Associations between prenatal and postnatal substance exposure and salivary C-reactive protein in early childhood

Highlights

• Pre and postnatal child tobacco and cannabis exposure were examined longitudinally

• Exposures were assessed via self-report, meconium, and salivary measures

• Prenatal tobacco and cannabis co-exposure was associated with child salivary CRP

• Child salivary CRP was 43% lower when meconium indicated prenatal co-exposure

• In boys, postnatal cannabis exposure was associated with higher inflammatory levels

Abstract

Background

Exposure to tobacco and cannabis during developmental periods of enhanced vulnerability (e.g., in utero and early childhood) may have long-lasting effects on child health. One potential mechanism underlying these associations is the alteration of inflammatory pathways. Using data from a longitudinal study of mother/child dyads, we examined the adjusted and combined associations of prenatal and postnatal tobacco and cannabis exposure with inflammation in early childhood. Furthermore, we explored the relations between different measures of exposure, partly reflecting differences in timing, dose, and level of fetal exposure (e.g., self-report vs. biomarker), and inflammation. Finally, we explored child sex as a moderator of prenatal and postnatal tobacco and cannabis exposure and inflammation.

Method

Women were recruited from a local hospital during their first prenatal appointment. Repeated assessments were conducted at each trimester, at birth, and when children were 2, 9, 16, 24, 36, and 60 months old (N = 215; 112 female children). To evaluate associations with different measurement approaches, prenatal tobacco and cannabis exposure were assessed using: 1) continuous dose-response variables of maternal self-reported tobacco and cannabis use during each trimester to assess associations with timing and severity of exposure, 2) categorization of children into exposure groups based on drugs and metabolites present in infant meconium reflecting later pregnancy fetal exposure, and 3) categorization into exposure groups using a combination of maternal self-report data and biomarker data derived from maternal saliva samples and infant meconium taking advantage of multiple methods of assessment to examine group differences. Postnatal exposure to tobacco (assessed using child salivary cotinine) and cannabis (assessed using maternal self-reported average joints smoked per day) was measured at each infancy/early childhood assessment. Adjusted pre- and postnatal exposure associations with child inflammation were assessed by including both measures as predictor variables in linear regression models predicting child salivary C-reactive protein (CRP) concentrations at 60 months. Interactions between pre- and postnatal exposure variables were then modeled to investigate the combined relations between pre- and postnatal substance exposure with child salivary CRP concentrations at 60 months.

Results

Adjusting for postnatal exposure variables, there was a significant interaction between the average daily cigarettes and the average daily cannabis joints smoked during the third trimester predicting salivary CRP concentrations in early childhood. At high tobacco exposure, the effect of cannabis on CRP concentrations was negligible, whereas at low tobacco exposure, the effect of cannabis exposure on CRP concentrations was positive. Adjusting for postnatal tobacco and cannabis exposure, children for whom meconium data indicated co-exposure to tobacco and cannabis showed approximately 43% lower CRP concentrations at age 60 months compared to children with no exposure. However, when mother/child dyads were categorized based on a combination of maternal self-report data and biomarker data from saliva samples and infant meconium, there were no differences in salivary CRP concentrations at age 60 months across the three groups (no prenatal exposure, prenatal tobacco exposure only, prenatal co-exposure to tobacco and cannabis), controlling for postnatal associations. Regardless of the measurement method used to assess prenatal exposures in adjusted analyses, prenatal tobacco exposure alone did not predict CRP concentrations in early childhood, nor did postnatal tobacco exposure. Among boys, postnatal cannabis exposure was associated with higher concentrations of CRP at age 60 months, controlling for prenatal exposure relations. There were no significant combined associations of pre- and postnatal exposure with CRP concentrations.

Conclusion

This study expands upon known relations between prenatal and postnatal substance exposure and immunological outcomes in early childhood, underscoring the importance of assessing cannabis exposure during gestation and early life in combination with tobacco exposure.

Introduction

In the United States, one in 14 women use either tobacco or cannabis during pregnancy¹ (Drake et al., 2016; Young-Wolff et al., 2017). Frequently, tobacco and cannabis are used in combination, with more women reporting co-use of tobacco and cannabis during pregnancy than cannabis alone (Coleman-Cowger et al., 2018). Co-use of cannabis and tobacco during pregnancy presents significant risks to the fetus, as ∆9-tetrahydrocannabinol (THC; the main psychoactive compound in cannabis) and nicotine cross the placenta (Blackard, 1984; Napierala et al., 2016; Perez-Reyes and Wall, 1982). THC concentrations in fetal blood are comparable to those in maternal blood and nicotine concentrations in fetal blood are typically 15% higher than maternal blood (Lambers and Clark, 1996; Thompson et al., 2019). Substantial research has demonstrated that exposure to tobacco and cannabis in utero can negatively affect neural function and health outcomes (for reviews, see Forray, 2016 and Gunn et al., 2016). More recently, this line of research has expanded to include the effects of continued exposure to tobacco and cannabis during the postnatal period (Luk et al., 2018).

Exposure to toxins in utero and in early childhood is particularly harmful for developmental outcomes (Falck et al., 2015), as these are periods of enhanced vulnerability when environmental exposures may have long-lasting effects on child health. The Fetal Origins of Health and Disease Model (Barker, 1990) and the Biological Embedding Model (Miller, Chen, & Parker, 2011) highlight the critical role of prenatal and early childhood adversity on children's health outcomes. For example, tobacco and cannabis use during pregnancy affects the maturation of the immune system (Dietert and Zelikoff, 2008; Dong et al., 2019). Tobacco exposure in utero causes reductions in anti-inflammatory markers (e.g., IL-10) and increases in pro-inflammatory markers (e.g., IL-6 and IL-1β) (Arnson et al., 2010). Similarly, research suggests that prenatal cannabis exposure contributes to dysregulation of the innate and adaptive immune systems (Dong et al., 2019; Molnar et al., 2018; Lombard et al., 2011; Zumbrun et al., 2015). Mothers who use substances during pregnancy are likely to continue use after childbirth (Kahn et al., 2002). Postnatal exposure in early life could also contribute to poor immune outcomes in children, given the potential for passive exposure from tobacco and cannabis. Postnatally, passive tobacco smoke exposure has been associated with higher concentrations of C-reactive protein (CRP; a protein involved in the systemic inflammatory response) in children and youth (Kang et al., 2017; Nagel et al., 2009; Wilkinson et al., 2007; Azar and Richard, 2011). Importantly, these studies did not account for prenatal tobacco use.

Despite research demonstrating associations between prenatal and postnatal substance exposure and immunological outcomes (Arnson et al., 2010; Dong et al., 2019; Kang et al., 2017) and studies suggesting that continued postnatal adversity may exacerbate the effects of prenatal adversity (Monk et al., 2012; Raineki et al., 2017), research on the effects of prenatal substance exposure rarely adjust for prolonged exposure postpartum, and vice versa. Similarly, there is a lack of research on co-exposure to both prenatal and postnatal tobacco and cannabis (DeGenna et al., 2019). The few studies that do account for polysubstance exposure often employ retrospective designs and lack robust measures of exposure (e.g., methodological issues related to recall bias) (Gesterling and Bradford, 2022). To address these gaps, the current study assesses prenatal tobacco and cannabis exposure using three different methods of assessment, given the unique information provided by each approach and the importance of multi-method assessments of substance exposure (e.g., Shisler et al., 2017).

Furthermore, a growing body of research examining sex-differential programming in utero suggests that male fetuses may be more vulnerable to adverse fetal conditions affecting immune system development than female fetuses (Goldenberg et al., 2006; Klein and Flanagan, 2016). There are also significant sex-related differences in the development and function of the endocannabinoid system (Craft et al., 2013), and early evidence suggests increased vulnerability to poor health outcomes among males prenatally exposed to cannabis relative to females (Bara et al., 2018; Benevenuto et al., 2017). Similarly, recent research has found that males, but not females, show impaired hypothalamic-pituitary-adrenal (HPA)-axis regulation following co-exposure to tobacco and cannabis in utero (Stroud et al., 2020). Due to the bidirectional communication network between the HPA-axis and immune system, we may expect similar increased vulnerabilities in immune system development among males following prenatal exposure to tobacco and cannabis exposure. However, no study to date has investigated sex-related differences in immunological outcomes following pre- and postnatal tobacco and cannabis exposure.

We employed data from a longitudinal study of mother/child dyads to address gaps in our understanding of 1) the associations of prenatal and postnatal tobacco and cannabis exposure with inflammation in early childhood, 2) the associations following co-exposure to tobacco and cannabis on inflammation in early childhood, 3) different methods of measuring exposure to examine prenatal timing and dose-response relations, associations with fetal exposure, and overall group differences and 4) the role of biological sex as a moderator of pre- and postnatal substance exposure relations with inflammation in early childhood. Specifically, we examined the associations of prenatal substance exposure (tobacco alone or co-exposure to tobacco and cannabis) and postnatal substance exposure (tobacco and cannabis) with child salivary CRP concentrations assessed during kindergarten. For each measure of prenatal substance exposure, we assessed the relation with CRP for tobacco use alone and for co-use of tobacco and cannabis during the prenatal period as well as the adjusted and combined relations of pre- and postnatal exposure. Finally, we examined the extent to which the relations linking pre- and postnatal tobacco and cannabis exposure with child CRP concentrations differed by child biological sex. We expected that: 1) children with prenatal tobacco exposure, with a higher dose of tobacco exposure, and those who were positive for nicotine analytes in meconium (a biomarker of third trimester fetal exposure also reflecting persistent smoking in pregnancy), would display higher CRP concentrations in early childhood compared to children with no prenatal substance exposure, adjusting for postnatal substance exposure; 2) children co-exposed to tobacco and cannabis pre- or postnatally, and to higher doses of the substances, and with meconium positive for analytes of both tobacco and cannabis, would display higher CRP concentrations than those with no exposure; and 3) the associations of both pre- and postnatal exposure to tobacco and cannabis with early childhood CRP concentrations would be stronger among boys compared to girls.