Can oestrogen use prevent the onset of psychosis?

To the Editor

The oestrogen protection hypothesis in psychosis is well recognised (Riecher-Rössler and Häfner, 1993). Oestrogen inhibits postsynaptic dopamine transmission and this may account for the proposed antipsychotic effect (Grossman et al., 2008). This hypothesis is further supported by studies whereby adjunctive oestrogen is effective in the setting of acute psychosis (Kulkarni et al., 2008). What is less well established is whether oestrogen could prevent psychosis in women at risk during menopause.

In this case report, we present a 38-year-old woman prescribed dexamphetamine for attention deficit disorder (ADD) for several years, who subsequently developed a psychotic disorder following cessation of oestrogen replacement therapy prescribed for primary ovarian failure when she was aged 36. The important question this report raises is whether the psychosis would have emerged if oestrogen replacement therapy was continued indefinitely.

Ms K was on dexamphetamine (30 mg/day) for an adult attention deficit disorder (ADD) for 5 years with no noted complications. Ms K did not have any known genetic predispositions. With the onset of symptoms suggestive of menopause, Ms K underwent investigations and was discovered to have primary ovarian failure.

Five months following this diagnosis, Ms K was started on Oestradiol Gel 1 sachet daily. Three months later, Oestradiol Transdermal Patch was added to the Oestradiol Gel. Ms K subsequently reported intolerable side effects (inter-menstrual bleeds and menorrhagia) which resulted in the Oestradiol Transdermal Patch being ceased. Ms K later discontinued the Oestradiol Gel. Approximately 8 weeks following cessation of all oestrogen replacement therapy, Ms K presented with her first episode of psychosis characterised by persecutory delusions. A full organic workup identified no physical aetiology for her psychosis. Ms K denied the overuse of dexamphetamine and her psychiatrist confirmed that this was unlikely, based on her close supervision in the community.

During Ms K’s first hospitalisation, her psychosis was initially suspected to be triggered by dexamphetamine in a milieu of low oestrogen related to premature menopause. Her paranoid psychosis was treated successfully with olanzapine. Her dexamphetamine therapy was ceased and Ms K declined oestrogen therapy as she feared side effects as experienced previously. Ms K ceased her antipsychotic treatment soon after discharge and was free of psychosis till she was recommenced on dexamphetamine (30 mg/day) for her ADD. Ms K had two further psychotic episodes warranting hospitalisation related to non-compliance with antipsychotic treatment and the ongoing prescription of dexamphetamine.

Ms K declined oestrogen therapy during all psychiatric hospitalisations, despite receiving psycho-education about its potential adjuvant antipsychotic benefits. On her last hospital discharge, Ms K’s psychosis was stable on quetiapine 600 mg a day.

In reviewing this case, the authors postulate that dexamphetamine use on the background of oestrogen depletion could have led to Ms K’s psychosis. Whether oestrogen replacement therapy can prevent psychosis in at-risk women warrants further research. In the case of Ms K, we speculate that exogenous oestrogen replacement following the diagnosis of primary ovarian failure prevented the onset of psychosis despite ongoing dexamphetamine use. The cessation of oestrogen therapy was possibly a significant factor in the timing of the onset of her psychosis.

This case adds to the literature in the area of oestrogen depletion contributing to psychosis. It suggests the potential benefit of oestrogen use in early perimenopausal women to prevent the onset of psychosis in vulnerable patients with genetic or exogenous predisposition. This has previously been proposed by other authors in the literature (Felthous et al., 1980).