Induced fetal demise

INTRODUCTION — Induced fetal demise (also called iatrogenic fetal demise, feticide, or fetocide) refers to the injection of a pharmacologic agent into a fetus or the amniotic fluid to cause fetal asystole. Various injection techniques (intracardiac, intraamniotic, intraumbilical, intrafetal), routes (transvaginal, transabdominal), and agents (potassium chloride [KCl], digoxin, lidocaine) are used. This procedure is sometimes an adjunct to induced abortion after the first trimester. It is also done for selective or nonselective reduction of one or more fetuses in a multiple gestation. (See "Multifetal pregnancy reduction and selective termination".)

The indications and procedure for induced fetal demise are discussed here. Second-trimester abortion and induced fetal demise in multiple gestations are discussed in detail separately. (See "Overview of second-trimester pregnancy termination" and "Second-trimester pregnancy termination: Induction (medication) termination" and "Second-trimester pregnancy termination: Dilation and evacuation" and "Multifetal pregnancy reduction and selective termination".)

RATIONALE AND BACKGROUND — The procedure requires a willing patient who provides informed consent to the procedure, a skilled practitioner, appropriate ultrasound equipment, and stocking of the feticidal medication of choice. Intracardiac injection requires more expertise than intraamniotic or intrafetal procedures. Injections may be more difficult in patients with obesity and those with fibroids.

Regulatory issues — In the United States, following the 2004 passage of federal legislation regarding second-term abortion techniques [1], feticidal injection was adopted as a practice by some second-trimester surgical abortion providers as a way to document adherence to the law.

Standard methods of induced abortion via labor induction do not ensure fetal demise. In some institutions, any birth over 20 weeks with signs of life mandates neonatal intervention and resuscitation. Induced fetal demise before labor induction abortion ensures stillbirth and thus avoids potential emotional distress or the possibility of resuscitation or litigation related to survival of a neonate delivered after labor induction abortion [2].

In 2007, the Supreme Court upheld the Partial-Birth Abortion Ban Act of 2003 [1,3], which defines "partial-birth abortion" as occurring when "the person performing the abortion deliberately and intentionally vaginally delivers a living fetus for the purpose of performing an overt act that the person knows will kill the partially delivered living fetus." The term "partial-birth abortion" is not a medically recognized procedure, and the law is medically vague, but the procedure that is most likely affected by this legislation is an intact dilation and evacuation (D&E). Like dividing the umbilical cord after amniotomy and waiting for demise before extraction, feticide before D&E ensures compliance with this law should the preparation for the D&E itself be interpreted as intention to provide the banned procedure (eg, more than one day of osmotic dilators).

Many clinicians providing second-trimester abortion care prefer or require induction of fetal demise prior to D&E or labor induction abortion. While some ethicists advise that this decision is the prerogative of the clinician [4], some clinicians may opt not to use feticide based on patient preference (autonomy) and shared decision-making.

The science addressing fetal pain — Some policies restricting access to abortion care at certain gestations have been based on the concept of fetal pain. Pain perception requires peripheral sensory neural tissue that senses painful stimuli, functional thalamocortical pathways that transmit pain signals to the brain, and development of higher cortical functions that can interpret the pain signals. The gestational age at which neural maturation is sufficiently developed to allow the fetus to perceive pain has not been established [5-7]. However, a systematic multidisciplinary review of available evidence concluded that fetal pain perception is unlikely before 29 weeks gestational age [6]. In the rare situations where abortion after 28 weeks is necessary and legal, induction of fetal demise prior to the procedure avoids the possibility that the fetus might experience pain during abortion care.

INDICATIONS AND OUTCOMES

Labor induction (medication) abortion — In our practice, we suggest feticidal injection for patients undergoing induction abortion at ≥20 weeks of gestation and generally require it at ≥22 weeks. We engage in shared decision-making with patients, and they may choose whether feticidal injection will be included in the procedure. However, some institutions may require feticide prior to induction abortion due to considerations outlined below.

The goal of feticide before induction abortion is to ensure fetal demise, which is not ensured by standard methods of labor induction abortion. In one series, transient signs of life after induction abortion at 20 to 24 weeks gestation were reported in 50 percent of cases, with lower gestational age and fetal anomalies being associated with higher chances of stillbirth [8]. Transient signs of life in this setting may be psychologically and emotionally difficult for the patient, their support person(s), and staff [9]. Feticide removes this possibility and avoids futile resuscitation efforts [10,11].

Feticide may also shorten induction to abortion times, but data are limited. In a retrospective case series including 68 patients between 18 and 24 weeks gestation undergoing medication abortion with prostaglandins, those receiving feticidal potassium chloride (KCl) prior to induction compared with no feticidal agent required fewer prostaglandin doses and had a shorter interval to expulsion (570 versus 890 minutes) [12]. Similarly, in a subsequent retrospective study of 118 patients between 17 and 28 weeks gestation undergoing medication abortion, those managed with versus without feticide also had shorter induction to abortion times (900 versus 1198 minutes) [13].

Feticide may also decrease the risk of transfusion, but the quality of evidence is low. In one case-control study including 15 patients with placenta previa undergoing second-trimester induction abortion, preinduction feticide decreased the risk of transfusion [14]. This finding needs to be confirmed in larger trials.

Patient experience and views of feticide vary. In a United Kingdom study, patients and their partners had varied views of the option of feticide before labor induction abortion. For those who chose feticide before induction, it was seen as one part of a stressful process. For those who declined feticide, they knew it was not right for them but viewed it as appropriate for others [9].

In the same study, clinicians performing feticidal injection before labor induction abortion described the procedure as necessary and as a way to spare patients from the emotional pain of transient signs of life at delivery [9]. In another United Kingdom qualitative study, the physicians took pride in performing a technically difficult procedure, felt most comfortable with providing feticide when they were part of the counseling about the need for abortion care (framed as "clinician over technician"), and found a supportive clinical team helpful in dealing with what could be a stressful part of their jobs [15].

Dilation and evacuation — Use of feticidal injection prior to dilation and evacuation (D&E) abortion was introduced with the aim of improving procedure safety and shortening its duration by making the tissue softer, but data do not support these outcomes [16]. Society of Family Planning guidelines do not recommend routine use [17]. When not required by institutional policy, we do not perform feticidal injection prior to D&E. For those clinicians who do perform feticidal injection before D&E, most do it after 20 weeks [18].

Data are inconsistent regarding whether feticidal injection prior to D&E reduces procedure duration; outcome measures vary across studies, and duration may be evacuation time or duration of all or part of the entire abortion process. Representative studies of feticide prior to D&E include:

●A randomized trial including 126 patients evaluated second-trimester D&E with preprocedure intraamniotic injection of digoxin compared with placebo [16]. The trial found no differences in procedure duration, estimated blood loss, pain scores, complications, or surgeon-reported case difficulty. But digoxin was associated with higher rates of emesis (16.1 versus 3.1 percent).

●A randomized trial including 178 patients found that, though complications were rare, inducing fetal demise with digoxin does not shorten procedure duration overall but may do so for D&Es at 22 to 24 weeks [19].

●A retrospective study including 548 patients from the British Pregnancy Advisory Service compared second-trimester abortion procedures with and without preprocedure feticide with intracardiac KCl [20]. While the average duration of evacuation was 3.5 minutes shorter with feticide, the injection itself took approximately 8 minutes, and the subjects who received feticide had more preprocedure pain and had higher rates of uterine atony.

Despite the lack of evidence of clinical benefit, many clinicians administer feticidal injection before some second-trimester D&E procedures. In one survey, approximately one-half of North American Family Planning subspecialists reported use of feticide before second-trimester D&E [18]. This is due to a variety of reasons, including regulatory issues and concerns such as preoperative extramural delivery [21].

Patient preferences vary regarding feticidal injection as part of the abortion process:

●In the randomized trial cited above, a majority of patients in both the digoxin and placebo groups (92 percent in each group) reported a preference for fetal death before the procedure [16]; this may be affected by selection bias of patients willing to participate in this study.

●In a study of 20 subjects undergoing feticidal digoxin injection prior to D&E, participants had both positive and negative responses [22]. Participants discussed negative experiences with the injection procedure itself (pain, fear of needles) and with carrying a dead fetus until the D&E but also discussed positive experiences with feeling relief at starting the abortion process and at preferring that the fetus be dead before the D&E.

●In a prospective study of transcervical digoxin injection before D&E in a practice that offered the alternative of cord transection during D&E, only 19 percent of eligible patients agreed to enroll. Of patients who declined study enrollment, 37 percent did not want preoperative induced demise, 36 percent did not want an unnecessary medication, and 14 percent were afraid of pain with the injection [23].

CONTRAINDICATIONS — Allergy to the feticidal medication is a contraindication to induced fetal demise. There are no other absolute contraindications.

INFORMED CONSENT — Patients have diverse views about the option of feticide before second-trimester labor induction abortion or dilation and evacuation (D&E). Patient-centered shared decision-making allows clinicians to present the option of feticidal injection at the start of abortion care, elicit the patient's values, and develop the treatment plan that is best for the patient. In our practice, prior to labor induction abortion after 20 weeks gestation, we offer feticide and, in most cases, present it as the preferred option. The informed consent discussion should include indications, contraindications, risk of complications, and alternatives to this procedure. (See "Counseling in abortion care".)

PROCEDURE

Timing — The timing of feticidal injection is based on the type of abortion (labor induction or surgical) and the agent to be injected. Potassium chloride (KCl) is immediately effective whereas digoxin is not and is generally injected one day before the abortion:

●Labor induction abortion – Injection may be performed at any time before labor induction abortion but is often elected one day before induction at the time of office counseling and mifepristone administration.

●Dilation and evacuation (D&E) – The feticidal injection is usually given one day prior to the D&E at the patient's preoperative visit for intracervical osmotic dilator insertion.

Equipment

●Sterile gloves

●Sterile drapes

●Sterilizing solution, such as povidone-iodine, chlorhexidine, or alcohol, to prepare the surgical site

●Local anesthetic (eg, 1% lidocaine solution)

●25- or 30-gauge needle attached to a 3 mL syringe to inject the local anesthetic; local anesthesia is used if the treatment needle is 18-gauge or greater

●18-, 20-, or 22-gauge spinal needle attached to a 15 mL syringe

●Digoxin 1 mg or KCl 5 to 15 mL of 2 mEq/mL solution

●Sterile saline solution to confirm needle location

Choice of agent — Digoxin, KCl, and lidocaine are the most common drugs used to induce fetal asystole.

In our practice, we prefer intracardiac injection of KCl over other agents; KCl is immediately effective and has a high success rate. However, KCl is not available in all settings (given the risk of fatal cardiac arrest when administered accidently) and some clinicians prefer injection of digoxin or lidocaine (given its wider availability and safety profile).

Potassium chloride — KCl can be injected into the fetal heart or umbilical vein to induce fetal death. It is immediately effective at inducing fetal asystole, and the procedure is not concluded until fetal asystole is confirmed.

Potassium chloride is the most frequently used method in the United Kingdom, and was the preferred feticidal agent in the 2010 Royal College of Obstetricians and Gynaecologistsꞌ Termination of Pregnancy for Fetal Abnormality guidelines [24,25].

In the largest series describing the use of KCl for feticide, 239 left ventricular injections at 17 to 24 weeks of gestation were 100 percent successful [26]. No procedure took longer than five minutes.

Three small series involving KCl injection into the umbilical vein reported success rates ranging from 87 to 100 percent [27-29]. In these series, no electrocardiographic modifications were observed, and plasma potassium levels did not show any significant variation throughout the procedure [27,28]. Our experience with KCl has been similarly successful and safe [30]. However, cardiac arrest in the pregnant person after attempted fetal intracardiac injection of KCl has been reported [31]. (See 'Side effects and complications' below.)

Digoxin — Digoxin can be injected into the amniotic fluid, the fetal heart or body, or the umbilical vein to induce fetal demise [32]; it is not immediately effective and can fail to induce fetal asystole in some cases. Intraamniotic injection is technically easier than intrafetal injection but may be less effective.

The injection is generally done transabdominally but can be done with high success rates transvaginally into the amniotic cavity or the fetal body at the time of osmotic dilator placement [23,33]. This has the benefit of combining the two procedures and saving the patient a transabdominal injection, which is sometimes painful.

Data vary regarding the efficacy of different routes of digoxin administration. One randomized trial showed that fetal demise occurred within 24 hours at higher rates with transabdominal intrafetal digoxin compared with intraamniotic (95 versus 82 percent) [34-36]. However, a randomized pilot study showed that rates of induced fetal demise were the same (87 percent) for both routes and the same for two doses (1 mg and 1.5 mg) of digoxin [37].

A dose-finding case series found that, when injected intrafetally, a dose of 1 mg had 100 percent success rate as defined by fetal asystole the following day, and a dose of 0.5 mg had 96.4 percent success [38]. In another large case series, intrafetal injection had a 99.7 percent success rate with most doses being 1 mg [33]. A subsequent study using a higher dose (2 mg) of digoxin intraamniotic injection at 21 to 30 weeks showed a success rate of 93 percent, with serum levels of digoxin in the therapeutic range at 6 and 10 hours after injection [39].

Lidocaine — Lidocaine may be used as a primary agent to induce fetal demise or as a secondary agent in the setting of digoxin failure; it can be injected into the fetal heart, thorax, amniotic fluid, and umbilical vein [40-44].

Lidocaine appears to be an effective method to induce fetal demise. Representative studies include the following:

●In a randomized trial of 26 patients undergoing termination of pregnancy between 24 and 38 weeks gestation, intracardiac lidocaine (10 mL in 2% preparation) compared with intracardiac KCL (6 mmol) was more effective at inducing fetal asystole at three minutes after injection (85.7 versus 57.9 percent, respectively) [45]. If fetal demise did not occur at three minutes, normal saline (10 to 20 mL) was injected into the fetal pericardium, leading to rapid fetal asystole in all cases.

●In a retrospective study including 338 pregnancies (median gestational age 27 6/7 weeks), a single dose of lidocaine (intracardiac or intrathoracic) successfully induced fetal demise in almost all (98 percent) cases; a second injection was required in five fetuses [40]. Asystole typically occurred within one to two minutes.

●In a retrospective study including 16 pregnancies (mean gestational age 22 weeks), a single dose of intracardiac lidocaine successfully induced fetal demise is all cases [44].

Injection sites and techniques — All feticide procedures are performed under continuous ultrasound visualization.

Intracardiac potassium chloride injection — To minimize discomfort from larger-gauge needles, as a first step, we use a 25-gauge needle to inject 2 to 5 mL of 1% lidocaine solution into the patient's abdominal wall. We use an 18- or 20-gauge needle for the intracardiac injection. Use of a 22-gauge needle minimizes patient discomfort but is more difficult to visualize and may bend in the fetus or a patient with obesity and thus be more difficult to maneuver.

The spinal needle is directed into the fetal heart (image 1). When the tip appears to be in the ventricle, we attempt to aspirate blood, which suggests but does not guarantee intracardiac placement. If the tip of the needle is not well visualized, we also inject sterile saline until turbulence is seen, as turbulence at the needle tip confirms an intraventricular location. It is important to confirm correct placement, as inadvertent injection into the pregnant patient rather than the fetal compartment can produce serious adverse sequelae. (See 'Side effects and complications' below.)

After correct placement is confirmed, we inject 5 to 15 mL of KCl solution (2 mEq/mL), keeping the needle in place until asystole occurs, which typically takes one to five minutes (image 2). KCl may be injected in increments of 5 mL over a number of minutes. Asystole may occur abruptly but usually takes two to three cardiac cycles. Digoxin (1 mg) is an acceptable alternative.

Intraamniotic digoxin injection — This is the easiest procedure technically, as noted above, but has a higher failure rate than intracardiac injection and may have a higher failure rate than intrafetal injection.

Intraamniotic digoxin injection is performed one to two days before the abortion because, unlike KCl injection, asystole is not immediate [37,38,46].

The procedure for transabdominal intraamniotic injection is similar to that for amniocentesis. Correct needle placement is confirmed by visualization of the needle in the amniotic cavity and aspiration of amniotic fluid, and for feticide, 1 mg of digoxin is then injected into the amniotic fluid. Sonographic visualization of the fluid jet further confirms intraamniotic injection. A 20- or 22-gauge needle is used, and analgesia is usually not necessary.

It may also be performed transvaginally at the time of osmotic dilator insertion. After speculum placement, vaginal preparation with an antiseptic solution, and paracervical block placement, additional local anesthetic can be injected at the cervicovaginal junction anteriorly in the midline or through the anterior cervical stroma. Then, an amniocentesis needle or an 18-gauge standard needle on a needle extender is placed at this site and advanced parallel to the axis of the cervix into the amniotic cavity under ultrasound guidance [23,33].

Intraumbilical vein injection — Feticide can also be accomplished by injection of KCl or digoxin into the umbilical vein. Compared with intracardiac injection, umbilical vein injection is technically more difficult and appears to be less successful in inducing asystole [27-29].

Intrafetal injection of digoxin — Intrafetal injection (ie, into any part of the fetus) requires more skill than intraamniotic but generally less than an intracardiac approach. It has been studied before D&E procedures and can be done either transabdominally or transvaginally at the time of osmotic cervical dilator placement.

Lidocaine injection — Lidocaine injection into the fetal heart, thorax, amniotic fluid, and umbilical vein have been described [40-43].

In one study, intracardiac injection of lidocaine was described as follows [40]: A 15 cm 20-gauge needle is attached to a 20 mL syringe containing 1% lidocaine and introduced through the skin and subcutaneous tissue. Approximately 0.5 to 1 mL of lidocaine is injected into the myometrium and the needle is then passed into the amniotic cavity and into the fetus, where 2 mL of lidocaine are injected directly into the fetus. After intracardiac position is confirmed by aspirating fetal blood from the fetal cardiac chamber, the remaining lidocaine (approximately 17 mL) is injected over 20 to 60 seconds. The physician then aspirates and reinjects 10 mL of fetal blood to further disrupt fetal blood flow. Asystole is confirmed, typically within one to two minutes and reconfirmed two to five minutes later. If asystole is not achieved, a second dose can be administered using the same technique after 20 to 40 minutes.

Variations may include utilizing 2% lidocaine and/or intrathoracic injection if intracardiac placement cannot be achieved [40].

SIDE EFFECTS AND COMPLICATIONS — Side effects and complications are rare. The range of findings is described below:

Potassium chloride

●In the largest series of 239 intracardiac potassium chloride (KCl) injections, no complications were reported [26].

●Three series of patients who underwent intraumbilical vein administration of KCl reported no changes in postprocedural electrocardiograms, potassium concentrations, or other adverse outcomes [27-29].

●One case of cardiac arrest during an attempt at intracardiac KCl injection has been reported [31]. Toxicity may have resulted from absorption of KCl when the needle passed through the fetus and into the uterus and pelvis of the pregnant patient.

Digoxin

●In one study, eight patients between 19 and 23 weeks of gestation received 1 mg digoxin via intraamniotic injection and then had serial serum digoxin levels determined for 48 hours, Holter cardiac monitoring performed for 24 hours, and laboratory assessments (digoxin level, prothrombin time, partial thromboplastin time, fibrinogen level) [34]. No rhythm or conduction abnormalities occurred, the peak digoxin concentration was in the low therapeutic range, and coagulation studies remained in the normal range.

●A case series of eight subjects showed that in the 24 hours after intraamniotic injection of 1 mg of digoxin, serum digoxin concentrations in the pregnant patient peaked in the low therapeutic range and were not associated with abnormalities on Holter monitor or clinically significant changes in coagulation parameters [34].

●In two retrospective studies, low rates of extramural delivery (0.3 to 1.9 percent) and infection (0.03 to 3.4 percent) were reported when intraamniotic or intrafetal digoxin injection was performed one to two days before the dilation and evacuation (D&E) procedure [32,36].

●In a randomized trial of intraamniotic digoxin compared with placebo, a higher prevalence of vomiting was noted in patients who received intraamniotic digoxin than in those receiving saline (16 versus 3 percent) [16].

●No adverse events suggestive of digoxin toxicity were noted in another study that included over 1700 patients between 17 and 24 weeks gestation who received intraamniotic or intrafetal digoxin [38].

Lidocaine

●In the retrospective study including 338 pregnancies described above (see 'Lidocaine' above), one patient experienced lidocaine-related side effects (ie, mild tinnitus) [40]. There were no major adverse events.

●In another prospective study including 50 patients undergoing pregnancy termination with lidocaine plus sufentanil, no side effects were reported [42].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Pregnancy termination".)

SUMMARY AND RECOMMENDATIONS

●Clinical significance – Induced fetal demise refers to the injection of a pharmacologic agent into a fetus to cause its demise before abortion. This may be performed using several injection techniques (intracardiac, intraamniotic, intraumbilical, intrafetal) and agents (potassium chloride [KCl], digoxin, lidocaine) (image 1). (See 'Introduction' above.)

●Indications

•For patients undergoing labor induction abortion at ≥20 weeks of gestation, we suggest use of feticidal injection (Grade 2C). The goal is to ensure fetal demise; feticide may also shorten the procedure duration. We engage in shared decision-making with patients, and they may choose whether feticidal injection will be included in the procedure. (See 'Labor induction (medication) abortion' above.)

•Feticidal injection before second-trimester dilation and evacuation (D&E) has not been shown to improve patient safety or shorten procedure duration. In our practice, we do not routinely offer feticide prior to D&E. Among those clinicians who perform feticidal injection before D&E, most do it after 20 weeks. (See 'Dilation and evacuation' above.)

●Informed consent – Patients have diverse views about the option of feticide before second-trimester labor induction abortion or D&E. Patient-centered shared decision-making allows clinicians to present the option, elicit the patient's values, and develop the treatment plan that is best for the patient. (See 'Informed consent' above.)

●Choice of agent (see 'Choice of agent' above)

•KCl, digoxin, and lidocaine are the most commonly used drugs for inducing fetal demise.

•We suggest intracardiac KCl rather than other agents (Grade 2C). While KCl is immediately effective and has a high success rate, KCl is not available in all settings (given the risk of fatal cardiac arrest when administered accidently) and some clinicians prefer injection of digoxin or lidocaine.

●Injection technique – All feticide procedures are performed under continuous ultrasound visualization. (See 'Injection sites and techniques' above.)

●Complications – Serious side effects and complications are rare as long as appropriate dosing is used and inadvertent injection into the pregnant person is avoided. (See 'Side effects and complications' above.)