21 Endoscopic Third Ventriculostomy

21.1 Introduction

Endoscopic third ventriculostomy (ETV) establishes an alternate route for cerebrospinal fluid (CSF) drainage from the ventricular system to the cortical subarachnoid space via connecting the third ventricle into the basal cisterns. The use of ETV for the treatment of hydrocephalus has evolved concurrently with the progress in endoscopic technology, and ETV is now a routine procedure at most major neurosurgical centers. The classical use of ETV is for the treatment of hydrocephalus arising from third ventricular outflow compromise, such as in tectal gliomas, pineal region tumors, and fourth ventricular tumors, or congenital aqueductal stenosis.1,2,3,4 These etiologies are natural candidates for ETV since the cortical subarachnoid space has preserved its resorptive function. However, the use of ETV for the treatment of other forms of hydrocephalus, such as posttraumatic hydrocephalus, postinfectious hydrocephalus, and posthemorrhagic hydrocephalus, has also been reported.5,6 In this chapter we will review the indications, surgical techniques, complications, and outcomes of ETV.

21.2 Indications/Contraindications

ETV is used to treat a variety of hydrocephalic states (Table 21.1) and is primarily indicated for hydrocephalus due to third ventricular outflow obstruction in children and adults from such pathologies as aqueductal stenosis, tectal gliomas, pineal region tumors, tumors or cavernous malformations of the midbrain, and tumors of the fourth ventricle (Fig. 21.1). ETV allows the egress of CSF from the third ventricle to the basal cisterns and cortical subarachnoid space, where it can be reabsorbed normally.

**Fig. 21.1** Representative examples of patients treated with endoscopic third ventriculostomy (ETV). **(a)** Sagittal T1-weighted magnetic resonance image (MRI) of a 50-year-old woman with a pineal region mass obstructing the aqueduct. **(b)** T1-weighted MRI of a 60-year-old woman with a low-grade glioma of the thalamus obstructing the third ventricle. **(c)** 69-year-old woman with shunt failure and ventriculomegaly. At the time of ventricular exploration, a web was seen obstructing the third ventricle. She was treated with ETV as part of a shunt removal protocol. (Reproduced with permission from Rekate HL. Selecting patients for endoscopic third ventriculostomy. Neurosurg Clin N Am 2004;15(1):39–49.)

Numerous studies have examined the variables that correlate with ETV success or failure. The etiology of hydrocephalus is a key element in predicting ETV success; dysfunction of the subarachnoid space such as with postinfectious or posthemorrhagic etiologies are associated with higher rates of ETV failure.3,7 Very young (under 6 months of age) patients have a high rate of ETV failure, which is likely due to a combination of the etiology of the hydrocephalus and the young age itself, with a presumed milieu of growth factors and lower rates of long-term ventriculostomy patency.8,9,10,11 Furthermore, patients with previous shunts are less likely to have long-term success with ETV; this may be correlated with hydrocephalus etiology or be due to loss of normal cortical CSF reabsortion mechanisms that occurs with long-term shunting.12 In pediatrics, the ETV Success Score has been validated by multiple groups to predict ETV success at 6 months and longer (up to 36 months) after surgery.7,13,14 This score is comprised of three numeric variables: age at surgery, etiology of hydrocephalus, and previous shunting procedure. Higher scores are associated with higher chances of long-term ETV success. However, no such scale exists for adult patients.

**Table 21.1** Indications for endoscopic third ventriculostomy and reported success rates
Etiology	Reported success rate
Idiopathic aqueductal stenosis	60–94%2,3,32,33
Tectal and pineal tumors	81–100%2,3
Fourth ventricular tumors	74–80%2,3
Arachnoid cyst	67–80%2,3

Although the primary indication for ETV is third ventricular outflow obstruction, numerous authors have reported the successful use of ETV in other etiologies such as postinfectious, posthemorrhagic, normal-pressure hydrocephalus, and long-standing overt ventriculomegaly (LOVA).5,6,15,16 The physiological basis by which ETV works in these applications is not fully understood. Small, randomized trials have demonstrated moderate clinical superiority of ventricular shunting over ETV in normal-pressure hydrocephalus and hydrocephalus following tuberculous meningitis, and the application of ETV in these and other pathologies with CSF absorption derangement must be undertaken cautiously.17,18 We remain skeptical of the efficacy of ETV in true cases of normal pressure hydrocephalus.

Preoperative imaging findings can yield insight into the success rate of ETV for a given patient. Multiple groups have found an association between third ventricular floor deformation, or bowing, on magnetic resonance imaging (MRI) and ETV success.19,20 ETV success is also associated with preoperative displacement of the lamina terminalis, and this can be measured via a standardized index using MRI that quickly changes after successful ETV.20

Relative contraindications to ETV include:

Basilar artery apex aneurysm
Postinfectious and posthemorrhagic hydrocephalus
Slit ventricles
Patient age < 6 months old

The risks of ventricular fenestration with any basilar apex vascular abnormality outweigh the benefits of ETV and these patients should receive traditional shunting. With postfungal meningitis hydrocephalus, such as in coccidioidomycosis (Fig. 21.2), or in tuberculous meningitis hydrocephalus, the arachnoid granulations are scarred and subarachnoid space CSF resorption is dysfunctional. The same physiology is observed in patients with intraventricular hemorrhage of prematurity or aneurysmal subarachnoid hemorrhage. Lastly, slit ventricles make ETV technically difficult and increase the risk of blunt injury to important nearby structures such as the deep veins, the fornix, and the thalamus. Patients under 6 months of age have a very high rate of ETV failure, and so the use of ETV in this population must be carefully weighed and if performed, close follow-up is essential.

**Fig. 21.2** A 24-year-old male with coccidioidomycosis fungal meningitis with extensive inflammation of the basilar cisterns. He may not be a good candidate for endoscopic third ventriculostomy (ETV because he has not only a cerebrospinal fluid [CSF] circulation block at the outlet of the fourth ventricle but also within the basal cisterns).

21.3 Operative Technique

21.3.1 Preoperative Preparation

The surgeon and assistant stand at the vertex of the patient’s head and look toward the surgical monitor and image guidance monitor near the patient’s feet. The anesthesiologist is positioned on the patient’s left and the surgical scrub assistant is positioned to the patient’s right. General anesthesia is induced in the usual fashion, and the patient’s head is positioned supine, with a small amount of flexion. If using a rigid endoscope, frameless stereotaxy is to be used (we recommend it), a Mayfield head holder is applied; otherwise, a horseshoe head holder or foam donut is sufficient. In the majority of cases, a right-sided approach is preferred. Only in unusual scenarios such as prior mesh cranioplasty or right frontal lobe pathology would a left-sided approach be chosen.

The proper placement of the bur hole for endoscope insertion makes the remainder of the procedure better tolerated by surgeon and patient alike. An ideal trajectory will allow the endoscope to pass from the bur hole into the foramen of Monro and visualize the floor of the third ventricle with minimal manipulation of adjacent structures. The fornix may be injured if excessive retraction on the foramen of Monro is required. Traditionally the bur hole for ETV is placed 2.5 to 3 cm lateral to the midline and immediately anterior to the coronal suture. However, our group has studied the ideal placement of the bur hole for ETV using neuronavigation and found the ideal trajectory entry point to vary significantly from patient to patient, with individual trajectories ranging up to 3 cm anterior or 3.5 cm posterior to the coronal suture. Furthermore, due to the natural course of the coronal suture, the bur hole site should be moved more posteriorly as it is moved laterally (Video 21.1).21,22 In sum, these studies demonstrate the usefulness of intraoperative neuronavigation in the routine performance of ETV due to significant variance in patient anatomy. If a flexible neuroendoscope is used, the standard Kocher’s point (2.5 to 3 cm lateral to midline and immediately anterior to the coronal suture) is usually adequate (Video 21.2).

Prior to the initiation of the surgical procedure, we inspect every piece of endoscopic instrumentation to be used during the ETV. This includes the endoscope itself, for image clarity, monitor functionality, self-irrigating features, etc., as well as the balloon catheter and forceps. This allows for effective troubleshooting prior to skin incision.