 |
|
||||
| Home | Login | |||||
| About | Editorial board | Articles
|
NSI Publications
|
Search | Instructions | Online Submission | Subscribe | Videos | Etcetera | Contact |
|
||||||||||||||||||||
|
|
|
Endoscopic Third Ventriculostomy And Choroid Plexus Coagulation in Infants: Current Concepts and Illustrative Cases
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.332270
Keywords: Choroid plexus coagulation, endoscopic third ventriculostomy, pediatric hydrocephalus
Every year, almost 400,000 new cases of pediatric hydrocephalus are diagnosed globally.[1] Since the introduction of silicone shunts in the late 1950s,[2] diversion of cerebrospinal fluid (CSF) to the peritoneal cavity or other distal sites has been the main surgical treatment for this disease in most countries. Ventriculoperitoneal shunting is an essential skill set among general neurosurgeons and pediatric neurosurgeons alike,[3] and the life-saving operation may be performed even in primary hospitals.[4] Despite advances in shunt technology, however, the failure rate during the first year of shunt implantation remains between 20 and 40%.[5] Implementation of shunt protocols and the use of antibiotic-impregnated catheters have reduced shunt infection rates to 2–6%, but figures vary widely across centers.[6],[7] To many children, these shunt complications pose a life-long threat to neurocognitive function and may significantly impair their quality of life.[8],[9] With the goal of reducing shunt dependency, there has been renewed interest in the combined use of endoscopic third ventriculostomy (ETV) and choroid plexus coagulation (CPC) among infants with hydrocephalus, gaining momentum in the last two decades. While many studies supporting the efficacy and safety of ETV+CPC were initially conducted in low-resource settings in East Africa,[10],[11],[12] completed and ongoing prospective trials in North America are helping define the role of ETV+CPC in the contemporary management of pediatric hydrocephalus.[13],[14],[15] In this article, we give an overview of ETV+CPC among infants. Illustrative cases are provided to demonstrate clinical decision making for pediatric neurologists and neurosurgeons who intend to incorporate ETV+CPC into their practice. ETV+ CPC in young infants The history of ETV and CPC for the treatment of hydrocephalus has been extensively discussed elsewhere.[2],[16],[17] In 2005, working at Cure Children's Hospital of Uganda (CCHU), Benjamin Warf published his analysis of 550 children who underwent either ETV alone or ETV+CPC.[11] These were mostly infants, with 58% of patients having postinfectious hydrocephalus. Warf showed that among patients less than 1-year old, the addition of CPC to ETV increased the proportion of successful treatment from 47 to 66%. Success was defined as being shunt-free on last follow-up, with a median duration of 9.2 months for the ETV+CPC cohort. Further, the intervention was more likely to succeed in infants with hydrocephalus that was not infectious in etiology, and in those with myelomeningocele. Subsequent studies in CCHU have demonstrated similarly good outcomes in patients with aqueductal stenosis,[18] Dandy Walker complex,[19] encephaloceles,[20] and congenital idiopathic hydrocephalus.[21] Driven by these encouraging numbers from Uganda, neurosurgeons from around the world have since published their own experiences with ETV+CPC. In Nigeria and Haiti, success rates of 75% (out of 20 patients) and 52% (out of 82 patients), respectively, were reported after 6 months of follow-up.[22],[23] Among patients that Warf treated in Boston, 57% did not require any additional surgery for hydrocephalus at 1 year.[24] In this first ETV+CPC series in North America, scarring of the prepontine cistern and prior CSF diversion were demonstrated to be additional predictors of poor outcome. A more recent prospective study from the Hydrocephalus Clinical Research Network involving 191 patients demonstrated a 48% success rate for ETV+CPC at 6 months, only slightly declining to 45% at 18 months.[14] ETV failure was more likely in patients less than 1 month of age and in those with hydrocephalus from intraventricular hemorrhage of prematurity. How does ETV+CPC work? In the bulk flow model of CSF circulation,[25] CSF secreted predominantly by choroid plexus flows within and exits the ventricular system to reach the intracranial and spinal subarachnoid space. CSF is subsequently absorbed in the dural venous sinuses through arachnoid granulations. When hydrocephalus is obstructive in nature (e.g., aqueductal stenosis), the rationale behind an ETV is fairly straightforward: an alternate pathway from the third ventricle to the prepontine cistern is created, allowing CSF to bypass the obstruction. However, as studies have shown, ETV alone is often inadequate for addressing hydrocephalus in young infants.[26],[27] This is presumed to be because of insufficient CSF absorption through the arachnoid villi in this age group, where other pathways such as periventricular capillary absorption may play a bigger role.[28] It is theorized that by cauterizing choroid plexus in lateral ventricles, CSF production is reduced while waiting for absorption pathways to mature, and over time, transition to adult circulation.[16] Where these mechanisms fall short is in explaining how good outcomes have been achieved with ETV+CPC, even when (1) hydrocephalus is idiopathic, where no obvious anatomic obstruction is apparent,[21] and (2) CSF production persists from residual choroid plexus and extrachoroidal sources such as ependyma and cerebral capillaries.[28] Using the hydrodynamic model of CSF,[25] it is postulated that pediatric hydrocephalus is not just a consequence of high-ventricular CSF pressure. Warf proposes that choroid plexus pulsation is also a significant driver of ventricular enlargement.[21] Thus, coagulating choroid plexus decreases these rhythmic, intraventricular forces during the cardiac cycle, especially important in a young infant whose brain parenchyma has high compliance. CPC also leads to a consequent reduction in cerebral venous pressure, thereby increasing capillary absorption of CSF. In keeping with this mechanism, the ETV stoma acts as a pulsation absorber, dissipating pulsatile forces in the ventricular cavity to the basal cisterns. The combined effects of ETV (“pulsation absorber”) and CPC (“pulsation reducer”) restore equilibrium between CSF production and absorption in the infant brain.[21] The goal is a reduction in intracranial pressure to enable normal brain growth and development. Identifying ideal patients for ETV+CPC Whether in East Africa or North America, it is evident from published literature that the benefits derived from ETV+CPC vary among subgroups of patients. Thus, the onus is on the neurosurgeon to identify which infants are suitable candidates for ETV+CPC. For clinicians, the ETV success score (ETVSS) developed by Kulkarni and colleagues is a useful and externally validated tool for estimating the likelihood of success at 6 months following an ETV.[29],[30] It considers three patient factors: age, etiology of hydrocephalus, and prior shunting. To illustrate, a 6-month-old infant with aqueductal stenosis and no prior shunt has an ETVSS of 70, whereas a 1-month-old infant with communicating hydrocephalus has an ETVSS of 20. Thus, it would be more reasonable to offer ETV+CPC as primary treatment for hydrocephalus in the former. While a low ETVSS (≤ 40) does not preclude ETV+CPC, parents must be counseled so that appropriate expectations are set. It is important to communicate that there may be a need for early reintervention particularly within the first 3 months; this risk is weighed against shunt dependency with its life-long risk of multiple shunt revisions. Neuroimaging is a requisite for safe neuroendoscopy. At the minimum, a computed tomography (CT) of the brain should be obtained. It allows the neurosurgeon to establish an etiologic diagnosis, plot the trajectory for the ETV, identify anatomic variations, and inspect the thickness of the third ventricular floor, adequacy of the prepontine space, and the basilar artery's position and course. If septations, loculations, and/or abnormal contrast enhancement are present, a transfontanel ventricular tap is useful to rule out active infection. Magnetic resonance imaging (MRI) of the brain is ideal but may not always be feasible. T2-weighted sequences in three planes provide sufficient information, supplemented by sagittal FIESTA/CISS sequence to show the anatomy of the Liliequist membrane and any cisternal scarring. If extensive scarring and/or redundant membranes are seen on MRI, it is prudent to proceed with VP shunting instead due to the high rate of ETV failure seen in these patients.[31] In limited-resource settings where only an ultrasound is available, neurosurgeons are advised to proceed with caution. The third ventricular floor, prepontine space, and other posterior fossa structures are not clearly delineated on an ultrasound—especially in cases of severe macrocephaly. Characteristics known to influence ETV success, which may otherwise be apparent on MRI, may not be observed until the time of intraoperative endoscopic inspection. It is reasonable in these limited facilities to adopt a “scope-first” paradigm: patients are offered an upfront endoscopic approach with the understanding that a shunt may ultimately be deemed most appropriate. In such cases, parents are counseled accordingly, and patients are positioned and prepped in the operative suite for a possible VP shunt insertion. A low threshold is maintained for aborting the ETV procedure when intraoperative findings warrant: an opaque third ventricle floor, heavy prepontine cistern scarring, or nonvisualization of the basilar artery. Equipment needs and surgical technique In the illustrative cases that follow, ETV+CPC was performed at the Philippine General Hospital using a Karl Storz flexible neuroendoscope, following the technique described by Warf.[11] The endoscope is held in place using a scope holder that attaches to the bed frame [See [Figure 1]]. Under general anesthesia, the patient is positioned supine with the head slightly elevated and turned 90° to the contralateral side. The head, neck, and abdomen are prepped and draped in case the procedure needs to be converted to shunting. A semilunar incision is made over the lateral portion of the anterior fontanel, corresponding to Kocher's point. The dural opening is kept to a minimum, and after coagulation of pia, the lateral ventricle is cannulated using a blunt trocar or 8 French feeding tube. CSF is obtained for analysis, and the neuroendoscope is introduced through this newly created tract. The ventricular anatomy is inspected before navigating to the floor of the third ventricle.
ETV is performed first. The stoma is created using the tip of the bugbee wire without cautery. It is then enlarged using a Fogarty balloon, or with experience, gentle traction in different directions, avoiding the surrounding neurovascular structures. In most cases, it is possible to insert the flexible scope through the stoma into the prepontine cistern. The aim is to completely perforate the Liliequist membrane, visualize the naked basilar artery, and clear any redundant arachnoid tissue. At this point, the decision is made whether to proceed to CPC or not. In patients ≥ 1-year old, ETV alone may be sufficient. In patients with cisterns that are extensively scarred, particularly in postinfectious or posthemorrhagic hydrocephalus, it is preferable to leave a shunt. To perform CPC, the neurosurgeon must let the tip of the bugbee wire glide on the surface of the choroid plexus, to and fro, applying monopolar diathermy until the vessels change in color from red/yellow-orange to pale yellow. CPC is accomplished in a logical manner: beginning at the foramen of Monro, tracing the choroid plexus to the atrium, coagulating the glomus, and finally, navigating to the temporal horn. The tip of the monopolar electrode must not be buried within the choroid plexus, and careful attention is given to the surrounding veins and ependyma. If necessary, a septostomy is created. CPC is then performed on the contralateral side following the same sequence. Whenever possible, the neurosurgeon should aim for bilateral complete CPC. Cauterizing >90% of choroid plexus in the lateral ventricles has been associated with a greater chance of treatment success.[13],[14] After CPC, the ETV is inspected to ensure patency and clear any blood clot or debris. The endoscope is gently withdrawn, checking for bleeding along its tract. Watertight dural closure and meticulous apposition of galea and skin are essential to reduce the risk of CSF leak and subsequent infection. Rigid endoscopes, with their better optics and greater familiarity to the operator, have also been used to accomplish ETV+CPC.[23] However, flexible endoscopes have the advantage of being able to maneuver to the temporal horns even when ventricles are only moderately enlarged.[32] There is a learning curve with the flexible endoscope, although it is less steep with prior experience in rigid neuroendoscopy. Formal training (i.e., working alongside CCHU neurosurgeons) has been associated with greater extent of CPC.[14] Complications and postoperative care A 2016 review of 11 studies with 524 patients showed that adverse events after ETV+CPC occurred in 3.7%, with a mortality rate of 0.4%.[33] Complications included hemorrhage, CSF leak, meningitis, ventriculitis, SIADH, subdural hygroma, and seizures. Perhaps the most dreaded intraoperative complication is bleeding, but this is usually controlled with irrigation. If needed, gentle tamponade may be applied using the endoscope tip or an inflated Fogarty balloon. In the event of major hemorrhage, the procedure is aborted, and an external ventricular drain left in place. Head elevation and intermittent irrigation using lactated Ringers help reduce the occurrence of pneumocephalus, overdrainage, and massive subdural collections. Because of the absence of hardware, the risk of infection following ETV is low (~2%).[34],[35] Prophylactic antibiotics are given during induction following institutional protocols, and layered closure is paramount. Vomiting and fever are not uncommon during the first 72 h, and these are managed medically. Because of the approximately 5% risk of postoperative seizures, some neurosurgeons have advocated for prophylactic antiepileptics.[36] In the authors' practices, prophylactic antiepileptics are not routinely prescribed. Patients are usually sent home within 3 days postop, unless they live far away, in which case they are kept admitted for a bit longer to monitor for early complications. Long-term follow-up Compared with shunting, ETVs have a higher rate of early failure.[37] Most ETVs fail within 3–6 months following surgery. In Warf's original ETV+CPC series, median time to failure was 1.4 months, and 75% occurred within 2 months.[11] Corollary to this, an ETV that is still functioning at 6 months is likely to remain patent beyond this period, while a shunt inserted at the same time in an identical patient will continue to be at risk for failure. Therefore, follow-up is generally recommended at 2 weeks to check for wound complications; subsequently, at 6 weeks, 3 months, and 6 months to monitor for early ETV failure; and then less frequently thereafter. The frequency of surveillance imaging depends largely on the setting. At Vanderbilt University Medical Center, fast-sequence MRI is obtained at 2 weeks, 6 weeks, 3 months, 6 months, and 1-year postop. In contrast, at the Philippine General Hospital, where capacity for imaging is limited and CT is more affordable and readily available than MRI, post-op scans are advised at 6 months, but may be reasonably delayed to 1 year in patients who are otherwise clinically improved. Patients with incomplete CPC, evidence of past infection, and cisternal scarring need to be monitored closely because they are at a higher risk of failure. These are reflected in CCHU's modified ETVSS[38] that has been used to reliably predict outcome after ETV+CPC in Uganda. When ETV failure is suspected, ultrasound may be requested initially to assess the thickness of the cortical mantle and size of the ventricles. It is known that ventriculomegaly may persist following an ETV, and the presence of large ventricles alone—in the absence of definite progression and signs of raised intracranial pressure—does not necessarily equate to ETV failure.[39] Dewan and colleagues recommend monitoring three signs: bulging fontanel, rapid increase in head circumference as plotted on normal curves, and progression of ventriculomegaly using either frontal and occipital horn ratio or Evan's index.[40] Failure occurs because of two reasons: stoma closure or inability of ETV+CPC to restore equilibrium between CSF production and absorption.[41] As such, many have advocated for repeat neuroendoscopy when a patient presents with signs of ETV failure. Intraoperatively, if the stoma is found to be closed or occluded, it is reopened (“redo/repeat ETV”), and any underlying arachnoid scarring or membrane is dissected free. On the contrary, a patent stoma suggests that the latter reason is responsible for ETV failure, and in this scenario, a VP shunt is inserted. ETVs that fail early (i.e., <3 months after initial surgery) are more likely to require shunt placement.[41] Of note, among patients who eventually need a shunt following a failed ETV+CPC, the endoscopic procedure does not seem to increase the subsequent risk of shunt infection or malfunction.[42] To this day, questions on the long-term durability of ETV+CPC and its effect on neurocognitive development of children with hydrocephalus have lingered, and this is the primary reason for the cautious adoption of ETV+CPC elsewhere around the world. A randomized trial of 100 Ugandan infants with postinfectious hydrocephalus showed no significant difference in cognitive outcome, using the Bayley Scales of Infant Development at 12 months after either ETV+CPC or VP shunting.[43] Patients who underwent endoscopic treatment had larger ventricles on postop imaging, but median brain volume was similar for both groups. This highlights the importance of overall clinical assessment in the continuing care of children after ETV+CPC. Illustrative cases Case 1: Congenital hydrocephalus An 11-month-old boy presented with macrocephaly and a bulging anterior fontanel [See [Figure 2]]. His pediatrician started acetazolamide and requested for an MRI, which showed a posterior fossa cyst, likely to be a Blake's pouch cyst, associated with supratentorial hydrocephalus. This relatively older infant had noninfectious hydrocephalus, and thus, a favorable ETVSS of 70. ETV+CPC was performed. On follow-up, his head circumference remained above average; however, the rate of growth had decelerated, and the measurements plateaued over time. The anterior fontanel remained flat and eventually closed, despite discontinuation of diuretics. Consistent with his clinical findings, repeat CT showed a decrease in the size of the ventricles with an increase in the thickness of the cerebral cortex. The child has remained well and on par with developmental milestones.
Case 2: Idiopathic communicating hydrocephalus A 6-month-old girl with no known history of perinatal infection was referred for evaluation due to progressive macrocephaly and preferential downward gaze (“sunset eyes”). CT showed communicating hydrocephalus with no abnormal enhancement [See [Figure 3]]. After discussion with the parents, ETV+CPC was offered (ETVSS: 40) and performed without complications. Postop, there was an initial reduction in head circumference. However, at 4 months post-ETV, the increase in head circumference began to accelerate once more, associated with a bulging fontanel. Repeat CT revealed progression of the ventriculomegaly. Redo endoscopy was contemplated, but due to the etiology of the patient's hydrocephalus, the decision was made to proceed with VP shunting. As expected, the head circumference stabilized, and surveillance scan showed regression of the ventriculomegaly. Even with failure of ETV+CPC, the treatment outcome remained favorable, and she had no delay in her neurocognitive development, emphasizing the importance of closely monitoring these patients.
Case 3: Shunt malfunction A 6-month-old infant who previously underwent lumbar myelomeningocele repair and shunt insertion presented with frontal bossing and a bulging fontanel, consistent with shunt malfunction [See [Figure 4]]. There were no signs of infection. His CT showed favorable anatomy for a safe ETV; hence, neuroendoscopy was performed (ETVSS: 50). After ETV, the shunt was removed under direct visualization, cauterizing the choroid plexus that had occluded the ventricular catheter and averting intraventricular hemorrhage. Bilateral CPC was accomplished. The signs of raised intracranial pressure resolved and the patient remained shunt-free on last follow-up. This illustrates that previously shunted patients, especially those with myelomeningocele and aqueductal stenosis should be evaluated if they can be candidates for ETV+CPC when they are readmitted for shunt failure, considering the positive outcomes this subgroup of children can realize.
In appropriately selected infants, combined ETV+CPC is an effective and safe treatment for hydrocephalus. A successful outcome is seen in approximately 50–70% of patients in most series, with greater likelihood of shunt freedom among infants older than 1 month, those with noninfectious etiology such as aqueductal stenosis and myelomeningocele, and those without prior shunting. ETV failure usually occurs within 3–6 months, particularly when CPC is incomplete, cisterns are scarred, and hydrocephalus is postinfectious in origin. Studies on long-term functional and developmental outcomes will guide clinicians in future decision making. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||
