Exposure to childhood maltreatment and systemic inflammation across pregnancy: The moderating role of depressive symptomatology

Abstract

Background: Childhood maltreatment (CM) has long-term consequences for dysregulation of the immune system which is particularly pronounced when mental and physical health sequelae have manifested. Higher proinflammatory state has been shown in non-pregnant state in association with CM as well as with depression, one of the most frequent and pernicious psychiatric sequelae of CM. During pregnancy, however, this association is less clear. Given the important role of maternal inflammatory state during pregnancy for fetal, pregnancy, and birth outcomes, we sought to examine the association between CM and proinflammatory state during pregnancy considering the moderating role of maternal depressive symptoms characterized serially across pregnancy.

Methods: A prospective, longitudinal study of 180 healthy pregnant women was conducted with serial assessments in early (12.98 ± 1.71 weeks gestation), mid (20.53 ± 1.38 weeks gestation) and late (30.42 ± 1.4 weeks gestation) pregnancy. Maternal history of CM was assessed with the Childhood Trauma Questionnaire (CTQ) and the total score was used as an indicator of CM experience. Maternal depressive symptoms were assessed at each pregnancy visit with the Center for Epidemiologic Studies Depression Scale (CES-D). Serum concentrations of tumor necrosis factor (TNF)-α and interleukin (IL)-6 were obtained at each pregnancy visit and combined to a composite maternal proinflammatory score. Linear mixed effects models were employed to assess the association between CTQ score, CES-D score, and proinflammatory score during pregnancy, adjusting for potential confounders.

Results: Gestational age was associated with the proinflammatory score (B = 0.02; SE = 0.00; p < .001), indicating an increase in inflammation across gestation. Neither CTQ score nor depressive symptoms were independently associated with the proinflammatory score (ps > 0.28). However, the interaction between CTQ score and depressive symptoms was associated with the proinflammatory score (B = 0.03, SE = 0.01, p < .05), indicating higher inflammation across pregnancy with increasing levels of depressive symptoms during pregnancy in women with higher CTQ scores. Exploratory analyses suggested that this interaction was mainly driven by CTQ subscale scores assessing experiences of abuse rather than neglect.

Conclusions: These findings suggest a moderating role of maternal depressive symptoms during pregnancy on the association of early life stress with inflammation and thus highlight the importance of the timely assessment of both CM exposure and depressive symptoms which might allow for the development of targeted and individualized interventions to impact inflammation during pregnancy and to ameliorate the detrimental long-term effects of CM. The current findings add to a better understanding of the prenatal biological pathways that may underlie intergenerational transmission of maternal CM.

Keywords: Childhood maltreatment; Depression; Inflammation; Pregnancy.

Figure 1.

Association between total CTQ score and pregnancy mean CES-D score. For illustration purposes, a cut-off of 20 or higher was applied to the CES-D scores to indicate clinically relevant depressive symptoms [59].

Figure 2.

Association between repeatedly-measured proinflammatory scores and gestational age in weeks.

Figure 3a.

Association between pregnancy mean proinflammatory score and pregnancy mean CES-D score in women without vs. with CM experiences. For illustration purposes, a cut-off for moderate CM experiences was used. Note. CM+ = moderate to severe score (EA ≥ 13, PA ≥ 10, SA ≥ 8, EN ≥ 15, and PN ≥ 10) on at least one CTQ subscale; CM- = scores below moderate cut-off on all CTQ subscales.

Figure 3b.

Pregnancy mean proinflammatory score in women without vs. with CM experiences and none vs. any clinically relevant depressive symptoms [59]. Note. CM+ = moderate to severe score (EA ≥ 13, PA ≥ 10, SA ≥ 8, EN ≥ 15, and PN ≥ 10) on at least one CTQ subscale; CM- = scores below moderate cut-off on all CTQ subscales.