Maternal variables associated with physiologic stress and perinatal complications in preterm infants

1 Introduction

Preterm birth is a major challenge in perinatal health care. The World Health Organization defines any infant born before 37 weeks of gestation as preterm [1], with further categorization into gestational age (GA) at birth (extremely preterm, 25 weeks; very preterm, 26–33 weeks; and late preterm, 34–36 weeks) for clinical and research analyses [2]. In the United States, one in eight babies (12%) is born prematurely; of these, 80% are born at 32–36 weeks gestation and 20% are born at 22–32 weeks [3]. Very preterm infants (<32 weeks GA) are at increased risk for neonatal mortality and morbidity [1, 2]. The most common complications leading to mortality and morbidity in preterm infants are bronchopulmonary dysplasia (BPD), intraventricular hemorrhage (IVH), retinopathy of prematurity (ROP), and necrotizing enterocolitis (NEC) [2].

BPD, or chronic lung disease, affects 25–42% of preterm infants in the United States, and involves progressive lung abnormalities such as pulmonary hypertension and abnormal vascular development. Severity of BPD is characterized by requirement of oxygen at 28 days of life and 36 weeks corrected gestational age, and changes on radiographic imaging, and is often associated with inflammatory response and stress mechanisms. IVH, which affects 30–60% of preterm infants, is characterized as injury to the white matter of the brain due to a microvascular event occurring in the germinal matrix, and puts the infant at increased risk of neurodevelopmental delays and additional brain injury. ROP is in ocular disease in which abnormal vascular growth in the retina can lead to vision impairments or blindness, and affects 7–15% of preterm infants. NEC is the leading cause of morbidity and mortality in preterm infants, with a mortality rate as high as 30% and increased risk of neurodevelopmental delays and liver or bowel transplant. NEC involves intestinal necrosis, signs and symptoms of shock, and pneumoperitoneum, and may be associated with stress and inflammatory response [4].

Moore, Berger and Wilson [4] developed a theoretical model of allostatic load as a mechanism for complications of prematurity, positing that these complications exhibit a common theme of physiologic dysregulation. Their model was derived from McEwen’s allostatic load model in adult disease, where allostasis involves the physiologic response patterns used to maintain homeostasis during change, and allostatic load is defined as the “wear and tear” on the body systems from collective allostasis [5]. Moore’s model suggests that the stress of prematurity leads to allostasis as the infant struggles to use functionally and structurally immature body systems for physiologic regulation. Their efforts to maintain a consistently threatened homeostasis lead to dysregulated physiologic coping mechanisms and increased allostatic load. As the infant’s allostatic load increases and complications develop, the infant becomes even more physiologically stressed and their allostatic load increases further [4].

Complications of prematurity may arise due to disruptions in the hypothalamic-pituitary-adrenal (HPA) axis that prevent the infant from mounting adequate physiologic responses to the stress of being born too early. Cortisol is known to modulate the maturation of endocrine and neuroendocrine systems in the developing infant, including intestinal enzymatic and transportation activity and surfactant production in the lungs [6, 7]. Normally, the infant experiences a surge of cortisol at birth. However, cortisol levels are found to be lower in preterm infants than term infants [4]. These dysregulated cortisol levels significantly increase the risk of complications, including BPD [8], NEC [9], and mortality.

Dysregulation of cortisol in preterm infants has been linked to prenatal maternal cortisol levels. Increased intrauterine exposure to maternal cortisol may lead to adverse prenatal programming of the infant HPA axis [10] and a dysregulated stress response after birth [11]. Excess maternal cortisol may be related to maternal stress from various social variables. Specifically, socioeconomic status, financial stress, and exposure to negative life events have been found to increase individual stress and can be predictive of poor health outcomes in these individuals [12]. A significant association between adverse psychosocial factors experienced by the mother and disruptions in HPA axis hormones adrenocorticotropic hormone (ACTH), corticotropin releasing hormone (CRH) and cortisol; as well as a correlation between CRH and preterm birth, has been demonstrated in previous research [13]. These findings suggest a complex relationship between maternal stress, the HPA axis, and preterm birth. Although excess prenatal exposure to maternal cortisol in the intrauterine environment has been associated with adverse outcomes for infants later in life [11], little research has examined the association of maternal stress and cortisol with complications in the first few weeks of life.

The present study will examine whether maternal social variables are associated with inadequate HPA axis response on day of life 1 and development of common complications of prematurity in very preterm infants. The present study included maternal social variables (marital status, employment status, insurance coverage, smoking habits), prenatal care, infant outcomes (NEC, BPD, IVH, ROP), and cortisol, as a biomarker of physiologic stress at birth in preterm infants (<32 weeks’ GA). Specific aims of the study are to compare maternal social variables and prenatal care and examine 1) their association with cortisol levels on day of life 1; and 2) the development of complications of prematurity in the preterm infant by discharge.

2 Methods

A prospective, correlational design was used. Approval was received from the Institutional Review Board. A convenience sample of very preterm infants was recruited from a level III NICU in a Midwestern US tertiary medical center. Infants <32 weeks GA at birth were included. Infants with major congenital anomalies were excluded. A parent of the eligible infant was approached by research personnel after admission to the NICU and within 24 hours of birth. The infant was enrolled after parental consent was obtained, and the day of consent was regarded as day 1 of the study.

3 Data analysis

Statistical analyses were performed using SPSS version 19 (SPSS Inc, Chicago, IL). Categorical maternal variables were compared with interquartile (25% –50% –75%) cortisol distributions and infant outcomes using 2×3 contingency tables and chi-square analyses. Interquartile cortisol distributions were analyzed to explore the correlation between maternal variables and very high or low cortisol levels. A one-way analysis of variance was used to compare continuous maternal variables with interquartile cortisol distributions and infant outcomes. Pairwise comparisons were done using a Bonferroni correction if a significant effect was found.

4 Results

5 Discussion

This is one of the first studies to compare maternal social variables with cortisol levels in very preterm infants (<32 weeks GA) on first day of life and subsequent development of complications of prematurity. The incidence of complications experienced in our study sample was consistent with previous reports [4]. In our sample, infants at an increased GA when their mothers initiated prenatal care had lower cord blood cortisol levels and an increased risk of developing NEC. Infants whose mothers smoked >10 cigarettes per day had lower salivary cortisol levels and an increased risk of IVH.

Infants whose cord blood cortisol was distributed in the lowest quartile had mothers who initiated prenatal care later in the pregnancy, at a median GA of 93 days (13.3 weeks) at initial prenatal visit, versus median GA of 62 days (8.8 weeks) for those in the middle quartiles. Additionally, infants whose mothers waited longer to initiate care trended a higher risk of developing NEC. Median GA at initial prenatal visit for infants who developed NEC was 98 (14 weeks), while median GA for infants who did not was 74 days (10.5 weeks). In both cases, this is the difference between initiating prenatal care during the first or second trimester of pregnancy. It is recommended that women who are pregnant seek care before the end of their first trimester [16]. Increased GA at initial prenatal visit could be due to any number of maternal factors, including age, cultural differences, health literacy, work status, comorbid diagnoses, limited financial resources, or lack of transportation [17]. Any or all of these factors have the potential to not only delay prenatal care but add to the mother’s actual or perceived stress. Maternal exposure to stressors may lead to prenatal programming of the HPA axis and increased risk of adverse outcomes such as NEC [10]. These findings, when considered with the trend that infants with cord blood cortisol in the lowest quartile had a higher risk of developing NEC, indicate an association between prenatal care, HPA axis response, and development of NEC. Previous studies have also found a link between HPA axis dysregulation and gastrointestinal complications in preterm infants as evidence of a brain-gut interaction [9]. This supports the research model of physiologic dysregulation and allostatic load as a mechanism for complications of prematurity.

Our finding that infants whose mothers smoked >10 cigarettes per day were more likely to have salivary cortisol levels in the lowest quartile and develop IVH is consistent with previous research. Spinillo et al. [18] found that preterm infants whose mothers smoked >10 cigarettes per day were three times as likely to develop IVH. Acute cigarette smoke inhalation can cause a slow, sustained increase of cortisol in the maternal circulation [19] and may significantly increase fetal ACTH and induce a state of hypoxia in the fetus [20], which is a well-known risk factor for IVH [21]. Exposure to increased cortisol in utero can lead to adverse prenatal programming of the fetal HPA axis [10]. This association between heavy smoking, HPA axis dysregulation, and development of IVH lends additional support to Moore’s theory of allostatic load as a mechanism for complications of prematurity.

Strengths of this study include the use of noninvasive biomarkers in the preterm population and the novel approach of comparing maternal social variables that may correlate with maternal physiological stress to cortisol and complications of prematurity, building on Moore’s theory of allostatic load and HPA axis dysregulation as a mechanism for complications of prematurity.

Limitations of the study include the maternal social variables used. Maternal data were collected retrospectively from the infant’s medical record; this limited the breadth of the project and the number of variables available. Additionally, the social variables collected are subjective measures of psychosocial stressors experienced by the mothers and may not correlate with actual maternal stress perceived or experienced. Another limitation was the use of cortisol as a marker of physiologic stress. Cortisol is a valuable indicator of HPA axis response, but can be highly variable and any comparison between cortisol levels using different sources or collection methods should be cautiously interpreted [14]. Other limitations include small sample size, missing specimen data, and methods of specimen collection.

Implications for future research include testing of the research model with larger samples and additional biomarkers. Further research is necessary to more specifically determine the effects that maternal social variables have on maternal perceived stress and physiologic stress. Continued investigation into the effects of maternal stress exposure in utero on HPA axis dysregulation and development of complications of prematurity in the preterm infant is warranted. Using biomarkers during the prenatal period and first week of life to identify infants at risk for physiological dysregulation and complications of prematurity may allow for preventative measures to be developed and tested.

Implications for clinical practice include the continued promotion of early prenatal care and ensuring access to care, especially for those who may be affected by the barriers to receiving early prenatal care such as low health literacy, cultural differences, age, and lack of financial resources or transportation. Additionally, providers should continue to advise pregnant women to stop smoking or at least reduce smoking during their pregnancy and after.

The association between maternal social variables associated with stress, HPA axis dysregulation, and increased risk of developing certain complications of prematurity supports the research model of physiologic dysregulation and allostatic load as a mechanism for complications in the preterm infant population. More research is warranted to investigate the particular effects these maternal social variables have on maternal stress levels, adverse prenatal programming of the HPA axis, and development of complications of prematurity.

Disclosure statement

This study was funded by the University of Nebraska Medical Center Clinical Research Center, the Foundation for Neonatal Research and Education, and the Sigma Theta Tau-Gamma Pi Chapter.