1 Section on Maternal-Fetal Medicine, Department of Obstetrics & Gynecology, Wake Forest University School of Medicine, Winston Salem, NC 27157, USA
Abstract
Objective: Preterm delivery remains the leading cause of neonatal
morbidity and mortality leading to a burden lasting well beyond the inherent
costs of caring for the premature neonate. Physician-scientists, scientists, and
clinicians have intensively studied associations, scoured every aspect to
determine modifiable risk factors, and trialed prospective interventions to
generate best practices. We aimed to generate a useful review for clinicians for
the identification of women at risk for preterm birth along with modifiable
factors and treatments to help reduce preterm delivery. Mechanism: We
performed a literature search for preterm birth prevention to facilitate
compilation of a narrative review. Findings in Brief: The PROLONG study
found that Makena did not significantly reduce the risk of preterm birth (PTB)
Keywords
- preterm birth
- prevention of preterm birth
- progesterone supplementation
- micronutrient intake
- adverse pregnancy outcome
- cervical insufficiency
- obesity
- periodontal disease
Preterm birth is the leading cause of neonatal morbidity and mortality and is defined as delivery at less than 245 days after conception or between 20 0/7 and 36 6/7 weeks’ gestation [1]. Preterm delivery can be subcategorized by gestational age at delivery with early preterm birth defined as delivery prior to 34 0/7 weeks of gestation or late preterm when delivery is between 34 0/7 weeks of gestation and 36 6/7 weeks of gestation [2]. The etiology of preterm birth is often multifactorial because of the many medical and obstetric conditions that are associated with it. Despite advances in medical care, rates of preterm birth continue to increase in the United States. According to the Centers for Disease Control (CDC), 10.4% of all pregnancies in the United States were affected by prematurity in 2022. These rates disproportionately affect African American women at 14.6% compared to 9.4% in white women and 10.1% in Hispanic women [1, 2].
Preterm delivery is associated with high rates of neonatal and infant morbidity and mortality. It is the leading cause of neonatal intensive care unit (NICU) admission and has a significant economic impact with an estimated cost of $25 billion dollars in the US in 2016 alone [3]. Common immediate neonatal complications of prematurity include respiratory distress syndrome (RDS), necrotizing enterocolitis (NEC), sepsis, intraventricular hemorrhage (IVH), bronchopulmonary dysplasia (BPD), patent ductus arteriosus (PDA), apnea, and retinopathy of prematurity (ROP). Neonatal morbidity relating to prematurity increases with decreasing gestational age, low birth weight, maternal race, location of delivery, use of group B strep prophylaxis, and use of antenatal corticosteroids [4]. There are also negative long term neonatal outcomes such as cerebral palsy, neurosensory impairment, reduced cognition, impaired motor performance, academic difficulties, and attention deficit hyperactivity disorders (ADHD) [5].
Preterm births can be separated into spontaneous or medically indicated preterm deliveries. Medically indicated preterm births account for ~25% of all preterm births in the United States and are accompanied by a medical condition that has increased maternal or fetal risk with continuing the pregnancy. Some of the most common causes are preeclampsia (~25%–40%), abnormal fetal antenatal testing (~25%), fetal growth restriction (~10%–20%), placental abruption (~7%–10%), and stillbirth (7%) [6]. The remaining 75% of preterm births are not related to any maternal or fetal illness and are preceded by spontaneous preterm labor, preterm prelabor rupture of membranes (PPROM), or cervical insufficiency with preterm delivery being the common endpoint. Many of the risk factors for spontaneous preterm labor such as low pre-pregnancy body mass index (BMI), tobacco use, substance abuse, and short interpregnancy interval are modifiable [7]. Temporary clinical factors during a current pregnancy associated with an increased risk of preterm birth include periodontal disease, vaginal bleeding, vaginal infection, and urinary tract infections [8, 9]. Unfortunately, treatment of these risk factors has not been shown to decrease rates of preterm birth. Other common risk factors such as short cervical length, surgical treatment of cervical dysplasia, multiple gestation, history of dilation and curettage, and a history of prior preterm birth are non-modifiable [10]. This review aims to generate a referent guide for clinicians for the identification of women at risk for preterm birth along with modifiable factors and treatments to help reduce preterm delivery. Accordingly, our review compiles risk assessment and intervention strategies that are evidence-based and show promise in reducing preterm delivery in the obstetric population.
To generate this narrative review, our group utilized PubMed with the search terms ‘preterm birth prevention’, ‘risk assessment for preterm birth’, ‘biomarkers and preterm birth’, ‘short interpregnancy interval and preterm birth’, ‘nutrition and preterm birth’, ‘depression and preterm birth’, ‘stress and preterm birth’, ‘smoking and preterm birth’, ‘substance use and preterm birth’, ‘infection and preterm birth’, ‘periodontal disease and preterm birth’, ‘vaginal pH and preterm birth’, ‘vaginal microbiome and preterm birth’, ‘progesterone and preterm birth’, and, finally, ‘cerclage and preterm birth’ in attempt to generate a comprehensive yet succinct review on identification of at-risk mothers and prevention strategies to reduce preterm birth. Given the large scope of the review, the comprehensive review demanded a narrative format with inherent limitations of a curated review intent on selecting representative literature on the subject matter of preterm birth prevention.
An important strategy in the prevention of preterm birth is accurately predicting which patients are at increased risk for this adverse outcome. Having a prior preterm birth is one of the strongest predictors of preterm birth. However, this predictor is only present in about 10% of all preterm births [11]. Several screening methods have been studied with variable success, including the development of risk-scoring systems, fetal fibronectin, biomarkers, and assessments of cervical length [12].
Risk-scoring systems have been developed to identify at-risk pregnancies based on history, physical exam findings, and social and economic risk factors. While some models have shown promise initially, systemic reviews have raised concerns about their quality and performance [12, 13]. These models are heavily weighted on history of prior preterm delivery and are less helpful for nulliparous patients.
The use of serum and vaginal biomarkers, including proinflammatory cytokines, have been studied extensively in hopes of identifying a measurable factor to distinguish pregnancies at high risk for preterm birth [14]. Unfortunately, due to the heterogeneity of existing studies, many of these biomarkers were never successfully translated into clinical practice [15]. Studies have shown that the presence of cervical fetal fibronectin (fFN) is a risk factor for preterm birth. However, the sensitivity and positive predictive value (PPV) of a positive fFN is low. Additionally, there are no successful interventions available for women with positive results, further limiting its clinical utility. In nulliparous patients, fFN has low predictive accuracy for spontaneous preterm birth and is not recommended as a screening test [16]. Conversely, the absence of cervical fFN has excellent negative predictive value (NPV 97%) and serves as a useful tool in stratifying patients into a low risk profile for imminent preterm birth and, in effect facilitates triaging patients appropriate for discharge with preterm labor precautions [17]. Proteomics is also being used to help identify proteins that may serve as predictors of spontaneous preterm birth. Studies have identified the ratio of insulin-like growth factor-binding protein 4 to sex hormone-binding globulin (IBP4/SHBG) as a predictor of spontaneous preterm birth [18, 19]. However, a recent prospective randomized intervention trial investigating whether a screening protocol using IBP4/SHBG did not show a reduction in spontaneous preterm birth [20].
A short cervix is one of the strongest predictors of preterm birth. An
assessment of cervical length in the second trimester can be used to identify
pregnancies at increased risk for preterm birth [16, 19]. Data from the Alabama
Preterm Prevention project showed high negative predictive value of a cervical
length more than 2.0 cm for preterm birth, particularly in women with prior
preterm birth less than 34 weeks (hazard ratio (HR) 2.8, p
Universal cervical length remains a subject of debate. However, the American College of Obstetrics and Gynecology (ACOG) recommends that the cervix be visualized at the time of the mid-trimester anatomy ultrasound in all pregnancies [2]. In the evaluation of the patient with threatened preterm labor, a cervical length more than 2.0 cm confers high negative predictive value (81.4% in nulliparas and 76.9% in multiparas) as opposed to short cervix. As with fFN, the positive predictive value of a short cervix for preterm delivery remains poor in the asymptomatic patient but has better utility in the patient with symptoms of threatened preterm labor effectively identifying those most suited for tocolysis and antenatal corticosteroid therapy (PPV 93.7% in nulliparas and 87.5% in multiparas) [24].
Primary prevention of preterm birth involves early recognition and mitigation of risk factors for preterm delivery in early pregnancy or prior to pregnancy. Primary prevention should start prior to conception. Postpartum and pre-pregnancy care provide opportunities to address and prevent modifiable risk factors associated with preterm birth such as short interpregnancy interval, defined as interpregnancy interval less than 18 months [25]. The easiest and most common tool to assess for risk factors for preterm birth is a detailed history and physical exam. This involves discovery and optimization of preexisting medical conditions that may have a negative impact on the pregnancy. The focus of this early assessment is to identify modifiable risk factors and those that have a treatment intervention. The key to primary prevention during pregnancy is early entry into prenatal care. This involves increasing access to care and removing barriers to care such as transportation difficulties, cultural differences, and financial burden [26].
Low maternal BMI and poor nutritional status have consistently been shown to
pose an increased risk of preterm birth. Women with a BMI less than 20 kg/m
Micronutrient deficiencies in zinc, iron, and folate are also associated with spontaneous preterm delivery but data are heterogeneous with respect to benefit gained and meta-analyses nullify results from small cohorts favoring recommendations for supplementation [28, 29]. Nonetheless, micronutrient deficiencies are common in women with a low BMI and women from low-income countries. Micronutrient supplementation may be beneficial in these specific populations but is not generalizable to most larger populations. Maternal obesity is also a risk factor for indicated preterm birth with conflicting evidence regarding the risk of spontaneous preterm birth [30, 31]. The increased rates of indicated preterm birth are attributed to the common pre-existing and obstetric co-morbidities associated with obesity such as hypertension and diabetes.
Maternal depression is one of the most common pregnancy complications, with up to 37% of pregnant women reporting depressive symptoms at some point during pregnancy [32]. Numerous studies have suggested an association between maternal depression during pregnancy and preterm birth [33, 34, 35]. Further, pregnant women whose depression symptoms worsen during pregnancy have been shown to have increased odds for preterm birth [36, 37].
There are several potential biologically plausible explanations for this association, including dysregulation of the hypothalamic-pituitary-adrenocortical axis [38] and the exaggerated inflammatory response that occurs in response to depression [39]. There are also social and behavioral factors which may mediate this observed association, including tobacco and substance use and underutilization of healthcare. Increased efforts towards identifying biopsychosocial interventions that target early identification and treatment of perinatal depressive symptoms may represent a strategy for preterm birth prevention.
The influence of factors related to social determinants of health cannot be understated when discussing preterm birth. Social and economic disadvantages are consistently associated with an increased risk of preterm birth. Strikingly, a profound disparity in preterm birth rates exist for Black women in the United States—in 2018, the rate of preterm birth among Black women (14%) was approximately 50 percent higher than the rate among white women (9%) [1]. Other specific factors including maternal age, income, educational attainment, social support, and housing instability have been associated with increased preterm birth rates [40, 41, 42, 43, 44].
Maternal stress during pregnancy has been shown to be associated with adverse pregnancy outcomes including preterm birth in both cohort studies and animal models [45, 46, 47]. This association is complex and multifactorial and involves both biologic and psychosocial factors [48]. While the underlying mechanisms that contribute to this observed association are not completely understood, research suggests that cortisol may be a mediator between maternal stress and pregnancy outcomes [49]. Studies have shown that women with higher concentrations of cortisol may be at higher risk of having a preterm birth [50, 51, 52]. Though more research is needed to further understand the relationship between maternal psychosocial factors and preterm birth, it is possible that biopsychosocial interventions to improve maternal depressive symptomology and stress levels could lead to a decrease in preterm births.
Tobacco use in pregnancy has been consistently shown to increase the risk for preterm birth, both spontaneous and iatrogenic. Though the exact mechanism is incompletely understood, research suggests that this observed association may be due to vasoconstrictive and hypoxia-mediated pathways. There is a dose-response relationship between tobacco use and preterm birth, with heavier smoking being associated with higher rates of preterm birth [53]. Cessation of tobacco use in pregnancy is one of the few preventable causes of preterm birth, and pregnancy has been shown to be the time in a woman’s life when she is most likely to quit smoking [54]. Efforts should be made to provide interventions including behavioral support to aid in smoking cessation during pregnancy.
Substance use is a common complication of pregnancy, with approximately 5% of pregnant women in the United States having reported using illicit drugs while pregnant [55]. Many drugs freely cross the placenta and cause vasoconstriction that restricts the fetal oxygen supply [56]. Use of substances including both stimulant and depressive drugs during pregnancy has been shown to increase the rate of preterm birth [57, 58, 59]. A study examining all live births in California from 2007 through 2012 found that nearly 3% of women who reported drug use during pregnancy delivered before 32 weeks’ gestation, while less than 1% of non-drug using women delivered this early [59].
Intrauterine infections and the fetal inflammatory response have a strong association with preterm birth. It has been estimated that between 40 to 80% of preterm births are associated with inflammation within the uterus, cervix, placenta, decidua, fetal membranes, or amniotic fluid [60, 61]. Screening for and treating infections has been identified as a target for the primary prevention of preterm birth.
Bacterial vaginosis (BV) has been associated with an increased risk for preterm birth. However, studies have not consistently shown that treatment of BV reduces the risk for preterm birth [62]. A 2013 Cochrane review of 21 trials of antibiotic treatment of BV in pregnancy showed no definitive evidence that treatment reduced the risk for preterm birth [63]. ACOG does not recommend screening for and treating asymptomatic BV to prevent preterm birth [2]. Screening for BV early in pregnancy for women with a history of previous preterm delivery may have some benefit, but data remains insufficient [64].
Untreated asymptomatic bacteriuria in early pregnancy has also been associated with increased risk for preterm birth. A Cochrane review including over 2000 women showed antibiotic treatment may be associated with a reduction in preterm birth (RR 0.34, 95% CI 0.13–0.88) with low-certainty evidence. However, it has been suggested that this risk reduction may be mostly related to prevention of the progression of asymptomatic bacteriuria to pyelonephritis [65].
Observational studies have identified periodontal disease as a risk factor for preterm birth, with risks increasing with progression of disease in pregnancy [66]. However, a large trial assessing treatment of periodontal disease was not able to show benefit of reduction in preterm birth [67]. Also, a 2017 Cochrane review of 15 trials concluded that it remains unclear if the treatment of periodontal disease during pregnancy impacts preterm birth and that insufficient data exists to determine which treatment option is best [68]. Based on current evidence, the treatment of periodontal disease to prevent preterm birth is not recommended [69]. However, no harm from the treatment of periodontal disease has been reported. While it may not reduce the risk for preterm birth, it remains reasonable to recommend treatment for a patient’s overall health.
Alkaline vaginal pH has been not only associated with malodorous discharge and
vaginal irritation due to overgrowth of Gardnerella vaginalis, a
condition known as bacterial vaginosis, but has a clear association with preterm
birth. Cervical shortening and preterm birth are increased 3-fold in setting of
vaginal pH
Conversely, Lactobacillus colonization in the vagina has an acidifying effect on the vaginal pH, which is protective against preterm delivery by way of inhibiting the growth of harmful microbes. Many studies show these bacteria generate lactic acid and hydrogen peroxide, which not only shifts the pH to mildly acidic promoting low diversity present species and domination by Lactobacillus but also protects against infections [73]. Vaginal samples of African American women have been demonstrated to have low Lactobacillus, less of a predominance of Lactobacillus crispatus, or increased Lactobacillus jensenii, which all have been postulated as contributory towards increased incidence of preterm birth [74]. A multitude of cohort studies have shown dysbiosis or colonization with pathogenic bacteria—Gardnerella vaginalis, Escherichiacoli, Pepto streptococcus—is associated with increased risk for late miscarriage, preterm premature rupture of membraned, preterm labor, and as would naturally follow, increased relative risk for having a premature birth [74, 75, 76, 77, 78, 79, 80, 81, 82]. Treatment of symptomatic as opposed to asymptomatic bacterial vaginosis mitigates the risk for preterm delivery as well as that of chorioamnionitis and endometritis in the puerperal period; this association has led many to question the absolute “sterility” of the intrauterine cavity in gestation [78, 79, 80, 81, 82].
Historically, obstetricians and physician-scientists have viewed the confines of the gestational sac—fetus, placenta, amniotic fluid, and the chorion and amnion—as being sterile. Over the past couple of decades, data have been collected that challenge this dogma. Meconium samples collected at birth have demonstrated colonization with Lactobacillus [83].
Similarly, growing evidence of amniotic fluid samples, placenta, and membrane colonization is accumulating. Moreover, variation in microbial flora colonizing these previously thought to be “sterile” products of the conceptus are correspondingly associated with preterm birth, noting that alterations of the vaginal and placental microbiome have been linked to both spontaneous preterm birth as well as with indicated preterm birth, namely preeclampsia [84]. Data are mounting that several health factors are inexorably intertwined as maternal mouth flora alterations have been shown to lead to alterations in meconium colonization of the fetus as sampled at birth suggesting hematogenous mechanism. Furthermore, maternal stress antepartum not only alters flora of maternal gut and vagina but also of the placenta and fetal gut; in turn such alterations, have been shown to have key association with term versus preterm birth [82].
Accordingly, the fetal-placental microbiome has been recently postulated as a target for the prevention of preterm birth [83, 84]. Tests that identify variance in the microbiome may prove useful in both identifying those at risk for preterm birth as well as those who may benefit from probiotics to increase colonization by protective bacteria like Lactobacillus as well as to eradicate or reduce colonization by vaginal group B streptococcus [83, 84, 85].
Animal and human studies suggest that functional progesterone withdrawal
precedes labor [86]. As such, progesterone supplementation has been and continues
to be evaluated for its ability to prevent preterm birth (PTB) among women at
risk. Until recently, 17 alpha-hydroxyprogesterone caproate (17-OHP), administered
as a weekly intramuscular injection from 16 to 36 weeks, was a cornerstone of
recurrent spontaneous PTB prevention. This intervention was established following
the findings of Meis et al. [87], a randomized, double-blind,
placebo-controlled study, which concluded that 17-OHP resulted in a statistically
significant reduction in PTB at
Vaginal progesterone has also been proposed to reduce the risk of PTB with the
evidence suggesting that its effect is most robustly observed in with women with
a short cervix,
In the light of the FDA’s withdrawal of 17-OHP, the American College of
Obstetricians and Gynecologists has updated its recommendations regarding
progesterone supplementation to reflect the most recent evidence. The Society for
Maternal Fetal Medicine (SMFM) and ACOG no longer advises 17-OHP for the
prevention of recurrent spontaneous PTB [94, 95]. It continues to support serial
cervical length measurements from 16 0/7 weeks to 24 0/7 weeks for those with
history of spontaneous PTB, with consideration of vaginal progesterone in the
event of cervical length (CL)
It is important to note that the exact mechanisms by which intramuscular 17-OHP and vaginal progesterone work remain elusive and in some instances are conflicting. While some evidence suggests that 17-OHP may promote myometrial quiescence, this finding has not been consistently shown [97, 98, 99]. Other proposed mechanisms of action surround the notion that 17-OHP acts as an anti-inflammatory agent [98, 100]. Available data has not consistently demonstrated that 17-OHP acts in this way. Progesterone, not 17-OHP, may act via multiple anti-inflammatory pathways [101]. In the effort to develop strategies to reduce the risk of PTB, it is critical to note that progestins are not interchangeable.
An estimated 0.5%–1% of pregnancies are affected by cervical insufficiency,
and for these patients, cerclage remains a mainstay of preterm birth prevention.
Those eligible for consideration of cerclage placement include those with and
without a history of PTB. Among those with singleton pregnancies and no history
of PTB, CL screening is performed at the time of the anatomic survey, and those
who are found to have CL
In his recent narrative review, Dr. Kent Heyborne asserts that clinicians may
consider foregoing serial CL screening in patients with a history of spontaneous
late preterm birth, as cerclage has not specifically been found to reduce
recurrent spontaneous PTB in those who prior PTB occurred at
Preterm birth remains the leading cause of neonatal morbidity and mortality. Clinicians need to have a set of tools and guidelines to frame their clinical practice. As like many observations in obstetrics history is a major indicator of at-risk pregnancies: having a prior preterm birth is one of the strongest predictors of preterm birth. However, this predictor is only present in about 10% of all preterm births, leaving most preterm births subject to finding modifiable behavioral and environmental factors that can be identified and mitigated.
Achieving appropriate gestational weight gain should not be overlooked as inappropriate gestational weight gain remains a modifiable risk factor for preterm birth. Likewise, reducing high risk behaviors (e.g., smoking, drug use) and treatment of depression/anxiety remain important tools to reduce risk for patient well-being and reducing adverse outcomes such as preterm birth.
Treatment of genitourinary infection and periodontal disease are important for patient health status and poses potential benefit in reduction of premature delivery. Vaginal pH and microbiome screening are showing potential for identifying those at risk of preterm birth and studies are accumulating on promoting vaginal Lactobacillus as a strategy for increasing ability to achieve term pregnancy.
While ACOG no longer advises 17-OHP for the prevention of recurrent spontaneous PTB, obstetricians should be aware of the areas in which preterm birth may be primarily prevented. Primary prevention of preterm birth involves early recognition and mitigation of risk factors for preterm delivery such as interpregnancy interval, tobacco use, access to prenatal care, optimizing nutritional status, and optimizing mental health.
Additionally, a short cervix is one of the strongest predictors of preterm
birth. An assessment of cervical length in the second trimester can be used to
identify pregnancies at increased risk for preterm birth. Hence, ACOG continues
to support serial cervical length measurements from 16 0/7 weeks to 24 0/7 weeks
for those with history of spontaneous PTB, with consideration of vaginal
progesterone in the event of CL
ACOG further recommends that those with singleton pregnancies and a history of
spontaneous PTB undergo serial transvaginal ultrasound for CL measurement from 16
0/7 weeks to 24 0/7 weeks; those with CL
All authors contributed to literature search, selection of salient information, contribution to writing, and editorial changes in the manuscript. Specifically, (1) SH wrote sections on universal cervical length screening, biomarkers, primary prevention, prenatal care, interpregnancy interval, nutritional interventions, nutritional supplementation, prenatal care, intergestational health, risk assessment of preterm birth, cervical length screening, conclusions, and recommendations, coordinated structure and outline of overall manuscript with JD and oversaw writing of all sections in the manuscript with each of the authors in the manuscript; (2) AG wrote sections on introduction, incidence of preterm birth, spontaneous preterm birth defined as opposed to indicated preterm birth, and risk factors for preterm birth; (3) SD performed literature search, wrote sections with statements on progesterone, cerclage, tocolytic medications, bedrest and hydration, and clinics for preterm birth prevention; (4) AS maternal psychosocial, smoking cessation, substance use, treatment of infections, and periodontal care; (5) KQ coordinated conclusions and recommendations with JD and SH, literature search, review and editing of all sections in the manuscript, and (6) JD performed literature search, wrote abstract, methods, sections on vaginal pH and microbiome, statements on nutrition and intergestational health, coordinated and wrote conclusions and recommendations with SH and KQ, review and editing of all sections in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.
Not applicable.
We would like to acknowledge the Wake Forest University School of Medicine (WFUSOM), Department of Obstetrics & Gynecology, Section on Maternal-Fetal Medicine for provision of protected time to provide a thoughtful review of the literature as invited by the journal Clinical and Experimental Obstetrics & Gynecology. Furthermore, we would like to thank our WFU for supporting any publication related fees in the production of this manuscript. Lastly, we would like to thank the peer reviewers for the opinions and suggestions to make this a successful referent review for clinicians in obstetrics & gynecology caring for women at risk for preterm birth.
This research received no external funding.
The authors declare no conflict of interest.
References
Publisher’s Note: IMR Press stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.