Table 2.

Summary of studies examining the relationship between prenatal stress, biomarkers of stress, and preterm birth

Study and country (region)	Design	Participants (n)	Measures			Results
			Scales	Specimen	Time, gestation (weeks)
Hobel et al. [81] USA (Los Angeles)	Prospective case–control studya	Subsample of 524 Cases: 18 (spontaneous onset of preterm labor) Control: 18 (delivered at term) Inclusion/exclusion criteria: not specified	PSS-8; STAIT-10	Plasma CRH, ACTH, cortisol	18–20 28–30 35–	Higher plasma CRH levels and ACTH levels were reported at all three time periods and elevated cortisol levels at 18–20 weeks’ gestation and 28–30 weeks’ gestation in women who delivered preterm when compared to those who delivered at term. Stress levels did not differ between 18–20 weeks’ gestation and 28–30 weeks’ gestation. Variance in CRH at 28–30 weeks’ gestation was explained by maternal stress level at 18–20 weeks’ gestation and maternal age.
Erickson et al. [82] Denmark (Odense)	Prospective case–control cohort designa	Subsample of 2,927 Cases: 84 (delivered preterm [idiopathic etiology] without complications) Control: 224 (delivered at term and matched, at time of enrollment, to within 10 days of due date of cases) Inclusion criteria: age >18 years, ability to understand Danish. Exclusion criteria: insufficient responses to the questionnaires, placental previa (diagnosed after 30 full gestational weeks), history of severe fetal congenital malformations in previous pregnancy, uterine cervix insufficiency treated with cervical circlage	Three questionnaires: (1) just before inclusion (past medical history); (2) 30 weeks’ gestation (social and demographic information); (3) birth; (urogenital and obstetric problems) If delivered preterm, completed second and third questionnaire at same time	Plasma CRH, CRH-binding protein, cortisol Venous blood sample taken during labor (delivered preterm), and 37–43 weeks’ gestation (delivered at term)	7–23 27–37	7–23 weeks: CRH and CRH-binding protein levels were higher in women who delivered preterm when compared to women who delivered at term. 27–37 weeks’ gestation: CRH and cortisol levels were higher but CRH-binding protein levels were lower in women who delivered preterm when compared to women who delivered at term. Previous preterm delivery and engagement in some risk-taking behaviors were associated with preterm birth
Ruiz et al. [83] USA (central Texas)	Prospective, longitudinal, observational study	Cases: 78 Inclusion: English speaking, <28 weeks’ gestational age, 18–40 years of age, singleton pregnancy. Exclusion criteria: Rh isoimmunization, cervical cerclage, use of tocolytic agents during current pregnancy, diabetes mellitus requiring insulin, thyroid disorders, chronic renal or heart disease, misses more than 1 monthly prenatal check for data collection	PSS-10 (23–26, and 31–35 weeks’ gestation)	Blood cortisol (all time points); vaginal swabs for fetal fibronectin, chlamydia, and bacterial vaginosis screen (23–26 and 27–30 weeks’ gestation)	15–19 20–22 23–26 27–30 31–35	Cortisol was a poor predictor of both preterm labor and preterm birth; however an increase in cortisol level was noted in women with genitourinary infection. Change is PSS score, that is decrease in perceived stress during the 2nd trimester, was significantly associated with increase in length of gestation
Mancuso et al. [84] USA (Los Angeles)	Case–control study nested in a prospective cohorta	Subsample of 688 Cases: 282 Inclusion criteria: singleton intrauterine pregnancy, gave birth to liveborn infant, received prenatal care in prenatal clinics and private practices in Los Angeles, California. Exclusion criteria: age <18 years, stillborn births, multiple gestation births, lack of birth outcome data, and incomplete psychosocial data	PSA	Plasma CRH	18–20 28–30	Women with high CRH levels and high maternal prenatal anxiety at 28–30 weeks gestation delivered earlier. CRH levels were significantly higher at both times points in women delivered preterm than women who delivered at term. Mediation effect of CRH
Kramer et al. [6] Canada (Montreal)	Prospective cohort and nested case–control design	Subsample of a larger study Cases: 207 Control: 444 Inclusion criteria: age ≥18 years, singleton gestation, and able to speak English or French. Exclusion criteria: severe chronic illness with ongoing treatment (note: other than hypertension, asthma, or diabetes), placenta previa, diagnosis of incompetent cervix in previous pregnancy, impending delivery, or fetus with congenital anomaly	DHS (lacked basic or essential needs subscale), MSS (chronic stress), AAS (conjugal violence), 5-item scale (injury, job related stress), MIS (intention of pregnancy), ASSIS (perceived social support), PLES (acute stressors), PSS, Dunkel-Schetter 4-item scale (pregnancy related anxiety), RSES, LOT (optimism and pessimism), CES-D, single item (woman’s perception of her risk of birth complications), 8-item scale (commitment to pregnancy)	Hair cortisol, histo-pathologic examination of vaginal swabs, placenta, and cord	24–26	Only pregnancy related anxiety was consistently and independently associated with spontaneous preterm birth and a dose–response was reported across quartiles. Hair cortisol was positively associated with gestational age but not CRH. Maternal plasma CRH, hair cortisol, placental histopathology (i.e., features of infection/inflammation, infarction, or maternal vasculopathy) were not associated with stress, anxiety, or distress measures
Pearce et al. [85] Denmark (Odense)	Case–control study nested in a prospective cohort	Subsample of 2,927 Cases: 60 [delivering preterm (<37 weeks) without a cause, as determined from clinical findings or laboratory investigations during pregnancy or at delivery] Control: 123 (delivering at term) Inclusion criteria: age >18 years, ability to understand Danish. Exclusion criteria: insufficient responses to the questionnaires, placental previa (diagnosed after 30 full gestational weeks), history of severe fetal congenital malformations in previous pregnancy, uterine cervix insufficiency treated with cervical circlage	Questionnaire (stressful life events, risk-taking behavior indicated by lack of seat-belt usage)	Serum measures of cortisol, MIF, CRP, CRH, interleukin-1 ß, interleukin-6, interleukin-10, tumor necrosis factor-alpha	<24	Individual biomarkers: MIF (strongest association), interleukin-10, CRP and tumor necrosis factor-alpha predicted preterm birth at various cutoff levels (e.g., 75th, 85th, and 90th percentile). Logistic regression models: MIF, CRP, risk-taking behavior, and low education consistently predicted preterm birth at various cutoffs; however, the 75th percentile cutoff was the best predictive model. MIF may be a psychobiological mediator
Shaikh et al. [55] Pakistan (Kirachi)	Prospective cohort study design	Cases: 132 (125 with complete data) Inclusion criteria: age 18–40 years, 28–30 weeks’ gestation. Exclusion criteria: diabetes mellitus, thyroid disorder, chronic renal or heart disease, or uterine and cervical abnormality, or on antidepressants or other psychotropic drugs, and did not deliver in setting where the study was based	A–Z Stress Scale, CES-D	Serum cortisol	28	A significant positive relationship was reported between maternal depression and stress. No relationship was noted between cortisol value and stress scale or depression scale. Preterm birth was associated with higher parity, past delivery of a male infant, and higher levels of paternal education

Adopted from Shaikh et al. [39]

AAS Abuse Assessment Screen (adapted), ACTH adrenocorticotropic hormone, ASSIS Arizona Social Support Interview Schedule, CES-D Centre for Epidemiology Studies Depression Scale, CRH corticotropin-releasing hormone, CRP C-reactive protein, DHS Daily Hassles Scale, LOT Life Orientation Test, MIF macrophage migration inhibitory factor, MIS Miller Intendedness Scale, MSS Marital Strain Scale of Pearlin and Schooler, PLES Prenatal Life Events Scale, PSA Pregnancy-Specific Anxiety Scale, PSS-8 Perceived Stress Scale 8-item version, PSS-10 Perceived Stress Scale 10-item version, RSES Rosenberg Self-Esteem Scale, STAIT Spielberger’s State Anxiety Inventory 10-item version

^aNot labelled