First-Onset Psychosis Occurring in the Postpartum Period: A Prospective Cohort Study

Objective: To prospectively characterize a cohort of patients for whom first lifetime episode of psychosis occurs in the postpartum period. Method: Included in the study were 51 women admitted to an inpatient facility for postpartum psychosis and a population-based control group (n = 6,969). All patients received naturalistic treatment using the sequential addition of benzodiazepines, antipsychotics, and lithium. A clinician-administered questionnaire and parallel history provided information about obstetric history, pregnancy, delivery, breastfeeding, neonatal outcomes, and onset of the disease. Clinical remission was defined as the absence of psychotic, manic, and depressive symptoms for at least 1 week, The primary outcome measure was the Clinical Global Impressions-Severity scale. The study was conducted from 2005 to 2009. Results: Compared to the general population sample, women with postpartum psychosis had a significantly higher incidence of primiparity (OR = 2.90; 95% CI, 1.49-5.67) but had no significant differences in delivery-related, lactational, or neonatal-related risk factors. The median onset of psychiatric symptoms occurred at 8 days' postpartum (interquartile range [IQR], 5-14), and median duration of episode was 40 days (IQR, 23-69). Patients with prominent depressive symptoms had a significantly later onset (P = .01) of psychosis and a longer duration of episode (P < .01) than patients without depressive symptoms. Psychotic symptoms were mood-incongruent in 64.7% of patients. Conclusions: In contrast to other findings related to postpartum psychosis in bipolar patients, no delivery-related, neonatal-related, or lactational risk factors could be identified. Further, our findings of a delayed onset and mood incongruence of postpartum psychotic symptoms markedly contrasts with that of patients with a previous history of bipolar disorder. These results suggest that women with psychosis limited to the postpartum period might have a distinct risk profile and phenomenology.