July 2007: Volume 33, Number 7
Pregnancy Risk Assessment
Improving Outcomes with Appropriate Lab Testing
By Michael Papez, MD and David G. Grenache, PhD
Every year in the U.S., more than 6 million women become pregnant, ultimately giving birth to approximately 4 million infants. But despite significant improvements in the health of newborns over the last century, our nation’s infant mortality rate remains relatively high compared to other industrialized nations. According to the National Center for Health Statistics, the rate at which babies less than one year of age die has steadily declined over the past several decades, from 26.0 per 1,000 live births in 1960 to 6.9 per 1,000 live births in 2000. Yet in 1998, the U.S. ranked 28th in the world in infant mortality, a measure used to compare the health and wellbeing of populations across and within countries.
While challenges in decreasing the infant mortality rate clearly remain, as laboratorians, we provide test results that are an integral part of prenatal and perinatal care and that have the potential to save infants’ lives. Laboratory support of a successfully managed pregnancy presents certain challenges, too. The breadth of laboratory screening and diagnostic studies performed in low-risk pregnancies is substantial, and the results require skilled interpretation. In addition, laboratorians will likely see higher demand for some screening tests based on revised recommendations from the American College of Obstetricians and Gynecologists (ACOG) that all women, regardless of age, be offered Down syndrome screening.
This article focuses on how laboratories can improve infant outcomes by providing an appropriate menu of tests. Specifically, we discuss maternal serum screening tests for open neural tube defects (ONTD), preterm delivery, fetal lung maturity, and new molecular tests that show promise for the management of neonatal Group B streptococcal infection.
Maternal Serum Screening
The purpose of maternal serum screening is to identify pregnant women who are at risk of having an infant with an ONTD or a chromosomal anomaly, the best known of which is trisomy 21 (Down syndrome). Women at increased risk are offered more invasive diagnostic testing that requires the collection of amniotic fluid by amniocentesis. Screening for ONTDs relies on the detection of α-fetoprotein (AFP), while screening protocols for trisomy 21 are more complex and utilize multiple markers, both biochemical and physical.
Traditionally, obstetricians order maternal serum screening tests for both ONTDs and trisomy 21 in the second trimester, typically between 16 and 18 weeks of gestation. This time frame doesn’t necessarily maximize sensitivity and specificity of the screening tests; however, it offers the most acceptable combination of screening sensitivity, risk associated with diagnostic testing, and ability to identify multiple congenital defects, as well as a detection window that is early enough to allow for possible intervention procedures. There are two current second trimester screening strategies. The triple screen includes AFP, human chorionic gonadotropin (hCG), and unconjugated estriol (uE3). Similarly, the quadruple screen uses the triple screen markers plus dimeric inhibin A (DIA). More recently, investigators have described first trimester screening for trisomy 21. This screening occurs between 11 and 14 weeks of gestation and combines both biochemical measurements of maternal serum hCG and pregnancy-associated plasma protein A (PAPP-A) in conjunction with ultrasound measurement of fetal nuchal translucency, the width of space between the fetal skin and the cervical spine.
Key to the laboratorian’s understanding of maternal serum screening is that the final result is not a distinct concentration of a particular analyte, but rather it is the modification of an age-based risk by the likelihood ratios derived from each of the analyte’s concentrations expressed as a multiple of the median (MoM). Obstetricians prefer the MoM score over the absolute concentration of analytes because it is simple to derive, more stable, and allows for interlaboratory variation.
|Laboratory Support of Pregnant Women|
|Disorder ||Incidence ||Screening Tests ||Comments|
|Open neural |
|1:1000 ||Maternal serum AFP ||Detects ~90% of open defects; will not detect closed defects|
|Down syndrome ||1:1000 ||Triple screen (AFP, hCG, uE3) ||Performed in 2nd trimester; ~70% detection rate.|
|Quad screen (AFP, hCG, uE3, DIA) ||Performed in 2nd trimester; ~80% detection rate.|
|Nuchal translucency, hCG, PAPP-A (combined screen) ||Performed in 1st trimester; ~80% detection rate.|
|Preterm delivery ||1:8 ||Fetal fibronectin ||~95% negative predictive value.|
|Inversely related to gestational age ||Fetal lung maturity tests (surfactant/ albumin ratio, lamellar body count, L/S ratio) ||Tests have high sensitivities and negative (mature) predictive values.|
|Group B |
|0.35:1000 ||Rectovaginal culture ||Not 100% sensitive; long turnaround time.|
|Real-time PCR ||Best utilized for patients in labor who have not been cultured.|
Open Neural Tube Defects
ONTDs result from the failure of the neural tube to close properly in early embryogenesis. Several types of ONTDs occur, but spina bifida is the most widely recognized type. As the second most prevalent congenital anomaly in the U.S., approximately 1 out of every 1000 babies is born with this type of birth defect.
AFP serves as a marker for fetuses with an ONTD. This glycoprotein is produced first by the yolk sac and then by the liver of the developing fetus, and its concentration predictably increases and then decreases in fetal serum, amniotic fluid, and maternal serum throughout gestation. In fetuses with an ONTD, AFP has direct access to the amniotic fluid compartment and increased amniotic fluid concentrations lead to increased concentrations in maternal serum, which can be assessed without harming the fetus. An AFP MoM greater than 2.0 is ~90% sensitive and 95% specific for ONTDs (1). Patients with a positive screening test require additional workup; typically obstetricians order amniotic fluid AFP and acetylcholinesterase tests, as well as high-resolution ultrasound. Abnormally low AFP levels are also significant, signifying a chromosomal abnormality such as trisomy 21. As such, AFP is one component of 2nd trimester screening tests for Down syndrome.
Trisomy 21, or Down syndrome (DS), is the most common form of mental retardation and occurs in 1 out of every 1000 live births. As the risk of having an infant with DS increases with increasing maternal age, biochemical screening for DS has historically been used for women older than 35 years of age. As mentioned earlier, ACOG recently recommended that all women, regardless of age, be offered DS screening, as early as the first trimester (2). Multiple screening strategies are currently available and a complete description of each is beyond the scope of this article. However, laboratorians should at least be familiar with the overall strategy of the 2nd trimester triple and quadruple screens, as well as the combined 1st trimester screen.
The triple screen consists of maternal serum measurements of AFP, hCG, and uE3. With an average MoM of 0.8, AFP is lower in DS pregnancies than in unaffected ones. hCG, which is produced by the placenta, is a dimeric glycoprotein composed of an α- and ß-subunit. It is responsible for maintaining progesterone concentrations in early pregnancy. Concentrations of hCG in pregnancy increase steadily and predictably after implantation of the embryo, peak at 10–12 weeks of gestation, and sharply decline over several weeks, and then remain at a plateau until delivery. Intact hCG is increased in DS pregnancies with a MoM of ~2.0. The final marker of the triple screen is uE3, the end result of steroid metabolism that begins in the fetal adrenal gland and ends with estriol production by the placenta. Estriol increases predictably during pregnancy, but lower concentrations (MoM ~0.7) are observed in DS pregnancies. Overall, the triple screen detects ~70% of DS fetuses, with a 5% false-positive rate (1).
The quadruple screen consists of the triple screen plus the addition of DIA. The inhibins are heterodimeric glycoproteins consisting of an α and one of two ß subunits (A or B) that function to inhibit pituitary gland release of follicle stimulating hormone. Maternal serum concentrations of DIA rise in early pregnancy, fall to a plateau at 15–20 weeks, and then peak again at term. Concentrations of DIA rise abnormally in DS pregnancies (MoM ~1.7), and the inclusion of DIA with the traditional triple screen markers increases the DS detection rate by ~10% (1).
First trimester screening for DS has several advantages over 2nd trimester screening, including earlier detection of affected fetuses, additional time for the mother and/or parents to make a decision about the pregnancy, and, if desired, safer methods of pregnancy termination.
Two large studies have confirmed that combining ultrasound measurements of nuchal translucency with maternal serum measurements of hCG (intact molecule or the free ß-subunit) and PAPP-A yields a detection rate for DS that is slightly better than the quadruple screen (3, 4). As previously mentioned, increased thickness of the width of space between the fetal skin and the cervical spine as measured by ultrasound, termed nuchal translucency, is strongly associated with DS. A placental glycoprotein, PAPP-A increases rapidly during the 1st trimester and is more than two times lower in DS pregnancies (MoM ~0.4). In contrast to hCG, which is a slightly more sensitive DS marker in the 2nd trimester, PAPP-A is more sensitive in the 1st trimester. One disadvantage of 1st trimester screening is the technical challenge inherent in consistent and reproducible ultrasound findings, which may require measurements by a reference facility to resolve. In addition, a proportion of DS pregnancies spontaneously abort between the 1st and 2nd trimester screening, implying that some subset of the 1st trimester screen-positive pregnancies have an inevitable outcome and negating many of the benefits of early screening in that subset.
The birth of an infant prior to 37 weeks of completed gestation is considered a preterm delivery. In 2004, one out of every eight live-born infants was born prematurely. Furthermore, the preterm birth rate has climbed 18% percent since 1990. Consequently, preterm delivery represents a significant healthcare problem and a major cause of neonatal morbidity and mortality. Fifty percent of infants born at 24 weeks of gestation die, and more than half of those who do survive live with a disability.
Predicting preterm labor has traditionally been based on methods that are neither sensitive nor specific, such as obstetrical history or symptoms and epidemiological risk factors. A test that could differentiate women with symptoms of preterm labor who are likely to deliver preterm infants from those who have apparent preterm labor and are unlikely to deliver prematurely would not only prevent unnecessary and potentially risky interventions but could also reduce treatment costs.
A laboratory test that detects fetal fibronectin (fFN) in the cervicovaginal fluid has been useful for determining if pregnant women with symptoms of preterm labor will deliver their baby prematurely. fFN is a member of the fibronectin family of proteins that are present in the extracellular matrix and plasma. During gestation, fFN’s location at the choriodecidual interface suggests a role for this protein as a “trophoblast glue” that anchors the fetal trophoblast in the uterus. Although fFN is detectable in cervicovaginal secretions during the first 24 weeks of pregnancy, it typically declines to undetectable concentrations between weeks 24 and 34. The protein is released into cervicovaginal secretions when the chorionic/decidual interface is disrupted, a fact that underlies the rationale for measurement of fFN as a predictor of preterm delivery. While the presence of fFN in symptomatic women during weeks 24 through 34 of gestation indicates an increased risk of preterm delivery, the absence of fFN is a much more reliable predictor, because it indicates that the pregnancy is likely to continue for at least another two weeks.
Cytyc (Sunnyvale, Calif.) manufactures the only FDA-approved test for cervicovaginal fluid fFN. The chromatographic immunoassay produces a positive or negative result that is interpreted by a meter. A positive result indicates a concentration of fFN that is >50 ng/mL, and studies show that the assay can be used to predict the risk of preterm delivery in symptomatic and asymptomatic patients. In a meta-analysis of 40 studies, fFN had sensitivities of 77%, 74%, and 70% for predicting delivery in symptomatic women within 7, 14, and 21 days of testing, respectively (5). Because sensitivities in asymptomatic women are lower, the National Academy of Clinical Biochemistry does not recommend fFN testing in this population. (6). Unfortunately, the positive predictive value of fFN is consistently low (<30%) and many women with positive results go on to deliver term infants. The real value of the assay appears to come from its high negative predictive value (~95%), which helps women with a negative result to avoid potentially dangerous or expensive interventions.
The FDA has also approved salivary estriol for use as a marker of preterm delivery. Like fFN, it has low sensitivity and positive predictive value but high negative predictive value. Estriol’s reduced effectiveness for identifying patients at risk for early preterm delivery (<30 weeks) and long analytical time present significant limitations, and therefore, ACOG does not recommend the test (7).
Cytokines, in particular interleukin-6 (IL-6), have also been investigated as markers of preterm delivery. In a study of 161 asymptomatic women between 24 and 36 weeks of gestation, levels of cervical IL-6 >250 pg/mL optimally identified patients with subsequent preterm deliveries, versus term deliveries (47% positive predictive value and 86% negative predictive value) (8). Another study of 165 patients demonstrated that cervical IL-6 levels perform as well as fFN in predicting preterm delivery (9). Additional studies investigating the clinical utility are needed before this marker can be used routinely.
Respiratory Distress Syndrome
To function as an effective organ of gas exchange upon delivery, the fetal lungs must overcome surface tension forces that promote collapse of the alveoli. Pulmonary surfactants are a complex mixture of phospholipids and proteins packaged into storage granules called lamellar bodies. The function of these structures is to decrease surface tension at the alveolar-air interface. Secretion of surfactant by type II pneumocytes is a late-stage event in the development of the fetal lung; therefore, respiratory distress syndrome (RDS) in newborn infants is nearly always associated with preterm birth. At 29 weeks, the risk of RDS is >60%, but it steadily declines to <5% at 37 weeks or more. In 2003, RDS was the seventh leading cause of infant death in the U.S., claiming 20.3 deaths per 100,000 live births.
Because fetal lung liquids contribute to amniotic fluid, laboratory tests that determine the biochemical or biophysical properties of pulmonary surfactant are performed on amniotic fluid. Several tests for fetal lung maturity (FLM) are available, all of which have thresholds that maximize diagnostic sensitivity for immaturity at the expense of specificity. This means that many infants born with an “immature” result do not develop RDS. Due to the combination of low RDS prevalence, high diagnostic sensitivity, and low diagnostic specificity, FLM test results have high negative (mature) and low positive (immature) predictive values.
In 1971, researchers introduced the first fetal lung maturity test based on the ratio of lecithin to sphingomyelin (L/S ratio) as determined by thin-layer chromatography. Lecithin is an abundant pulmonary surfactant that gradually increases from 28 weeks of gestation until delivery, while sphingomyelin remains relatively constant during the last trimester. Therefore, sphingomyelin serves as a useful internal standard against which the increase in lecithin can be monitored. The L/S ratio increases with gestational age and correlates with fetal lung maturity, and labs frequently use a result of >2.0 as a cutoff to indicate maturity. Because the method is labor-intensive, technique-dependent, and demonstrates considerable inter-laboratory variation, it has been recommended that laboratories receiving <15 requests per week not perform the test (10).
Most labs perform the TDx FLM II assay, a fetal lung maturity test manufactured by Abbott Laboratories (Abbott Park, Ill.). This assay uses fluorescent polarization to determine the ratio of surfactant to albumin. As an automated method, it is rapid and simple to perform, and has low intra-assay variability. Its major disadvantage, however, is a wide quantification scale: values ≤39 mg/g are considered immature, values ≥55 mg/g are considered mature, and those between the two cutoffs are considered indeterminate. Because of this gap, some investigators have suggested use of gestational age in conjunction with the surfactant/albumin ratio result as a predictive model for assessing RDS risk (11).
At 36 weeks, phosphatidylglycerol (PG) is the last surfactant to appear in the amniotic fluid, making it suitable as a late marker of fetal lung maturity. Like the determination of the L/S ratio, labs can detect PG by thin-layer chromatography, and some labs use this measurement in an effort to increase the specificity of the L/S ratio. An immunochemical agglutination test is also available to qualitatively detect PG. The Amniostat-FLM test from Irvine Scientific (Santa Ana, Calif.) uses polyclonal anti-PG antibodies that agglutinate PG-containing lamellar bodies into macroscopic clusters. The major advantage of this assay is its ability to produce rapid results even with specimens contaminated by blood or meconium. However, the late appearance of PG in gestation means that the failure to detect it often occurs in a high proportion of infants that do not develop RDS.
As lamellar bodies are similar in size to blood platelets, labs can enumerate them from amniotic fluid using automated hematologic cell counters. Known as the lamellar body count (LBC), labs also use this test to estimate surfactant production and predict the degree of fetal lung maturity. Because the LBC varies depending on the type of cell counter used, labs should establish instrument-specific reference intervals and confirm them with outcome-based investigations (6). To date, only one study has compared the clinical utility of the LBC to that of the surfactant/albumin ratio, and the investigators found no difference in their clinical utility (12).
Because a mature result is highly predictive of lung maturity, ACOG recommends a sequential approach to fetal lung maturity testing (7). Using this approach, labs should first perform a rapid test, such as surfactant/albumin ratio or LBC, and then other lung maturity tests only if a mature result is not obtained.
Group B Streptococcal Infection
Group B streptococcus (GBS) has long been acknowledged as an etiologic agent of neonatal sepsis, pneumonia, and death. Clinically, GBS disease in neonates is divided into early-onset disease, which occurs within 5 days of birth after microbial transmission in utero or during passage through the birth canal, and late-onset disease, which occurs between 7 days and 3 months with bacterial transmission via birth or by caregivers after birth. Perinatal GBS infection remains a leading cause of neonatal death, with a 2–8% mortality rate in term neonates that have early-onset disease and a 10–15% rate in babies with late-onset disease. Because appropriate intrapartum antibiotic treatment reduces the risk of developing GBS disease in neonates, in 1996 the Centers for Disease Control and Prevention (CDC) issued recommendations for health care providers to enact either culture or risk-based screening and treatment of expectant mothers. The CDC modified these guidelines in 2002, recommending universal screening by culture at 35–37 weeks gestation (13).
Rectovaginal culture, however, remains an imperfect screening tool. Early-onset GBS disease can occur in culture-negative expectant mothers, suggesting that the sensitivity of antepartum culture is not perfect. In addition, traditional culture techniques yield results in 1–3 days. This time frame may not be useful should an expectant mother present in labor prior to having been cultured. Antibiotic treatment requires at least 4 hours to take effect, which may be insufficient to protect babies who are delivered unexpectedly.
Some labs have turned to molecular-based methods to overcome these problems. A new test, the IDI-StrepB assay (Cepheid, Sunnyvale, Calif.), recently received FDA-approval as a real-time PCR assay that can give results in a matter of hours with a sensitivity of ~85%. As its sensitivity is less than that of culture, labs still need to perform an intrapartum culture, and even in PCR-positive patients, cultures for antibiotic sensitivity testing may be required. The clinical utility of this test is best realized in patients who present in labor without having been cultured and those whom obstetricians estimate will not deliver for at least 6 hours. This time frame takes into account 1–2 hours for the analytical time required to perform the assay and 4 hours necessary for pre-delivery treatment in case of a positive result (14).
While no systematic approach to improving the sensitivity of population-based screening has been clinically accepted yet, molecular techniques show promise. In one study, researchers tested intrapartum vaginal swabs that had been incubated in the appropriate liquid media overnight by traditional (not real-time) PCR analysis. Using this method, researchers identified more positive samples than what eventually grew on intrapartum culture and in a significantly shorter time frame, <1 day versus 1–3 days for culture. In fact, if the grouped intrapartum PCR results are considered the gold standard for GBS detection, then traditional antenatal culture may have a sensitivity as low as 42% (15). The implication here is that antenatal screening via molecular methods, which could be batched, may be a more sensitive predictor of GBS colonization than traditional antenatal culture. Further study is necessary to link this possibility to clinical outcomes.
One drawback to molecular-based screening is that specialized equipment and trained personnel are necessary, and in the case of intrapartum screening, they must be available all hours of every day. Such hurdles, however, are likely temporary in this age of growing molecular-based diagnoses.
The Goal: Improving Outcomes for Mothers and Babies
Laboratory testing is an important and fundamental aspect of the medical care received by pregnant women and their infants. Importantly, the discovery of biomarkers and the development of new technologies require that laboratorians remain attentive to the remarkable advances in pregnancy risk assessment and judiciously employ the tools available.
1. MacRae AR, Canick JA. Prenatal Screening for Fetal Defects. In: Gronowski AM, ed. Handbook of Clinical Laboratory Testing During Pregnancy. Totowa, NJ: Humana Press, 2004:71–137.
2. American College of Obstetricians and Gynecologists. ACOG Practice Bulletin: Screening for fetal chromosomal abnormalities. Obstet Gynecol 2007; 109: 217–27.
3. Wald NJ, Rodeck C, Hackshaw AK, Walters J, Chitty L, Mackinson AM. First and second trimester antenatal screening for Down’s syndrome: the results of the serum, urine and ultrasound screening study (SURUSS). J Med Screen 2003; 10: 56–104.
4. Malone FD, Canick JA, Ball RH, Nyberg DA, Comstock CH, Bukowski R et al. First-trimester or second-trimester screening, or both, for Down’s syndrome. N Engl J Med 2005; 353: 2001–11.
5. Leitich H, Kaider A. Fetal fibronectin—how useful is it in the prediction of preterm birth? BJOG 2003; 110 Suppl 20: 66–70.
6. Sherwin JE, Lockitch G, Rosenthal P, Rhone S, Magee LA, Ashwood ER, Goldsmith BM, Lee CR, Geaghan S, Millington D, and Bennett M. National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines: Maternal-Fetal Risk Assessment and Reference Values in Pregnancy. Washington, D.C.: AACC Press, 2006.
7. American College of Obstetricians and Gynecologists. ACOG Educational Bulletin. Assessment of fetal lung maturity. International Journal of Gynecology & Obstetrics 1996; 56: 191–198.
8. Lockwood CJ, Ghidini A, Wein R, Lapinski R, Casal D, Berkowitz RL. Increased interleukin-6 concentrations in cervical secretions are associated with preterm delivery. Am J Obstet Gynecol 1994; 171: 1097–1102.
9. Grenache DG, Hankins K, Parvin CA, Gronowski AM. Cervicovaginal interleukin-6, tumor necrosis factor-alpha, and interleukin-2 receptor as markers of preterm delivery. Clin Chem 2004; 50: 1839–1842.
10. Ashwood ER. Standards of laboratory practice: evaluation of fetal lung maturity. Clin Chem 1996; 43: 211-214.
11. Parvin CA, Kaplan LA, Chapman JF, McManamon TG, Gronowski AM. Predicting respiratory distress syndrome using gestational age and fetal lung maturity by fluorescent polarization. Am J Obstet Gynecol 2005; 192: 199–207.
12. Haymond S, Luzzi VI, Parvin CA, Gronowski AM. A direct comparison between lamellar body counts and fluorescent polarization methods for predicting respiratory distress syndrome. Am J Clin Pathol 2006; 126:894–899.
13. Schrag S, Gorwitz R, Fultz-Butts K, Schuchat A. Prevention of perinatal group B streptococcal disease. Revised guidelines from CDC. MMWR Recomm Rep 2002; 51: 1–22.
14. Davies HD, Miller MA, Faro S, Gregson D, Kehl SC, Jordan JA. Multicenter study of a rapid molecular-based assay for the diagnosis of group B Streptococcus colonization in pregnant women. Clin Infect Dis 2004; 39: 1129-1135.
15. Rallu F, Barriga P, Scrivo C, Martel-Laferriere V, Laferriere C. Sensitivities of antigen detection and PCR assays greatly increased compared to that of the standard culture method for screening for group B streptococcus carriage in pregnant women. J Clin Microbiol 2006; 44: 725–728.
Michael Papez, MD, is a resident in the Department of Pathology and Laboratory Medicine at the University of North Carolina School of Medicine in Chapel Hill, North Carolina.
David Grenache, PhD is an assistant professor in the Department of Pathology at the University of Utah and medical director of special chemistry at ARUP Laboratories, Salt Lake City, Utah. Email: email@example.com