Evidence Based Treatment of Pre-Menstrual Syndrome in Family PracticeThis article by Dr. Sue Douglas is more than "just a berry". It is, however such an excellent review of this topic that we felt that it should be presented in "Just the Berries". Objectives - To evaluate the strength of evidence for various treatments for premenstrual syndrome (PMS) to arrive at a set of practical guidelines for its management in a Family Practice setting. The use of natural therapies will also be discussed. Quality of evidence - Advanced MEDLINE was searched from January 1990 to October 1999. The Cochrane library and Natural Medicine comprehensive data bases as well as personal contacts were also used. Main Message - There is little scientific evidence to support the effectiveness of vitamin B6, oral contraceptives, natural progesterone and thiazide diuretics as treatments for PMS. There are a wide range of treatments that may be effective including: aerobic exercise, a healthy diet rich in complex carbohydrates, stress reduction, vitamin E, magnesium, evening primrose oil, non steroidal antiinflammatories, spironolactone and ovulation supression with various hormonal regimes. There is good evidence that calcium carbonate and the selective serotonin inhibitors (SSRIs) are effective. To date there is little scientific evidence that various complementary products are effective, but some of which are potentially toxic. Conclusions - Aerobic exercise, a healthy diet and stress reduction are reasonable first line therapies for mild to moderate PMS. Calcium carbonate should also be considered as first line treatment for mild to moderate PMS. Initial treatment with SSRIs continuously or during the luteal phase may be considered as initial treatment for women with severe affective symptoms. Family physicians are in an ideal position to detect and treat this challenging condition. Pre-menstrual syndrome (PMS) is a common cause of significant psychological and physical distress in women during their reproductive years. Forty percent of women have symptoms that are severe enough to disrupt some aspect of their daily lives during the luteal phase of their cycle. Five percent of women are incapacitated by their symptoms.1 Family physicians are in an ideal position to diagnose and treat this common but often overlooked condition. Despite the magnitude of this problem, a lot of confusion exists in the medical and lay community alike about what is and is not effective for the treatment of PMS. This in part reflects the fact that the cause of PMS is still unknown, although a number of theories have been proposed including hormonal imbalances, micronutrient deficiencies and neuroendocrine dysfunction. A range of treatment options are currently available that reflect this theoretical diversity.2 There has also been a rise in the use of alternative therapies with the increasing popularity of complementary medicine. Given the diverse range of treatment options, including complementary therapies, it has become increasingly difficult for Family Physicians to confidently counsel patients on what is safe and effective for the treatment of PMS. Based on the above search strategies possible treatments were categorised as one of the following: not effective, possibly effective, or effective. Treatments were categorised as "Not effective" if there were good quality studies that showed no significant benefits, and/or studies showed that the risks/costs clearly outweighed the potential advantages, and/or the studies had serious methodological flaws that invalidated any positive findings. Treatments were classified as possibly recommended if there were some studies that showed positive benefits but their interpretation was limited by methodological limitations including: small study size and/or number of trials, lack of true control groups, and questionable clinical significance of positive findings. Finally the evidence for recommended treatments came from high quality randomised controlled trials that showed unequivocal positive benefits of treatment. It was not possible to use the above criteria to assess the effectiveness of complementary therapies because of the lack of published clinical trials in the scientific literature. Consequently, the Comprehensive Natural Products Database was used to obtain information of the possible effectiveness of various complementary therapies, which are summarised in their own category. NOT EFFECTIVE Progesterone - Progesterone has been widely publicized in the lay literature as a treatment for PMS.4 ProgesteroneÕs role in the treatment of PMS probably arose from the theory that the syndrome is caused from a lack of progesterone which was popular back in the 1950s up until the 1980s. Treatment with high doses of "natural" progesterone vaginally became popular in the 1970s after the publication of a large number of case reports in the lay press, none of which had any true control groups.4,5 Since then, several randomised-controlled trials have failed to show any benefit from topical or oral micronized progesterone over placebo.6-10 Topical progesterone preparations are also expensive. Given the lack of efficacy and the expense of the product, Progesterone can not be recommended as a treatment of PMS. Pyridoxine (vitamin B6) - Pyridoxine or vitamin B6 is the most widely used supplement used to treat PMS.4 It has been proposed that vitamin B6 may help to correct a "deficiency" in the hypothalamic pituitary axis.4 Vitamin B6 is a cofactor in the synthesis of tryptophan and tyrosine, which are the precursors of serotonin and dopamine respectively.11 Theoretically, low levels of vitamin B6 may lead to high levels of prolactin which in turn could underlay the edema and psychological symptoms associated with PMS.12 The Cochrane Library data base collaboration is currently in its final phases of publishing the most comprehensive review to date on the efficacy of vitamin B6 in the treatment of pre-menstrual syndrome.13 The heterogeneous nature of the studies precluded the performance of a metaanalysis, consequently the studies were analyzed individually.13 They reviewed 16 trials in total. Six of the trials found some positive effects from using vitamin B6, however, 10 trials showed no difference between treatment and control groups. The quality of the trials showing positive effects were judged to be of questionable quality according to the JadadÕs quality scale.13 Consequently it would appear that there is very limited evidencve to support the generalized use of vitamin B6 for the treatment of PMS. Vitamin B6 can also cause significant toxicity and unpleasant side effects. It can produce a progressive sensory ataxia taken at doses as low as 500 mg. a day and can also cause a number of gastrointestinal side effects, particularly nausea.12-14 Consequently, given the lack of clear scientific evidence for its effectiveness, and the associated risks of treatment, vitamin B6 can not generally be recommended as a treatment for PMS. Bromocriptine - Another theory that was popular in the 1970s was that PMS was caused by increased levels of, or an increased sensitivity to, Prolactin.4 Consequently Bromocriptine was commonly prescribed to women suffering from PMS, particularly those with significant fluid retention and breast tenderness. An extensive review by Andersch et. al., which analysed 14 studies, found no improvement in general PMS symptoms compared to placebo.15 One exception is severe cyclical mastalgia for which Bromocriptine may be effective.15 Bromocriptine is also expensive and has a number of side effects. Consequently its use can not be recommended for the general treatment of PMS. Combination Oral contraceptives (OCPs) - Combination oral contraceptives are also widely used to treat PMS. Despite their popularity, there has only been one randomised-controlled trial to date that has compared an OCP (triphasic) to placebo. This trial found some improvement in physical symptoms but not mood changes.(16) There are also some descriptive studies that suggest that women with PMS are more frequently intolerant of OCPs because of mood swings.17,18 Consequently, given the lack of scientific evidence for their effectiveness along with the associated expense and potential risks, OCPs can not be recommended for the treatment of PMS. POSSIBLY EFFECTIVE Diet - Dietary recommendations are commonly recommended to help alleviate the physical and psychological symptoms of PMS. The most common dietary recommendations are to restrict sugar and increase the consumption of complex carbohydrates.4 Of these measures, only an increase in carbohydrate consumption has been studied in a randomised controlled trial. One small trial involving 24 women found that women who consumed a carbohydrate rich beverage daily during the late luteal phase reported less mood changes in the hours following consumption, than women who consumed an isocaloric beverage.19 This beverage was specially formulated to boost tryptophan levels. Nonetheless, encouraging women to eat a diet rich in complex carbohydrates during the latter half of their cycle may help alleviate some of the psychological symptoms of PMS. Aerobic exercise - Women who have PMS are often encouraged to increase their activity level. It has been hypothesised that exercise; particularly aerobic varieties increase endorphin levels, which in turn improves mood.4 There are a number of descriptive studies that do indicate that women who exercise on a regular basis have less PMS symptoms than women who do not exercise regularly.(20) Similarly, one study has shown that women who increased their activity levels reported a decrease in PMS symptoms in previously sedentary women as well as women who already exercised regularly.21 Finally, one randomised controlled non blinded trial involving 23 women, did find that women who were randomised to an aerobic exercise group did report less PMS symptoms than women who were in a non aerobic exercise group after 3 cycles.22 Given the known physical and psychological benefits of exercise, it would seem reasonable to recommend an aerobic exercise program to alleviate PMS symptoms. Psychological approaches - Epidemiological evidence does suggest that increased stress levels do aggravate PMS symptoms.23 A number of trials do suggest that relaxation and cognitive therapies do help alleviate PMS symptoms. In one trial woman were randomised to a group instructed to practice a relaxation technique for 20 minutes a day, or to a 20 minutes of "quiet time" group. The women in the relaxation response group reported less PMS symptoms than the women in the "quiet time" group.24 In another study women who reported "severe" PMS symptoms were randomised to a cognitive behavioural coping skills, "non specific behavioural" and waiting list group. Women who were in the cognitive behavioural group reported significantly less PMS symptoms than the women in the other groups immediately following treatment and at nine months follow up.25 In another study both cognitive behavioural and information focused therapy led to significant reductions in symptoms.26 While the above studies involved relatively small numbers and often lacked "true" control groups; their cumulative results do suggest that various psychological approaches including instruction on relaxation techniques, cognitive behavioural strategies and information giving may all help relieve PMS symptoms. Magnesium - Studies have found that women who suffer from PMS have lower levels of erythrocyte and monocellular magnesium during their menstrual cycles than women who do not have PMS.27 Accordingly, magnesium supplementation has been used as a potential therapy. There are a small number of randomised controlled trials that do indicate that magnesium may in fact be helpful. Facchinetti et. al, did a randomised controlled cross over study using 28 women diagnosed with "severe" PMS.(28) He found that women who received 360 mg. daily (divided dose) during the luteal phase reported less physical and emotional symptoms than women who received placebo. In another study involving 54 subjects, women who received 200 mg. of magnesium daily reported less fluid retention than women receiving placebo.29 They did not have any improvement in mood, cramping or food cravings however. Interestingly it took two cycles before the women noted any benefits from the therapy. While the sizes and numbers of these trials are small they do support a possible benefit from taking magnesium, particularly for women who have significant fluid retention. Magnesium is considered safe at doses up to 483 mg. per day in healthy adults. It must be used with caution, however, in people with significant heart and renal disease.30 Evening Primrose Oil (EPO) - Evening Primrose Oil is used extensively to alleviate PMS symptoms. EPO contains two essential fatty acids: linoleic and gamma linoleic acids. It has been hypothesised that women with PMS are deficient in gamma linoleic acid which is necessary for prostaglandin formation.31 There is a wide body of literature on this topic, however, the quality and methodological limitations of the studies has made it difficult to draw any firm conclusions about the efficacy of EPO.4 Buderi et. al. conducted a review of seven controlled clinical trials done up until 1995.32 Only two of these trials were judged to be of high methodological quality. These trials showed at best marginal benefits for EPO. The Cochrane data base group is also in the final phase of their review on the subject which will be the most comprehensive review of the topic to date.33 Their preliminary findings suggest that EPO may be of some benefit for cyclical mastalgia but little else.34 Evening Primrose oil is generally well tolerated but occasionally it can produce nausea, dyspepsia, and headache. Long-term use may be associated with an increased risk of inflammation, thrombosis, and immunosupression.33 Finally, EPO is relatively expensive. In summary, EPO may be of some benefit to those women with cyclical mastalgia but is probably of limited if any benefit to women who have significant mood and cognitive symptoms. Vitamin E - Vitamin E has been used to treat PMS and general breast tenderness. There have been only a few studies that have addressed this issue. London et. al. did a RCT involving 46 women who were randomised to receive 400 mg. of vitamin E or placebo daily for three cycles.35 Five of the women dropped out. The treatment group did report a reduction in various mood and physical symptoms, which ranged from 28 to 42%. These results did not reach statistical significance however, which is probably not surprising given the size of the study. Chuong et. al. did not find any differences in vitamin E levels in women diagnosed with severe PMS compared to controls.36 Spironolactone - Diuretics have been used to treat the fluid retention associated with PMS for over 50 years. Despite their wide spread use, there is no evidence that the thiazide diuretics are of any benefit.4 These medications are also associated with significant side effects including hypokalemia, secondary aldosteronism and cyclical edema.4 Consequently they can not be recommended for the treatment of PMS. There are some studies that do indicate that Spironolactone may be of some benefit in alleviating the physical and affective symptoms of PMS. To date there have been three small randomised controlled trials. Two have found that Spironolactone given daily (100 mg./day) does result in improvements in irritability, depression, breast swelling and food cravings.36,37 Another trial found that Spironolactone given during the luteal phase also reduced swelling but did not have an impact on psychological symptoms.38 Consequently, the evidence to date does suggest that Spironolactone may be of some benefit to women with PMS although larger studies are needed to confirm this. Non Steroidal Anti-inflammatories (NSAIDS) - There is some evidence that NSAIDS given during the luteal phase does help relieve the physical and affective symptoms of PMS. Mefenamic acid (500 mg. T.I.D.) has been studied in two RCTs involving 15 and 37 women, and has been found to reduce the incidence of headache, fatigue and generalised pain as well as tension, irritability and mood swings.39,40 Naproxen has also been found to produce similar results when used at a dose of 550 mg. B.I.D.41 In all of these studies the NSAIDS were started one week before the onset of menses and continued during the first few days of bleeding. Again, these results come from relatively small studies. Nonetheless, the consistency of the findings does suggest that NSAIDS may help to relieve the affective and physical symptoms of PMS when administered during the luteal phase of the cycle. Ovulation Suppression - The use of Danazol and Gonadotrophin Releasing Hormone Agonists to suppress ovulation have been shown to reduce the symptoms of PMS.4 The significant side effects associated with these treatments however, makes them generally unacceptable for use in Primary Care. Some small studies do indicate that ovulation suppression with Estradiol and Progestin is also effective. In one double blind cross over study (N = 48) women were given 15 mg. of oral medroxyprogesterone, another progesterone norethisterone, or placebo.42 Women reported significant improvements in their psychological and physical symptoms when taking the medroxyprogesterone. Women also reported less bloating on the norethisterone but did not have any decrease in affective symptoms. Unfortunately there was a 74% breakthrough-bleeding rate in the progestin group, which could significantly influence patient acceptability. Watson et. al did a randomised controlled clinical comparing 0.2 mg. Estraderm patch and cyclical progestin with placebo.43 Women in the study group reported significant reduction in affective and physical symptoms. A subsequent study showed that 0.1 mg. dose was as effective as the 0.2 mg.44 These studies do indicate that hormonal treatment with some forms of estrogens and progesterone may be of benefit in the treatment of PMS. Nonetheless, given the potential side effects and risks involved, this treatment should probably be reserved for women for whom other measures are ineffective or contraindicated. Hormonal measures may be of particular benefit however, to women in the perimenopausal age group who are contemplating HRT. It is important to appreciate that the synthetic hormones vary in their chemical composition and effects from each other and the natural products. Consequently differences in chemical compositions, even relatively subtle ones, may underly the differences in response to various hormonal treatments including hormonal regimes that have been found to be effective and the OCPs and natural progesterone which have not been found to be effective. EFFECTIVE Calcium - Thys-Jacobs et. al. of the Pre-menstrual Syndrome Working group conducted a large multicentred trial at 12 sites involving 466 women diagnosed with moderate to severe PMS.45 Women were randomised to a calcium carbonate (1200 mg/day) or placebo group. Women recorded their symptoms daily over three cycles. Compliance with treatment was also measured. The main outcome measure was a 17-parameter complex score, which included 17 core symptoms and four symptom factors including negative affect, food cravings, pain, and water retention. There was no significant reduction in symptoms reported after the first cycle. However by the third cycle women reported a 48% reduction in their total symptom scores compared to baseline, compared to a 30% reduction in the placebo. In addition all four-symptom factors were significantly reduced by the third cycle. Given the studyÕs sound methodology, large size, and size of the treatment effect, these findings provide good evidence for the effectiveness of calcium carbonate as a treatment for PMS. Calcium is also relatively inexpensive and plays an important role in the prevention of osteoporosis, therefore it is recommended for the treatment of PMS. Selective Serotonin Reuptake Inhibitors (SSRIs) - Another significant advance in the treatment of PMS is the use of the SSRIs. PMS has been linked with dysfunctional serotonin metabolism and there is experimental evidence that hormonal fluctuations do affect central serotonin levels.46 Dimmock et. al. have recently published a comprehensive systematic review on the topic including a metaanalysis involving 15 randomised controlled trials.47 The results of the metaanalysis strongly support the effectiveness of SSRIs in the treatment of PMS. Interestingly, the study group also found no difference in the effectiveness of continuous compared to intermittent therapy during the luteal phase. The effectiveness of SSRIs administered on an intermittent basis has been attributed to differences in receptor sites involved in affective and PMS disorders. The doses used for PMS also tend to be lower than that used for depression. Consequently the incidence of side effects tend to be lower as well.48 The use of the SSRIs is not with out its drawbacks. A host of side effects have been reported including headache, nervousness, insomnia, drowsiness, fatigue, sexual dysfunction and gastrointestinal complaints. The SSRIs are also relatively expensive especially the brand name formulations. Nonetheless given their proven efficacy, they are recommended, particularly for women with severe affective symptoms for whom other measures have not been effective. NATURAL THERAPIES Following is a description of some of the more commonly used herbal preparations used to treat PMS. Our current knowledge about these substances is largely based on pharmacological and descriptive data, which significantly limits our ability to draw conclusions about their effectiveness and long term safety. Black Cohosh - This herbal remedy is derived from the rhizome and root of the plant. Its action is related to the binding of estrogens receptors and suppression of leutinizing hormone although it is not thought to increase the risk for endometrial and breast cancers. It has been rated as "possibly effective" for the treatment of pre-menstrual discomfort. It is likely safe when taken in low doses (0.3 to 2 mg. T.I.D.) for less than six months. Black Cohosh also contains Salicylic acid and consequently should not be taken by people who should avoid aspirin or who are at risk of bleeding. Similarly, it should be avoided in women in whom estrogen is contraindicated. Overdose of Black Cohosh can cause nausea, vomiting, dizziness, visual disturbance, and decreased heart and respiration rates.30 Borage Seed oil - Borage seed oil contains 26% gamma linoleic acid and is used as a replacement for evening primrose oil. It is "likely safe" if used orally as directed. Gamma linoleic acid can prolong bleeding time and therefore should be used with caution in people at risk of serious bleeding including those who are taking other medications and herbal products that can prolong bleeding times.30 Dandelion - Dandelion is used for a variety of medicinal purposes. It has been shown to have mild diuretic and anti-inflammatory properties in animal studies. It has been rated as "possibly effective" for promoting diuresis and may be of some benefit in treating the fluid retention associated with PMS. Theoretically dandelion can have hypoglycemic effects and therefore should be used with caution in individuals taking diabetic medications. Individuals who have environmental allergies to members of the Asteracae family, which includes ragweed, chrysanthemums, marigolds and daisies, should also avoid this herb.30 Dong Quai - Dong Quai is a commonly used herb used for a variety of gynecological symptoms including PMS. It contains a number of different constituents, which are thought to have vasodilating, antispasmodic, and anti platelet activities. Dong Quai does have carcinogenic and mutagenic properties and can cause severe photodermatits especially when used in large amounts. It is rated as "possibly unsafe" by the Natural Medicine Comprehensive Database. It may also interact with several medications and other herbal remedies.30 RECOMMENDATIONS How do we organise the above information into a practical concise set of guidelines for Family Physicians? The following recommendations are based on the author's interpretation of the strength of evidence for effectiveness of the various therapies, as well as the potential costs, adverse effects and long term risks involved. The nature of the symptoms was also taken into account. Johnson describes a similar but not identical approach in her very comprehensive review article on the subject.4 MILD/MODERATE PMS FIRST LINE TREATMENTS:
If not effective after three cycles consider adding a second line treatment. SECOND LINE TREATMENTS:
SPIRONOLACTONE - 100 mg. O.D daily MAGNESIUM - 360 mg./day
NSAIDs - Mefenamic acid 500 mg. T.I.D.or Naproxen sodium 550 mg.B.I.D. starting one week before menses starts; to continue for the first few days of bleeding.
HORMONAL TREATMENT - Methyprogesterone acetate - 15 mg. daily or Estraderm patch (0.1 or 0.2 micrograms x 2 per week) plus cyclical medroxyprogesterone acetate 5 mg. from day 17 to 26.
SSRIs - Fluoxetine (20 mg. O.D.). Sertraline (50 mg. O.D.), Paroxetine (20mg. O.D.) or Fluvoxamine - Two month trial during luteal phase only. Switch to daily administration if response inadequate.
EPO - 500 mg. three to four times daily
VITAMIN E - 400 I.U. daily SEVERE PMS
CONCLUSIONS In closing, the management of PMS requires more than just scientific knowledge of effective treatment strategies. It also tests the breadth of the family physician's patient centred clinical and communication skills. Identification of the problem requires a systematic approach to screening and diagnosis. Effective communication is required to educate patients about the nature of the condition, determine the patient's ideas about their symptoms and assess the impact of symptoms on the quality of life. The Family physician must also be able to negotiate a treatment plan that is acceptable to the patient, particularly where lifestyle changes are concerned. Finally, continuity of care is essential for followup, especially where it may take months before any positive benefits are apparent. Despite the challenges, Family Physicians are still in the best position to help women alleviate their symptoms. A patient centred approach along side the principles of evidence based medicine can provide Family Physicians with the tools they need to meet this challenge. Thanks to Dr. Don Wescott, Department of Obstetrics and Gyncology at St. Martha's Regional Hospital in Antigonish, Nova Scotia for reviewing the draft copy of this article. References:
You can search for abstracts of the above references by following this link: PubMed Return to Archives Page ] [ Berries Home Page |