Chapter 3

Amniocentesis is a procedure that involves the aspiration of amniotic fluid from the amniotic sac during pregnancy. It is generally carried out with a spinal needle (20–22 gauge) in a transabdominal approach, using a sterile technique under continuous ultrasound guidance. ^∗

Amniocentesis can be classified by the time in the pregnancy when it is done and by its indication. In the second trimester amniocentesis is most often performed for genetic indications. Before 15 weeks’ gestation the procedure is called “early” amniocentesis, and 1 milliliter of amniotic fluid per week of gestation is obtained. However, early amniocentesis is gradually being abandoned because it is associated with a high rate of subsequent amniotic fluid leakage (premature rupture of membranes). The majority of amniocenteses for prenatal diagnosis are done between 15 and 20 weeks’ gestation.

In the third trimester amniocentesis is most often performed for fetal lung maturity testing. In the setting of preterm labor or preterm rupture of the membranes, amniocentesis can be used to evaluate possible intraamniotic infection or inflammation. It can also be done for special diagnostic procedures such as polymerase chain reaction for cytomegalovirus in the setting of intrauterine growth restriction (IUGR) or hydrops. Amniocentesis can be helpful in reducing amniotic fluid volume in the setting of polyhydramnios with either premature labor or maternal respiratory difficulty. It is also used for twin-twin transfusion associated with polyhydramnios in one fetus. This type of amniocentesis is often called reduction amniocentesis.

In Rh and other blood group isoimmunizations, amniocentesis has traditionally been used for bilirubin assessment using ΔOD 450, but it is being done far less frequently now because middle cerebral artery Doppler has been found to be extremely accurate in predicting the degree of fetal anemia. In this setting amniocentesis is used to determine whether the fetus is Rh positive or positive for the sensitized antigen so that testing can be avoided if the fetus is not at risk. ^∗

Immediate and preliminary (1- to 3-day) results can be obtained for cytogenetics using fluorescence in situ hybridization. Definitive chromosome studies require cultured amniocytes (cells from amniotic fluid) and therefore usually require 10 to 14 days.

Third-trimester hemorrhage refers to any bleeding from the genital tract during the third trimester of pregnancy. In practice, it refers to any bleeding that occurs from the time of viability, (i.e., 23 to 24 weeks’ gestation). The common causes are classified as placenta previa (7%), placental abruption (13%), and other bleeding (80%), including local lesions of the lower genital tract, vasa previa, early labor, trauma, neoplasia, and marginal placental separation. Such bleeding complicates about 6% of pregnancies. ^∗

Ultrasound visualization is the method of choice for diagnosis of placenta previa. Multiple reports show a transvaginal approach to be safe and superior in its accuracy compared with transabdominal ultrasound. ^∗

Digital vaginal examination is not recommended when bleeding occurs until placenta previa is excluded by performing an ultrasound examination.

Placenta previa occurs in 1 in 200 deliveries at term. Complete placenta previa is detected in 5% of second-trimester gestations, with 90% resolving by term; partial placenta previa is seen in 45% of second-trimester gestations and resolves in more than 95% of cases. This apparent resolution is most likely related to the growth of the lower uterine segment in late pregnancy, so the placenta appears to move away from the os.

Placental abruption is the separation of the normally implanted placenta before the birth of the fetus. It results from bleeding from a small arterial vessel into the decidua basalis. It is termed a revealed abruption when vaginal bleeding is present (90%) and a concealed abruption if no bleeding is visible (10%). It is uniquely dangerous to the fetus and the mother because of its serious pathophysiologic sequelae. The incidence varies but averages about 0.83% or 1 in 120 deliveries. Abruption severe enough to cause fetal death is less common (approximately 1 in 420 deliveries). ^∗†

In subsequent pregnancies the recurrence risk of placental abruption is between 6% and 16%; after two consecutive abruptions the risk is 25%. Women who have a placental abruption severe enough to cause fetal death have a 7% risk of a similar outcome in a subsequent pregnancy.

Maternal complications

Fetal and neonatal complications

Fetomaternal hemorrhage is caused by a disruption of the normal barrier at the placental-decidual interface. It may occur with abruptio placentae; however, it occurs more commonly with abruptio placentae associated with maternal trauma, with maternal trauma without abruptio placentae, or spontaneously without an apparent precipitating event. Approximately 5% of stillbirths without apparent cause are the result of fetomaternal hemorrhage. The diagnosis is made by performing a Kleihauer–Betke test on maternal blood, which allows quantification of fetal cells in maternal serum. In patients with spontaneous fetomaternal hemorrhage, the presenting symptom is decreased fetal movement. If the fetus is still alive and the hemorrhage is severe enough, the diagnosis is often made because of a sinusoidal fetal heart rate (FHR) tracing. Treatment can consist of immediate delivery if the fetus is near term or intrauterine transfusion if the fetus is premature and no abruption is apparent. ^∗

The Centers for Disease Control and Prevention (CDC) protocol, as well as the American Academy of Pediatrics (AAP) and American College of Obstetricians and Gynecologists (ACOG) guidelines, recommend prophylactic treatment with penicillin or ampicillin for women in labor who are positive for Group B streptococcus (GBS). There has been a 50% reduction in the rate of neonatal GBS infection in institutions since the CDC protocols were introduced, which underscores the usefulness of appropriate chemoprophylaxis. There is some evidence, however, that part of this reduction in GBS disease has been offset by an increase in gram-negative perinatal infections.

For pencillin-allergic women with high risk of anaphylaxis (as in the present case), the CDC’s 2010 guidelines recommend susceptibility testing against both erythromycin and clindamycin. Prophylaxis with erythromycin is no longer recommended, even if sensitivity is documented. Prophylaxis with clindamycin is recommended if GBS is proved sensitive to both clindamycin and erythromycin and if there is no inducible resistance to clindamycin using D-zone testing. If sensitivity is unknown, or if all these requirements are not met, vancomycin is recommended. ^∗†‡

Preterm labor is defined as regular painful uterine contractions associated with a change in cervical dilation and effacement before 37 weeks of gestation. Often there is concern that by waiting for substantial cervical change before implementing treatment, the delay will result in failed treatment. Furthermore, regular contractions are common in patients who later go on to deliver at term. Thus in randomized series, as many as 50% of episodes of preterm labor do not progress with placebo treatment, and in practice as many as 80% of patients who are treated are not truly in preterm labor.

Fetal fibronectin is an extracellular matrix protein, the presence of which in cervicovaginal secretions is a predictor of preterm birth. This predictor has a high negative predictive accuracy (>99% negative predictive value; i.e., the absence of fetal fibronectin indicates <1% chance of delivery within 2 weeks) but only a mediocre positive predictive accuracy. ^∗†

Most commonly, this test is used in patients with preterm contractions in which the diagnosis of preterm labor is uncertain. A negative test result allows greater than 99% reassurance that the patient will not deliver in the next 2 weeks and often prevents unnecessary treatment. ^∗

See Table 3-1.

See Table 3-2. ^∗

There is no question that tocolysis is effective over short-term intervals; however, clinical trials have not consistently demonstrated that gestation can be prolonged significantly or that respiratory distress syndrome can be consistently prevented with tocolysis. ^∗

The obstetric definition of PROM is rupture of membranes before the onset of labor. “Premature” in this context means only “before onset of labor” and does not imply a preterm gestational age. Thus the term is actually a misnomer. More recently the more accurate term “prelabor rupture of membranes” has been used,especially in the obstetric literature, but it has not been generally adopted in clinical practice. PROM can occur at term or preterm. The latter is commonly abbreviated pPROM. Prolonged PROM (>24 hours) in the term patient is associated with an increased risk of neonatal infection and mortality; however, the duration of membrane rupture does not alter the rate of infection or mortality in the preterm patient. Complications of PROM include the following:

Multiple courses of antenatal steroids (more than three) are associated with suppression of the fetal adrenal gland and decreased response to stress in a critically ill neonate. In addition, animal and human data suggest less brain growth and developmental delay in childhood after multiple doses of steroids. The benefit of more than one course of antenatal steroids is controversial. A National Institutes of Health consensus conference on antenatal steroids recommended that only a single course of steroids be used and that the use of subsequent courses be limited to patients in research studies that address this question. Several clinical trials tested weekly repeated courses of steroids versus a single course. A Cochrane review concluded that repeated courses may result in a modest reduction in neonatal respiratory distress syndrome. Still, more than three courses can result in other problems, as noted previously. A reasonable compromise is the use of a “rescue course” of steroids—that is, a single repeat course targeted at those most likely to deliver within a week. ^∗†

Many strategies have been used to identify patients who are destined to deliver prematurely. Risk assessment scoring using the modified Creasy score ( Table 3-3) or other similar systems works well in some populations but not in others. The Creasy score looks at a series of variables in an attempt to define clinical indicators that are likely to result in preterm labor. A major limitation of most clinical risk scoring systems is that they rely heavily on a history of preterm birth in a prior pregnancy, yet the majority of preterm births occur in women without such a history.

Endovaginal ultrasound screening can detect cervical shortening several weeks before the onset of preterm labor in some patients. If a short cervix is found at 18 to 24 weeks, treatment with vaginal progesterone therapy reduces the risk of preterm birth by 40% to 50%.

Fetal fibronectin screening can identify a subgroup of women at high risk for preterm birth, but there is no known therapy that will consistently prevent preterm delivery in women with positive fibronectin screening. ^∗

Since 2003, there have been over a dozen trials evaluating prophylactic use of progesterone agents, either vaginal or oral micronized progesterone or intramuscular 17-hydroxyprogesterone caproate (17Pc). In women with prior preterm birth, weekly 17Pc reduced the recurrence of preterm birth by 33% to 45% and vaginal micronized progesterone showed similar benefit in one large trial but not another. In women with short cervix detected by endovaginal ultrasound screening, vaginal micronized progesterone reduced early preterm delivery by 40% to 50% in two large trials. The ACOG has endorsed progesterone therapy in such patients. Several trials showed that these agents are not effective in twin or triplet pregnancies. ^{∗†‡§¶}

Monozygotic twins (identical twins) arise from the division of a single fertilized egg. Depending on the timing of the division of the single ovum into separate embryos, the amnionic and chorionic membranes can be shared (if division occurs more than 8 days after fertilization), separate (if it occurs less than 72 hours after fertilization), or mixed (separate amnion, shared chorion if 4 to 8 days after fertilization). Sharing of the chorion, amnion, or both is associated with potential problems of vascular anastomoses (and possible twin-twin transfusion), cord entanglements, and congenital anomalies. These problems increase the risk of IUGR and intrauterine death. They also increase the risk of preterm delivery. Dizygotic twins, however, result from two separately fertilized ova and, as such, usually have a separate amnion and chorion.

Conjoined twins are classified according to the degree and nature of their union. These are listed below in order of decreasing frequency:

Few terms evoke more trepidation from obstetricians and neonatologists (particularly in a court room, not to mention the delivery room) than perinatal asphyxia. The term perinatal asphyxia, however, is so vague and so arbitrarily applied that it is virtually meaningless. In actuality there are two definitions. One is strictly the presence of hypoxia and metabolic acidosis, and the other includes the presence of metabolic acidosis and organ damage. ACOG has suggested that the term not be used, except when all of the following criteria are clearly met:

In labor electronic FHR monitoring is the primary modality used to determine fetal oxygenation. Although this method is quite reliable when results are normal, when marked abnormalities occur on the FHR tracing the infant is more often vigorous and not acidotic at birth. Thus the term fetal distress is more often than not inaccurate. A more accurate expression is that the FHR is no longer reassuring and that either other information must be used to establish fetal well-being or, failing that option, the fetus must be delivered (Fig. 3-1).

Certain fetuses are at risk for antepartum fetal death and asphyxial injury. The purpose of evaluating fetal well-being before labor is to prevent such adverse outcomes by first identifying hypoxia and impending damage or death and then either reversing the process if possible or, failing that, executing delivery in the hope that the baby’s condition will improve in the nursery.

Many obstetricians teach their patients to assess fetal movement regularly in the latter half of pregnancy. There is no consensus regarding the best method to count fetal movements. However, when fetal movement reaches an alarmingly low level (e.g., fewer than two in an hour), the mother should come in for a test to assess the fetus.

The FHR accelerates in response to fetal activity. This responsiveness forms the basis of one of the most widely used assessments of fetal well-being, the NST. In the NST the presence of FHR accelerations occur in response to fetal movement. The NST is usually performed in an outpatient setting. The patient is connected to a standard tocodynamometer while the FHR is monitored (by Doppler ultrasound transduction). In general, the clinician looks for at least two accelerations of the FHR of greater than 15 bpm amplitude lasting at least 15 seconds in a 20-minute period of monitoring. If reactivity standards are not met, the tracing is considered nonreactive and a second period of 20 minutes may be observed to eliminate the possibility of fetal sleep. Vibroacoustic stimulation may also be used to rouse the fetus from sleep. If the study is still nonreactive, it should be followed by a CST or a BPP to further assess fetal well-being.

A CST was one of the earliest techniques to assess fetal well-being. In this test uterine contractions are induced, either by maternal nipple stimulation or by an intravenous infusion of oxytocin (oxytocin challenge test [OCT]). The former method may be quicker and removes the need to establish an intravenous infusion; the latter is the traditional, time-honored technique. Results are interpreted the same, regardless of the method of inducing contractions. The mother is monitored with a tocodynamometer and a FHR transducer while uterine contractions are stimulated until adequate, which is defined as three contractions within 10 minutes. In a negative test result, there are no late decelerations of the FHR. In a positive test result, in which there are late decelerations, the risk of mortality and morbidity for the fetus increases, with some reports of mortality as high as 15%. There are, however, many false-positive instances of CST results. In such situations the obstetrician often faces a difficult decision of how aggressively to proceed with delivery of the fetus because the cervix may not be in a favorable condition at that time, and a cesarean section may be required. If the test results are equivocal, it may be reasonable to wait an additional 24 hours to repeat the test.

At present, the CST is primarily used to back up a nonreactive NST. When the NST is persistently nonreactive, and the CST result is positive, the false-positive rate is virtually nil, and delivery is almost always warranted.

The BPP is a more extensive biophysical assessment of fetal well-being. It includes five parameters that are scored as 2 points each as normal or present, or 0 as abnormal or absent:

Scores of 8 or 10 are considered normal and reassure the clinician that the fetus will not die or experience damage resulting from a chronic process for the next week. A score of 6 is equivocal and requires repeat testing in 1 day. Scores of 0 to 4 require further evaluation and consideration of delivery.

In large trial it became apparent that the NST alone was associated with a false-negative rate (fetal death within a week of a normal test) that was considerably higher than that of the CST or BPP (1% as opposed to 0.2%). The reason for this is that the loss of fetal movement, and thus reactivity, occurs very late in the process of fetal deterioration and death. Amniotic fluid volume generally declines well before reactivity is lost. Thus the combination of amniotic fluid volume assessment by ultrasound and the NST done weekly has become the test used by many clinicians in testing fetal well-being; the full BPP or the CST is used when the modified BPP result is abnormal.

Waveform analysis of umbilical artery flow using ultrasound-guided Doppler warns the clinician of increased resistance to flow within the placenta. This test is expressed as systolic-to-diastolic ratio. When the situation is severe enough, the flow during diastole either becomes absent or goes in the reverse direction, indicating marked resistance to flow. This form of testing is principally of value in the severely growth-restricted fetus and can give a very early warning of impending fetal demise.

A nonreassuring FHR monitoring optimally requires a backup method, except in extreme circumstances where the pattern is clearly indicative of hypoxia and acidosis. Scalp pH has been the gold standard. For various reasons, however (especially the technical difficulties involved in the procedure), recent surveys have indicated that fewer than 5% of clinicians in practice actually use this procedure and choose to err on the side of operative intervention when the FHR pattern is nonreassuring.

Neonatal depression is not always caused by hypoxia and acidosis. Furthermore, given the litigious environment surrounding the issue of perinatal brain damage, the issue of documenting the fetal blood gas status at birth is critical to an objective assessment of the baby’s condition. The following are indications for assessing blood gas status at birth:

Although some clinicians believe that all babies should have cord pH determinations, there is little medical value (babies with normal Apgar scores but low pH cord values have normal nursery and follow-up outcome), and it can be argued that a low pH with a normal Apgar could be more harmful than helpful in a medicolegal setting.

Virtually all drugs cross the placenta to some degree, but few produce any significant problems for either the fetus or the neonate. Large organic ions such as heparin and insulin do not cross the placenta and are therefore safe. There are some drugs taken by the mother, however, that can be problematic.

Although this list is relatively complete for many of the drugs known to produce significant fetal problems, the practitioner should always review the most recent medical literature for any updates that might reflect changes in awareness of potential risks of drugs during pregnancy. As was demonstrated by the maternal DES story, the full teratogenic potential of some medications may not be known for many years.