Unusual manifestations of premenstrual syndrome



    1. Women’s Mental Health Program, Department of Psychiatry, Chang Gung Memorial Hospital,
    2. Chang Gung University School of Traditional Chinese Medicine and
    Search for more papers by this author

    Corresponding author
    1. Women’s Mental Health Program, Department of Psychiatry, Chang Gung Memorial Hospital,
    2. Institute of Clinical Behavioral Sciences, Chang Gung University, Tao-Yuan, Taiwan
    Search for more papers by this author

Chia-Yih Liu, MD, Department of Psychiatry, Chang Gung Memorial Hospital, 5 Fu-Shin Street, Tao-Yuan, Taiwan. Email: liucy752@cgmh.org.tw


Abstract  The authors have identified six unusual manifestations of premenstrual syndrome: (i) typical premenstrual dysphoric disorder symptoms during ovulation; (ii) upper airway infection in the premenstrual phase that remits after onset of menstruation; (iii) recurrent perimenstrual erythematous and edematous skin plaques; (iv) hypomanic or manic state 2–3 days before onset of menstruation; (v) auditory hallucinations and delusion of reference, present only premenstrually; and (vi) remitted bipolar disorder presenting with premenstrual psychotic features. Psychiatrists should be aware of menstrual cycle-related changes in mental illness, physical illness, or both in women of reproductive age, if cyclic trends are suspected.


Premenstrual syndrome (PMS) is believed to affect up to 80% of women of reproductive age. Severe premenstrual syndrome, or premenstrual dysphoric disorder (PMDD), affects 3–8% of women, imposing an enormous burden on these individuals and their families.1

In everyday clinical practice, we encounter the prototypes described in textbooks, but presentation is sometimes atypical. Most of the original studies excluded these unusual cases, with only the diagnostically distinct cases receiving evaluation and management. In the current report we present six patients with unusual premenstrual syndrome from one or more of three aspects: unusual onset of menstrual period, concurrent infection or immune-mediated disorder, or clinical course characterized by atypical bipolar or psychotic symptoms.


This study was conducted as part of the Chang Gung Depression and PMS Program (CDPP). Subjects were recruited from women who visited the women’s mental health clinic in a medical center in Taiwan.2 The study was approved by the institutional review board and all six participants provided written informed consent.

All women except patient 1 met ICD-10 diagnostic criteria for premenstrual syndrome.3 Psychiatric diagnoses were made by psychiatrists using the structured Mini-International Neuropsychiatric Interview (MINI).4 The subjects were given a copy of the Prospective Record of the Impact and Severity of Menstrual Symptomatology (PRISM) calendar,5 and a series of laboratory tests were scheduled including levels of hemoglobin, thyroid-stimulating hormone, fasting glucose, estradiol, progesterone, prolactin, and follicle-stimulating hormone. The diagnosis of PMDD was confirmed when no abnormal laboratory data were noted and there were two prospective cycles of charting for these subjects.

Daily PRISM ratings were summed for each subject in each cycle to obtain premenstrual (2–6 days before menses) and postmenstrual scores (cycle days 7–11). All subjects except patient 1 demonstrated an increase of at least 50% in PRISM severity score (PSS) when postmenstrual ratings were compared with premenstrual ratings for two screening cycles. Treatment outcome measures included scores on the PRISM calendar and the Clinical Global Impression Scale for Severity (CGI-S) and Global Impression Scale for Improvement (CGI-I).


The six case reports are summarized in Table 1.

Table 1.  Case study of atypical premenstrual syndrome
Patient dataCurrent diagnosisUnusual PMS featuresPRISM calendar PSSCGI-STxPost-Tx CGI-S/CGI-IPost-Tx calendar PSS
  1. CGI-S/CGI-I, Clinical Global Impression Scale (severity and improvement); NA, not applicable; PMDD, premenstrual dysphoric disorder; PMS, premenstrual syndrome; PRISM calendar, Prospective Record of the Impact and Severity of Menstrual Symptomatology calendar; PSS, PRISM calendar severity score: (total scores of luteal phase–scores of follicular phase)/(scores of follicular phase); Tx, treatment.

25 years, single studentNATypical PMDD symptoms for 2 days during ovulationNA6Alprazolam 0.5 mg daily in the 2 days during ovulation2/2NA
32 years married housewifePMDD, frequent premenstrual upperUpper airway infection only in premenstrual phase, remitting respiratory tract infection75%6Paroxetine 20 mg daily for 6 months after onset of menstruation2/221%
33 years married school teacherPMDD, autoimmune progesterone dermatitisRecurrent perimenstrual erythematous and edematous skin plaques84%6Prednisolone 10 mg daily premenstrually and venlafaxine 150 mg daily full cycle for four cycles2/1 8%
24 years single unemployed borderline mental retardationBipolar disorder not otherwise specified, PMS, dysmenorrheaHypomanic or manic state 2–3 days premenstrually87%5Sulpiride 200 mg daily for 3 days premenstrually1/1 0%
42 years married school staffPMDD, recurrent brief psychotic disorderAuditory hallucinations, delusion of reference, only premenstrually93%5Sulpiride 200 mg premenstrually1/1 0%
34 years married bank clerkManic episode, full remission, post-partum onsetIdeas of reference and persecution, elementary auditory hallucinations88%6Valproate 750/1000 mg and sulpiride 200/400 mg daily full cycle/premenstrually2/217%

Patient 1 was a 25-year-old woman whose typical core PMDD symptoms had occurred only in the periovulatory period (2–3 days) during the past 2 years. She had a totally symptom-free premenstrual period. After treatment with anxiolytics, her periovulatory PSS scores decreased from 90% to 10%. The ‘periovulatory dysphoric syndrome’ was still noted after 2 years of follow up.

Patient 2 was a 32-year-old woman who had had PMDD since the age of 28 years. She had experienced frequent premenstrual upper respiratory tract infections with postmenstruation remission for over 1 year. After 20 mg daily paroxetine for 6 months, she reported that the frequency and severity of her respiratory tract infections had both decreased (CGI severity reduced from 6 to 2). She was followed at our service for more than 2 years with luteal-phase paroxetine treatment. Her PMDD symptoms decreased over time, with only occasional mild upper respiratory tract infection symptoms before menstruation.

Case 3 involved a 33-year-old woman with PMDD who had an uncommon comorbid disorder, autoimmune progesterone dermatitis, characterized by the appearance of skin eruptions during the premenstrual period. Her routine biochemical laboratory tests and serum levels of immunoglobulins (IgG, IgA, IgM, and IgE) and complement components (C1, C2, C3, C4) were within normal limits. Prednisolone 10 mg daily premenstrually and venlafaxine 150 mg daily during the full menstruation cycle were prescribed. Her PMDD symptoms responded well to antidepressant treatment and the dermatitis was progressively relieved over four menstrual cycles. This patient was lost to follow up 8 months later.

Patient 4 was a 24-year-old woman with regular menstrual cycles who had been diagnosed with bipolar disorder with recurrent manic episodes for 2 years previously. Her manic symptoms occurred periodically every month. The manic or hypomanic symptoms responded to an antipsychotic agent, sulpiride 200 mg daily given only on symptomatic days. Synchronization of mania/hypomania with menstruation was suspected. We discontinued her psychotropics for two menstrual cycles after gaining informed consent. She and her father completed the PRISM calendar for the following two cycles, resulting in a high severity score (87%). She developed restlessness, insomnia, talkativeness, hyperactivity, and various disinhibited behaviors 2–3 days before menstruation, with all symptoms remitting on the second day after menstruation began. She was treated with premenstrual doses of the same antipsychotic.

Patient 5 was a 42-year-old woman who reported a 3-year history of bewildering auditory hallucinations and delusion 3–5 days before every premenstrual cycle, with all symptoms remitting on the second day of her menstrual period. All biochemistry and computed tomography studies produced negative findings. Fortunately, sulpiride 200 mg daily was effective when given on symptomatic days. After 2 years of follow up, the patient still experienced premenstrual psychotic symptoms that responded to sulpiride 200 mg daily given over the symptomatic period. The patient was generally quite well and maintained a good functional level.

Case 6 involved a 34-year-old woman diagnosed with bipolar disorder, single manic episode with post-partum onset. After full remission of the typical manic episode, the patient developed ideas of reference and persecution and elementary auditory hallucinations 1 week before almost every menstrual period, with full remission in the 3–4 days after menstruation began. The mood stabilizer valproate was administered together with the antipsychotic sulpiride, with dosages increased during the luteal phase (valproate 750/1000 mg and sulpiride 200/400 mg daily in usual days and premenstrual week, respectively). After 3 months of treatment, the severity of psychotic symptoms decreased, and the PSS score decreased from 88% to 17%. One year later the patient was still followed at our clinic with sulpiride 400 mg daily. Premenstrual psychotic symptoms were still noted but were not disturbing.


A severity score elevation at ovulation as well as during the premenstrual phase has been noted in some patients with premenstrual syndrome.6 It seems that ovulation and the premenstrual phases are both times of increased vulnerability to mental disorder. However, there have been no reports of patients who presented with typical symptoms of PMDD occurring exclusively during ovulation (as opposed to the premenstrual period), as was the case with patient 1. One possibility is that the patient’s vulnerability to mental disturbance during the premenstrual phase was at the subclinical threshold. Interestingly, the patient continued to report no premenstrual discomfort after 2 years of follow-up.

A highly significant clustering of infectious illness symptoms during the perimenstrual period compared with the mid-menstrual cycle has been reported.7 Some authors have indicated that these conditions arise from immunological origins rather than entirely from hormonal causes.8 It is interesting that patient 2 had a decreased frequency of, and symptom severity of, premenstrual upper airway infection after a period of continuing paroxetine treatment. Neveu and Castanon have noted that depression may include different subtypes with different immune alterations that may be epiphenomena or contributory to pathophysiological processes.9 Thus, improvement in immune alterations with antidepressant treatment in depressed patients may be related to the effects of these agents.

The autoimmune progesterone dermatitis present in patient 3 is a rare disorder, which may be due to progesterone sensitivity.10 Various treatment modalities have been described, including administration of conjugated estrogens, antihistamines, corticosteroids, and tamoxifen.9 We treated this patient with a short-term prednisolone adjuvant and long-term venlafaxine. The response was satisfactory. Evaluation of long-term efficacy may require further follow-up study.

Patient 4 demonstrated mania-like symptoms after documentation of true symptom cyclicity through daily charting of the PRISM calendar; we found that the onset period was typically in the late luteal phase of the menstrual cycle. However, a literature review showed that most manic episodes occur near menstruation or during the follicular phase.6–11 Whether the impression of mania was truly atypical premenstrual syndrome in this woman or whether the diagnoses were comorbid requires further investigation, for instance, therapeutic trial. The patient refused mood, or even a fluoxetine trial (a good treatment in the general population with severe premenstrual syndrome or PMDD) due to a good response to sulpiride. However, it is important to attempt to distinguish cyclical behavioral changes from menstruation-related mood disturbances. Daily calendar charting during the entire menstrual cycle by subjects and family members or other caregivers may prove helpful.

There are many reports of periodic psychosis with each episode followed by complete remission within 2 weeks,12 but nosology and long-term prognosis in such cases have not been elucidated. Abe and Ohta prospectively followed 11 cases of periodic psychosis (involving nine girls and two boys).13 The episodes were linked to one phase of the menstrual cycle in only two of the six girls with regular menses. Of nine patients followed for 5–14 years after first onset, three were well, three had become bipolar, and three were still experiencing brief mood episodes (as in patient 6). Recurrent brief mood episodes tend to show a near-monthly rhythm and psychotic features.

A small subset of women who experience psychotic symptoms in the premenstrual phase has been identified.14 There are reports in the literature of gender differences in bipolar illness, and issues relevant to the treatment of women with this illness have been discussed.15 Although it is clear that bipolar women are at high risk for post-partum episodes, the effects of other reproductive system events (such as puberty, menstrual cycle, pregnancy, menopause, use of oral contraceptives or hormone replacement therapy) on the course or treatment of bipolar illness have received little systematic study. Gender differences in bipolar illness and effects of the female reproductive system on the course and treatment of the illness deserve further investigation.

In general, the psychiatrist should be aware of gender-based differences and menstrual cycle-related changes in mental illness. If cyclic trends are suspected, patients should be encouraged to use a menstrual calendar (e.g. the well-designed PRISM calendar) to track signs and symptoms for at least two cycles. And vice versa, unusual presentations of severe premenstrual syndrome or PMDD have important implications in terms of reconsideration of the diagnosis for most female mental illness.


This study was supported in part by grants from the National Science Council of Taiwan (NSC-090-2314-B-182-054, NSC-91-2314-B-182-049) and Chang Gung Memorial Hospital (CMRP-1267, CMRPG32106).