Estrogen Administration Does Not Reduce the Rate of Recurrence of Affective Psychosis After Childbirth

Background: High rates of postpartum relapse occur in women with histories of bipolar or schizoaffective disorder. These relapses may be triggered by the postdelivery fall in circulating estrogen through alteration of central neurotransmitter (especially dopaminergic) systems. This study tested the hypothesis that estrogen administration after childbirth would prevent postpartum relapse and would alter dopamine receptor sensitivity.

Method: Twenty-nine pregnant women with a Research Diagnostic Criteria diagnosis of hypomania (bipolar II), mania (bipolar I), or schizoaffective disorder participated in an open clinical trial. Three transdermal dose regimens of estrogen (17beta-estradiol) were tested. Starting doses were 200 (N = 13), 400 (N = 3), and 800 (N = 13) micrograms/day, beginning within 48 hours after delivery and reduced by one half every 4 days for a total of 12 days. On the fourth day after starting estradiol therapy (before relapse occurred), subjects participated in a neuroendocrine challenge test that measured the sensitivity of the central nervous system (tubero-infundibular) dopaminergic system (plasma prolactin and growth hormone responses to apomorphine).

Results: Estradiol at all dose regimens did not reduce the rate of relapse. However, of the 12 women who relapsed, those who had taken the highest dose of estradiol (800 micrograms/day) needed less subsequent psychotropic medication (fewer chlorpromazine equivalents) and were discharged sooner than those who had taken either of the 2 lower doses. No differences in neuroendocrine responses to apomorphine were detected between women receiving the high-dose and the lower-dose regimens.

Conclusion: The results do not support the hypothesis that a fall in circulating concentrations of estrogens precipitates relapse in subjects at risk of postpartum affective psychosis. The use of prophylactic estrogen in such circumstances is therefore highly questionable.