Maternal Immune Activation During Pregnancy and Obstetric Outcomes: A Population-Based Cohort Study

Abstract

Objective: Maternal immune activation has been proposed as a mechanism for adverse pregnancy outcomes, yet the mechanisms and effects of timing remain unclear. Immune disruption in early gestation may be particularly detrimental as this is an important period for placental development, which has been associated with the pathology of adverse obstetric outcomes. To increase our understanding of risk factors for adverse obstetric outcomes, we aim to investigate the association between multiple inflammatory and angiogenic markers during early pregnancy and adverse pregnancy outcomes in a large population-based cohort.

Design: Prospective population-based pregnancy cohort study (n = 7513).

Setting: Rotterdam, the Netherlands.

Population: Pregnant women in Rotterdam between April 2002 and January 2006.

Methods: Serum inflammatory markers (high-sensitivity (HS)-C-reactive protein (CRP), interleukin (IL)-1β, IL-6, IL-17a, IL-23, interferon (IFN)-γ) and angiogenic factors (sFlt-1 and PlGF) were analysed in repeated measures around 13-20 weeks gestation. A cytokine index was created using principal component analysis.

Main outcome measures: Hypertensive disorders of pregnancy, spontaneous preterm birth and small for gestational age at birth.

Results: HS-CRP, but not the cytokine index, was associated with increased risk of spontaneous preterm birth after multiple testing correction. We found no association of HS-CRP or the cytokine index with hypertensive disorders of pregnancy and small for gestational age at birth after multiple testing correction. Inflammatory and angiogenic factors were associated with each other, yet effect sizes were small.

Conclusions: We found no strong evidence of a link between early gestation typical inflammatory marker levels and the risk of adverse pregnancy outcomes.

Keywords: CRP; cytokines; maternal immune activation; preterm birth; pre‐eclampsia.

FIGURE 1

Study design, sample demographics and pregnancy outcomes. (A) Study design. Blood samples were collected at median 13 and 20 weeks of gestation in 7513 pregnant participants of the Generation R study in Rotterdam, the Netherlands. A total of 12 348 blood samples were collected. HS‐CRP and cytokines (IL‐1β, IL‐6, IL‐17A, IL‐23 and IFN‐γ) were measured and a cytokine index was created using principal component analysis. In addition, sFlt‐1 and PlGF were measured at both timepoints and were used to create the sFlt‐1/PlGF ratio. The first aim of the study was to assess the association between HS‐CRP/cytokine index and adverse pregnancy outcome (a composite outcome including hypertensive disorders of pregnancy, SGA and/or spontaneous PTB). Aim 2 was to investigate the association between HS‐CRP/cytokine index and the sFlt‐1/PlGF ratio. (B) Flowchart of participant selection. (C) Participant demographics maternal age, maternal BMI, parity (nullipara/multipara), national background (Dutch/non‐Dutch) and birth outcomes gestational age at delivery (weeks) and birthweight (grams). (D) Adverse pregnancy outcomes. The total number of participants with each outcome is shown with horizontal bars. The number of cases with combined adverse outcomes, as specified by the black dots, is shown with vertical bars.

FIGURE 2

(A) HS‐CRP across gestational age (weeks) compared between individual adverse pregnancy outcomes (pink) GHTN (n = 278), spontaneous PTB (n = 224) and PE without severe symptoms (n = 165), SGA (n = 431) compared to a reference group of healthy pregnancies (blue; n = 6505). (B) HS‐CRP across gestational age (weeks) compared between adverse pregnancy outcomes (pink; n = 1008) and healthy pregnancies (blue; n = 6505). Shaded areas indicate 95% confidence intervals. Plots were created using a generalised additive model (GAM) with thin‐plate regression splines.