KEY QUESTIONS
What are the indications for inpatient amniocentesis?
How does patient counseling before amniocentesis change with the indication?
What steps should be followed to minimize the risk of complications associated with amniocentesis?
Amniocentesis is the procedure by which amniotic fluid is removed from the uterine cavity via a transabdominal approach (Fig. 57-1). While the vast majority of amniocenteses are performed in the outpatient setting by perinatologists in the previable period, there are indications for amniocentesis in the inpatient setting as well. Therefore the ability to perform an amniocentesis is a very useful tool in the obstetric and gynecologic (OB/GYN) hospitalist’s arsenal. This chapter focuses on reviewing the critical aspects of the amniocentesis procedure as it is performed for inpatient indications.
Amniotic fluid is an incredibly complex substance; the origins, composition, properties, and functions of which evolve dramatically over the course of the pregnancy. Prior to fetal skin keratinization, which occurs gradually between approximately 19 and 25 weeks gestation, amniotic fluid composition closely mirrors fetal plasma composition due to relatively free diffusion between the amniotic and fetal extravascular compartments. In the second half of pregnancy, amniotic fluid is produced largely by fetal urination, accounting for 75% to 80% of amniotic fluid volume, with the remainder contributed by secretions of oral, nasal, tracheal, and pulmonary fluid.1 The tests on amniotic fluid discussed in this chapter assess the properties of the secretions and sloughed cells from these compartments.
Overall, the most common purpose for amniocentesis by far is to obtain cells for fetal karyotype or genetic testing. As the implications and interpretation of these types of results are complicated and carry significant potential medicolegal complications, these tests generally should not be ordered unless the patient has undergone genetic counseling, effectively limiting such testing to the outpatient realm. Similarly, testing for evidence of fetal TORCH infection typically is performed only after consultation with a perinatologist and based on specific, concerning fetal findings on ultrasound. However, if a patient who has received counseling and declined amniocentesis later develops an additional indication for the procedure and elects to proceed, these studies may be ordered by an OB/GYN hospitalist in addition to the tests needed for the primary indication.
Amniocentesis procedures are generally divided into diagnostic and therapeutic categories. A diagnostic amniocentesis is performed to obtain amniotic fluid on which to perform diagnostic tests. A therapeutic amniocentesis is performed in order to relieve signs or symptoms caused by amniotic fluid volume and pressure. Inpatient indications that might reasonably arise within the scope of practice of an OB/GYN hospitalist are listed in Table 57-1.
| Diagnostic Indications |
|
| Therapeutic Indications |
Amnioreduction
|
Ultrasound-guided, intra-amniotic instillation of indigo carmine dye, with subsequent evaluation of a vaginal tampon or gauze for the presence of dye, remains the gold standard for diagnosing premature rupture of membranes. Because of its invasive nature, its use is limited to cases where standard, noninvasive evaluation for preterm premature rupture of membranes (PPROM) is inconclusive.2 Unfortunately, this test has recently become impossible to perform since production of the drug was discontinued due to a shortage of raw materials. As of July 2017, a new manufacturer (American Regent, Inc.) has begun producing indigo carmine, so the drug may again be available for use. Meanwhile, there are currently no safe alternatives to indigo carmine for this indication.3
In patients undergoing induction of labor after diagnosis of stillbirth, it is reasonable to offer amniocentesis to obtain cells for karyotype, especially if the patient receives epidural anesthesia and will not suffer any discomfort from the procedure. This is based on evidence that the chances of a successful karyotype improve dramatically if an amniotic fluid sample is provided, rather than a postnatal biopsy of fetal tissue, placenta, or cord. One such study of stillbirth after 12 weeks gestation compared 81 patients who underwent amniocentesis to 126 patients who had postnatal fetal skin biopsies sent, and found that 90% of amniocentesis specimens (but only 13.5% of skin biopsy specimens) resulted in a successful karyotype.4
However, a dilemma does exist when choosing between karyotype by amniocentesis and chromosomal microarray by postnatal biopsy (Table 57-2). The choice between these two approaches depends on multiple factors, including the likelihood of genetic abnormalities based on the presence or absence of fetal structural anomalies. Ideally, the decision should be made in consultation with a geneticist, a perinatologist, or both. (See Chapter 35 for more information about stillbirth.)
| Chromosomal Microarray | Karyotype | |
| Cost | Higher; may not be covered by insurance | Lower |
| Sample needed | Postnatal tissue biopsy (no proliferating cells needed) | Amniocentesis strongly preferred (needs proliferating cells for culture) |
| Ease of interpretation | Complicated; may require blood draws from parents | Relatively straightforward |
| Pitfalls | Variants of unknown clinical significance; may miss mosaicism, balanced translocations | Misses microdeletions, microduplications |
| Diagnostic yield—all stillbirths | 8.3% | 5.8% |
| Diagnostic yield—stillbirths with congenital anomalies | 29.9% | 19.4% |
The potential utility of amniocentesis for the detection of subclinical intra-amniotic infection in guiding the management of patients admitted with either preterm labor or PPROM is an area of active investigation. Amniotic fluid assays with rapidly available results predicting the presence or absence of intra-amniotic infection have two major advantages over amniotic fluid culture: First, the results are available in hours instead of days, and second, the results do not depend on the ability to culture atypical microbes such as Ureaplasma and Mycoplasma. Nonrandomized studies suggest that using amniocentesis to assess for intra-amniotic infection as part of the decision-making process for timing of delivery may improve neonatal outcomes in extremely premature patients (< 28 weeks gestation) and in cases of PPROM.5,6 However, overall data supporting improved neonatal outcomes when a decision to end the pregnancy is based on amniocentesis results suggesting intra-amniotic infection are very limited. Protocols surrounding whether to use amniocentesis in this way, and if so, which tests and thresholds to use to make a presumptive diagnosis for or to exclude the possibility of intra-amniotic infection (Table 57-3), vary by institution according to the availability of certain diagnostic tests and the opinions of local perinatologists and neonatologists.
| Predictive Value (%) | ||||
| Amniotic Fluid Tests | Sensitivity (%) | Specificity (%) | Positive | Negative |
| Gram stain | 63.6 | 99.1 | 87.5 | 96.4 |
| IL-6 ≥ 11.3 ng/mL | 100 | 82.6 | 36.7 | 100 |
| WBC ≥ 50/mm3 | 63.6 | 94.5 | 53.8 | 96.3 |
| Glucose ≤ 14 mg/dL | 81.8 | 81.6 | 31.0 | 97.8 |
| Gram stain + WBC ≥ 50/mm3 | 90.9 | 93.6 | 58.8 | 99.0 |
| Gram stain + glucose ≤ 14 mg/dL | 90.9 | 80.7 | 32.3 | 98.9 |
| Gram stain + IL-6 ≥ 11.3 ng/mL | 100 | 81.6 | 35.5 | 100 |
| Gram stain + glucose ≤ 14 mg/dL + WBC ≥ 50/mm3 | 90.9 | 78 | 29.4 | 98.8 |
| Gram stain + WBC ≥ 50/mm3 + IL-6 ≥ 11.3 ng/mL | 100 | 79.8 | 33.3 | 100 |
| Gram stain + glucose ≤ 14 mg/dL + IL-6 ≥ 11.3 ng/mL | 100 | 71.6 | 26.2 | 100 |
| Gram stain + WBC ≥ 50/mm3 + glucose ≤ 14 mg/dL + IL-6 ≥ 11.3 ng/mL | 100 | 69.7 | 25 | 100 |
Once relatively common, amniocentesis for assessment of fetal lung maturity has fallen rapidly out of favor over the past decade. Amniotic fluid testing for fetal lung maturity should never be performed in an attempt to justify an elective delivery prior to 39 weeks gestation. The American College of Obstetricians and Gynecologists (ACOG) recommends specifically against the routine use of amniocentesis to assess fetal lung maturity in decision-making regarding the timing of delivery for both well-dated and suboptimally dated pregnancies.7,8 These recommendations are based on retrospective studies suggesting that a mature pulmonary profile on amniocentesis is predictive of a lower risk only of the specific complication of respiratory distress syndrome (RDS) in the neonate, and that overall, it is a very poor surrogate for the assessment of fetal maturity.9–11 In addition, the risk of RDS in a late-preterm infant with amniotic fluid testing consistent with pulmonary maturity has been shown to be as high as 10%.10 Therefore the ability of fetal lung maturity testing to significantly affect the clinical assessment of the relative risks of continuing a pregnancy vs. delivery < 39 weeks gestation is very limited.
There is no evidence supporting the routine use of amnioreduction in the setting of polyhydramnios, as it is probably not effective in reducing the rates of preterm delivery in such situations. However, uncontrolled maternal abdominal pain, maternal dyspnea due to uterine distention by polyhydramnios, or both are accepted indications for amnioreduction, and available data suggests no increased risk of complications in this population compared to expectant management.12
Women presenting with painless cervical dilation prior to viability and in the absence of clinical evidence of infection or abruption may benefit from the placement of exam-indicated cerclage (also known as emergency or rescue cerclage).13 In a subset of these women with membranes prolapsing into the vagina, removal of approximately 250 mL of amniotic fluid via transabdominal amniocentesis at the time of cerclage placement may improve technical feasibility (and even outcomes) by allowing retraction of the membranes into the uterine cavity.14
