Premenstrual syndrome, a common but underrated entity: review of the clinical literature

The role of genetic factors in the predisposition to PMS and PMDD has been an active and interesting area for many researchers yet definitive conclusions have not emerged. Some studies suggest a possible association with estrogen receptor alpha (ESR1) gene ( 13 , 14 , 15 ). In one report, cells from women with and without PMDD appeared to show different response patterns to the components of the ESC/E (Z) complex containing the ESR1 gene ( 16 ). Other risk factors for PMDD development include lower levels of education and smoking ( 17 ), history of traumatic events or anxiety disorder, and higher daily difficulty scores ( 18 ).

Premenstrual syndrome (PMS) is an entity characterized by the presence of psychiatric symptoms such as mood swings, depression, loss of confidence, anxiety and irritability, without any underlying psychiatric disorder, accompanied by physical symptoms. Typical complaints include bloatedness and mastalgia encountered at the luteal phase of the menstrual cycle (LPMC), that deteriorates the well-being of the women and then subsides or disappears with menstruation ( 1 ). PMS affects a huge proportion of women at reproductive age and is characterized by the cyclic recurrence of a range of symptoms shown in Table 1 during the LPMC ( 2 , 3 , 4 ). Symptoms occur mostly in women of 25-35 years old, although it may be observed at any age between adolescence and menopause. Premenstrual dysphoric disorder (PMDD) is at the more severe end of the PMS spectrum that is characterized by the cyclic recurrence of psychological manifestations including irritability, nervousness, agitation, anger, insomnia, difficulty in concentrating, severe fatigue, depression, anxiety, and confusion. In addition, neurologic and vascular conditions are present which may include headache, dizziness, numbness, heightened sensitivity of arms and/or legs, palpitations, gastrointestinal and ocular symptoms that disrupt the daily life and functioning of the affected women. The mood disorder symptoms that are experienced both in PMS and PMDD disappear within the first days of menstruation. The Diagnostic and Statistical Manual of Mental Disorders-5 (DSM-5) established seven criteria (A through G) for the diagnosis of PMDD and at least five of these symptoms should be present, and should include one of the first four (A-D) ( 5 ). Both PMS and PMDD show a cyclic pattern with affective, behavioral and somatic symptoms beginning at the LPMC and disappearing within a few days after the onset of menstruation. These cyclic symptoms can also be observed in amenorrheic, reproductive aged women who had a hysterectomy but have functioning ovaries.

A thorough evaluation of patients with PMS or PMDD is required ( Table 3 ). The cyclic occurence of symptoms should be confirmed by careful questioning. It might be difficult to interpret in women who have PMS or PMDD with irregular menses. The etiology of the irregular menses should be investigated. In women who are using combined oral contraceptives the timing of the initiation of the symptoms and the relation to the onset or termination of the combined pill should be queried. Various screening tools or applications can be used for self-reporting the symptoms. In cases where the self-reported dairy is not conclusive, the RCOG recommends the use of GnRH testing. PMS can be diagnosed if the symptoms subside when ovarian hormonal suppression is obtained with GnRH analogues.

As medical treatment of PMDD is usually successful, surgery [bilateral oophorectomy/bilateral salpingoophorectomy (surgical menopause)] is considered only in a very few patients who failed to respond to all the medical therapies described ( 44 ). Before surgery, all the pharmacological treatment modalities, especially administration of GnRH analogues, should be considered ( 80 , 81 ). In young women HRT must be commenced after bilateral oophorectomy ( 44 ).

Although various hypotheses have been put forward, the etiology of PMS and PMDD is not fully understood ( 19 ). The best-known hypothesis is the presumed role of circulating gonadal steroids in the development of PMS symptoms, as suppression of ovulation has a beneficial effect on PMS ( 20 , 21 , 22 ). However, cyclic changes in ovarian steroids do not appear to be the only cause of PMS symptoms, as daily serum progesterone and estrogen concentrations are shown to be similar in women with and without PMS. An early study by Andersch et al. ( 23 ) demonstrated that, apart from prolactin levels, which were lower in the follicular phase of the PMS group, all the sex steroids were present in similar concentrations in both the control and the PMS groups in the LPMC. In two studies no correlation was found between serum progesterone levels and the affective and/or somatic symptoms ( 24 , 25 ). Deficiencies in progesterone, progesterone metabolites (some of which have anxiolytic effects) and progesterone receptor function have also been proposed to be a possible cause of PMS/PMDD. However, as mentioned, serum progesterone concentrations are normal in women who have PMS. In addition, serum concentrations of the progesterone metabolites allopregnanolone and pregnenolone are similar in women with PMS and without PMS ( 25 ). It has been shown that blocking the effect of progesterone in the LPMC with a progesterone receptor antagonist, such as mifepristone, does not alleviate PMS symptoms ( 26 ). Women with PMS may have an abnormal response to normal ovarian hormonal changes, even though serum progesterone and estrogen concentrations are within normal limits ( 21 ). Current evidence suggests that PMS is a disorder triggered by changes in gonadal steroids during the LPMC in susceptible women. This is thought to be due to the interaction between cyclic changes in ovarian steroids and the functioning of central neurotransmitters. One of the most frequently investigated neurotransmitters in PMS pathogenesis is serotonin, but beta-endorphine, gamma-aminobutyric acid (GABA) and the autonomic nervous system are also part of the pathogenesis of PMS. Based on in vitro data and animal studies, there is evidence that cyclic variation in circulating estrogen and progesterone result in marked changes in the opioid ( 27 ), GABA ( 28 ) and serotonin ( 29 ) systems. The potential role of the GABAergic system in PMS has not been extensively investigated previously. The main effect of GABA is to reduce cellular excitability through a chloride system. The hypothesis that suggests a modulatory role of progesterone in the GABAergic system is supported by the improvement observed in PMS symptoms when agents such as benzodiazepine and alprazolam, that increase GABAergic activity, are used ( 30 ). Additionally, GABA-A enhances receptor function and has anxiolytic effects. Low levels of the progesterone metabolite allopregnanolone are shown to produce a similar anxiolytic effect ( 31 ). Current studies highlight the pivotal role of serotonin in the etiology of PMS. In a number of studies, patients with PMS have been shown to have lower levels of whole blood serotonin, platelet serotonin, and imipramine binding and serotonin metabolites in the LPMC ( 32 , 33 , 34 , 35 , 36 ). In cerebrospinal fluid, 5-hydroxyindoleacetic acid has been shown to be present in higher levels compared to the dopamine metabolite homovanillic acid ( 37 ). In a study by Brzezinski et al. ( 38 ) and Menkes et al. ( 39 ), improvement in PMS symptoms was achieved with the serotonin agonist fenfluramine ( 38 ) and aggravated with acute reduction of the serotonin precursor tryptophan ( 39 ). In addition, selective serotonin reuptake inhibitors (SSRI) that increase the serotonin level in brain, such as fluoxetine, are one of the most effective drugs for treatment of PMS. The role of minerals and vitamins in the etiology of PMS is still controversial. In previous studies, vitamin E, vitamin A or vitamin B6 levels were not different in women who have PMS ( 40 , 41 ). Several vitamins and dietary supplements have been studied as therapeutic agents for PMS, including vitamin E, B, vitex agnus castus, calcium, zinc, and magnesium. However, the evidence showing any of these are more effective than placebo is scarce. A Cochrane review that described a protocol for vitex agnus castus use for PMS was withdrawn (Cochrane Database of Systematic Reviews 2018, issue 3. art. no.: CD004632.). Also, although several studies related to iron, magnesium, zinc, zinc to potassium ratio, and calcium are present in the literature, their role in the treatment and etiology of PMS/PMDD is still debatable ( 41 , 42 ).

There is no specific finding on physical examination in women with PMS/PMDD. Physical examination might be useful in ruling out other entities, such as endometriosis.

Women with PMS experience a wide range of cyclic and recurrent emotional, physical, behavioral and cognitive symptoms that begin in the LPMC and resolve shortly after the onset of the menstrual period (follicular phase). The number of symptoms seen in the majority of patients is much more limited ( 47 ). However, core symptoms include emotional symptoms such as depression, irritability and anxiety, and somatic symptoms such as breast pain, bloating and swelling and headache ( Table 1 ). For most women, the types of symptoms are consistent between periods and usually last an average of six days a month ( 48 ). Besides affective symptoms, women with PMDD, also have physical symptoms. Analysis of prospective symptom studies in women with PMDD shows that mood and physical symptoms are usually the most severe (and with functional impairment) within four days before menstruation in the first two to three days ( 49 ). The most common emotional or behavioral symptom in PMS is mood swings. Other non-physical behavioral symptoms include irritability, anxiety/tension, sadness or depressed mood, increased appetite, sensitivity to rejection, and decreased interest in activities ( 47 ). The most common physical symptoms in PMS are abdominal bloating and excessive fatigue followed by breast tenderness, headache, dizziness and flushing ( 47 ). Hot flushes, similar to menopausal hot flushes, that occur before menstruation suggest PMS or PMDD in women who are not in postpartum or menopause ( 50 ). Many symptoms associated with PMS significantly affect quality of life, ranging from a moderate to a severe effect and PMS symptoms have been associated with decreased health-related quality of life ( 51 , 52 ).

CBT has been used previously in the treatment of depression and anxiety disorders in women, but data on its use for PMS/PMDD are limited. While some studies report that it is useful ( 76 , 77 ), others failed to show any statistically significant benefit ( 78 ). In a randomized study covering 174 women with PMDD, women in the internet-based cognitive behavioral intervention group experienced a decrease in symptom severity compared to a waiting list control group ( 76 ). The CBT course may be useful for some women, but data on this are limited and qualified health-service providers are required for delivering this treatment.

For women with moderate to severe symptoms seeking hormonal contraception, treatment with a COC is recommended, as these medications suppress the hypothalamic-pituitary-ovarian axis and ovulation. Monophasic preparations should be preferred as multiphasic preparations can worsen mood symptoms ( 44 ). COC’s containing drospirenone, especially with a 24 4+ regimen [3 mg drospirenone (DRSP)/20 mcg ethinyl estradiol] with a shortened drug-free interval of four days are effective and approved by the Federal Drugs Agency for management of PMDD ( 71 ). In case of persistence of the PMS/PMDD symptoms and/or presence of intermediate bleeding that does not improve after three month of COC use, the dosage of ethinyl estradiol can be increased and a 21+7 regimen with 3 mg DRSP/30 mcg ethinyl estradiol COC can be commenced. However, the 24/4 regimen is associated with better cycle control ( 72 ). The COC can be switched to an alternative formulation if drospirenone-containing COC fails to improve the symptoms or an SSRI can be added to COC monotherapy in order to improve the treatment results. Continuous use of COC is better than intermittent use. In two randomized studies ( 73 , 74 ) and in a meta-analysis ( 75 ), it was shown that COCs containing 20 mcg ethinyl estradiol/3 mg drospirenone with a four-day confinement interval are more effective than placebo in reducing PMDD symptoms.

Many groups have published diagnostic criteria for premenstrual diseases, including the World Health Organization, American College of Obstetricians and Gynecologists (ACOG), Royal College of Obstetricians and Gynecologists (RCOG), International Society for Premenstrual Disorders (ISPMD) and the American Psychiatric Association (APA; DSM-5) ( 1 ). ACOG describes symptoms consistent with PMS as: 1- The symptoms should be restricted to the LPMC; 2- The symptom pattern should be confirmed by prospective evaluation; 3- symptoms should cause functional impairment; and 4- other diagnoses that could better explain the symptoms should be excluded ( 43 ). The ACOG definition involves the presence of at least one of the six affective symptoms (angry outbursts, depression, anxiety, confusion, irritability and social withdrawal) and one of the four somatic symptoms (abdominal bloating, headache, breast tenderness, and swelling of extremities) reported five days prior to the onset of menses in the three prior menstrual cycles which have ceased within 4 days of onset of menses ( 43 ). ACOG has added more symptoms in the patient-available resources which include emotional symptoms such as depression, irritability, angry outbursts, crying spells, social withdrawal, anxiety, confusion, poor concentration, insomnia, increased nap taking, and changes in sexual desire, while physical symptoms include thirst and appetite changes (food cravings), weight gain, breast tenderness, bloating and headache, swelling of the hands or feet, aches and pains, abdominal pain, fatigue, skin problems, and gastrointestinal symptoms (https://www.acog.org/patient-resources/faqs August 2020). The RCOG recommends keeping a self-reported symptom diary by the women for recording the symptoms for diagnosis of PMS and that the symptoms should be observed prospectively over two cycles ( 44 ). ISPMD published a consensus article on the management of PMD and cited the classification published by O’Brien et al. ( 45 ) in 2011. According to this classification ( 46 ) two entities are described which are: 1) Core premenstrual disorder (CPD); and 2) Variant premenstrual disorders. The symptom characteristics of CPD can be somatic and/or psychological and were stated as occurring in ovulatory cycles, being absent after menstruation, and before ovulation, recurring in the luteal phase and be prospectively rated for at least two cycles. In addition, CPD symptoms should cause significant impairment of daily life (work, school, social activities, hobbies, interpersonal relationships, and distress). PMDD was defined as a sub-group of CPD. The variants of PMD were: a) Menstrual exacerbation of symptoms of an underlying somatic, psychological or medical disorder that significantly worsens premenstrually; b) PMD due to non-ovulatory ovarian activity; c) Progestogen-induced PMD related to exogenous progestogen administration; and d) PMD with absence of menstruation due to ongoing ovarian activity. PMS is currently being diagnosed when any one of the four symptoms that may be physical, behavioral, or emotional/psychological (with at least one being emotional) or physical or behavioral and five or more symptoms are present. However, if a woman has five or more symptoms and one of them is an “emotional symptom”, it would be better to diagnose PMDD instead of PMS ( Table 1 ). PMDD, classified by the APA, should include five of the 11 symptoms required to meet the diagnostic criteria, defined in DSM-5 and at least one of these should involve mood swings ( 5 ). Currently the APA DSM-5 system, that defines PMDD criteria, is used for diagnosis. These criteria specify that: a) prospective documentation of behavioral and physical symptoms (using diaries) for most of the previous year should be available; b) accompanied by five or more symptoms that occur in the week before the onset of menstruation and disappear within a few days of the onset of menstruation. These criteria also indicate that PMDD can overlap with other psychiatric disorders but in these patients the symptoms do not exacerbate in the luteal phase. According to the DSM-5 criteria, one or more of the following must be present for the diagnosis of PMDD: a) anger/irritability b) sudden sadness, increased sensitivity, mood swings to rejection; c) tension and anxiety; and d) depressed mood, feeling hopeless, self-critical thoughts. To achieve a total of five symptoms, one or more of the following symptoms must also be present: 1. Premenstrual irritability, 2. Concentration difficulty, 3. Appetite change, overeating, food cravings, 4. Decreased interest in ordinary activities, 5. Easily fatigued, having reduced energy, 6. Feeling overwhelmed or out of control, 7. Bloating, breast tenderness, weight gain, or joint/muscle pain, 8. Sleeping too much or not getting enough sleep. These symptoms should occur in the luteal phase and must be severe enough to deteriorate daily functions (work, school, social life).

There is no specific biochemical or any other tests to diagnose PMS. Daily gonadotropin and sex steroid serum concentrations are not different from women with and without PMS ( 23 , 24 ). Thyroid tests might be useful in order to exclude hypothyroidism.

Percutaneous estradiol (100 mcg estradiol patches twice weekly) opposed with a cyclical 10-12 day course of oral or vaginal micronized progesterone or long-term progestogen with the LNG-IUS 52 m are recommended regimens for PMS ( 44 ). However, careful monitoring is required. Danazol 200 mg twice daily is also effective in reducing the symptoms, although virilizing side-effects including weight gain, acne, hirsutism and deepening of the voice limits its use ( 44 ). Effective contraception must be provided during Danazol treatment as Danazol may cause virilization of the female fetus.

PMS and PMDD symptoms compromise the well-being of women and have a negative impact on quality of life. Definitive diagnosis is based on prospective self-reporting of the symptoms. Most cases are unrecognized as the presence of the symptoms is not usually questioned during gynecological exams and routine check-ups. Exercise, a healthy diet rich in vitamins and minerals and CBT are the first-line treatment modalities in mild cases. SSRIs and/or COCs are first-line pharmacologic treatments as they are effective in the majority of the women with PMS and PMDD symptoms. Alternative hormonal therapies can be utilized when the standard therapies fail. Surgery is the last resort and should not be considered until all the alternative medical treatment modalities have been tried and are ineffective.

For differential diagnosis of PMS and PMDD the entities given in Table 2 should be ruled out ( Table 2 ) ( 43 , 44 , 46 ).

During the LPMC at least one physical or psychiatric symptom is observed in 80% of the women. However most of them do not report any change in their daily life and activities ( 6 ). In a study of 2800 French women 12% demonstrated the diagnostic symptoms of PMS while only 4% had more severe symptoms ( 7 ). Although most women experience one or a few of these symptoms during the LPMC, only 3-8% demonstrate clinically significant PMS symptoms ( 8 ). Three prospective, population-based studies that utilized the strict criteria for diagnosis of PMDD reported a prevalence of approximately 2% among the study population ( 8 , 9 ). In a study conducted at Switzerland in 2007, a total of 3,913 women aged 15 to 54 responded to a questionnaire inquiring about PMS symptoms and 3,522 (90%) reported that PMS affected their daily life. While 90% had at least one symptom, 10.3% had PMS, and 3.1% met the PMDD criteria ( 10 ). Also, in a meta-analysis covering 18,803 women, the overall prevalence of PMS was 47.8% [95% confidence interval (CI): 32.6-62.9]. The lowest and highest prevalence were reported as 12% (95% CI: 11-13) in France and 98% (95% CI: 97-100) in Iran, respectively. The authors emphasized that the prevalence of PMS was studied more in Asia than in other continents ( 11 ). A global study involving 7,226 women from South America, Europe and Asia, investigating the frequency of PMS symptoms and found the frequency to be parallel between countries and the regions, but in some countries, such as Pakistan, women were found to be less familiar with the term PMS compared to European women ( 12 ).

In women with severe symptoms who cannot respond or tolerate SSRIs or COCs, the next step might be considering GnRH that provides a reversible medical oophorectomy ( 78 ). Although GnRH agonist therapy is effective in PMDD ( 79 ), it cannot be used for a long period due to the menopausal side-effects related to estrogen deficiency, such as bone loss and genital atrophy. Depot leuprolide acetate 3.75 mg per month is the first choice. Add-back therapy with COC is recommended if GnRH analog will be used longer than six months ( 44 ).