Measures of Maternal Stress and Mood in Relation to Preterm Birth

Abstract

OBJECTIVE To investigate the relationships between psychological and physiologic measures of stress, mood and gestational age at delivery and preterm birth (PTB). METHODS This prospective cohort study recruited healthy women in the early second trimester, 18–45 years of age. Validated psychological measures of perceived stress, depressive symptoms, and anxiety were completed at 16, 22, 28, 34 and 40 weeks of gestation. Cortisol concentration was measured in maternal hair at 16, 28, and 40 weeks of gestation to approximate 1st, 2nd, and 3rd trimester levels of physiologic stress. Statistical methods included: analyses of variance (ANOVA), t-tests, Chi-square, Pearson correlations, regression modeling and mediation analysis as appropriate. Hair cortisol concentrations were natural log transformed to normalize values. RESULTS Eleven (12%) out of 90 included women had a spontaneous preterm birth or preterm premature rupture of the membranes. Perceived stress at 16 weeks of gestation correlated with both 2nd trimester cortisol concentration (r= 0.28, p=0.007) and earlier gestational age at delivery (r= −0.30, p< 0.01). Gestational age at delivery was also negatively correlated with cortisol concentration in the 2nd trimester (r= −0.25, p=0.02) and 2nd trimester cortisol concentration was higher in preterm (2.7 ± 0.4 LN pg/mg) vs term (2.0 ± 0.7 LN pg/mg, P<0.001) delivered women. Using mediation statistics, the association between the psychological measure, the physiologic measure, and gestational age at delivery was mainly driven by increased physiologic stress (hair cortisol concentration) in the 2nd trimester (difference in coefficients (standard error)= −0.05(0.02)),. CONCLUSION Higher perceived stress in the 2nd trimester is associated with both elevated 2nd trimester hair cortisol concentration and gestational age at delivery. Physiologic measure of stress in the 2nd trimester appears most strongly associated with preterm birth. Identification and amelioration of early pregnancy stressors may attenuate physiologic stress and ultimately affect preterm birth.

Childhood stress and birth timing among African American women: Cortisol as biological mediator

“…While we controlled for stress in adulthood, it is possible that pregnancy itself serves as a “stressor.” This is an interesting concept that may be best addressed through inclusion of non-pregnant comparison groups in future studies. We also provide preliminary support for the accelerated “placental clock” theory, which is consistent with some (Guendelman et al, 2008; Hobel et al, 1999; Hoffman et al, 2016; Mancuso et al, 2004; Moog et al, 2016), but not all (Himes and Simhan, 2011; Karakash et al, 2016; Kramer et al, 2009; Kramer et al, 2013; Noll et al, 2007; Owen et al, 2017), prior studies. It may be that inconsistencies are largely related to the considerable variability in the timing, duration, and nature of the stressors assessed.…”

“…However, HPA output can be measured in numerous ways (e.g., CRH, ACTH, plasma cortisol, salivary cortisol, hair cortisol). Markers tend to correlate well during pregnancy (e.g., Glynn et al, 2007; O’Keane et al, 2011), with each linked to early birth in prior work (i.e., serum and plasma CRH (Guendelman et al, 2008; Hobel et al, 1999; Mancuso et al, 2004; Moog et al, 2016) plasma ACTH (Hobel et al, 1999), plasma cortisol (Hobel et al, 1999), diurnal salivary cortisol pattern (Patacchioli et al, 2013), hair cortisol (Hoffman et al, 2016)). Given our interest in the peripheral actions of cortisol, use of plasma as opposed to saliva may also be considered a strength given the notable individual variability in passive salivary diffusion and rate of conversion of active cortisol to inactive cortisone by type II 11β-hydroxysteroid dehydrogenase (Perogamvros et al, 2010).…”

“…With a 28–32 week sampling window, we witnessed a negligible effect of gestational age on plasma cortisol, which may have increased our ability to link individual variation at a single time point to birth timing. In longitudinal studies, various HPA markers in the 2 nd and 3 rd trimesters provide some predictive power for birth timing (e.g., Hobel et al, 1999; Hoffman et al, 2016). Additional work is required to determine which HPA markers and at which time points, including potentially the rate of rise, are most useful in risk prediction.…”

“…Of studies assessing effects of stress in adulthood on HPA activity and birth timing, formal, prospective testing of biological mediation is rare (Hoffman et al, 2016; Mancuso et al, 2004). As such, it is not surprising that some (Guendelman et al, 2008; Hobel et al, 1999; Hoffman et al, 2016; Mancuso et al, 2004; Moog et al, 2016), but not all (Himes and Simhan, 2011; Karakash et al, 2016; Kramer et al, 2009; Kramer et al, 2013; Noll et al, 2007; Owen et al, 2017), studies provide support for the “placental clock” theory. The two studies formally testing mediation are in favor of the proposed pathway.…”

Development of a culturally informed protocol for hair cortisol sampling in Black women

“…Participants were notified of the option to receive their own cortisol results if desired. Women who gave written consent were escorted individually to a private room, where the cosmetologist asked the participant about her usual hair care routine using a questionnaire and recorded the answers. The cosmetologist assessed the hairstyle to determine the most cosmetically optimal sampling site within the posterior vertex target zone and hair was sampled following standard hair cortisol collection procedures (Hoffman, Mazzoni, Wagner, Laudenslager, & Ross, ). During the procedure, the cosmetologist informed the participant of each step. After sampling, the cosmetologist restyled the participant's hair to its original condition. Three weeks after sampling, participants received an individualized hair analysis from the cosmetologist with recommendations for products and practices to enhance hair care.…”