Literature
Clinical Pearls & Morning Reports
Published November 30, 2022
How often do CSF shunts to treat hydrocephalus fail in children?
Hydrocephalus is a well described but incompletely understood disorder in infants and children. Management strategies are being developed as alternatives to cerebrospinal fluid (CSF) shunting but are not universally effective and may have adverse consequences. Read the NEJM Review Article here.
Clinical Pearls
Q: What are some well-known causes of hydrocephalus?
A: Various impediments, congenital or acquired, can derange the CSF circulation and result in hydrocephalus. Common congenital causes are myelomeningocele, aqueductal stenosis, and posterior fossa malformations (particularly Chiari and Dandy–Walker malformations). Acquired disorders that commonly cause hydrocephalus are tumors, hemorrhages, and infections. The most common cause of hydrocephalus globally in infants and children is infection. Hydrocephalus often follows neonatal sepsis and can be caused by a variety of pathogens. In the United States, the most common cause of hydrocephalus in the pediatric population is intraventricular hemorrhage associated with prematurity.
Q: Can hydrocephalus be diagnosed and treated in utero?
A: Fetal ventriculomegaly, defined as abnormal enlargement of the atria of the lateral cerebral ventricles (to a diameter >10 mm), can be detected on ultrasonography during the midtrimester of pregnancy. This finding may be due to hydrocephalus, but it can also result from abnormal brain development, particularly if there is global cerebral atrophy that is accompanied by passive enlargement of the ventricles. The difficulty of making an accurate diagnosis based solely on imaging during the fetal period complicates decisions about early treatment for fetal hydrocephalus. Interventions for fetal ventriculomegaly, such as early delivery and in utero drainage of CSF, are associated with high complication rates and no clear clinical benefits and therefore have not been recommended.
Morning Report Questions
Q: How often do CSF shunts to treat hydrocephalus fail in children?
A: Approximately 30 to 40% of shunts placed in infants fail in the first year, 40 to 50% fail by 2 years, and 80 to 90% fail by 10 years. The average patient undergoes 2.5 shunt revisions during childhood, with 5% requiring 10 or more revisions. Shunts fail for various reasons, but the most common cause is obstruction at some point in the system. Obstruction most often occurs in the ventricular catheter and is caused by reactive tissue that occludes the holes in the tubing and is composed predominantly of astrocytes and macrophages. Infection, the second most common cause of failure, usually occurs as a result of bacterial contamination during shunt insertion. Many studies have been conducted in an effort to identify risk factors that could be modified to reduce shunt failure; however, little progress has been made.
Q: What procedure is the main long-term alternative to CSF shunting for hydrocephalus?
A: The main long-term alternative to CSF shunting is endoscopic third ventriculostomy (ETV). In this procedure, an endoscope is passed through a frontal scalp incision into the third ventricle, and an opening is made in the ventricular floor. This procedure allows CSF to flow between the third ventricle and the subarachnoid space just below it. A successful outcome of this procedure is generally defined as the resolution of hydrocephalus symptoms, a modest reduction of ventricular size, and evidence of CSF flow through the fenestration on magnetic resonance imaging. Failures occur when the signs and symptoms of hydrocephalus reappear or ventriculomegaly fails to stabilize. Most ETV failures happen within 6 months after the procedure; subsequently, the failure rate is less than 5%. Neonates and infants have the highest rates of ETV failure. In patients under 6 months of age, the chances of success have been estimated to be 50% or less, depending on the cause of the hydrocephalus. Many neurosurgeons do not attempt the procedure in this age group.