Fetal Preoperative Imaging
Thierry A. G. M. Huisman
Imaging of pregnant women is currently performed with various imaging modalities that include:
Radiography, scintigraphy, computed tomography, ultrasound (US), and magnetic resonance imaging (MRI).
Technical advances now allow not only the prenatal diagnosis of anomalies but also surgical and/or medical in utero fetal therapy in some cases.
Although US is the primary diagnostic choice for evaluating fetuses and pregnant women, the combination of US and MRI has elevated our capability to carefully select those fetuses who may benefit from fetal intervention.
Over the past 30 years, fetal MRI has become essential in the evaluation of selected fetal anomalies, especially after MRI technology matured to the point of rapid image acquisition that reduces its susceptibility to fetal motion. The excellent soft tissue resolution and detailed visualization of fetal and extrafetal structures frequently confirms or changes diagnoses that influence parental counseling (1).
Fetal MRI is usually performed after 17 weeks of gestation based on a recommendation from the American College of Obstetricians and Gynecologists (ACOG).
Although the safety of field strengths >1.5 Tesla (T) in the early stages of pregnancy is not assured, currently there are no known risks or identified delayed sequelae from 1.5 T fetal MRI. Studies in which a phantom was used demonstrate an increase in fetal body temperature after continuous MRI scanning for 7.5 minutes. Putting this in context, a comprehensive diagnostic fetal MRI examination that includes cranial, thoracic, abdominal, and placental imaging, usually takes 30 to 45 minutes, of which 11.4 minutes in total is required for sequence acquisition (2).
The International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) has provided guidelines for the use of MRI in the assessment of fetal anomalies, which sets out the minimum technical MR protocol requirements for a state-of-the-art fetal MR examination.
Because gadolinium-based contrast agents cross the placenta, cannot be effectively cleared from the fetal system, and have been reported to be associated with postnatal skin and rheumatic conditions, they are not used prenatally.
In this chapter, we focus on imaging some of the most common fetal anomalies amenable to fetal surgical intervention—namely, spinal dysraphism (SD), congenital diaphragmatic hernia (CDH), congenital neck and sacrococcygeal teratomas (SCT), and lung masses.
SD is used as a generic term for the most common central nervous system birth defect that occurs owing to a failure of closure of the neural tube by the fourth week of gestation. Although terms such as neural tube defect and spina bifida are commonly used, the most correct term for this defect is spinal dysraphism (3,4).
The incidence of SD is persistently between 3 and 4/10,000 live births in the United States despite folic acid fortification.
SD is classified as open (nonskin covered) or closed (skin covered).
The open SD is subclassified according to the morphology of the spinal defect.
The most common form of SD is myelomeningocele (MMC) (Figures 1.5.1 and 1.5.2) in which the splayed open spinal cord, known as the neural placode, is extruded dorsally through a defect in the osseous spinal
canal. The neural placode, associated meninges, and cerebrospinal fluid (CSF) bulge beyond the skin of the fetal back.
A less common form of open SD is myelocele (also called myeloschisis) in which the neural placode does not bulge dorsally and is instead level with the adjacent skin.
In Chiari II malformation:
The posterior fossa is abnormally small.
Parts of the cerebellar vermis are below the level of the foramen magnum and within the cervical spinal canal.
Chiari II malformation is frequently accompanied by secondary aqueductal stenosis, which leads to obstructive hydrocephalus and enlargement of the lateral ventricles.
Hydrocephalus may need to be treated postnatally by placing a ventriculoperitoneal shunt or performing a third ventriculostomy (with or without choroid plexus ablation).
As originally described by John Cleland and Hans Chiari, the “hindbrain” is not actively “herniated” into the spinal canal; rather, it is almost exclusively the cerebellar vermis that is developing “ectopically” within the spinal canal. This concept is important because the pathophysiology of Chiari II malformation is frequently misinterpreted as active herniation of cerebellar and hindbrain structures into the spinal canal. The currently accepted unified theory—which combines genetic predisposition with the mechanical consequences of CSF leaking out of the open neural tube defect—is the most rational explanation for the complex etiology of Chiari II malformation (3).
Surgical closure of open SD in early gestation results in a complete or partial reversal of vermian displacement as demonstrated by serial MRI scanning (Figure 1.5.2).
The level and degree of SD is of prognostic value because it correlates with both (i) lower limb function and (ii) need for therapy for hydrocephalus.
At this time, because of the availability of expert first trimester prenatal US, markers of Chiari II malformation (which include compression of the fourth ventricle, intracranial translucency, obliteration of the cisterna magna, or increase in the distance between the brainstem and the occipital bone) are seen as early as 12 weeks, allowing early diagnosis of SD.
In the second trimester, ˜95% of SD cases can be diagnosed with US alone. Classic fetal head US findings are based on the following:
Morphology of the fetal head
Appearance of the cerebellum
Size of the lateral ventricles
The characteristic imaging signs are, respectively, as follows:
“Lemon sign” (inward bowing of the frontal bones on axial views)
“Banana sign” (semicircular-shaped cerebellum on axial views)
In addition, US examination of the fetal spine in the second trimester can accurately localize the site of the osseous and soft tissue defects, which are most commonly seen in the lumbosacral region. Additional spinal findings include the following:
Intracranial findings such as the presence or absence of ventriculomegaly and hindbrain herniation may also be identified.
The level and severity of neurologic compromise can be estimated in most cases by detailed second trimester US examination of the lower extremities, by identifying talipes and other deformities, as well as noting the presence or absence of flexion and extension motions at the hip (L1 and 2), knee (L3 and 4), and ankle (L5 and S1) joints.
Second trimester prenatal US can provide most of the data required to decide whether or not to operate on a fetus with open SD; however, in our opinion, fetal MRI adds dimension of accuracy and precision, particularly when US is limited by large maternal body habitus or unfavorable fetal positioning.
First, the level of the defect and the characteristic intracranial findings can be beautifully demonstrated by using echoplanar and/or balanced steady-state free precession (bSSFP) sequences, allowing a whole-body view that includes the entire fetal spine.
Second, the exact position of the displaced/ectopic cerebellar vermis can be clarified, which is important because there are times when closed lesions (i.e., skin covered myelocystoceles) with little to no vermian ectopia can masquerade as SD, and in utero surgery in such fetuses is unnecessary and places the mother at additional risk.
Third, several specific intracranial findings that may be important when assessing predicted cognitive outcome can be reliably seen with fetal MRI. These findings include the following:
Presence of the corpus callosum which is dysplastic in at least 25% of cases
The patency of the Sylvian aqueduct which might be relevant when assessing the likelihood of persistent CSF circulation abnormalities
Subtle migrational disorders such as subependymal nodular heterotopia
Minor intraventricular hemorrhages that can be sensitively detected by T2*-weighted and echoplanar MRI sequences and then specifically differentiated from subependymal heterotopia based on signal characteristics
Detailed visualization of the posterior fossa and cerebellar vermis
Finally, the larger field of view of fetal MRI contributes to the determination of lesion level of the SD, which is expected to be within one level of concordance between pre and postnatal MRI in about 80% of cases.
US and MRI seem to perform equally well in correctly predicting the postnatal motor deficit level, showing agreement within two segments in up to 80% of cases. Additionally, MRI adds to the visualization of the repair site and allows in utero follow-up after fetal surgery, which is useful in delivery planning.
The pertinent positive MRI findings that we have found most useful in terms of preoperative and 6-week postoperative counseling, pregnancy management, and follow-up are the following:
Presence/absence/distortion of the corpus callosum
Persistent Chiari II malformation
Obliteration of the cisterna magna
Shape/size of the fourth ventricle
Distinction between an open versus closed defect
Regardless of the excellent resolution of US and MRI and the functional assessment capabilities of real-time imaging, babies with SD may ultimately present with worse functional and cognitive outcomes than were predicted prenatally, and caution is advised when counseling parents on expected outcomes based on prenatal findings.
In utero repair with hysterotomy has become more widespread since the management of myelomeningocele study (MOMS) trial showed it to be of value in:
Reducing hindbrain herniation and the need for postnatal shunting
Improving motor outcomes
There are non-SD-related maternal and fetal risks of open hysterotomy in utero surgery, which include preterm birth, premature rupture of membranes, uterine rupture, and need for initial and repeat cesarean section. However, recent advances in fetoscopic surgical approaches have significantly improved maternal and obstetric outcomes while providing the same fetal benefits.
Closed SD rarely requires intrauterine treatment and is consequently routinely managed postnatally as the spinal nerves and cord are typically functional.
CONGENITAL DIAPHRAGMATIC HERNIA
CDH occurs as a consequence of partial or complete agenesis of the diaphragm during embryogenesis leading to the displacement of fetal abdominal organs (stomach, bowel, liver) into the thoracic cavity (6).
This condition is associated with:
Altered lung development (diminished bronchiolar branching, decreased overall arterial cross-sectional area, and abnormal, thickened muscular walls of the peripheral pulmonary arteries) resulting in
You may also need