Daryl H. Chinn

Prenatal diagnosis and natural history of the fetus with a congenital diaphragmatic hernia: Initial clinical experience

To study the accuracy of prenatal diagnosis and define the natural history of fetal congenital diaphragmatic hernia (CDH), we reviewed experience with CDH at The University of California, San Francisco (UCSF) over the last three years. All nine babies born in our institution (inborns) and six of 11 babies referred from other hospitals after birth (outborns) died, an overall mortality of 75%. All had pulmonary hypoplasia. Forty percent had associated malformations or chromosomal abnormalities, a higher incidence than generally reported. Prenatal sonograms were available in all nine inborn cases. CDH was correctly diagnosed prospectively in only five, but could be recognized retrospectively in all nine cases using the sonographic criteria developed from the study. Polyhydramnios was present in all nine cases; in seven cases sonography was performed because the woman was large-for-dates clinically. There were no false positive interpretations, and when necessary the diagnosis was confirmed by amniography. All nine cases of CDH detected in utero died. Seven deteriorated so rapidly that surgical repair could not even be attempted. Two who had optimal care (maternal transport, immediate resuscitation and operation) died after repair despite maximal intensive care including vasodilator therapy. Despite the theoretical advantages of maternal transport to pediatric surgical specialty centers, a majority of fetuses with a prenatal diagnosis of CDH will die because their lungs are inadequate to support extra-uterine life even at term.

Management of ventriculomegaly in the fetus

We studied 24 human fetuses with cerebral ventriculomegaly by serial obstetric ultrasound to define the natural history of fetal ventricular enlargement and to develop a management strategy. In 10 fetuses, ventriculomegaly was associated with other severe anomalies; nine of these families chose to terminate the pregnancy. In three other severely affected fetuses in whom ventriculomegaly was detected serendipitously late in gestation, routine obstetrical management was performed; none survived. Eleven fetuses had ventriculomegaly without associated severe anomalies. Ventriculomegaly remained stable or of moderate severity throughout gestation in nine, resolved gradually in one, and progressed in one who did not have signs of increased intracranial pressure at birth. All of these fetuses were viable; three patients required shunting in the neonatal period, and two others by 5 months of age. Although obstetric ultrasound usually can detect anomalies associated with fetal ventriculomegaly, three fetuses with isolated ventriculomegaly had midline brain malformations that could not be distinguished in utero from hydrocephalus, even in retrospect. Prenatal diagnosis improves perinatal management by allowing counseling, and selective pregnancy termination, or selection of the timing, mode, and place of delivery to optimize outcome. Most fetuses with ventriculomegaly do not require intervention before birth.

Correction of congenital hydrocephalus in utero I. The model: Intracisternal kaolin produces hydrocephalus in fetal lambs and rhesus monkeys*

In the fetus with congenital hydrocephalus, obstruction to the flow of cerebrospinal fluid (CSF) results in ventricular dilation and neurologic impairment. Decompression of the obstructed ventricles before birth may ameliorate the damage and allow normal development to proceed. Although appealing, this pathophysiologic rationale has not been adequately tested because a satisfactory fetal model has not been available. We have developed a model of obstructive hydrocephalus in the fetal lamb and rhesus monkey by injecting kaolin into the cisterna magna through the posterior atlanto-occipital membrane early in the last trimester. Preliminary studies injecting silicone oil were unsuccessful. The development of fetal ventriculomegaly was followed using prenatal ultrasonography. Massive hydrocephalus developed in six sheep, three liveborn at term and three stillborn after premature vaginal delivery, and in 2 fetal rhesus monkeys. All treated animals had external signs of hydrocephalus with marked cranial enlargement. Neuropathologic examinations demonstrated fibrosis of the leptomeninges and subarachnoid spaces around the fourth ventricle. Dilation of the lateral and third ventricles resulted, with attenuation of the cerebral white matter. On histologic examination, the grey matter was relatively well preserved, while the white matter was severely attenuated. This model mimics the clinical and pathologic picture seen in human infants and should allow us to study the pathophysiology of congenital obstructive hydrocephalus and the efficacy and feasibility of its correction in utero.

Ultrasound diagnosis of fetal genitourinary tract anomalies

Specific diagnoses of fetal genitourinary anomalies can be made with current high resolution ultrasound. Accurate diagnosis is essential if fetal intervention is contemplated.