The Role of Environmental Hazards in Premature Birth: Workshop Summary (2003)

Preterm birth is a complex, multifactorial public health issue that is a growing problem in the United States and Canada. It continues to be a significant public health burden that has received little attention outside the prematurity research community. In the future, we need to think more critically about the impact on health if we are going to improve our performance in meeting the challenges posed by preterm births, asserted Donald Mattison.

Many of the participants agreed that preterm birth is not an acute event. The roots of preterm birth may start in early pregnancy, while some believe that the path to preterm birth starts before pregnancy and may go back as far as the parent’s early life, according to Mark Klebanoff, National Institute of Child Health and Human Development. Similarly, researchers are recognizing that preterm birth is a “catch-all” phrase for the multiple pathways leading to a clinical presentation. At the fundamental level, Allen Wilcox, National Institute of Environmental Health Sciences, suggested that we may be looking at two different types of prematurity. The first would be a fetus that is in trouble from the early stages of pregnancy (e.g., problems of placental implantation, a fetus in distress).

The second situation would be a perfectly healthy fetus that for some reason is subjected to the natural events of delivery too soon. A number of presenters speculated that perhaps not all preterm birth is bad, and in fact, it may serve as a protective mechanism in some cases for the fetus or the mother. Yet, overall, the reason for the health burden from preterm birth is summarized as increased mortality and morbidity.

This workshop represented a unique opportunity for the Roundtable and participants to look at the field of premature birth and the research opportunities that intersect with environmental health. The insight provided by bringing together investigators from diverse disciplines, including social and behavioral sciences, toxicology, and reproductive biology, clearly points out the heterogeneous and complex nature of the factors associated with preterm birth. While planning for future research, some participants suggested that we need to look at the complexity of the process and that preterm birth will have to be treated as a chronic disease—one might see this as being similar to our approach to studying hypertension, cardiovascular disease, or diabetes.

Many speakers alluded to the idea that we are on the verge of understanding the factors underlying preterm birth at the molecular, individual, and community levels. Armed with this knowledge, we will be in a better position to develop intervention strategies to address many types of preterm birth (see Box 6.1). However, the real questions that need to be asked, according to one panelist, are how are we going to get there and how are we going to get there with great dispatch? We must bring together biological, psychological, and social research to form a partnership for science—similar to the 1960s National Aeronautics and Space Administration (NASA) project to put a man on the moon. This would require a commitment of resources to make a significant impact. Box 6.1 is a compilation of the strategies identified during these discussions.

Many speakers pointed out the variety of ways to classify premature delivery. They suggested that the criteria for defining term and preterm labor should be reassessed because they can alter the interpretation of the data as well as of the risks. For example, the current data are blurred by a lack of distinction between indicated preterm labor and spontaneous preterm labor. Determinations of gestation length could be improved by the use of an additional measurement beyond calculations based on the last menstrual period. Several speakers questioned whether the appropriate measure might be preterm birth or gestational

length, and whether these variables should be discussed as dichotomous or continuous variables.

Because it is expensive and time-consuming to create new birth and newborn registries, Karla Damus suggested that efforts should be made to improve already existing registries. We need to review the content and types of information we are collecting to make them more relevant to issues of prematurity by adding well-thought-out case definitions to selected maternal and child health outcome registries.

During the opening remarks and the closing discussions, participants suggested that we do not have a full understanding of the consequences of prematurity. They noted that current attempts have looked at short-term outcomes and that the full spectrum of the impact of preterm birth over the life of the individuals has not been fully studied with respect to disabilities of all types, including neurodevelopmental disabilities. For this reason, some individuals suggested that it would be important to describe more completely the impact of preterm delivery on the life course, including individual, family, and social impact.

Some participants expressed amazement that they could not describe the initiation of labor and delivery in the normal pregnancy and suggested that this continues to impede our ability to understand mechanisms underlying preterm

birth. Without basic understanding of normal labor and delivery processes, we will not be able to study premature birth systematically. Others, however, believed that we are very close to describing the basic mechanisms and that we will be able to make significant progress by the end of the decade. Charles Lockwood discussed the need to understand gene–environment interactions (see below). The current risk factors do not accurately predict which women will deliver early. For example, maternal stress, infection, decidual bleeding (abruption), and uterine abnormalities have been implicated in preterm birth. However, the majority of women with these risk factors do not deliver prematurely.

In his closing summation, Mattison pointed to a number of the risk factors as a starting point for addressing preterm birth. He suggested that researchers should explore why prior preterm birth is one of the strongest risk factors for subsequent preterm delivery. Further, researchers may have to understand how several or multiple risk factors interact to increase the risk of preterm delivery. Better definitions and fully characterizing the interactions between birth weight and gestational age, the interactions of genetic factors and the environment, and differential mechanisms of preterm birth as a function of gestational age provide some direction to tease apart the complexity alluded to during the course of the workshop.

The study of genetics, environmental agents (including chemical, physical, and biological exposures), and gene–environmental interactions in relation to preterm birth is at a very early stage. Recent advances in molecular biology and the sequencing of the human genome will help to guide the field further. Researchers pointed to the fact that there are three genomes and environments involved—those of the mother, the father, and the fetus. Understanding these genetic roles and their interactions with the environment will be important for making the next leap in addressing issues of prematurity, concluded Mattison. To begin to address these issues, participants discussed the intersections between the environment and the genome.

Participants and panelists discussed a variety of environmental factors that may contribute to the problem of premature birth. People had listed a number of toxicants and/or classes of toxicants that warranted further study. Endocrine disruptors, such as 1,1,1-trichloro-2,2-bis (p-chlorophenyl) ethane (DDT) and its metabolites, were discussed as one area of interest for the field of preterm birth. Participants suggested that further study was needed to understand the impact on preterm birth. Of special concern would be the endocrine disruptors that fall into the estrogen agonist/antagonist categories because of the delicate balance of hor-

mones during pregnancy. Some participants wondered if a shift in the balance between estrogen and progesterone could explain some preterm birth.

During the course of the workshop, participants spoke of the environmental effect on the myometrium, but they also said that we know little about the effects of exposures on the cervical collagen or rupture of the amnion. Understanding how these environmental factors influence the cellular and molecular pathways will be of continued importance in the next few years.

On the first day of the workshop, participants discussed the growing influence of those social environmental factors that influence preterm birth. Participants questioned of the impact of the environment and life-style factors on all aspects of implantation, fetal timing, and the following initiation of parturition. Some areas that might help to shape the field include the following:

• reframing research questions about the etiology and mechanisms of preterm delivery to take advantage of new knowledge about community, family, individual, social, economic, and environmental factors on the organism, as well as cellular, molecular, and genetic mechanisms of preterm delivery,

• understanding the impact of social factors on the biological mechanism of preterm delivery,

• understanding the impact of stress on maternal physiology on premature delivery,

• encouraging cross-disciplinary approaches that would allow more critical evaluation of the role of stress and of the biological mechanisms produced by stress as they influence preterm delivery, and

• exploring the role of stress on androgen and estrogen levels in pregnancy and their influence on gestational length, especially given the growing knowledge concerning environmental endocrine disruptors.

Participants discussed the tremendous opportunities for future research. As this research avenue moves forward, gene–environment interactions in this area must be combined with rigorous evaluation of reproductive health end points and potential confounding factors. This means that researchers have to know more about biological markers and their significance. Two areas that were highlighted included exploring (1) the interactions of fetal and maternal genetic factors contributing to preterm delivery, and (2) the interactions of environmental exposures such as tobacco and genetic factors. In his summation, Mattison recognized the importance of these areas and suggested that it will be necessary to fund research to explore both individual factors, and their interactions.

The workshop highlighted the tremendous complexity of prematurity that needs to be addressed. Participants noted that there is tremendous variability between nations and ethnic groups, even among developed countries. For example, Caucasians in the United States have approximately twice the incident rate of Caucasions in France for reasons that are not obviously apparent. To begin to understand the population differences both here and abroad will require additional research.

Fernando Guerra suggested that we must address changing demographics. There is still a considerable amount of research that has to be done that will enhance the discussion of relevance to public health, including at the community level. He further suggested that we must look at special populations such as recent immigrants to the United States. Many of the participants echoed a need for a surveillance system that will allow us to follow various populations over time. Mark Klebanoff suggested that we should not get too focused on biology. Any biological mechanism must be reconciled with preterm births being a disease of poverty. Spontaneous preterm birth occurs in almost every place that has been looked at, but it is more common in cases of social adversity.

To begin to address some of the research questions, we will also have to develop good animal models for prematurity, according to Carole Kimmel. It is difficult to cause prematurity in the rat, which suggests that we must look at other animals, such as the rabbit, to better understand the process. Further, others suggested that we will need to develop animal models for each of etiologies underlying prematurity. Developing these models will help screen for potential environmental interactions and establish priorities for toxicological evaluations. In addition, this research will provide data about receptor interactions that will help guide future research.

Molecular classifications of exposure have much to offer since true prospective studies that begin prior to or early in pregnancy allow for assessment during the etiologically relevant time interval, according to some panelists. Biological markers of nutritional status (e.g., serum folate, ferritin, transferrin receptor saturation), stress (e.g., corticotropin releasing hormone, cortisol), tobacco use (e.g., urinary and serum cotinine), cocaine use (e.g., cocaine and benzoylecognine in hair and urine), and infection or inflammation (e.g., fetal fibronectin, cytokines) have all been incorporated into epidemiological studies of preterm birth to great benefit. In particular, participation in prenatal care offers unusually favorable

opportunities to incorporate the collection of biospecimens in large clinically based populations, as opposed to community samples. The collection of such biospecimens has clear applicability to the study of environmental agents because biospecimens are key in determining the toxic mechanisms of suspected agents as well as for measuring exposures.

Exposure studies have methodological limitations that must be addressed before useful interpretations can be made about how exposure affects preterm labor, noted participants. Some participants suggested that we must increase our basic knowledge of risk assessment and toxicology as they relate to preterm birth, including exposure (e.g., quantity, timing, and dosage) and outcome measurements (e.g., gestational age, initiation of labor, birth outcome, status of the child into early childhood). In addition, improved study designs and approaches are needed to better measure exposures in longitudinal studies. For example, it would be useful to know the range or types of exposures that are significant for fetal implantation problems or the range of possible exposures necessary to initiate maternal complications. We also need to know how these complications overlap and affect one another.

Strategies developed to address prematurity have focused in the past on means of reducing mortality; however, as Maureen Hack discussed, we may be at the limits of viability with present neonatal technology. Further, these strategies have failed to make a significant impact on morbidity, which suggests that researchers and clinicians must focus more clearly on other data and issues to address problems of morbidity.

Many panelists reminded participants that the goal of any pregnancy was to have healthy mothers and healthy babies. They suggested that we would have to determine those incidents in which when medically indicated, it would be better to have the baby born premature than to prolong the pregnancy. However, for cases in which it would be advantageous to prolong the pregnancy, some speakers felt that we were close to understanding some of the pathways leading to prematurity and, thus, close to developing intervention strategies. Mattison suggested that by the end of the decade we should have effective intervention strategies for many of the potential pathways. He further suggested that we need to be able to think strategically about intervening. This was echoed by another participant, who further asserted that the basic science has to be integrated with applied science. Too often during the workshop, speakers and participants discussed intervention strategies that had been implemented but had no benefit, or even had ill effects. A number of speakers asserted that we must study interven-

tions in randomized control trials before we implement such strategies on a population base.

Influences on preterm birth have been more difficult to identify than causes of reduced birth weight. The predictors of reduced birth weight are sometimes shared with predictors of preterm birth, but often have weaker associations. Some predictors of preterm birth, such as socioeconomic status and tobacco use, seem to have strong influences on birth weight, but only modest influences on preterm birth, indicating that the pathways, although unidentified, probably overlap to some extent. Direct examination of preterm birth, low birth weight, and small-for-gestational-age deliveries suggests that modest overlap occurs.

One reason for the more limited success in the study of preterm birth may be the markedly greater uncertainty involved in the measurement of duration of gestation as compared to birth weight, reflected in the greater magnitude of error in vital records, the lower quality of maternal reports of gestational age compared to birth weight, and the more limited availability of gestational age from less developed countries. Duration of gestation is usually based on the last menstrual period—a fallible marker of the time of ovulation—although the widespread use of ultrasound for dating has been helpful in increasing accuracy in many settings and has been of benefit to those studies that can incorporate this information.

Isolating specific causal agents from nonspecific influences of a healthy life-style and favorable socioeconomic conditions will be very difficult, cautioned David Savitz. For example, consistent findings have emerged that identify an association between cocaine use and preterm birth, a reduced risk associated with leisure time physical activity during pregnancy, a reduced risk associated with favorable nutritional status and the use of prenatal vitamins, and an increased risk associated with physically demanding occupations. However, none of these associations is necessarily causal, and distinguishing a true etiologic effect from a spurious association due to other unmeasured or unknown factors has been unsuccessful thus far. Isolating environmental agents from the circumstances that give rise to exposure will pose a serious challenge for the identification of the many environmental agents associated with socioeconomic deprivation and less favorable life-styles.

This page intentionally left blank.