Skip Navigation
  Print Page

Romero Lab: Perinatology Research Branch

Skip sharing on social media links
Share this:
​​​​​​​​​​​​​ Skip Internal Navigation

Research Areas

Prospective randomized clinical trial of vaginal progesterone vs. placebo in women with a short cervix

This year, we reported the results of a multicenter, randomized, double-blind, placebo-controlled trial to determine the efficacy and safety of micronized vaginal progesterone gel (90mg) to reduce the risk of preterm birth and associated neonatal complications in asymptomatic women with a singleton pregnancy and a sonographically detected short cervix (10–20 mm) at 19–23.9 weeks of gestation. Women were allocated to receive vaginal progesterone gel or placebo daily from 20–23.9 weeks until 36.9 weeks, rupture of membranes, or delivery. Randomization sequence was stratified by center and history of a previous preterm birth. Approximately 32,000 women were screened worldwide. Of 465 women randomized, 458 were included in the analysis (vaginal progesterone gel, n=235; placebo, n=223). Women allocated to receive vaginal progesterone had a lower rate of preterm birth, i.e, at less than 33 weeks (primary endpoint), than those allocated to placebo (8.9% vs. 16.1%). The effect remained significant after adjustment for co-variates. Vaginal progesterone was also associated with a significant reduction in the rate of preterm birth at less than 28 weeks (by 50%) and less than 35 weeks (by 35%), respiratory distress syndrome (by 61%), any neonatal morbidity or mortality event and birth weight of less than 1500 g. There were no differences in the frequency of treatment-related adverse events between the groups. This is the first randomized clinical trial to demonstrate that the administration of vaginal progesterone in women with a short cervix can reduce the rate of early preterm birth and neonatal morbidity.

Universal sonographic screening of pregnant women for short cervical length and treatment with vaginal progesterone to prevent preterm birth is cost-effective: a decision and economic analysis

Prior to the availability of the results of our randomized clinical trial described above, our group conducted a decision-analytic and cost-effectiveness analysis to determine which strategy for the prevention of preterm birth and associated morbidity is the most cost-effective. Four strategies were tested (using quality-adjusted life-years, cost in US dollars, and number of preterm births prevented). Universal sonographic screening for cervical length and treatment with vaginal progesterone was the most cost-effective strategy and was the dominant choice over the three alternatives, which were: (i) sonographical cervical screening limited to women with a previous history of preterm birth and treatment with vaginal progesterone; (ii) treatment of women with a history of previous preterm birth with 17 alpha-hydroxyprogesterone caproate (17-OHP-C); and (iii) no screening or treatment. Universal screening would represent savings of $1339 ($8325 vs. $9664), when compared with treatment with 17-OHP-C, and would lead to a reduction of 95,920 preterm births annually in the United States. We concluded that universal sonographic screening for short cervical length and treatment with vaginal progesterone appears to be cost-effective and yields the greatest reduction in preterm birth at less than 34 weeks' gestation. The results of this study provide compelling evidence for a change in healthcare policy in the United States—to prevent preterm delivery, reduce neonatal morbidity, and decrease the economic burden of the leading cause of infant mortality.

Identification of feto-placental rejection as a mechanism of disease in preterm birth

We had previously proposed that an abnormal allogeneic reaction was responsible for several obstetrical syndromes, including spontaneous preterm labor and delivery. The mammalian fetus is a semi-allograft, as 50% of its genome is of paternal origin. The mechanisms responsible for tolerance of this allograft have been subject to investigation. Our Program focused on determining whether "rejection" of the semi-allograft may be identified by studying the human placenta and the characterization of associated lesions. We published several studies that support the view that chronic chorioamnionitis and villitis of unknown etiology (VUE) are closely related immunologic inflammatory lesions, harboring features of allograft rejection of the mother and graft-versus-host disease of the fetus. We demonstrated that a subset of women with chronic chorioamnionitis have circulating maternal anti-fetal antibodies and that such antibodies can cross the placenta and induce complement activation and a novel form of systemic fetal inflammation never described before. We reported a cross-sectional study to examine the frequency and significance of anti–HLA antibodies in maternal and fetal sera according to the presence of chronic chorioamnionitis in the context of spontaneous preterm birth (PTB). Maternal anti–HLA class I seropositivity in PTB was significantly higher than in normal term births. Chronic chorioamnionitis was significantly associated with a higher maternal anti–HLA class I seropositivity in both preterm and term births. VUE was significantly associated with increased maternal and fetal anti-HLA class I and II seropositivity. C4d immune complex deposition on umbilical vein endothelium was more frequent in PTB than at term (77% vs. 11%). Logistic regression analysis revealed that chronic chorioamnionitis (OR=6.1), maternal anti-HLA class I seropositivity (OR=5.9), and C4d deposition on umbilical vein endothelium (OR=36.2) were associated with preterm labor/delivery. Collectively, this evidence suggests that a subset of spontaneous preterm births is the result of maternal anti-fetal cellular and antibody-mediated rejections with links to fetal graft-versus-host disease and alloimmune reactions. Our future studies will focus on identifying biomarkers for this mechanism of disease as well as elucidating the mechanisms responsible for spontaneous preterm labor in these patients.

Molecular imaging for the detection of fetal microglial activation

We established a role for intrauterine inflammation/inflammation in the development of periventricular leukomalacia (PVL) and cerebral palsy (CP). Neuroinflammation is characterized by activation of the microglial cells, which play a key role in the pathogenesis of white matter injury and PVL. A major challenge is the identification of neuroinflammation at the time of birth. Diagnosis of clinically "silent" neuroinflammation would allow early treatment and prevention of motor disorders. We conducted a series of studies to determine whether microglial activation can be detected by molecular imaging. PK11195 is a ligand that is selective for the translocator protein expressed on activated microglia and macrophages, and [11C]PK11195 can effectively be used as a ligand in positron emission tomography (PET) for the detection of activated microglial cells in various neuroinflammatory conditions. We generated a model of fetal inflammation by administering bacterial endotoxin to pregnant rabbits at 28 days of gestation. Kits were born spontaneously at 31 days and underwent PET imaging on the day of birth and neurobehavioral testing. Using ROC curves, we identified an optimal cutoff for detecting abnormal neurobehavioral scores consistent with the phenotype of CP in neonatal rabbits. We observed that an increased uptake of [11C]PK11195 in the brain over time had a sensitivity of 100% for all parameters tested. The magnitude of [11C]PK11195 binding measured in vivo by PET imaging correlated strongly with the severity of motor deficits in the neonatal rabbit, suggesting that molecular imaging in the neonatal period can identify neuroinflammation (microglial activation) that is clinically significant. This observation would allow identification of the newborn at risk for neurologic injury and early treatment.

A link between fetal systemic inflammation and abnormalities in the CNS/serotonergic system: a link between cerebral palsy and autism-spectrum disorders

It is now clear that a subset of infants are affected by both cerebral palsy and autism-spectrum disorders, yet the mechanisms responsible for this association remain to be elucidated. Our previous work provided a link between fetal systemic inflammation and cerebral palsy (clinical and experimental evidence). We propose that neuroinflammation may also affect the serotonergic system and therefore be responsible for a group of neurodevelopmental disorders with a somatosensory component (e.g., autism-spectrum disorders). Serotonin plays a crucial role in the regulation and maturation in the developing brain, and a disruption of serotonin metabolism has been implicated as a mechanism of disease in autism-spectrum disorders. To test the hypothesis that fetal inflammation may result in disruption of the serotonergic system in the perinatal period, we administered saline or bacterial endotoxin into the intrauterine compartment of pregnant rabbits at term. We used PET imaging with a [11C]methyl-L-tryptophan (AMT) to study the newborns. Endotoxin-exposed newborns in utero demonstrated a decrease in the uptake for AMT and serotonin concentration in the frontal and parietal cortices. We also found significant reduction in serotonin-immunoreactive fibers and expression of the serotonin transporter (5HTT) in the somatosensory cortex and a three-fold increase in the number of apoptotic cells in the ventrobasal (VB) thalamus. Collectively, these data suggest that fetal inflammation induced by microbial products results in cortical serotonin disruption of thalamocortical development in the newborn brain. Our observations shed light on the role of fetal infection in the genesis of autism, and also provide an explanation for the coexistence of cerebral palsy and autism in a subset of children.

Maternal plasma concentrations of angiogenic/anti-angiogenic factors are of prognostic value in patients presenting to the obstetrical triage area with the suspicion of preeclampsia.

Preeclampsia (PE) remains one of the leading causes of medically-indicated preterm birth and maternal mortality/morbidity worldwide. Many signs and symptoms associated with PE are subjective, non-specific, and have poor diagnostic performance for PE. A growing body of evidence suggests that an imbalance of angiogenic/anti-angiogenic factors is involved in the pathophysiology of PE. We have shown that patients who presented to the obstetrical triage area with a diagnosis of "suspected PE" and subsequently were diagnosed with mild PE or severe PE requiring preterm delivery had significantly different mean plasma multiple of the median (MoM) concentrations of angiogenic/anti-angiogenic factors from those who did not require delivery until term. Plasma concentrations of sEng, PlGF, PlGF/sEng, and PlGF/sVEGFR-1 performed very well in identifying patients who developed severe PE requiring preterm delivery. Among patients who presented before 34 weeks' gestation, a plasma concentration of PlGF/sVEGFR-1 of ≤0.033 MoM was associated with a shorter interval-to-delivery than those above the cutoff. These biomarkers were useful in identifying patients who required delivery within two weeks. We propose that plasma concentrations of the PlGF/sVEGFR-1 or PlGF/sEng ratios may assist clinicians in the management and disposition of patients suspected to have PE.

Four-chamber view and "swing technique" (FAST) echo: a novel and simple algorithm to visualize standard fetal echocardiographic planes

Novel algorithms using three- and four-dimensional ultrasound have been generated to simplify fetal echocardiography. Using a new display technology (OmniView) for three-dimensional ultrasound, we developed an algorithm using spatiotemporal image correlation (STIC) volume datasets by drawing four dissecting lines through the longitudinal view of the ductal arch. We tested the algorithm in 50 normal fetal hearts between 15.3 and 40 weeks of gestation and calculated the visualization rates of cardiac diagnostic planes. We also tested the algorithm in five cases with proven congenital heart defects. In normal cases, the algorithm was able to generate the diagnostic planes individually in 100% of cases (except for the three-vessel view, which was seen in 98%), and simultaneously in 98%. The "swing technique" was able to generate the three-vessels and trachea view, five-chamber view and/or long-axis view of the aorta, four-chamber view and stomach in 100% of normal cases. In cases of congenital heart defects, the images displayed views that deviated from what was expected from the examination of normal hearts. This study demonstrated that the FAST echo algorithm simplifies examination of the fetal heart, thereby reducing operator dependency.

Simple targeted arterial rendering (STAR) technique: a novel and simple method to visualize the fetal cardiac outflow tracts

A major challenge in fetal echocardiography is the visualization of the outflow tracts (aortic and pulmonary arteries). It is now clear that a fetal cardiac examination without adequate imaging of the outflow tracts is incomplete and may miss major anomalies such as transposition of the great vessels. Imaging of the outflow tracts is operator-dependent and widely accepted as unsatisfactory. We used OmniView to generate an algorithm that would reliably identify the fetal outflow tracts. Three dissecting lines were drawn through the four-chamber view of the heart contained in a STIC volume dataset. We tested the technique in 50 normal fetal hearts between 15.3 and 40.4 weeks of gestation and abnormal cases. The STAR technique was able to generate the necessary diagnostic planes in all normal cases. In fetuses with congenital heart defects, the STAR technique identified a ventricular septal defect and great vessel anomalies. The practical implication of this work is that physicians can now obtain fetal outflow tract views in most cases when a STIC volume is available.

Evaluation of fetal cardiac function with 4DUS using STIC and virtual organ computer-aided analysis (VOCAL)

Evaluation of the basic parameters of cardiac function in the human fetus is challenging and has relied on two-dimensional ultrasound examination, yet this approach relies on many assumptions about the anatomy of the fetal heart that have not yet been validated. 3DUS and 4DUS allow volumetric examination of the fetal heart and therefore determination of key parameters such as stroke output and cardiac volume. We conducted a cross-sectional study in normal pregnancies (19-42 weeks of gestation) to evaluate ventricular volume, stroke volume (SV), cardiac output (CO), and ejection fraction (EF). The CO was also expressed as a function of estimated fetal weight and biometric parameters. We analyzed 184 STIC datasets. With advancing gestation, ventricular volume, SV, CO, and adjusted CO increased while EF decreased. The RV volume was larger than the left ventricular (LV) volume in systole (0.50 vs. 0.27 mL) and diastole (1.20 vs. 1.03 mL). There were no differences between the left ventricle (LV) and right ventricle (RV) in SV, CO, or adjusted CO. LV EF was greater than the RV EF (72.2 vs. 62.4%). This study also provided reference ranges for the evaluation of fetal cardiac performance.

Collaborative study on 4-dimensional echocardiography for the diagnosis of fetal heart defects: the COFEHD study

4DUS with STIC is used to perform fetal echocardiography, and mounting evidence indicates that this may facilitate examination of the fetal heart. We conducted a multi-center study to determine the accuracy of 4DUS in the diagnosis of congenital heart defects. Fetuses with and without confirmed heart defects were scanned between 18 and 26 weeks of gestation, and their volume data sets were uploaded onto a centralized file transfer protocol server. Ninety volume data sets were randomly selected for blinded analysis by international experts. Overall, the median (range) sensitivity, specificity, positive and negative predictive values, and false-positive and false-negative rates for the identification of fetuses with congenital heart defects were: 93% (77%–100%), 96% (84%–100%), 96% (83%–100%), 93% (79%–100%), 4.8% (2.7%–25%), and 6.8% (5%–22%), respectively. There was excellent intercenter agreement. This study demonstrates that 4D volume data sets can be remotely acquired and accurately interpreted by experts located at great distances from where patients are examined. These data suggest that tele-fetal echocardiography can be performed around the world.

Last Updated Date: 11/30/2012
Last Reviewed Date: 11/30/2012

Contact Information

Name: Dr Roberto Romero
Senior Investigator
Perinatology Research Branch
Phone: 313-993-2700
Fax: 313-993-2694
Email: romeror@mail.nih.gov

Staff Directory
Vision National Institutes of Health Home BOND National Institues of Health Home Home Storz Lab: Section on Environmental Gene Regulation Home Machner Lab: Unit on Microbial Pathogenesis Home Division of Intramural Population Health Research Home Bonifacino Lab: Section on Intracellular Protein Trafficking Home Lilly Lab: Section on Gamete Development Home Lippincott-Schwartz Lab: Section on Organelle Biology