(265) Characteristics and Adverse Postpartum Psychiatric Outcomes in Women With Bipolar Disorder Who Discontinue Lithium During Pregnancy: A Swedish Cohort Study

Background: Lithium treatment during pregnancy may be continued or discontinued depending on individual assessment of the benefits of psychiatric relapse prevention against the risks associated with fetal exposure. However, there are inconsistencies between guidelines for management of bipolar disorder during pregnancy.

Objectives: To determine whether discontinuation of lithium near or during pregnancy in women with bipolar disorder affects postpartum psychiatric health.

Methods: In this Swedish population-based cohort study, pregnant women with bipolar disorder and pre-pregnancy lithium use were identified. Natural language processing of the prescription free-text information was used to identify daily dose, and with the number of pills per prescription fill, treatment episodes were constructed. Women with an ongoing lithium treatment episode at six months before the last menstrual period were followed and classified either as ‘discontinuers’ and ‘continuers’. Clinical characteristics of the continuers and discontinuers were described, as well as the crude relative risks of adverse psychiatric outcomes in the 90 days postpartum.

Results: Of 662 women, 438 (70%) discontinued lithium treatment, the majority pre-pregnancy or in the 1st trimester. Discontinuers, compared with continuers, had a higher prevalence of comorbid recurrent depressive disorder (7.1% vs. 3.8%), substance use disorder (7.8% vs. 2.7%), personality disorder (11.4% vs. 6.0%), autism spectrum disorder (3.0% vs. 1.1%), attention deficit hyperactivity disorder (13.5% vs. 7.6%) and recent history of suicide attempt (3.4% vs. 0.5%). Among discontinuers, there was no indication of increased use of other mood stabilizers during pregnancy. In the 90 days postpartum, discontinuers had a similar risk of psychiatric hospitalization compared to continuers (unadjusted risk ratio (RR) 0.82, 95% confidence interval (CI) 0.50-1.35); the risk of hospitalization due to psychosis was approximately three times higher (RR 3.36, 95% CI: 0.42-26.67); and the risk of hospitalization due to depression or anxiety was almost halved (RR 0.42, 95% CI: 0.06-2.96).

Conclusions: A large majority of women with bipolar disorder discontinue their lithium treatment around the start of pregnancy. Discontinuers, compared with continuers, have a higher burden of psychiatric co-morbidity. Preliminary results indicate an elevated crude risk of postpartum hospitalization due to psychosis in lithium discontinuers. Future plans include incorporating data from other Scandinavian countries to increase the number of women in the cohort.