25 Endoscopic Shunt Placement

25.1 Introduction

The role of endoscopic techniques in the treatment of neurosurgical conditions continues to grow. Neuroendoscopy can be used as a primary technique, as in endoscopic third ventriculostomy, or as an adjuvant technique to assist in an open procedure, and to provide the advantage of direct visualization to target anatomical structures in procedures previously performed using known external landmarks and their relationship to internal structures to locate target structures. One such procedure is the insertion of the proximal ventriculostomy catheter during ventriculoperitoneal shunting for cases of hydrocephalus. One of the first reports using endoscope guidance for assistance of proximal catheter insertion noted the benefit of accessing the ventricular system with certainty prior to insertion of the ventricular catheter.1 However, a multicenter randomized trial of endoscopic shunt insertion for all cases of ventriculoperitoneal shunting demonstrated no improvement in rates of shunt failure over traditional shunt-insertion methods.2,3 Nonetheless, perhaps neuroendoscopy’s greatest advantage is for the subset of patients with complex ventricular anatomy, a sentiment echoed by several authors.4,5,6 In this chapter, we will review the indications, operative technique, potential benefits, and complications of endoscope-assisted shunt placement.

25.2 Indications/Contraindications

Endoscope assistance for proximal catheter insertion during ventriculoperitoneal shunting is indicated for patients with hydrocephalus in whom the ventricular anatomy has been distorted from acute or chronic pathologic states or in whom ventricular access is challenging (such as those with slit ventricles). This distortion precludes the usefulness of traditional external landmarks for localization of structures in the ventricular system during blind catheter insertion and increases the risk from freehand catheter placement.

A multicenter, randomized trial evaluating the utility of endoscope-assistance for all-comers demonstrated that routine use of neuroendoscopy does not reduce the incidence of shunt failure at 1 year (42% versus 34%).2 However, smaller studies evaluating the utility of neuroendoscopy in cases of complex ventricular anatomy have found increased ease of successful ventriculostomy with image guidance. Roth et al found that neuroendoscopy aided in appropriate ventricular catheter location in 13 of 16 cases where the children had complex ventricular anatomy and concluded that the primary value of endoscopy during shunt procedures is achieved in patients with small and distorted ventricles.4

Prior attempts at ventriculoperitoneal shunting, endoscopic third ventriculostomy, or other endoscope-assisted or non–endoscope-assisted procedures are not contraindications to neuroendoscope-assisted shunt insertion. Similar to any surgical procedure, foreign material inserted into the ventricular system increases the theoretical risk of infection. Further, neuroendoscopy would likely not be useful in cases of assessment of shunt viability for patients presenting with ventriculoperitoneal shunt failure. Attempts to cannulate a pre-existing proximal shunt catheter are likely to be unsuccessful because choroid plexus overgrowth and scarring into the inner lumen of the catheter could lead to undue trauma during endoscope insertion. Given the small caliber of the endoscope, the image quality is significantly less clear than what we routinely see with endoscopic procedures, and therefore its use is limited in patients with significant hemorrhage or protein content in the cerebrospinal fluid (CSF).

Neuroendoscopy may also be useful in cases requiring nontraditional shunting of CSF. Two reports describe the ability of the endoscope to aid in achieving proper location of catheters for cases of symptomatic posterior fossa cysts.7,8

Given that we typically only use endoscope-assisted shunt placement with patients who have complex or unusual ventricular anatomy, we recommend the use of the endoscope as confirmation of correct catheter placement and in assistance of adjustments, but advocate using concurrent assistance with ultrasound or neuronavigation to target and pass the catheter.

Torres-Corzo et al described the technique of transventricular endoscope-assisted catheter placement in the fourth ventricle for the treatment of trapped fourth ventricle using a flexible neuroendoscope (Video 25.2).9 They treated 11 patients with diagnosis of trapped fourth ventricle on whom previous endoscopic procedures (third ventriculostomy and/or aqueductoplasty) had failed. The procedure was successful in all patients and with no complications. All patients were asymptomatic at a long-term follow-up, except for two patients that remained with their baseline neurologic symptoms (one patient with VI cranial nerve palsy and one with Parinaud’s syndrome).9

25.3 Operative Techniques

The patient is brought into the operating room and placed under general anesthesia. Preoperative antibiotics are administered, and the patient is positioned in a routine fashion for the desired shunt. Prior to incision, the endoscopic equipment should be tested to ensure proper function, and the endoscope should be focused. One should note that the quality of the image is limited due to the small caliber of the scope and the optimal image is pixilated (Fig. 25.1). The patient is then prepped and draped in sterile fashion; a generous surgical field from the head to the abdomen is prepared. The operation then proceeds per the standard ventriculoperitoneal shunting protocol, with diligent attention to minimization of infection by practicing careful surgical technique.

**Fig. 25.1** Recommended method for incising the proximal shunt catheter prior to insertion. The proximal catheter is incised parallel to its long axis, creating a “fish-mouth” opening when the endoscope is inserted through the catheter.

At the time of proximal shunt insertion, the proximal shunt catheter is prepared by incising a longitudinal slit at the most proximal tip of the catheter (Fig. 25.2). Surgeons should take care to avoid incising the catheter perpendicular to the long axis of the catheter, as this creates a blunt end for the proximal tip of the catheter. Passing this blunt end through the brain parenchyma may increase the risk of cerebral injury. By incising the catheter parallel to the long axis of the catheter, the cut ends of the proximal tip will reapproximate, enabling a rounded tip to be passed into the brain.

**Video 25.1** Endoscopic shunt catheter placement. The bottom right picture depicts the endoscope progress through the catheter and the “fish-mouthed.” The ependyma is then visualized and the endoscope can be seen in and out of focus to illustrate maximal focus of choroid plexus and associated vasculature. Note the pixelated appearance when in focus.

Once the proximal catheter is believed to be in appropriate position, the stylet is withdrawn while taking care to avoid undue movement of the catheter. The endoscope is then inserted into the proximal catheter and advanced to the tip. To minimize risk of inadvertent neural injury, the approximate length of the endoscope should be labeled in relation to the proximal catheter while also verifying the position of the endoscope tip directly, as it is advanced, via the camera. As the endoscope is advanced through the tip, slight increased resistance may be felt and this creates a fish-mouth opening of the cut end of the catheter. Visualization of the location of the catheter via the endoscope enables confirmation of ventricular access and appropriate catheter positioning. Further adjustment of the endoscope focus may be required at this time (Video 25.1). The surgeon should be wary of manipulating the endoscope too vigorously and instead rely on identifying the normal choroid plexus as a landmark. The endoscope can be advanced or withdrawn to expand the field of view as well. From a frontal approach, the target of the catheter tip should be direct visualization of the foramen of Monro whereas the occipital target will rely on indirect visualization of the optimal position by following the choroid plexus. If the catheter is felt to be in a suboptimal location, the surgeon may decide to advance, withdraw, or in other fashion adjust the endoscope and subsequently the catheter to improve positioning. The endoscope is then withdrawn from the catheter, and the operation proceeds per the usual shunting protocol.

**Fig. 25.2** Intraoperative screenshot of endoscopic digital image. Note the limited image quality and the limited scope of view provided by the small caliber camera.

Surgeons should note that in the absence of any spontaneous CSF egress with initial placement of the catheter, the endoscope should not be advanced, and only very cautiously if so, given the risk of catheter malpositioning and further neural injury by advancing the endoscope.

25.3.1 Tranventricular Transaqueductal Endoscope-Assisted Catheter Placement in the Fourth Ventricle Technique

With the patient in supine position, make a standard Kocher’s point bur hole. Insert the flexible neuroendoscope into the lateral ventricle. Do an endoscopic inspection of the lateral and third ventricles. If necessary, perform an aqueductoplasty (see Chapter 22 for details). Withdraw the endoscope and insert the ventricular catheter into the lateral ventricle. The length of the ventricular catheter is longer than for a standard VP shunt. Usually a 12- to 13-cm long catheter is sufficient. Once the catheter is in the lateral ventricle, insert the flexible neuroendoscope through the same bur hole, and under direct endoscopic visualization, guide the catheter into the third ventricle and cerebral aqueduct until it reaches the center of the fourth ventricle (Video 25.2). Connect the catheter to a valve and continue the procedure as for a standard VP shunt placement.