• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

The Diagnosis of Premenstrual Syndromes and Premenstrual Dysphoric Disorder - Clinical Procedures and Research Perspectives

Posted on: Tuesday, 22 February 2005, 03:00 CST

Key words: PREMENSTRUAL SYNDROMES, PREMENSTRUAL DYSPHORIC DISORDER, MENSI-RUAL CYCLE, DEPRESSION, DIAGNOSIS, DSM-IV, ICD-IO

ABSTRACT

Premenstrual syndromes (PMS) are quite prevalent among women of reproductive age. In up to 20% of women they are severe enough to warrant treatment, which is available and marketed as such. The impact of the cumulative burden of PMS is substantial and is in the same magnitude as affective disorders. Nevertheless, the definitions and diagnoses ofPMS are still fragmented, not widely accepted and, if accepted, not always applied in day-to-day clinical practice. In the present paper, the current diagnostic entities are critically reviewed, problems with the current definitions are delineated and a unified definition is proposed. For clinical purposes, the recommended clinical practical diagnostic process and differential diagnosis are described. For clinical trials of medications for treatment of PMS/premenstrual dysphoric disorder, research diagnostic criteria, inclusion and exclusion criteria, as well as well-defined outcome measures, are of utmost importance; they are described here. The gaps of knowledge in the description and diagnosis of PMS are described, with suggestions for future directions for research.

INTRODUCTION

It is quite indicative that, despite several decades of research and progress in knowledge on premenstrual syndromes (PMS), acceptance by most lay people in Western countries and the availability of reasonably efficacious treatment modalities, the definitions and diagnostic criteria for PMS are still controversial. We are still far from a biomedical and/or biopsychosocial model of a diagnostic entity based on etiology, pathophysiology, phenomena, time course and treatment response. Even descriptive diagnoses are fragmented, with at least two main pathways: gynecological and psychiatric. Furthermore, there is still a group that doubts the existence of PMS - mostly on political-psychosocial grounds.

Prior to delineation of diagnoses and diagnostic criteria, the clinical relevance of PMS as well as its public health impact should be emphasized. According to the criteria of the American Psychiatric Association's (APA) Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV)' and DSM-IV Text Revision (DSM- IV-TR), the prevalence of the dysphoric premenstrual syndrome, premenstrual dysphoric disorder (PMDD), is widely cited as being 3- 8% among women of reproductive age' " . However, if the clinically relevant impairment, suffering and distress are considered as important, as opposed to the arbitrary cut-off number of five symptoms, the 12-month prevalence of dysphoric PMS is closer to 20% among women of reproductive age (Table 1).

When viewed in the context of the prevalence of other mental disorders , it has been demonstrated that in the USA conservatively 9.1-14 million people suffer from dysphoric PMS with impairment each year, compared with 9.1 million sufferers of a major depressive episode, 1.0 million with bipolar disorder, 4.0 million with generalized anxiety disorder and 4.3 million with obsessive- compulsive disorder. The impact of PMDD on quality of life and social adjustment is similar to that of dysthymic disorder, and is comparable to the impact of major depressive disorder on marital and parental social adjustment as well as work and leisure functioning11,18.

The public health impact of PMDD may be demonstrated by using the global burden of disease (G13D) model and calculating the number of years of healthy life lost to disability, or the disabilityadjusted life years (DALYs) 9. Detailed conservative considerations and calculations have previously been published . The calculated burden of PMDD for the USA as well as the European Union is summarized in Table 2.

Table 1 Recent studies on the prevalence of premenstrual dysphoric disorder (PMDD). Adapted from reference 11

The actual total DALYs are probably much higher, because the calculations in Table 2 are limited to the estimate of 5% of women who would meet strict criteria for PMDD and do not take into account women with premenstrual impairment who have only three or four symptoms or who have physical symptoms only. If global DALYs for PMS would be calculated, the number would be staggering.

Considering the clinical and public health impact of PMS, an interdisciplinary, internationally accepted diagnosis is urgently needed.

CURRENT DIAGNOSTIC ENTITIES OP PREMENSTRUAL SYNDOMES

The World Health Organization's (WHO) International Classification of Disease, 10th edition (ICD10)21 includes premenstrual tension syndrome (PMTS) in its section of gynecologic disorders, as a disorder of the female genital organs. The significance of the ICD-10 diagnosis is that it focuses on the two main aspects of PMS: the association with the menstrual cycle; and cyclicity and timing. It requires that symptoms occur exclusively during the premenstrual period and remit following menses. It de- emphasizes the nature of specific symptoms or phenomena. It only gives examples, such as tension or migraine or any other molimen (any menstrually related symptom). It does not specify any level of required severity or impairment, no degree of change from the nonsymptomatic phase of the menstrual cycle and no differential diagnosis or exclusion criteria (Table 3). The ICD-9 code for PMS is 025.4.

Conceptually similar, but a step forward from the ICD-10 criteria, are the American College of Obstetricians and Gynecologists' (ACOG) diagnostic criteria for PMS22 (Table 3). The ACOG diagnostic criteria require report of at least any one of six affective and four somatic symptoms that should exist during the 5 days before menses at each of three prior menstrual cycles and be relieved within 4 days of onset of menses. They should not exist until the periovulatary phase of the menstrual cycle, be present in the absence of any pharmacologic therapy, be associated with impairment or dysfunction in social or economic performance and be prospectively confirmed during two menstrual cycles.

Table 2 The burden of premenstrual dysphoric disorder (PMDD)/ premenstrual syndromes (PMS) in the United States and the European Union in terms of disability-adjusted life years (DALYs) lost. Adapted from reference 19

The significance of the ACOG criteria is that they acknowledge the importance of dysfunction and impairment, specify the symptomatic and the asymptomatic periods, and require prospective confirmation of retrospective reports. In my opinion, the emphasis on timing with vagueness of the nature of symptoms, and the requirement of at least one symptom without a threshold of a minimal number of symptoms, are conceptual and practical strengths of the ACOG criteria.

The most elaborate diagnostic criteria of a subtype of PMS are the APA DSM-IV criteria for PMDD1,2. This set of criteria requires an initial retrospective report that symptoms, their lutealfbllicular cyclicity, luteal occurrences and fbllicular absence have been present for the majority of cycles during the previous year. At least five out of 11 listed symptoms must be present, at least one of which should be a major mood symptom depression, anxiety, irritability or affective lability. Symptoms should cause impairment and interfere with work, social activities and/or relationships, not be an exacerbation of another chronic disorder and be prospectivcly confirmed by daily ratings during at least two consecutive cycles. Because of varied opinions during the APA's DSM-IV process, PMDD is listed in the DSM-IV and DSM-IV-TR as one of the criteria sets and axes provided for further study, and is designated as 'depressive disorder not otherwise specified' ICD-9 code 311.

Table 3 Current diagnostic criteria for premenstrual syndromes (PMS) /premenstrual dysphoric disorder (PMDD)

The three main diagnostic criteria are summarized in Table 3.

UNSOLVED ISSUES WITH CURRENT DIAGNOSTIC CRITERIA

The main shortcomings of the current diagnostic criteria are the lack of universal agreement on the nature of PMS as well as the lack of universal acceptance of the criteria per se. The closest to universal acceptance is the ICD-10 entity. However, the PMTS description is vague and does not provide a specific diagnostic process and criteria. Actually, it just acknowledges the existence of the entity. Universal acceptance of a diagnostic entity is needed for unified coding and for regulatory purposes. It has financial ramifications for service delivery and reimbursement, as well as for prescriptions and their reimbursement.

The lack of clear acceptable definitions slows the development of drugs and other treatments. Regulatory agencies, like the United States' Food and Drug Administration (FDA) and the European Union's European Medicines Agency (EMEA), require clear clinical indications for pivotal clinical trials and clear projected outcome measures. In the USA, the indication of PMDD is acceptable mainly because it is included in the influential DSM-IV (although in an Appendix) and follows its descriptive principles. PMS, though, is still not acceptable despite the weight of ACOG.

The implied dichotomy between the physical symptoms of PMS and the mental symptoms of PMDD is arbitrary. The association between physical and mental symptoms is still unclear, especially when any of them is sever\e enough to cause dysfunction and/or distress. The repeated and widely presumed reference to PMDlT) as a severe form of PMS is not supported by data, either. It also assumes a continuum of basically same manifestation of an entity along severity gradation and attributes lesser severity to the perception of physical and other non-dysphoric symptoms.

A major unsolved issue is the definition of PMDD as a diagnostic entity, independent of PMS. This may well be correct, but so far data to the contrary are no less convincing. Most catamenial disorders (Table 4) are characterized as an episode whose phenomena and etiology are similar to generally occurring disorders, but the timing of the episode is entrained to the menstrual cycle, mostly due to the hormonal fluctuations. Therefore the menstrually related changes trigger the occurrence of the specific episode but do not influence the basic etiology and pathophysiology or the vulnerability to the parent disorder. Only a few physical symptoms of PMS or catamenial disorders (e.g. breast tenderness) may be attributed to the influence of estrogen or progesterone per se.

Review of the literature (e.g. references 24-29) demonstrates that PMDD may probably be a catamenial disorder like any other catamenial disorder. The phenomena are similar to other depressions and anxiety disorders and there is a statistical association with them. In some cases, treatment interventions and responses are similar.

A vulnerability-trigger conceptualization29-31 of the etiology and pathophysiology of PMS/PMDD suggests that some women are vulnerable to develop specific phenotypes of dysphoric disorders. Vulnerability implies a symptomatic threshold in response to hormonal or situational triggers. In many people symptoms appear even without a currently known trigger and the course of illness depends on currently unknown mechanisms.

Table 4 Catamenial episodes. Adapted and expanded from reference 23

Therefore, the main difference between DSM-IV affective and anxiety disorders and subtypes of PMS/PMDD would be the added vulnerability to menstrually related triggers and probably also an intact or strong normalization process that prevents the episode from extending for a long time beyond expiration of the menstrually related transitory stimulus.

According to this conceptualization, the distinction between PMDD (or PMS) and premenstrual exacerbations (PME) of other mental or physical disorders may not be substantial, especially when the exacerbations are limited to or occur mostly during the premenstrual period. Biologically, PMS and PME may represent stages on a continuum of vulnerability-(threshold)-stimulus interactions.

Even if PMDD is viewed as a separate entity, it presents a plethora of unsolved issues. Despite acceptance of the PMDD entity by the FDA and several PDA-approved medications for that indication, the definitions of PMDD are far from being perfect. The main weakness ofthat diagnostic entity is the requirement of endorsing five of 11 listed symptoms. The reason for choice of the specific numeric threshold is unclear and is not necessarily clinically relevant. A substantive percentage of women report severe impairment but only three or four symptoms16. They seek and certainly warrant treatment no less than the 5% of women who report the highly specific and restrictive five symptoms. Even though a multitude of premenstrual dysphoric symptoms25 has been reported, only 11 are listed. There is no latitude for accounting for other symptoms, even though they may be no less severe than the listed ones for a specific individual. The requirement of impairment of functioning does not consider the possibility of women who manage to continue reasonable day-to-day functioning despite having severe symptoms, but do so with severe distress.

As is the case with most DSM-IV clinical entities, the putatively quantitative measures are actually descriptive and are vaguely defined (e.g. 'most menstrual cycles', 'most of the time', 'within few days', 'markedly' interfere, 'marked' anxiety) and therefore are subject to additional subjective interpretations.

The contribution of inclusion of PMDD in the DSM-IV to recognition of the entity and to improving the treatment of women suffering from PMS should not be underestimated. The media coverage of that disorder as well as industry promotion of PDA-approved medications certainly increased the alertness of patients and clinicians as well as acceptance of the concept, but this does not negate the need for further improvement and refinement of the diagnostic entity.

The ACOG criteria of PMS consider the individual diversity of symptoms and avoid the trap of a threshold of a specific number of symptoms. They follow previous suggestions (and Mortola, unpublished) for the logic of PMS diagnosis but do not carry it to specificity and well-rounded inclusiveness.

DURATION OF THE PREMENSTRUAL PERIODS

The DSM-IV criteria require qualified symptoms to occur during the week before menses, and the ACOG criteria require them to occur within 5 days before menses. Both do not specify for how many days the symptoms should exist, nor does the DSM-IV specify how soon should symptoms remit ('within few days after the onset of follicular phase').

For clinical trials, inclusion criteria usually specify 'severe symptoms for at least 4 days' or 'average severity during 7 days32. When duration of premenstrual symptoms is reported, it is apparent that there is large individual variability in the number of premenstrual symptomatic days. The mean has been reported to be 6.1 days. However, in the USA, the UK and France, 44%, 36% and 1 9%, respectively, reported in retrospective telephone interviews that their PMS lasts no longer than 3 days. Up to 10% of women with PMS report duration of symptoms for the entire luteal phase, and 1-2% report that their symptoms actually include the periovulatory period. Some 12-18% report only one or two symptomatic days. My clinical experience is that some women may have extremely severe PMS symptoms for only 1-2 days, especially those with premenstrual manic or psychotic episodes' . The possibility that different lengths of premenstrual period (PMP) may be associated with different phenotypes and underlying mechanisms has not yet been fully elucidated. This is partially because women with very short PMP do not meet criteria for current clinical trials, as is many times the case with women with very long PMP (Figure 1).

The different time-related patterns of PMP and their differentiation from the chronic non-PMS pattern are demonstrated in Figure 2.

PROPOSED DEFINITION AND DIAGNOSTIC CRITERIA FOR PREMENSTRUAL SYNDROMES

If the diagnosis of PMS/PMDD is to be widely accepted, two complementary conceptual and procedural operational steps should be taken. First, an international authoritative organization should take the lead in formulating and establishing the diagnosis and its criteria. Indeed, the WHO should perform that task and update the ICD-IO definition and criteria of PMTS, as a component of the future ICD-II. Second, it should be recognized that PMS is an interdisciplinary domain. It involves many aspects of endocrinology, gynecology, mental health, clinical neurosciences and several sub- disciplines of internal medicine, as well as social and developmental sciences. It is the domain of all of the above and none of them can claim it exclusively.

It appears that the main aspects of a proposed definition of PMS for wide acceptance are already currently widely accepted:

(1) The entity is distinguished from other similar entities mostly by its timing. Symptoms appear cyclically mostly during the luteal phase of the menstrual cycle and disappear shortly following the beginning of menses. They are temporally entrained to the menstrual cycle.

(2) To be considered a disorder, symptoms should cause impairment and/or dysfunction. I suggest adding 'and/or distress'.

It is also quite widely accepted that individual vulnerability is an important contributor to the development of PMS. Issues involved in vulnerability to PMS have been discussed at length elsewhere25,27,28,30,31. It should be recognized that (as is increasingly suggested for affective disorders in general) there are probably diversified vulnerability traits to PMS27. The diversified vulnerabilities are probably associated with diversified genotypes32, pathophysiological processes (e.g. serotonin, yaminobuytric acid31) and ensuing phenotypes. Once the concept of multiple phenotypes or subtypes of PMS is accepted, this will be important not only for the diagnosis of PMS but also for phenotype- targcted treatment, which is the ultimate purpose of appropriate diagnosis.

Figure 1 Duration of premenstrual symptoms. In telephone interviews, 44% of US, 36% of UK and 39% of French women reported duration of ≤ 3 days (adapted from reference 20)

The main arguments in support of diversified vulnerabilities to PMS arc:

(1) Over 200 symptoms have been reported as appearing premenstrually, they involve many body systems and none of them is exclusively related to the menstrual cycle per se24,32,33.

(2) The nature and clusters of symptoms are quite consistent from cycle to cycle within individual women, although severity may fluctuate33.

(3) Individual women may tend to present with similar symptoms at other periods of physical, psychosocial or hormonal stress.

(4) The rate of treatment response to any currently approved medication for PMDD (currently only selective serotonin reuptake inhibitors) is barely 20% better than placebo35. One may suggest that only a subgroup of women with PMS respond to these medications. Other subgroups with other vulnerabilities may respond to other treatment modalities36.

The suggested diversified genotypes and phenotypes lead to a need de-emphasize the descriptive approach to PMS and to replace it with a generalized approach that has beenalready partly adopted by ACOG. Accordingly, the following diagnostic criteria for PMS have been proposed29,37:

Figure 2 Premenstrual syndrome (PMS) and non-PMS fluctuations of symptoms

(1) Any mood, behavior or physical symptom(s) or cluster(s) of symptoms that recurrently and cyclically occur during the luteal phase of the menstrual cycle;

(2) The symptom(s) remit(s) shortly following the beginning of menses and consistently do not exist for at least one week of the follicular phase of most menstrual cycles;

(3) The symptom(s) cause emotional or physical distress, suffering or impairment in daily functioning;

(4) The recurrence, cyclicity and timing in the cycle and severity of symptoms, as well as the existence of a menstrually related symptomfree period, are documented by daily monitoring and/ or reports.

Whether or not exclusively premenstrually repeated episodes of any disorder may be considered as PMS or catamenial disorder is a matter of definition. This may be addressed as part of the differential diagnosis of PMS.

THE CLINICAL DIAGNOSTIC PROCESS

Over 20 years ago it had already been documented that about 50% of women who complain of PMS do not confirm their retrospective reports with prospective daily ratings38. Therefore prospective confirmation of cyclicity and seventy is commonly advocated. As is the case with many diagnostic procedures, there is a difference between the elaborate requirements of research procedures and clinical practices.

The clinician in a busy outpatient clinic is under pressure to arrive at a diagnosis, code a reimbursable entity and dispense a prescription within a very brief office visit. In addition, the woman patient usually wants an immediate treatment decision. Data that accurately reflect the diagnostic process and decision-making of PMS in clinical practice are, to my knowledge, lacking. There is indirect information on the outcome of the process in terms of statistics on patterns of prescriptions of drugs for PMDD as well as its coding.

Based on the vast literature on failure of structured interviews for DSM-IV diagnoses to be used by primary-care physicians (reviewed in reference 39), which require a one-time relatively short interview, it is reasonable to assume that the diagnostic prospective requirements for PMDD which are lengthier and more complicated - are not being followed in clinical practice, either.

Therefore there is a need for a clinically relevant practical diagnostic process that will focus on the unique distinction of PMS: menstrual-cycle-entrained repeated on and off cyclicity .

The first step in the diagnostic process is the reasonably applicable prospective confirmation of retrospective reports. This may be performed in two partially complementary ways. When a patient calls for an appointment for complaints of PMS, the date of her last menses should be inquired and her appointment scheduled for the next mid-fbllicular phase - provided that she would have late luteal and perimenstrual periods in between. A short daily rating form should be mailed to her, with verbal and written instructions. She should bring it with her to the office visit. In this way the clinician will first see the patient in a presumably non-symptomatic phase of the menstrual cycle and have information on the most recent symptomatic period. If practically feasible, a second office visit may be scheduled during the anticipated symptomatic late luteal phase. Based on clinical experience, most women and clinicians would not comply with prospective monitoring of symptoms for two menstrual cycles. The short monitoring should focus on the individual's complaints and not necessarily on generalized lists. They should include reports on distress and dysfunction as well as bleeding, to ascertain the link to the menses.

During the initial visit, physical and mental histories should be taken and a physical examination performed. Differential diagnosis and ruling out of other conditions (Table 5(23)) is an integral component of the diagnostic process. Obstetric and gynecologic history should include inquiries about any symptoms during pregnancy and postpartum, as well on possible adverse effects of hormonal contraceptives - all may suggest vulnerability to PMS. Most importantly, the cyclicity of symptoms and their entrainment to the menstrual cycle should be confirmed (Figure 2).

Currently there are no objective laboratory tests for PMS/PMDD. However, blood and urine tests to rule out the general disorders listed in Table 5 should also be performed at the first visit. For women in their forties, serum level of folliclestimulating hormone should be determined to evaluate a possible contribution of perimenopausal status.

There is no consensus on the role of catamenial episodes (Table 4), as menstrually related disorders (MRD), and their relation with PMS. If the definition of PMS is based mostly on timing and not on descriptive phenomena, and if genotypes and dynamically evolving vulnerability31,41 contribute to specific phenotypes of PMS, then one may argue that catamenial episodes appearing during the premenstrual period are subtypes of PMS (if they do not appear during other periods of a woman's life). If the catamenial episodes consistently appear during other periods, the non-premenstrual phase of the menstrual cycle (e.g. periovulatory), then they are a part of the broader definition of MRD.

As is the case with subtypes or phenotypes of PMS, this is not just a heuristic discussion. It has direct implications for clinical treatment. In the case of catamenial episodes, treatment should involve the disorder-specific intervention as well as suppression of hormonal cyclicity. This may also be the case with specific phenotypes of PMS.

In that context, clinical follow-up and clinical evaluation of efficacy of treatment should also be confirmed. It seems unrealistic to expect most women to conduct daily monitoring of their symptoms for lengthy periods. Therefore it is productive to schedule follow- up visits for the first 1-2 days of the anticipated menstrual cycle and evaluate the presence of symptoms during the past week, which includes the premenstrual period.

Table 5 Differential diagnosis of premenstrual syndromes

RESEARCH DIAGNOSTIC CRITERIA FOR PREMENSTRUAL DYSPHORIC DISORDER/ PREMENSTRUAL SYNDROMES (INCLUSION AND EXCLUSION CRITERIA)

Owing to the vagueness on severity and other definitions of the DSM-IV PMDD, quantitative, replicable definitions and procedures for diagnosis and outcome measures had to be developed - to be translated into inclusion and exclusion criteria for enrollment of patients in clinical trials.

The research procedures aim at quantification of PMDD criteria of severity (quantify 'marked'), duration of symptomatic period, duration of nonsymptomatic period, degree of impairment and non- existence of other diagnostic entities. In addition, degree of change from non-symptomatic to symptomatic periods and cyclicity are sometimes measured.

The most acceptable research method to establish the severity of symptoms, their entrainment to the menstrual cycle, their fluctuation and cyclicity is by daily rating forms (DRFs). Several DRFs have been designed to assess the PMDD items, for instance the Daily Record of Severity of Problems (DRSP)42 and the Penn Daily Symptom Rating (DSR)43. The Calendar of Premenstrual Experiences (COPE) and the Premenstrual Symptom Diary (PMSD) may be used to assess PMS symptoms, but not necessarily PMDD.

Most widely applied DRFs provide a range of severity for each item, from non-existent to extremely severe. There are some anchor definitions for each level of severity. For research purposes a cut- off point is introduced, mostly between mild and moderately severe. However, that distinction is subjective and varies from individual to individual. If there was an objective measure of severity (there is none at present), it would be subjectively described by some women as not severe and by others as very severe - according to their tolerance, coping style, ability, personality, perception and subjective definition of'what is severe'. The difference between 'mild' and 'moderate' may determine a woman's ineligibility for enrollment in a clinical trial, especially if it is a symptom that would make the fifth required symptom of PMDD or is a crucial perception of impairment.

Since the symptoms should be absent in the week post menses, the same cut-off (although as an upper ceiling) applies to the mid- follicular period as well. Here another issue is pertinent. Some women might be subject to external stressful situations during the mid-follicular phase and consequently rate some of the symptoms higher than 'mild' and have to be disqualified, if clinical judgment is not exercised. An average severity across 5 days may control for incidental stressful days, but even then, a more flexible realistic schedule (e.g. allowing for several external stress related items above 'mild' severity) may reflect a representative patient group.

Since the duration of PMS varies among patients, a reasonable required duration of the severe symptoms may be at least 2 days, with lesser severity during the rest of the PMP. Some DRFs provide an overall daily score for the questionnaire, which may be averaged for the entire designated PMP (usually 7 days) as well as for the mid-follicular phase allowing for calculation of total change from midfbllicular to late-luteal phase and hence a measure of overall fluctuation and cyclicity. At least two issues may be considered in this regard. First, a total score does not reflect severity of PMS in a woman who complains of only few but very severe premenstrual symptoms. This would also be the case with the determination of percentage change from midfbllicular to late-luteal periods. second, the percentage change assumes a linear continuous score from non- existent to extremely severe that is identical for al\l individuals, which is not necessarily the case with the highly subjective description at focus here. Furthermore, a combined requirement of a minimal late-luteal total score and a minimal percentage increase from mid-fbllicular to late-luteal phase may be too strict for some women and very permissive for others (who have many mild symptoms that do not exist at mid-fbllicular phase).

The same considerations apply also to the impairment items. However, there are women who do not have actual premenstrual impairment of their performance but maintain their function with a high level of distress. This should be considered as a measure of severity no lesser than impairment in function.

For research purposes regular menstrual cycles are obligatory, usually 21-35 days are considered to be 'regular'. However, currently there is no requirement of limit of individual cycle-to- cyclc duration stability within this wide range.

Prospective subjects should sign an institutional review board- approved consent form, be assessed in compliance with all regulations and be compliant with all research procedures.

Main exclusion criteria should be:

(1) Not menstruating, including not pregnant;

(2) Past history or family history of vulnerability to possible adverse effects of the medication studied;

(3) Physical disorders severe enough to warrant treatment;

(4) Any chronic major mental disorder;

(5) Alcoholism or substance abuse within 6-12 months prior to the trial;

(6) Oral contraceptives or other hormonal interventions within 3 months prior to screening;

(7) Use of most medications within a time frame of five half- life periods of each medication;

(8) Laboratory values outside the well-defined normal range.

OUTCOME MEASURES IN CLINICAL TRIALS

Diagnostic enrollment criteria for studies of PMDD are logically, first and foremost: being diagnosed as having PMDD according to the DSM-IV criteria, meeting quantitative measures and threshold of seventy, and reported temporal menstrual-cycleentrained change in severity.

In most clinical trials for most medical indications, the measure of successful treatment is remission or not having the situation any more. In cases where the diagnosis of a disorder is not categorical but is based on a threshold along a continuum of severity (e.g. blood pressure for hypertension), success is assessed as bringing the measure within the normal range. In cases of exacerbation of non- curable chronic disorders, remission from the acute episode is a success measure.

It is of interest that this is not the case with PMDD. Even PDA- approved clinical trials do not consider not meeting criteria of PMDD at the completion of the trial as the primary outcome measure. Currently, proof of efficacy of an active component is considered as demonstration of better-than-placebo improvement in an overall daily rating measure (measured as a percentage decrease from baseline). In some large-scale trials there is not even a threshold to define 'responders'. As such, there is no report of percentage of responders to active medications compared with those who responded to placebo treatment.

Not only is the seemingly simple question 'Does the patient have or not have PMDD following treatment?' not answered, but also I am not aware of a single daily monitoring form of PMS symptoms that may not be challenged on psychometric grounds.

Since there is currently no objective measure of PMS, and in most cases no semi-objective behavioral observation by clinicians is available, diagnosis and assessment of clinical improvement depend on patients' own self-reports, whether they do so verbally, as part of an interview, or with written forms and daily checklists. The daily checklists may serve two functions: detailed subjective confirmation of the patient's overall subjective statement; and confirmation of the patient's report of cyclicity of symptoms and their being entrained to the menstrual cycle. It may also be suggested that the assignment of numerical values to subjective severity levels is quite arbitrary and does not necessarily represent a continuum.

Even though cyclicity and timing are essential components of the PMS/PMDD diagnoses, I am unaware of applying them as a primary or secondary outcome measure. Furthermore, in most cases, the daily rating overall score averaged over a specified general period of time (5-7 days) serves as a primary outcome measure even though that score is the sum of up to 21 items/symptoms when only five are required to meet entry criteria. It may be demonstrated that an overall decrease in the total score may not reflect improvement in the key symptoms. Moreover, patients may be in full remission and not suffering from any of their initial symptoms in higher than mild level with no impairment, but still would not meet the criteria of 'responder' based on a percentage decrease in total score (usually 50%).

It is recommended that in clinical trials the primary outcome measure should be based on 'not meeting entry criteria post- treatment', the summary of which is riot having PMDD any more.

Since even with the current criteria for PMDD (and especially for PMS symptoms) qualified symptoms may vary from individual to individual, outcome may more accurately be assessed based on change in severity of these symptoms and not on general change or no change in symptoms that were not clinically significant to start with.

Since impairment or dysfunction is the essential component of PMS/ PMDD, its improvement should be an essential part of the primary outcome measure.

Secondary outcome measures should be the components of the PMS/ PMDD criteria, since elimination of one of them (e.g. impairment) would preclude meeting the full diagnostic criteria even when others still exist.

Cyclicity of symptoms should be an outcome measure, especially in clinical trials in which the main demonstrated action of the active compound(s) is elimination of hormonal cyclicity by suppression of ovulation (e.g. hormonal contraceptives, gonadotropin-releasing hormone analogues) or any other hormonal or other interventions.

FUTURE DIRECTIONS FOR RESEARCH

The main diagnostic issues that, in my opinion, are still unsolved are:

(1) Are there multiple diversified premenstrual syndromes that also include diversified premenstrual dysphoric phenotypes? This issue is also of crucial importance for understanding the pathobiology of PMS and eventually for efficacious treatments (beyond the current efficacy ceiling of 60%).

(2) If there are diversified phenotypes, then can we move beyond the DSM-IV style descriptive arbitrary cut-off points towards a diagnosis based on symptoms' pattern, past history and time course, biological changes (etiology and pathobiology) and treatment outcome? Once phenotypes are established, are they specifically associated with specific genotypes?

(3) If our present understanding of PMS involves the concept of vulnerability and menstrually related trigger(s) of symptoms, then is it justified to distinguish between PMS and PME or catamenial disorders? The difference may be the degree of the threshold and the magnitude of the trigger needed to cause expression of symptoms, but not necessarily a fundamental difference between PMS and PME. If the vulnerability level changes over time, increases or decreases according to life events and their perception, as well as repeated assaults (kindling and dynamically evolving vulnerability), then there may be a continuum between at least some PMS, PME and more chronic major disorders. This notion deserves investigation.

(4) What is clinically relevant PMS (or PMDD)? When do women warrant and benefit from treatment?

(5) Can we develop clinically relevant diagnostic procedures that would not require prospective monitoring of symptoms for two consecutive menstrual cycles (which are too much of a burden for many women and their primarycare physicians)?

(6) Do we really need prospective confirmation of severe cyclic symptoms of two consecutive cycles, when most people have 'bad' months as well as 'good' ones? Should a woman with two symptomatic cycles out of three be diagnosed as suffering of PMS that warrants treatment? What is 'most menstrual cycles'?

(7) Are women enrolled in clinical trials of PMS/ PMDD representative of the general population of women with these disorders (who may benefit from treatment)?

(8) What are the characteristics and treatment response of women who did not meet enrollment criteria at screening due to reporting an insufficient number of severe symptoms (reporting, say, three or four severe symptoms), lesser severity and/or cycle-tocycle fluctuations in seventy but not in nature of symptoms?

(9) Is PMS different from perimenstrual disorders? Is the temporal pattern of symptoms phenotype-specific? Are there different clinically relevant MRD?

(10) Are there periovulatory mood changes? Many women report such changes; some of them are associated with stress and changes in stress hormones.

(11) Are women who report positive premenstrual changes (e.g. increased energy, appetite, creativity assertiveness or sex drive) different from those who report negative changes? Is there a continuum between positive changes, hypomania and premenstrual manic episodes?

There are many more unsolved diagnostic issues of PMS even before we tackle issues of the underlying mechanisms (which may vary from phenotype to phenotype).

It is of value to know what we do not know. Even more valuable is to know what we believe we know but actually do not.

REFERENCES

1. American Psychiatrie Association. Diagnostic and Statistical Manual of Mental Disorders, 4th edn. Washington, DC: American Psychiatric Press, Inc., 1994

2. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th edn, Text Revision (DSM-IV-TR). Washington, DC: American Psychiatric Press, Inc., 2000:771-4

3. Angst J, Sellaro R, Merikangas KR, Endicott J. The epidemiolog\y of perimenstrual psychological symptoms. Acta Psychiatr Scand 2001; 104:110-16

4. Johnson SR. The epidemiology and social impact of premenstrual symptoms. CHn Obstet Gynecol 1987;30:367-76

5. Merikangas KR, Foeldenyl M, Angst J. The Zurich Study. XIX. Patterns of menstrual disturbances in the community: results of the Zurich Cohort Study. Bur Arch Psychiatry Clin Neurosci 1993;243:23- 32

6. Ramcharan S, Love EJ, Pick GH, Goldfien A. The epidemiology of premenstrual symptoms in a population-based sample of 2650 urban women: attributable risk and risk factors. J Clin Epidemiol 1992;45:377-92

7. Rivera-Tovar AD, Frank E. Late luteal phase dysphoric disorder in young women. Am J Psychiatry 1990; 147:1634-6

8. Sveindottir H, Backstrom T. Prevalence of menstrual cycle symptom cyclicity and premenstrual dysphoric disorder in a random sample of women using and not using oral contraceptives. Acta Obstet Gynecol Scand 2000;79:405-13

9. Wittchen HU, Hoyer J. Generalized anxiety disorder: nature and course. J Clin Psychiatry 2001;62(Suppl. 11):15-19; discussion 20- 1.

10. Woods NF, Most A, Dery GK. Prevalence of perimenstrual symptoms. Am J Public Health 1982;72:1257-64

11. Halbreich U, Borenstein J, Pearlstein T, Kahn LS. The prevalence, impairment, impact, and burden of premenstrual dysphoric disorder (PMS/PMDD). Psychoneuroendocrinology 2003;28:1-23

12. Robinson RL, Swindle RW. Premenstrual symptom severity: impact 011 social functioning and treatment-seeking behaviors. J Womens Health Gend Based Med 2000;9:757-68

13. Steiner M, Brown E, McDougall M. The burden of illness of premenstrual symptoms. J Womens Health Gend Based Med 2002;11:324.

14. Cohen LS, Soares CN, Otto MW, et al. Prevalence and predictors of premenstrual dysphoric disorder (PMDD) in older premenopausal women. The Harvard Study of Moods and Cycles. J Affect, Disord 2002;70:125-32

15. Chawla A, Swindle R, Long S, et al. Premenstrual dysphoric disorder: is there an economic burden of illness? Med Care 2002;40:1101-12

16. Wittchen HU, Becker E, Lieb R, Krause P. Prevalence, incidence and stability of premenstrual dysphoric disorder in the community. Psychol Med 2002;32:119-32

17. Narrow WE, Rae DS, Robins LN, Regier DA. Revised prevalence estimates of mental disorders in the United States: using a clinical significance criterion to reconcile 2 surveys' estimates. Arch Gen Psychiatry 2002;59:115-23

18. Pearlstein TB, Halbreich U, Batzar ED, et al. Psychosocial functioning in women with premenstrual dysphoric disorder before and after treatment with sertraline or placebo. J Clin Psychiatry 2000;61:10l-9

19. Murray CJL, Lopez AD, eds. The Global Burden of Disease: A Comprehensive Assessment of Mortality and Disability from Diseases, Injuries, and Risk Factors in 1990 and Projected to 2020. Boston, MA: Harvard School of Public Health, on behalf of the World Health Organization and the World Bank, 1996

20. Hylan TR, Sundell K, Judge R. The impact of premenstrual symptomatology on functioning and treatment-seeking behavior: experience from theUnited States, United Kingdom, and France. J Womens Health Gend Based Med 1999;8:1043-52

21. World Health Organization (WHO). International Classification of Diseases, 10th revision. Geneva: WHO, 1996

22. American College of Obstetricians and Gynecologists (ACOG). Premenstrual Syndrome. Washington, DC: ACOG, 2000

23. Isaacson K, Balasubramanyam A. Relationship of medical illness to the menstrual cycle. In Smith S, Schiff I, eds. Modern Management of Premenstrual Syndromes. New York and London: WW Norton and Company, 1993:82-97

24. Halbreich U, Endicott J. The relationship of dysphoric premenstrual changes to depressive disorders. Acta Psychiatr Scand 1985;71:331-8

25. Halbreich U, Endicott J, Schacht S, Nee J. The diversity of premenstrual changes as reflected in the Premenstrual Assessment Form. Acta Psychiatr Scand 1982;65:46-65

26. Halbreich U. Menstrually related disorders: what we do know, what we only believe that we know, and what we know that we do not know. Crit Rev Neurobiol 1995;9:163-75

27. Halbreich LJ. Premenstrual dysphoric disorders: a diversified cluster of vulnerability traits to depression. Acta Psychiatr Scand 1997;95:169-76

28. Halbreich U. Premenstrual syndromes. Ballieres Clin Psychiatry hit Pract Res 1996;2:667-86

29. Halbreich U. Menstrually related changes and disorders: conceptualization and diagnostic considerations. Neuropsychopharmacology 1993;9:25-9

30. Halbreich U, Alt IH, Paul L. Premenstrual changes. Impaired hormonal homeostasis. Neural Clin 1988;6:173-94

31. Halbreich U. The etiology, biology and evolving pathology of premenstrual syndromes. Psychoneuroendocrinology 2003;28:55-99.

32. Halbreich U, Endicott ], Nee J. Premenstrual depressive changes. Value of differentiation. Arch Gen Psychiatry 1983;40:535- 42

33. Halbreich U, Endicott J. Methodological issues in studies of premenstrual changes. Psychoneurocndocrtnology 1985:10:15-32

34. Bloch M, Schmidt PJ, Rubinow DR. Premenstrual syndrome: evidence for symptom stability across cycles. Am J Psychiatry 1997;154:1741-6

35. Halbreich U, Marshall-Hobika D, Garde J. Kesponse rates to selective serotonin reuptake inhibitors for treatment of premenstrual dysphoric disorder. In Proceedings of the 2nd World Congress of Women's Mental Health, Washington, DC, 17-20 March 2004 p. 150

36. Halbreich U. Premenstrual syndromes: closing the 20th century chapters. Curr Opin Obstet Gynccol 1999; 11:265-70

37. Halbreich U. Menstrually related disorders towards interdisciplinary international diagnostic criteria. Cephalalgia 1997;17(Suppl. 20): 1-4

38. Endicott J, Halbreich U. Retrospective report of premenstrual depressive changes: factors affecting confirmation by daily ratings. Psychopharmacol Dull 1982;18:109-12

39. Kahn L, Halbreich U, Hershey C, Bloom M., Bidani R, Rich E. Screening for mental illness in primary care clinics. Submitted for publication

40. Halbreich U. Premenstrual syndromes. In Rackel RE, Dope ET, eds. Conn's Current Therapy. Philadelphia, PA: WB Saunders and Company, 2004:1110-14

41. Halbreich U. Future directions for studies of women's mental health. Psychopharmacol Bull 1998:34:327-31

42. Endicott J, Harrison W. Daily Rating of Severity of Problems Form. New York: Department of Research Assessment and Training, New York State Psychiatric Institute, 1990

43. Freeman EW, DeRubeis RJ, Rickeis K. Reliability and validity of a daily diary for premenstrual syndrome. Psychiatry Res 1996;65:97-106

44. Mortola JF, Girton L, Beck L, Yen SS. Diagnosis of premenstrual syndrome by a simple, prospective, and reliable instrument: the calendar of premenstrual experiences. Obstet Gynecol 1990;76:302-7

45. Thys-Jacobs S, Alvir JM, Fratarcangelo P. Comparative analysis of three PMS assessment instruments - the identification of premenstrual syndrome -with core symptoms. Psychopharmacol Bull 1995;31:389-96

U. Halbreich

Biobehavioral Program, State University of New York at Buffalo, Buffalo, New York, USA

Correspondence: Dr U. Halbreich, State University of New York at Buffalo, Uiobehavioral Program, Hayes C, Suite 1, 3435 Mam St. Bldg. 5, Buffalo, NY 14214-3016, USA

Copyright CRC Press Dec 2004

Source: Gynecological Endocrinology

More News in this Category