The EXIT Procedure: Experience and Outcome in 31 Cases
By Sarah Bouchard, Mark P. Johnson, Alan W. Flake, Lori J. Howell, Laura B. Myers, N. Scott Adzick,
and Timothy M. Crombleholme
Philadelphia, Pennsylvania
Background: The EXIT (ex utero intrapartum treatment) pro-
cedure, although initially designed for reversal of tracheal
occlusion in fetuses with congenital diaphragmatic hernias
(CDH), has been adapted to treat a variety of fetal conditions.
Methods: A retrospective chart review of all consecutive
EXIT procedures since 1996 was conducted.
Results: Thirty-one women underwent the EXIT procedure,
with an average maternal age of 29 years (range, 20 to 38),
and average gestational age of 34 weeks (range, 29 to 40).
The indication was airway obstruction from fetal neck mass
in 13, and reversal of tracheal occlusion from in utero clip-
ping in 13. Singular indications included an EXIT-to-ECMO
(extracorporeal membrane oxygenation) procedure for a fe-
tus with CDH and a cardiac defect (n ⫽ 1), congenital high
airway obstruction syndrome (CHAOS, n ⫽ 1), resection of a
very large congenital cystic adenomatoid malformation of
the lung (CCAM) on uteroplacental bypass (n ⫽ 1), unilateral
pulmonary agenesis (n ⫽ 1), and thoracoomphalopagus con-
joined twins. The mean duration on uteroplacental bypass
(from uterine incision to umbilical cord clamping) was 30.3 ⫾
14.7 minutes (range, 8 to 66). No fetus experienced hemody-
namic instability during uteroplacental bypass as recorded
by fetal heart rate (FHR), pulse oximeter, and fetal echocar-
diography, except for one instance of reversible bradycardia
T HE EXIT (ex utero intrapartum treatment) proce-
dure was designed initially for reversal of tracheal
occlusion performed in fetuses with severe congenital
diaphragmatic hernias (CDH).1 It ensured an optimal
controlled environment on uteroplacental bypass while
the tracheal occlusion was reversed at the time of deliv-
ery. The proven utility of the EXIT procedure has been
applied to the management of fetal neck masses. We and
others reported the successful application of the EXIT
From the Divisions of Pediatric General, Thoracic, and Fetal
Surgery and Anesthesia and Critical Care, The Center for Fetal
Diagnosis and Treatment, The Children’s Hospital of Philadelphia and
The University of Pennsylvania School of Medicine, Philadelphia, PA.
Presented at the 32nd Annual Meeting of the American Pediatric
Surgical Association, Naples, Florida, May 20-23, 2001.
Address reprint requests to Timothy M. Crombleholme, MD, Divi-
sion of Pediatric General, Thoracic, and Fetal Surgery, The Center for
Fetal Diagnosis and Treatment, The Children’s Hospital of Philadel-
phia, 34th St and Civic Center Blvd, Philadelphia, PA 19104.
Copyright © 2002 by W.B. Saunders Company
0022-3468/02/3703-0021$35.00/0
doi:10.1053/jpsu.2002.30839
418
from umbilical cord compression. The mean FHR and fetal
saturation were 153.0 ⫾ 38.5 beats per minute and 71.2% ⫾
19.9%, respectively. Five fetuses required a tracheostomy.
Only 1 death occurred during an EXIT procedure because of
inability to secure the airway secondary to extensive involve-
ment by a lymphangioma. The average cord pH and pCO2
were, respectively, 7.20 ⫾ 0.11 and 63.2 ⫾ 14.6. Two mater-
nal complications occurred: bleeding from a hysterotomy
site and dehiscence of an old hysterotomy scar noticed at a
subsequent cesarean section. The average maternal blood
loss was 848.3 ⫾ 574.1 mL.
Conclusion: The EXIT procedure was used successfully to
ensure uteroplacental gas exchange and fetal hemodynamic
stability during a variety of surgical procedures performed to
secure the fetal airway or ensure successful transition to
postnatal environment.
J Pediatr Surg 37:418-426. Copyright © 2002 by W.B.
Saunders Company.
INDEX WORDS: Ex utero intrapartum treatment, extracorpo-
real membrane oxygenation, congenital diaphragmatic her-
nia, congenital cystic adenomatoid malformation, teratoma,
congenital high airway obstruction syndrome, pulmonary
agenesis, intrinsic high airway obstruction, conjoined twins,
neck masses.
procedure in the management of giant fetal neck masses.2
Because of the prolonged stable fetal hemodynamic
environment it provides, this procedure has been adapted
further and used to treat a variety of fetal conditions at
delivery. As our experience with EXIT procedures has
grown, the anesthetic and surgical techniques have been
refined, indications have been broadened, and complica-
tions reduced.
In this study, we present our consecutive experience
with all the EXIT procedures performed at the Children’s
Hospital of Philadelphia since 1996 to highlight lessons
learned in the experience with 31 cases.
MATERIALS AND METHODS
Patient Selection and Data Collection
We reviewed our consecutive experience with the use of the EXIT
procedure in the management of 31 fetuses referred to The Center for
Fetal Diagnosis and Treatment at The Children’s Hospital of Philadel-
phia from March 1996 to March 2001. This study was approved by the
Committee for Protection of Human Subjects Institutional Review
Board at the Children’s Hospital of Philadelphia.
Data collected retrospectively from charts included maternal and
Journal of Pediatric Surgery, Vol 37, No 3 (March), 2002: pp 418-426
EXIT PROCEDURE
fetal demographics (maternal age, obstetrical history of current preg-
nancy, gestational age at diagnosis and at EXIT procedure), indications,
results of prenatal imaging studies including fetal ultrasound or mag-
netic resonance imaging (MRI), and echocardiography, chromosomal
studies, maternal triple screen, hemodynamics (fetal heart rate and
saturation, cord blood gas), duration of uteroplacental bypass, tech-
niques used to secure airway, and maternal or fetal complications.
Some of these cases have been the subject of previously published case
reports.2-5
Thirty-one women underwent the EXIT procedure at an average
maternal age of 29 years (range, 20 to 38). These women underwent
close follow-up during gestation with fetal ultrasonography, fetal ul-
trafast magnetic resonance imaging with HASTE (half-Fournier single-
shot turbo spin echo) sequences, and fetal echocardiography. A fetal
karyotype was obtained in 16 women to rule out major genetic
abnormalities. In some, a maternal triple screen analysis has been
performed, whereas in others the nature of the congenital anomaly was
not associated with karyotype abnormality, and there was no indication
for genetic amniocentesis based on maternal age.
Description of the EXIT Procedure
A multidisciplinary team is used for each procedure, involving 2 or
3 pediatric surgeons, an obstetrician, a neonatologist, an anesthesiolo-
gist, and 2 scrub nurses. Maternal monitoring consists of invasive
arterial blood pressure measurement, continuous electrocardiography,
pulse oximetry, and end-tidal CO2 monitor. The tocolytic drug indo-
methacin (50 mg intrarectally) usually is administered the morning of
surgery. On arrival, if the mother did not receive the morning dose, it
is administered after induction of general anesthesia. The mother
receives 30 mL of 0.3 mol/L sodium bicitrate orally to reduce gastric
acidity, and 10 mg of metoclopromide intravenously to enhance gastric
emptying. The mother is positioned with left uterine displacement.
Anesthesia is induced by a rapid sequence technique with thiopental (5
mg/kg), succinylcholine (2 mg/kg), and fentanyl (1 to 2 ␮g/kg) given
intravenously, followed by endotracheal intubation. Anesthesia is
maintained with 0.5 MAC (minimum alveolar concentration) of des-
flurane or isoflurane in oxygen. Before maternal skin incision, the
volatile anesthetic is increased to 2 MAC, and before hysterotomy it is
adjusted as needed to decrease uterine tone. It is of primary importance
to ensure adequate uterine relaxation during the entire duration of
utero-placental bypass to preserve maternal-fetal gas exchange at the
placental interface. Important factors that affect uterine blood flow, and
hence placental gas exchange, are uterine tone, maternal hypotension or
hypertension, and myometrial vasoconstriction from noradrenergic
activity. To maintain uteroplacental perfusion, maternal arterial pres-
sure must be maintained during the anesthetic. Ephedrine is used
because of its primarily beta-adrenergic effects and its minimal effect
on uterine blood flow. Muscle relaxation is maintained with intrave-
nous vecuronium titrated by peripheral nerve stimulation.
Although the fetus is anesthetized as a result of placental transfer of
maternally administered anesthetics, fetal analgesia and paralysis are
ensured with vecuronium (0.2 mg/kg), fentanyl (10 to 20 ␮g/kg), and
atropine (20 ␮g/kg) given intramuscularly either before hysterotomy
under ultrasound guidance, or immediately after exposure of the fetal
head and shoulders.
At the end of the EXIT procedure, before umbilical cord clamping
and delivery of the baby, coordination between the surgery and anes-
thesiology teams is essential to prevent uterine atony and excessive
maternal bleeding. The desflurane or isoflurane is decreased to 0.5
MAC and oxytocin 20 U in 500 mL of normal saline intravenous bolus
followed by 10 U in 1,000 mL drip is administered and titrated to
uterine response. Because of the risks of uterine atony after deep
anesthesia or uterine distension from polyhydramnios, methergine
(0.25 mg) and carboprost (250 ␮g F2-alpha prostaglandin) are on hand
419
and available to anesthesia for intramuscular or intrauterine injection if
needed. After skin closure, the desflurane or isoflurane and air/O2
mixture are discontinued, and 100% oxygen is administered. Glycopy-
rrolate (10 ␮g/kg) and neostigmine (0.7 mg/kg) are given intrave-
nously, and the patient is extubated after obtaining spontaneous breath-
ing, adequate ventilation, and the return of protective airway reflexes.
Airway Control and Delivery of the Fetus
After induction of deep anesthesia, a low-transverse abdominal
incision is performed to expose the uterus. Sterile ultrasonography is
used to map the position of the placenta and guides the site of
hysterotomy. A transverse lower-uterine-segment hysterotomy is per-
formed unless an anterior placenta is present, in which case the
hysterotomy is placed near the uterine fundus or posteriorly. In addi-
tion, the location of the hysterotomy is planned so that there will be no
concerns about the ability to adequately position the fetal head and
neck. The use of a uterine stapling device (US Surgical Corporation,
Norwalk, CT) is essential to perform a hemostatic hysterotomy given
the state of complete uterine relaxation.
It is important in cases associated with polyhydramnios that amniotic
fluid be reduced to near-normal range before EXIT procedure. In the
face of significant polyhydramnios, ultrasound scan may underestimate
the proximity of the placental edge to the hysterotomy. When indicated,
decompression of fetal ascites or cystic mass decompression in cases of
lymphangioma or teratoma is done using a 20- or 22-gauge spinal
needle before hysterotomy. After the hysterotomy, the fetal head and
neck are delivered, but the torso and umbilical cord remain in the
uterus. Infusion of warm Ringers Lactate using the level I warming
device assures maintenance of uterine temperature and normal amniotic
fluid volume and prevents cord compression. Fetal hemodynamics are
monitored using a specially adapted reflectance pulse oximeter that is
placed on the fetal hand and wrapped in aluminium foil and that
measures fetal heart rate (FHR) and PaO2. In addition, continuous fetal
echocardiography allows direct visualization of cardiac hemodynamics
during the entire length of the EXIT procedure. In cases in which a
more extensive surgical procedure is planned, an intravenous line is
inserted in the fetal hand to administer fluid, blood, or medication.
Direct laryngoscopy with a Miller 0 or Miller 1 laryngoscopy blade
and, when necessary, rigid bronchoscopy with a 2.5 mm rigid bron-
choscope is performed followed by endotracheal intubation when
possible. When endotracheal intubation is impossible because of air-
way distortion from external compression by a giant neck mass or to
abnormal airway anatomy such as congenital high airway obstruction
syndrome (CHAOS), a surgical airway is established. A transverse
neck incision usually is used, although extension into a “hockey stick”
configuration sometimes is required to expose the trachea when neck
anatomy is distorted by a neck mass. Once the trachea is exposed, fine
stay sutures are placed, and a tracheotomy is performed. When feasible,
a 6F feeding tube is passed in a retrograde manner up the trachea to the
oral cavity. An endotracheal tube (ETT) then is attached to the feeding
tube and pulled down the trachea establishing endotracheal intubation.
The tracheotomy opening then is closed, and the ETT is sutured to the
mouth to secure the airway. If reverse endotracheal intubation cannot
be achieved, a formal surgical tracheostomy is performed.
Once the airway is established and secured, the neonate is hand
ventilated by ambu bag, and room temperature surfactant is adminis-
tered (Infasurf, 3 mL/kg) by passing a feeding tube in the ETT. While
fetal ventilation is started, the umbilical cord is delivered into the
wound, one umbilical artery is isolated, and a 3.5F umbilical artery
catheter inserted. Similarly, a purse-string 5-0 Prolene is placed in the
umbilical vein and secured as the last step before the cord is clamped.
After inserting umbilical catheters and achieving good fetal saturation,
the umbilical cord is clamped and the baby delivered. The newborn
then is taken to a separate operating room for resuscitation by the
420
neonatology team, and, if necessary, surgery is performed once the
baby is stabilized.
In fetuses with CDH who underwent in utero tracheal clipping, the
EXIT procedure is performed through the site of the previous abdom-
inal and uterine incisions, the fetal neck incision is reopened, and the
trachea and surgical clips are identified. The occluding clips are
removed from the trachea under direct visualization with a broncho-
scope. After clip removal, the bronchoscope is advanced further into
the trachea distal to the occlusion site for inspection, instillation of
surfactant, and endotracheal intubation.
RESULTS
The mean gestational age at the time of surgery was
34.2 ⫾ 3.3 weeks (range, 29 to 40), and it varied
according to indications (Table 1). The EXIT procedure
was performed at an average of 31.8 ⫾ 2.7 weeks’
gestation for reversal of tracheal occlusion in fetuses
with CDH, whereas the average gestational age of the
remaining babies at the time of the EXIT was 35.9 ⫾ 2.6
weeks.
Indications
Reversal of fetal tracheal occlusion. Thirteen fetuses
with severe CDH underwent reversal of tracheal occlu-
sion with surgical clips at The Children’s Hospital of
Philadelphia at a mean gestational age of 26.6 weeks. At
a mean of 31.8 weeks’ gestation, the EXIT procedure
was successful in reversing the tracheal occlusion and
establishing an airway without hemodynamic instability
while on placental bypass in all 13 neonates. A detailed
description of the overall outcomes of these neonates has
been published previously.7
Fetal neck masses. Thirteen fetuses with a large fetal
neck mass were delivered using the EXIT approach
because of the high likelihood of a difficult airway
control at birth with standard delivery techniques. The
average gestational age of these neonates at the time of
the EXIT was 36.0 weeks. The mass consisted of a
lymphangioma in 5, a teratoma in 7, and a foregut
duplication cyst in 1. To help deliver the head and neck,
one fetus with a large cystic teratoma underwent aspira-
tion of 1,820 mL of fluid before the hysterotomy,
whereas 120 mL were removed from a foregut duplica-
Table 1. Indications and Gestational Age at Time of EXIT Procedures
Indications
Reversal of tracheal occlusion
Giant fetal neck mass
EXIT-to-ECMO
Resection of CCAM
Unilateral pulmonary agenesis
Bridge to separation for thoracoomphalopagus
conjoined twins
CHAOS
Overall
Abbreviation: N/A, not applicable.
BOUCHARD ET AL
tion cyst after hysterotomy but before delivery of the
head in another fetus. As previously reported, we have
used the EXIT procedure to control the airway in a fetus
with an extensive cervical lymphangioma in a twin at 35
weeks’ gestation.3 The normal twin was intubated and
delivered first followed by the twin with the large neck
mass who could be intubated uneventfully for a total
time on utero-placental bypass of 35 minutes.
EXIT-to-ECMO. A woman carrying a fetus of 20
weeks’ gestation with a left CDH and a tetralogy of
Fallot was referred for pre- and postnatal management.
The lung-to-head ratio measured at 23 weeks’ gestation
was 1.15. Given the virtual certainty for neonatal extra-
corporeal membrane oxygenation (ECMO) with the
combination and severity of congenital malformations
present in this fetus, an EXIT-to-ECMO strategy was
used at 36 weeks’ gestation to insert arterial and venous
ECMO cannulas in the internal carotid artery and inter-
nal jugular vein, respectively, while utero-placental gas
exchange was maintained. ECMO support then was ini-
tiated before clamping the cord, thus, avoiding a period
of hypoxia, hypercapnea, or hemodynamic instability
during neonatal resuscitation (Fig 1). The baby remained
hemodynamically stable throughout the procedure and
never experienced acidosis or hypoxemia while on utero-
placental bypass for 90 minutes.
CHAOS. A fetus was referred at 31 weeks’ gestation
after a 12 week history of massive ascites secondary to
presumed bilateral congenital cystic adenomatoid mal-
formation (CCAM). A prenatal diagnosis of CHAOS
was made on further imaging studies at our center. Fetal
ultrasound scan showed huge bilateral echogenic lungs, a
dilated proximal airway, and everted diaphragms. The
fetus had massive ascites, anasarca, brain edema, placen-
tomegaly, and polyhydramnios. Fetal MRI (Fig 2) con-
firmed the ultrasonographic findings and localized the
region of airway obstruction to the larynx or proximal
trachea. Because of the severe hydrops, the EXIT strat-
egy was used to deliver the baby and secure the fetal
airway after 2 doses of betamethasone. After hysterot-
omy but before delivering the baby’s head, 700 mL of
No. of Gestational Age
Patients (wk) Range
13 31.8 ⫾2.7 29-37
13 36.0 ⫾ 2.5 32-40
1 36 N/A
1 38 N/A
1 39 N/A
1 34 N/A
1 31 N/A
31 34.2 ⫾ 3.3 29-40
EXIT PROCEDURE
Fig 1. This 36-week-gestation baby underwent an EXIT-to-ECMO
delivery strategy to avoid birth asphyxia and potentially harmful
neonatal resuscitation. After obtaining control of the airway via
endotracheal intubation and placing an intravenous line, arterial and
venous cannulas were inserted. Shown here is the fetus under ECMO
support. Arterial and venous umbilical catheters then were inserted
after which the umbilical cord was clamped and divided.
fetal ascites were aspirated. The head and neck were
exposed, and laryngotracheoscopy showed complete
atresia in the immediate subglottic trachea prompting the
need for a fetal surgical tracheostomy. Surfactant then
was administered and ventilation initiated before clamp-
ing the umbilical cord. The fetus was kept on utero-
placental bypass for a total of 25 minutes.4
CCAM. A fetus was referred at 26 weeks’ gestation
for evaluation of a large right cystic lung lesion. Fetal
ultrasound scan and magnetic resonance imaging (MRI)
showed a complex tumor with solid and cystic compo-
Fig 2. This MRI of the fetus with CHAOS shows huge lungs,
dilated airways, and massive ascites with floating liver and intestine.
421
nents crossing the midline posteriorly. A large systemic
feeding vessel suggested the presence of a hybrid lesion.
The mass maintained a constant size until 36 weeks’
gestation when it doubled in size rapidly achieving a
total volume of 91 mL and a CCAM volume ratio (CVR)
of 2.03, which is obtained by dividing the CCAM vol-
ume by the head circumference to correct for differences
in fetal size. A CVR greater than 1.6 is associated with a
higher chance of hydrops.8 Despite this high CVR, there
was no evidence of hydrops in this fetus. An EXIT
procedure with possible chest mass resection was
planned because of the likelihood of difficult ventilation
at birth. After hysterotomy, the fetal head was delivered,
and the baby was intubated but could not be ventilated
adequately. The mass was resected while the baby was
on placental bypass (Fig 3) with resultant improved
ventilation. The cord was clamped and the baby deliv-
ered after being on utero-placental bypass for 66 min-
utes.
Unilateral pulmonary agenesis. A fetus was referred
to The Center for Fetal Diagnosis and Treatment 19
weeks after unilateral pulmonary agenesis, polyhydram-
nios, and suspected esophageal atresia were diagnosed
on ultrasound at 19 weeks’ gestation. Because of the
association between pulmonary agenesis, esophageal
atresia, and airway abnormalities, an EXIT procedure
was planned in this mother who had undergone previous
cesarean section. After delivery of the head, the baby
was intubated with a size 3.0 endotracheal tube. Subse-
quently, the baby underwent esophageal atresia repair
and rib graft reconstruction of a stenotic bronchus to the
solitary lung.
Bridge to separation in conjoined twins. A set of
thoracoomphalopagus conjoined twins diagnosed at 13
Fig 3. This figure shows the large CCAM resected from a 38-week-
gestation fetus though a right thoracotomy while on utero-placental
gas exchange. At the time of the EXIT, he could not be ventilated
adequately because of this large mass. After resection, there was no
difficulty in ventilation.
422
weeks’ gestation was followed up closely at The Center
for Fetal Diagnosis and Treatment. One twin was shown
to have a rudimentary heart on echocardiography, receiv-
ing its arterial blood supply through a vascular structure
spanning the twins respective thoracic aortas. The preg-
nancy also was complicated by severe polyhydramnios
and progressive ventricular enlargement with cardiac
dysfunction in the normal twin. In addition, the fetuses
shared the liver and a portion of the umbilical vein. An
immediate separation of the twins was planned at birth
with sacrifice of the abnormal baby with the rudimentary
heart. An EXIT strategy was planned at birth because of
the added burden on the healthy twin’s heart and the risk
of losing both babies on interruption of the placental
circulation. The EXIT allowed ample time to access the
fetal airway and place umbilical catheters, and although
ECMO support was considered to sustain life in the good
twin until completion of separation, it was not required.17
Stability of Maternal and Fetal Hemodynamics
The average time on placental bypass from uterine
incision to umbilical cord clamping was 30.7 ⫾ 17.8
minutes (range, 8 to 66). Table 2 depicts the average time
for EXIT procedure for reversal of tracheal clipping and
for airway control in cases of giant neck masses. Also
shown are the individual minutes on utero-placental
bypass for singular indications.
Table 3 illustrates the mean hemodynamic values
recorded during the EXIT procedure. The mean FHR and
fetal PaO2 and oxygen saturation recorded were 153.0 ⫾
38.5 beats per minute and 71.2 ⫾ 19.9, respectively. The
mean cord pH and pCO2 were, respectively, 7.20 ⫾ 0.11
and 63.2 ⫾ 14.6. Among all 31 cases, only 1 incidence
of hemodynamic instability was recorded in a fetus while
on utero-placental gas exchange. It was noted on contin-
uous echocardiographic monitoring that this 36-week-
gestation fetus was becoming bradycardic associated
with a decrease in cardiac contractility. The bradycardia
was secondary to a kink in the umbilical cord and
quickly resolved after the cord was straightened and
uncompressed without sequelae. The fetus received a
Table 2. Duration of Uteroplacental Gas Exchange
Minutes on Uteroplacental
Indications Support (range)
Reversal of tracheal occlusion 26.7 ⫾ 6.3 (18-38)
Giant fetal neck mass 29.2 ⫾ 16.4 (8-54)
EXIT-to-ECMO 58
Resection of CCAM 66
Unilateral pulmonary agenesis 14
Bridge to separation on
thoracoomphalopagus
conjoined twins 43
CHAOS 25
Overall 30.3 ⫾ 14.7 (8-66)
BOUCHARD ET AL
Table 3. Fetal Hemodynamic Data Monitored During
the EXIT Procedure
Mean fetal heart rate 153.0 ⫾ 38.5 beats per minute
Mean PaO2 and oxygen saturation 71.2 ⫾ 19.9%
Mean Cord pH 7.20 ⫾ 0.11
Mean Cord pCO2 63.2 ⫾ 14.6
Estimated blood loss 848.3 ⫾ 574.1 ML
dose of epinephrine through the previously inserted pe-
ripheral intravenous line. There were no recorded epi-
sodes of maternal hemodynamic instability in this series.
Airway Control
The EXIT procedure proved very successful in estab-
lishing an airway in this series of patients with poten-
tially catastrophic airways. An airway was secured in
97% (30 of 31) of neonates. Seventy-seven percent (24
of 31) of patients were intubated successfully endotra-
cheally after direct laryngoscopy or rigid bronchoscopy
with ETT ranging from sizes 2.5 to 3.5 mm.
In six fetuses, it was impossible to intubate the trachea
even using rigid bronchoscopy. Four neonates required a
formal surgical tracheostomy during the EXIT proce-
dure. The first case was a 36-week-gestation female fetus
with a very large cystic teratoma extending from the
floor of the mouth to the anterior mediastinum and
laterally beyond the carotid sheaths and could not be
intubated from above because of the extensive soft tissue
involvement by the lymphangioma. A tracheostomy was
performed during the EXIT procedure, and she under-
went immediate resection of the mass subsequent to
which endotracheal intubation was possible followed by
closure of the tracheotomy site. A tracheostomy once
again was performed at 2 months of age because of
severe tracheomalacia preventing extubation. The second
case was that of a mother who went into premature
labour secondary to polyhydramnios carrying a 33-week-
gestation fetus with a giant heterogenous neck mass. A
tracheostomy was required to secure the airway; how-
ever, once off placental bypass, the baby could not be
resuscitated. The carina was in the neck, and both lungs
were severely hypoplastic from the lungs being pulled up
into the apices of the thoracic cavity. The third case was
a 31-week-gestation male fetus with CHAOS, and the
fourth case was a 33-week-gestation fetus with a very
large cervical teratoma with laryngeal hypoplasia who
required a tracheostomy with tunneling of the tracheos-
tomy tube through the soft tissue of the neck.
In 2 fetuses with very large cervical teratomas (Fig 4),
a retrograde endotracheal intubation was performed
through a temporary tracheotomy. After an unsuccessful
attempt at endotracheal intubation caused by external
compression by the neck mass, the trachea was exposed
through a neck incision, and the teratoma was retracted
EXIT PROCEDURE
Fig 4. Thirty-three-week-gestation fetus with a giant neck tera-
toma. Endotracheal intubation was impossible in the fetus who
underwent retrograde intubation through a tracheotomy after partial
dissection of the mass away from the trachea.
away from the trachea to relieve airway compression. A
tracheotomy allows for the retrograde passage of a feed-
ing tube from the trachea to the mouth. The feeding tube
then is attached to the endotracheal tube, which is pulled
down into proper position. After suturing the ETT se-
curely to the mouth, the tracheotomy is closed with fine
absorbable sutures.
One 36-week-gestation fetus with a gigantic facial and
neck lymphangioma could not be intubated and died
during the EXIT procedure secondary to the extensive
involvement of the face and neck soft tissues by the giant
cystic mass. The family had given strict directives before
surgery not to perform a tracheostomy if endotracheal
intubation could not be performed.
Morbidity and Mortality
Maternal. The average maternal blood loss was
848.3 ⫾ 574.1 mL. Two maternal complications oc-
curred in this series. An immediate complication oc-
curred at the time of the EXIT in a 26-year-old women
carrying a fetus with left CDH who had undergone fetal
tracheal occlusion at 26 weeks’ gestation. During the
EXIT performed at 30 weeks’ gestation, massive bleed-
ing from the placental edges occurred and was controlled
with pledget sutures after a 2.5-L blood loss. Her preg-
nancy also was complicated by polyhydramnios. She did
not require a transfusion and had no recorded hemody-
namic instability. A late complication was noted in a
woman at a subsequent cesarean section for delivery of
her next child. The old hysterotomy scar had dehisced,
and the placenta was visible through the dehiscence.
There was no untoward effect to her new baby. This
woman had undergone 2 fetal procedures: in utero tra-
cheal clipping for her 29-week-gestation fetus with a left
CDH followed by the EXIT procedure for reversal of
423
tracheal occlusion at 37 weeks’ gestation. There were no
maternal deaths in this series.
Fetal. Two fetal complications were noted in this
series. One incidence of fetal bradycardia was mentioned
previously in the hemodynamic section. In addition, after
tracheal clip removal, a 1-mm tear was repaired in the
trachea of a 31-week-gestation fetus with a left CDH
who had undergone tracheal occlusion at 26 weeks’
gestation. Only one fetal death occurred during an EXIT
procedure in a fetus with a large cervical lymphangioma
who could not be intubated and whose parents had
declined a tracheostomy.
DISCUSSION
The EXIT procedure was designed and refined ini-
tially for reversal of tracheal occlusion in fetuses who
underwent fetal tracheal clip application for severe con-
genital diaphragmatic hernias.9 A period of sustained
utero-placental gas exchange was required to partly de-
liver the baby, expose the trachea, and reverse the airway
occlusion. Given the prolonged stability of fetal hemo-
dynamics observed during the EXIT procedure, its use
was expanded to care for fetuses with giant neck mass at
delivery.2,9,19
We have expanded further the use of the EXIT pro-
cedure to care for fetuses affected with a wide variety of
pathologies as shown in this series. An EXIT procedure
may be considered when a condition raises concern
about cardiorespiratory instability at birth using standard
delivery techniques. The EXIT procedure offers the abil-
ity to maximize the chances of survival and to convert a
potentially disastrous emergency situation into a con-
trolled one.
An EXIT procedure is not just a cesarean section, and
the differences between them are striking. During a
cesarean section performed under general anesthesia, a
major goal is to minimize the time from anesthetic to
cord clamping, thereby minimizing the exposure of the
fetus to inhalational anesthetics, and, thus, minimizing
depression of the newborn. In addition, lower MAC of
inhalational agents will prevent loss of uterine tone and
will minimize maternal hemorrhage. In sharp contrast,
during the EXIT procedure, deep inhalational anesthesia
is used to maintain uterine relaxation and, thus, preserve
uteroplacental gas exchange. The duration of placental
support required will vary according to the interventions
performed on the fetus. In fetuses undergoing the EXIT
procedure, fetal depression is not an issue, as all are
intubated and ventilated during and after the EXIT. In
fact, fetal anesthesia and paralysis are supplemented by
direct intramuscular administration of muscle relaxants
(vecuronium) and narcotics (fentanyl) to the fetus to
ensure a motionless, well-anesthetized fetus. There have
been several anecdotal reports of intrapartum laryngos-
424
copy or bronchoscopy for fetuses with neck masses in
which the fetus was delivered (vaginally or by cesarean
section), but the cord was not clamped.10-16 However, in
these cases there was no attempt to prevent return of
uterine tone or uterine contraction. In most, the fetus was
removed from the uterus resulting not only in uterine
contraction but placental separation and nearly immedi-
ate cessation of uteroplacental gas exchange. Maintain-
ing prolonged uterine relaxation with deep inhalational
anesthesia leading to preserved utero-placental blood
flow and gas exchange is the hallmark of the EXIT
procedure. To prevent uterine contraction and placental
detachment, uterine volume is maximized by delivering
only the necessary parts of the fetus, and by constant
infusion of warm Ringers’ Lactate solution to prevent
cord compression. Using these techniques, we have
achieved adequate utero-placental gas exchange in this
series for up to 66 minutes as shown by cord blood pH
and PCO2 levels. Despite prolonged deep inhalational
anesthesia, there was no evidence of fetal cardiovascular
compromise during the EXIT procedure. The single
episode of bradycardia was caused by mechanical kink-
ing of the umbilical cord, and not by fetal hemodynamic
compromise. Fetal hemodynamic compromise, which
may be manifested by bradycardia or fetal SpO2 less than
50%, usually results from hypoperfusion with low car-
diac output or umbilical cord kinkage. Earlier signs of
fetal hemodynamic instability, such as decrease ventric-
ular filling and contractility, may be visualized using
continuous echocardiographic monitoring. Early echo-
cardiographic recognition of the problem prompts a
search for the etiology and rapid intervention before fetal
compromise can occur.
Prematurity is not a contraindication to proceed with
an EXIT procedure. We had fetuses at gestational ages
varying from 29 to 40 weeks, and all tolerated the
procedure well. Fetuses undergoing the EXIT for rever-
sal of tracheal occlusion were, on average, 4 weeks
younger because, delivery was prompted by nonimmune
hydrops from “liver lock” caused by rapid lung growth7
or from preterm labor after fetal surgery. Indications for
proceeding with an EXIT in these babies was preterm
labor, evidence of fetal hydrops, or continuation of preg-
nancy to 37 weeks’ gestation. Despite their prematurity
and the extent of their disease, they tolerated the proce-
dure well. The EXIT also afforded the opportunity to
start surfactant replacement therapy before delivery.
In fetuses with giant neck masses, delivery near or at
term is ideal but seldom achieved. Rapid increase in the
growth velocity of a neck mass may result in severe
polyhydramnios, which may precipitate preterm labor
and delivery. Controlled delivery of the fetus with airway
management is indicated, followed by immediate resec-
tion of the mass if necessary. If signs of fetal compro-
BOUCHARD ET AL
mise develop or preterm labor becomes a problem, an
elective EXIT procedure should be scheduled, even if
early in gestation, rather than to wait for an emergency
procedure performed under less ideal conditions.
The EXIT procedure was well tolerated by the moth-
ers and their fetuses as shown by the low morbidity rate.
The average maternal blood loss of 848 mL is consistent
with the expected blood loss at the time of cesarean
section.6,18 However, the risk of post-EXIT uterine atony
is real and remains a significant potential complication.
The prolonged uterine relaxation maintained during the
EXIT procedure makes the hemostasis provided by the
uterine stapling device indispensable.
Distorted anatomy should be expected when perform-
ing endotracheal intubation or a tracheostomy in fetuses
with giant neck masses. Despite very distorted neck and
airway anatomy, the airway could be secured without
cardiorespiratory compromise in all but one patient
whose family declined a surgical airway: In giant fetal
neck masses, the fetal head often is extended and the
trachea pulled up into the neck. The surgeon must be
aware of this possibility when performing a tracheotomy
that may be quite low distal in the trachea. Additionally,
severe extension can pull the carina out of the chest into
the neck, jamming the lungs into the apices of the
thoracic cavity resulting in pulmonary hypoplasia.
As we have gained experience in performing the
EXIT procedure, important lessons have been learned
and pitfalls addressed. The operative management of
patients with polyhydramnios with amnioreduction
and meticulous placental mapping has helped reduce
the risk of placental hemorrhage from inadvertent
injury during hysterotomy. In addition, in contrast to a
cesarean section in which bleeding from the hysterot-
omy will be controlled rapidly by return of uterine
tone after a quick delivery of the fetus, the use of the
uterine stapler is essential to provide hemostasis dur-
ing a prolonged period. Deep inhalational anesthesia is
crucial to achieve a state of uterine relaxation that
allows prolonged uteroplacental gas exchange. We
now use desflurane instead of isoflurane because of its
more rapid onset of action and elimination to regain
uterine tone after cord clamping. Close coordination is
mandatory between the surgical and anesthesiology
team to achieve proper uterine relaxation during the
EXIT procedure and rapidly regain uterine tone at the
time of actual delivery. We have optimized our means
of fetal monitoring to better detect early signs of
placental shut down using pulse oximetry and contin-
uous fetal echocardiography. Hand ventilation and the
use of surfactant are continued until satisfactory fetal
oxygen saturations are obtained before cessation of
placental support. The placement of umbilical arterial
and venous catheters before clamping the cord ensures
EXIT PROCEDURE
adequate access to the neonatal circulation for post-
natal resuscitation. Further, to avoid the rigor of po-
tentially harmful neonatal resuscitation and birth as-
phyxia, we have used the EXIT-to-ECMO strategy in
1 case. Finally, a lower uterine segment hysterotomy
is used when possible because it allows for future
vaginal deliveries. However, in cases of anterior pla-
REFERENCES
1. Crombleholme TM, Albanese CT: The Fetus with Airway Ob-
struction. The Unborn Patient (ed3). Philadelphia, PA Saunders, 2001;
pp 357-371
2. Liechty KW, Crombleholme TM, Flake AW, et al: Intrapartum
airway management for giant fetal neck masses: The EXIT (ex utero
intrapartum treatment) procedure. Am J Obstet Gynecol, 177:870-874,
1997
3. Liechty KW, Crombleholme TM, Weiner S, et al: The ex utero
intrapartum treatment procedure for a large fetal neck mass in a twin
gestation. Obstet Gynecol 93:824-825, 1999
4. Crombleholme TM, Sylvester K, Flake AW, et al: Salvage of a
fetus with congenital high airway obstruction syndrome by ex utero
intrapartum treatment (EXIT) procedure. Fetal Diagn Ther 15:280-282,
2000
5. Gaiser RR, Kurt D, Crombleholme TM: The cesarean delivery of
a twin gestation under 2 minimum alveolar anesthetic concentration
isoflurane: One normal and one with a large neck mass. Anesth Analg
88:584-586, 1999
6. Letsky EA: Anaemia in obstetrics, in Studd J (ed): Progress in
Obstetrics and Gynaecology. Edinburgh, Scotland, Churchill Living-
stone, 1987, pp 23-58 (chap 2)
7. Flake AW, Crombleholme TM, Johnson MP, et al: Treatment of
severe congenital diaphragmatic hernia by fetal tracheal occlusion:
Clinical experience with fifteen cases. Am J Obstet Gynecol 5:1059-
1066, 2000
8. Crombleholme TM, Coleman B, Hedrick H, et al: Cystic adeno-
matoid malformation volume ratio predicts outcome in prenatally
diagnosed cystic adenomatoid malformation of the lung. J Pediatr Surg
37:331-338, 2002
9. Mychaliska GB, Bealer JF, Graf JL, et al: Operating on placental
support: The ex utero intrapartum treatment procedure. J Pediatr Surg
32:227-231, 1997
Discussion
From the Floor: This is a wonderful series from
CHOP and is a nice extension of the technique described
by Tim and his colleagues in his recent textbook. This is
much-needed information especially with respect to air-
way problems. The 6 patients that did not do well and
required tracheostomy, if you look back, did they all
have MRIs and were there any indications either by sono
or by MRI that they were going to need a tracheostomy
so that all the steps for laryngoscopy and attempt at
bronchoscopy could have been obviated and saved you a
little bit more time?
S. Bouchard (response): Thank you. There was 1 fetus
with congenital high airway obstruction syndrome or
CHAOS, for whom we knew that a neck dissection
425
centa or when adequate access to the fetus is not
possible, the hysterotomy will be placed accordingly.
Situations may be encountered during an EXIT pro-
cedure in which a surgical airway will be impossible to
achieve. A controlled, well-planned and organized pro-
cedure in which one is prepared to face the most difficult
situations will maximize the chance of success.
10. Catalano PJ, Urken ML, Alcarez M, et al: New approach to the
management of airway obstruction in “high risks” neonates. Arch
Otolaryngol Head Neck Surg 118:306-309, 1992
11. Kelly MF, Berenholz L, Rizzo KA, et al: Approach for oxygen-
ation of the newborn with airway obstruction due to cervical mass. Ann
Otol Rhinol Laryngol 99:179-182, 1990
12. Schulman SR, Jones BR, Slotnick N, et al: Fetal tracheal
intubation with intact uteroplacental circulation. Anesth Analg 76:197-
199, 1993
13. Langer JC, Tabb T, Thompson P, et al: Management of prena-
tally diagnosed tracheal obstruction: Access to the airway in utero prior
to delivery. Fetal Diagn Ther 7:21-16, 1992
14. Tanaka M, Sato S, Naito H, et al: Anesthetic management of a
neonate with prenatally diagnosed cervical tumor and upper airway
obstruction. Can J Anesth 41:236-240, 1994
15. Schwartz MZ, Silver H, Schulman SM: Maintenance of placen-
tal circulation to evaluate and treat an infant with massive head and
neck hemangioma. J Pediatr Surg 28:520-522, 1993
16. Stock RM, Egerman RS, Woodson GE, et al: Airway manage-
ment of neonates with antenatally detected head and neck anomalies.
Arch Otolaryngol Head Neck Surg 123:641-645, 1997
17. MacKenzie TC, Crombleholme TM, Jonhson MP, et al: The
natural history of prenatally diagnosed conjoined twins. J Pediatr Surg
37:303-309, 2002
18. Andrews WW, Ramin SM, Maberry MC, et al: Effect of type of
anesthesia on blood loss at elective repeat cesarean section. Am J
Perinatol 9:197-200, 1992
19. Skarsgard ED, Chithara U, Krane EJ, et al: The OOPS (opera-
tion on placental support): In utero airway management of the fetus
with prenatally diagnosed tracheal obstruction. J Pediatr Surg 31:826-
828, 1996
followed by placement of a surgical airway would be
required. A fetal MRI was performed in all cases of
cervical masses, and there were no specific factors iden-
tified to predict which baby would require a surgical
airway. The EXIT procedure was performed in all these
cases because of the concern of the inability to secure an
airway. An oro or endotracheal intubation is preferred
when possible, but the EXIT allows enough time to
proceed to a surgical airway if necessary.
P. Hirschl (Ann Arbor, MI): You had one EXIT to
ECMO. For these cases do you always have ECMO
available? And, if not, how do you decide when ECMO
should be on standby given that it takes time to get the
ECMO circuit prepared?
426
S. Bouchard (response): That is a good question. The
ECMO circuit is not on standby for all of these cases. It
is prepared only if the prenatal evaluation raised a con-
cern of circulatory collapse during the EXIT procedure
or very shortly after birth such as our patient with a
severe diaphragmatic hernia and the cases of conjoined
twins with shared circulation.
C. Albanese (San Francisco, CA): Excluding patients
who had a prior hysterotomy or a low anterior placenta,
how many of these EXIT procedures were you able to do
BOUCHARD ET AL
with a low transverse uterine segment hysterotomy;
therefore, potentially allowing them to deliver vaginally
in the future?
S. Bouchard (response): That is a very good point,
thank you Dr Albanese. As a matter of fact, all cases
without an anterior placenta or previous hysterotomy
were done through a lower uterine segment incision,
unless the fetal positioning did not allow for the
proper delivery of the head and neck to perform the
EXIT.