Prenatal Maternal Depression and Neonatal Immune Responses

Abstract

Objective: The aim of the study was to examine the association of lifetime maternal depression with regulation of immune responses in the infant, measured by cytokine levels and lymphocyte proliferation (LP) in cord blood mononuclear cells collected at delivery.

Methods: We studied women recruited in early pregnancy into the Project Viva longitudinal cohort who had cord blood assayed after delivery (N = 463). Women reported about depressive symptoms in midpregnancy (Edinburgh Postnatal Depression Scale) and depression history by questionnaire. Immune responses were assayed by an index of LP, and concentrations of five cytokines (interleukin [IL]-6, IL-10, IL-13, tumor necrosis tumor necrosis factor factor α, and interferon γ) after incubation of cord blood mononuclear cells either in medium alone or stimulated with phytohemagglutinin (PHA), cockroach extract, or house dust mite extract. We examined associations of maternal depression with these sets of cytokine measures using multivariable linear or tobit regression analyses.

Results: After adjustment for confounders (mother's age, race/ethnicity, education, household income, season of birth, and child sex), levels of IL-10 after stimulation with cockroach or dust mite allergen were lower in cord blood from ever versus never depressed women, and a similar trend was evident in IL-10 stimulated with PHA (percentage difference: cockroach extract = -41.4, p = .027; house dust mite extract = 1-36.0, p = .071; PHA = -24.2, p = .333). No significant differences were seen in levels of other cytokines or LP.

Conclusions: Maternal depression is associated with offspring immune responses at birth, which may have implications for later life atopic risk or immune function.

Figure 1:

Percent difference (95% CI) in concentration of IL-10 in children of ever vs. never depressed women, for unstimulated cord blood lymphocytes or for those stimulated with PHA, cockroach antigen, or dust mite antigen. Outcomes were log transformed for analysis and exponentiated to arrive at % difference. Unstimulated IL-10 was analyzed using tobit regression; others were analyzed using linear regression. Models adjusted for maternal age, race/ethnicity, and education; household income; and child sex and season of birth. Sample sizes: 442 (unstimulated, PHA), 368 (cockroach antigen), 362 (dust mite antigen).

Figure 2:

Percent difference (95% CI) in lymphocyte SI or cytokine concentration in children of ever vs never depressed women. Outcomes were log transformed for analysis and exponentiated to arrive at % difference. IFN-γ was analyzed using tobit regression; others were analyzed using linear regression. Models adjusted for maternal age, race/ethnicity, and education; household income; and child sex and season of birth. For sample sizes for each analysis, see Figure, Supplemental Digital Content 1).

Figure 3:

Percent difference (95% CI) in concentration of IL-10 for all depression measures, from unstimulated cord blood samples or for those stimulated with PHA, cockroach antigen, or dust mite antigen. Outcomes were log transformed for analysis and exponentiated to arrive at % difference. Unstimulated IL-10 was analyzed using tobit regression; others were analyzed using linear regression. Models adjusted for maternal age, race/ethnicity and education; household income; and child sex and season of birth. For sample sizes for each analysis, see Figure, Supplemental Digital Content 1.