(332) Antenatal depressive and anxiety symptoms in women with eating disorders who continue or discontinue antidepressant treatment in pregnancy - results from the Norwegian Mother, Father, and Child Cohort Study

Background: Lack of evidence about the effectiveness of antidepressants in pregnant women with eating disorders (ED) poses challenges for pharmacotherapy management.

Objectives: To investigate the effectiveness of antidepressant continuation into early and mid-late pregnancy on depression and anxiety symptom severity in pregnant women with ED, overall and by individual Polygenic Scores (PGS) for antidepressant response and psychiatric illnesses.

Methods: We used data from Norwegian Mother, Father and Child Cohort Study (MoBa) linked to Medical Birth Registry of Norway and MoBa genetics. MoBa questionnaires Q1 (week 17) and Q3 (week 30) measured the type of ED and exposure to antidepressants six months before and throughout pregnancy. We assessed symptoms of depression and anxiety using the Symptom Checklist scale (SCL5 and its depression and anxiety subscales) in Q1 and Q3 as numeric outcomes. We fit crude and weighted generalised linear models and adjusted for measured confounding via overlap weights based on propensity scores. In a subsample, we stratified our analyses by PGS tertials for antidepressant response, depression, anxiety and ED separately. Results are presented as mean differences in symptom scores with 95% Confidence Intervals (CI).

Results: We identified 5920 pregnancies with ED: 1.4% Anorexia, 12.4% Bulimia, 84.0% Binge Eating Disorder (BED), 1.5% recurrent self-induced purging in the absence of binge eating and 0.7% unspecified type. Of these, 96.2% did not use antidepressants. Of the pregnancies exposed to antidepressants pre-pregnancy, 1.4% (n=85) discontinued treatment before and 1.6% (n=93) during pregnancy, while 0.8 % (n=48) continued during pregnancy. Pregnancies with continued antidepressant had lower depressive and anxiety symptoms compared to discontinued use before or during pregnancy, albeit the effect size was small. Among women with BED, this association was specific to lower anxiety symptoms (β: -0.44 (-0.85, -0.03)). When stratifying by the PGS (total subsample n=2272), there were substantial differences in depression and anxiety symptoms (range of β: -0.44 to -1.37) with continued use relative to discontinued in women having greater PGS for antidepressant response.

Conclusions: In pregnancies within women with ED, continued antidepressant in pregnancy is associated with reduced antenatal depressive and anxiety symptom severity, compared to discontinued treatment before or during pregnancy, albeit the effect size is small. The observed reduction in symptom severity is, however, substantial in women having higher genetic susceptibility towards antidepressant response, making important advances towards personalised pharmacotherapy in pregnancy.