INTRODUCTION — Amniocentesis is a diagnostic technique for withdrawing amniotic fluid from the uterine cavity using a needle via a transabdominal approach. Laboratory tests to evaluate fetal health can be performed on amniotic fluid since this fluid is largely composed of fetal substances: urine, secretions, exfoliated cells, and transudate.
This topic will discuss technique and complications of diagnostic amniocentesis. Chorionic villus sampling is another commonly used invasive technique for fetal diagnosis and is reviewed separately. (See "Chorionic villus sampling".)
INDICATIONS — The most common diagnostic indications for obtaining amniotic fluid are prenatal genetic studies. Other indications include, but are not limited to, evaluation of the fetus for infection, degree of hemolytic anemia, blood or platelet type, hemoglobinopathy, and neural tube defects. Assessment of fetal lung maturity, which used to be a common indication for amniocentesis, is now rarely performed.
Amniocentesis is also performed as a therapeutic procedure to remove excess amniotic fluid, such as in symptomatic polyhydramnios or twin-twin transfusion syndrome, or to reduce volume and pressure of amniotic fluid in cases of prolapsed fetal membranes in the second trimester to facilitate placement of a physical examination-indicated cerclage . This procedure is termed amnioreduction and is discussed separately. (See "Twin-twin transfusion syndrome: Management and outcome", section on 'Amnioreduction' and "Transvaginal cervical cerclage", section on 'Replace prolapsed membranes, if present' and "Polyhydramnios: Etiology, diagnosis, and management", section on 'Amnioreduction (decompression amniocentesis)'.)
COUNSELING — Amniocentesis, like other potentially morbid procedures, should be preceded by appropriate counseling regarding the:
●Purpose of the procedure.
●Potential complications, including technical problems that might necessitate a second procedure.
●Time required before results will be available.
●Accuracy and limitations of the diagnostic test(s) planned, including possible inability to make a diagnosis.
●Alternatives that may yield the same or similar information.
Depending on the condition under investigation, invasive alternatives to amniocentesis may include chorionic villus sampling (CVS) and fetal blood sampling. (See "Chorionic villus sampling" and "Fetal blood sampling".)
Cell-free DNA in maternal blood is commonly used to screen for the most common autosomal and sex chromosome aneuploidies, and has substantially reduced the number of diagnostic amniocenteses performed for this indication (see "Prenatal screening for common aneuploidies using cell-free DNA"). Sampling maternal blood for cell-free DNA is a noninvasive method for diagnosing some DNA-based fetal conditions, primarily fetal Rh type when there is the potential for hemolytic disease of the fetus/newborn; other diagnostic applications remain investigational. (See "RhD alloimmunization in pregnancy: Management", section on 'Cell-free DNA testing'.)
Timing — Amniocentesis for prenatal genetic studies is technically possible at any gestational age after approximately 11 weeks of gestation but is optimally performed at 15+0 to 17+6 weeks of gestation. Procedures performed before 15 weeks (ie, early amniocentesis) are associated with higher fetal loss and complication rates, including culture failure, and should be avoided . Later second-trimester procedures are safe but can be problematic if termination of pregnancy is planned based upon abnormal results. Late second- and third-trimester procedures for genetic studies are performed in some cases, such as when fetal abnormalities are discovered late in gestation, because the information can be useful for counseling and preparing parents, as well as for planning the optimal time, route, and site of delivery.
Most of the cells floating in amniotic fluid are morphologically and biochemically characterized as epithelioid; fibroblastoid and amniotic fluid-specific cells are also present . Cells shed from the amnion and the lower fetal urinary tract comprise the largest proportion of proliferating amniotic fluid cells. Although the amniotic fluid contains more than 200,000 cells/mL at 16 weeks of gestation, only a small number of the floating cells (average 3.5±1.8 cells/mL of fluid) are capable of attaching to a culture substrate and yielding colonies. Cells derived from amniotic fluid before 15 weeks and at 24 to 32 weeks show a significant decline in cloning efficiency (fewer than 1.5 clone forming cells/mL fluid) . Indeed, laboratory-related failure rates of conventional G-band karyotyping increase significantly with advancing gestation so that, in the third trimester, karyotyping is more likely to fail than microarray .
Genetic tests — The approach to genetic evaluation of pregnancies with an anomalous fetus or positive results on aneuploidy screening can be found separately. (See "Prenatal genetic evaluation of the fetus with anomalies or soft markers" and "Down syndrome: Overview of prenatal screening", section on 'Screen-positive test result'.)
The choice of genetic test to be performed on the cells obtained via amniocentesis depends in part on the indication for testing. Chromosomal analysis (ie, karyotype) takes 7 to 14 days to obtain results. Interphase fluorescence in situ hybridization (FISH) provides a limited karyotype within 24 to 48 hours; the most frequently used probes detect aneuploidy of chromosomes 13, 18, 21, X, and Y, which are the most common causes of aneuploidy. The American College of Medical Genetics monograph Standards and Guidelines for Clinical Genetics Laboratories states that "abnormalities detected by interphase FISH that are also considered reliably detectable by chromosome analysis should be confirmed by conventional cytogenetic analysis" . (See "Tools for genetics and genomics: Cytogenetics and molecular genetics", section on 'Fluorescence in situ hybridization'.)
Use of chromosomal microarray increases diagnostic yield over conventional karyotyping because it can identify microdeletions and microduplications. It also has a faster turnaround time because studies can be performed directly on high quality DNA extracted from isolated cells so time for culture is not required (results take 3 to 5 days with direct testing and 10 to 14 days when cultured cells are used ). The lack of a requirement for culture increases the diagnostic yield in the evaluation of stillborns. (See "Prenatal diagnosis of chromosomal imbalance: Chromosomal microarray", section on 'Most common uses of CMA in prenatal diagnosis' and "Stillbirth: Maternal and fetal evaluation", section on 'Options for genetic testing'.)
Antibiotic prophylaxis — The efficacy of prophylactic antibiotics for reducing procedure-related pregnancy loss has not been evaluated extensively.
One trial randomly assigned 33,748 patients to receive either azithromycin 500 mg daily for three days before amniocentesis or no antibiotic therapy . All of the procedures were performed by a single operator. There were significantly fewer fetal losses in the four weeks after the procedure in the prophylaxis group than in the control group (0.03 versus 0.28 percent, relative risk [RR] 0.11, 95% CI 0.05-0.26). Nearly half of the fetal losses (21/43) occurred in the setting of preterm prelabor rupture of membranes, which also occurred less often in the prophylaxis group (0.06 versus 1.12 percent, RR 0.06, 95% CI 0.03-0.10).
However, there were several limitations to this trial, such as lack of follow-up until delivery, an unusually high rate of membrane rupture within four weeks of the procedure, an unusually high rate of fetal death after membrane rupture (14 percent) in the control group, and the use of a single dedicated operator performing all procedures (raising the issue of reproducibility of the results). Thus, it is unlikely that the conclusions from the trial will change the standard of care, which is not to use antibiotic prophylaxis prior to amniocentesis. Results from a case control study support this conclusion .
Site selection — An obstetric ultrasound examination is initially performed to determine fetal viability, position, biometry, location of the placenta, and anatomic survey.
The presence and location of an amniochorionic separation, which may be seen as late as 16 to 17 weeks of gestation, should also be noted if present. If the pregnancy is 15 to 16 weeks of gestation, we delay amniocentesis for a week, if possible, to allow more time for membrane fusion in these cases.
The optimal needle insertion site is then selected, avoiding the placenta if possible. Some studies have suggested a greater risk of fetal complications with transplacental amniocentesis [8-13]; however, this concept has been challenged by other studies [14-19].
If a transplacental approach is required, options include traversing the placenta or delaying the procedure by one week to allow the greater intrauterine volume with advancing gestation to potentially create a placenta-free window. If the first option is chosen, the needle should be directed through the thinnest portion of the placenta. Color flow mapping may be helpful to identify and avoid the umbilical cord insertion site and large chorionic vessels. If the maternal bladder interferes with the path of the needle, the patient should void prior to needle insertion.
Site preparation — The maternal abdomen is prepared with an antiseptic solution and draped. It is prudent to prepare a larger area of maternal abdomen than the site identified for needle insertion since fetal movements may change the location of the optimal pocket of fluid.
Local anesthesia — Local anesthesia is optional and usually unnecessary, as most patients have no or mild discomfort, administration of anesthetic is itself somewhat painful, and no intervention has been proven to reduce this discomfort [20-23]. Risk factors for pain at amniocentesis include maternal anxiety, needle insertion into the lower part of the uterus, history of menstrual cramps, and previous amniocentesis .
When our patients request anesthesia, we tell them that amniocentesis is associated with two moments of discomfort, one at skin insertion and the other when the needle crosses the uterine muscle; local anesthesia may relieve the former sensation, but is ineffective for the latter .
Needle — A 20 or 22 gauge spinal needle is generally used for amniocentesis. In a randomized trial, use of the larger bore needle lowered the risk of intrauterine bleeding if the needle was passed through the placenta and allowed faster fluid retrieval compared with use of the 22 gauge needle but caused more maternal discomfort . The authors hypothesized that the unexpected reduction in bleeding with a large bore needle may be related to the shorter duration of the procedure.
The standard length of a spinal needle is 8.9 cm (excluding the hub), but longer needles are available (15 cm). The length of the needle should take into account the thickness of the maternal panniculus, location of the target pocket of fluid, and the possibility that intervening events, such as uterine contractions, may increase the distance between the skin and target.
Needle insertion — The use of concurrent ultrasound guidance rather than pre-amniocentesis ultrasound evaluation has not been associated with a reduced rate of fetal losses in controlled studies . Nevertheless, we believe ultrasonographic monitoring with continuous visualization of the needle should be performed throughout the procedure to avoid direct fetal injury (see below) and to reduce the number of punctures and the incidence of bloody fluid .
The probe is either held by an assistant in an area of skin away from that prepared with antiseptic or, more commonly, it is inserted in a sterile pouch (eg, a glove). Nonsterile gel is placed in the pouch in contact with the probe and sterile gel is placed on the outer surface of the pouch in contact with the maternal skin.
A "free-hand technique" is preferred by most operators because it permits adjustments in the path of needle insertion. However, a needle-guiding device attached to the transducer may facilitate reaching the optimal pocket of fluid, particularly with less experienced operators. Most real-time ultrasound machines are equipped with an on-screen template of the needle tract that is used to target the sampling site.
The maternal bowel and bladder, if identified, should be excluded from the path of the needle. Traversing the bowel, which is rare, can introduce bowel flora into the amniotic fluid, leading to infection and pregnancy loss.
Most operators prefer to insert the needle while holding the transducer themselves; once the needle tip is in place, an assistant holds the transducer and maintains visualization of the needle tip while the operator draws amniotic fluid samples. Other operators prefer to be guided throughout the procedure by an assistant.
The operator should be familiar with the appearance of the needle on the screen. If a "parallel" approach is taken to needle insertion, the needle enters the skin at the end of a curvilinear or sector probe and can be imaged along its entire extent from skin to target location (image 1). If a "perpendicular" approach is taken, the needle is inserted adjacent to the middle of a linear or curvilinear probe, and only the area of the needle tip is imaged.
The needle should not be advanced if the position of its tip cannot be identified on the screen. Using the operator's wrist, a sudden thrust of the needle as it passes from the uterine muscle into the amniotic cavity may help to avoid membrane tenting (see below). Entrance into the amniotic cavity is often accompanied by a tactile popping sensation.
Dry tap — Amniotic fluid typically appears at the needle hub when the stylet is removed. If no amniotic fluid is visible at removal of the stylet, fetal membranes may have tented over the needle tip, thus preventing needle penetration into the amniotic cavity (ie, impingement of the membranes without piercing them). This occurs more often with insertions prior to 15+0 weeks of gestation due to incomplete physiological "fusion" of the amnion, chorion, and decidua parietalis (image 2).
The needle may be rotated to redirect the bevel and puncture the membranes. However, if this maneuver is unsuccessful and the needle must be repositioned, the stylet should be reinserted before further thrusts are made. These subsequent attempts at redirecting the needle should be guided ultrasonographically.
Alternatively, the needle may be withdrawn within the maternal abdomen and redirected to an area of the uterus where there is no amniochorionic separation. A second needle insertion may also be attempted.
If a dry tap occurs, many clinicians will postpone another attempt until a week has passed, although procedure-related fetal losses are not related to the number of punctures .
Failure to obtain amniotic fluid after 15 weeks of gestation by experienced operators using ultrasound guidance occurs in less than 1 percent of patients [8,13,26]. The rate of failure to aspirate fluid was much higher in older studies that did not use ultrasound guidance.
Withdrawing and handling the sample — The initial drop of amniotic fluid may contain maternal cells that stuck to the needle during passage through the maternal abdominal and uterine walls. Most operators discard the initial sample of fluid containing this drop because inclusion with the sample for testing may be responsible for low level mosaicism with maternal cells in cytogenetic studies. Others send the initial sample of fluid for amniotic fluid alpha-fetoprotein determination and then change syringes before aspirating more fluid for other fetal diagnostic studies.
Twenty to 30 mL of amniotic fluid are aspirated into sterile syringes by gentle traction on the barrel or by use of vacuum tubes . The amount of amniotic fluid withdrawn should be based upon gestational age since amniotic fluid volume increases across pregnancy (figure 1). One general rule for removing an appropriate amount of amniotic fluid is to remove a volume in milliliters equivalent to the gestational age in weeks (eg, 15 mL at 15 weeks).
A few case reports have described growth inhibition in cell cultures of fluid obtained by amniocentesis due to prolonged contact of the fluid with the syringe stopper (the black, rubber-like, grommet on the end of the plunger rod). To minimize this risk, we recommend keeping the syringe in a tip down position until the fluid is transferred to the culture tubes, and minimizing the time that the amniotic fluid is kept in the syringe.
Postprocedural care — The fetal heart rate should be assessed sonographically after the procedure and recorded. Uterine cramping, transient spotting, and vaginal loss of a few drops of amniotic fluid may occur immediately after the procedure. The patient should be instructed to report any persistent loss of vaginal fluid or bleeding, severe uterine cramping lasting for several hours, or fever. Limitation of physical or sexual activity after the procedure is unnecessary.
Electronic fetal and uterine monitoring following procedures performed after fetal viability are an option to assess for uterine contractions triggered by the procedure and to provide reassurance regarding fetal status. However, we believe that sonographic demonstration of normal fetal heart rate and activity postprocedure is adequate in most cases.
Nonalloimmunized RhD-negative patients should receive Rh(D) immune globulin (eg, RhoGAM) after the procedure to prevent RhD sensitization. The American College of Obstetricians and Gynecologists recommends a dose of 300 mcg. (See "RhD alloimmunization: Prevention in pregnant and postpartum patients".)
●Blood – Blood-tinged amniotic fluid is aspirated in less than 1 percent of amniocenteses when the procedure is performed under ultrasound guidance [8,26]. The blood is almost always maternal and does not adversely affect amniocyte growth. Two studies have reported an increased risk of fetal loss after bloody taps (odds ratios of 2.2 and 6.5) [9,16]. In contrast, a series of 5948 procedures performed under ultrasound guidance by the same operator reported a 3.7 percent rate of "hemorrhagic" amniotic fluid, and found no association between bloody fluid and fetal loss up to 24 weeks .
●Green or brown fluid – The presence of green or brown pigment in second-trimester amniotic fluid is more common (approximately 2 percent) than gross blood (picture 1) [8,26]. This type of discoloration indicates intraamniotic hemorrhage antedating amniocentesis [28,29], since the fetus does not pass pigmented meconium in the first half of pregnancy.
Green or brown fluid is associated with a higher rate of miscarriage or fetal death than clear fluid [29-32]. As an example, one randomized trial reported the relative risk of miscarriage of a euploid fetus after amniocentesis was increased 10-fold (95% CI 4.3-22.6) if discolored fluid was withdrawn . The rate of miscarriage was particularly high (occurring in one-third of pregnancies) if there were both abnormally colored fluid and an unexplained high maternal serum alpha-fetoprotein.
Discolored amniotic fluid is more likely to be associated with chromosomal abnormalities , culture failure , and microbial contamination of the amniotic fluid (particularly Mycoplasma species) than clear fluid [35,36]. Detection of Mycoplasma may indicate chronic endometrial infection, and possibly a higher risk of premature rupture of membranes  and preterm delivery . Antibiotics have been reported to be useful in these cases, but data are insufficient to allow a recommendation for routine culture of discolored amniotic fluid and treatment of positive results [38-40].
COMPLICATIONS AND ADVERSE OUTCOMES — The safety of genetic amniocentesis has been addressed by several case-control studies [11,26,41-43] and one randomized trial . The major complications of amniocentesis are rupture of membranes, direct fetal injury, indirect fetal injury, infection, and fetal loss. Maternal complications related to the procedure, such as amnionitis, are rare, occurring in less than 1/1000 procedures.
The following data regarding procedure-related complications of amniocentesis were derived from series in which ultrasound guidance was used concurrently to visualize needle placement. There is no evidence that specific techniques (eg, uterine relaxants) are associated with a reduced risk of complications .
Leakage of amniotic fluid — Temporary loss of amniotic fluid occurs more frequently in pregnancies that undergo amniocentesis than among controls not experiencing amniocentesis (1.7 versus 0.4 percent of pregnancies) . Fluid loss is almost always a small volume and usually spontaneously stops within one week, with reaccumulation of fluid and restoration of normal amniotic fluid volume in an average of three weeks (range one to seven weeks) [45,46]. Cessation of leakage is probably not due to actual repair and regeneration of the membranes, but rather to changes in the decidua and myometrium that block further leakage . Temporary loss of fluid is generally associated with a good pregnancy outcome .
Chronic leakage for the duration of pregnancy (ie, gross membrane rupture) rarely occurs but can also be associated with a good outcome [48,49]. In a study comparing preterm prelabor rupture of membranes after amniocentesis (PPROMa) with spontaneous preterm prelabor rupture of membranes (sPPROM), patients experiencing PPROMa delivered at more advanced gestational ages than those with sPPROM (34.2 versus 21.6 weeks), despite similar gestational ages at membrane rupture; perinatal survival was also higher (91 versus 9 percent) .
Given the high potential for a good outcome, conservative management with serial monitoring of amniotic fluid volume, fetal growth, and signs of maternal infection is an appropriate and reasonable option. However, patients with chronic fluid loss should be apprised of the increased risks of preterm delivery, skeletal deformation, and pulmonary hypoplasia, especially if anhydramnios is present. (See "Prelabor rupture of membranes before and at the limit of viability", section on 'Pediatric outcomes'.)
Chorioamniotic separation — Chorioamniotic separation can occur after amniocentesis. However, it does not appear to affect pregnancy outcome unless it extends along the entire chorioamniotic surface .
Direct fetal injury — Direct fetal needle injury is rare during amniocentesis under ultrasound guidance [8,41,51,52]. In a randomized trial including 2239 pregnancies in which ultrasound guidance was used for amniocentesis, no direct fetal injuries occurred .
Fetal injuries that have been attributed to midtrimester amniocentesis, primarily in isolated case reports, include exsanguination, skin dimples, ocular injury, and intracranial and bowel abnormalities . These attributions have mostly been based on association rather than direct evidence.
Indirect fetal injury — A few prospective studies have reported an increased rate of orthopedic malformations and respiratory problems in infants born after genetic amniocentesis compared with controls who did not undergo amniocentesis. The reported odds of talipes equinovarus or congenital hip dislocation in cases versus controls range from odds ratio (OR) 0.6-6.1; pooled data from over 10,000 pregnancies show the frequency of these problems in cases and controls are 0.76 and 0.56 percent, respectively [8,41,53]. For the risk of respiratory distress, the corresponding odds range from OR 2.1-3.4 and the pooled actual frequencies in cases and controls are 1.2 and 0.45 percent, respectively.
The mechanism underlying both complications is thought to be fetal compression as a consequence of decreased amniotic fluid. Support for this hypothesis comes from observations that postural malformations are more common after early amniocentesis, when the amniotic fluid volume is less than later in gestation, and following persistent amniotic fluid leakage after amniocentesis . In addition, there appears to be a correlation between the amount of amniotic fluid removed at amniocentesis and the risk of talipes . Finally, talipes and suboptimal lung growth can be reproduced experimentally in animals undergoing amniocentesis .
A cohort study of consecutive live born offspring of patients who had amniocentesis and matched controls whose mothers had not had amniocentesis found that exposure to amniocentesis was not associated with a significantly increased rate of major disabilities at follow-up (7 to 18 years after birth) .
Taken together, these studies show an increased risk of transient indirect fetal injury; the possibility of these risks should be addressed during the process of informed consent. Even if the increased risk is statistically significant, the absolute excess risk is small and may be minimized by avoiding the removal of an excess amount of fluid for gestational age (ie, amount of fluid in mL should not exceed the gestational age in weeks) and the performance of early (<15+0 weeks) procedures.
Vertical transmission — Case reports have described mother-to-infant transmission of infections, such as hepatitis virus, cytomegalovirus, toxoplasmosis, and HIV, where transmission was presumed to be related to amniocentesis [58,59]. Indirect evidence also suggests higher rates of neonatal infection in the offspring of chronically infected patients who undergo amniocentesis compared with controls who do not have the procedure [60-62].
Before highly active antiretroviral therapy (HAART), invasive prenatal diagnostic procedures were avoided in patients with HIV infection because of concerns about mother-to-child transmission. Available data on frequency of transmission were limited by very small and heterogeneous samples. Subsequently, three series of prenatal diagnostic procedures in patients on HAART reported no cases of vertical transmission, but the combined series included <200 such patients [62-64]. Guidelines recommend performing invasive prenatal diagnostic procedures after HAART has been started and viral load is undetectable, if possible . (See "Prenatal evaluation of women with HIV in resource-rich settings", section on 'Invasive diagnostic procedures'.)
In large controlled studies, amniocentesis was associated with increased vertical transmission rates of hepatitis B virus (HBV) in patients who were highly viremic (HBV DNA levels ≥7 log(10) copies/mL) and hepatitis B e-antigen (HBeAg) positive [66,67]. Antiviral therapy appeared to reduce the risk of mother-to-child transmission in highly viremic patients . (See "Pregnancy in women with pre-existing chronic liver disease".)
No association between amniocentesis and vertical transmission of hepatitis C virus (HCV) was found in two small case series . Limitations of the published data include small sample sizes and lack of addressing the potential impact of viral load on vertical transmission.
It is biologically plausible that mother-to-child transmission can occur during amniocentesis, particularly in patients with high viremic loads and if the needle crosses the placenta. Epidemiologic evidence is scant. Until more data become available on the safety of amniocentesis in such patients, patients should be counseled that data on the risk of vertical transmission are reassuring but limited, informed of the potential risk of fetal infection, and offered alternatives for prenatal diagnosis of aneuploidy using tests with high sensitivity and low false-positive rates. The latter include cell-free DNA testing in maternal blood, combined screening (nuchal translucency measurements and biochemical markers) between 11 and 14 weeks of gestation, and/or second-trimester maternal serum biochemical screening prior to proceeding with amniocentesis. (See "Prenatal screening for common aneuploidies using cell-free DNA".)
Inoculation by bowel flora — As discussed above, traversing the bowel, which is rare, can introduce bowel flora into the amniotic fluid leading to intrauterine infection, pregnancy loss and, rarely, septic shock.
Cell culture failure — Amniocytes fail to grow in culture in 0.1 percent of samples .
Mosaicism — True mosaicism is defined as one or more abnormal cell lines plus a normal cell line in at least two primary cultures from the same individual; it occurs in 0.1 to 0.2 percent of pregnancies undergoing amniocentesis [70,71]. Pseudomosaicism (ie, an abnormal cell line confined to one culture flask) is more common, occurring in up to 8 percent of pregnancies .
When mosaicism is detected on amniocentesis, fetal blood sampling can be performed to see if the same abnormal cell line is present in fetal blood, thus confirming the fetal mosaicism. However, even if the fetal blood sampling result is normal, the abnormal cell line may be present in other fetal tissues, such as skin. In either case, phenotype can be difficult to predict; we recommend consultation with a geneticist and serial ultrasound examinations of the fetus.
Fetal loss — Our interpretation of published data is that the procedure-related risks of amniocentesis are very small: probably around 0.2 to 0.3 percent when performed by a skilled operator under ultrasound guidance. The American College of Obstetricians and Gynecologists' practice bulletin on prenatal diagnostic testing for genetic disorders cites a procedure-related pregnancy loss rate of 0.1 to 0.3 percent when performed by experienced operators .
In the author's review of published controlled studies with concurrent real-time ultrasound guidance, the excess rate of spontaneous fetal loss attributed to the amniocentesis procedure (ie, procedure-related loss rate) ranged from 0.06 to 1.0 percent (0.6/1000 to 10/1000) (table 1). An increase in the rate of fetal loss was observed in most study groups following amniocentesis compared with controls. It should be noted that different studies used different definitions of fetal loss; as an example, the FASTER trial  and other studies  used fetal loss at less than 24 weeks of gestation as the criterion, whereas other studies included all fetal losses at less than 28 weeks  or all losses until delivery [16,73].
Most fetal losses occur within the four weeks following amniocentesis [8,72,73]. The cause of such losses remains unknown because no study has consistently examined the placenta and fetus in these cases.
A variety of factors has been studied to determine whether they affect the risk of procedure-related fetal loss:
●Operator experience may not play a significant role in the fetal loss rate [8,74,75], but it may affect sample quality [76,77].
●Most series report that the number of punctures is not independently associated with an increased risk of fetal loss [13,16,19,31]. However, one contemporary study found that the number of procedures and attempts significantly increased the risk of fetal loss from 0.3 percent for 1 procedure and 1 attempt to 5.2 percent for >1 procedure and >1 attempt .
●Several indications for amniocentesis are themselves risk factors for fetal losses and may explain the higher fetal loss rates after amniocenteses performed in those clinical settings:
•Higher loss rates were observed when an elevated maternal serum alpha-fetoprotein (MSAFP) value was the indication for amniocentesis. In one study, the relative risk for fetal loss was 8.3 (95% CI 2.4-19.8) after amniocentesis performed for MSAFP greater than 2.0 multiples of the median (MoM) .
•Presence of fetal anomalies as indication for diagnostic amniocentesis is associated with increased risk of fetal loss .
•Vaginal bleeding during the current pregnancy and history of abortion (spontaneous or induced) are additional risk factors for spontaneous losses after amniocentesis. In one report, vaginal bleeding prior to the procedure was associated with a significantly higher rate of spontaneous fetal loss before 28 weeks of gestation compared with controls without vaginal bleeding (5.9 versus 3.8 percent) . A history of three or more abortions in the first trimester or one or more abortions in the second trimester was also associated with a higher rate of pregnancy loss after amniocentesis (6.9 versus 3.5 percent in controls without a history of abortion) . The observation has been confirmed by a large cohort study .
Limitations of available data — Although amniocentesis is one of the most commonly performed procedures for prenatal diagnosis, a precise assessment of its safety is elusive for several reasons. Randomized trials are ideally positioned to provide the best estimates of the procedure-related risk of miscarriage, fetal death, and preterm delivery after amniocentesis. However, only one such trial has been conducted . In this trial, outcome of pregnancy was studied in 4606 patients 25 to 34 years of age without known risk of genetic disease randomly assigned to undergo amniocentesis or routine care (ultrasound and determination of MSAFP). In both groups, 99 percent of the participants followed the examination schedule of the group to which they had been allocated. The total spontaneous abortion rate was 1.7 percent in the study group and 0.7 percent in the control group (relative risk 2.3, 95% CI 1.3-4.0). Eleven of the 15 pregnancies with chromosomal aneuploidies in the study group were terminated, whereas in the control group, only one chromosomally abnormal pregnancy was diagnosed and terminated, suggesting that the 1 percent excess risk of spontaneous fetal loss in the amniocentesis group is underestimated.
Several large case-control and cohort studies have attempted to fill the gap in information on amniocentesis-related risks and complications, and their findings have undergone several meta-analyses. However, biases may occur in such studies both among cases and controls. The following factors are major biases in cases. The indication for amniocentesis may select patients at increased risk for adverse outcome. Indeed, the rate of miscarriage and other complications is significantly affected by maternal age [78-80] and, to a lesser extent, by other maternal factors (eg, abnormal serum screening results) that may constitute indications for the procedure. Another bias is that, in some series, outcome information was not available in all cases, the rate of loss to follow-up was significantly different between cases and controls , or cases lost to follow-up were excluded from analysis. This introduces a bias since pregnancies that are the most difficult to follow up are often associated with fetal loss .
The following factors are major biases in controls. In all of the nonrandomized studies, fetal losses in controls included both normal and abnormal karyotypes, whereas those after amniocentesis mainly involved euploid fetuses (since aneuploid pregnancies were usually terminated). This bias leads to an underestimation of the post-amniocentesis spontaneous fetal loss rate and it explains the paradoxical finding of the FASTER trial, in which amniocentesis was associated with significantly lower odds of spontaneous fetal loss before 24 weeks than controls who did not undergo amniocentesis (adjusted OR 0.4, 95% CI 0.3-0.7) . Not surprisingly, the rate of termination of pregnancy was significantly higher among patients undergoing amniocentesis than in those who did not (2.9 versus 0.2 percent).
In the FASTER trial, some patients who were screen negative in both trimesters elected to undergo amniocentesis anyway . This subpopulation of patients with screen-negative results represents a group in which underestimation of procedure-related spontaneous losses before 24 weeks is minimized because these patients would be expected to have a low risk of fetal aneuploidy and thus a low rate of active pregnancy termination. The excess spontaneous loss rate with amniocentesis in screen-negative patients was 0.15 percent (95% CI -0.25 to 0.81) or a procedure-related loss rate of 1/600 to 1/700 procedures. Of note, certain aneuploidies with increased risk of spontaneous midtrimester loss or active pregnancy termination, such as some cases of trisomy 21, 18 and 13 and Turner syndrome, would be expected to be present among screen-negative patients.
The choice of the control population also differs from study to study: In one study, the authors chose to compare the observed rate of fetal losses after amniocentesis in their cohort with the expected rate of spontaneous losses based on published data adjusted for ethnicity and maternal age . Some studies used control groups consisting of gestations that did not undergo second-trimester ultrasound scan to confirm fetal viability and exclude major anomalies [18,81].
Meta-analyses have reported procedure-related rates of pregnancy loss from 0.3 to 1.0 percent, depending on whether only randomized trials were included or cohort studies were included as well [82,83]. Meta-analyses inclusive of nonrandomized controlled studies are plagued by significant heterogeneity among studies, which accounts for 88 percent of the observed variation . Use of a random-effect model to pool the results cannot resolve the methodologic differences between studies .
Obstetric complications — Studies that have examined whether exposure to first- or second-trimester amniocentesis is associated with an increased risk of subsequent obstetric complications have generally found no significant increase in risk of pregnancy-related hypertension , low birth weight or very low birth weight , abruption , preterm delivery, preterm prelabor rupture of membranes (PPROM), stillbirth, neonatal mortality, or perinatal mortality (table 2).
Third-trimester amniocentesis — The safety and efficacy of third-trimester, ultrasound-guided amniocentesis is illustrated by two controlled studies:
●In one study, patients undergoing amniocentesis for fetal lung maturity studies at >32 weeks were matched with controls undergoing antepartum testing based on gestational age at testing and maternal age . There was no significant difference between groups in the rate of obstetric complications within 48 hours of testing (0/167 versus 1/167).
●In the other study, the outcome of pregnancy in 91 patients with PPROM who underwent amniocentesis was compared with the outcome of 46 controls with PPROM who did not undergo the procedure . There was no significant difference between groups in interval to delivery.
The safety of third-trimester amniocentesis is also supported by several uncontrolled observational studies [90-93]. These studies reported low (<1 percent) rates of obstetric complications proximate to the procedure. In studies in which third-trimester amniocentesis was performed for genetic testing, small sample size and confounding variables related to indications for amniocentesis affect the interpretation of the safety of the procedure [93-98]. One case-control study found that corticosteroid administration prior to amniocentesis did not reduce the rates of neonatal complications, which were similar in both groups and not higher than the general population .
Multineedle technique — Most operators perform separate procedures on each sac for genetic studies of multiple gestations, using separate and sequential insertion of a new needle for each amniotic cavity. The multi-needle technique does not appear to increase the risk of adverse outcome compared with the single needle technique .
Sampling only one sac usually will be adequate in monochorionic twins with no sonographic evidence of abnormalities and with concordant growth , but reports of discordant karyotype and microarray in monozygotic twins have been described so the one specimen approach is fallible [101,102].
With the multi-needle technique, amniotic fluid is aspirated from one sac, and blue indigo carmine dye (2 to 3 mL) is injected prior to withdrawal of the needle. If indigo carmine is not available, methylene blue should not be used because it can cause methemoglobinemia in the neonate, increase the risk of small bowel atresia, and stain the skin. Alternative agents, including indocyanine green (unknown dose) and fluorescein (2 to 5 mL of 10 percent solution), have been proposed, but there is no published information on their use and safety in obstetrics . If a marker dye is available, amniocentesis is subsequently performed in the other sac by inserting a new needle and aspirating amniotic fluid, which should be free of dye, thus confirming that a different sac has been sampled. Experienced operators often feel comfortable performing the multi-needle technique without dye injection, particularly when the dividing membranes are clearly visualized. However, misdiagnosis using such a technique has been reported in as many as 3.5 percent of samples .
If triplets or higher order gestations are present, indigo carmine is commonly injected into the second sac after obtaining the amniotic fluid sample from that sac and then a third needle is inserted into the third sac (and so forth until each sac has been sampled). Withdrawal of clear fluid confirms that a new sac has been successfully entered, while aspiration of blue-tinged fluid indicates that a previously tapped sac was reentered.
Single needle technique — Aspiration of both twin sacs using a single needle insertion into the uterus is an alternative technique [104-107]. This technique reduces maternal discomfort.
The needle is inserted into the more anterior sac, choosing a pocket of fluid that is contiguous to the deepest twin's sac. After aspiration of fluid from the first sac, the needle is advanced under ultrasound guidance through the dividing membrane into the second sac. No dye is used for confirmation of the origin of the fluid.
Theoretic concerns with this procedure include the potential for mixing the two fluids, leading to inaccurate interpretation of genetic tests, and shearing of the dividing membrane, resulting in a functionally monoamniotic pregnancy with its attendant risks. Placental location or fetal position may make this procedure either too difficult or too risky. Data using a single needle insertion are too limited to allow an adequate assessment of the associated risks and safety of this approach. (See "Twin pregnancy: Labor and delivery", section on 'Monochorionic/monoamniotic twins'.)
Loss rate — The risk of amniocentesis-related loss in twin pregnancies is likely increased by approximately 1 percent above the baseline risk of loss among twin gestations, but the exact risk is uncertain [108-110]. The American College of Obstetricians and Gynecologists estimates a procedure-related loss rate of 1 percent in twins .
In two systematic reviews, the rate of total pregnancy loss in twin pregnancies undergoing amniocentesis was higher than in those not undergoing amniocentesis (one review: 3.9 versus 3.1 percent, pooled odds ratio 1.5, 95% CI 1.0-2.1 ) [108,110]. While the pooled loss rate for one or both fetuses before 24 weeks of gestation following amniocentesis was similar to the natural loss rate in twin pregnancies in both reviews, these results are difficult to interpret because of variations in the definition of pregnancy loss (fetal death or miscarriage/delivery), whether one or both fetuses were lost, and the timing of loss after the procedure.
There are inadequate data from large, well-controlled studies to conclude whether the risk of early procedure-related fetal loss is higher in twin than singleton pregnancies undergoing amniocentesis. Confounders that need to be addressed include maternal age, chorionicity, ultrasound findings, and maternal analyte screening results. Other important factors include the completeness of ascertainment of pregnancy outcome and how fetal loss is defined (all losses up to term versus losses up to 24 weeks versus losses within 4 weeks of the procedure).
The only two observational studies that compared amniocentesis with CVS in twin pregnancies did not find a significant difference between techniques in pregnancy loss rate [111,112].
Amniocentesis for signs of infection — Only the lower-most (presenting) sac needs to be sampled in procedures done in the late second or early third trimester for evaluation of infection in preterm prelabor rupture of membranes or preterm labor, since infection generally follows an ascending route from the vagina.
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Amniocentesis (The Basics)")
●Beyond the Basics topics (see "Patient education: Amniocentesis (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●Amniocentesis should be performed after 15 weeks of gestation because earlier procedures are less likely to be successful, are associated with higher rates of cell culture failure, and carry greater fetal risks. (See 'Prenatal diagnosis' above.)
●We suggest not using antibiotic prophylaxis for amniocentesis (Grade 2C). (See 'Antibiotic prophylaxis' above.)
●The true rate of pregnancy loss associated with amniocentesis is uncertain. Our interpretation of published data is that the procedure-related risks of amniocentesis are very small: probably around 0.2 to 0.3 percent when performed by a skilled operator under ultrasound guidance. The American College of Obstetricians and Gynecologists practice bulletin on prenatal diagnostic testing for genetic disorders cites a procedure-related pregnancy loss rate of 0.1 to 0.3 percent when operators are experienced. (See 'Fetal loss' above.)
●Green or brown fluid is associated with an increased risk of subsequent spontaneous abortion or fetal death compared with clear fluid. Chromosomal abnormalities, culture failure, and microbial contamination of the amniotic fluid (particularly Mycoplasma species) are also more common when the fluid is discolored. (See 'Discolored samples' above.)
●The risks of direct fetal injury or maternal injury during amniocentesis are minimal. There may be a small risk of indirect fetal injury, such as orthopedic abnormalities and respiratory problems. (See 'Direct fetal injury' above and 'Indirect fetal injury' above.)
●Temporary leakage of amniotic fluid occurs more frequently in pregnancies that undergo amniocentesis than among controls (1.7 versus 0.4 percent), but is generally associated with a normal pregnancy outcome. (See 'Leakage of amniotic fluid' above.)
●Mother-to-child transmission of infection may occur in patients with viremia/bacteremia at the time of the procedure. Although the risk is low, noninvasive methods of prenatal fetal risk assessment are preferable, using tests with high sensitivity and low false-positive rates. (See 'Vertical transmission' above.)
●When performing amniocentesis in a multiple gestation, we suggest insertion of a new needle for each amniotic cavity. The risk of amniocentesis-related loss in twin pregnancies is likely increased above the baseline risk of loss among twin gestations, but the exact risk is uncertain. (See 'Multiple gestation' above.)
1 : Amnioreduction in emergency cerclage with prolapsed membranes: comparison of two methods for reducing the membranes.
6 : Antibiotic prophylaxis before second-trimester genetic amniocentesis (APGA): a single-centre open randomised controlled trial.
9 : Spontaneous abortion following mid-trimester amniocentesis. Clinical significance of placental perforation and blood-stained amniotic fluid.
10 : Procedure-related fetal losses in transplacental versus nontransplacental genetic amniocentesis.
11 : Incidence of spontaneous abortion after amniocentesis: influence of placental localization and past obstetric and gynecologic history.
13 : Amniocentesis-related adverse outcomes according to placental location and risk factors for fetal loss after midtrimester amniocentesis.
17 : Transplacental needle passage and other risk-factors associated with second trimester amniocentesis.
21 : Does local anesthesia at mid-trimester amniocentesis decrease pain experience? A randomized trial in 220 patients.
24 : Comparison between 20G and 22G needle for second trimester amniocentesis in terms of technical aspects and short-term complications.
28 : Significance of brown and/or green amniotic fluid at the time of second trimester genetic amniocentesis.
29 : Analysis of the significance of discolored amniotic fluid detected at midtrimester amniocentesis.
31 : Revisiting the fetal loss rate after second-trimester genetic amniocentesis: a single center's 16-year experience.
35 : Isolation of Mycoplasma hominis and Ureaplasma urealyticum from amniotic fluid at 16-20 weeks of gestation: potential effect on outcome of pregnancy.
37 : Mycoplasma hominis and Ureaplasma urealyticum in midtrimester amniotic fluid: association with amniotic fluid cytokine levels and pregnancy outcome.
39 : Ureaplasma/Mycoplasma-infected amniotic fluid: pregnancy outcome in treated and nontreated patients.
41 : Midtrimester amniocentesis: is it safe? A single centre controlled prospective study of 517 consecutive amniocenteses.
44 : Technique modifications for reducing the risks from amniocentesis or chorionic villus sampling.
47 : Fetal membrane healing after spontaneous and iatrogenic membrane rupture: a review of current evidence.
50 : Chorioamniotic separation after second-trimester genetic amniocentesis: importance and frequency.
53 : Infant morbidity following amniocentesis and chorionic villus sampling for prenatal karyotyping.
59 : Amniocentesis as a possible risk factor for mother-to-infant transmission of hepatitis C virus.
60 : Obstetric factors and mother-to-child transmission of human immunodeficiency virus type 1: the French perinatal cohorts. SEROGEST French Pediatric HIV Infection Study Group.
62 : Amniocentesis and mother-to-child human immunodeficiency virus transmission in the Agence Nationale de Recherches sur le SIDA et les Hépatites Virales French Perinatal Cohort.
63 : Amniocentesis and chorionic villus sampling in HIV-infected pregnant women: a multicentre case series.
66 : Risk of vertical transmission of hepatitis B after amniocentesis in HBs antigen-positive mothers.
67 : Mother-to-child transmission of hepatitis B virus after amniocentesis: A retrospective matched cohort study.
68 : RETIRED: Prenatal invasive procedures in women with hepatitis B, hepatitis C, and/or human immunodeficiency virus infections.
70 : European collaborative study on prenatal diagnosis: mosaicism, pseudomosaicism and single abnormal cells in amniotic fluid cell cultures.
71 : Proposed guidelines for diagnosis of chromosome mosaicism in amniocytes based on data derived from chromosome mosaicism and pseudomosaicism studies.
74 : Importance of complete follow-up of spontaneous fetal loss after amniocentesis and chorion villus sampling.
77 : Factors associated with maternal cell contamination in amniocentesis samples as evaluated by fluorescent in situ hybridization.
81 : Risk of amniocentesis in women screened positive for Down syndrome with second trimester maternal serum markers.
82 : Risk of miscarriage following amniocentesis or chorionic villus sampling: systematic review of literature and updated meta-analysis.
84 : Trials and tribulations of meta-analyzing procedure-related risks of amniocentesis and chorionic villus sampling.
85 : Invasive procedures for fetal karyotyping: no cause of subsequent gestational hypertension or pre-eclampsia.
86 : Second trimester amniocentesis is not a risk factor for very low birth weight and extremely low birth weight.
87 : Maternal complications following amniocentesis and chorionic villus sampling for prenatal karyotyping.
90 : Amniotic fluid insulin testing in gestational diabetes: safety and acceptance of amniocentesis.
99 : Procedure-to-delivery interval after late amniocentesis and the need for routine antenatal corticosteroids.
100 : Amniocentesis and chorionic villus sampling in twin gestations: which is the best sampling technique?
105 : Single-needle insertion: an alternative technique for early second-trimester genetic twin amniocentesis.
106 : Genetic amniocentesis in biamniotic twin pregnancies by a single transabdominal insertion of the needle.
109 : Pregnancy loss after chorionic villus sampling and genetic amniocentesis in twin pregnancies: a systematic review.
110 : Risk of fetal loss following amniocentesis or chorionic villus sampling in twin pregnancy: systematic review and meta-analysis.
111 : Second-trimester amniocentesis vs. chorionic villus sampling for prenatal diagnosis in multiple gestations.
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟