation significantly lowered mild symptoms of fluid retention (ie, weight gain, breast tenderness, swelling of extremities, and abdominal bloating) compared with placebo in the second cycle of administration (P = .0009), but not in the first cycle.7

Another study conducted by Freeman and colleagues included 53 women with premenstrual symptom rates that were 30% higher during the late luteal phase than in the follicular phase.9 Patients were randomized to a commercial carbohydrate-rich beverage or to an isocaloric placebo beverage taken twice daily for 5 days before the anticipated onset of menses. Mood symptoms were decreased in approximately one third of women consuming the carbohydrate- rich beverage, compared with 5% of the placebo group.

A study indicated that vitamin B6 had some clinical benefit in reducing premenstrual symptoms,10 but doses in excess of 100 mg/day can be harmful.1 Additionally, the herbal product evening primrose oil has not been shown to be effective in treating premenstrual symptoms11; however, the ACOG practice bulletin indicated that it may decrease breast tenderness.1 Finally, with regard to nutrition, reductions in salt, sugar, alcohol, and caffeine intake are often suggested for relieving premenstrual symptoms, but these approaches have not been investigated extensively in controlled studies.12

Pharmacotherapeutic Options

Compared with the nonpharmacologic approaches, pharmacotherapeutic options for managing premenstrual disorders have been investigated in greater detail. However, the study techniques employed have varied widely, including methods of diagnosis, outcomes analyzed, and methods of outcome measurement. Studies should include procedures for recording improvements in psychological symptoms and physical symptoms as well as overall improvement. Daily self-report diaries constitute the primary measurement, but some clinician-rated scales have also been validated. In addition, symptom assessment should include several months of tracking to confirm the diagnosis before entry into the study and placebo runin periods to exclude placebo responders.

Antidepressants. Of numerous options available, antidepressants from the class of the SSRIs may be considered the therapy of choice for PMDD in many patients. Currently, the only agents with an FDA indication for PMDD are fluoxetine hydrochloride, sertraline hydrochloride, and paroxetine hydrochloride. Unlike tricyclic antidepressants, which interact with several receptors, the SSRIs interact minimally with receptors other than the serotonin (5-HT) reuptake receptor.13 Fluoxetine has a recommended dose of 20 mg/day (Figure 1); in clinical studies, no added benefit was observed with increasing the dosage to 60 mg/day. Sertraline is initiated at a dose of 50 mg/day and can be increased up to 150 mg/day for daily dosing or up to 100 mg/day for dosing only during the luteal phase of the cycle. Paroxetine is initiated at a dose of 12.5 mg/day and can be increased to 25 mg/day.14 Clinical trials that formed the basis for approval of these 3 SSRIs for managing PMDD symptoms and additional trials with other SSRIs are listed in Table 1.



Several adverse effects are associated with daily use of the SSRIs that have received an FDA indication for PMDD (Table 2). Individual response to these side effects may lead to poor adherence or discontinuation of these medications.


In a study to investigate compliance to antidepressant agents prescribed for PMS, Sundström-Poromaa and colleagues noted reasons given by these patients for discontinuing antidepressant use.15 A total of 170 (84.2%) of the 202 women who were prescribed an SSRI or a tricyclic antidepressant for PMS during a 4-year period completed a written questionnaire. The 22 (12.9%) women who never started treatment listed their primary reasons as fear of negative side effects (54.5%) and not wishing to take this type of drug (54.5%). (A woman could give more than 1 reason for not initiating antidepressant therapy.) Of the 148 (87.1%) women who did start therapy, 91 (61.5%) had discontinued the antidepressant by the end of 2 years. Table 3 lists the reasons for discontinuation of therapy.


Women who experience severe side effects can be advised to switch to a different SSRI. Should they choose to switch to a different drug class, the SSRI dose must be tapered slowly to avoid discontinuation symptoms. In addition, the FDA recently revised the safety labeling for SSRIs to advise against their use in patients younger than 18 years and to warn patients with major depressive disorder of the risk for worsening symptoms and/or for suicidal ideations.

Studies with non-SSRI/selective norepinephrine antidepressants have had less favorable results compared with SSRIs. A comparative study of treatment with sertraline and desipramine (flexible dosage range 50-150 mg/day) vs placebo in 189 subjects with PMS/PMDD showed that sertraline was significantly more effective than placebo on the Penn Daily Symptom Report (>50% symptom decrease in 65% of subjects in the sertraline groups), whereas desipramine was not (symptom decrease in 36% of subjects in the sertraline group and 29% in the placebo group).16 Another comparative study investigated fluoxetine 20 mg/day, bupropion 100 mg/day, and placebo in 34 women with PMDD. Fluoxetine was superior to both bupropion and placebo in efficacy by Global Clinical Impression ratings. Posttreatment Hamilton Rating Scale for Depression and Global Assessment Scale rating scores were intermediate between but not significantly different from fluoxetine or placebo.17

On the whole, studies conducted with antidepressants in women with PMS and PMDD indicate that serotonergic activity is required for efficacy. The SSRIs have been investigated for both continuous and lutealphase (intermittent) administration, and fluoxetine, paroxetine CR (controlled release), and sertraline are approved for use in PMDD without specification of the regimen, so can be employed continuously or intermittently. Onset of efficacy is rapid: therapeutic benefit is seen in the first menstrual cycle after initiation of treatment with these agents. Patients on systemic hormonal contraceptives were excluded from many of the SSRI trials, so the efficacy of SSRIs in combination with systemic (including oral) hormonal contraceptives for the continuous daily treatment of PMDD is unknown.

Anxiolytics. The anxiolytic agent alprazolam has not shown consistent results in studies evaluating its effectiveness in alleviating premenstrual symptoms, according to the ACOG practice bulletin.1 In a study conducted by Evans and colleagues, women with premenstrual symptoms were given either alprazolam (0.25, 0.50, or 0.75 mg) or placebo during both the luteal and follicular phases of the cycle under controlled laboratory conditions.18 It was observed that acute doses of alprazolam did not improve negative premenstrual mood and actually were associated with an increase in negative mood in the follicular phase.18 Also, alprazolam impaired task performance in both phases of the cycle. As a result, acute administration of alprazolam was not deemed a clinically useful treatment for premenstrual symptoms. In addition, continued use of alprazolam can lead to dependency, and some users develop a tolerance to this agent.

GnRH agonists. A third treatment option is the class of GnRH agonists, which use a hormonal approach to suppress ovarian steroid hormone production and prevent ovulation, in effect by inducing medical oophorectomy.4 These agents have demonstrated efficacy in alleviating several premenstrual symptoms. For example, in a small study conducted by Mortola and colleagues,19 GnRH monotherapy was associated with at least a 75% improvement from baseline in the Calendar of Premenstrual Experiences (COPE) scores for behavioral (P <.01), physical (P <.05), and total (P <.01) symptoms. However, because GnRH agonists induce medical menopause, estrogen and progestin must be added back to prevent bone loss and potentially for cardioprotection.4 According to an additional, small study, 10 women with PMS symptoms given leuprolide acetate at 3.75 mg/mo for 3 months had a significant decrease in symptoms measured using a Daily Rating Form and the observer form of the Rating Scale for Premenstrual Tension Syndrome. Addition of either progesterone vaginal suppositories or a 17β-estradiol patch or the leuprolide regimen resulted in significant return of symptoms. No changes in mood occurred in 15 normal women who received the same regimen. The authors conclude that in women with PMS, the occurrence of symptoms represents an abnormal response to normal hormonal changes.20

Synthetic androgens. Danazol, a synthetic androgen indicated in the United States for the treatment of endometriosis, menorrhagia, fibrocystic breast disease, and hereditary angioedema, has also been investigated for the management of PMS and premenstrual mastalgia, with moderate results. Luteal-phase-only danazol was not effective for treating the general symptoms (daily analogue scale scores) of premenstrual syndrome, but appeared highly effective for relieving premenstrual mastalgia in a study conducted on 100 women.21 A smaller study, conducted in 31 women meeting rigorous criteria for a diagnosis of severe PMS, evaluated effects of danazol treatment using the Premenstrual Tension Self-Rating Scale, the Beck Depression Inventory, and a visual analogue scale. Danazol 200 mg bid provided greater symptom relief than did placebo.22 Potential adverse effects of danazol are a cause for concern with this agent.23

Diuretics. Another therapeutic consideration is spironolactone, an aldosterone receptor agonist derived from 17α-spirolactone. According to the ACOG practice bulletin, thiazide diuretics have not been demonstrated to be beneficial in alleviating premenstrual fluid retention, but spironolactone has, in fact, been shown to have benefit in PMS.1 Spironolactone also has been shown to relieve other symptoms associated with the premenstrual phase of the cycle,4 as noted in a double-blind, parallel-group study over 3 cycles conducted by Vellacott and colleagues. 24 Sixty-three women, aged 16 to 45, with a history of at least 6 months of cyclic symptoms were randomized to spironolactone 100 mg/day or to placebo, given from day 12 of the menstrual cycle to the onset of menses. Spironolactone was significantly superior to placebo in relieving the general feeling of bloating (P = .001). By cycle 3, more than half of the 26 women using spironolactone also experienced improvement of abdominal swelling, swelling of the hands and feet, breast discomfort, irritability, depression, anxiety, tension, and increase in sexual interest.

Spironolactone was also examined in a double-blind, crossover study conducted by Wang and colleagues in 35 women with PMS.25 Two pretreatment cycles were used to diagnose PMS, after which the treatment phase began, consisting of two 3-month periods. Women were randomized to 100 mg/day of spironolactone or placebo administered daily from day 14 of the menstrual cycle until the onset of menses. The primary outcome measure was the prospective daily visual analogue scale. During the intervals when women were taking spironolactone, they experienced significant improvements in negative symptoms (ie, anxiety and tension, irritability, fatigue, and depression) and in physical symptoms (ie, headache, feelings of swelling, cravings for sweets, and breast tenderness) compared with baseline values and with placebo (P <.01 for all measures). Spironolactone treatment was also associated with an improvement in positive symptoms (cheerfulness, well-being, friendliness, feeling energetic) compared with baseline values (P <.01).

Oral contraceptives. Another hormonal option, combination OCs, is a popular choice for helping to relieve several premenstrual symptoms. In the United States, OCs contain estrogen as ethinyl estradiol (EE) in combination with a variety of progestins. EE causes a rise in serum aldosterone levels, which lead to sodium and water retention, thereby contributing to bloating and breast tenderness.26 All progestins have progestogenic activity, but they can differ in terms of other pharmacologic effects. The pharmacologic profiles of progesterone and various progestins used in OCs (as found in animal models) are demonstrated in the work of Krattenmacher.27

Until recently, very few controlled studies had evaluated the efficacy of OCs in reducing premenstrual symptoms, and those that had been conducted yielded mixed results. For example, in 1992 Graham and Sherwin studied 82 women with moderate-to-severe premenstrual symptoms in a double-blind, placebo-controlled trial.28 The women charted daily symptoms for 1 cycle, after which they were randomly assigned to a triphasic OC containing EE 35 μg and norethindrone (0.5 mg, 1.0 mg, and 0.5 mg) or to placebo for 3 cycles. A total of 23 women (28%) dropped out of the study (18 in the OC group and 5 in the placebo group). Com pared with placebo, the OC significantly reduced premenstrual breast pain and bloating (P <.03) but did not have significantly better effects on mood symptoms.

Similarly, in a double-blind crossover study of 3 OCs, 36 women aged 20 to 40 who either had PMS or experienced symptoms throughout their entire menstrual cycle with premenstrual aggravation were examined.29 The study consisted of 2 treatment cycles followed by a crossover to the alternate preparation for 2 additional treatment cycles. Nineteen women were randomly assigned to a monophasic OC containing EE combined with either desogestrel (DSG) or levonorgestrel (LNG), and 17 women were randomized to treatment with either monophasic EE/DSG or a triphasic OC containing EE plus LNG. Mood scores improved from baseline for all 3 OCs, but Bäckström and colleagues concluded that the beneficial effect observed in the study was no higher than that reported for placebo in other studies. The monophasic DSG pill resulted in less change in mood parameters than did the monophasic and triphasic LNG OCs. However, physical complaints were less frequently reported during the use of the triphasic preparation as compared to the monophasic DSG preparation.29

Finally, a nested case-control study conducted by Joffe and colleagues examined a cohort of 976 women, aged 36 to 45.30 Of the 658 women who had previously used a variety of OCs, 107 (16.3%) reported pill-related premenstrual mood deterioration, 81 (12.3%) reported premenstrual mood improvement, and 470 (71.4%) reported no effect of OCs on premenstrual mood. Therefore, it was concluded that OCs do not affect premenstrual mood in most women.30

All but one of the progestins being used in OCs in the United States are derived from 19-nortestosterone. The exception is the progestin drospirenone, which is derived from 17α-spirolactone and is an analogue of spironolactone. The pharmacologic profile of drospirenone closely resembles that of natural progesterone in that it has potent progestogenic, antiandrogenic, and antimineralocorticoid activities, and no androgenic activity.26 The antimineralocorticoid activity of 3 mg of drospirenone is comparable with 25 mg of spironolactone.31 Drospirenone acts by binding to aldosterone receptors, blocking aldosterone action in the kidneys, resulting in a substantial rise in sodium and water excretion and some retention of potassium (Figure 2).27


In recent years, several studies have examined the efficacy of the OC formulation containing EE 30 μg plus drospirenone 3 mg (30EE/drospirenone) on premenstrual symptoms. For example, a double-blind trial included 82 women with PMDD who were randomized to 30EE/drospirenone (n = 42) or placebo (n = 40).32 The drospirenone-containing OC was observed to have a positive effect on symptoms of PMDD, with the between-group differences reaching statistical significance in appetite, food cravings, and acne (P = .03). In addition, Apter and colleagues conducted an open, 6-cycle study of 336 women to evaluate the actions of 30EE/drospirenone on fluid-related symptoms during the luteal phase of the cycle and the effects of these symptoms on overall well-being.33 Use of 30EE/drospirenone was associated with a significant reduction in the incidence and severity of the abdominal bloating and breast tenderness (both P <.001) associated with the menstrual cycle. Also, the significant beneficial effect of 30EE/drospirenone on psychologic well-being (P <.0001), as measured by the Psychological General Well-Being Index, observed at cycle 3 was maintained at cycle 6.

Further, an open-label, 26-cycle study was conducted by Foidart and colleagues of 627 evaluable women: 310 were randomized to 30EE/drospirenone and 317 to EE 30 μg plus DSG 150 μg.34 Compared with the EE/DSG group, women who were given 30EE/drospirenone experienced a greater incidence of premenstrual symptoms before the study and a lesser incidence throughout the study. The between-group differences were not statistically significant. More recently, Sangthawan and Taneepanichskul conducted an open-label, 6-cycle study of 99 evaluable women who were randomized to either 30EE/drospirenone or EE 30 μg plus LNG 150 μg.35 The prevalence of premenstrual symptoms was reduced from 58.0% at baseline to 32.0% at cycle 6 in the 30EE/drospirenone group and rose from 59.2% at baseline to 61.2% at cycle 6 in the EE/LNG group. The between-group difference was statistically significant (P = .005).

Positive results with drospirenone were also noted by Borenstein et al, who analyzed responses of 858 women who completed a survey when initiating 30EE/drospirenone and again after 2 cycles of treatment.36 Compared with baseline values, 30EE/drospirenone use was associated with significant reductions in premenstrual symptoms (P <.001) and improvement in the women's sense of well-being (P <.05). Finally, Sillem and colleagues conducted an observational study of 1433 women using 30EE/drospirenone, 175 of whom continuously took this OC between 42 and 126 days using an extended regimen.37 Although it was not designed specifically to evaluate premenstrual symptoms, this study did monitor some symptoms associated with PMS and PMDD. A reduction in edema was experienced by 31% of new users and 40% of the switchers and by 34% of the women receiving the standard regimen compared with 49% of the women receiving the extended regimen (P <.001). A decrease in breast tenderness was reported by 40% of new users and 42% of the switchers and by 40% of the women receiving the standard regimen compared with 50% of the women receiving the extended regimen (P = .046). A reduction in bloating was experienced by 31% of new users and 30% of the switchers and by 29% of the women receiving the standard regimen compared with 37% of the women receiving the extended regimen. Table 4 summarizes the results of the studies of OCs in women with premenstrual symptoms.


The beneficial results observed in the studies cited should be considered in light of potential side effects associated with OC use in some women, including nausea, breakthrough bleeding, weight gain, breast tenderness, and headache, as well as contraindications with certain coexisting medical conditions. However, as with SSRIs, women can switch to a different OC if side effects make this necessary. Given the known noncontraceptive health benefits of OCs, especially their favorable effects on several premenstrual symptoms, they are strong candidates for patient use.

Recent Research Findings for OC Formulations and Regimens

PDF is available on the last page.
Adult ADHD Compendium
COPD Compendium
Dermatology Compendium
Diabetes Compendium
GI Compendium
Immuno-oncology Compendium
Lipids Compendium
MACRA Compendium
Oncology Compendium
Pain Compendium
Reimbursement Compendium
Rheumatoid Arthritis Compendium
Know Your News
HF Compendium
Managed Care PODCAST