Importance: Clinical decision-making on antidepressant treatment during pregnancy, particularly selective serotonin reuptake inhibitors (SSRIs), is challenging, as both prenatal SSRI exposure and maternal depressive symptoms may be associated with negative outcomes in offspring.
Objective: To investigate the association between intrauterine SSRI exposure and maternal depressive symptoms and structural brain development in offspring from mid-childhood to early puberty.
Design, setting, and participants: This prospective, population-based cohort study was embedded in the Generation R Study in Rotterdam, the Netherlands. All pregnant individuals with an expected delivery date between April 1, 2002, and January 31, 2006, were invited to participate. Data were analyzed from February 1 to September 30, 2022.
Exposure: Maternal-reported SSRI use verified by pharmacy records. In mid-pregnancy and 2 and 6 months after delivery, participants reported depressive symptoms using the Brief Symptom Inventory and were divided into 5 groups: SSRI use during pregnancy (n = 41; 80 scans), SSRI use only before pregnancy (n = 77; 126 scans), prenatal depressive symptoms without prenatal SSRI use (n = 257; 477 scans), postnatal depressive symptoms only (n = 74; 128 scans), and nonexposed control individuals (n = 2749; 4813 scans).
Main outcomes and measures: The main outcome was brain morphometry in offspring, including global and cortical brain volumes, measured at 3 magnetic resonance imaging assessments from 7 to 15 years of age.
Results: The study included 3198 mother-child dyads. A total of 3198 mothers (100%) identified as women; mean (SD) age at intake was 31.1 (4.7) years. Children (1670 [52.2%] female) underwent brain imaging assessment from 7 to 15 years of age with 5624 total scans. Most brain gray matter volumes showed an inverted U-shaped trajectory. Compared with nonexposed controls, children prenatally exposed to SSRIs had less cerebral gray matter (β [SE], -20 212.2 [7285.6] mm3; P = .006), particularly within the corticolimbic circuit, which persisted up to 15 years of age. Children exposed to SSRIs prenatally showed a steeper increase in volumes of the amygdala (age interaction: β [SE], 43.3 [13.4] mm3; P = .006) and fusiform gyrus (age interaction: β [SE], 168.3 [51.4] mm3; P = .003) from 7 to 15 years of age. These volumetric differences in the amygdala and fusiform observed in childhood did not persist until early adolescence. Prenatal depression was associated with a smaller volume in the rostral anterior cingulate gyrus (β [SE], -166.3 [65.1] mm3; P = .006), and postnatal depression was associated with a reduced fusiform gyrus (β [SE], -480.5 [189.2] mm3; P = .002). No association of SSRI use before pregnancy with brain outcomes was observed.
Conclusions and relevance: The results of this cohort study suggest that prenatal SSRI exposure may be associated with altered developmental trajectories of brain regions involved in emotional regulation in offspring. Further research on the functional implications of these findings is needed.