Challenges in diagnosis, treatment and coordination of care of menstrual psychosis

Background

Menstrual psychosis has been a documented phenomenon since the 1800s yet remains largely described in case studies and generally under-recognised. It is characterised by recurring episodes of psychosis in rhythm with the menstrual cycle, where the timing within the cycle may differ among those affected. Repeated attempts have been made to classify menstrual psychosis by timing, first temporally, then by the stage in reproductive life. Most recently, the classification has been restructured by timing within the menstrual cycle (premenstrual, catamenial, paramenstrual, mid-cycle or epochal) and the reproductive epoch (single episodes at menarche, prepubertal, during amenorrhoea, onset after childbirth or after menopause) in which they occur.1 To diagnose menstrual psychosis, presentations must demonstrate an acute onset from a normal baseline, brief duration with full recovery, features of psychosis or mania, and an established rhythm with the menstrual cycle.2 Menstrual psychosis can be difficult to differentiate as it can also appear similar to other conditions such as brief psychotic disorder, schizophrenia and substance/medication-induced psychosis.3 While menstrual psychosis episodes fulfil many of the criteria for bipolar disorder, it is the rhythmicity with the menstrual cycle and hormone changes that set it apart.

Further diagnostic ambiguity is created when episodes of menstrual psychosis are inconsistent and do not occur in each menstrual cycle, making it difficult for an acute treating physician to identify a pattern and coordinate care. In a 2018 case report, Thippaiah et al described the proposed aetiologies of menstrual psychosis over the years, including: psychodynamic theory, genetic theory, hormonal theory, oestrogen hypothesis, and interaction of serotonin/dopamine and oestrogen/progesterone with other biological systems in the brain.4 While several of these theories point toward potential coordination of treatment with gynaecology, only a handful of inpatient psychiatric hospitals have integrated care to address gynaecological needs, and outpatient collaboration between psychiatry and gynaecology remains low.5

Case presentation

A young female in early adolescence presents with recurrent episodes of acute manic and psychotic symptoms beginning with menarche. Family history is significant for bipolar disorder, anxiety and multiple family members with dysphoric mood during menstruation. During each admission to the hospital, the presentation appeared to be consistent with acute emergence of agitation, aggression, labile mood, disorganised thinking and behaviours, hallucinations, poor sleep and paranoia, with return to baseline mental status prior to discharge. During the third episode, the mother and the patient reported that all episodes had begun about 3–4 days prior to onset of menses, with the natural timing of her return to baseline being unclear given that no episodes occurred without the intervention of psychotropic medications. The average length of stay during each of the four admissions was 12.5 days. The patient’s mother describes the first noticeable changes from baseline as a decreased need for sleep and depressed mood, followed by increasing lability, disorganisation of thoughts and behaviours, and eventual escalation to psychosis and aggression.

Differential diagnosis

Several diagnoses were considered, including bipolar mood disorder, early-onset schizophrenia and premenstrual dysphoric disorder. Our team’s initial working diagnosis was bipolar mood disorder, a chronic disorder characterised by episodes of at least three of the following: distractibility, increased goal-directed activities, grandiosity, pressured speech, decreased need for sleep and racing thoughts, in the context of an elevated, irritable or expansive mood. Manic episodes may also involve impulsivity, lability, oftentimes with psychotic features including disorganised speech and behaviours. Due to the temporal relation of episodes with her menstrual cycles, frequency of episodes, brief duration of each episode and improvement during periods of amenorrhoea, her presentation aligned more with menstrual psychosis. Based on the timing of symptoms within the menstruation cycle, our patient’s presentation would be classified as premenstrual psychosis based on Brockington’s classification. Such cases of psychosis typically commence in the latter half of the menstrual cycle and can end soon after menstruation begins.1 Both the patient and her mother reported that symptoms began 3–4 days before menstruation with variable time to symptom resolution.

Schizophrenia was another consideration, which has a chronic continuous course, including at least two of the following symptoms: hallucinations, delusions, disorganised speech and behaviours, as well as negative symptoms often preceded by a prodrome of less severe symptoms and social withdrawal. Typical age of onset is in the late 20s for females, with childhood onset being much rarer. Our patient presented with an acute change from baseline over the course of a few days with rapid resolution, making schizophrenia a less likely diagnosis. Premenstrual dysphoric disorder encompasses multiple symptoms of depression, physical symptoms and labile mood for most menstrual cycles, resolving soon after the cycle ends. While symptom emergence and duration seemed to be in rhythm with the menstrual cycle, our patient’s symptoms aligned with mania or psychosis rather than depression.

Treatment

During the patient’s first episode, she was taken to a crisis centre outside our system and started on aripiprazole. Symptoms appeared to resolve, and she was discharged with a plan to follow up with outpatient psychiatry. Despite compliance with her medication, the patient had a second episode 1 month later. Aripiprazole was increased to 10 mg/day, symptoms stabilised and she was discharged to her established outpatient psychiatrist at an outside hospital system. During admission for her third episode 3 months later, aripiprazole was switched to risperidone 3 mg/day, with the addition of diphenhydramine 25 mg and lithium 150 mg in the evening. Brain MRI, lead levels, urine drug screen, thyroid panel, complete blood count, comprehensive metabolic panel and heavy metal urine screening were unremarkable. Gonadal hormone assay was not included at this time. The patient’s mother reported that these past three episodes seemed to occur 3–4 days prior to menses. Due to a lack of access to inpatient gynaecology consultation at our mental health centre, an informal discussion with outpatient gynaecology recommended a combined oral contraceptive (OCP) that could be taken continuously over 3 months. The patient was discharged with a follow-up to gynaecology and endocrinology with concerns of menstrual psychosis.

Between admission for her third and fourth episodes, about 17 months, the patient experienced galactorrhoea and amenorrhoea while on risperidone. Prolactin levels were elevated, and the gynaecologist recommended reducing the medication. Additionally, the patient had decided not to take the OCP due to the rare chance of blood clots related to OCP use. Risperidone was reduced by her outpatient psychiatrist. Shortly thereafter, she presented with her fourth episode. She was switched to olanzapine and titrated to 20 mg/day. Lithium, increased to 900 mg/day prior to hospitalisation, was continued.

After discharge from inpatient psychiatry, she was stepped down to partial hospitalisation and intensive outpatient hospitalisation, along with continued outpatient follow-up with gynaecology. Over the next 3 months, she experienced an approximate 10-pound weight gain, significantly elevated prolactin level with suppressed oestradiol and irregular menses. Galactorrhoea was not present. It was decided to taper off the olanzapine and continue lithium monotherapy during the intensive outpatient programme. Meanwhile, to obtain baseline hormone levels, gynaecology recommended holding any hormonal therapies. Shortly following completion of the intensive outpatient programme and the taper off olanzapine, she had her fifth episode and subsequent hospitalisation, coinciding with her menstrual cycle. Symptoms were greater in intensity and olanzapine was restarted and titrated to 30 mg/day to achieve stabilisation. She was discharged at her baseline mental status.

Outcome and follow-up

Following the last hospitalisation, she was hospitalised for 2 days at an outside hospital for an episode that her mother reported had occurred outside the rhythm of her usual menstrual cycle. Based on the much shorter admission, this episode was presumed to be less intense than prior episodes. She continues to refuse the recommended OCP but does follow-up with her gynaecologist, who report that oestrogen and prolactin levels have normalised. She has had regular menstrual cycles while remaining on oral olanzapine, reduced to 15 mg at bedtime, and lithium 900 mg at bedtime per the established outpatient psychiatrist.