Abstract
‘The diagnosis of premenstural syndrome continues to lack widely accepted criteria and remains controversial as a medical disorder’
There have been more studies of premenstrual syndromes (PMS) in the past 26 years than in all preceding years combined. Much of this research has focused on treatment and contributed to greater recognition and more effective therapy for the disorder. However, the diagnosis of PMS continues to lack widely accepted criteria and remains controversial as a medical disorder, in spite of the many women who seek treatment for the problem.
Surveys indicate that PMS is among the most common health problems reported by women of reproductive age. Current estimates of the prevalence of severe PMS range from 12.6 to 31% [1], while the severe, dysphoric form termed premenstural dysphoric disorder (PMDD) is limited to 5–8% of women [2]. Wittchen and colleagues used PMDD criteria to define cases in population-based data and found that in addition to the 5–8% who met criteria for PMDD, another 18% were one symptom short of the five symptoms required for the diagnosis [2]. This higher estimate of 23–26% is consistent with the numbers of women who believe they need treatment for PMS, as reported in other survey studies.
PMS is characterized by mood, behavioral and physical symptoms that occur several days to 2 weeks before menses and remit with the menstrual flow. The term PMS as used by both clinicians and the general public is generic, imprecise and commonly applied to numerous symptoms that range from the mild and normal physiological changes of the menstrual cycle to clinically significant symptoms that limit or impair normal functioning. While other terms have been employed to indicate a clinical syndrome, such as premenstrual tension, premenstrual dysphoria, cycle-related symptoms, menstrually related symptoms, late luteal phase dysphoric disorder and PMDD, none have been universally accepted.
The absence of a widely accepted diagnosis for PMS has significant economic costs resulting from the increased use of healthcare. Medical costs for outpatient visits to clinicians and the emergency room, laboratory testing, and treatment are higher for PMS patients than for women without the disorder [3]. These costs are, in part, due to poor recognition or diagnosis of the disorder, the resulting use of ineffective treatments that do not control the symptoms, and the limitations of effective treatments that do not help all women with PMS.
In spite of the evidence that the PMDD diagnosis is limited to 5–8% of menstruating women and excludes a sizeable number who seek treatment for PMS, there is no consensus for the diagnostic criteria of PMS [4]. The controversies partly result from focusing on the myriad of symptoms observed in the premenstrual phase, while disregarding the wide range in severity of the symptoms. The absence of systematic criteria for the diagnosis of PMS maintains acceptance of the observations of more than 200 premenstrual complaints associated with the disorder, and results in a PMS population so heterogeneous that it hinders scientific advances in effective treatments and understanding of the pathophysiology of pure PMS.
‘The controversies partly result from focusing on the myriad of symptoms observed in the premenstrual phase while disregarding the wide range in severity of the symptoms.’
A further problem with PMS diagnosis is that physiological abnormalities have not been identified. Numerous studies have reported that there is neither an excess, nor a deficiency in the reproductive hormones of PMS patients, and the etiology of the disorder remains unknown.
There are several diagnostic approaches to PMS at this time, but none are universally accepted. The WHO International Classification of Diseases (ICD-10) lists premenstrual tension syndrome (PMS or PMTS) as a gynecological disorder [5]. The ICD-10 classification for premenstrual symptoms requires no specific symptoms, no specific number of symptoms, no specified level of severity, and no differential diagnosis. While this classification is intended to address premenstrual tension/anxiety, numerous other symptoms that arise in the premenstruum and remit postmenstrually could be included.
The most extensive attempt to formulate diagnostic criteria for the disorder was undertaken by the American Psychiatric Association. The resulting criteria identify a severe, dysphoric form of PMS and were included in the section for further research in the Fourth Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) under the term PMDD [6]. The PMDD criteria require five of 11 specified symptoms at a severe level in the premenstrual phase of the cycle, symptom remission in the follicular phase of the cycle, marked functional impairment, absence of other diagnoses that would account for the symptoms and prospective confirmation of the symptoms for at least two consecutive menstrual cycles.
The PMDD criteria have been used by many researchers in recent years and have been important in defining samples for clinical trials. However, the European Committee for Proprietary Medicinal Products determined that “PMDD is not a well-established disease entity across Europe,” highlighting the lack of consensus for this diagnosis.
‘Accumulating evidence, primarily from clinical trials, suggests that PMS differs from depressive disorders and is a distinct diagnosis.’
A further problem with the PMDD diagnosis is that only about 5% of women who seek treatment meet the PMDD criteria [2]. PMDD emphasizes dysphoric symptoms and excludes women with fewer than five of the required symptoms, who may nonetheless experience severe distress or impairment in their daily functioning. Furthermore, in primary care, where most women seek treatment for PMS, few clinicians employ the PMDD requirement for prospective daily symptom ratings to confirm the reported symptoms and cycle-related pattern.
A third diagnostic approach was recently offered by the American College of Obstetricians and Gynecologists (ACOG) [7]. The ACOG criteria for PMS require one of ten specified symptoms during the 5 days before menses, remission during the menstrual flow without recurrence of the symptoms until at least cycle day 13, identifiable impairment or distress, absence of other diagnoses that would account for the symptom(s) and prospective confirmation of the symptom(s) for three menstrual cycles. These criteria define specific cycle days for the symptoms, but if literally observed, the cycle days may be too limiting. More importantly, the extent to which these criteria appropriately diagnose PMS in women who seek treatment has not been demonstrated.
There is a large overlap between clinically significant PMS and PMDD. At this time, these diagnoses are differentiated primarily by the PMDD diagnosis, which requires specified symptoms and a specified number of symptoms. Those not meeting the PMDD criteria due to fewer symptoms or other symptoms may be termed PMS.
Neither PMDD nor the ACOG guidelines for PMS provide definitions for quantitatively operationalizing their diagnostic criteria. As a result, study samples continue to differ with respect to the predominant symptoms and the required severity level of the symptoms. Other factors in the heterogeneous PMS population, such as age or history of depression, also differ between and within study samples. These differences in study samples contribute to conflicting results that are particularly evident in studies of hormones or underlying mechanisms of the disorder.
A dominant perspective in PMS studies is the assumption that PMS is a variant of depression. However, accumulating evidence, primarily from clinical trials, suggests that PMS differs from depressive disorders and is a distinct diagnosis:
The onset of response to selective serotonin reuptake inhibitors (SSRIs) is rapid, within hours or a day or two as reported by Eriksson [8], and as demonstrated by the efficacy of luteal-phase dosing, rather than several weeks that may be required for depression;
The maximal response of PMS patients to SSRIs is at low doses, with little or no additional improvement at the high end of the dose ranges and may not be due to the inhibition of serotonin (5-HT or 5-hydroxytryptamine) reuptake that is believed to improve major depressive disorder [9];
The cardinal PMS symptom is irritability, not depression, and the syndrome includes physical symptoms such as breast soreness and swelling, which are not characteristic of depression but respond to SSRI treatment for PMS [10];
Neither a history of depression nor depressive symptoms account for the response to SSRIs in PMS patients [11]; conversely, women with comorbid depressive disorders do not respond to treatment with a gonadotropin-releasing hormone (GnRH) agonist, which is very effective for PMS [12];
PMS does not respond to tricyclic antidepressants that have less selective serotonergic activity but are nonetheless very effective for depressive disorders;
PMS appears to be a chronic disorder that does not reach remission without treatment [2]. Symptoms may return swiftly, within months after discontinuation of medication in small pilot studies and in clinical reports, indicating the chronic nature of the disorder.
The numerous symptoms that have traditionally been attributed to PMS contribute to the controversy regarding its diagnosis. The plethora of over 200 premenstrual complaints partly results from the failure to distinguish PMS from other comorbid conditions. Many disorders, both physical and psychiatric, are exacerbated premenstrually or occur as a comorbid disorder with PMS [13].
When a careful diagnosis is made to distinguish PMS from other conditions, a much smaller group of symptoms appear to be typical of the disorder. Irritability, anxiety, abdominal bloating or swelling, food cravings and depressed mood are predominant in numerous PMS studies. However, only the timing of the symptoms in the menstrual cycle has been reliably demonstrated, and whether a small number of core symptoms can diagnose PMS remains unanswered. In addition to the almost limitless number of symptoms linked with PMS, there is no consistent evidence that any of the symptoms occur only in the luteal phase. In a current review of the gonadal steroid regulation of mood, Rubinow proposed that the focus be changed from “the futility of cataloging symptoms for PMS” to addressing disturbances in affective information processing and regulation [14].
Another controversy emerges from the clinical observation that PMS worsens with age, inasmuch as women who seek treatment are primarily in their late twenties or thirties. It is believed that PMS can occur in conjunction with ovulatory cycles, that is, from menarche to menopause. However, there is scant information in the literature on PMS for either adolescents or women in the late reproductive years, when symptoms associated with menopause are not limited to the luteal phase as occurs with PMS.
A commonly accepted belief in medicine is that PMS is unusual in adolescents, who frequently suffer from dysmenorrhea but do not have PMS. In contrast to clinical beliefs, epidemiological data suggest that PMS is not rare in the adolescent population. Among females aged 14–24 years, the prevalence of threshold and subthreshold PMDD was 20.4% at baseline and increased to 31.0% 2.5 years later [15]. In addition to the likelihood that PMS is not differentiated from dysmenorrhea by either clinicians or young women, evidence indicates that adolescents may not apply the term PMS to their own symptoms and may have no knowledge of treatments for the disorder. The lack of differentiation between dysmenorrhea and PMS in young women may compromise effective treatment for these disorders.
‘In contrast to clinical beliefs, epidemiological data suggest that PMS is not rare in the adolescent population.’
At the older end of the age spectrum, PMS may decrease as ovulation becomes less regular and women enter the transition to menopause. Women reported that PMS decreased when they were in the menopausal transition, which was defined by changes in cycle length [16,17]. However, these women were also more likely to experience menopausal symptoms such as hot flashes, depressed mood or poor sleep that were not limited to the luteal phase [18]. Attributing a worsening of symptoms with age to PMS rather than to menopause may also compromise effective treatment. A careful diagnosis to determine the timing of the symptoms is necessary in order to select effective treatment approaches.
In conclusion, the absence of a widely accepted diagnosis of PMS reflects the difficulties surrounding a problem characterized by a plethora of nonspecific symptoms, a wide range in the severity of the symptoms, and the likelihood that symptom control is at the brain level and not fully understood at this time. Nonetheless, a widely accepted diagnosis is essential to address the many questions regarding the treatments and causes of PMS that remain unanswered. Efforts are being made to reach a consensus and define criteria for the diagnosis of PMS [4]. Attaining this goal would contribute to better treatment, reduced healthcare costs and improvments in the quality of life of women who suffer from this disorder.