Ventricular endoscopy in the pediatric population: review of indications

Indications

Applying ETV to the correct clinical indication is the single greatest factor influencing its success in bringing relief to the pediatric patient. While ETV is applicable to both communicating and non-communicating hydrocephalus, ideal candidates are those with an obstructive etiology due to a variety of possible causes, including primary congenital anomalies such as aqueductal stenosis, myelomeningocele, and idiopathic causes [15, 16, 17]; obstruction secondary to pineal region tumors for which ETV and biopsy may be coupled [18, 19, 20]; aqueductal stenosis secondary to tectal gliomas [21]; and giant retrocerebellar cysts [22]. Some have found ETV to be effective in managing hydrocephalus in children with posterior fossa tumors prior to tumor resection [23, 24]. ETV also provides great utility in managing patients with obstructive hydrocephalus previously treated by ventriculoperitoneal shunting who present with shunt failure secondary to obstruction, infection, abdominal CSF pseudocyst, or other complications [25, 26]. In cases where shunting can give rise to slit-ventricle syndrome, ETV has also been proven effective in assessing brain compliance. Specifically, if the brain is sufficiently compliant, the existing shunt is removed and ETV is performed during the same operation [27]. Such indications have made ETV a very valuable tool.

Operative technique

The operative technique for endoscopic third ventriculostomies has been extensively documented [28]. In our experience, use of a classical right frontal burr hole is typically employed. Access to the frontal horn of the lateral ventricle is subsequently achieved using an Elsberg cannula followed by insertion of a 12- or 14-French peel-away sheath. A 0-degree endoscope is passed through the peel-away sheet under constant irrigation with physiologic saline at body temperature. The tip of the balloon catheter or the stiff end of a vascular guide wire is used to perforate the floor of the third ventricle in the area of the tuber cinereum. Fogarty balloons or NeuroBalloons (Integra Life Science) are used to perform the ventriculostomy [29] (Fig. 1). Based on our practice, the use of diathermy is not recommended.

Open image in new window

Success rates and determinants of success

Regarding indicators of outcome, Kulkarni et al. sought to validate a success score based upon age at ETV, cause of hydrocephalus, presence of a previous CSF shunt, and center with high operative volumes (>100 ETV cases performed) and stratified patients based on their scores to retrospectively evaluate the likelihood of ETV failure during long-term follow-up. They concluded that a high ETV success score correlated with a greater reliability of ETV compared to CSF shunts immediately after surgery, whereas this benefit was not attained among lower ETV success scorers until 3 to 6 months after surgery [30]. In determining the ETV success score, patient age was implicated as the strongest predictor of success; older children (>6 months) typically experienced superior outcomes versus their younger counterparts [31]. Peretta et al. also reported similar results for both primary and secondary ETV procedures, as a repeat ETV was found to be more successful in establishing long-term shunt-free resolution of hydrocephalus following initial ETV failure when the initial ETV was performed at age 2 or older and in the absence of arachnoid scarring [32]. For children under the age of 2 years, probability of ETV success increases gradually over the first few months of life [16]. Ogiwara et al. reported similar observations, noting ETV to be an effective first treatment option for hydrocephalus in patients under 6 months old, particularly for those between 3 and 6 months of age [33]. However, with results in direct contrast to aforementioned sources in such an age group, the controversy as to whether ETV is superior to shunt placement in infants continues. Although precise etiologies remain controversial, some posit that ETV failure in the first few months of life may be related to CSF pathway reclosure due to a tendency to form new arachnoid membranes in younger infants. Another theory suggests that the open cranial suture in infants with a relative low intraventricular pressure leads to a limited differential pressure gradient across the third ventricular floor. This would favor failure of the stoma. A limited CSF absorptive capacity in the first few months could also potentially contribute to this ETV failure [34].

Other factors influencing ETV success in infants include gestational age at birth, etiology and manifestations of the first episode of hydrocephalus, degree of choroid plexus cauterization, intraoperative assessment of cerebral aqueduct patency and cisternal scarring, and thickness of the third ventricle floor, with thinner floors correlating with improved outcomes [35, 36, 37, 38]. Additionally, more recent outcome analyses based on the work of Warf in Uganda seem to suggest that ETV alone is generally not as effective in young infants under the age of 1 year versus their older counterparts, with the sole exception being post-infectious aqueductal obstruction, where ETV was equally effective across age groups [39]. In all other instances, as described below, Warf found ETV in combination with choroid plexus cauterization (CPC) to be superior to ETV alone, potentially avoiding the long-term dangers of shunt dependency in both industrialized and developing settings [14, 39].

Identifying metrics of improvement is important to assessing operative success rates. Considerable controversy exists regarding which metrics are most appropriate. A reduction in third ventricular volume as calculated on CT imaging is a helpful tool to follow improvement in patients after ETV. Schwartz et al. noted that degree of change correlated well with the duration of preoperative symptoms [40]. Third ventricular width and mid-sagittal cross-sectional area measured on magnetic resonance (MR) imaging can also be used to assess the success of ETV [41]. Ventricular volume after ETV decreases postoperatively but remains at supranormal values stabilizing at 3 to 6 months [42]. Volumetric assessment of CSF spaces in post-ETV MR imaging shows enlargement of subarachnoid spaces in concordance with decreases in ventricular volume as an indicator of ETV efficacy [43]. Garg et al. found that decreased third ventricular volume post-ETV does not appear to correlate well with clinical improvement and suggested that improvement in cerebral perfusion on SPECT imaging may be a better surrogate measure [44].

Quantitative analyses also prove vital when comparing ETV to other techniques such as ventriculoperitoneal shunts. Drake et al. conducted a large retrospective comparison of these techniques with results corroborating the general hypothesis that ETV success correlates with patient age. However, they also noted that that ETV provides no additional benefit to patients in terms of quality-adjusted life years (QALY) 1 year following surgery when compared to shunting [45]. Similar studies seem to report a similar lack of benefit in quality-of-life following ETV versus shunting [46]. Such findings underscore the need for continued research on quality indicators and reviews to optimize successful outcomes for pediatric hydrocephalus.

Complications

Though less invasive than non-endoscopic approaches, ETV nonetheless carries an appreciable risk of intraoperative and postoperative complications. Earlier reports of postoperative complications in the early stages of pediatric neuroendoscopy included hyperphagia, diabetes insipidus, CSF leaks, and amenorrhea [47]. Our review of recent literature (Table 1) reveals that some of the more common complications include CSF leaks, ventriculitis, meningitis, intracranial hemorrhage, and early ETV failure. Of these, hemorrhage is perhaps the most challenging intraoperative complication to manage due to the difficulty of obtaining hemostasis with endoscopic instruments [48]. As reported by Peretta and colleagues, the overall complication rate for such operative procedures is roughly 7.3 %, with CSF leaks, aborted procedures, and intraventricular hemorrhages reported as the most common causes of morbidity [49]. A number of other reported complications, albeit far more rare, include subdural hygromas, wound infections, and endocrine disturbances [48, 49, 50, 51, 52]. Regarding associated endocrine dysfunction secondary to intraoperative manipulation, the precise mechanism remains unknown, but possible involvement of the tuber cinereum, a component of the hypothalamic–pituitary axis, is implicated [53]. Similar complications and complication rates have been reported in the septostomy literature, including injury to structures on the lateral ventricle floor and third ventricle roof, meningitis, and significant intraventricular hemorrhage [54, 55]. Ultimately, when employed in appropriate clinical settings by highly trained and experienced personnel, and in conjunction with surveillance to preempt emerging morbidities, procedures such as ETV and endoscopic septostomy may be performed with fewer and less severe complications, thereby enhancing benefits to patients.

ETV and choroid plexus cauterization

Early attempts at hydrocephalus management involved choroid plexus resection or cauterization. In 1922, Dandy utilized a rigid ventriculoscope to endoscopically extirpate the entire choroid plexus in a case that was ultimately unsuccessful [3]. However, such efforts led to the development of endoscopic CPC techniques for hydrocephalus management, as first performed by Putnam in 1934 [56]. Putnam, Scarff, and Feld subsequently developed improved endoscopes enabling them to perform CPC in a limited number of patients with success rates as great as 80 % between 1946 and 1952 [57]. Despite these attempts, mass appeal to implement endoscopic CPC remained limited for several decades given mixed long-term results, poor technological capacity, lack of appropriate patient selection, and a high degree of mortality.

With advances in endoscopic technology, endoscopic CPC enjoyed a re-emergence in the 1970s. Pople et al. published their results on the use of CPC in 104 pediatric patients with hydrocephalus where they were able to achieve long-term, shunt-free hydrocephalus control in 35 % of patients [58]. The low success rates of CPC alone may be enhanced if used in combination with ETV. More recently, Warf has been credited for introducing CPC in combination with ETV for hydrocephalic infants with a low degree of mortality and infection, even in resource-limited settings [14, 39, 59]. In his work, comparing outcomes of 266 children undergoing ETV-CPC compared to 284 children undergoing ETV alone, the success rate for ETV-CPC was significantly higher than ETV alone among infants younger than 1 year of age. Use of a flexible endoscope and electrocautery attached to a Bugby wire enabled cauterization of the choroid plexus in the lateral and third ventricles in addition to the temporal horns. A septostomy was performed to gain access to the contralateral choroid plexus [14]. A subset of infants with non-postinfectious hydrocephalus with closed aqueducts as well as infants with post-hemorrhagic hydrocephalus do not seem to respond well to ETV-CPC procedures. Chamiraju et al. studied ETV-CPC in 27 infants with post-hemorrhagic hydrocephalus of infancy and found a 63 % failure rate. The majority of such failures occurred within the first 3 months of treatment [60]. Some authors have suggested using MRI FIESTA imaging to identify the subset of post-hemorrhagic hydrocephalic infants that could respond well to ETV-CPC, especially those with a lack of cisternal scarring [61]. Patient selection in terms of aqueductal patency and cisternal scarring appears to be paramount in achieving success with using ETV-CPC procedure [38].

Future avenues

Despite centuries of experience with hydrocephalus and decades of experience with ETV, much remains to be uncovered in order to achieve significant improvements in patient outcomes following therapeutic intervention. Important recent advances to standard clinical technique, particularly in terms of navigation and visualization, include virtual neuroendoscopy using magnetic resonance and 3-D ultrasound images to plan approaches for complex anatomy, electromagnetic guidance to obviate the need for rigid head fixation in the pediatric setting, and the introduction of high-definition camera equipment for superior visualization of the operative field [62, 63, 64].

Collaboration across disciplines is required to enhance our basic science understanding of the pathophysiology of hydrocephalus, to develop more powerful mathematical models to identify appropriate candidates for the various treatment options, and to devise the technologies required to implement those treatment strategies in the safest, most effective way possible [65]. Recent literature suggests the possible utility of transforming growth factor beta-1 (TGF-β1) in differentiating between obstruction versus hyporesorption as the etiology of post-hemorrhagic hydrocephalus in premature newborns, acting as a biomarker to both highlight a potentially important factor in condition development as well as treatment guidance to resolve it [66]. A collection of advanced neuroendoscopes, imaging platforms, and robotic devices, in addition to broadened training opportunities, will hopefully continue to improve outcomes and extend available techniques to patients in more limited clinical settings. Similarly, other exciting possibilities in hydrocephalus treatment on the horizon include in utero neuroendoscopic treatment of the pathology, which, if successful, could halt the disease process prior to the onset of irreversible brain damage [13, 67].

Arachnoid cysts

Arachnoid cysts are congenital, typically asymptomatic cystic collections of cerebrospinal fluid within the subarachnoid space first described by Bright in 1831 [73, 74]. They account for 1 % of all non-traumatic intracranial mass lesions, usually arising within and expanding the margins of CSF cisterns rich in arachnoid [75, 76]. Given such a mechanism of pathophysiology, arachnoid cysts may occur anywhere, but most commonly present within the middle cranial fossa [77, 78, 79].

When symptomatic, arachnoid cysts typically present with local mass effect on neural tissue and obstruction of CSF flow and macrocephaly in younger children in addition to other potential symptoms including headache, head bobbing, focal neurologic deficits, seizures, and even psychomotor retardation to varying degrees. Additionally, there remains a risk of post-traumatic extradural and subdural hemorrhage and subdural hygromas in such instances.

In symptomatic arachnoid cysts, treatment indication is usually given without much debate. In infants, a pathologic increase (symmetric and asymmetric) in head circumference would be considered a surgical indication. However, considerable debate continues to exist in the management of asymptomatic arachnoid cysts, in infants and older children [80]. In determining whether asymptomatic cysts merit intervention, whether by endoscopic or open approach, given possible risks such as the hindrance of normal brain development, subdural hematomas, and subdural hygromas, surgeons must carefully consider each case individually. Current evidence remains limited on predictive factors that may individuate those cases that may benefit most from intervention [77, 81, 82, 83, 84, 85, 86]. The author does not advocate treatment in asymptomatic arachnoid cysts. Especially in older children, the causality of minor symptoms must be analyzed in detail to avoid unnecessary surgery.

Arachnoid cysts can be treated with a number of neurosurgical options dependent upon location. The underlying treatment goal is to drain or divert the contained CSF within the arachnoid cyst. The described approaches include stereotactic aspiration, craniotomy with excision of cyst, cyst fenestration, cystocisternostomy, ventriculocystostomy, and cystoperitoneal shunting [87]. Recent advances in neuroendoscopy have resulted in most of the arachnoid cysts being treated with endoscopic fenestration [88, 89]. This approach has the advantage of significantly reducing the size of the cyst and precludes the need for cystoperitoneal shunt placement. The fenestration must be completed either into a nearby cistern or into the closest abutting ventricle in an attempt to create a single CSF space, thus providing the greatest chance of cure.

Success of endoscopic fenestration in the setting of middle fossa arachnoid cysts is well-documented [91, 98, 99]. In work by Spacca et al., 90 % of the pediatric cohort treated demonstrated clinical improvement, and 73 % of cysts treated reduced in size or completely resolved following intervention [79]. Charalampaki et al. described a group of 13 pediatric patients who underwent neuroendoscopic suprasellar cyst fenestration; seven of these patients who had prior shunts placed remained shunt-free afterwards with no complications noted [100]. Similarly, Ersahin et al. suggested endoscopic surgery as first-line treatment for suprasellar cysts after reporting a series of 17 patients, where only three required ventriculoperitoneal shunting following ventriculocystocisternostomy [75]. These studies seem to indicate that neuroendoscopy is a safe approach for the fenestration of suprasellar arachnoid cysts with a high success rate and low likelihood of necessary postprocedural shunting. For infants with large hemispheric arachnoid cysts, the success rate of endoscopic fenestration is less favorable than in older children; therefore, some authors advocate shunting those primarily in light of low endoscopic success rates [34, 60]. The author will advocate a neuroendoscopic approach if possible. It is critical to discuss the limited success rate in infants with large cysts. The possible need of a secondary cysto-peritoneal shunting in case of neuroendoscopy failure is always discussed prior to the intervention.

While these success rates for endoscopic fenestration are encouraging, endoscopic techniques face similar postoperative challenges as open techniques for treatment of arachnoid cysts. In the Spacca series, five patients developed subdural hygromas while an additional five patients developed postoperative subdural hemorrhage requiring surgical treatment [79]. Cinalli et al. used neuroendoscopic techniques for treatment of interhemispheric arachnoid cysts in a series of five patients and noted success with a single procedure in only three patients. However, three patients (60 %) developed subdural CSF hygromas postoperatively due to a thin cerebral mantle and associated hydrocephalus [96].

For less common arachnoid cysts, such as those within the quadrigeminal cistern, endoscopic cystostomy demonstrates a high success rate in combination with ETV, particularly in patients greater than 6 months of age [87]. In instances where cysts may extend into the third ventricle, a bipolar fenestration into the ventricle and basal cistern can help decompress the cyst and leave the patient shunt-free [94]. While endoscopy works well for most cysts where fenestration is an option, the technique may pose a problem in the presence of intervening brain tissue. In such scenarios and for small-sized cysts within hemispheric, temporal, or posterior fossa regions, microsurgery or shunting may prove to be the most reasonable option. However, generally, neuroendoscopy is first-line therapy in the treatment of arachnoid cysts due to high success rates with minimal rates of reoperation, shorter hospitalization courses, and avoidance of additional procedures such as shunt placement or craniotomy [101, 102].

Intraventricular cysts

Previously, microsurgical resection or ventriculoperitoneal shunting were the only means of treatment for patients with colloid and other intraventricular cysts. However, with advances in endoscopy and the ability to cannulate the lateral ventricle, navigate through the cerebrospinal fluid, and achieve greater visualization of capsular vascularity, neurosurgeons are now able to treat a number of intraventricular cysts in the pediatric population, including colloid, pineal, and porencephalic cysts in an additional manner (Table 3) (Figs. 2 and 3). Most lessons in this regard have been learned from larger series derived from adult neurosurgical literature due to a paucity of literature in the pediatric population [103].

Table 3

Review of case series for neuroendoscopic therapy of intraventricular cysts

Authors	Age range (mean)	Cohort size	Cyst type	Results	Complications
El-Ghandour 2008 [154]	2 months to 3 years (12.5 months)	24	Multiloculated hydrocephalus	Hydrocephalus improved in 75 % cases; shunt avoided in 3 patients; shunt revision rate decreased from 2.9 to 0.2 per year	CSF leak (n = 2), minor bleeds (n = 2)
Gaab and Schroeder 1999 [129]	4–61 years	9	Colloid, epidermoid, pineal	All colloid cysts and epidermoid cysts resected; hydrocephalus resolved in all patients	Major bleed (n = 2), meningitis (n = 1), mutism (n = 1), memory loss (n = 2)
Delitala et al. 2004 [155]	9–72 years (50.1)	7	Craniopharyngioma	Complete drainage achieved in 85 % cases; 1 recurrence 25 months following initial procedure	None
Maqsood et al. 2006 [112]	7–18 years (13 years)	18 (4 endoscopic)	Colloid	All resected with endoscopy with no recurrence	Extradural hematoma (n = 2), seizures (n = 1), diabetes insipidus (n = 1)
Nakahara et al. 2004 [156]	46–72 years (66.2 years)	5	Rathke’s cleft, craniopharyngioma	No regrowth of cyst in 80 % cases	Mild left hemiparesis (n = 1)
Hellwig et al. 2003 [104]	16–64 years (43 years)	20	Colloid	18 excellent outcomes; 1 cyst recurrence requiring reoperation	Intraoperative hemorrhage (n = 1), CN III palsy (n = 1), meningitis (n = 1), short term memory loss (n = 3)
Lewis et al. 1994 [106]	14–57 years (40 years)	15 (7 endoscopic)	Colloid	Patients undergoing endoscopy with short length of surgery, shorter hospital stays, and faster rates of recovery versus microsurgery cohort; no difference in complication rates	Basal ganglia infarction (n = 1), memory loss (n = 1)
Charalampaki et al. 2006 [114]	12–71 years (41 years)	28 (endoscope-assisted)	Colloid	Total removal of cyst in all cases; clinical improvement noted in 96 % of cases	Memory disturbances (n = 7), psychomotor disturbance (n = 1), midbrain syndrome (n = 1)
Gangemi et al. 1999 [157]	6 months to 40 years	19	Intraventricular, paraventricular, arachnoid	All successfully treated with no shunt requirement; 1 patient underwent subsequent craniotomy for CSF leak	Rhinoliquorrhea (n = 1)

Open image in new window

Open image in new window

Colloid cysts, which are posited to derive from the aberrant migration of endodermal elements to the velum interpositum during embryonic development, can produce obstruction most commonly at the foramen of Monro, thereby leading to obstructive hydrocephalus and risk of sudden death [104, 105]. Such cysts tend to most commonly occur between the second and fourth decades of life and are relatively rare in children. A number of surgical techniques have been described in the resection of colloid cysts, including shunting, microsurgery (interhemispheric, transcallosal, or transventricular), stereotactic aspiration, and endoscopic resection.

Endoscopic approaches, including both primary endoscopy and endoscopy-assisted keyhole approaches, have enjoyed general success when indicated for the resection of colloid cysts with evidence of shorter hospital stays, rapid postoperative recoveries, and a lesser degree of complications and recurrence versus microsurgically resected tumors even during long-term follow-ups [106, 107, 108, 109, 110]. Powell et al. reported the first such successful resection in 1983 [111]. More recently, further case series have confirmed the feasibility and efficacy of lesion resection [112]. Teo described his experience with 18 patients, including one pediatric patient who underwent complete endoscopic removal of colloid cysts with no recurrence. Two patients developed aseptic meningitis postoperatively without any neurologic sequelae [113]. In the case of endoscope-assisted microsurgical approaches, Charalampaki et al. described colloid cyst resections in 28 patients with gross total resection and no recurrence. Postoperative complications included mild memory disturbances, psychomotor retardation, midbrain syndrome, and meningitis in limited cases [114].

Cystic suprasellar tumors

A wide variety of surgical approaches exist in the treatment of cystic mass lesions, but many are technically difficult and fail to remove such masses in full without significant risk to surrounding structures dictating neurologic and endocrine function. As such, endoscopic transventricular approaches, particularly for craniopharyngiomas and cystic optic hypothalamic tumors, have evolved. While literature on the indication is generally sparse, published small case series suggest endoscopic management of suprasellar cystic lesions to be an effective alternative to open approaches, with negligible rates of complication and recurrence. Recent evidence suggests that a multimodal approach incorporating neuroendoscopy alongside microsurgery and radiosurgery may lead to improved tumor control and cognitive outcomes versus radical resection alone [115, 116]. Primary treatment of the cystic component can lead to decreased mass effect and restoration of open CSF pathways, thus delaying the need for radiation therapies especially in young children.

Surgical management of pediatric craniopharyngiomas remains a hotly contested issue, with a recent shift from gross total resection, given the potential for hypothalamic insult and associated complications, to limited resection in conjunction with multimodal therapy [117, 118, 119]. One such therapy—placement of intracystic catheters to treat craniopharyngiomas via direct injection of chemoradiotherapeutic agents—can be facilitated by endoscopic guidance, thus providing a minimally invasive method to deliver potent antineoplastic agents (i.e., bleomycin) directly to the tumor site [120, 121]. The safety and efficacy of such an approach was recently documented in three cases by Mori et al. in which no complications occurred with placement of Ommaya reservoir catheters over the outer surface of a transparent endoscopic sheath [122]. Given that 90 % of craniopharyngiomas are reported to have a cystic component, such a technique provides an effective approach in lesion management [123].

Technique for cyst fenestration and decompression

The patient’s head is positioned so that it allows the shortest linear distance to the cyst. Stereotactic image guidance can be used to find the best site for the surgical incision. A 2-cm linear incision is made over the identified site of insertion. A burr hole is placed over the area, and the dura is bipolared and opened in a standard cruciate fashion. A small cortical opening is completed that allows a peel-away sheath to be introduced into the ventricular cavity. Typically, the endoscope is introduced through the peel-away sheath and the cyst wall is identified under endoscopic view. Endoscopic forceps and scissors are subsequently employed to make a small opening in the cyst wall, which is further enlarged using a double balloon catheter. In instances of a vascularized surface, cautery can be used prior to incision of the cyst wall. It is critical to have an ample opening to allow cyst drainage into a ventricle or a CSF cistern. Use of an endoscope allows visualization of the cyst contents, helps determine the site of cyst fenestration, and allows avoidance of critical neurovascular structures (Fig. 2).

For Sylvian cysts and other superficial cysts, it is important to plan the burr hole such that it creates an ideal trajectory to the basal cisterns. Stereotactic navigation can be utilized to plan optimal burr hole positioning as necessary. Particularly in the case of tumor-associated cysts (i.e., craniopharyngiomas), biopsies of the cyst wall and CSF sampling for further analysis are recommended.

Technique

The technical approach for endoscopic biopsy of pediatric brain tumors is determined by the location in or adjacent to the ventricular system. Lesions near the foramen of Monro and the frontal horns of the lateral ventricle can be approached with a standard ETV frontal burr hole just anterior to the coronal suture (Fig. 4). Lesions located more posteriorly within the third ventricle require a more anterior frontal burr hole to allow easy access with the rigid endoscope. The degree of angulation and maneuverability is limited with the rigid endoscopes, and, in some cases, two burr holes may be required for ETV and biopsy, respectively (Fig. 5). Stereotactic image guidance can be extremely useful in planning trajectories and the best location to place burr holes. Biopsy forceps and scissors can be used through the endoscope port to obtain an adequate biopsy specimen.

Open image in new window

Open image in new window

Some tumors may be hemorrhagic and bleeding may be encountered at the time of biopsy. In such situations, we recommend active irrigation with saline at the site of the biopsy to allow the blood to be washed away and to allow clear visualization of the bleeding site. Electrocautery can be extremely useful in these scenarios.

Flexible endoscopes are an alternative to rigid endoscopes. They allow greater degrees of freedom but this is at the expense of the optic quality [125]. One approach for biopsies in the posterior part of the third ventricle in combination with ETV is to place a single more frontal burr hole. The biopsy is performed using the rigid endoscope with benefit from excellent optics. The ETV can be performed through the flexible endoscope [14].

When the feasibility of a pure endoscopic approach is limited, an open microsurgical approach may be merited in the resection of an intraventricular mass. An endoscopic-assisted endoport approach allows good microsurgical access combined with excellent lighting and visualization through the endoscope. The utility of such approaches has been noted in the resection of colloid cysts, arachnoidal cysts, primitive neuroectodermal tumors, Rathke cysts, hypothalamic hamartomas, and other intraventricular masses, and seems to represent a new approach to the management of complex ventricular lesions [100, 114, 126, 127, 128].

Success rates

In experienced hands, neuroendoscopic biopsy and resection of pediatric brain tumors appears to be an increasingly safe and effective procedure [51, 55, 129, 130, 131, 132]. The risks of clinically significant intraoperative hemorrhage necessitating abandonment of the procedure and postoperative bleeds are reported to be 2.3 and 3.5 %, respectively [133]. The results of endoscopic biopsy in the setting of hydrocephalus are largely favorable, with recent studies citing success rates as high as 96.0 % [55].

In a recent multicenter study, Constantini et al. found that the technique provided meaningful pathological data for the majority of patients analyzed across a wide range of tumor types and locations with minimal morbidity and mortality. In analyzing 293 patients who underwent endoscopic biopsy for lesions across 13 centers in nine countries, biopsies were noted to be informative in 265 patients (90.4 %). In 14 cases (17.9 %) of those who underwent subsequent open surgery, a discrepancy was noted between sample pathological diagnoses, particularly in cases where biopsy results seemed to indicate a low-grade glial tumor. Of note, only one death was reported, which was attributed to significant intraoperative bleeding [134]. Such findings supporting the diagnostic accuracy of neuroendoscopic biopsy have also been reported elsewhere, particularly as advances in navigation-assisted systems have come to the forefront over the past decade [135, 136].

Limitations

While the advantages of combined ETV and tumor biopsy are obvious, the technique is limited to tumors that are intraventricular or protrude into the ventricular space. Other locations require the use of a conventional stereotactic biopsy, open craniotomy, or a transsphenoidal approach. A significant limitation for tumor resection is the narrow working channel with small instruments and their very limited degrees of freedom. Previous reports recommend such tumors should not exceed 2 cm in diameter given the degree of intraoperative time involved to resect such masses [129]. In cases where a pure endoscopic option is limited by anatomical barriers, such as small lateral and third ventricular spaces, combined microsurgical and endoscopic resection may prove of greater utility [126]. However, absence of hydrocephalus is not a contraindication to endoscopic surgery [137].

Technique

Kellnar et al. described the use of a 30-degree optic to visualize the ventricular anatomy and to achieve ideal catheter tip placement anterior of the choroid plexus at the level of the foramen of Monro [139]. Further development of this technique led to the introduction of disposable flexible endoscopes that can be inserted into the ventricular catheter instead of the stylet [140]. The optics and visualization through these disposable endoscopes are, however, limited.

Additional applications

In addition to the insertion of new catheters for hydrocephalus [140] or for the administration of antineoplastic agents directly to tumor sites [120], endoscopes can be used for the retrieval of failed ventricular catheters. Particularly in the setting of multiloculated hydrocephalus where multiple catheters were placed in the past, retrieval of nonfunctional catheters can be accomplished in an endoscopically assisted manner (Fig. 6). This can be especially important in cases of meningitis and ventriculitis where removal of the potential nidus of infection is of paramount importance [141].

Open image in new window

Success rates