Maternal anesthesia for EXIT procedure: A systematic review of literature

Abstract

The ex utero intrapartum treatment (EXIT) procedure is performed in cases of fetal congenital malformation. The anesthetic management is much more challenging and involves providing profound uterine relaxation, maintenance of Uteroplacental blood flow and fetal anesthesia. The aim of the article is to review the literature and compare the efficacy of both the anesthetic techniques with respect to maternal and fetal outcomes. The literature source for this review was obtained via PubMed, Medline, Google scholar and Cochrane database of systematic reviews until January 2017. In our literature review we found that both GA and Regional anesthesia were successfully described for EXIT procedure but GA was performed in the majority of cases. Consideration for anesthetic technique should be done on a case-by-case basis.

Introduction

The ex-utero intrapartum treatment (EXIT) is a rare surgical procedure to manage fetal airway anomalies. It was first described in 1989 to manage a fetus with complete tracheal obstruction secondary to a prenatal diagnosed cervical teratoma.[1]

Since then, the indications for EXIT procedures have expanded to include a variety of fetal congenital abnormalities such as congenital high airway obstruction syndrome (CHAOS), laryngeal or tracheal atresia/stenosis, intrathoracic lesions (lung and mediastinal masses, pulmonary arteriovenous malformation with pulmonary hypoplasia), as well as congenital diaphragmatic hernias.[2,3,4]

It is a specialized surgical procedure that involves partial delivery of fetus during a cesarean section allowing the fetal airway to be secured while fetal oxygenation is maintained by placental perfusion. It has also been described as “operation on a placental support (OOPS).”[5,6]

Anesthetic management of the EXIT procedure is challenging and different from a standard cesarean section and involves providing profound uterine relaxation, maintenance of uteroplacental blood flow, and fetal anesthesia.[2,3,4,7,8,9]

Traditionally, general anesthesia (GA) has been the technique of choice for EXIT procedures due to the ease of titration of inhalational agents to achieve satisfactory uterine relaxation and to provide fetal anesthesia. In recent years, several reports have described neuraxial anesthesia use in combination with uterine relaxants to achieve satisfactory operating conditions for EXIT. The aim of this article is to review the literature and compare the efficacy of both anesthetic techniques with respect to maternal and fetal outcomes.

Material and Methods

We searched PubMed, Medline, Google scholar, and Cochrane Database of Systematic Reviews till January 2017 using the terms “Ex-utero intrapartum treatment” and “Anesthesia/anaesthesia.” The search was limited to the English language. The reference lists of retrieved articles were examined to identify further articles. Articles were reviewed by the authors independently and evaluated for eligibility. This review includes all reports describing anesthetic techniques as well as maternal and fetal outcomes.

Results

In the literature search [Figure 1], we found 224 articles, from which 167 were excluded due to duplication or because sufficient details were not available regarding anesthetic management. A total of 43 eligible articles of 224 patients were considered for analysis. Thirty-seven reports described the use of GA in 214 patients and six reports described regional anesthesia (RA) as the primary anesthetic technique in 10 patients. Primary anesthetic techniques are described in Table 1.

Figure 1.

Methods and Results

Table 1.

Maternal Outcome

Anesthesia technique	GA	Regional
Total number of patients	214	10
Mean placental bypass time	14.37 min (n=117)	7.6 min (n=10)
Mean estimated blood loss (ml)	1128.34 (n=174)	918 (n=8)
Estimated blood loss >1000ml	18 (n=214)	1 (n=10)
Blood transfusions	17 (n=214)	0 (n=10)
Use of multiple uterotonics	23 (n=214)	4 (n=10)
Invasive monitoring	121 (n=214)	1 (n=10)
Use of vasopressors (Ephedrine, phenylephrine, dopamine)	139 (n=214)	7 (n=10)
ICU transfer	1 (n=214)	0 (n=10)
Maternal deaths	0 (n=214)	0 (n=10)
Conversion of anesthetic techniques	0 (n=214)	0 (n=10)
Conversion of EXIT	5 (n=214)	0 (n=10)

All parturients were monitored with standard monitors. Seventy-four parturients received arterial lines for continuous blood pressure monitoring in the GA group and one in the RA group. Fetal monitoring was achieved with SpO₂ monitoring in most cases, whereas fetal ultrasonography was used by thirteen authors.

General anesthesia

All parturients receiving GA were given aspiration prophylaxis and placed in the left lateral decubitus prior to induction. Rapid sequence induction was performed for endotracheal intubation.

Thiopentone or Propofol was used as induction agents and in a majority of patients; succinylcholine was the muscle relaxant of choice except for the two Malignant hyperthermia susceptible patients.

GA was the main anesthetic technique in 37 of the 43 reports, involving a total of 214 patients. Epidural anesthesia was used for 72 patients for postoperative pain control[1,2,8,13,14] and one patient received intrathecal morphine. Thirty-five reports described inhalational anesthesia for maintenance in 212 patients with a target MAC of >2. Isoflurane, sevoflurane, and desflurane were used with good success. Desflurane was the most commonly used halogenated agent (141/212 patients) [Figure 2].

Figure 2.

Anesthesia technique distribution by percentage

Three authors described Total intravenous anesthesia (TIVA in 20 patients, 18 of which were combined TIVA and desflurane.[10,11,12] Remifentanil and propofol were the most common combination with only one patient receiving TIVA with propofol alone and one receiving a remifentanil and N₂O combination; of these patients, two were MH susceptible.[11,12] Duration of placental support under GA ranged from 3 to 93 minutes.

Regional anesthesia

Gagnon et al.[13] published the initial case report using RA as the primary anesthetic technique in 1998. They successfully used epidural anesthesia with IV nitroglycerin (NTG) to achieve adequate surgical conditions. Subsequently, five other reports of RA have been published[14,15,16,17,18] involving 9 patients [Figure 2]. Eight patients received a combined spinal-epidural (CSE) anesthetic with bupivacaine in doses ranging from 10 to 12 mg and intrathecal fentanyl and epimorphine in doses of 15–25 mcg and 150–300 mcg, respectively. Benonis et al.[18] described the successful use of a subarachnoid catheter following a failed CSE in one patient with a history of failed spinal for the previous cesarean section. This patient achieved a T4 sensory level and adequate surgical anesthesia with a total bupivacaine of 11.25 mg, epimorphine 150 mcg, and fentanyl 10 mcg. All patients received IV NTG for uterine relaxation. Duration of placental support under RA ranged from 1 to 21 min. None of the patients with RA required a conversion to GA.

Uterine relaxation

NTG IV was the uterine relaxant of choice in all RA and TIVA[10,11,12,14,15,16,17,18,19] cases; it was required as an adjunct to inhalational agents in 25 cases.[1,5,8,20,22,23]

We found that NTG was most commonly used as a loading dose, ranging from 25–100 mcg followed by an infusion at a rate of 1–20 mcg/kg/min to keep the uterus adequately relaxed. Pharmacokinetic studies have demonstrated that placental transfer of NTG has no significant fetal hemodynamic effects. This is most likely due to the short half-life and rapid placental metabolism of the drug.[17,18]

In a majority of the patients, maternal hypotension was treated with intermittent boluses of phenylephrine and ephedrine as well as with judicious fluid management. Four patients in the RA group[15,16] required phenylephrine infusions of 50 mcg/min. Three patients in the GA group[20,21,22,23] required dopamine infusions to maintain adequate mean arterial pressures while receiving nitroglycerine (NG). Shih et al.[23] used an angiotensin II infusion ranging from 10–55 ng/kg/min to maintain a systolic BP >100 mmHg without concurrent use of NG.

Uterine tone

Oxytocin was the primary uterotonic in all cases, ranging from boluses of 5 U IV to infusions of 10 U–50/L [Figure 3].

Figure 3.

Distribution of uterotonic in general anesthesia and regional anesthesia groups

Carboprost 200–250 mcg IM was used in 9 patients in the GA group[1,20,21,24,25,26] (4.2%) and 3 patients (30%) in the RA group.[17]

Methylergonovine 200–250 mcg IM was used in 14 patients (6.5%) in the GA group[1,26,27,28] and 3 patients (30%) in the RA group.[15,16,17,18,19]

Two patients in the RA group[15] and one patient in the GA group[26] required oxytocin, carboprost, and methylergonovine concurrently.

None of the patients receiving more than one uterotonic received any blood transfusions.

Maternal outcome

Eight maternal postpartum hemorrhages (blood loss >2000ml) were reported in the GA group.[1,5,13,29,30,31,32,33] Of these, 5 required blood transfusion and 1 patient required ICU admission overnight.[33] Five EXIT procedures were converted to cesarean section secondary to significant bleeding and inadequate placental perfusion.[1,33] There were no postpartum hemorrhages in the RA group. There were no maternal deaths in either the RA or GA group [Table 1].

Fetal outcomes

The overall fetal/neonatal mortality rate was 10.31% (23/223). Five out of the 23 fetuses died intraoperatively[1,29,30,34] due to various causes which included failure to oxygenate and intubate, parents refusal for tracheostomy, and intraoperative bleeding. Eleven neonates who required complex surgery died due to pulmonary hypoplasia in the neonatal intensive care unit.[1,5,13,29,35,36] Five had complex congenital heart disease and died due to sever PHT and congestive heart failure.[1,5,29,30] One died because of postoperative bleeding and of unknown causes.[1,34]

There were no maternal or fetal complications related to anesthesia in either GA or RA groups [Table 2].

Table 2.

Fetal outcomes

Anesthesia techniques	GA	Regional
Total number of patients	214	10
APGAR score at 1 and 5 min	5.9 and 8.3 (n=8)	3.9 and 7.2 (n=9)
Mean umbilical cord pH (arterial)	7.21 (n=54)	7.23 (n=3)
Neonatal death	21 (n=214)	2
Need for supplemental anesthesia	149 (n=214)	1

Fetal monitoring and anesthesia

Fetuses were monitored using different techniques at various times during the surgical procedure. Sterile SpO₂ probes were the most common fetal monitor described fetal sonography, and ECG were used additionally by some authors. Sixteen of the 42 authors used additional medications for infants, including atropine, analgesics, and NDMB. Fentanyl was the most commonly used narcotic.[1,5,7,8,16,25,26,29,31,34,37,38,39,40,41,42,43,44]

Discussion

Our literature review has shown that successful anesthetic management of the EXIT is possible with both GA and RA. However, there is a significant preponderance towards GA. The preference for GA for EXIT procedures can be explained by a number of reasons. A major goal of EXIT is profound uterine relaxation. Inhalational anesthetics provide reliable and titratable uterine relaxation at MAC >2 with relatively quick reversibility. Neuraxial techniques require additional IV NTG use for uterine relaxation that can lead to problematic hypotension and uterine hypoperfusion. We found increased use of vasopressors in the RA group as evidenced by the use of phenylephrine infusions in 50% of the patients receiving RA. We found few descriptions of vasopressor infusions in the GA group.

Fetal anesthesia is an important component of the anesthetic goals for EXIT and can usually be achieved by transplacental transfer of inhalational anesthetics. Remifentanil has been used successfully to achieve fetal anesthesia and should be considered in parturients receiving RA. A retrospective review of medical records by Boat et al.[10] comparing maintenance with desflurane alone to desflurane and TIVA with propofol and remifentanil in women having both open fetal procedures or EXIT showed a reduction in acute fetal cardiac dysfunction in the desflurane plus TIVA group as well as an overall decrease in fetal interventions. There are no other studies comparing TIVA and inhalational agents.

Remifentanil rapidly crosses the placenta providing excellent fetal immobilization and is rapidly metabolized by fetal nonspecific blood and tissue esterase. It also acts as a good anxiolytic for an awake parturient, and is easily titratable to avoid excessive sedation in the presence of a high risk of aspiration.

Although most procedures described lasted between 5 and 25 min with relatively uncomplicated airway manipulations, either direct laryngoscopy or tracheostomy, there were some procedures that lasted over 50 min. Calculation of the mean duration of EXIT was difficult because some case reports lacked an explicit time of placental bypass. Due to this potentially unpredictable length of time required for placental support coupled with the emotional stress of witnessing airway manipulations on the baby, it may be more difficult for an awake parturient to tolerate the procedure.

Postpartum atony and hemorrhage (PPH) is an important intraoperative complication. There were eight PPH cases reported in our review, and of these five patients who received blood transfusions all were in the GA group. However, given the large difference in the number of GA vs RA patients, it is hard to draw any definitive conclusion regarding the effect of anesthetic technique on the incidence of PPH in EXIT. It is also interesting to note that the percentage of patients receiving more than one uterotonic was much higher in the RA group, which may have a direct impact on PPH incidence.

Unfortunately, APGAR scores were not available for infants, especially as some received NDMB and narcotics IM. Hence, it was impossible to use this method to assess fetal outcomes. Similarly, cord pH was not available for all patients and could not be used reliably.

Other potential shortfalls of our review include the disparity between the level of detail provided by each author regarding drugs used, including inhalational agents, vasopressors, uterotonics, and blood loss, making it more difficult to draw definitive conclusions.

Conclusion

In summary, we found in our literature review that both GA and RA were successfully described for EXIT procedures, but GA was performed in a majority of cases. The goals of uterine relaxation, maintenance of placental perfusion, and fetal anesthesia must always be considered when choosing the anesthetic technique. Consideration for anesthetic technique should be done on a case-by-case basis.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.