Clinical Course of Illness in Women With Early Onset Puerperal Psychosis: A 12-Year Follow-Up Study

Objective: To complete a follow-up analysis at a mean of 12 years after patients had presented with an early onset puerperal psychotic index episode.

Method: A retrospective design was used. Patients with puerperal psychosis and onset within 4 weeks after childbirth who had been referred to the Psychiatric Department of the Ludwig Maximilian University of Munich, Munich, Germany, between 1975 and 1995 (maximum: 24 years, minimum: 7 years) were followed up after a mean of 12 years post index episode. Ninety patients were included in the study. Before the index episode, 35 of the patients had previous nonpuerperal psychoses, while 55 patients presented their index episode as the first manifestation of a psychotic illness. Diagnostic evaluation at follow-up was performed by the Structured Clinical Interview for DSM-IV Axis I Disorders according to DSM-IV-TR. Differential rates of risk of psychotic relapse were calculated. Data on some gynecologic variables (postpartum blues, premenstrual tension, psychiatric symptoms triggered perimenstrually, mood symptoms while taking oral contraceptives) were collected. Clinical and psychosocial outcomes were measured by the Global Assessment Scale and Disability Assessment Scale.

Results: Patients who presented with major depression and bipolar affective disorder with psychotic features at the initial index episode showed overall diagnostic stability. Many patients with initial brief psychosis (cycloid psychosis) shifted to a clear bipolar affective disorder. The general risk of a psychotic relapse was high (previous psychosis = 0.77 vs first psychotic manifestation = 0.56; not significant). The risk after further pregnancies was 0.57 versus 0.48, respectively (not significant), and the risk regarding at least 1 other psychotic nonindex episode was 0.71 versus 0.44, respectively (P = .015). Gynecologic variables did not significantly discriminate between the groups. In some patients, a possible link to a hormonal susceptibility was discussed. Patients who remained without any further psychotic relapse (n = 24) had a favorable outcome.

Conclusions: Puerperal psychosis of an early onset seemed to be of a prevailing affective nature. Brief psychosis (cycloid psychosis) during a puerperal index episode showed a strong link to bipolar affective disorder in the further course of illness. Outcome was excellent in patients without a further psychotic relapse.